key: cord-0010206-r0tl01kq
authors: nan
title: Dublin Pathology 2015. 8th Joint Meeting of the British Division of the International Academy of Pathology and the Pathological Society of Great Britain & Ireland
date: 2015-09-15
journal: J Pathol
DOI: 10.1002/path.4631
sha: f8fee87468d60e19bc12d9a88c189abcda28b98c
doc_id: 10206
cord_uid: r0tl01kq

nan

The Christie NHS Foundation Trust, Manchester, UK One controversy in preparing the RCPath "Dataset for tumours of the urinary collecting system [2nd edition])", April 2013 was which WHO grading scheme (1973 or 2004) to use for urothelial carcinoma. Since there is a split within grade 2 (1973) between low grade & high grade, use of both schemes in parallel was recommended. This has the advantage of better indicating where a particular patient lies in the grading continuum & minimises the consequences of 'grading error' for cases close to the threshold between low & high grade in the 2004 scheme, (a critical distinction for management if 2004 WHO grading is used in isolation). The 2015 NICE guidance on bladder cancer (http://www.nice.org.uk/guidance/ ng2) incorporates risk stratification tables for Ta/T1 bladder cancer utilizing this parallel WHO grading recommendation in a multiparameter formula that also considers tumour size, pT classification and the presence/absence of CIS or aggressive subtype(s). Amongst the more aggressive bladder cancer subtypes is invasive micropapillary urothelial carcinoma and the nested variant. A bladder origin can be difficult to recognise at metastatic sites, especially for micropapillary and discohesive/plasmacytoid subtypes. Uroplakin II has recently become available, is more sensitive than uroplakin III & may assist within a panel. Spindle cell lesions of the bladder are often difficult & loss of cytokeratin expression is common in sarcomatoid carcinoma. There is potential to mistake inflammatory myofibroblastic tumour for a malignant tumour. Two molecular pathways for bladder cancer are recognised. FGFR3 at 4p16 is the most frequently mutated oncogene in bladder cancer & is prevalent in low grade papillary tumours. p53 loss of function mutations and loss of RB1 are prevalent in CIS. No prognostic molecular test is currently validated for clinical use in bladder cancer, though some (e.g. FISH) are in use as an adjunct in diagnosis.

Indiana University School of Medicine, Indianapolis, USA Some dogmas in testicular pathology do not hold up to scrutiny. The belief that all pure, postpubertal teratomas are malignant is invalid. Within this group there is a small subset that is benign; these may be divided into dermoid and non-dermoid types that, however, share features that distinguish them from the usual teratoma of adults. These include absence of: atypia, regressive parenchymal changes, intratubular germ cell neoplasia, and i(12p). Also they usually show organoid arrangements, and prominence of ciliated epithelium, squamous cysts and smooth muscle. Patients do not need further intervention beyond excision sufficient to establish the diagnosis. Cases often interpreted as isolated testicular polyarteritis nodosa because of the presence of fibrinoid vascular necrosis are mostly attributable to chronic, intermittent torsion. They usually present as pain-associated, palpable or ultrasound-detected masses, with the "mass" corresponding to infarct and/or hemorrhage. There are associated chronic vascular changes, with frequent marked intimal hyperplasia of arteries, mural fibrosis of veins, dilated venules and arteriolar hyalinization, consistent with torsion-induced venous outflow obstruction and secondary arterial hypertension. These patients do not develop systemic vasculitis on follow-up. Many "sarcomas" in patients with germ cell tumours, especially after chemotherapy, are more correctly regarded as sarcomatoid yolk sac tumours. They are reactive for cytokeratin and glypican 3. They often show characteristic features: nodular growth, myxoid and fibrous stroma, spindled and epithelioid cells, abrupt changes in cellularity, and tumor "ringlets." Most are high grade and aggressive. Most regressed germ cell tumours can be recognized through a combination of findings, although the only diagnostic ones are a scar with coarse intratubular calcifications or with intratubular germ cell neoplasia.

Alcohol consumption has doubled in Ireland and the UK in the last 60 years. The causes of this increase include increased affordability of alcohol and its widespread availability. As a consequence, the health harms associated with alcohol have dramatically increased. Binge drinking and alcohol consumption among by women have risen especially dramatically. For example, the mortality from cirrhosis has doubled in the last 20 years in both men and women. In response to this, the medical profession on both islands have led informal and later formal campaigns to encourage policy change regarding alcohol at a national level. In Ireland, this was driven by RCPI. The involvement of the medical profession has had a powerful influence, as doctors do not have a conflict of interest in this matter, in contrast to the alcohol industry. The policy changes advocated include particularly Minimum Unit Pricing (MUP), which has been shown to be effective in reducing alcohol consumption, alcohol-related admission to hospital and crime in Canada. Modelling of the data suggests it would have similar benefits in UK and Ireland. This is regarded as the single most important first step. Other steps which will help include actions around alcohol labelling, availability and breaking the link between alcohol and sports and leisure promotion. Turning off the tap of cheap alcohol will hopefully soon reduce alcohol health harms in UK and Ireland  Barrett's Oesophagus: An Evolving Challenge for the Gastroenterologist P DOT O'Toole

Cost-effective surveillance programmes for Barrett's oesophagus (BO) needs to be focused at risk groups. It is not only a question of identifying more individuals with BO (initial screening) but screening and subsequent surveillance has to identify at-risk individuals with BO who can benefit most from surveillance or therapy. Advances in endoscopic imaging (high resolution endoscopy ([HRE] with dye-based chromoendoscopy, electronic chromoendoscopy, and autofluorescence,...) certainly prove beneficial in better detecting dysplasia within known BO and can guide sampling and subsequent therapy. Dysplasia can be patchy and easily missed during routine biopsy sampling of BO and adequate training with high resolution instruments is needed to increase detection rates. Once dysplasia is detected, endoscopic ablation is recommended. Until recently, the standard treatment for HGD was oesophagectomy but endoscopic resection and ablation techniques are now available to eradicate dysplasia and mucosal adenocarcinomas. Resecting visible lesions (using endoscopic mucosal resection [EMR] or endoscopic submucosal dissection techniques) allows full pathological T staging. When invasive cancer is eliminated at multidisciplinary review (i.e., purely mucosal neoplasia confirmed with a nodal risk <2%), further endotherapy to ablate residual metaplasia in BO can be performed using radiofrequency ablation (RAF). In a tertiary centre use of staging EMR is frequently necessary (>60%) in patients referred for endotherapy and expert endoscopy is required to ensure safe and complete oncological resection. Conversely pT1 submucosal cancers detected following EMR are confidently triaged for oesophagectomy. Combination of EMR and RFA in expert groups exceeds >95% for eradication of neoplasia and metaplasia. Adverse events are quite low (stricture form healing, 4%; self-limited haemorrhage, very rare perforations ...) and recurrence of BO is also low (<10% at 5 years). Careful follow-up endoscopies is necessary at 3 to 6 months initially; intervals theerafter probably yearly.

 The Pathologist's Role in the Diagnosis and Management of Neoplasia in Barrett's Oesophagus P C Muldoon St. James's Hospital, Dublin, Ireland The last decade has seen a revolution in the management of neoplasia in Barrett's oesophagus. More detailed biopsy protocols, along with the advent of sophisticated local resection and ablation techniques, have radically altered the management of this expanding cohort of patients. These new treatment modalities, coupled with a massive increase in the numbers of cases of Barrett's being diagnosed, have significantly altered the demands placed upon pathologists involved in this area. These changes have presented pathologists with an opportunity to challenge our existing practices, to improve the reproducibility of our analysis and to increase the clinical relevance of the way in which we report neoplasia in this setting, where the pathologist plays a critical role in the multidisciplinary management approach. This talk aims to outline a practical approach to the handling of these specimens and to provide clear guidelines as to how to report them in the most clinically useful way.

 Modern Management in IBD P MWR Vieth 1 ; H Neumann 2 1 Klinikum Bayreuth, Pathology, Bayreuth, Germany; 2 University Hospital Erlangen, Medical Clinic I, Erlangen, Germany Ulcerative colitis and Crohn's make a distinct histological picture. Around 80% of an IBD diagnosis is the clinical information and about 20% derives from histology, only. For routine purposes it is recommendable in case of a first manifestation of an IBD to make a diagnosis such as: "picture of ulcerative colitis or picture of crohn's disease" and to recommend a follow-up endoscopy with biopsies not prior to 8 weeks after the first endoscopy and than confirm the diagnosis later to exclude mimickers of IBD. In Crohn's disease it is helpful to take biopsies from the upper GI-tract to get further hints of Crohn's disease. In case of neoplasia, the guidelines leave some room for local endoscopic treatment of low grade dysplasia whereas high grade dysplasia is still seen as an indication for operation since there is a high probability of detecting a carcinoma in the operation specimen afterwards. Operation means on normal complete proctocolectomy. This has been indivdually questioned in the last time. There are exceptions for discussing the indication of an operation in IBD: cases with numerous pseudopolyps that cannot be searched for neoplasia, low grade lesions that cannot be completely removed, multiple neoplastic lesions and unresponsiveness to medical treatment.Operation and endoscopic specimen in IBD need a subtile search for neoplastic lesions. A microscope with reverse light may help to identify suspicous lesions and may help to decide where exactely to cut a specimen. In conclusion a tight cooperation between clinical and histopathological partners is recommended to reach a high standard for patient care. Second opinions may help to achieve and fuel the own learning process esp. in an institution with a lower number of IBD patients during the year.

Colorectal or bowel cancer screening (BCS) is commonplace and organised national screening programmes have been developed in many countries, most notably in western Europe. Traditionally, faecal occult blood (FOB) detection has been the screening method of choice, those testing positive being selected for subsequent colonoscopy, but this is changing, with alternative or additional screening tests gaining favour. Most of the problems in BCS pathology practice are particularly related to FOB-based screening programmes, as these are enriched for large, bleeding sigmoid adenomas, in comparison to programmes utilising endoscopy as the primary screening modality. Experience within the closely related UK BCS programmes to date has yielded several recurring problems: the diagnosis of stage pT1 or 'polyp' cancers, in particular distinguishing common epithelial misplacement from 'true' invasion; the management of stage pT1 cancers, in relation to indications for surgical intervention after such a diagnosis; and the minimum criteria for a biopsy diagnosis of colorectal adenocarcinoma. These issues will be discussed with illustrative examples, along with the somewhat more mundane but highly important practical issue of measuring various parameters related to BCS pathology. The importance of quality assurance measures to ensure high standards within BCS pathology is emphasised.

With the introduction of colorectal cancer screening in various countries of the EU there is a sharp increase in the incidence of early colorectal cancer. A significant part of these early tumours presents in a pedunculated polyp. In most cases, these carcinomas are already completely removed by polypectomy. Classic risk factors that suggest a high risk for lymph node metastases include Haggitt level 4, positive resection margins, poor differentiation and lymphatic or vascular invasion. However, the evidence is rather thin. Most pT1 studies are performed on sessile polyps. Risk factors are more firmly established and include differentiation grade, lymphatic invasion, Kikuchi level sm3 or the presence of budding. However, for a clinical useful decision model, we will need an integrated approach, and both specificity and sensitivity of the various factors should be taken into account. Radical surgery seems overtreatment for a large number of polyp cancers.

Leeds University, Leeds, UK Minimum datasets have changed cancer reporting. This talk will explain the decision making processes behind the latest datasets both for cancer reporting and bowel cancer screening. It will also look at where we may be going in the future for staging and molecular reporting.

The intestines play host to a broad spectrum of infective organisms ranging from viruses, through bacteria , fungi and unicellular parasites to worms. The spectrum of infections seen varies with geographical location, due to socioeconomic factors and due to changes in human behaviour. Immunocompromisation due to infections (in particular HIV), malignancy (especially haematological tumours) and the use of immunosuppressive drugs also has an important role in determining the infections commonly seen in the GI tract. The typical pathological features of intestinal infections will be discussed together with suggestions on how to optimise the diagnosis of such pathologies.

Erasme University Hospital, Brussels, Belgium

Pathologists are confronted with different types of colitis, most commonly infectious colitis and inflammatory bowel disease (IBD) followed by microscopic colitis and ischaemic colitis. Several other forms of colitis, however, exist and might be underrecognised; these diseases include segmental colitis associated with diverticulosis, diversion colitis, eosinophilic colitis and Behcet's colitis. Clinical presentations of these rare types of colitis vary, and laboratory data are often non-specific; mucosal biopsy is essential in establishing the diagnosis. Segmental colitis associated with diverticulosis (SCAD) is mainly characterised by the involvement of the sigmoid colon with sparing of the rectum and proximal colon. SCAD often mimics IBD at endoscopic and histological examination; since SCAD has a self-limited course that resolves without further recurrence or need for treatment, the implications of an inaccurate diagnosis are obvious. Diversion colitis is a non-specific colonic inflammation following surgical diversion of the faecal stream. Is is characterised by a chronic lymphoplasmacytic infiltrate, and the existence of lymphoid follicular hyperplasia is considered to be a hallmark feature. The development of diversion colitis is attributed to a lack of short chain fatty acids. Eosinophilic colitis is etiologically obscure and can be associated with involvement of other sections of the gastrointestinal tract. An infiltrate of eosinophilic granulocytes is found to varying degrees in all wall layers. A history of food intolerance or allergy is present in most of the patients, and peripheral eosinophilia is present in 80% of cases. Gastrointestinal involvement has been reported in up to 25% of patients with Behcet's disease. In cases with ileocolonic involvement, it is often difficult to distinguish Behcet's disease from other inflammatory bowel diseases. The diagnosis, therefore, often depends on clinical manifestations and intestinal ulcerative lesions.

S·8  How to Write a Paper and Get it Published P CS Herrington 1 ; P DM Berney 2 1 University of Edinburgh, Edinburgh, UK; 2 Barts Health NHS Trust, London, UK Scientific papers have a predetermined structure, and writing in this way requires practice. Most Journals accept only a small fraction of submitted papers and it is important that any paper has something specific to say; and says it in a clear and concise way that can be understood by editors, reviewers and readers, all of whom play a role in assessment of its contribution. Editors look for novelty and significance in the context of the aims and scope of their Journal; and scientific rigour, which expert reviewers help them to assess. Writing a paper and having it assessed by a Journal is an iterative process. During the writing phase, the scientific rigour of the argument can be refined; and following submission and peer review, reviewers and editors often make constructive comments that help to improve it still further. The peer review process therefore acts not only as a quality filter but also as a mechanism for quality improvement. Writing papers and submitting them for publication is therefore generally a positive experience, particularly if one remembers that the process is iterative and (inevitably) not all papers will be accepted for publication by the first Journal that they are sent to.

 Large-Scale Routine Diagnostics Using Whole-Slide Imaging in Sweden -the Linköping Experience P C Lundström CMIV, Linköping University, Linköping, Sweden This presentation will describe the large-scale routine usage of WSI at Linköping University Hospital, Sweden. Since 2011 all histology slides are scanned, amounting to more than half a million slides to date. To a significant extent the digital images are used for primary review. The initial implementation led to several of the benefits foreseen with digital pathology, but it could also be concluded that further development was needed to unlock the full potential, in particular within the IT solutions. Therefore, a consortium led by CMIV, Linköping University was formed in 2012 to create innovations for a new generation of digital pathology. This triple helix consortium also includes industry and more than half of Sweden's health care providers, an engagement that reflects the dominating view in Swedish pathology that large-scale adoption of WSI practice is possible and desirable. This talk covers the experiences made during the initial digitization, including laboratory process adjustments, and the later additions to the digital pathology toolbox accomplished by the ongoing innovation project. Apart from obvious targets such as the pathologists' workstation, the developments also touch upon other areas including grossing and enterprise image management.

 Digital Pathology -Are We There Yet? P SM Hewitt National Cancer Institute, Bethesda, Maryland, USA

The implementation of whole slide imaging for diagnostic histopathology is far more complex than connecting an instrument to a server, and placing a computer on a pathologist's desk. The technology is additive to the histology workflow, with additional cost beyond the current practice of review with a microscope. The adoption of Digital Pathology for histomorphologic diagnosis requires the restructuring of the workflow, additional technology advances beyond the imaging instrument, and development of new tools to assist the pathologist. The end goal is to improve pathologist's productivity and provide additional diagnostic information. Digital Pathology, to succeed must become a value-added proposition. The adoption of Digital Pathology requires: 1)Improvements in scanner performance as measured by defined quality metrics. 2)Advancement in server and networks to distribute images to the desktop efficiently. 3)Software to facilitate review and diagnosis, beyond presenting only and image of the slide. Evolution of the current technologies is required to provide an economic impetus for widespread adoption and use of Digital Pathology in the diagnostic setting.

To a significant extent, the distinction between melanocytic naevi and malignant melanomas is based on tissue architecture. Amongst the best known architectural features pointing to malignancy are absence of lesional symmetry and maturation, and presence of melanocyte ascent. However, each of these three features has significant pitfalls. As a rule, naevi are 'roughly symmetrical' and melanomas are not, but there are asymmetrical naevi (traumatized naevi, most larger congenital naevi; some combined naevi; some large acral and genital naevi) and symmetrical melanomas (including many small melanomas, especially small nodular melanomas; some spitzoid melanomas). In addition, it is not always clear whether a lesion should be considered 'roughly symmetrical' or not. I suspect that not uncommonly, a diagnosis is reached first, and the verdict regarding symmetry is adjusted according to that diagnosis. Similar caveats relate to absence of maturation as an indicator of malignancy. It is seen in blue naevi and all its variants; deep penetrating naevi; some BAP1 naevi. Melanomas not uncommonly feature smaller cells in their deeper parts, or there may be an underlying naevus remnant with smaller cells. Naevi with ascent include many Spitz naevi; Reed naevi; some naevi in early infancy; traumatized naevi; naevi of acral skin. Over-interpretation of ascent may result from inexperience with Melan-A and some other immune stains. Melanomas devoid of ascending melanoma cells comprise a wide variety of subtypes including, desmoplastic melanomas and, vexingly, some spitzoid melanomas. These architectural features must, therefore, be evaluated in the context of all other findings, and with a 'splitter's' mind set, taking into account the individual characteristics of the specific naevus and melanoma variants that are of relevance to the case under study.

Most melanomas are fairly easy to diagnose on histological grounds. However, melanoma is a tumour that can histologically mimic almost any other tumour including epithelial and mesenchymal neoplasms. Pathologists need to familiarize with the wide histological appearances of melanoma to avoid serious misdiagnoses. Of crucial importance is the knowledge that a number of melanomas can closely mimic benign naevi. Some variants of melanoma represent distinctive clinicopathological entities and these include desmoplastic melanoma, "malignant" blue naevus, pigment synthethizing melanoma, naevoid melanoma, spitzoid melanoma and epidermotropic metastatic melanoma. Tumoral melanosis refers to complete regression of a melanoma, a diagnosis that it is often missed because of the absence of tumour cells within the regressed area. A small percentage of melanomas display focal or extensive histological changes that closely mimic other neoplasms and often a combination of histological features with immunohistochemistry is necessary to arrive to the correct diagnosis. Microscopic variants of melanoma include adenoid (pseudoglandular), angiotropic and angiomatoid, signet ring cell, balloon cell, clear cell, rhabdoid and follicular (with exclusive involvement of hair follicles). Some melanomas display heterologous differentiation also known as transdifferentiation. The latter should not be confused with the so-called collision tumour in which a melanoma co-exists with a neoplasm of different lineage. A wide variety of heterologous differentiation has been described in melanoma including osteosarcomatous and chondrosarcomatous (mainly seen in acral melanomas), meiomysarcomatous, rhabdomyosarcomatous, neuroendocrine, ganglioneuromatous and even epithelial. Except for desmoplastic and pigment synthetizing melanoma, all other variants of the tumour have the same behaviour as ordinary melanomas.

Melanocytic tumours with Spitzoid features represent one of the most challenging and controversial areas in Dermatopathology. What is currently known as Spitz naevus was initially reported as "juvenile melanoma" by Sophie Spitz on 1948. She recognized the relatively indolent but somewhat unpredictable behaviour of these distinctive melanocytic lesions that are particularly common in young children. Over the years, the histological spectrum of these tumours was expanded, and it has become clear that classical Spitz naevi follow an entirely indolent disease course. The prognosis of tumours with atypical histological features remains somewhat unpredictable. This presentation will give an overview of the morphological spectrum of Spitzoid melanocytic tumours, their behaviour and recent advances of their molecular characteristics.

Optimisation is a core tenet in radiography and involves the radiographer ensuring that images of diagnostic quality are produced with minimum radiation dose burden to patient and staff [1, 2] . In paediatric practice this is particularly important due to the more radiosensitive nature of the child [1] . In alignment with the ISRRT 2013 World Radiography Day theme 'Radiographers Optimise Dose', radiography students in an institution submitted clinical case study coursework that focused on paediatric radiation dose optimisation.

The purpose of the current study was to analyse these case studies as examples of prevailing radiographic practice and to compare students' perception of optimisation with the evidence within each case. The evidence of optimisation was established through independent and objective image analysis along with thematic analysis of the case commentaries. The case study evidence demonstrated that optimised techniques were generally well implemented. The exception was collimation, which was sub-optimal in 84% (n=31) of the examinations, and on average irradiating an area 27% larger than necessary. Students were generally able to correctly identify techniques as optimal or not. However, when appraising exposure, positioning and collimation, between 9% and 33% of students were inaccurate in their assessment of what is optimal. Overall the study reflects positively on current Irish paediatric radiography with regard to dose optimisation, although more accurate collimation needs to be practised. Similarly student perceptions show good understanding of optimal techniques, although appreciation of exposure, positioning and collimation errors could be improved. The aim of this project was to design and prototype immobilisation devices for children who are unable to independently maintain upright sitting posture during radiographic investigations. While current market devices exist, they are seldom used by radiographers -particularly in Europe as their methods of restraint have been deemed 'culturally unacceptable' with some claiming that they are in violation of the human rights of the child [1] . The design challenge was to create devices that were functional (fit for purpose [2], radio-lucent, compliant with infection control and easy to use) while minimising discomfort and intimidation. A search of the literature, prior art, patent landscape and current market devices was performed in order to identify product requirements. TRIZ methodologies -a problem solving, analysis and forecasting tool derived from the study of patterns of invention in the global patent literature were used to identify the physical contradictions underlying the design challenge and generate potential solutions. Eight unique concept designs were identified from these methods. These were then evaluated using Pugh Criteria -a ranking system of the relative merits of each concept based on design requirements identified. Four of the eight concepts were chosen to be prototyped: a 3-D printed seat, a swing based template, an acrylic-based support and an adaptable wheelchair. The prototypes were made in collaboration with the UCD School of Engineering and tested using paediatric phantoms in UCD Radiography department. The final prototypes will be trialled in Crumlin Children's Hospital with a view to future use and development. Thrombotic microangiopathy (TMA) is a pathology that results in thrombosis of capillaries and arterioles due to endothelial injury. It is usually characterized by an atypical haemolytic syndrome (aHUS) or thrombotic thrombocytopaenic purpura (TTP). TMA is considered to be caused by infections, drugs, autoimmunity, tumours, pregnancy, transplants and inherited abnormalities involving the alternate complement pathway. This presentation describes the pathology of TMA. It includes a retrospective 15 year study (1999 -2013) of all renal biopsies reported by one pathologist. All renal biopsy request forms and reports, where cases included light (LM), fluorescence(FM) and electron microscopy(EM), were reviewed. Cases without all 3 modalitIes (LM, FM, and EM) were excluded. 6639 biopsies were reported in the study period (328 in 1999 (328 in to 629 IN 2013 . 2105 were transplant biopsies. 284 biopsies were insufficient (LM, FM and EM all not possible. This resulted in 4250 native renal biopsies as the study group. Following review of the reports 641 cases were reported as TMA. 66 were associated with thin membrane nephropathy, 41 with minimal change disease and 24 with plasma cell dyscrasia/ B cell malignancy. This resulted in 510 cases with TMA as the only pathology reported which represents 12% of all adequate native medical renal biopsies. Clinical indications included proteinuria in 73%, nephrotic syndrome in 15%, increased creatinine in 43%, increased blood pressure in 55% and haematuria in 43% of the cases. Acute renal failure was described in 10% and HUS in just 1% of the cases. Pathological changes were predominently arteriolar sclerosis and glomerular double contours on LM with chronic subendothelial injury on EM. The conclusions of this presentation are 1. TMA is overwhelmingly a chronic lesion as seen in renal biopsy pathology. 2. It is a very common pattern of injury. 3. It is not usually associated with clinical HUS or TTP features at presentation. The era of targeted cancer therapeutics has brought forth new challenges for molecular diagnostic laboratories. The list of genes, and indeed specific mutations, that predict drug responses keeps growing, and with it grows the demand for molecular sub-classification of tumors. In colorectal carcinoma, for example, recent studies support expanding testing beyond KRAS to include NRAS and BRAF in predicting resistance to EGFR-targeted therapies. Similarly, in non-small cell lung carcinoma standard screening for EGFR mutations and ALK gene fusions may be insufficient when actionable alterations involving ROS1, RET, HER2, MET, BRAF and other genes are being targeted (successfully) in ongoing clinical trials. Fortunately, the introduction of next-generation sequencing (NGS) into the clinical laboratory is meeting the demand. Due to its quantitative output, NGS not only provides precise mutant allele ratios, but it can also be used to detect gene gains and losses. Furthermore, when applied to RNA, NGS supports the detection of gene fusions and serves in assessing gene expression levels. While NGS is a powerful tool for molecularly characterizing solid tumors, the quality of the results in large part rests on the selection of appropriate input material; therefore, review by a pathologist prior to testing remains a cornerstone to success. Other growing uses of NGS include monitoring minimal residual disease in the setting of hematologic malignancies, and in the detection of targetable mutations in cell-free DNA within the plasma. The advent of next generation sequencing has ushered in an era of tremendous potential for identifying the molecular causation of simple and complex disorders, both rare and common. The successes of next generation sequencing reflect the combined interpretive skills of geneticists, bioinformaticians and clinicians working in close collaboration. In the realm of muscle disease, we have witnessed both the strengths and the weaknesses of next generation sequencing technologies. The use of exome sequencing in patients with rare muscle disease who have been carefully phenotyped has proven to be a successful strategy for identifying causative variants in new genes as well as in known genes. In fact, exome sequencing has significantly expanded both the clinical and the histological phenotypic spectra of muscle conditions associated with causative variants in known genes. In the absence of careful phenotyping or large genetic reference data sets for identifying variants of interest, causative variants may be missed, however. The use of RNA sequencing-using RNA extracted from muscle biopsy specimens-has proven to be a powerful tool for finding causative variants affecting gene splicing or expression, which may be missed with next generation sequencing. Muscle pathology plays an essential role in complementing next generation sequencing. The deep phenotyping of patients with muscle disease relies heavily on muscle histological and immunohistochemical findings in combination with muscle imaging, clinical history and neuromuscular examination findings. Examples of how next generation sequencing coupled with careful clinical and histological phenotyping has uncovered causative variants in new genes as well as in known genes will be discussed in this talk.

Medicine, diagnostic pathology, technology, and diagnostic tests are evolving at an extremely rapid pace. Drug developers and diagnostic developers each face unique challenges. Companion diagnostic development is a key component of pharma drug development strategy. We will discuss the importance of companion diagnostics in the success of personalized medicine, the FDA position on companion diagnostics, and the role, advantages and disadvantages of tissue based companion diagnostics. Other technologies, such as Next Gen Sequencing, will increasingly be utilized in a complementary fashion along with traditional slide based immunohistochemical and in situ hybridization. The scope and limitations of available technologies will be reviewed. Diagnostic technologies of all types will complement each other to provide the most accurate diagnostic information for clinicians and patients.

Following the 2008 WHO classification of haematological malignancies there has been a greater emphasis on the integration of molecular information with clinical and morphological data not just for diagnostic purposes but also to help convey both prognostic and therapeutic information. This talk will concentrate on routine testing in the work-up of common haematological malignancies focusing specifically on clonality and translocation analysis in lymphoproliferations and mutational testing in BCR-ABL negative myeloproliferative neoplasms. Using case studies to illustrate common indications for testing this talk will also highlight some of the practical points and pitfalls in the interpretation of these tests.

Beaumont Hospital, Dublin, Ireland "Should I keep the brain?" is one of the most frequent questions addressed to neuropathologists by surgical pathology colleagues. Fears relating to inappropriate organ retention coupled with decreasing availability of expert neuropathology opinion and the widely held belief that advances in neuroimaging have replaced the brain autopsy, have all contributed to a decline in the post mortem study of human brain tissue. Leaving aside the critical relevance of neuropathology to forensic medicine, the vital role played by careful examination of the post mortem brain extends far beyond pathology and has contributed greatly to science and medicine. In general, prolonged retention of entire brains may be avoided. In hospital practice it is uncommon for a patient to die without brain imaging. Access to pre-mortem brain imaging will guide the surgical pathologist in careful and appropriate sampling of calvarial, dural, meningeal, vascular and parenchymal central nervous system components. Spinal cord examination requires prior experience in spinal cord removal but most post mortem technologists are expert in cord extraction. Sampling and appropriate processing of nerve and muscle requires prior experience or neuropathology advice. High quality photography obtained at all phases of post mortem brain examination including the coronally sectioned individual cerebral hemispheres, with retention of blocks from each of the brain lobes together with cerebellum, brain stem, vessels and dura -meninges will ensure that in the event of a neuropathology opinion benign required -that opinion will not be compromised. Specific issues which will be addressed will include the death of patients with epilepsy, dementia, stroke and undiagnosed neurological disease. The key learning objective will be to ensure that pathology trainees approach post mortem examination of the nervous system with interest and excitement. Following several high profile misdiagnoses in Ireland a national quality assurance (QA) programme for cellular pathology in 2009 was initiated with a vision of establishing a patientcentred pathologist-led framework that would enhance the quality of patient care with timely, accurate and complete pathological diagnoses and reporting. National QA guidelines were developed based on 17 key quality activities generating a total of 51 Key Quality Indicators (KQI). Examples of the quality activities include turnaround time, monitoring of amended reports, frozen section correlation and various elements of peer review. All 25 cellular pathology departments in the public state-funded hospitals participate in the programme in addition to 7 laboratories within privately-run hospitals. Each laboratory enters codes on individual cases designed to capture the relevant KQI and the anonymised encrypted QA data is then electronically extracted from the laboratory information system to a national database. This national central database is managed by a novel information technology system, the National Quality Assurance Intelligence System (NQAIS)-Histopathology, that was designed to process and display the QA data so that each individual laboratory can analyse their own data and also compare their performance to the national average for each KQI. Since 2013 complete national data has been inputted into the NQAIS system and in 2014 initial QA targets were agreed for turnaround time, frozen section correlation and rate of intra-departmental consultation (IDC). In 2015 additional targets for autopsy ICD, frozen section deferral rate and turnaround time have been added. To our knowledge this programme has enabled Ireland to be the first country to publically report national metrics on the quality of their pathology services.

 The Politics of EQA: The NHS England QA Review and its Consequences P DE Hughes

In 2013, a review of External Quality Assessment processes was carried out on behalf of NHS England. The main recommendations of this review related to the strengthening of governance of EQA, both nationally and within pathology provider organisations. Recommendations specifically affecting Cellular Pathology were: (i) Professional bodies, led by RCPath, should develop methodologies for assessing the performance of individuals in EQA schemes. (ii) All pathologists reporting pathology results and providing clinical advice should be participate in EQA schemes relevant to their practice, should achieve levels of performance determined by the professional bodies and this performance should be noted at annual appraisal. (iii) Where a need to improve performance is identified, additional remedial training should be carried out, or practice in the area of concern should be stopped until appropriate retraining has been undertaken and revalidation achieved. This process should supported and resourced by the employing organisation, as should EQA scheme participation. (iv) Interpretative EQA schemes are designed to assess and improve individual performance, and attempts at collusion are considered matters of professional probity. The professional response to these recommendations is expected to be led by the RCPath under the guidance of a newly-established national oversight group working on behalf of NHS England and should be more clear at the time of the Pathsoc conference. A key part of this response will be to separately consider the implications of this review for technical schemes, affecting laboratories, and interpretative schemes, affecting individual practitioners.

Roche Tissue Diagnostics, Tucson, Arizona, USA Tumour samples to guide treatment decisions have become of increasing significance. Most importantly the results of companion diagnostic testing directly influences the management of individual patients as more drugs are approved for treatment of specific molecular distinct subgroups. Reporting suboptimal quality test results may be harmful to the patient and cause the mismanagement of a prescribed companion drug. The consequences of unsatisfactory performance and measures for improvement are the responsibility of the laboratory. Presently, there are number of EQA schemes for molecular testing available in Europe however, their results clearly indicate the need for EQA since 10%-15% of laboratories do not carry out according to the standard set by the EQA provider or utilize standardized procedures. Continual improvement programs, internal quality control and validation program assist laboratories however; by using standardized quality practices such as ISO 15189 and the use of external quality assurance schemes can provide essential feedback to the laboratory to assure accurate molecular testing results.

 How to Run a Histopathology EQA in the Digital Age P NJ Mayer 1 ;P JD Oxley 2 1 Cork University Hospital, Cork, Ireland; 2 Southmead Hospital, Bristol, UK Interpretative EQA schemes in Histopathology were first introduced in the UK in the mid-1980s, well before the advent of the Internet and high resolution digital images. In this lecture we will outline key developments, as EQA schemes have evolved into the digital era, with particular emphasis on the National Urological EQA scheme, which we have run since 2007. We will outline the main practical issues involved in running an EQA scheme and share our personal experience of the development and introduction of the web-based EQAlite software, which is being utilised by increasing numbers of schemes. We will address the pros and cons of traditional glass slide-based circulations versus virtual circulations using scanned digital images and show how the digital archive generated from old EQA circulations has become a valuable educational and teaching resource. We will also briefly explore, from an Organiser's perspective, the major issues facing EQA schemes in the future, as EQA performance becomes more embedded into revalidation and fitness to practice.  Molecular Pathology: The Future? P CS Herrington

Molecular pathology is already central to stratified medicine. And the ability of pathologists to understand disease phenotype is essential for interpretation of the current explosion in '-omics' data. Moreover, the future of stratified medicine will require integration of information from different sources, in the context of disease phenotype, to inform patient management: pathologists are ideally placed to lead this integration. This applies not only to data derived from ex vivo cells and tissues but also to molecular imaging data, which require accurate correlation with cell and tissue phenotype for accurate interpretation. Molecular pathology is key to the future of pathology; and this future extends beyond the traditional light microscope

The following Plenary, Oral and Poster abstracts have been subjected to peer review. Hypothesis: Suppressor of cytokine signalling (SOCS) family members play a vital role in the activation of the JAK/STAT signalling pathway via a negative feedback loop and have been implicated in the development of cancers. In breast cancer (BC), SOCS2 mRNA has been correlated with oestrogen receptor (ER) positive tumours favouring a good prognosis (BMC Cancer 2007, 7:136) . This study aimed to determine whether SOCS2 at the protein level correlates with tumour morphology and low grade in BC. Methods: Differential expression analysis between tubular and grade matched NSTs were undertaken in the METABRIC cohort. Primary breast cancer tissue microarrays (n=1041) were immuno-stained for SOCS2 and expression patterns correlated with clinico-pathological and molecular variables including outcome. Results: Differential gene expression analysis on the METABRIC data identified SOCS2 as the top gene with a significant overexpression in the tubular type as compared to low grade NSTs (adjusted p value=0.004). Immunohistochemistry on the Tenovus series showed positive nuclear SOCS2 expression to correlate with tumours of low grade (p<0.0001), low proliferation (Ki67 p<0.0001), ER/PR positive (p<0.0001) phenotype and tubular morphology (p<0.0001); as well as negative HER2 status (p=0.005) and non-triple negative status (p<0.0001). Survival analysis revealed significant associations with long term breast cancer specific survival (p=0.019). Positive SOCS2 correlations were also observed with the expression of androgen receptor (AR) (p<0.0001) and STAT3 (p=0.001), further indicating its role in these two signalling pathways. Conclusions: Results from this study suggest SOCS2 to be a marker of favourable prognosis: identifying low grade, ER positive breast tumours with particular correlations to the tubular histological tumour type. Background: Ovarian cancer is the fifth leading cause of cancer in women and has poor long-term survival, in part, due to chemoresistance. Tumour hypoxia is associated with chemoresistance in ovarian cancer. However, relatively little is known about the genes activated in ovarian cancer which cause chemoresistance due to hypoxia. This study aimed to firstly identify genes whose expression is associated with hypoxia-induced chemoresistance, and secondly select hypoxia-associated biomarkers and evaluate their expression in ovarian tumours. Methods: Cisplatin-sensitive (A2780) and cisplatin-resistant (A2780cis) ovarian cancer cell lines were exposed to combinations of hypoxia and/or cisplatin as part of a matrix designed to reflect clinically relevant scenarios. RNA was extracted and interrogated on Affymetrix Human Gene arrays. Differential gene expression was analysed for cells exposed to hypoxia and/or treated with cisplatin. Potential markers of chemoresistance were selected for evaluation in a cohort of ovarian tumour samples by RT-PCR. Results: A wide range of genes associated with chemoresistance were differentially expressed in cells exposed to hypoxia and/or cisplatin. Selected genes [ANGPTL4, HER3 and HIF-1α] were chosen for further validation in a cohort of ovarian tumour samples, n=35. High expression of ANGPTL4 trended towards reduced progression-free and overall survival. High expression of HER3 trended to increased progression-free but reduced overall survival, while high expression of HIF-1α trended towards reduced progression-free and increased overall survival. Conclusion: This study has further characterized the relationship between hypoxia and chemoresistance in an ovarian cancer model. We have also identified many potential biomarkers of hypoxia and platinum resistance and provide initial validation of a subset of these markers in ovarian cancer tissues. Methods: Analysis of Affymetrix™ Human Exon 1.0ST microarray data revealed differentially expressed genes (DEGs) between a BC group and a control non-BC group. Ingenuity Pathway Analysis (IPA; bioinformatics software) was used to identify networks of the DEGs. The expression of a micro-network was validated using immunohistochemistry. Double immunofluorescence was undertaken to identify the lineage of cells expressing components of the network. Results: We identified a network of interacting genes that were upregulated in FCDIIb compared to normally formed cortex or FCD without balloon cells (FCDIIa). Some components of this network were expressed in BCs but others were expressed in novel cell populations. Double immunofluorescence identified a cell with the phenotype of a glial progenitor that was only present in FCDIIb but not in normally formed cortex. Conclusions: We have identified a novel population of glial progenitors found frequently adjacent to BCs in FCDIIb. Paracrine signaling between BCs and the novel CHI3L1 positive cells is likely to be involved in the pathogenesis in FCDIIb. Further investigations into the role of these cells would give us a better understanding of the molecular abnormalities underlying FCD and possibly provide novel therapeutic targets. Excellent anatomical knowledge of the anal sphincter complex (ASC) is essential for the treatment and understanding of low rectal and anal pathology. Some of the current descriptions of the ASC are contradictory. In this study, the three-dimensional (3D) anatomy of the ASC is described with relevance to low rectal and anal surgical pathology. Six human adult cadaveric specimens (three males, three females) were obtained from the Leeds GIFT Research Tissue Programme. Paraffin embedded mega-blocks containing the ASC were serially sectioned at 250 µm intervals. Sections were stained with haematoxylin & eosin, Masson's trichrome and Millers' elastin, from which 3D reconstructions were developed. The ASC is a complex structure, varying between individuals in the size and distribution of its layers with intermingling of fibres and inconsistency of the longitudinal smooth muscle affecting the creation of the surgical intersphincteric plane. Longitudinal fibres penetrate the internal and external anal sphincter to anchor in the submucosa and ischiorectal fossa. Striated muscle fibres from the external sphincter were identified in the submucosa in four of six specimens. The ASC is highly complex due to the degree of variation in its structure and intermingling of smooth and striated muscle fibres and their penetration of major structures. This creates potential tissue planes for the spread of infection, fistula extension and tumour spread. The complex anatomy of the ASC also impacts on the staging of low rectal cancers in this region, which requires further investigation.

P H Thorpe; A Asiri; M Akhlaq; D Jackson; M Ilyas

Cten is upregulated in a number of tumour types and in colorectal cancer expression is associated with advanced Dukes stage, poor prognosis and distant metastasis. Cten is localised at focal adhesions and regulates cell motility but knowledge of underlying signalling mechanisms is sparse. Epithelial to mesenchymal transition (EMT) is a process whereby cells acquire an invasive phenotype to aid cell migration and is found to occur in a number of biological processes including cancer metastasis. We investigated whether Cten increases cell migration through EMT pathways in colorectal cancer. Cten was forcibly expressed in colorectal cell lines and Snail expression determined by qPCR and western blot. The cycloheximide pulse chase assay was used to assess any changes in Snail protein stability. Further to this, the Transwell migration assay was performed to investigate changes in cell motility. Forced expression of Cten was shown to increase Snail protein expression in HCT116 and Caco2 cell lines. There was no change in the level of Snail mRNA suggesting that Cten regulates Snail at a post transcriptional level. Inhibition of protein synthesis confirmed this and showed that Cten regulates the stability of Snail protein. Simultaneous forced expression of Cten and knockdown of Snail demonstrated that this relationship was functionally active. Forced expression of Cten increased cell migration (p<0.05) which was subsequently lost when Snail was knocked down (p<0.001). We are the first to identify Snail as a downstream target of Cten signalling. This finding advances the understanding of cancer cell motility regulatory networks and further highlights Cten as a potential therapeutic target in colorectal cancer. Work supported by a Pathological Society grant.

Treatment Strategies for Patients with Advanced Colorectal Cancer P SD Richman 1 ; GJ Hemmings 1 ; P Chambers 1 ; M Taylor 1 ; HM Wood 1 ; E Tinkler-Hundal 1 ; K Southward 1 ; JM Foster 2 ; A Ouime 2 ; KG Spink 2 ; P Quirke 1 1 Leeds Institute of Cancer and Pathology, Leeds, UK; 2 Affymetrix, High Wycombe, UK Treatment for advanced colorectal cancer is moving to combination therapies, targeting multiple signalling pathways. Indeed, MRC FOCUS4 has been designed to assess this. We determined pTEN protein expression, and assessed this in relation to other biomarkers associated with signalling downstream of the epidermal growth factor receptor. Tissue microarrays were constructed from 2 advanced colorectal cancer (aCRC) clinical trials (FOCUS and PICCOLO) for immunohistochemistry (IHC). Mutation status of KRAS, NRAS, PIK3CA and BRAF was assessed by pyrosequencing. Copy number variation was assessed on Oncoscan® FFPE Assay Kit (Affymetrix Inc.). pTEN protein expression was correlated with mutation status, MMR status, primary tumour location and copy number. pTEN protein expression for 1288 patients showed complete loss of expression in 85/787 (10.8%) -FOCUS and 64/501 (12.8%) -PICCOLO. BRAF mutation status was significantly different between the pTEN negative and pTEN positive populations (p<0.0001), with significantly more pTEN negative tumours having the BRAF V600E mutation. Loss of pTEN expression correlated with genomic deletions involving the pTEN gene. 20/30 (66%) of pTEN negative tumours exhibited loss of the pTEN region (10q), half of which were focal deletions. Only 54/202 (26.7%) pTEN positive tumours showed deletions of this region, and none were focal events. There was no significant difference in either primary tumour site or MMR status (p=0.1765) between the pTEN negative and pTEN positive populations. Signalling pathways do not stand in isolation; they are interlinked in a complex signalling network. Current treatment interventions must target the correct pathway combinations if patients are to benefit from targeted therapy. Our data suggests a subset of patients may require dual AKT and MEK pathway inhibition, in addition to anti-EGFR monoclonal antibody therapy and inhibition of BRAF.

Zonal Differences in PD1 Expression in Centre of Tumour Versus Periphery in Microsatellite Stable and Unstable Colorectal Cancer P GM O'Kane 1 ; M Lynch 2 ; J Aird 3 ; S Hooper 3 ; C Muldoon 1 ; N Mulligan 3 ; C Loscher 2 ; DJ Gallagher 3 1 St. James's Hospital, Dublin, Ireland; 2 Dublin City University, Dublin, Ireland; 3 Mater Misericordiae University Hospital, Dublin, Ireland Colorectal cancers (CRC) that show evidence of microsatellite instability (MSI-H) are marked by a high tumour infiltrating lymphocyte (TiL) population which is thought to be prognostic. Programmed cell death 1(PD-1) is a negative regulator of the immune system and targeting the interaction with its ligand PD-L1 offers a potential therapeutic target. We aimed to characterize CD8 and PD-1 expression in both the tumour centre (cT) and tumour periphery (pT) of microsatellite stable (MSS) and unstable CRC. Methods: Paraffin-embedded tumour blocks were cut at 5um, prepared and stained using specific antibodies for CD8 and PD-1. The pT was defined as the area within a 400x high power field (HPF) from the outline of the tumor. The cT was defined as the area at least one 400x HPF apart from the tumor outline toward centre of the tumor. Images were taken at 40x, 100x, 200x and 400x. Positive cells were averaged across 3 high power fields and classified as high or low positivity. Results: Forty-two specimens have been analysed to date including 28 MSI-H and 13 MSS tumours. Sixty-eight percent of MSI-H were stage II and 69% of MSS were stage III. In the MSI-H group, a high CD8 count in the cT and pT correlated with and earlier tumour size and stage. PD-1 positivity was seen in 61% of MSI-H cT compared to 0% positivity in the cT of MSS tumours. The periphery of both MSS and MSI-H specimens showed significant PD-1 expression with 71% and 85% of samples showing positivity respectively. There was no association between high or low densities of staining and stage. Conclusions: Zonal differences exist in the expression of CD8 and PD-1 in microsatellite stable and unstable tumours. A high proportion of MSI-H tumours show PD-1 activity in the centre of the tumour despite an improved prognosis. Further profiling of other T cell populations may help to further understand this expression which may act as a biomarker or provide a therapeutic target Biomarkers that are able to distinguish stage II and III colon cancer patients at high risk of developing disease recurrence, who may benefit from adjuvant chemotherapy, are still lacking. Genome-wide profiling of somatic aberrations, including gene point mutations, DNA copy number aberrations (CNA) and structural variants (SV), is expected to provide better insight into the molecular pathology of tumour progression and clinical outcome. Genome-wide analysis of CNAs was performed using high-resolution comparative genomic hybridization for microsatellite stable (MSS) stage II and III primary colon cancer samples (n=114). In addition, the prevalence of genes suffering from CNA-associated chromosomal breaks, indicative for SVs, was determined. The mutation status of commonly affected APC, TP53, KRAS, PIK3CA, FBXW7, SMAD4, BRAF and NRAS genes was examined for 60 samples using targeted massive parallel sequencing. Associations of genomic aberrations with disease-free survival (DFS) rates were explored by log-rank tests using 10,000 permutations. Disease recurrence and DFS rates differed significantly for several CNA-regions (P<0.05). A total of 267 genes were recurrently affected by CNA-associated chromosomal breaks (FDR<0.1), among which 168 genes (66%) that were also identified in a previously analysed cohort of 352 metastatic colorectal cancers. Gene point mutation frequencies were in concordance with literature. In a univariate analysis, none of the individual mutated genes appeared to be significantly associated with DFS. In summary, several associations are found between highly prevalent genomic CNAs and disease recurrence in this cohort of MSS stage II and III colon cancers. Further in-depth analysis is required to unravel underlying biology that contributes to disease recurrence.

KM Sutton 1 ; D Bottomley 1 ; D Morton 2 ; P Quirke 1 ; P NP West 1 Accurate and reliable methods for assessing the molecular profile of clinical tumour samples are important for the delivery of personalised medicine. When adopting a targeted amplicon sequencing method in combination with next generation sequencing (NGS), it is ideal to call mutations against a control sample to enable artefacts to be removed from the analysis. Blood is considered the gold standard control but may not always be available. We compared the use of histologically normal mucosa to blood as a control in colon cancer. We examined mutations in 40 colon cancers from the NCRI FOxTROT trial using the Fluidigm Access Array for NGS library preparation. We assessed the use of both blood and normal colonic mucosa as a control for assessing mutations in 11 genes. All samples were tested in duplicate. The work was partly funded by a PathSoc Career Development Fellowship and is presented on behalf of the FOxTROT Collaborative Group. Mutation calls made using normal mucosa as a control compared to blood were in good agreement; a Mathew's Correlation Coefficient above 0.7 was seen for all of the genes where agreement could be assessed. We found that false positive mutations were due to poorer amplification of the normal mucosa samples and false negatives were due to mutation calls in the normal mucosa. Overall we found that when assessing mutations in hotspot oncogenes, testing in duplicate and the use of a normal control tissue is not required to make mutation calls. However, where a normal control is required, normal mucosa from the resection margin is a suitable alternative to blood where it is not available.

We report a series of four unusual ovarian or extraovarian neoplasms composed of an admixture of adenosarcoma and a predominant component comprising a sex cord tumour. The neoplasms occurred in women aged 50 to 69. Three cases arose within the ovary and one was extraovarian (pelvis and abdomen) in location. In all four cases, there were minor areas with morphological features of adenosarcoma with a phyllodes-like architecture and periglandular increased cellularity with mitotic figures. In two cases, the stromal component was morphologically in keeping with a juvenile granulosa cell tumour. In one case, the stromal component had some features of both adult granulosa cell tumour and Sertoli cell tumour within a fibromatous background. The fourth case morphologically could not be categorised as any of the usual types of ovarian sex cord tumour and was categorised as an unclassifiable sex cord tumour. In all four cases, there was immunohistochemical evidence of sex cord differentiation. In each case, we propose that the sex cord tumour arose from a pre-existing adenosarcoma thus representing an unusual form of sarcomatous overgrowth of sex cord elements which can occur within adenosarcomas. This phenomenon is not well described in the literature. Background: Human epididymis protein 4 (HE4) is a secreted protein that is overexpressed in some cancers. HE4 is emerging as a useful biomarker in diagnosis and follow-up of endometrial cancers. The aim of this study was to evaluate the potential role of serum HE4 in the diagnosis and management of endometrial cancer. Methods: Patients undergoing surgery for endometrial disease were recruited into this study and had pre-operative serum samples taken, n=157. Demographic, clinical, radiological and laboratory data were reviewed. HE4 and CA125 serum levels were analysed using the Fujirebio Diagnostic ELISA Kits and results correlated with clinicopathological details. Standard cut-off points of 70 pmol/L for HE4 and 35 U/ml for CA125 were used. Results: HE4 showed a sensitivity of 64% and specificity of 97.50% for detection of endometrial cancer. CA125 had a very low sensitivity of 14% for endometrial cancer diagnosis. HE4 was elevated in all stages of endometrial cancer and demonstrated the ability to distinguish between benign and malignant groups. HE4 also provided information about myometrial space invasion. Conclusion: HE4 has a role in endometrial cancer diagnosis and prognosis and has the potential to be used in a screening setting or as a triage marker in the primary care setting. For women diagnosed with endometrial cancer, HE4 has the potential to stratify them into treatment regimens where the most appropriate treatment can be delivered resulting in improved quality of life and outcome for endometrial cancer patients.

Platelets Drive Metastatic Changes in Ovarian Cancer Cells P CD Spillane 1 ; NM Cooke 2 ; S O'Toole 3 ; D Kenny 2 ; O Sheils 1 ; JJ O'Leary 1 1 Histopathology Department, Trinity College Dublin, Dublin, Ireland; 2 Department of Molecular and Cellular Therapeutics, RCSI, Dublin, Ireland; 3 Department of Obstetrics and Gynaecology, Trinity College Dublin, Dublin, Ireland Background: Ovarian cancer is the 5th leading cause of cancer related deaths in women. Previously we described a dynamic interaction between ovarian cancer cells and platelets in vitro, involving platelet adhesion, activation and induction of pro-survival and pro-angiogenic signals in the cancer cells. This study looked to further investigate this phenomenon in ovarian cancer cells by assessing the molecular changes it induced. Methods: Cell lines 59M and SKOV3 were used as in vitro models of metastatic ovarian cancer. Platelet cloaking of cells was quantified by flow cytometry. Cells co-cultured with/ without platelets for 24hrs were examined by RT-PCR for EMT related changes and by Affymetrix Gene2.0ST arrays for whole transcriptome changes. Results: Significantly more platelets adhered to SKOV3 cells than 59M cells. While there were different rates of adhesion, the platelets induced similar changes in EMT related genes in both. There was a significant loss in expression of epithelial genes and an increase in mesenchymal genes, indicating the induction of EMT. Whole transcriptome analysis showed that there were a greater number of gene expression changes occurring in SKOV3 cells compared to 59M cells, correlating with the adhesion data. A 32 gene panel of commonly affected genes in both cell lines was identified, many of which form part of an interlinking pathway that is regulated by TGFβ1 and associated with cell adhesion/ECM remodelling. Though only 32 genes overlapped, the biological processes affected in both cell lines were very similar, with 103 of the 148 processes enriched in the 59M data set also seen in the SKOV3 data set. Conclusion: This study shows that platelets can enhance the metastatic potential of ovarian cancer cells through the induction of EMT and ECM changes. In addition, it has identified a set of 32 genes that hold potential to be in vivo markers of this interaction. Background: During the metastatic cascade, circulating tumour cells rapidly and efficiently adopt a platelet cloak. Platelet cloaking of tumour cells promotes metastatic disease by promoting cellular proliferation, angiogenesis and EMT while inhibiting autophagy and apoptosis. The aim of this study is to examine whether the platelet cloak contributes to tumour cell evasion of NK cell mediated immune surveillance. Methods: Freshly isolated PBMCs were harvested from healthy donors and stimulated for 18 hours with IL-2 (500U/mL). PBMCs were co-incubated with ovarian (59M and SKOV3), melanoma (Sk-Mel-28) and CML (K562) cell lines that were either uncloaked, or cloaked with washed platelets from healthy donors. The NK-tumour cell receptor ligand systems, NKG2D-MICA/MICB and CD96/CD226-CD155 were examined using NK cell CD107a expression and interferon-gamma production to quantify NK cell mediated recognition and 'killing' of cancer cells. Results: We first demonstrated that ovarian and melanoma cancer cell lines when cloaked with washed platelets strongly inhibited NK cell antitumor reactivity. Platelet cloaking induced down-regulation of the stress ligands MICA and MICB on the tumour cell coupled with their release into the microenvironment, a known NK cell immune decoy strategy. In addition, platelets significantly down-regulated both CD96 (NK cell) and CD155 (tumour cell), inhibiting NK cell activity. Both mechanisms occur in tandem to comprehensively incapacitate NK cells and promote tumour immune evasion. Conclusions: Ovarian and melanoma tumour cells are efficiently cloaked by platelets, which facilitates immune evasion by actively suppressing NK cell cytotoxicity and cytokine production. Purpose of the study: Glioblastomas (GBM) are the most common and most aggressive primary malignant brain tumours in adults. One of their histopathological hallmarks is the microvascular proliferation; these tumours are among the most angiogenic of malignancies by displaying the highest degree of microvascular proliferation. IGFIIR/Man-6-P is a receptor that belongs to the insulin-like growth factor (IGF) system. The involvement of IGF-IIR/Man-6-P in the process of angiogenesis has been postulated in rare earlier studies. To our knowledge, the role of IGF-IIR/Man-6-P in the neovascularisation of human GBM has never been studied. Methods: IGF-IIR/Man-6-P expression was evaluated in the vascular compartment from 322 human GBM and from 10 normal adult brain samples by means of quantitative immunohistochemistry on tissue microarray sections. In vitro cell line experiments were carried out in order to characterise the IGFIIR/Man-6-P role in angiogenesis. Summary of results: IGF-IIR/Man-6-P was strongly expressed in the cytoplasm of endothelial cells in hyperplastic vessels and exhibited a dot-staining pattern. We found a higher expression of IGF-IIR/Man-6-P in GBM vessels compared to normal brain vessels (p=0.05). Furthermore, preliminary in vitro experiments suggest a role of IGF-IIR/Man-6-P in tube formation but not in growth of the EA.hy926 endothelial cell line. Conclusions: This work shows a possible role of IGFIIR/Man-6-P in the process of neovascularisation of GBM angiogenesis. Additional investigations are required to confirm the role of this receptor as a direct actor of angiogenesis in GBM. Purpose of the study: Vasa vasorum (VV) are microvessels which supply vessels that cannot be nourished by diffusion from their own lumina. VV are believed to be a key element in the pathogenesis of vascular diseases. A number of different imaging methods have been used to study the VV but there is still no definitive consensus on their structure. The aim was to describe the normal microvessel anatomy of temporal arteries. Methods: Human temporal artery, obtained following routine biopsy with ethical approval and patient consent. Samples were embedded into paraffin blocks and serially sectioned at 5 micron intervals. Alternate sections were stained with H&E and scanned to create virtual slides. The slides were aligned, VV were segmented (annotated) and iso-surfaced to generate 3D reconstructions. Summary of results: The reconstruction shows the structural arrangement of the VV as a complex plexus. No connection to the vascular lumen was visualised. In this segment a hierarchical branching structure was not observed. VV were almost exclusively restricted to the adventitia of the vessel wall. Mean ± SD area of the VV (n = 5283) is 2287.23µm2 (±4956.03). The mean ± SD number of vessels per slide is 60.76 (±15.37). These metrics are based on one arterial specimen. Conclusion: This method allows us to study the three-dimensional spatial relationships of microvessels within arterial specimens. Furthermore, metric data generated in the process can support the 3D images to study the microvasculature. This method will be applied to diseased arteries in future to generate novel hypotheses about the inflammatory process. Acknowledgements: This research was supported by a PathSoc intercalated studentship. Purpose of the Study: It is controversial whether mesothelioma can be diagnosed with confidence in effusion cytology and therefore an ancillary marker of malignant mesothelial cells would be clinically valuable. BRCA-1 associated protein (BAP1) is a tumour suppressor gene which shows biallelic inactivation in approximately half of all mesotheliomas. BAP1 expression is commonly lost in mesothelioma. We investigated whether loss of BAP1 expression can be used to support a diagnosis of mesothelioma in effusion cytology. Methods: Immunohistochemistry (IHC) for BAP1 was performed on cell blocks from effusions associated with confirmed mesothelioma cases, effusions containing mesothelial cell atypia, benign effusions, and effusions from patients with lung adenocarcinoma. Results: IHC for BAP1 was performed on 75 cases of confirmed mesothelioma. 43 (57.3%) showed negative staining in the presence of an internal positive control. In 57 effusions considered to have atypical mesothelial cells in the absence of definitive diagnosis of mesothelioma, 8 cases demonstrated negative staining for BAP1. On follow up, 6 of these patients received a definitive diagnosis of mesothelioma in the subsequent 14 months (2 were lost to follow up immediately). Only 5 of 100 consecutive benign effusions were interpreted as BAP1 negative. 47 patients with confirmed adenocarcinoma demonstrated positive staining for BAP1. Conclusion: We conclude that loss of BAP1 expression in effusion cytology is quite specific for mesothelioma. Whilst it is not definitive, it can be used to support the diagnosis of mesothelioma in atypical effusions. We caution that interpretation of BAP1 IHC on cell block may be difficult and that convincing positive staining in non-neoplastic cells is required before atypical cells are considered negative. We also note that BAP1 loss is not a sensitive test and cannot be used to exclude mesothelioma.

The South-East of Scotland Experience on the Molecular Detection of EGFR, KRAS and ALK Mutations in Lung Adenocarcinomas P Y Kheng 1 ; L Williams 2 ; K Walsh 1 ; J Fairley 1 ; S Camus 1 ; L Gilroy 1 ; K Gilmour 1 ; D Stirling 1 ; W Wallace 1 ; D Harrison 1 ; A Oniscu 1 1 Royal Infirmary of Edinburgh, Edinburgh, UK; 2 The University of Edinburgh, Edinburgh, UK

The approval of novel targeted treatments for EGFR-positive and ALK-positive non-small cell lung cancer (NSCLC) has led to the increased requirement for mutation testing services in South East of Scotland. EGFR mutations are typically found in females, Asians and never smokers whereas KRAS mutations are associated with smoking. ALK rearrangements are commonly found in younger patients and never smokers. This study aimed to determine the prevalence of EGFR, KRAS and ALK mutations in South East of Scotland and to evaluate our experience in testing of ALK with IHC and FISH. Data of all patients tested were collected retrospectively from clinical records. From January 2011 to May 2014, we reported mutation rates of EGFR, KRAS and ALK to be 10.4% (67/643), 35.8% (86/240) and 2.3% (7/304) respectively. In our cohort, an increase in one pack years of smoking resulted in a decrease in the odds ratio of EGFR-positivity (OR 0.94, 95% CI 0.92 -0.96, p<0.001). KRAS-positivity was associated with a history of smoking, with rates in both former (OR 6.26, 95% CI 2.00-19.56, p=0.002) and current smokers (OR 6.82, 95% CI 2.18-21.35, p=0.001) significantly higher than in non-smokers. The number of smoking pack years had no influence on the rates of KRAS-positivity. ALK-rearrangements were found to be associated with never smokers (p<0.001) and younger patients (≤50 years old) (p<0.001). To date, no false positives were reported for parallel testing of ALK with IHC and FISH. We observed 100% sensitivity (7 IHC+/7 FISH+) and 96.6% specificity (113 IHC-/117 FISH-) when comparing IHC with FISH. In conclusion, the prevalence of EGFR mutation in South East of Scotland has reflected mutation rates reported in West of Scotland. Our findings further support the use of ALK-IHC as a diagnostic screening tool. Purpose of the Study: The molecular mechanisms of metastasis and progression of penile squamous cell carcinoma (PSCC) are unclear. Nobody, to our knowledge, has investigated the expression of cell-cycle proteins in advanced or metastatic PSCC. We aimed to determine the extent of HPV infection in patients with advanced PSCC and its effect on the expression of the key cell-cycle proteins p53, p16INK4A and retinoblastoma (RB). Methods: Archival paraffin embedded blocks were obtained from 27 primary penile cancers, all patients having developed locally-advanced or metastatic disease. All patients were treated in the Phase II Trial of docetaxel, cisplatin & 5-fluorouracil (TPF) chemotherapy CRUK/09/001 (Nicholson et al. BJC 2013; 109: 2554-9) . Samples were analysed immunohistochemically for p16INK4A, p53 and RB protein expression on a tissue microarray. All tumours were HPV typed using PCR. Summary of Results: HPV DNA was detected in 8/22 (36%) with HPV 16 present in 7/8 (88%). 5 cases were not suitable for analysis. No association was found between HPV and expression of either p16INK4A (p= 0.3426), p53 (p= 0.1365) or RB (p= 1) using Fisher's exact test. Conclusions: HPV DNA is detected in less than half of progressive PSCC, suggesting either the loss of HPV in advanced disease or that non-HPV related cancers progress more commonly. The lack of correlation between HPV and these cell-cycle proteins suggests that they may undergo somatic mutation that is not driven by HPV, leading to increased growth and invasiveness. Treatment strategies may be hampered by this genetic diversity, which requires further investigation. Prostate cancer is the second most common form of cancer in males, and the incidence of this disease is predicted to double globally by 2030. More than 1.1 million new cases of prostate cancer are diagnosed each year and two thirds of these patients are from the Western world. Current diagnostic tests for prostate cancer are limited in both sensitivity and accuracy, and a method for accurate prognosis in these patients is yet to be developed; therefore, there is a need for a sensitive and specific prostate cancer test to implement early and appropriate therapy. The recent discovery of altered endosomal-lysosomal biogenesis in prostate cancer cells has identified a fundamental change in the cell biology of this cancer that holds great promise for the identification of novel biomarkers that can predict disease outcomes. Investigation of the endosome compartment and endosome biogenesis revealed elevated gene and expression of critical machinery components that are required for endosome biogenesis and endocytosis. Here we demonstrate significantly altered expression of endosomal and lysosomal genes in mRNA microarrays of prostate cancer tissue compared to non-malignant tissue, and that specific endosomal and lysosomal genes are predictive of patient outcomes. Two endosomal tri-gene signatures were identified that had a significant capacity to stratify patient outcomes. Changes in the expression of these genes was further ascertained by qPCR in fresh-frozen prostate tissue specimens, which further implicated altered endosome biology during disease progression, with significant changes in expression observed between aggressive prostate cancer and indolent disease or normal prostate tissue. These findings support the initiation of a retrospective trial to determine if these new biomarkers can accurately predict clinical progression in prostate cancer patients.

M Craze; C Joseph; C Nolan; A Green; EA Rakha; IO Ellis; P A Mukherjee University of Nottingham, Nottingham, UK Introduction: Lymphovascular invasion (LVI) is an important step in the metastatic cascade. Identification of a molecular signature for the LVI positive phenotype will help identify relevant drivers and pathways. This study aimed to investigate determinants of LVI from a biomarker database. Methods: Biomarkers (n >200) from a well annotated series (n=1929) were analysed for correlations with LVI [clinical/IHC (D2-40) supplemented]. Proteins with significant associations with LVI were interrogated for pathway enrichment analysis [corrected for false discovery rate (FDR)], using the STRING 9.1 platform incorporating Gene Ontology (GO), KEGG and NCI. Results: Biomarker analysis related to both clinical/IHC determined LVI identified 35 positively associated markers, 14 in both clinical and IHC categories (e.g. ADA3, CD8, FOXP3, KPNA2). A further 21 markers were negatively associated, 8 in both categories (e.g. Bcl2, BRCA1, MAGE3 and SOX10). Significant pathways (p<0.001) unifying the positively associated proteins include metabolism, immune responses (T-cell regulation and differentiation), cell activation and transcription [GO] ; T-cell receptor signalling pathways and pathways in cancer and haematopoietic cell lineages [KEGG] . For negatively associated proteins, the following were significant: ubiquitination processes, regulation of the mitosis [GO]; p53 pathways [KEGG] and apoptotic and cell cycle pathways [GO & KEGG] . On cross-validating a subset included in the METABRIC cohort, there were overlapping enrichments for immune response regulation (GO) and haematopoietic cell lineages (KEGG). Conclusions: These preliminary findings are the first to unify biomarkers for LVI pathway analysis in BC, using protein based data. Within the constraints of selection bias, data mining from immunohistochemistry of multiple biomarkers in relation to biological processes hold promise. *AM supported by the NIHR and the Academy of Medical Sciences ABSTRACTS S·17 Assessment of HER2 Status on Needle Core Biopsy of Breast Cancer: Impact of Histopathological Concordance P M Pigera; AHS Lee; IO Ellis; EA Rakha; Z Hodi Nottingham City Hospital, Nottingham, UK One of the key recommendations introduced in the ASCO/CAP update guideline recommendation on HER2 testing is the novel concept of "histopathological concordance." It is proposed that certain tumour morphological features such as histologic type and grade should trigger repeating a molecular test in cases of "discordance". In this study we have we have reviewed 3104 breast cancer cases consecutively reported in routine practice in Nottingham in the last 4 years. Data on HER2 status was collected and cases with HER2 assessed on resection specimens (RS) were analysed in details. Results: of all cases, 98 patients (3%) had HER2 status assessed on core biopsy and the corresponding tumour RS. The main reasons for a repeat were tumour multifocality and morphologically different or heterogeneous tumours. A few cases were repeated because of borderline negative FISH results or neoadjuvant therapy. 18 cases were repeated due to insufficient tumour in the core biopsy. In this study the HER2 status of the index tumour was changed in 2 cases and both were in the borderline result category. HER2 testing of different tumour foci of multifocal or morphological heterogeneous tumours was consistent with that of the index tumour assessed on the core biopsy apart from two cases; one positive and one negative. 17 tumours were upgraded from grade 2 on core to grade 3 on excision and HER2 status did not change. No contribution of hormone receptor or tumour type was identified. Conclusion: There is excellent agreement between HER2 assessed in core biopsy and RS. Histopathological discordance seems to play a minor role which does not justify test repeat in routine practice. Tamoxifen prevents breast cancer in a sub-set of high-risk women in a mechanism that appears to be dependent on reduction of MD. Animal model studies suggest that tamoxifen remodels the mammary stroma to a tumour-inhibitory phenotype. This study aims to analyse the effect of tamoxifen on breast fibroblast function and identify potential protumourigenic pathways contributing to density-associated risk. Methods: Primary human breast fibroblasts were treated with hydroxytamoxifen (100nm-5µM). Fibroblast function was analysed by measuring: proliferation; expression of stromal proteins fibronectin (FN), LOX and collagen 1; effects on TGF-β signalling via SMAD phosphorylation and upregulation of the myofibroblast marker SMA. Genome wide analysis was performed using RNA-Seq. Summary of Results: Fibroblasts from 25 patients were treated with tamoxifen. All patients showed reduced proliferation with treatment. In 62% of patients tamoxifen treatment resulted in reduced expression of FN. TGF-β-mediated upregulation of SMA and FN were consistently inhibited by tamoxifen, as was fibroblast contraction of collagen gels. RNA-Seq analysis revealed modulation of a number of metabolic pathways by tamoxifen, including significant upregulation of DHCR7, part of the microsomal antioestrogen binding site (AEBS). Conclusions:These data indicate that tamoxifen can directly remodel the stromal microenvironment, generating a less 'reactive' stroma. Modulation of AEBS activity has been proposed to be anti-tumourigenic, and also is implicated as a suppressor of Hedgehog signalling. Thus, tamoxifen impacts on multiple pathways to create a tumour inhibitory phenotype. This work was supported by the Pathsoc Small Grant Scheme. Purpose of the Study: The Phenotypic features of basal like (BL) breast cancer (BC) resemble those occurring in BRCA1-germline mutation carriers. Several lines of evidence suggesting the overall tendency of basal-like/triple negative BC to spread through vascular rather than lymphatic routes. The latter has recently been attributed to the activation of cadherin switch, an EMT-like phenomenon, in BLBC. This study aims at studying the cadherin switch expression profile TGFB1, a key EMT-trigger, expression in BRCA1 mutated compared to sporadic BC.

The expression of E-cadherin, N-cadherin and TGFB1 were studied in a subset of germline BRCA1 mutated BC (n= 47) compared to non-selected cohorts of non-lobular sporadic invasive BLBC (n= 422) and non-basal BC (n=1190) using IHC and TMA. Summary of results: Compared to sporadic BC, BRCA1 mutated cases were of younger age, more grade 3, with more medullary-like tumours, and more LVI positive. E-cad was significantly less expressed in BRCA1 cases than in the sporadic non-basal and in the BLBC. However, N-cad was not significantly expressed in BRCA1, non-basal, and BLBC. TGFB1 was significantly less expressed in sporadic BC, both non-basal BLBC than BRCA1 mutated BC. E-cad/N-cad combinatorial expression phenotypes were significantly different between BRCA1 mutated and non-basal and BLBC. Higher proportions E-cad-/N-cad+ were significantly observed BLBC than non-basal BC. BRCA1 mutated cases displayed the least E-cad+ expression and the highest E-cad-/N-cad+ in the studied series. Conclusions: Despite the known similarities between BRCA mutated and BLBC, results of this study demonstrate the more occurrence of cadherin switch in BRCA1 mutated breast cancer. E-cad repression appears to contribute more than N-cad gain in BLBC than non-basal BC.

Exploring Molecular Mechanisms Underlying Lymphovascular Invasion in Breast Cancer P SN Sonbul 1 ; A Mukherjee 1 ; R Russell 2 ; OM Rueda 2 ; M Aleskandarany 1 ; AR Green 1 ; E Provenzano 3 ; C Caldas 3 ; IO Ellis 1 ; EA Rakha 1 The development of tamoxifen resistance (TR) in oestrogen-dependent breast cancer (BC) is a therapeutic challenge. Insulin-like growth factor binding proteins (IGFBPs) may play a role in this process. We have investigated the role of IGFBP proteins in TR BC. IGF axis genes were evaluated in MCF-7 (wt) cells and tamoxifen-resistant (TamR) variants using qRT-PCR and confirmed by ELISA, Western, and Ligand blotting. IGFBP-2 & -5 were knocked down by shRNA transfection, and subsequent sensitivity to 4-hydroxytamoxifen (4-HT) was determined via WST-1. Cell migration was investigated by using the Incucyte system. IGFBP-2 expression was evaluated in 424 BC cases by TMA immunohistochemistry. Five out of 10 genes of the IGF axis (IGF-IR, IGF-2R, IGFBP-2, -4 and -5) had the highest expression levels by both parental wt and TamR cells. IGFBP-5 was down-regulated by ~7-fold while IGFBP-2 was up-regulated by ~2-fold in TamR versus wt cells (mRNA and protein levels). Significantly, a knockdown of IGFBP-2 in TamR cells restored sensitivity to (4-HT), reduced ERα expression to 45 ± 11.9% and enhanced cell migration. Expression of IGFBP-2 was significantly (P< 0.001) associated with survival advantage in TR patients. IGFBP-2 and IGFPB-5 are reciprocally regulated in the acquisition of TR by MCF-7 cells. IGFBP-2 may play a role in the development of TR in vitro and its high levels in clinical samples may predict TR. Purpose of the Study: Several lines of evidence are currently suggesting that the morphologic heterogeneity of breast cancer is mirrored at the genetic level. Understanding the molecular genetic evolution of BC would contribute further insights into the molecular derangements driving disease progression. Moreover, varied clinical outcome and response to similar therapeutic regimen is attributed, at least in-part to intratumoural heterogeneity. NGS can reliably study the genetic events using miniscule amounts of genomic DNA. Methods: gDNA was extracted from FFPE tissue sections from a case of invasive duct carcinoma (3 primary tumour samples and 3 samples from positive axillary lymph node metastases). Sample preparation and exome enrichment was performed using Nextera Rapid Capture exome kits (illumina, FC-140-1000 ). Exome sequencing was performed using illumina MiSeq with 15x depth of coverage (following adapter/barcode trimming). Exploratory analyses and data mining were executed regarding variant (s) concordance/ discordance between primary tumour samples and their respective metastatic variants. Summary of Results: Initial findings revealed 37 candidate indels common to all three axillary lymph node samples yet absent from the three primary tumour samples. Several genes have been identified as having frameshift mutations caused by indels. Molecular players previously linked to anti-angiogenesis are amongst the genes affected by indel mutations in their coding sequences that may lead to potential abrogation of protein function. Conclusions: These initial findings provide the framework for detailed molecular analyses for assessing molecular evolutionary events in primary breast cancer and their corresponding metastases. C-MYC is amplified in approximately 15% of breast cancers (BC) and is associated with poor outcome. c-Myc protein is multi-faceted and participates in many aspects of cellular function and is linked with therapeutic response in BC. We hypothesised that the functional role of c-Myc differs between molecular subtypes of BC. We therefore investigated the correlation between c-Myc protein expression and other proteins involved in cell cycle control, proliferation, apoptosis and DNA damage together with clinicopathological parameters, outcome and treatments in early invasive primary BC (n=1,106) using immunuohistochemistry. The METABRIC BC cohort (n=1,980) was evaluated for c-Myc mRNA expression. In whole series, there was significant association between c-Myc protein expression with higher tumour grade, lymph node(LN) positivity and medullary-like tumours. C-myc showed differential association with other proteins in the molecular classes. In luminal A tumours, c-Myc was associated with ATM (p=0.005), Cyclin B1 (p=0.002), PIK3CA (p=0.009) and Ki67 (p<0.001). In contrast, in basal-like tumours, c-Myc showed positive associated with Cyclin E (p=0.003) and p16 (p=0.042) expression. c-Myc was an independent predictor of a shorter distant metastases free survival in luminal A LN+ tumours treated with endocrine therapy (ET; p=0.013). c-Myc expression did not predict patient outcome in the other molecular subtypes with respect to adjuvant treatment. High c-Myc mRNA expression was associated with higher grade and basal phenotype (p<0.001). In luminal tumours treated with ET, c-Myc mRNA expression was associated with BC specific survival (p=0.001). c-Myc function is associated with specific molecular subtypes of BC and confers resistance to ET. The diverse mechanisms of c-Myc function, particularly in luminal A BC, warrants further investigation.

Metasin Axillary Predictive Score (MAPS): A Measure of Axillary Nodal Disease Prediction to Provide an Informed Choice for Breast Cancer Patients and Surgeons P PP Gopinath 1 ; D George 1 ; P Sai-Giridhar 2 ; S Jader 1 ; E Arkoumani 1 ; S Holt 2 ; G Francis 3 ; C Yiangou 3 ; S Al Ramadhani 4 ; S El Sheikh 5 ; N Agrawal 3 ; V Sundaresan 1 Purpose of study: Intra-operative sentinel lymph node sampling and molecular analysis empowers the surgeon to carry out axillary clearance as a one-step process. We have recently completed the clinical validation of Metasin, an intraoperative molecular assay for sentinel lymph node analysis in breast cancer patients (1836 cases). Method: The assay uses 2 positive predictive markers and is quantitative, enabling the prediction of tumour volume using 2 markers CK19 and Mammaglobin. In this study group, 439 patients had positive sentinel nodes and 444 cases underwent axillary clearance. Of the axillary clearance cases, 26% contained positive lymph nodes. 84% were sentinel node (SNB) macrometastases, 5% were SNB micrometastases and 11% were SNB negative or contained isolated tumour cells. Informative data was available for sentinel nodes from 125 positive cases. Results: Using the qPCR values (from Metasin assays using standardised pre-mixes) and clinical axillary clearance data, the cases have been stratified on the basis of the involvement of other axillary nodes. We have shown a three-tiered predictive grouping exists: Group A includes low tumour volume disease with a nodal positivity of 25% within the axilla (n=16): Group B with a 44% positivity of other nodal involvement (n=80) and Group C with positivity of 73% of axillary clearances (n=29).

The clustering of the Metasin data is dependent on the qPCR results and shows that the cases can be sub-grouped to provide a probability basis for prediction of axillary nodal involvement; dependent on the qPCR cut offs. This gives the patient and surgeon a statistical basis for determining the likelihood of other axillary nodal disease. Sequencing of the BRCA1 and BRCA2 genes has long been used in genetics laboratories to identify cases of familial breast and ovarian cancer. However, the advent of chemotherapy for ovarian cancer based on PARP inhibitors, which requires the presence of a BRCA1 or BRCA2 mutation, is turning this specialist test into a commonly-applied companion diagnostic. At the same time, the introduction of new DNA sequencing technologies is posing challenges even for experienced genetics laboratories. EMQN has been providing EQA of BRCA1 and BRCA2 gene sequencing world-wide for 15 years. The rate of serious diagnostic errors has varied from year to year, but the mean has hovered stubbornly around 3%. In EQA, just 3 samples per year are sent out, and the quality and experience of participating laboratories varies greatly. We recently carried out a collaborative study to measure the quality of BRCA gene sequencing by traditional and new methods in 20 experienced, expert laboratories from 11 countries. Ten DNA samples (8 with pathogenic mutations, 2 with normal DNA sequence) were sent to each laboratory. Ten labs used next-generation sequencing (NGS) alone, 3 used Sanger sequencing alone, and the others used combinations of Sanger sequencing, NGS, MLPA and other technologies. Seventeen (85%) of labs identified all clinically-significant variants on all 10 samples. Four false negative results were reported by 3 labs. Two were due to deficiencies in the bioinformatics pipeline of the NGS process, while 2 were attributed to a sample swap, and incorrect interpretation of a melting profile. No significant trend was identified with respect to the genotyping accuracy of the different methodologies used. The observed error rate of 2% amongst expert laboratories indicates the complex and challenging nature of this kind of testing. Caution will be required when applying these technologies to sub-optimal FFPE samples in Pathology laboratories.

P N Wolstenholme 1 ; SJ Patton 1 ; Z Deans 2 ; S Abbs 3 ; J Coxhead 4 ; K Brugger 3 ; P Westwood 5 ; K Thomson 6 ; H Scheffer 7

Next Generation Sequencing (NGS) is increasingly being introduced into clinical diagnostic laboratories worldwide. The huge amount of data generated by NGS cannot be duplicated by alternative methods for laboratories to internally validate all results, therefore external assessment of data is required. The UK National External Quality Assessment Scheme (UKNEQAS) for Molecular Genetics and the European Molecular Genetics Quality Network (EMQN) have developed a joint EQA scheme for NGS, with the aims to: (a) assess and improve quality; (b) enable laboratories to benchmark their NGS service against others and against best practice; (c) work towards consistency of reporting clinical results generated by NGS; and (d) contribute towards best practice. EMQN and UKNEQAS offer numerous disease-specific, molecular pathology and technical EQA schemes. The objectives for developing NGS EQA were to make it generic (independent of genes, diseases, platforms, and testing context (e.g,. Somatic, germline etc)) and applicable all users. Two pilot EQAs have been run and 157 labs from 32 countries participated. These labs were sent a genomic DNA sample and asked to sequence either their smallest gene panel or largest single gene which the lab tested, submit technical details, and genotypes at known SNPs. The results were compared against a "consensus EQA genome" established by multiple validations of the DNA. 12187 different genes were tested. Most labs are using small panel of 1-10 genes. 60% of all variants were detected by every lab which tested for them. A detailed summary of the key findings will be presented. Both pilots have proved to be challenging to meet our objectives, however the results have enabled clinical diagnostic labs to start to address the quality of their NGS testing. Tumours invade the vasculature, which transports circulating tumour cells (CTCs) to distant sites enabling growth of secondary tumours. CTCs hold the potential to monitor: therapeutic response, emergent mutations and act as a screening tool for the early detection of cancer. There are numerous methods to isolate CTCs. Once isolated, EpCAM and/or panCK positivity and CD45 negativity are used to verify CTC status. However, due to the metastasis associated process of Epithelial-Mesenchymal Transition, epithelial markers may be ineffective at identifying all CTCs. To overcome such protein marker based limitations, we have developed a novel staining pipeline (CTC-5) that combines histochemical staining (giemsa) with immunofluorescene (DAPI, EpCAM/panCK, HER3 and CD45) staining and whole slide imaging for robust identification, enumeration and characterisation of CTCs from cancer patients. CTCs are isolated from whole blood using ScreenCell Cyto devices. Cyto devices are then slide mounted, giemsa stained and digitised. Giemsa Staining is washed out and slides are immunofluorescently stained for EpCAM/panCK, CD45, HER3 and counter stained with DAPI. Fluorescently stained slides are digitised. Giemsa stained and four colour immunofluorescent digital slides are processed in silico generating a single z-stacked digital slide for pathological assessment. The CTC-5 staining pipeline has been experimentally validated via CTC characterisation of peripheral blood from Lung, Breast and Ovarian cancer patients, with respect to healthy donor and spiked-in controls. The CTC-5 pipeline overcomes recognised weaknesses in CTC characterisation. Histochemical staining is added to the current gold standard of EpCAM/panCK and CD45 staining, while also preserving a fluorescent channel for assessment of biomarker status (e.g. HER3, apoptosis or platelet cloaking). Such advancements enable robust pathological assessment of CTCs in the clinic.

Thorough interrogation of diseased tissue requires the use of multiple biomarkers in order to investigate biological pathways. Unless fluorescent technology is used, multiple sections are required from each tissue block as each section can only be tested for a limited number of markers. Histogenic Molecular Mapping (HMM) is a technique which used digitized images to evaluate multiple biomarkers. Although each section cut from a block is slightly different from the immediately preceding section, the similarity is sufficient to allow non-linear registration of images of successive sections. If the order is known, multiple sections can be mapped onto each other by registering each with the immediately preceding section. This allows several biomarkers to be mapped into a single "composite" section thereby giving a representation of the pathways activated/expressed in the tissue. We used HMM to investigate the mismatch repair pathway in colorectal cancer. Sequential tissue sections were stained for MLH1, PMS2, MSH2 and MSH6 and then scanned. Bespoke computational algorithms were used for image registration and composite images were binned as either "mismatch repair proficient" or "mismatch repair deficient". Validation of each category could be obtained by quantification of pixels in binarized images or pixel distribution using stereology. Our data show that HMM can be used for interrogating biological pathways in tissue sections and, ultimately, automated diagnosis of disease states.

Personalising Whilst pre-operative radiotherapy is the standard of care in locally advanced rectal cancer (LARC), only half of patients respond. Individualised treatment based on a predictive test could avoid unnecessary radiation exposure in poor responders. Macrophages in the tumour microenvironment with tumoricidal M1 and tumour protective M2 phenotypes could be modulating this response. This study investigated the possible predictive value of M1 and M2 subpopulations in identifying the response to short-course radiotherapy (SCRT). Pre-treatment biopsies and post-treatment resection samples were taken from 29 patients with LARC given SCRT. Dual-staining immunohistochemistry was performed with CD68, HLA-DR (M1 marker), and CD163 (M2 marker). Samples were scored for hot-and-random spots by Nuance software (version 3.0.2) and compared with tumour response measured by reduction in tumour-cell density. The work was partly funded by a PathSoc Career Development Fellowship. Samples showing a low score for HLA-DR positive M1 macrophages exhibited a better response to SCRT with a median 80% reduction in tumour cell density (IQR 47 to 85). Those with a high score exhibited a poor response with only a 20% reduction (IQR 0 to 49, p=0·017). No such trends were observed for CD163+ M2 macrophages. The ratio of HLA-DR+ to CD163+ macrophages for biopsy and resection samples was significantly different showing a drop in the HLA-DR positive macrophages in the resection samples (biopsy median 2·53, IQR 1.98 to 3.08; resection median 1·38, IQR 0.96 to 1.80; p=0·024). Assessment of macrophage subpopulations in pre-treatment biopsies appears to predict the degree of response to SCRT in LARC. Further investigation to validate these findings is now required prior to developing a predictive test for use in routine clinical practice. Patients with a poor predicted response could avoid toxic and costly radiotherapy and undergo alternative strategies including chemotherapy. Next-generation sequencing technologies (e.g. 16S profiling) are increasingly used to investigate complex bacterial communities. They have advantages over classical methods, as a significant proportion of bacteria are 'non-culturable'. However, they do not distinguish 'viable' and 'non-viable' populations, which may skew results, particularly following antibiotic exposure. Here we report culture and 16S data from a clinically reflective human gut model, describing changes in the gut microbiota following exposure to multiple antibiotics. A triple-stage chemostat model was inoculated with pooled human faeces from healthy volunteers to establish gut microbiota populations. The model was sequentially exposed to clindamycin (33.9 mg/L, QDS, 7days), vancomycin (125mg/L, QDS, 7days) and fidaxomicin (200 mg/L, BD, 7 days). Specific bacterial populations were enumerated daily on selective agars. Periodically, 16S profiling of gut model samples was performed; DNA was extracted on a QIAXtractor, 16S V4 PCR products were sequenced on an Illumina MiSeq, and resulting data were analysed using QIIME. Both culture and 16S profiling demonstrated marked alterations in gut microbiota populations following antibiotic exposure. For many populations, notably bifidobacteria and enterobacteria, changes seen by culture correlated with 16S profiling. However, as culture describes numerical changes in populations, and 16S profiling describes proportional changes, results are not always directly comparable. 16S profiling greatly increased microbiome coverage, particularly for clostridia. Population diversity (number of observed species and Shannon index) decreased with sequential antibiotic exposure. Use of culture and molecular methods in tandem can greatly increase understanding of changes occurring in complex microbial populations. Barrett's oesophagus is the erosive replacement of the normal squamous oesophageal lining with a glandular epithelium and is the major precursor of oesophageal adenocarcinoma. Barrett's patients are enrolled into active surveillance programmes in order to detect and treat oesophageal cancer at an early stage. Surveillance however is costly and burdening to patients. To improve screening efficacy there is an acute need for accurate biomarkers of cancer progression risk in Barrett's patients. Understanding the pattern and pace of clonal evolution that occurs within the Barrett's segment is a key step towards achieving this goal. Opinion is divided over whether goblet cells (intestinal metaplasia) on oesophageal biopsy are required for a diagnosis of Barrett's oesophagus. This is based on the unproven assumption that goblet cell differentiation marks increased cancer risk in Barrett's oesophagus patients. We have investigated the clonal structure of non-dysplastic and neoplastic Barrett's oesophagus by combining state-of-the-art 3D modeling and genetic lineage tracing. By tracing the clonal origin of an early oesophageal adenocarcinoma through whole-exome sequencing and mitochondrial DNA sequencing, we find that this cancer developed from non-goblet columnar epithelium, whereas the adjacent goblet-bearing mucosa was free of oncogenic mutations. Our results have important implications for the harmonization of the clinical diagnosis of Barrett's oesophagus.

Long-course chemoradiotherapy (CRT) is used to down-stage locally-advanced rectal cancer (LARC) prior to resection. An interval period prior to surgery allows for tumour shrinkage to facilitate surgical removal. The optimal time interval remains unclear, with little high-quality evidence to guide clinical decisions about when to operate. This study explores the pathological outcomes from a pilot randomised controlled trial comparing an interval of 6 weeks versus 12 weeks between CRT and surgery. Thirty one patients were recruited from seven UK centres between June 2012 and May 2014. Photographs were taken of the specimens and assessed by a blinded histopathologist for the quality of the mesorectal dissection. Rates of pathological complete response (pCR), down-staging, and circumferential resection margin (CRM) involvement were determined. Response was also assessed using novel tumour cell density (TCD) assessment where the slides from the resected specimen and baseline biopsy were scanned at 400x magnification, the tumour area selected and 285 to 315 data-points analysed by a blinded expert to describe the percentage of different tissue components. The work was partly funded by a PathSoc Career Development Fellowship and is presented on behalf of the STARRCAT Trial Investigators. Twenty three patients underwent surgery (10 from the 6-week arm and 13 from the 12-week arm). The mesorectal fascial plane was intact in 7 specimens from the 6-week arm (70%) and 8 from the 12-week arm (62%). Three patients at 6-weeks and two patients at 12-weeks showed a pCR. Only one patient (from the 12-week arm) had an involved CRM. TCD was 0.3% for the 6-week arm and 4.3% for the 12 week arm (p=0.12). In this small randomised trial, rates of mesorectal quality, CRM status, pCR and TCD were similar following either a 6 or 12 week interval after CRT. Further studies are now needed to clarify whether a longer interval does facilitate on going down-staging.

The Role of Tissue Factor Pathway Inhibitor (TFPI) in Liver Injury P G Petts 1 ; H Kudo 1 ; A Dorling 2 ; M Thursz 1 ; R Goldin 1 1 Imperial College London, London, UK; 2 Kings College London, London, UK Introduction: Studies have demonstrated that inhibition of the coagulant cascade is associated with less advanced liver fibrosis and better outcome in acute liver injury. TFPI is a serine protease inhibitor that acts as a homeostatic inhibitor of the coagulation cascade and may be a target to modify outcome in liver disease. Methods: Transgenic mice carrying a genetic modification that allows cells expressing a-smooth muscle actin (aSMA; e.g. activated hepatic stellate cells) to simultaneously express TFPI were used in models of chronic liver injury (carbon tetrachloride, CCl4) or acute liver injury (paracetamol) and culled at set time points after dosing. Results:Chronic liver injury: At 24 hours after the last dose of CCl4 the transgenic mice had significantly decreased aSMA expression and tissue inhibitor of metalloproteinase (TIMP) -1 gene expression but no difference in matrix metalloproteinase (MMP) -2 and -9 gene expression compared to wild types. This suggested a microenvironment that would promote fibrosis resolution. However after 24 hours this difference was lost. At all time points there was no significant difference between fibrosis in transgenic and wild type mice as demonstrated by Sirius red staining, hydroxyproline assay and collagen 1a1 gene expression. Acute liver injury: In paracetamol induced liver injury there was a significant difference in parenchymal necrosis in transgenic mice compared to wild types at 24 and 48 hours after dosing (24 hours: mean necrosis 6% vs. 30% respectively, Mann-Whitney test p=0.008. 48 hours: mean necrosis 2% vs. 20% respectively, Mann-Whitney test p=0.036). Conclusion: These results suggest that TFPI is an unlikely therapeutic target in chronic liver injury. However in acute paracetamol induced liver injury TFPI appears to rescue the injured liver in a sustained manner from 24 hours after the initial insult and suggests a role for TFPI in managing acute liver injury. (Research funded by the Pathological Society). 

Analysis of adenocarcinoma and non-small cell lung cancer (NOS) for EGFR mutations now forms part of the Royal College of Pathologists' lung cancer dataset. Identification of patients harbouring these mutations facilitates delivery of targeted therapies with superior efficacy. Testing of these tumours for ALK has also been introduced in our centre. We assessed our compliance with the College guidelines in this area for 2013 and 2014. In those tumours positive for EGFR or ALK mutations, we examined the original sections to assess any correlation between mutation status and morphological subtype. 96 of the 116 appropriate cases (83%) diagnosed histologically in 2013 were sent for EGFR mutation analysis, increasing to 183/190 (96%) in 2014. 20 of the cases over this time (7%) were positive for an EGFR mutation. Of these, 12 showed an acinar growth pattern, 3 were solid, 1 lepidic, 1 papillary and 1 micropapillary. It was not possible to characterise the growth pattern in two of the cases analysed as cell blocks. The most common mutation, a missense mutation at codon 858 of exon 21, was most frequently associated with an acinar growth pattern. Of the 2013 cases, 22 (19%) were sent for ALK mutation analysis, compared with 152 (80%) in 2014. Both of the two cases with an ALK translocation (2p23 rearrangement) showed an acinar growth pattern. Our compliance with College guidelines in sending appropriate lung specimens for mutation analysis is improving. The correlations between mutation status and morphological subtype add to, and are in keeping with, the current body of evidence in this area.

Primary Synovial Sarcoma of the Heart -An Interesting Case Report and Review of Literature P S Venkatesan 1 ; P Sloan 1 ; S Kendall 2 ; M Giles 2 1 Royal Victoria Infirmary, Newcastle, UK; 2 James Cook University Hospital, Middlesbrough, UK Seventy five percent of primary cardiac tumours are reported to be benign atrial myxomas. The remaining are malignant tumours with most of them being sarcoma, particularly angiosarcoma and malignant fibrous histiocytoma. Synovial sarcoma of the heart is a very rare malignancy accounting for less than 1% of all primary cardiac tumours. Most of them arise from the pericardium and the right side of the heart and is considered to be highly aggressive with reduced survival rates. Diagnosis in these rare locations is also challenging. We report a 42-year-old gentleman who presented to us with productive cough, chest pain and paroxysmal nocturnal dyspnea. Echocardiography revealed a calcified left atrial mass arising from the posterior leaflet of the mitral valve and radiologically was thought to be a benign atrial myxoma. Excision was planned with histology showing a malignant biphasic spindle cell tumour exhibiting marked cellular atypia and numerous mitoses. On immunohistochemistry, the glandular component expressed diffuse positive staining for Bcl-2 and EMA with focal positive staining for pancytokeratins. The spindle cell component expressed CD99 and EMA and was found to be negative for CD34, S100, desmin, Melan-A and HMB-45. Cytogenetic testing revealed SS18-SSX1/2 gene fusion with SS18 rearrangement confirming the diagnosis of synovial sarcoma in this rare location. A postoperative computed tomography was performed which showed no evidence of metastasis or primary lesions elsewhere. There was excellent postoperative surgical recovery and adjuvant chemotherapy was considered in the multi disciplinary meeting. Primary cardiac synovial sarcoma is an extremely rare malignancy especially when arising from the left atrium posing diagnostic difficulty mimicking atrial myxoma. In contrast to the poor prognosis mentioned in the literatures, there was excellent recovery in this gentleman. Swyer-James-MacLeod syndrome (SJMLS) is a rare lung condition that manifests radiologically as unilateral hemithorax lucency as a result of post-infectious obliterative bronchiolitis, leading to small airways obstruction and secondary emphysema. The histological features of SJMLS are poorly and infrequently described. We present three cases of the syndrome that underwent lobectomies in our institution from 2013 to 2015, in three women, aged, 33, 46 and 22 years, presenting with recurrent lower respiratory tract infection, shortness of breath and pleuritic chest pain. Two underwent left upper lobectomies, one left lower lobectomy. The first case demonstrated hyperlucency of the affected lobe with markedly reduced blood vessel attenuation. The radiological findings of the second case were of extensive bronchiectasis, hyperlucency, mucus plugging and hypervascularity. The radiological findings of the third case were of an apical bulla and upper lobe cavitating lesion with lobar hypolucency and hypoperfusion. The main histological findings were bronchiolar changes with bronchiolectasis, mucus plugging, constrictive / obliterative bronchiolitis and various degree of peribronchiolar inflammation. Emphysema was mild and diagnosed as loss of attachment of alveolar walls. In addition, Case 1 had dystrophic, hypoplastic or absent branches of the pulmonary arteries. Case 2 showed prominent bronchial arterioles and abnormal tortuous dilated pulmonary arteries and veins. Case 3 had established bronchiolar scars in the bronchovascular bundles, pleural arteries showed medial hypertrophy and the interlobular septa contained dilated prominent veins, as well as cystically dilated inflamed peripheral bronchus. These cases highlight the importance of vascular changes in SJMLS, likely secondary to the bronchiolar inflammation and destruction leading to capillary bed destruction from secondary emphysema and reactive pulmonary and arterial changes. Mediastinal nodal staging with EBUS is recommended for patients with resectable non-small cell lung cancer and has emerged as a safe tool to establish granulomatous pathology in suspected sarcoidosis.

We conducted a retrospective analysis of the outcomes of EBUS performed in a large teaching university hospital with a rapid access lung clinic over a 12 month period and correlation with endobronchial and CT guided biopsies, and surgical resections, when available, and compared the adequacy of EBUS when performed with and without rapid on-site evaluation (ROSE Background and aims: In Interstitial lung disease (ILD), when an aetiological factor appears absent and clinical-radiological correlation is non-contributory, histology is required. The traditional surgical lung biopsy (SLB) is not without risks. Cryotechnically obtained specimens contain more alveolated lung tissue and less crush artefact than conventional transbronchial biopsies and may offer an alternative to SLB in selected cases. We aimed at studying the complications of cryoprobe transbronchial lung biopsy (CPBx) and the quality and pathological characteristics of the tissue obtained.

Methods: This is a prospective study of patients who were selected for CPBx including cases of possible/probable idiopathic pulmonary fibrosis (IPF). Complications of the procedure as well as the quality and pathological characteristics of the tissue are studied. Results: Twenty-seven procedures were performed in 24 patients, 20 of which were radiologically IPF. A total of 77 biopsies were obtained (Average 2.85 biopsies per procedure).

Only one was inadequate initially. Fibroblast foci and features consistent with usual interstitial pneumonia (UIP) pattern were present in 18 biopsies from 16 patients (66.7% of total; 80% of suspected IPF cases). Granulomas were identified in 4 patients (16.7%), 3 of which were radiologically suspected IPF (15% of suspected IPF cases). Two patients (8.3%) had organizing pneumonia; both were inconsistent with IPF radiologically. The findings in the remaining 3 patients were nonspecific; two of these were radiologically IPF (10% of IPF cases). Seven patients (25.9% of procedures) developed pneumothorax, only 2 of them (7.4%) required chest tube drainage. Five patients (18.5%) developed bleeding (moderate in 3 (11.1%) and mild in 2 (7.4%)). Conclusion: CPBx was useful in this cohort at potentially identifying features not typical of IPF and displayed an acceptable complication rate.

Cardiomyopathy ZJ van der Klooster 1 ; S Sepehrkhouy 1 ; M Harakalova 1 ; R Goldschmeding 1 ; N de Jonge 1 ; AJH Suurmeijer 2 ; RA de Weger 1 ; F Asselbergs 1 ; P A Vink 1

Introduction: Genetic dilated cardiomyopathy is a heterogenous group of diseases caused by mutations in various genes. Several types of cardiomyocyte cell death have been implicated in dilated cardiomyopathy: (macro)autophagy-related cell death, apoptosis, necroptosis and oncosis. One plausible mechanism of genetic cardiomyopathy is proteotoxicity of accumulated protein aggregates. We investigated the association of such aggregates as sign of autophagy-related cardiomyocyte cell death with specific pathogenic mutations. Methods: Hearts from 30 patients with a genetic dilated cardiomyopathy or a combined phenotype of dilated and arrhythmogenic cardiomyopathy were included. Microscopic slices from 8 regions were immunohistochemically stained for P62, a marker for aggregated proteins destined for autophagy. Results: Sporadic P62 positive cells were seen in control hearts (0.5% of cardiomyocytes, range 0.1-0.8%). Troponin mutations (TNNT2 and TNNI3; 0.7%, range 0.2-1.2%, n=3) showed hardly any increase in P62. Titin (1.6%, range 0.7-2.7% ,n=5) and lamin A/C (1.7%, range 1.1-2.5%, n=5) mutations showed a threefold increase in P62 staining. A tenfold positive staining was found in desmosomal mutations (PKP2 and DSP; 3.8%, range 3.7-4.0%, n=3) and myosin mutations (MYH7 and MYBPC3; 4,6% range 3.3-5.7%, n=3). Phospholamban mutations (8.8%, range 4.1-16%, n=8) and desminopathies (desmin and Alpha-B crystallin; 17% of cardiomyocytes, range 4.0-31%, n=3) showed the highest number of P62 positive cells.

Conclusion: Accumulation of P62 positive protein aggregates is associated with the type of mutation underlying the dilated cardiomyopathy. Titin, lamin A/C and troponin mutations revealed little protein aggregation, whereas desminopathies, phospholamban, desmosomal and myosin mutations show abundant aggregates. This suggests that the type of mutation plays an important role in determining distinct mechanisms of cardiomyocyte cell death.

Major Trauma Centre Status and its Impact on the Department of Cellular and Anatomical Pathology in a large Tertiary Referral Centre P RA Hadden

Background: Major Trauma has been centralised into Major Trauma Centres which act as the focus of Major Trauma Networks. In April 2012, Derriford Hospital in Plymouth, Devon became operational as the regional Major Trauma Centre for the South West Peninsula. As a result, there was potential for an increased number of trauma-related deaths to be referred to the local coroner, as well as surgical specimens, potentially increasing the work load on pathologists. The case mix could include post-operative cases, neurosurgical cases, polytrauma cases and forensic cases. Methods: On admission, all eligible trauma patients are recorded onto the Trauma Audit & Research Network (TARN) database. The TARN data was retrospectively analysed and cross referenced with the Department of Cellular and Anatomical Pathologies database to determine how many patients had died, how many had post-mortem examinations were performed and how many surgical specimens were sent, on patients from outside the region or transferred from smaller Major Trauma Units. Results: Over the first two years, there was a small increase in workload from patients who, prior to Trauma Centre status would have gone to other centres. Conclusions: In recieveing patients from elsewhere in the region, there was an increase in workload for both autopsy and non-autopsy work. This excluded some neurosurgical cases, which traditionally would have been referred (as Derriford is the neurosurgical centre). There are several areas for implication including, APT time, mortuary space and non-autopsy surgical work. Although the workload increase is small, at a time when services are being stretched it is important to ensure any increase in work will not be the "straw that broke the camels back" and can be dealt with accordingly.

Derriford Hospital, Plymouth, UK

To attempt to streamline general pathologist's approach to the investigation of potentially asbestos-related deaths Methods Turnaround times, tissue sampling protocol and frequency with which samples were sent for formal fibre counts was investigated for 100 consecutive coronial autopsies at the author's institution. Colleagues at other institutions were questioned about their own practice when investigating cases of potential asbestosis, lung cancer or mesothelioma.

The author found no consensus in opinion on methods of sampling of the lungs in potential asbestosis, lung cancer or mesothelioma. The most common indication for samples to be sent for asbestos fibre counts was for malignant mesothelioma. Sending tissue for fibre counts led to considerable delays in the authorisation of postmortem reports and to significant cost implications. The author presents a pragmatic algorithmic guide to approaching potentially asbestosis-related deaths with suggestions for sampling the lungs and tumour in all cases. In general terms, malignant mesothelioma previously confirmed premortem with histology and immunohistochemistry should not require extensive postmortem histological sampling. Lung cancer and asbestosis require widespread sampling of lung tissue to determine amphibole count according to Helsinki criteria in the former, and in the latter, assessment of the distribution and degree of fibrosis in addition to fibre count. One or more of these tissue blocks can be sent for formal counts in equivocal cases after following the algorithmic approach.

Conclusions Although predominantly intended as a pragmatic approach to assist the busy practicing autopsy pathologist, the author believes that the algorithm presented will help departments streamline their approach to these cases and help the relative of the deceased gain access to compensation when appropriate in a more timely fashion.

Purpose of the Study: This is a case report of a three year old girl who died suddenly at home. An autopsy was performed in order to determine the cause of death. Method: An autopsy was conducted which showed no gross abnormalities. Microscopy of the main organs and microbiological samples were taken for further assessment. Results: Histological assessment of the heart showed multiple small foci of lymphocytes around vessels and within the interstitum of the epicardium, myocardium and subendocardium. These lymphoid aggregates consisted of 20-30 lymphocytes up to larger numbers of 100 lymphocytes collectively. Several foci were present within virtually all of the sections taken in both right and left ventricles. There was no evidence of myocyte necrosis. Immunohistochemistry confirmed they were of T lymphocyte cell origin admixed with smaller numbers of macrophages. Histology from the respiratory system showed a diffuse subepithelial lymphocytic infiltrate in the larynx and trachea, and the nasopharyngeal samples detected Coronavirus, Adenovirus and two types of Parainfluenza virus. However, viral polymerase chain reaction (PCR) from the cardiac tissue was negative. Conclusion: An unequivocal diagnosis of a myocarditis could not be made in this case due to the lack of myocyte necrosis and the absence of viral DNA within the cardiac tissue. Genetic testing was strongly advised as splenic material had been taken at autopsy and following molecular genetic techniques a mutation was detected in the sodium channel indicating an inherited ion channelopathy. Further genetic counselling and testing of the remaining siblings and family members is being performed.

Varicose veins affect a third of the UK population. Isolated case reports and small series of fatalities resulting from varicose vein haemorrhage appear in the literature infrequently. Some of the earliest reports of fatality we have found appear in British Medical Journal (1958) and the Lancet (1973), more recently they have appeared in journals of forensic pathology. Our purpose is to establish and bring attention to the rarity of fatality resulting from varicose vein haemorrhage and the importance of the scene of death and autopsy findings. A literature review was undertaken, we obtained relevant Office of National Statistics (ONS) mortality data for the years 2011-2013, and reviewed our own post-mortem records for demographic, clinical and scene of death information in cases we have encountered. Our findings confirm that fatality resulting from varicose veins remains a rare cause of death. Some of these deaths are preventable and in 2013 NICE (National Institute of Health and Care Excellence, UK) issued guidelines in which haemorrhage from varicose veins constitute a vascular emergency. Importantly emphasis on first aid is required, simply elevating the limb stops bleeding and is life saving, whereas direct pressure and tourniquets do not.

Pathologists should be aware of potential findings at autopsy in these cases. In particular, awareness that even obscure minor injury to a varicose vein could have resulted in significant blood loss leading to death. Blood lost at the scene will not be apparent at autopsy, and details of blood loss could be variably recorded on the scene of death information provided, therefore vigilance is required.

Histopathologists practice in an era of ever advancing medical treatments for a wide variety of oncological, neurological, haematological and rheumatological diseases. Immune modulating therapies are taking a more prominent place in clinical practice. However, with such great advances in therapy comes great risk, with the potential of life threatening opportunistic infections in our patients. We present a series of 9 immunosuppressed patients who acquired such infections and in whom the diagnoses were made by histopathological examination. The spectrum of these pathogens ranges from viral (CMV, EBV, Herpes), parasitic (strongyloides) to fungal (P. jirovecii, cryptococcus), and the range of infections is diverse. Our series includes 4 males and 5 females, with an age range of 32 -72 (mean age = 57 years). Unsuspected infectious diagnoses were made at post mortem in 6 of the 9 cases. Organs affected included lung (n = 5), brain (n = 1) and haematological system (n = 1). In one case both colon and lung were affected (n=1) and in a further case both liver and lung were affected (n=1). Immunohistochemistry and/or histochemistry was invaluable in making the diagnoses and was used in all 9 cases (n=9). Treatments leading to immunosuppression included chemotherapeutic agents, monoclonal antibodies, steroids and methotrexate. We believe that with the ever increasing use of immunosuppressive therapies (both new and old) for a wider number of disorders, vigilance should be paid to their potential to cause life threatening side effects. Histopathologists play a pivotal role in the recognition of this risk and in the diagnosis of these diseases.

Audit of Hospital-Based Adult Autopsy Practice in a University Hospital from July 2013-2014 P D Abu-Sinn; F MacSweeney

The contribution of hospital-based autopsy practice to improvements in patient care is substantial; however, there remains a void in the processes of audit and raising quality of standards in autopsy services. We aim to assess the current autopsy practice compared to RCPath guidelines and identify areas for achieving a high quality autopsy service. All adult autopsy cases performed at a university hospital mortuary between July 2013 and 2014 were reviewed. A total of 522 adult autopsies were performed by 5 Consultant Histopathologists. Ninety nine percent were Coroners' cases. The median turnaround time was 38.5 days, with a range of 3-123 days, excluding 31 outlier cases (complex timeconsuming cases). There was considerable variation in turnaround times in complex cases and between the various reporting pathologists. Eighty five percent of cases were compliant with RCPath Minimum Dataset for Autopsy Practice. The remainder were lacking clinical information only. Histology and toxicology contributed to cause of death in 34.4% and 14.8% respectively. No organs were retained. Further review of the cases not compliant with RCPath guidelines (15%), identified that the possible reasons were the inaccuracy, and sometimes irrelevance to the cause of death, of the information received by the pathologists. In many instances, the clinical information given to the pathologist may be controversial, and a certain degree of caution needs to be implemented to avoid including misleading information in the autopsy report. The turnaround times could be improved if a preliminary report is issued within a set time frame, to be followed by the complete report when the histology and toxicology results are available. However, this practice is not acceptable to some coroners who prefer one complete final report. Variations in autopsy practice are to be expected as each autopsy involves substantial case-specific information to which a case-specific answer to the cause of death is expected. Prostate cancer is the second most common form of cancer in males, and the incidence of this disease is predicted to double globally by 2030. More than 1.1 million new cases of prostate cancer are diagnosed each year and two thirds of these patients are from the Western world. The current PSA-based test for the diagnosis of prostate cancer lacks specificity, results in missed-diagnoses, over-diagnosis and unnecessary biopsies/treatment. There is an urgent need for a method that enables early accurate detection of prostate cancer. Endosomes and lysosomes are cellular compartments that degrade and turnover macromolecules in order to maintain cellular homeostasis. These organelles are directly involved in the critical processes of energy metabolism, cell division, and intracellular signalling, which are all hallmarks of cancer pathogenesis. Endosomes have a critical role in controlling the secretion of proteins into extracellular fluids, making them an ideal system to identify new biomarkers that are released from cancer cells. We have discovered that endosome biogenesis (formation and function of endosomes) is altered in prostate cancer. There were significant changes in the gene and protein expression for 19 endosomal proteins and differential distribution of endosome subsets in prostate cancer cell lines. There were also changes to the endosomal traffic and signalling of the transferrin receptor in prostate cancer cells. These fundamental changes in the cell biology of prostate cancer have allowed us to identify a specific set of endosomal proteins that have diagnostic potential. We are developing ELISA's to quantify these endosomal proteins in patient samples and antibodies for immune histology applications. The objective for this project is to develop an effective method for the early and specific diagnosis of prostate cancer, which is important as this will have a major impact on patient outcome and survival. The incidence of malignant melanoma has rapidly increased in recent times and melanoma currently represents the second most common cancer diagnosed in young adults. Diagnosis is based predominantly on histological assessment; however, due to the wide spectrum of morphological characteristics and lack of firm diagnostic criteria, accurate diagnosis can be challenging. Some atypical melanocytic lesions do not display clear-cut morphological features to allow distinction of benign from malignant tumours, making diagnosis and treatment difficult. Among these atypical melanocytic lesions blue nevi, Spitz nevi and dysplastic lesions are common. From histological features alone, it can be difficult to exclude a diagnosis of melanoma and therefore aggressive surgical strategies may be employed in cases were they are unnecessary, highlighting the need for improved diagnostic techniques. Both mRNA and miRNA profiling have been shown to be able to distinguish benign nevi and primary melanoma tumours. Studying miRNA expression levels is an attractive strategy as miRNAs are highly resistant to degradation and can be easily analysed in FFPE samples. We have studied miRNA expression levels in a cohort of benign, blue, Spitz and dysplastic nevi versus primary melanoma tumours and their derived metastases. Expression levels of key melanoma miRNAs, including miRNA 21, miRNA 211, miRNA 205 and miRNA 200c can be used to distinguish between nevi and malignant melanomas. We propose an easy to implement, simple and robust molecular method based on miRNA expression ratio that, in combination with histological assessment, allows diagnosis of difficult to classify atypical melanocytic lesions. Background: Diagnosis of Lynch syndrome (LS) traditionally relies on clinical criteria to guide diagnostic genetic testing. MMR status of the patient's tumour can help detect Lynch syndrome families as well as having other recognised applications for the patient's management including prognostic and predictive significance. As such, the 'Dataset for colorectal cancer histopathology report' recommendations from the Royal college of pathologists were updated in July 2014 to include screening of colorectal cancer patients under the age of 50 and molecular testing for abnormalities in the mismatch repair genes. In South-East of Scotland we introduced molecular testing to identify individuals at risk of LS. To widen our screening in line with revised guidelines set by European experts, we expanded our cohort criteria to include those between the age of 50 and 60. Methods: Molecular analysis was carried out on 553 individuals: 446 via 'reflex testing' (newly diagnosed colorectal carcinoma ≤60 yrs, or clinical/ pathological features associated with MMR defects, such as pre-menopausal endometrial carcinoma, multiple tumours and medullary-type carcinomas) and 107 via 'request testing' (clinical criteria and referral dependent). 'Molecular-positive' profiles for LS were identified for genetic pre-testing counselling/diagnostic testing. Results: 41 patients with potential LS were identified, 24 (58.5%) underwent genetic counselling/testing and 13 cases were confirmed LS with germline pathogenic mutations in the MMR genes. Eight of these were identified using reflex testing. Conclusion: This is the first UK study to show that screening for LS in patients with colorectal cancer under the age of 60 is effective at identifying families with LS. The testing protocol is in line with the recent recommendations.

The human microbiome is rich and diverse, especially in the oral cavity and gastro-intestinal tract, where it has been shown to be more stable in adults, although various factors such as diet and antibiotics mays influence its composition. This pilot study aimed to examine and compare the oral and gut microbial composition in four individuals using a culture-independent approach. Methods: Saliva and faecal samples were collected from volunteers within the same day on two separate occasions. The V4 region of the 16S rRNA gene was amplified in all samples and PCR products sequenced on an Illumina MiSeq. Unique barcodes were used to sequence 24 multiplexed libraries together. The data were analysed using the Quantitative Insights into Microbial Ecology (QIIME) software. A second series of 42 samples of faeces from 3 individuals were run to investigate consistency over time.

Operational taxonomic units (OTUs) were assessed and showed 5 major phyla represented in the saliva samples: Firmicutes, Proteobacteria, Bacteroidetes, Fusobacteria and Actinobacteria. Similar phyla except forFusobacteria, were found in the stool samples. The weighted Unifrac PCoA analysis displayed a clear separation of the 2 sample groups, and also showed a more disperse bacterial profile for the saliva samples, based on population sizes, whereas rarefaction curves and unweighted analysis indicated higher bacterial diversity in the stool samples. Each individual could be distinguished either by oral or faecal microbiome. One volunteer who had had previous radiotherapy to the mouth displayed a particularly distinct oral microbiota.

The microbial community profiles of saliva and faecal samples of four individuals were found to be distinct from each other, despite sharing similar phyla. Analysis of multiple samples from each volunteer clearly separated each sample by volunteer and by sample type.

Are Current Automated Approaches for Determining the Phylogeny of Multiple Deposits Capable of Interpreting the Complexity of Cancer Evolution? P TG Palmer; HM Wood; M Taylor; W Fateen; IM Carr; P Quirke

Tumour heterogeneity is central to chemotherapy resistance and disease progression in advanced malignancy. This heterogeneity arises due to the evolution of clones within the tumour cell population; the advent of high throughput sequencing has allowed the detection of different tumour cell clones within and between primary tumours and their metastases, potentially allowing mapping of tumour evolution. Several, automated bioinformatic approaches have been devised for determining tumour phylogeny from changes in genomic copy number (CN); either by the overall similarity of genomic changes between tumour deposits or by examining the occurrence of shared breakpoints.

We have compared these automated approaches with a manual determination of phylogeny based upon shared breakpoints identified from four cases of metastatic colorectal cancer consisting of between 6 and 53 deposits. We illustrate several recurrent issues identified with the use of automated systems for the determination of tumour phylogeny associated with an inability to correctly identify and interpret changes in ploidy, an inability to identify heterogeneity within tumour deposits, the masking of smaller events by larger ones, overinterpretation of convergent, but unrelated events, over calling sequencing artefacts as changes in CN, and non-calling of genuine CN changes due to low tumour cell content or low sequencing depth. We conclude that manual interpretation of bioinformatics data is still required to determine the phylogeny of metastatic cancer within an individual. Results: Levels of agreement between each sample size and the 'gold standard' were evaluated using Bland-Altman plots. 7 separate pairwise comparisons were performed. Some small sample sizes were shown to have small mean difference and narrow limit of agreement. The Ki-67 PIs were then translated into grades and similar comparisons were performed by calculating the kappa score for categorical variables. Additionally, the interobserver variation between the two independent researchers were calculated. Conclusion: Smaller sample sizes (below 1000) tend to overestimate the Ki-67 PIs, possibly due to the effect of concentric counting starting from the center of the hotspot. However, the Ki-67 PIs do start to stabilise closer to 2000 (e.g. 1500). The interpretation of whether a lower sample size can replace the current standard would be a subjective decision, but the kappa score gives a rough idea of how much it affects the clinical grading. Updated data will be presented. Purpose of the Study: Targeting the stem cell properties of tumor-initiating cells is an avenue through which cancer treatment may be improved. Before this can be achieved, so-called cancer stem cell (CSC) models must be developed and characterized in specific malignancies. Methods: In this study, holoclone formation assays were used to characterize stem-like molecular signatures for prostate cancer (PCa) cells. Summary of Results: LNCaP and PC3 parent cells were capable of responding to stem cell differentiation morphogen retinoic acid (RA), suggesting the presence of inherent stemlike properties. LNCaP cells, which represent early, androgen-responsive disease, formed holoclones after twenty six days. PC3 cells, which represent advanced, metastatic, castrationresistant disease, formed holoclones after only six days. Holoclones displayed decreased expression of RA-genes, suggesting a more immature, less differentiated phenotype. Gene and mircoRNA arrays demonstrated that holoclones downregulated a number of stem cell differentiation regulators while displaying enhanced regulation of G2 to M transition and the mitotic spindle checkpoint components of the cell cycle. PC3 holoclones displayed pronounced downregulation of known regulators of osteoblast differentiation from mesenchymal stem cells and Epithelial Mesenchymal Transition. Conclusion: Our results suggest that some PCa cells retain the ability to transition to a more immature state in which differentiation and metastatic mechanisms are changed. The highlighting of osteoblast differentiation regulators in this mechanism is particularly notable, considering the propensity of PCa to metastasize to bone. We examined by flow cytometetry the interaction in vitro between platelets and 15 human cancer cell lines of different origin and metastatic potential. The EMT profile of cells 24hr post platelet exposure was assessed by morphology and gene expression analysis (RT-PCR).

Here we showed that platelet cloaking of cancer cells is universal, occurring across all 7 tumour types examined. However, it is heterogeneous with adhesion rates varying both across and within tumour types, from 35% (PC3-metastatic prostate cancer) to 83% (SKMES1-metastatic lung cancer). Changes indicative of EMT were seen in all cell lines. However, again they were heterogeneous in nature; with morphology changes akin to EMT observed at varying degrees across the cancer types. Also, there was no consistent pattern to the EMT-like gene expression changes seen, with one exception a significant increase in the expression of plasminogen activator inhibitor 1 (PAI-1)was observed in 93% of the cell lines examined.

In this study we describe the universal nature of platelet cloaking and that even though the interaction is not inducing precisely the same molecular changes in all the cancer cells; overall it is driving these cells into a mesenchymal phenotype. Giant cell tumours of bone (GCT) are primary locally aggressive bone tumours with a recurrence rate of up to ~30%. The tumour is characterised by numerous osteoclasts and neoplastic stromal cells. Making a diagnosis can be challenging because the differential diagnosis incudes an array of benign osteoclast-rich tumours but also osteoclast-rich osteosarcoma. Recently the occurrence of H3F3A p.Gly34Try (G34W) and G34L mutations was reported in 96% of GCT, the latter occurring rarely. These mutations occur in less than 2% of >400 other benign and malignant bone tumours. It has been emphasised that a diagnosis of GCT should be made with caution in the absence of detection of G34W substitution. Given the diagnostic importance of G34W mutation in GCT, we have developed a simple, quick and cost-efficient diagnostic test to detect this recurrent alteration in FFPE DNA using droplet digital PCR (ddPCR). The ddPCR data from DNA of >100 GCT have been compared with previous 'genotype' data generated using Sanger sequencing, and a number of next generation sequencing approaches (whole exome, whole genome and targeted panels). The G34W and G34L can both be detected in a single assay. We have demonstrated the sensitivity, specificity, repeatability and robustness of the test to be very high with a turnaround time of no more than 5 working days. As ~30% of GCT recur locally following curettage a blood test would be valuable to monitor patients. To this end the ddPCR test also shows that the mutation can be detected in plasma. Typically resistant to chemotherapy and radiotherapy, high grade disease has been treated by surgery for more than 50 years. It has recently been shown that IDH1 and IDH2 mutations are present ab initio in ~60% of chondrosarcoma cases and that these are retained throughout disease progression. This has opened up a number of new potential diagnostic, biomarker and therapeutic options. Digital PCR is currently the most sensitive and accurate method for detecting and quantifying mutant DNA molecules. The BioRad QX200 digital PCR platform is also both cost effective and scalable. Using the QX200 platform, we have developed assays for the 5 common IDH1 mutations and the 1 common IDH2 mutation. We have developed the IDH1 assays both in singleplex and multiplex. We have optimised and validated all assays in tissue samples demonstrating both high sensitivity and specificity when compared to previously genotypes samples. We have demonstrated that the assays are quantitative over 4 orders of magnitude and in high quality DNA we can detect IDH mutations at below 1 mutant molecule in 10,000 wild type molecules. In a pilot study, we have used digital PCR to analyse circulating tumour DNA levels in plasma taken pre-surgery from 14 patients whose chondrosarcoma harbour an IDH1 mutation. It was possible to detect IDH1 mutant molecules in plasma of all Grade III samples, 50% of Grade II and none of the Grade I samples. In 4 of these cases where the ctDNA was also measured post-operatively, the levels of ctDNA dropped dramatically.

Tumour Necrosis Factor Receptor, CD40, Gene Functions as an Oncogene and Promotes Cell Proliferation in Colorectal Cancer Cell Lines P HAA Almasmoum; H Thorpe; M Ilyas

Introduction: CD40 is a tumour necrosis factor (TNF) receptor which regulates a range of cellular responses. CD40 is activated by its ligand CD40L and may promote tumourigenesis in haematological cancers. However, CD40 functions as a tumour suppressor in solid cancers. CD40 maps to chromosome 20q13, a region which is amplified in 40-50% of colorectal cancer (CRC). The functional activities of CD40 were tested in CRC cell lines for cell proliferation and motility. Methods: Expression of CD40 was screened in CRC cell lines by western blot. To define the role of CD40 in human CRC, we knocked down CD40 using small interfering RNA (siRNA) and the knockdown was confirmed by qPCR and western blot. The PrestoBlue assay was used to study proliferation in colorectal cell lines, and flow cytometry to study the cell cycle. Transwell migration and wound healing assays were performed to investigate the effect of CD40 on cell motility in CRC.

Result: CD40 was expressed in CRC cell lines HCT116, RKO, DLD1 and HT29, and not expressed in SW480 and SW620 cell lines. Knockdown of CD40 reduced cellular proliferation in HCT116 (p=0.0158) and DLD1 (p=0.0020) cell lines. Knockdown of CD40 showed a higher number of cells in the sub G0 phase (dead cells) in the cell cycle analysis compared to the control. However, knockdown of CD40 in HCT116 did not have an effect on cell motility in both the transwell migration (HCT116= p=0.253) and wound healing assays. Discussion: CD40 exhibited oncogenic activity in CRC cell lines. CD40 enhanced cell proliferation but not cell motility in CRC cell lines.

The expression of CD45 (Common Leucocyte Antigen) and cytokeratin is thought to be mutually exclusive with CD45 expression largely restricted to haematological malignancies and cytokeratin expression largely restricted to carcinomas. We report two clinically relevant cases. The first case is a urinary bladder biopsy showing a high grade malignant tumour with cells that had scanty cytoplasm, hyperchromatic stippled nuclei and high mitotic activity. Nuclear molding was present. The tumour cells showed focal strong positivity for CK7 and diffuse positivity for CD56 and synaptophysin. Focal positivity for CD45 was present and confirmed on repeat staining. The morphology and immunoprofile was consistent with a small cell carcinoma showing aberrant CD45 expression. The second case is a maxillary tumour biopsy composed of medium/large atypical lymphoid cells with hyperchromatic nuclei, small nucleoli and scanty cytoplasm. Mitoses and apoptotic cells were noted. Immunohistochemistry showed the atypical cells to express CD20, CD79a, Bcl6 and MUM1 but not CD45, CD5, CD10, cyclin D1, CD30, TdT, ALK1, CD2, CD3, neuroendocrine or melanocytic markers. A high Ki67 proliferation fraction was present. The appearances were consistent with a diffuse large B-cell lymphoma. The above cases highlight the possibility of aberrant expression as well as loss of expression of immunohistochemical markers by neoplastic cells in undifferentiated malignancies. Attention to tumour morphology may provide diagnostic clues. Interpretation of immunohistochemistry in the context of tumour morphology as well as awareness of aberrant expression/loss of expression can help avoid diagnostic error.

Accurate, timely diagnosis is the ultimate aim in surgical pathology. Numerous histological pitfalls and lesional mimics exist, with the need to maintain an awareness of such entities vital if potentially serious misdiagnoses are to be avoided. This case report describes two distinct lesions within the same lymph node, both of which are potential mimics of each other. A 43 year old female presented with a three week history of a left breast lump. She had no known previous breast disease or any associated risk factors. A needle core biopsy of this clinically and radiologically suspicious mass yielded a diagnosis of grade 2 invasive lobular carcinoma. Left axillary sentinel node biopsy was thus undertaken. Two hot and blue sentinel lymph nodes were excised. One was free of neoplasia, the second contained benign naevus cell inclusions within the capsule together with a micrometastasis. Immunocytochemistry confirmed the presence of two distinct cell populations; the benign naevus inclusion cells stained positively for S100 but not for AE1/3, the reverse pattern was observed in the invasive lobular carcinoma cells. Heterotrophic benign inclusions within lymph nodes are an infrequent yet well recognised entity. Ridolfi et al reviewed the lymph nodes from 909 axillary surgery patients and found 0.017% of lymph nodes contained benign naevus cell inclusions. Small benign naevus cells within the capsule of a lymph node can resemble the 'Indian file' pattern of classic invasive lobular carcinoma. This case is unusual in that both metastatic carcinoma and benign naevus inclusion cells were present within the same lymph node, enabling a clear comparison of the cytomorphology and immunoprofile of these two distinct lesions. An awareness of benign inclusions within lymph nodes helps to avoid the potential for misdiagnosis. The judicious use of immunocytochemistry can be useful in distinguishing benign inclusions from carcinoma. Recently there has been increasing recognition of distinct breast cancer phenotypes. Of these, Basal phenotype breast cancer (BBC) has attracted particular interest since the majority are triple negative (TN); have an aggressive natural history and can be associated with BRCA1 germline mutation. This area is mired in difficulty, as a precise unifying definition of BBC remains elusive. Several morphological features more prevalent in BBC have been identified. In our practise we noticed variable use of 'basal' in reports. Given this, whilst no specific therapies to BBC currently exist, we felt it necessary to understand how accurate our designations have been and whether this is a worthwhile practise. Method: The diagnostic database was searched for all malignant TN breast resections or reports containing the word 'basal' within 12-months. TN was defined as Allred score ER 0-2/8, PR 0-2/8 and HER2 0, 1+ or 2+ negative on FISH. For completeness, we considered including all breast cancers, but pragmatically this was not possible. We carried out CK5 and CK14 staining on all cases where not performed. Results: Of the 618 invasive breast cancers, 69 cases (11%) were identified, of which 88% were TN and 28% were designated BBC in the report. 16% were both TN and BBC. Where a diagnosis of BBC was made, 37% of cases had additional markers requested. Preliminary results showed 94% were CK5+ and CK14+. This is higher than other studies, implying specificity but not sensitivity in suspecting BBC amongst reporting pathologists. Discussion: A limitation of this review is that it cannot identify the rare non-TN BBC not diagnosed as such. It also represents current practise in a single institute and may not reflect national practise. We identified patchy use of the designation BBC, with overall under-reporting of this subtype. We recommend that if it becomes necessary to distinguish BBC lesions, additional markers studies, such as CK5 and 14 be consistently performed. Purpose: Neoadjuvant chemotherapy (NACT) is increasingly used for the management of large but operable, inflammatory, and locally advanced breast cancer (LABC). Little is known about predictors of response/survival following NACT. The topoisomerase IIα (TOP2A ) gene, a key regulator of DNA repair and modelling, is thought to be target for anthracyclin and other chemotherapeutic agents. The aim of this study is to assess the role of TOP2A as marker for response/resistance to NACT and patient outcome. Methods: Patients who underwent NACT, predominantly anthracyclin, for primary and operable invasive carcinoma or LABC in the period between 2005 to 20013 at a single large tertiary referral breast unit were identified. Comprehensive data on chemotherapy regimen, surgical treatment, pathological response and survival were collected. Pre-treatment tumour samples were stained for standard predictive and prognostic markers and TOP2A. Results were correlated with pathological response (PR) and patient survival. Results: 252 patients fulfilled inclusion criteria. Mean age was 48.94ys. Complete PR was achieved in 15.4%. The mean expression level of TOP2A in pre-treatment core biopsies was 75.4%, range 0-95%. There was significantly higher expression in high grade tumours (p=0.04) and positive correlation with ki67 expression (r=0.405, p<0.001). There was no correlation with nodal status, PR or HER2 expression. Cases with high expression (>50%), had significantly worse overall survival (mean 38 vs 52 months, p=0.01). This was also identified in the endocrine non responsive group (ER Allred score≤4), mean 74 vs 33 months, p=0.04. On multivariate analysis, TOP2A was not an independent factor for overall survival. Conclusions: TOP2A protein is expressed in high grade breast carcinoma with high Ki67 proliferation index. Its expression in pre-treatment biopsies predicted patient outcome in the neoadjuvant setting. This strong adverse effect on survival warrents further prospective investigation as a marker of outcome in NACT patients. The risk of circumferential resection margin (CRM) involvement is confined to tumours of the rectum with the risk of peritoneal involvement increasing the further a tumour is located above the peritoneal reflection. There is no internationally accepted definition of the upper limit of the rectum, and the term 'rectosigmoid' is frequently applied to tumours in this area leading to confusion around the risks and whether radiotherapy can be given. The photographs from 331 abdominoperineal excision specimens were available for quantitation using Aperio ImageScope. Both fresh and fixed specimen images were included where available. The position of the anal verge, top of the sphincters, anterior peritoneal reflection, mesorectal apex (defining the limit of the mesorectum) and high vascular tie were identified and the distances between each point measured. The work was supported by a PathSoc bursary. There was wide variation in the length of the mesorectum in both fresh (median 172 mm, IQR 146 to 199 mm) and fixed (166 mm, 140 to 196 mm) specimens. The length of the anal canal also showed variation (fresh 66 mm, 49 to 78 mm; fixed 66 mm, 53 to 75 mm). The height of the anterior peritoneal reflection was lower in females compared to males (fresh 125 vs. 132 mm, p=0.288; fixed 111 vs. 126 mm, p=0.034). There is marked variability in the anatomy of the rectum between individuals and genders. This potentially affects the risk of either CRM or peritoneal involvement and whether radiotherapy could be offered. A fixed definition of the upper limit of the rectum for all patients is not helpful. This should be determined for individual patients on the basis of the MRI findings. The term 'rectosigmoid' should be abolished and more accurate definitions based on the position of the mesorectal apex and commencement of the sigmoid mesentery should be used to define the boundaries of the rectum and sigmoid colon and determine subsequent risks to the patient. Pre-operative chemoradiotherapy (CRT) with anti-EGFR antibodies may change the status of EGFR pathway mutations. We assessed the mutational status of a number of EGFR pathway genes before and after CRT in the NWCOG EXCITE trial. Patients with MRI-threatened surgical margins were given pelvic radiotherapy (45Gy) with capecitabine, irinotecan and cetuximab followed by surgery after 8 weeks. DNA was retrospectively extracted from the pre-treatment biopsy and resection specimen by macrodissecting areas of greatest residual tumour. The mutational status of KRAS (codons 12/13/61/146), NRAS (12/13/61), PIK3CA (542/545/546/1047) and BRAF (V600E hotspot) were determined by pyrosequencing. The work is presented on behalf of the NWCOG EXCITE trial investigators and was part-funded by a PathSoc fellowship. 80 patients commenced treatment and 76 underwent surgery with pathological complete response in 14 (18%) and near-complete in 6 (8%). Pre-treatment testing (n=78) detected mutations in KRAS (n=34), BRAF (n=3), NRAS (n=3) and PIK3CA (n=10). Any EGFR pathway mutation was detected in 58%. Following CRT, cases with residual tumour able to be tested (n=54) showed mutations in 32 patients (59%). There was a discrepancy compared to pre-treatment biopsy in 18 cases (33%): from wild-type (wt) to mutant (mut) in 9, from mut to different mut in 1 and from mut to wt in 7. One patient changed in 3 codons (mut to wt in KRAS 146/PIK3CA 545 and wt to mut in KRAS 12). In 12 patients (22%) this changed their overall EGFR pathway status (6x wt to mut and 6x mut to wt). Intratumour heterogeneity may explain some of the differences in EGFR pathway mutations reported between biopsies and resections presenting a challenge to personalised medicine. However, cetuximab may also drive the growth of undetectable mutant clones to detectable levels on pyrosequencing. Further assessment using more sensitive sequencing technologies is currently being employed to investigate these differences. There is a vast amount of historical FFPE material held in archives, but due to variations in fixation and processing this presents several challenges when applying newer genomic technologies to it. In this study we compared the genomic information obtained with the OncoScan® FFPE Assay Kit (OncoScan) and next generation sequencing (NGS). Samples from 378 patients were obtained from 10 centres taking part in the MRC CR07 trial of short course radiotherapy versus selective long course chemoradiotherapy in rectal cancer. DNA was prepared using Agilent SureSelect kits and sequenced using Illumina platforms in parallel to analysis using the OncoScan assay. For both methods, quality control (QC) data was generated and the sample classified as a 'pass' if it fell within the pre-defined QC boundaries. For the OncoScan assay, copy number (CN) and somatic mutation (SM) data was further investigated. This study was part funded by a PathSoc Fellowship. In total, 272 cases (72%) passed the NGS QC and 232 (61%) passed the OncoScan QC. A total of 186 (49%) passed QC on both platforms with marked variability in sample pass rates between the 10 centres for the NGS (range 0% to 100%) and OncoScan (ranges 33% to 84%). When assessed manually, the OncoScan SM data was considered acceptable for 273 cases (72%), which included 40 initially classified as 'failed' by the QC data. Similarly, the OncoScan CNV data was interpretable for the majority of cases. This study has shown that whilst historical DNA held in the FFPE blocks of archival clinical trials like MRC CR07 can present challenges when using new genomic technologies, a large proportion of samples can still yield valuable genomic data. Marked variation exists in the quality of genomic material between centres confirming that differences in specimen handling affect DNA quality. Prospective trials must address this by standardising fixation and processing protocols. The plane of colon cancer resection has recently been shown to predict survival. Complete mesocolic excision (CME) with central vascular ligation (CVL) produces an oncologically superior specimen and appears to be related to optimal outcomes. We aimed to assess whether a regional educational programme in CME with CVL led to an improvement in the quality of colon cancer specimens. Following a regional educational programme in CME with CVL in the Capital and Zealand areas of Denmark, 686 cases of primary colon cancer resected across six hospitals were assessed by grading the plane of surgery and undertaking tissue morphometry. These were compared to 263 specimens resected prior to the educational programme. This work was partly supported by a PathSoc undergraduate bursary. Across the region, the mesocolic plane resection rate improved from 58% to 77% (p<0.0001). Hillerød hospital had implemented CME with CVL as standard prior to the educational programme and continued to produce optimal specimens. Three of the other hospitals showed a significant improvement in the plane of surgical resection. Hillerød specimens continued to be more radical with a greater distance between the tumour and the high tie, area of mesentery and lymph node yield compared to the other five hospitals. A multidisciplinary regional educational programme in CME with CVL has improved the oncological quality of colon cancer specimens as assessed by mesocolic planes, however, there has been no significant effect on the amount of tissue resected. Surgeons at Hillerød continue to produce more radical specimens suggesting that such educational programmes are not alone sufficient to increase the amount of tissue resected around the tumour.

Hillerød have recently published their long term outcomes with survivals being 10% higher when compared to other hospitals across the region. Further engagement is now necessary to ensure that optimal outcomes are achieved across the region.

Investigating the Faecal Microbiome in Formalin Fixed Paraffin Embedded (FFPE) Material P ITR Jobling 1 ; M Taylor 2 ; C Young 2 ; HM Wood 2 ; P Quirke 2 1 University of Leeds, Leeds, UK; 2 Leeds Institute of Cancer and Pathology, Leeds, UK Purpose: Research into the faecal microbiome has shown a diverse population with a high level of variability between individuals. Altered faecal microbiomes are present in a range of diseases but work remains to understand their role in gastrointestinal disease. Current research into the microbiome makes use of fresh or frozen faecal samples. This restricts researchers to predominantly prospective study designs. One potential method for rapidly increasing and diversifying research is the retrospective study of FFPE material. We aimed to investigate the feasibility of typing the microbiome in FFPE faecal samples using next generation sequencing (NGS) technology. Methods: Material from six faecal samples was divided and stored as frozen or fixed and paraffin embedded creating two matched sub-groups. To assess assay sensitivity one sample was diluted to eight different concentrations before fixing and embedding. The V4 and V6 regions of the 16s rRNA gene were amplified. Primer pairs created approximately 240bp and 98bp targets in E.coli respectively. PCR products were multiplexed and sequenced on an Illumina MiSeq. QIIME software was used for analysis. Results: Analysis of alpha (within sample) diversity showed a significant difference between sub-groups when targeting V4 (p=0.05) but not V6. Analysis of beta (between sample) diversity showed a significant difference between sub-groups when targeting V4 (p=0.01) while the V6 region showed a reduced, but still significant (p=0.02) difference. The sensitivity assay showed comparable results down to 0.5% concentration levels. Conclusion: To our knowledge this is the first feasibility study generating NGS data on the microbiome from FFPE faecal material. Variation between matched frozen and FFPE faecal material was less when targeting V6 compared to V4. We hypothesise this may be due to the shorter amplicon undergoing less DNA fragmentation in FFPE material. There are several platforms available for DNA mutation detection in formalin-fixed paraffin-embedded (FFPE) material, all with their relative strengths and weaknesses. We investigated the OncoScan® FFPE Assay Kit (OncoScan) in comparison to pyrosequencing in patients with operable colon cancer recruited to the phase II component of the NCRI FOxTROT trial of pre-operative vs. post-operative chemotherapy. FFPE samples of tumour from the resection specimens of 132 cases were tested for KRAS 12/13/61 and BRAF V600E mutations using pyrosequencing. The OncoScan assay allows for the interrogation of 74 mutations across nine genes. Pre-extracted DNA was analysed on the OncoScan assay and quality control (QC) scores generated, indicating confidence in mutation calling results. The mutational status of all samples was automatically assessed in the Affymetrix SM Viewer, and then manually confirmed. This work is presented on behalf of the FOxTROT Collaborative and was part funded by a PathSoc fellowship. Out of 132 samples, 22 failed OncoScan QC thresholds, however, only 10 of these were deemed inconclusive by manual interrogation. 130 samples were interpretable by pyrosequencing. Of the 120 samples that produced conclusive results on both platforms, the concordance rate was very high at 95.8% when calling a mutated versus non-mutated KRAS/BRAF status. Mutations were 'missed' by pyrosequencing in only 1 case (0.8%) and by OncoScan in 4 cases (3.4%). In addition, the OncoScan assay provides mutational data in additional genes along with copy number (CN) and loss of heterozygosity (LOH) information. In patients with colon cancer recruited to the NCRI FOxTROT trial, the OncoScan FFPE assay shows good correlation with pyrosequencing when determining the mutational status of KRAS/BRAF. Although pyrosequencing has a slightly lower failure rate, the OncoScan has the added advantage of targeting more mutations, producing genome wide CN, and LOH information in one assay. Excellent anatomical knowledge of the rectum and surrounding structures is essential for total mesorectal excision (TME). Denonvilliers' fascia (DVF) has been frequently studied, though the optimal anterior plane in TME is still disputed. The relationship of the lateral edge of DVF to the autonomic nerves is also unclear. We studied whole-mount microscopic sections of en-bloc cadaveric pelvic exenteration specimens and describe implications for TME. Four human adult cadaveric specimens (two males, two females) were obtained from the Leeds GIFT Research Tissue Programme. Paraffin-embedded mega-blocks were produced and serially sectioned at 50 and 250 µm intervals. Sections were stained with haematoxylin & eosin, Masson's trichrome and Millers' elastin. Additionally, a developmental series of eleven human fetal pelvic specimens (embryonic age of 9-20 weeks) were studied. DVF consisted of multiple fascial condensations of collagen and smooth muscle fibres and was indistinguishable from the anterior mesorectal fascia and the capsule of the prostate or posterior vaginal wall. The lateral edges of DVF appeared fan-shaped, and the most posterior part was continuous with the mesorectal fascia. Peri-rectal fasciae were not identified in fetal specimens. DVF is adherent to and continuous with the mesorectal fascia. Optimal surgical dissection during TME should be carried out anterior to DVF to ensure radical removal, particularly for anterior tumours. Autonomic nerves are at risk, but can be preserved by following the mesorectal fascia along the anterolateral mesorectum. The lack of evident fasciae in fetal specimens suggests that these might be formed in later developmental stages. The perineal body (PB) is poorly understood. In abdominoperineal excision (APE), there is no natural dissection plane through the PB. Knowledge of the PB is essential to avoid straying in to incorrect planes leading to tumour perforation and unnecessary urogenital and anorectal injuries. This study describes the anatomy of the PB and the implications for APE. Six human adult cadaveric specimens (three males, three females) were obtained from the Leeds GIFT Research Tissue Programme. Paraffin-embedded mega-blocks containing the PB were produced and serially sectioned at 50 and 250 µm intervals. Sections were stained to reveal collagen and elastin, and with an antibody against α-smooth muscle actin. The PB is formed of a fibromuscular mass, which was thicker and wider in female specimens compared to males, extending from the external anal sphincter to the rectogenital septum. Muscles from the urogenital diaphragm and anterior rectal wall anchored into the PB. The longitudinal muscle (LM) of the rectal muscularis propria extended in anterolateral directions and intertwined with the somatic pelvic floor muscles to create strong fixation of the anorectum. The LM plays a dominant role in the formation of the PB. Surgeons should be aware of the complex course of the LM through the PB to prevent injuries to the urogenital organs and perforation of the anterior rectal wall. The perineal phase of an APE starts with excellent exposure followed by proper tension on the PB to allow safe dissection through the densely-packed fibromuscular mass.

Introduction: Ki67 is a proliferation marker that is exclusively present in dividing cells and absent in resting cells. Its expression has already been studied in different cancers and used to understand the cellular organisation of Barrett's epithelium (Lavery 2014). However, very little is known about the cellular organisation based on Ki67 expression patterns in upper GI sequence. This study aims to examine the cellular organisation as defined by Ki67 expression patterns in upper GI cancer sequence. Methods: Ki67 expression within Barrett's crypts was assessed in 40 cases (NBDE 11, LGD 15, HGD 14) . The Barrett's crypts were divided into three equal regions: crypt base (bottom third), middle region and the surface (upper third), respectively. Ki67 was scored using the Allred system and analysed using one-way ANOVA with Bonferroni post-hoc analysis. Results: One-way ANOVA showed significant difference across the three groups (p < 0.0001). Bonferroni post-hoc analysis showed significant difference in the surface architecture between NBDE and HGD (p < 0.0001) and LGD and HGD (p < 0.0001). For the middle region, although there was no statistical significance between the groups, NBDE and LGD and

LGD and HGD showed statistical trends (p = 0.079 and p = 0.089 respectively). For the basal compartment there was significant difference between NBDE and LGD (p = 0.035).

This study showed for the first time a significant difference in the Ki67 expression between NBDE, LGD and HGD in the basal and surface regions. Middle compartments showed trends but additional NDBE, LGD and HGD groups need to be analysed to increase the statistical power. The results warrant further molecular analysis between the various groups and show a clear role for proliferation in the maintenance of the cellular architecture and organisation across the upper GI groups which might help in the understanding of the origin and development of OAC. • Frequency of serosal involvement in rectal cancers (suggested contributing factors for this include effect of pre-operative therapy, tumour regression and recent changes in surgical practice).

• Turnaround times: suggested contributing factors include increased departmental workload, retirements and reduced reporting capacity.

The following action plan was implemented to improve compliance with standards:

• Ensure all pathologists are aware of results via presentation/dissemination of audit report • Identify issues affecting turn-around times and improvement strategies • Support recruitment to increase reporting capacity.

• Maintain awareness of the need to recognise serosal involvement in rectal excisions.

• Re-audit in 1 year.

The design and maintenance of a pilot online digital archive of archetype colorectal polyps and gastrointestinal (GI) teaching cases for the National Bowl Screening programme (BowelScreen) in the Republic of Ireland. Methods: Suitable internal and referral cases were identified by BowelScreen consultants at Saint Vincent's University Hospital. These cases were subject to both internal and external review, by the Mater Misericordiae University Hospital, and represented typical examples of lesions seen in a National Bowel Cancer screening programmes (e.g adenomas, SSLs, adenomas with misplacement, TSAs). Representative slides, including immunohistochemistry, were anonymised and digitised using the Hamamatsu NanoZoomer Digital Pathology (NDP) whole slide scanner platform and associated software packages (NDP scan and view). Whole slide images (WSI) were uploaded to secure cloud storage using a generic file transfer protocol program. WSI were collated into 18 cases and accessible via the PathXL gateway (pathXL.co.uk) by approved users via an online case referral and reporting system. Users were notified of pending cases via email and the viewing of WSI occurred within the user's web browser utilising an online version of NDP view program and did not require local use of propriety software. Each case was referred across the two participating sites and scored in four areas; diagnosis concordance, quality of WSI, web interface and the online referral and reporting system. Conclusion: With the maturation of technology involved in digital microscopy a digital archive program is now a feasible approach to the standardisation of diagnosis and a useful adjunct to traditional optical microscopy in education within the National Bowel Screening programme. Conclusions: A committed and conscientious BMS can learn how to report histopathology cases. However, if this is to be achieved, the department in which he or she works must also be committed and supportive.

Carcinoid Tumour of the Appendix: A Case Report P AAE Shalaby; P AAE Shalaby

A case of a 24 years male operated on for acute appendicitis and an incidental finding of a carcinoid tumour at the tip is reported. The tumour was less than 2 cm in greatest dimension but it infiltrates through the wall of the appendix into the surrounding fat. It stains positive for the neuroendocrine markers. Carcinoid tumour of the appendix is unusual, but it has to be looked for during examination of appendectomy specimens done for appendicitis (0.5%). Women are more frequently affected than men (3:1) and the tumour is usually small less than 1 cm in diameter and frequently located at the tip. It is usually diagnosed incidentally after an operation for acute appendicitis and sometimes during other procedures (colectomy, cholecystectomy and others). The tumour rarely metastasis to the liver and this is usually related to the tumour diameter) and can cause a "carcinoid syndrome": flush, diarrhea bronchoconstriction, cardiac valve disease. Diagnosis is made by the pathologist and staging by conventional radiologic procedures (TAC, US), dosage of neuroendocrine mediators such as 24 hours urinary 5-HIAA. Simple appendectomy is adequate treatment for appendicular carcinoids less than 1 cm in diameter. Adequate treatment for tumours greater than 2 cm is right hemicolectomy. The mangement of tumours 1 to 2 cm range is controverisal, but generally, appendectomy alone is sufficient except when meso-appendix is invaded. Carcinoid tumour of the appendix has a good prognosis with a 5-year-survival rate, of 85-100%.

The Prevalence of Epithelial Changes in Helicobacter Pylori-Associated Gastritis in Oman: A Retrospective Study P AAE Shalaby; A Al Saadi

There is Strong association between H. pylori gastric infection and epithelial changes and progression to cancer. It has been shown that H pylori infection is strongly associated with high proliferative activity and it could be a risk of initial step of gastric carcinogenesis. The aim of this study was to examine the association between epithelial changes in the gastric mucosa and gastric H pylori infection in Oman by retrospective examination of the gastric biopsies for patients presented to Sultan Qaboos University hospital (SQUH) in 2013. A total of 697 biopsies were studied with a prevalence of H pylori infection in 34% with about 13% showing epithelial changes, mainly intestinal metaplasia in 10% out of the H pylori positive cases, a few cases with low grade dysplasia and reactive atypia. In cocnlusion intestinal metaplasia was the main epithelial change that was related to H pylori infection. Further studues are required to investigate the relation between H pylori infection and the progression to gastric carcinoma. Purpose of the study: Low rectal carcinoma may require abdomino-perineal excision of the rectum (APER), which has been associated with higher rates of tumour perforation and circumferential margin (CRM) involvement than anterior resection. This increases the risk of local recurrence and may necessitate adjuvant treatment. The Extralevator Abdominoperineal Excision of Rectum (eLAPE) in the prone position has been found to improve these outcomes and has been encouraged by the Low Rectal Cancer National Development Programme (LOREC). We aimed to assess the effect of increasing the practice of eLAPE on the histological and oncological outcomes in these cases in the Mid-Yorkshire NHS trust, a large District General Hospital.

In 2011 the number of surgeons routinely performing APER was reduced and all those performing the procedure had been trained in the cylindrical resection technique. Joint operating and laparoscopic procedures were encouraged. A retrospective review of case notes and histological reports between 2009 and 2012 was performed (before and after sub-specialisation). Patient demographics, histological findings and complications including local recurrence were recorded. Summary of Results: Between 2009 and 2011, 39 APERs were performed, with tumour perforation in 5 (13%) and CRM involvement in 8 (21%) of cases. After sub-specialisation, 28 were performed. None were perforated and 2 cases (7%) showed margin involvement. Local recurrence occurred in two cases before specialisation and none after 2011 at the time of follow-up. Joint operating and subspecialisation increased the number of cases performed by each surgeon, and the number performed laparoscopically. Conclusions: eLAPE in conjunction with departmental restructuring significantly improves immediate oncological outcomes in a DGH setting, with no effect on 30 day mortality. The technique may reduce local recurrence, although longer follow-up would be required. Background: Systems biology uses computational and simulation approaches to interrogate gene expression datasets and explore biological pathways. By employing systems biology and data mining tools we can identify new biomarkers. Our objective was to ascertain the utility of a novel panel of systems biology derived biomarkers in cervical pre-cancer for more accurate grading and stratification of CIN disease. Methods: This project is conducted within the framework of an FP7 funded programme "SYSTEMCERV". Gene pathways were analysed using MATLAB and SIRENE. Along with accessing KEGGS online database for gene prediction and DAVID for gene functional classification, we identified a novel panel of biomarkers. Gephi software was used to visualise communities of genes related to cervical pre-cancer and cancer progression. Clinical validation was performed by immunohistochemistry on a range of cervical LLETZ specimens (Normal, CIN1, CIN2 and CIN3). All patients gave written informed consent. In parallel, p16 IHC was performed on all specimens as a benchmark stain. The mortality associated with cervical cancer can be reduced if the disease is detected at the early stages of development or at the pre-malignant stage. The Pap smear is the current screening method, but is highly subjective and can often exhibit low specificity and sensitivity. For this reason, either a replacement or supportive technique is necessary to improve the quality of cervical cancer screening. Raman spectroscopy is a powerful tool that can generate a biochemical fingerprint of a sample in a rapid and non-destructive manner. In this study, Raman spectroscopy has been applied to the investigation of cervical cells from PreservCyt specimens. Raman measurements were taken from the nuclei of cervical cells from normal, CIN1, and CIN3 samples. These spectra were processed, analysed and used to define a spectral signature for each grade of cervical disease. Principal Component Analysis (PCA) was used to discriminate between the two data sets. Distinct Raman spectral differences were detected between normal, CIN1 and CIN3 cells. Notably, it was possible to observe spectral peak shifts representing fluctuations in Guanine (DNA/RNA), CH deformation in proteins and carbohydrates, Carbon-carbon double bonds in Phenylalanine, Tyrosine and Tryptophan, and Amide I. The PCA showed an excellent discrimination between the data sets. This study has shown that Raman spectroscopy can detect subtle changes between cervical cells, and may be a powerful tool for improved diagnosis of cervical dysplasia. Background: MyD88 and MAD2 are two potential prognostic biomarkers that have been investigated in ovarian cancer. High MyD88 and Low MAD2 IHC staining is associated with reduced PFS, both markers are also linked to paclitaxel chemoresistance.

The main objective of this study was to assess the in vitro relationship between MAD2 and MyD88, through alteration of MAD2, MyD88 or its receptor TLR4 in two ovarian cancer cell lines using siRNA targeting MAD2, TLR4 or MyD88 and a MyD88 overexpression plasmid vector. Following overexpression/siRNA knockdown procedures, MyD88, TLR4 and MAD2 expression was assessed through qPCR and Western Blot analysis. Mir-433, Mir-21 and Mir-146a gene expression was also assessed by qPCR. Furthermore the effect of TLR4/ MyD88 knockdown on chemoresponse was assessed in SKOV-3 cells using the CCK-8 assay. Results/Discussion: It was found that knockdown or overexpression of MyD88 in SKOV-3 or A2780 cells respectively or knockdown of TLR4 in SKOV-3 cells had no effect on MAD2 expression or the expression of Mir-21, Mir-433 and Mir-146a. Interestingly however knockdown of MAD2 in both cell lines induced a 3 fold increase in TLR4 expression, furthermore knockdown of TLR4 in SKOV-3 cells was shown to restore chemosensitivity to paclitaxel.

The results demonstrate a potential in vitro link between TLR4 and MAD2 and support a role for TLR4 in paclitaxel chemoresistance. Background: The prognosis of epithelial ovarian cancer is poor in part due to the high frequency of chemoresistance. Recent evidence points to the Toll-like receptor-4 (TLR4), and particularly its adaptor protein MyD88, as one potential mediator of this resistance. Downregulation of MAD2, a key component of the spindle assembly checkpoint complex, has also been linked with paclitaxel resistance . Both markers have individually been shown to be associated with poor outcome in ovarian cancer. High MyD88 and low MAD2 immunohistochemical staining is associated with reduced progression free survival. The main objective of this study was to assess the combined utility of MAD2 and MyD88 in predicting patient prognosis. Methods: Two tissue microarrays composed of cores from 51 high grade serous epithelial ovarian cancers patients were constructed and stained for MAD2 and MyD88. Staining was scored based on previously derived scoring schemes for MyD88 or MAD2 staining. The mean overall score from triplicate cores was then used to classify patients into high and low staining categories. Results: A trend towards reduced progression free and overall survival was observed in patients with high myd88 and low mad2 expression.

The results demonstrate the combined utility of MAD2 and MyD88 as predictors of prognosis in ovarian cancer. Purpose: Ovarian cancer is characterised by high rates of terminal, chemoresistant recurrence. Although chemoresistance is known to be a property of Cancer Stem Cells (CSCs), the mechanism is poorly understood. We have previously identified a novel four-member CSC Stem-Progenitor cell hierarchy for ovarian cancer. The aim of this study was to characterise the contribution of each member of the ovarian CSC hierarchy to chemoresistance. Methods: The CSC hierarchy was assessed for tolerance to chemotherapy drug cisplatin (MTT assay) both as components of the parent population (A2780 cell line) and as isolated cell types. The hierarchy was additionally assessed in the long-term cisplatin-adapted 'A2780cis' cell line. Cell types were analysed and isolated via flow cytometry and assessed for stem cell characteristic via single-cell asymmetric division and murine xenograft tumourigenicity assays, and molecularly characterised (whole transcriptome arrays). Results: Cisplain dose-response assays from A2780-derived CSC sub-populations indicated that only one of the four populations within the hierarchy had a high cisplatin-tolerance (IC50=10uM) compared to the other populations (IC50=4uM). This was notable as the relative cisplatin IC50s for the A2780 and A2780cis parent cell lines are 4uM and 11uM respectively. Treatment of the parent A2780 cell line with the IC50 (48 hours) resulted in a proportional 50% loss in each of the four cell types, suggesting that this specific CSC subpopulation adapts to cisplatin over a longer period of time. Conclusion: Although CSCs are known to be chemoresistant, the mechanism though which this is achieved is poorly understood. Our data indicates that only some members of a CSC hierarchy are responsible for chemoresistance. Notably, this sub-population appears to possess inherent chemoresistance in pre-treatmentcells. As such, it should be possible to target these CSCs in the primary malignancy to prevent chemoresistant recurrence. Mixed sex cord-stromal tumours of the ovary are very rare. We report a case of mixed sex cord -stromal tumour (also referred to as gynandroblastoma) containing both Sertoli-Leydig cell tumour and adult granulosa cell tumour in a female 61 years old who presented with postmenopausal bleeding. On histology, the majority of the tumour represented an unsual form of well differentiated Sertoli-Leydig cell tumour with a pseudoendometrioid appearance. Minor foci of classic adult granulosa cell tumour were present. On immunohistochemistry, the tumour was diffusely positive for inhibin and SF1 and focally for calretinin, ER and CD56. EMA, PAX8 and CK7 were negative. As far as we are aware, this is the first report of an ovarian mixed sex cord-stromal tumour containing a component of pseudoendometrioid Sertoli-Leydig cell tumour. A 37 year old female patient presented with bilateral painful warty lesions on the labia majora. The patient had had HIV for a long time and was on highly active antiretroviral therapy. She also suffered chronic renal failure requiring haemodialysis three times weekly. Clinically, the lesions were highly suspicious of vulval cancer. The lesions increased significantly in size over a short period of time (2 months) requiring surgical resection under general anaesthesia. Histological examinations revealed polypoid lesions with prominent pseudoepitheliomatous hyperplasia and dense inflammatory infiltrate, composed mainly of lymphocytes and plasma cells, extending to the hypodermis. Numerous abscesses with large numbers of eosinophils were present withinin the hyperplastic epithelium. The typical intranuclear inclusions of herpes simplex virus (HSV) were identified. HSV immunohistochemistry was positive. This is a rare case of vulval HSV warts mimicking cancer. Oral acyclovir was administered following surgery and resulted in good control. Literature review shows only 6 previously described cases of verrucous HSV, Types 1 and 2, simulating neoplasia in patients with AIDS on antiretroviral therapy. Primary mucinous eccrine adenocarcinoma of the skin is a rare adnexal neoplasm, typically involving the head and neck region in the elderly population. Here we present a case of primary mucinous eccrine adenocarcinoma of the vulva; occurrence at this site is extremely rare, with only five cases published in English literature. A 62 year old female presented with a 10mm vulval lesion, clinically suspected to be an inclusion cyst. The lesion was removed and sent for histopathological assessment. Histological examination revealed a well circumscribed, partly encapsulated tumour composed of rounded and irregular nests of polygonal epithelial cells with scattered lumina, suspended in pools of extracellular mucin. The epithelial cell nuclei displayed a uniform chromatin pattern with small distinct nucleoli. The mucin pools stained positive for Alcian blue and dPAS. Immunohistochemical staining demonstrated positivity for CEA, CK7, GCDFP, oestrogen receptor, progesterone receptor, synaptophysin and chromogranin. Immunostaining was negative for CK20, CDX2, CA-125, TTF-1, CH5/6, CK14, HER-2, WT-1, CD56 and S100. Ki-67 proliferation fraction was approximately 5%. Overall, the findings were those of a mucinous eccrine adenocarcinoma with neuroendocrine differentiation. Following multidisciplinary discussion, and negative imaging of the breasts and gastrointestinal tract, a diagnosis of primary mucinous eccrine adenocarcinoma of the vulva was reached. Only a handful of cases of primary mucinous eccrine adenocarcinoma of the vulva have been reported. Metastatic disease, particularly from breast and colon, must be excluded. Follow up data from patients with primary mucinous eccrine adenocarcinoma of the skin suggests a high local recurrence rate (29.4%), necessitating close follow-up. However, risk of metastasis is low (9.6%).

Royal Shrewsbury Hospital, Shrewsbury, UK Introduction: Primary Extraskeletal Myxoid Chondrosaroma (EMC) of the vulva is a rare mesenchymal neoplasm. The myxoid tumour differential diagnosis on a core biopsy can be quite challinging. To date, few cases have been reported in the literature. Case Report: A 42-year old woman noticed a swelling on the right side of the labia, thought to be a Bartholin's cyst in 2011. She was managed conservatively. She had drainage and marsupialization under general anaesthesia. This resulted in extreme bruising of the vulva. This was managed with antibiotics and non-steroidal anti-inflammatory medication, and it resolved after 3 weeks. Six months later, the patient presented again with a persistent vulval mass. A biopsy was obtained under general anaesthesia, and it showed a myxoid tumour with differential diagnosis of low grade chondroid tumour. An MRI was performed to assess the extent of the disease. The tumour was excised. At surgery, a 7 x 5 cm lobulated, extremely vascular vulval tumour was found. The tumour was inseparable from the inferior pubic ramus of the pelvic bone. A complete macroscopic resection was obtained. Histology confirmed low grade myxoid chondrosarcoma. Conculsion: Vulval lesions with unusual characteristics or insidious evolution in the labia majora or Bartholin's glands area should be carefully and promptly investigated. Differential diagnosis of myxoid tumours in the vulva should include myxoid chondrosarcoma amongst other diagnoses. (FISH) showed the presence of a BCL2 rearrangement in a proportion of cells. Therefore this case is best regarded as a composite lymphoma of diffuse large B cell lymphoma with hairy cell leukemia rather than blastic transformation of hairy cell leukemia. To the best of our knowledge, simultaneous occurrence of diffuse large B cell lymphoma and hairy cell leukemia in a lymph node has not yet been reported in the literature.

Bone marrow examination by aspirate and trephine biopsy is an important haematological investigation. Ideally, aspirate findings should inform examination of the trephine biopsy, but if the two modalities are separate the aspirate report can be delayed and histopathologists may assess the trephine biopsy without being aware of the aspirate findings. We audited the availability of aspirate results to the histopathologist examining trephine biopsies, over the period in which our department implemented an integrated haematopathology reporting system. The effects on diagnostic concordance, turnaround times and immunohistochemistry requesting were also assessed. The setting was a regional specialist haematopathology centre. Prior to integration, a prospective audit of 101 consecutive trephine biopsies received by a senior haematopathologist was carried out, against standards set by the International Committee for Standardisation in Haematology. Data were collected from hospital computer systems. The move to integrated reporting involved the installation of new software (HiLIS) to specifically handle integrated haematopathology data. Ten months later, a retrospective analysis of a further 100 cases was performed using HiLIS data. Prior to integration, 35% of aspirates were reported within 3 days, and access to the aspirate report was available at time of examination for 61% of trephine biopsies. After integration, 96% of aspirates were reported within 3 days, and reports were available at time of examination for 100% of trephine biopsies. Diagnostic concordance was 72% initially, and 100% after integration. The mean number of immunostains requested per case was unchanged (5.9 vs 5.7). Our findings show integrated reporting has markedly increased the availability of aspirate reports to the histopathologist, and improved diagnostic concordance. This new model benefits the haematologist, histopathologist and patient.

Introduction: Clonality studies are carried out when the diagnosis of lymphoma is particularly challenging. The detection of clonality in lymphoproliferative lesions suspicious for lymphoma can be a valuable supplementary tool as it has a high positive predictive value. Clonal studies can also help to distinguish recurrent or residual disease from reactive inflammation. However false positive and negatives are common and can be attributable to several factors, including poor DNA quality. Methods: 206 cases reported over a six month period (Jun-Nov 2014) were retrospectively reviewed, 62% of which were referral cases. We investigated various aspects of clonality studies; including DNA quality, fixation method, clinical information provided and correlation between the morphological/immunophenotypical findings and clonality results using Euroclonality/BIOMED-2 primers (IgH, IgK, TCR-Β and gamma-delta). Results: 166 (81%) had adequate DNA quality, 29 (14%) poor DNA and 11 (5%) had inadequate DNA quality. External cases had better DNA quality in the majority of cases. Clinical information was provided in 75% of local cases and 53% of external cases. In 56% of cases clonality results supported the initial histological report. 9 cases showed clonal expansion despite a benign process on histology. 21 suspected cases lymphoma (13 B-NHL and 8 T-NHL) showed no clonality, 8 of which yielded poor DNA quality. Skin cases although had good DNA quality, usually had low number of neoplastic cells resulting in poor PCR products.

Conclusions: DNA quality is very variable and clinical information is often not provided precluding adequate assessment of clonality findings. DNA was worse locally (decalcified marrow trephines using formic acid and Peloris system with high temperatures that can cause DNA degradation). Standardisation of fixation methods and interpretation of peaks/ bands in the clinical context of the patient is essential for clonality to be informative.

Kikuchi-Fujimoto Disease: A Novel Diagnosis by Transbronchial Biopsy of Mediastinal Lymphadenopathy P PM Ellery; N Archard; A Ramsay UCL Hospitals NHS Foundation Trust, London, UK Objectives: Kikuchi-Fujimoto disease (KFD) is a rare, self-limiting form of necrotising lymphadenitis that most commonly affects young Asian women, and classically presents with fever, malaise and lymphadenopathy. The cervical lymph nodes are involved in around 85% of cases, with other sites rarely involved. We report an unusual case in which an unexpected diagnosis of KFD was made via transbronchial biopsy of mediastinal lymph nodes.

A 15 year old boy of Pakistani origin presented with a 4 month history of lethargy, neck stiffness and weight loss, with fever (up to 40°C) and night sweats. Chest X-ray, Mantoux test, blood cultures and viral PCR were negative. Lumbar puncture was normal, with no acid fast-bacilli. CT showed enlargement of the deep cervical lymph nodes (PETpositive on further imaging), and mediastinal lymphadenopathy. He was transferred to our hospital for further management, with a differential diagnosis of TB, lymphoma, rare infection or autoimmune disease. He underwent transbronchial biopsy of the mediastinal lymph nodes. Results: His biopsy showed blood clot and cores of lymph node, with focal collections of crescentic macrophages, admixed lymphocytes and prominent apoptotic debris. Immunohistochemistry demonstrated a population of CD123-positive plasmacytoid dendritic cells and granular MPO positivity in macrophage cytoplasm. The background lymphocytes were mainly CD8-positive T-cells. The features were those of KFD. Conclusion: Involvement of deep lymph nodes is unusual in KFD, and to our knowledge, this is the first case diagnosed via transbronchial biopsy. Such biopsies often produce scanty diagnostic material, and here the detection of the characteristic immunoprofile of KFD helped confirm the diagnosis. This case highlights that KFD should be considered at sites other than the cervical lymph nodes, and demonstrates the value of immunohistochemistry in reaching a definitive diagnosis. We describe a benign intravascular proliferation of atypical polytypic CD30 positive T cells, co-expressing follicular T helper cell lineage markers coincidental to local sepsis of the buttock. A 43 year old female presented with a 5x5cm buttock abscess at the site of a longstanding palpable lump. Peripheral blood showed only a neutrophilic leucocytosis. Immunohistological examination showed large aggregates of atypical CD30 positive, ALK negative lymphoid cells expressing a pan T helper phenotype with CD7 partially downregulated. Podoplanin proved that the atypical T cells were primarily, but not exclusively, localised within lymphatic channels. A T cell receptor clone was not detected using PCR. To find intravascular concentrations of atypical lymphoid cells is uncommon in skin biopsies and raises the possibility of leukaemia or intravascular lymphoma. Intravascular lymphoma is a rare variant of non Hodgkin Lymphoma with a minority possessing T or NK cell lineage but frequently involving skin. CD30 is a transmembrane glycoprotein and a member of the TNF superfamily involved in regulating proliferation. It is considered a reliable marker of lymphoma. Primary cutaneous CD30 positive TLPDs encompass a spectrum of biological aggressiveness and include primary cutaneous anaplastic large-cell lymphoma and lymphomatoid papulosis (LyP). CD30 can also be up regulated in activated B and T cells and it has been proposed that CD30 positive ivTLPDs are equivalent to an intravascular form of LyP. Intravascular proliferations of atypical CD30 positive T cells have been linked with chronic inflammation and abscess formation. Furthermore atypical CD30 positive TLPD expressing a CD4 positive T helper phenotype and exhibiting an indolent clinical course have been reported in the arm, trunk, neck and prepuce. Ultimately ivTLPD may require follow up based upon clinical features and natural progression due to overlapping diagnostic features.

P G Laing; S Craig; L Moss; C Crichton; P Johnston

The investigation of lymphoid neoplasia requires multiple sections, in our practice consisting of twelve antibodies and thirteen single stained slides. By selecting particular antibodies for double staining, spatial relationships between cell types and overall tissue organisation can be more easily visualised. This pilot study aimed to optimise the staining intensity and specificity to provide accurate diagnostic information, reduce slide number, material and consumables costs, preparation time in the laboratory and storage space to enhance costbenefit. Twelve antibodies were chosen and paired: kappa/lambda, CD3/CD20, MUM-1/CD10, Cyclin D1/CD30, BCL-6/CD5 and CD5/PAX-5. Firstly, these combinations were applied to normal tissue and then known tumours to optimise technique. Once the staining protocols were finalised they were run on eleven consecutive cases with conventionally stained Non-Hodgkin Lymphoma (NHL) panel requests. The slides were then reviewed by the Lymphoma team for quality and diagnostic accuracy compared to the standard single stained slides. The results demonstrate that double staining is possible in the diagnosis of NHL. The combinations chosen have proved successful and have provided interpretable results; for example, the relationship of light chain staining in plasma cells, MUM-1 and CD10 positive cells in Diffuse Large B-cell Lymphoma proves positive. We feel time will be saved cutting sections to improve efficiency in Lymphoma investigation and reduce panel storage space by around 30%. In conclusion the outcomes from this pilot study have been positive for medical and scientific staff, has shown that double staining in the diagnosis of NHL is possible and that optimising this protocol with a view to live diagnosis is worthwhile.

In an Age in which Novel therapies are not necessarily defined by their ability to kill malignant cells, understanding the biology of malignant cells after treatment is extremely important. This is particularly true of Ibrutinib (Pcl-32765) therapy in chronic lymphocytic leukaemia which is characterised by lymphocytosis. Imagej/Fiji image analysis software could therefore be used to analyse cell shape and grouping characteristics. We used a novel assay in which Chronic Lymphocytic Leukaemia cells, cultured for 5 days, were seeded onto fibronectin coated glass coverslips and then had their B cell receptors ligated with goat anti human Igm. They were compared with cells simultaneously inhibited with Ibrutinib. We tested multiple staining techniques and found that using either Rose Bengal or Texas Red Phalloidin staining produced the most reproducibly analysable data when using Imagej/ Fiji. We demonstrated, using the assumption that the outline of interacting cells would be larger than cells which were alone that Homotypic interactions were increased after B Cell receptor cross linking (54 groups larger than 2 cells against 35 group larger than 2 cells per 500nm X 250nm Field (P=0.019)). Nuclei were also significantly larger when cross linked. (Mean 33.23 (+-0.43) pixels without cross linking and 44.15(+-0.57) (P=≤0.0001)) suggesting increased nuclear spreading. These effects were reversed by the addition Of Ibrutinib with 18 groups larger than 2 cells per the same field(P=0.0252 compared with cross linked sells) and Nuclear size mean being 33.83 (P≤0.0001 compared with cross linked cells). This study demonstrates a novel, easily reproducible assay to assess a variety of cellular responses to Ibrutinib therapy and suggests a method of quantifying activity both when stimulated and inhibited. This technique could easily be scaled up to further investigate cellular behaviour following Ibrutinib therapy.

Purpose: Gliomas represent 43% of all solid intracranial tumours and are associated with a poor prognosis. Recent studies indicated that the human cytosolic branched chain aminotransferase protein (hBCATc), which metabolises the branched chain amino acids (BCAA), was significantly upregulated in IDH1/2 wild type (WT) glioblastomas, correlated with methylation patterns in the BCAT1 promoter and is associated with a worse prognosis compared with IDH mutant gliomas. The diagnostic and prognostic significance of markers of BCAA metabolism is currently under investigation. Methods: 64 glioma tumour samples were compared for hBCATc, hBCATm and BCKDC expression using western blotting and immunohistochemistry. DNA was extracted from fresh frozen tissue. Sanger sequencing of the p.Arg132 region of IDH1 and p.Arg172 region of IDH2 was undertaken using a 3730 DNA analyser (Applied Bio-Systems). SUMMARY: In IDH WT tumours, like hBCATc (p=0.007), the expression of the mitochondrial isoform (hBCATm) is significantly (p=0.036) expressed relative to IDH mutant gliomas. hBCATm additionally shows a more significant correlation with patient survival than hBCATc on Kaplan-Meier analysis. In IDH WT tumours, low hBCATm expression is a positive prognostic factor (p = 0.003). hBCATm expression additionally correlated with WHO grade. Although previous reports indicate that increased hBCATc occurs exclusively in IDH-WT tumours, our studies demonstrate that 30% of IDH mutant tumours express comparable levels of hBCATc. Although hBCATc alone has been suggested as a putative therapeutic target, it is important to evaluate the expression of hBCATm in glioblastomas as its expression may impact the efficacy of new treatments targeting hBCATc. Conclusions: IDH WT high grade gliomas traditionally have a poor prognosis. However we demonstrate for the first time that relatively low hBCATm may select for a better performing clinical cohort and may be a possible candidate target for drug therapy.

A 38 year old female presented with an occipital mass, presumed to be a lymph node and underwent fine needle aspiration of the lesion. FNA yielded two air dried slides, upon which a diagnosis of mesenchymal neoplasm was made. The patient underwent a subsequent incisional biopsy allowing a formal histological diagnosis of myxoinflammatory fibroblastic sarcoma to be made. Myxoinflammatory fibroblastic sarcoma is a low-grade neoplasm usually occuring on the distal extremities and only rarely presents as a head and neck neoplasm. FNA is a useful tool in the diagnosis and subsequent management of head and neck neoplasia and we describe here the cytological features and subsequent histological diagnosis of myxoinflammatory fibroblastic sarcoma occuring in the occipital scalp. Objective: Presentation of giant cell fibroblastoma (GCF). Because of it has a dilemma of microscopic appearances; it is mandatory to differentiating it from atypical dermatofibroma (ADF), fibrous hamartoma of infancy (FHI) and vascular lesions. Methods: Eighteen month-male Egyptian child presented with painless slowly expanding subcutaneous back swelling at the left scapular area. Results: Histologically, the lesion is poorly circumscribed and range from cellular to myxoid in a dense to loose collagenous stroma. The tumours composed of mixture of spindle shaped or stellate cells admixed with multinucleated giant cells with occasional pleomorphism and very low mitotic index (<1 per 50 high-power fields). These cells infiltrate around adnexal structures and through subcutaneous fat. A distinctive finding is cracking artifact of the stroma simulating angiectoid spaces. These pseudovascular spaces are lacking a true endothelial lining and lined by discontinuous layer of enlarged multinucleated giant cells. Immunostains, including factor VIII, CD31, CD1a, SMA, S100 and CD68 were negative. Ki 67 labeling index is very low. All the cellular components show positive immunoreactivity for CD34. Conclusions: First case of GCF reported in Egypt. We recommend a wide scaled study to categorize this tumour with molecularly similar lesions.

The Royal County Sussex Hospital, Brighton, UK

A 60 year old man being investigated for obstructive hydronephrosis was incidentally found to have a 3.2cm splenic mass on computed tomography (CT). No lymphadenopathy was present and the mass remained stable on sequential CT and ultrasound scans. On positron emission tomography (PET) the lesion had a low signal with moderate uptake. All haematological investigations were within normal limits including a negative Epstein Barr Virus (EBV) test. His past medical history included previous immunosuppressive therapy for inflammatory bowel disease. A core biopsy under CT guidance was performed. The cores showed a paucicellular spindle cell lesion with bland, blunt ended nuclei, no cytological atypia and a sparse chronic inflammatory infiltrate. There was no necrosis. The spindle cells stained positive for smooth muscle actin (SMA) and H-Caldesmon indicating this to be a splenic leiomyoma. Splenic lesions are uncommon and within their differential include, lymphoma, inflammatory pseudotumour, harmatomas and leiomyomas. (1) A splenic leiomyoma is an unusual and rare benign smooth muscle tumour with an unknown pathogenesis. They are thought to arise from the capsule and blood vessel walls of organs.

(2) They have been documented in immunosuppressed states (constitutional or acquired), in those with EBV infection and in children with ataxia-telangiectasia.

(2) Leiomyomas within the spleen are rarely reported in the literature, especially in those patients over the age of eighteen. In this case there was historical immunosuppression, however leiomyomas should be considered in the differential diagnosis of well-defined solitary splenic lesions Purpose of the Study: The Zucker Diabetic Fatty (ZDF) rat is extensively used as a model of Diabetic Kidney Disease (DKD) associated with obesity and progressive insulin resistance ('diabesity'). This study aimed to validate qualitative ultrastructural parameters of glomerular injury in the ZDF animal model and apply these criteria to an interventional study investigating the effects of Roux-en-Y gastric bypass (RYGB) on DKD. Methods: Superficial renal cortices were immersion-fixed in 2.5% glutaraldehyde, post-fixed in 1% osmium tetroxide, processed and embedded in epoxy resin prior viewing under a Technai 12 transmission electron microscope. Glomerular basement membrane (GBM) thickness, podocyte foot process diameter (PFPD) and podocyte foot process frequency (PFPF) per unit length of GBM were determined for each group (Sham and RYGB operated ZDF fa/fa diabetic animals vs non-operated non-diabetic ZDF fa/+ lean controls). Statistical analysis was performed using a Mann Whitney U test and an unpaired t-test where appropriate. Summary of Results: Selected TEM parameters (GBM thickness, PFPD and PFPF) demonstrated significant differences between specified Sham-operated ZDF fa/fa vs fa/+ samples, p=0.017. Analysis of RYGB interventional study samples still in progress. Early post-operative glucose measurements showed a significant improvement in glucose homeostasis in the RYGB group (RYGB vs SHAM, P=0.0001) occurring independently of weight loss. Urinary albumin:creatinine ratios were lower in the RYGB group vs Sham operated positive controls (P=0.0079) and were comparable with age-matched lean control fa/+ samples. Conclusions: Preliminary findings support a beneficial role for RYGB in an animal model of 'Diabesity'. Validated ultrastructural parameters should assist in elucidating changes in podocyte activation and differentiation as mediators of the observed remission of albuminuria following RYGB surgery.

Nottingham University Hospital, Nottingham, UK Spitz naevus is a benign melanocytic lesion that shares many histological features with malignant melanoma. Although the morphological criteria differentiating the two entities are well established however, some cases can be challenging. Many isolated markers have been proposed to help in differentiating Spitz naevus from melanoma, albeit none has been shown to be definitive. Aim: This is a preliminary study looking at the immunohistochemical expression of 3 markers that are known to have important role in cell cycle regulation, proliferation and melanocytic differentiation (P16, Ki67, and HMB45). The aim is provide to a combination of proteins that can help in differentiating Spitz naevus from malignant melanoma.

The study included 12 cases of Spitz naevi, 6 benign compound naevi and 6 cases of malignant melanoma. Immunohistochemical expression of p16, Ki67, and HMB45 has been accessed and compared with the morphological features of these lesions. Results: It is noted the mean P16 expression is higher in compound and spitz naevi than melanoma (83, 91, and 36 respectively). Proliferation activity as measured by KI67 index is higher in melanoma in comparison with compound and spitz naevi (30.8, 2.8, and 1.6 respectively). HMB45 shows only junctional positivity in 9 out of 11 cases of Spitz naevi while in the other two it shows week dermal component. HMB45 is constantly positive at the deep dermal component of melanoma, albeit the staining intensity is variable.

The immunoprofile of Spitz naevus is different from that of a malignant melanoma. A combination of biological markers as (p16, Ki67, and HMB45), can provide a potential tool to differentiate between the two entities. Nevertheless, expanding the biomarker repertoire on a large number of cases is necessary to further establish a reliable panel to differentiate among difficult cases.

Direct Immunofluorescence in a Tertiary Referral Centre: An Audit of Local Guidelines and Usage P LJ Lumsden; L Motta; R Green Salford Royal Hospital, Salford, UK Direct immunofluorescence (DIF) forms an important and costly adjunct to conventional haematoxylin and eosin (H&E) histology in dermatopathology, particularly in bullous diseases and other immune-mediated diseases. We aim to assess the usage and diagnostic yield of DIF in our dermatopathology department. 134 requests for DIF on skin biopsies received over a 5 month period met the inclusion criteria. Each individual report was assessed with regard to the indication for DIF, whether DIF was deemed to be indicated or not indicated on assessment of the clinical history supplied on the request card, the results of DIF and whether DIF was contributory to the final diagnosis. We also collected data on the usage of DIF over the last 2 years to assess changes in practice. All 134 requests for DIF were granted in line with current departmental policy. The indication categories were divided as follows: bullous 41, alopecia 4, lupus 32, vasculitis 23, dermatitis herpetiformis (DH) 15 and 'other' 19. All requests for DIF were deemed to be indicated in both the bullous and DH categories by our panel, but indicated requests varied from 8.7% to 50% in the remaining categories. In 45.5% (61 out of 134) of cases DIF was deemed to be contributory to the final diagnosis. Our analysis also showed that usage of DIF in our department is escalating, with a 23.2% increase in requests from 2013 to 2014. Our departmental policy with regard to DIF is inclusive and operates solely on the basis of clinician request. With the increasing usage of DIF, established departmental guidelines and/or a protocol for DIF usage should be mutually agreed with dermatology colleagues in order to ensure effective use of this expensive test.

Overview of Merkel Cell Carcinoma in an Irish Population P A Cooper 1 ; J Thorne 2 1 Royal College of Surgeons in Ireland, Dublin, Ireland; 2 Beaumont Hospital, Dublin, Ireland Purpose of Study: Merkel cell carcinoma (MCC) is an uncommon but highly aggressive primary cutaneous malignancy of neuroendocrine cells with a propensity for regional and distal metastases. Due to it's rarity, information relating to it's epidemiology in an Irish population is limited, mainly owing to difficulty in gathering large patient series. Our aim was to identify all cases of MCC in our institution in a defined 10 year period and review the patient demographics compared to internationally available data.

A search was carried out on the hospital laboratory system to identify all cases of MCC from 01/01/2005 to 21/12/2014. All histology reports were reviewed and any information pertaining to patient demographics was recorded in an excel spreadsheet. A literature review was performed relating to the patient profile of MCC internationally and the results were compared. Results: A total of 33 reports pertaining to 25 individual patients were recovered. All patients were of Caucasian Irish ethnicity. The incidence of MCC was higher in men (56% of cases, n=14) than women. The median age at diagnosis was 80 years (range 57-90). Men presented at an earlier age (median 78 years) than women (median 83 years). Regarding the anatomic site of the tumours, 64% (n=16) were on the head or face, 20% (n=5) were on the lower limb and 12% (n=3) were on the upper limb. All were on sun-exposed sites. Of note, the majority of tumours in the male population were on the head (78.6%, n=11), while the female population showed an equal distribution between the head and the lower limbs (45%, n=5 for each sub-site).

The subset of patients we identified show demographics consistent with published literature for US, Australian and other European cohorts. Merkel Cell Carcinoma is a disease of the elderly affecting sun-exposed sites. We note some variation in the dominant anatomic sites between genders and conclude this is due to differing environmental exposure. Background: A novel cell-dispensing instrument referred to as a Single Cell Manipulator (SCM) device was developed with the following features: i) rapid optical and fluorescent detection of single cells ii) generation of picoliter sized droplets encapsulating the isolated single cell and iii) printing of the single cell in an "ink-jet" like manner onto a chosen substrate. This technology was used to isolate cells of interest from i) heterogeneous mixed populations of cells, ii) co-cultures of cells and iii) clinical patient samples for subsequent downstream biological analysis. Methods: Cells were injected into a reusable silicon dispenser chip that was coupled to a live cell camera for image capture and display of cells approaching the chip's exit nozzle. An optical detection mechanism determined the presence of single, fluorescent cells within the selected region of interest close to the chip exit nozzle. A sorting algorithm ensured that only droplets containing the single cells of interest were selected for printing to the prescribed location and user-chosen substrate. Results: Fluorescently labelled HPV16 CaSki cervical cells were spiked into a cervical liquid based cytology sample and printed onto a glass slide using the SCM. Undifferentiated NTera2 human embryonal cancer stem cells were isolated from a mixture of differentiated and undifferentiated cells based on fluorescent tagging of the cell surface receptor, stagespecific embryonal antigen 4 (SSEA4). The thyroid stimulating hormone receptor (TSHR) was expressed in anaplastic V600E mutated thyroid cancer cell lines that were treated with the MEK inhibitor PD0325901. Treated cells were isolated using the SCM.

The SCM PASCA technology allows isolation of single cells from heterogeneous populations of cells and clinical samples for downstream analysis at a single cell level. This study aims at quantifying immunohistochemistry (IHC) stained human cell lines for protein biomarkers by manual pathologist review. Staining analyses are used to calibrate tissue microarrays of tumour cores, against quantitative protein concentration allowing a systems-based data analysis. As a proof of concept, FFPE human cell line pellets (n=13) were IHC stained for Smac protein and analysed using Aperio image analysis software. Staining quantification manually performed by pathologists provided parameters including average staining intensity, percent total cell positivity and H-score. These data were enriched by qualitative parameters pointing out possible histological artefacts. A calibration curve was plotted using H-score data and protein concentrations, previously determined by Western blotting. The panel of cell lines provided a range of strong and weak/absent IHC staining using a highly specific Smac antibody. The calibration curve showed a strong correlation between absolute protein concentrations and manual H-scores. Expression amounts in cell lines correlate with IHC staining intensities determined by pathologist review. The linear correlation between manual H-score and absolute protein values provides an avenue to indirectly determine protein expression. Further analysis will be performed on additional antigens and analysis outcomes will be then compared to digital results. This data will provide the basis for deterministic systems-biological data analysis approaches.

Purpose of the Study: 'Lean' is a management framework for maximising value and minimising waste. It originated in the automotive manufacturing industry and has been utilised successfully in non-manufacturing processes. One such application in our department was the 'leaning' of the molecular test requesting process using a smart-phone app. This study will look at the potential utility of this application within the National Health Service (NHS), wherein approximately twenty different molecular test request forms are currently in use.

A mobile application to facilitate molecular test requesting was developed using Xcode and the objective C programming language. The application was built around an email based system. Patient anonymity was paramount in the design; NHS numbers are used as identifiers. The application generates a molecular test request form and can also generate a national cancer drugs fund application form for each request. Administrative staff use colour coded flags to represent the progress of each email request through the workflow process to facilitate tracking.

The app reduced the administrative staff workload by reducing the number of steps and paperwork involved in the molecular test requesting process. A threefold reduction in time taken by clinicians to request molecular tests was noted. A survey of staff involved with molecular test requesting revealed a 90% reduction in 'lost requests' after the introduction of the application.

We present a 'lean' method for requesting molecular tests using a smart-phone app. This application can be used to standardise molecular test request forms within the NHS along with automatic generation of a national cancer drugs fund application form for each request. Purpose of Study: Automated approaches for quantitative digital image analysis (DIA) of tissues are becoming increasingly popular in pathology due to advances in whole slide scanning hardware and digital imaging technology. It is essential that DIA is standardised to ensure accuracy and reproducibility of results. Very limited published data exists on the effect of scanner hardware variations on the accuracy and reproducibility of DIA results.The aim of this study was to test the following variables: variation in light source intensity during the day; presnap calibration & white balance of scanned images; variation in DIA due to debris on peripheral parts of the section or coverslip edges. Methods: Immunohistochemistry stained sections from 3 patient samples were scanned on the same Aperio CS scanner, hourly, for 2 consecutive days to generate 15 scanned images of each sample, representing 45 images in total. All scanned images were run through Aperio software using a macro to quantify positive cell counts. For a subset of images, a region of interest was drawn around the tissue to exclude any debris/coverslip edges from peripheral parts of the slide in the subsequent DIA. Statistical analysis was performed to calculate the coefficient of variation between DIA results from scans on different days. Results: Variation in light source intensity accounted for 1.2% to 2.1% variation in cell counts between repeat scans of the same slide. Exclusion of debris/coverslip edges accounted for 1-2% variation in cell counts between repeat scans of the same slide. Subjective analysis revealed no significant difference in appearance of different scans of the same slide. Conclusion: Variation in light source intensity, presnap calibration and overall white balance, plus debris in peripheral areas of the section account for minimal variation in resultant DIA results. Technical advances in scanner hardware have reduced variability in scanning operations. Further investigations are ongoing.

The Complexity of Ovarian Cancer Resistance Mechanisms: A Novel, Clinically Relevant, in-vitro Investigation P S Busschots 1

G Blackshields 1 ; BT Hennessy 3

Novel carboplatin and taxol resistant cell lines were developed from UPN251 OC cells in a clinically relevant selection strategy to better understand resistant mechanisms in OC. UPN251-7C models carboplatin resistance and UPN251-7T models taxol resistance. UPN251-6CALT and UPN251-6TALT were exposed to alternating treatments of both agents during development. Affymetrix arrays were used to characterise gene/miRNA signatures linked with the development of chemoresistance in OC cell lines UPN251-7C and UPN251-7T. Bioconductor software, DAVID v6.7 and miRNA-target interactions (MTIs) analysis was carried out to identify de-regulated genes/miRNAs, gene pathways and gene/miRNA interactions involved in resistance. UPN251 sublines developed using taxol were significantly resistant to taxol, vinblastine and olaparib (P-gp substrates), and reversible with elacridar (P-gp inhibitor) treatment. Significant up-regulation ABCB1 was seen in UPN251-7T which was reflected at the protein level. SRPX2 was highly up-regulated in UPN251-7T. GLI3 and CCL20 were up/down-regulated respectively in UPN251-7C. GLI3 had a validated interaction with miR-205 down-regulated in UPN251-7C. LIN28B was highly deregulated in UPN251-7C and UPN251-7T and had a validated interaction with let-7i, down-regulated in UPN251-7C. P-gp over-expression is a dominant mechanism for taxol resistance in our cell lines. Mechanisms for carboplatin resistance are more complicated. The top deregulated genes are involved in numerous pathways including apoptosis, cellular transformation, signal transduction, and cell migration

Cervical Glandular Neoplasia: The Influence of Excision Procedure on Margin Status

Although cold knife cones (CKC) have been traditionally advocated for treatment of adenocarcinoma in situ (AIS), large loop excisions of the transformation zone (LLETZ) are increasingly used. We analysed excisions from 111 patients where AIS was confirmed prior to procedure over 5 years to assess the influence of excision procedure on final margin status. We tabulated whether margins were involved, close (lesional tissue less than 5mm from margin) or excised. Results: LLETZs were performed in 70% (78 of 111 patients), CKC excision in 30% (33 of 111). Women who had LLETZs were younger than those having CKC excision (32.8 years versus 35.4 years). Positive margins were present in 21% (16 of 78) and close margins in 27% (21 of 78) LLETZ cases. For CKC, margins were positive in 9% of cases (3 of 33) and close in 6% (2 of 33)

Conclusion: Although complete excision is more frequently observed when CKC is performed, compared to LLETZ, for the treatment of AIS

0%) patients, discordant low-grade lymphoma was identified in the bone marrow during staging investigations. 4 were classified as Follicular Lymphoma, 2 Marginal Zone Lymphoma, 2 Chronic Lymphocytic Lymphoma, 1 Lymphoplasmacytic Lymphoma, 1 Non CLL -like Monoclonal B-cell Lymphocytosis. Five had an accompanying a low-grade component that could not be classified. None of the transformed high-grade lymphoma cases were EBV positive. In our cohort, we did not observe statistically significant difference in survival between the Transformed and Non-Transformed cases. An equal proportion of cases transformed from Follicular Lymphoma and Marginal Zone Lymphoma. 19.1% of Transformed patients had a previous history of another cancer, compared to 12.1% of Non-Transformed cases. Age, gender and a history of autoimmune disease were not associated with transformation. Transformed Chronic Lymphocytic Lymphoma

An Investigation Into the Presentation and Nature of Diffuse Large B Cell Lymphoma Within a Large Patient Cohort P HR Freer 1

T Cell/Histoiocyte-Rich Large B Cell Lymphoma (4.2%) was the next most common subtype, followed by Primary DLBCL of the CNS (3.0%). A modified R-IPI (the patient performance score was unknown) was used to stratify the patients into four risk groups and was found to be predictive of patient outcome. The average LDH level was 733.5 IU/L, well above 480, the upper limit of the normal range. Of the cohort of patients, 52.8% achieved remission, 15.5% were alive with disease at the end of the study and 31.7% are now deceased. The majority of patients that did die did so within a year of diagnosis. In addition, 27 Bcl-2 negative patients were identified, with a mean age at diagnosis of 66.8 years. 11 of these patients were female and 16 male. The mean LDH level was 747

Some researchers included additional clinical features that were common when there was malignancy in the asymmetrical tonsils. In this study we aim to evaluate the cancer detection rate in this setting including those who possess high risk clinical features such as age (> 45), pain and history of smoking. Methodology: In total, 119 consecutive tonsillectomy cases, clinically labelled as asymmetrical tonsils were analysed. Out of these, 28 cases (Group 1) were additionally labelled with investigation for unknown primary, obvious lesion or suspicious ulcer seen. The remaining 91 cases with a sole clinical indication of asymmetry were stratified as group 2. Results: In the first group, 7 cases had a histological confirmation of tonsillar primary (25%). While, in the second group, the histological analysis showed 78 cases with benign reactive pathology (89%) while, one case was diagnosed as malignant lymphoma (1%), 6 cases with mild dysplasia (7%)

Conclusion: Our results do correlate with the considerable agreement amongst otolaryngologists that the appearance of an asymmetrically enlarged tonsil in the presence of associated risk factors is considered an indication for tonsillectomy and histological examination given the significant rate of tonsillar malignancies in this group. Anatomical difference in the depth of tonsillar fossa and asymmetry of the anterior tonsillar pillar may give a false impression of a clinically asymmetrical or unilateral enlarged tonsil

Metastatic Adenoid Cystic Carcinoma to the Lung and Kidney: A Single Case Report P RM Doyle

The tumour has three prognostically significant subtypes which form the basis for tumour grading; cribriform, most frequent, tubular and solid, associated with a more aggressive clinical course and metastasis. In the majority of cases (74%) the tumour has an insidious onset and patients have locally invasive disease at first presentation, which coupled with adjacent, important anatomical structures means a complete primary surgical resection is often not feasible. Distant metastases are frequent and predominately involve the lung and bone with renal metastasis a rare occurrence. An optimal treatment regime for ACC has yet to be established and while local management with combination surgery and adjuvant radiotherapy is currently favoured there is conflicting evidence regarding the use of radiotherapy and no formal surveillance guidance for local and regional recurrence or distant metastasis exists. We report a single case of a 52 year old male who presented in 2004 with an asymptomatic right neck mass and underwent right radical neck dissection and adjuvant radiotherapy for ACC. The patient represented in 2014 with aspiration pneumonia on a seven month history of dysphagia and dyspnea. Xray and subsequent computed tomography (CT) imaging identified large, multiple right sided lung lesions, maximum 3cm, and a 5.6cm left renal upper pole mass. CT guided biopsy of the right lung lesion and ultrasound guided biopsy of the renal mass were reported as metastatic adenoid cystic carcinoma

Regan 3 ; CM Martin 2 ; CV Timon 3

Cytokeratin 7 (CK7) is a junctional biomarker with a SEQIKA fragment which stabilises HPV-16 E7 transcripts. We assessed the expression pattern of CK7 protein in tumour specimens from patients diagnosed with oropharyngeal squamous cell carcinoma (SCC) presenting at two major Irish head and neck centres, within the last 10 years. Methods: Archived tumour specimens together with epidemiological data were collected from patients presenting with new primary oropharyngeal SCC at two main head and neck centres in Ireland, within the last 10 years. Briefly, DNA was extracted from tissue blocks and HPV testing carried out using SPF10 HPV PCR

Immunohistochemical staining for CK7 [Clone SP52, Ventana] was performed on tissue blocks following optimisation on the Ventana BenchMark Ultra Immunostainer. Slides were analysed by light microscopy and scored using the H scoring system

with 95% of these identified as HPV-16 subtype. CK7 expression was observed in the tonsillar crypt epithelium of both normal tonsils and tumour specimens. 56% of cases were positive for CK7, with 38% of cases demonstrating H score of >60. CK7 expression in the tumour cells was significantly linked to HPV status and

Our results suggest that the expression of CK7 in normal tonsillar crypt epithelial cells provides a selective advantage to HPV-related carcinogenesis at this site, possible due to the unique propensity of CK7 to bind and stabilise HPV-16 E7 transcripts

Regan 3 ; CM Martin 2 ; CV Timon 3

Methods: Archived HPV-positive tumour specimens and epidemiological data were collected from patients presenting with new primary oropharyngeal SCC at two head and neck centres in Ireland over a one year period. Briefly, DNA was extracted from tissue blocks and HPV testing carried out using SPF10 HPV PCR. The INNO-LiPA HPV Genotyping Extra test [Fujirebio] was used to determine genotype. Immunohistochemical staining for CK7, GDA, MMP-7, AGR-2, PD-1 and PD-L1 was performed following optimisation. Slides were analysed by light microscopy and scored using the H scoring system (junctional biomarkers)

Frozen Section Reporting of Necrotising Granuloma of the Liver Following Percutaneous Instrumentation of The Biliary Tree: A Case Series EH Hadjimichael; P DF Fielding; MI Ilyas; AZ Zaitoun; DL Lobo; PK Kaye Nottingham University Hospital, Nottingham, UK Percutaneous transhepatic cholangiography (PTC) is an interventional radiological technique for both diagnostic imaging and therapeutic decompression of the proximal biliary tract in malignant distal obstruction when retrograde techniques fail. Recognised complications of PTC include sepsis, haemorrhage and pneumothorax. We describe four cases where necrotising granuloma, apparently secondary to previous PTC, has resulted in frozen section examination at the time of subsequent planned cancer resection, to exclude tumour metastasis. Four cases of necrotising granuloma in the liver have been identified between January 2013 and February 2015. All cases were planned Whipple's procedures for pancreatic cancer where initial intraoperative evaluation revealed solitary subcapsular liver lesions. Biopsy and intraoperative frozen section examination were performed to exclude metastatic disease. All frozen sections except one were reported as showing benign necrotizing granuloma formation. The first case was initially reported as malignant and the operation was abandoned. A benign diagnosis was confirmed on paraffin sections in all four cases with the first patient undergoing successful surgical resection at a later date. To the best of our knowledge these are the first reported cases of necrotising granuloma in the liver secondary to prior instrumentation of the liver and leading to intraoperative histological assessment. We highlight this as a potential pitfall in frozen section interpretation undertaken ahead of planned potentially curative surgery which can lead to overstaging of otherwise resectable disease or to the interpretation of a potential diagnosis of tuberculosis. These risks can be reduced with greater surgical and pathological awareness of this entity.

Inclusion body fibromatosis, also known as infantile digital fibroma, is a benign, predominantly myofibroblastic tumour primarily found on the digits of infants. Clinically, these lesions present as asymptomatic cutaneous nodules, rarely larger than 2cm in size, classically on the dorsal or dorsolateral aspect of the second, third and forth digits. They have a high recurrence rate, reported as between 61 and 75%, although this can be reduced by undertaking complete wide local excision. We report a case of inclusion body fibromatosis in an 11 month-old boy, presenting with an enlarging, firm lesion on his left second toe. Following surgical excision, the lesion showed classical histological features of inclusion body fibromatosis -spindle cells arranged in interlacing fascicles in collagenous stroma and numerous pink intracytoplasmic inclusions. The lesion appeared incompletely excised and the patient will be kept under review on account of the high risk of recurrence.Decreased Expression of the Mitochondrial BCAT Protein Correlates with Improved Patient Survival in IDH Wild-Type Gliomas ME Conway 1 ; J Hull 1 ; M El Hindy 1 ; SC Taylor 1 ; F El Amraoui 1 ; C Paton-Thomas 1 ; P White 1 ; M Williams 2 ; P HR Haynes 3 ; SM Hutson 4 ; KM Kurian 3