key: cord-0947748-ejpb9se9
authors: Mégarbane, B.
title: Specific treatments in the critically ill COVID-19 patients: benefits and toxicities
date: 2021-09-30
journal: Toxicology Letters
DOI: 10.1016/s0378-4274(21)00277-0
sha: d5929366e96a3c925b92b2a99541c87926b72897
doc_id: 947748
cord_uid: ejpb9se9

nan

The use of computer modeling and simulation in medicine, also called in silico medicine, is taking big strides towards fulfilling its promise: contributing to a faster, safer and more personalised healthare. Developments in in silico medicine are very relevant to the field of (in silico) toxicology -and vice versa -as we face similar technical, societal and regulatory challenges. This presentation discusses the latest (non-technical) advances in the area of in silico medicine. It demonstrates how building a strong community and stakeholder interaction allows to bring the whole field forward.

In silico technologies are increasingly adopted in all aspects of healthcare, from prevention over diagnostics to treatment prediction and clinical follow-up. This progress is owed to a collaborative effort of a wide variety of stakeholders, from researchers to clinicians, from industry to regulators, and from policy makers to funders. In recent years, FDA and ASME published guidelines and standards on reporting, verification and validation of computer models used in the context of medical devices. In Europe, regulators are working together with academia and industry to establishe guidelines for the evaluation of in silico technologies (beyond the classical well-regulated pharmacometrics) in drug design and development. Organizations such as the Virtual Physiological Human institute (VPHi, the international scientific society for in silico medicine) and the Avicenna Alliance (the alliance of VPHi and industry) are driving initiatives that will further accelerate the adoption of in silico medicine. One such initiative is the work on 'Good Simulation Practice' which will be a quality standard defining how to assess and approve in silico tools before they can be used to produce regulatory evidence on the safety/efficacy of a new medical product. Another initiative is the Community Challenge for the adoption of consensus protocols for the characterization of biological tissue properties (C4bio), which will lead to a 'certificates of birth' for the generated data meeting regulatory requirements. Key to all these initiatives is the involvement of multiple stakeholders, which ensures a wide basis for acceptance and uptake of the developed standards and guidelines and which is the only way to remove the remaining barriers for the full adoption of in silico medicine. Introduction: Since 18 months, the world has been fighting against a terrible pandemic due to severe acute respiratory syndrome coronavirus-2, responsible for the coronavirus disease-2019 (COVID-19). Rapidly, in addition to supportive care mainly based on oxygen administration and mechanical ventilation when requested, several data. We have also started to apply our algorithms to the prediction of key events further down the adverse outcome pathway, allowing them to provide more mechanistic understanding of toxicity. Chemical risk assessment requires information on molecular potency, uncertainty and toxicity mechanism, and decision-makers and regulators need to have confidence in the tools being used. The models we have developed provide additional understanding and confidence, vital for their future use in risk assessment.

Machine learning in silico models in chemical hazard identification

In silico models can be developed by use of Machine learning (ML), a form of artificial intelligence (AI), where knowledge is automatically learned from training data. Many ML modeling techniques exist. Once a model has been developed and validated for a given endpoint, it can be applied to make predictions with high computational speed for new substances similar to the training data. A large number of free and commercial in silico models for many human health-related endpoints are available, and many new models are being developed, especially for adverse outcome pathway (AOP) molecular initiating events (MIEs) and other key events (KEs). Together with in vitro and in chemico assays, in silico approaches are part of so-called new approach methodologies (NAMs). Besides being used for screening, prioritisation, classification and labelling, benign-by-design chemistry etc., in silico NAMs are expected to be increasingly used in e.g. integrated approaches to testing and assessment (IATA) and next generation risk assessment (NGRA) to support regulatory decision making. To enhance regulatory acceptance of in silico models, the OECD defined principles for their validation >15 years ago. The EU has for more than a decade promoted the development and use of non-animal approaches including in silico models under different chemical regulations including REACH, and in the U.S. the aim to significantly reduce animal testing by 2025 and eliminate it completely by 2035 has been announced. This talk will go briefly through the above outlined topics and give some examples of our own in silico systems like the Danish (Q)SAR Database and Models free websites as well as examples of our in silico screening and assessment projects for the Danish EPA.

The pandemic has also crystallized the role of clinical toxicology and identified a new role for clinical toxicologists. Poison Control Centers (PCC) and clinical toxicologists have been committed on several fronts: prevention, including campaigns, collaboration with Government Agencies, toxicovigilance, etc. Our activity in addition to prehospital triage of intoxicated patients helped to stabilize the flooded emergency system. An ad hoc EAPCCT Covid-19 Working Group was established so as to evaluate the impact of Covid-19 on European PCCs activities. A pilot study was performed in 4 European PCCs [Copenhagen, Pavia, Utrecht, Zurich] to evaluate critical points on methods, data collection and resources. The study was then extended to all European PCCs to investigate the effect of the Covid-19 pandemic during the first wave on the activities of European PCC. All 65 European PCCs listed in the WHO directory were asked to supply epidemiological data on poisonings (e.g., number of calls, patients, type of caller (medical professional or public), type of exposure (accidental, intentional (all), intentional suicide attempt), and age groups). Investigated exposures were disinfectants, household cleaning products (according to the European Chemicals Agency (ECHA) classification) and drugs, including antivirals. Data was analyzed during a 4-month period (March-June 2020) and compared with the previous two years. Furthermore, all members of EAPCCT were asked to participate in a survey on organizational changes during the COVID-19 pandemic.Results: The study included data from 36 PCCs from 21 countries (55% of EU PCCs). 60 % of Head of PCCs from 24 countries submitted pandemic related organizational data. Twenty per cent of PCC saw an increase in length of shifts while 42% saw an increase of total number of shifts. 25% indicated an increase of time spent on duties other than PCC activities, such as activities in the emergency departments (35%), intensive care units (12%), and nursing wards (9%). Over 50% of PCCs lacked protocols to manage PCC staffing upon massive sick leave. In contrast safe working protocols were in place in nearly every PCC. No centers reported receiving special funds for activities arising from the pandemic. Conclusion: European PCCs activities vary in different countries. A substantial improvement can be made in many areas such as data harmonization, establishing a European database of poisonings, and an improved network and cooperation of European PCC, and timely toxicovigilance especially in times of global emergencies.

anti-COVID-19 therapies including drugs with demonstrated and drugs with only hypothetical benefits have been used. Many unapproved therapies including cheaper drugs available as over-the-counter have been marketed. We aimed to review the benefits and toxicities of the main specific drugs used in the critically ill COVID-19 patients.

A narrative review based on published articles in PubMed or as preprints.

Results: Anti-COVID-19 pharmacotherapy mainly consisted in antiviral and immunomodulatory drugs given either in an early stage to strengthen the anti-COVID-19 immunity or in a late stage to limit the extent of the deleterious anti-COVID-19 immune response responsible for the so-called cytokine storm. Lopinavir/ritonavir combination and remdesivir have been initially used despite significant liver and renal toxicities but rapidly abandoned due to the absence of limited effectiveness, respectively. The use of hydroxychloroquine and its combination to azithromycin has been also stopped due to the absence of established efficacy in the randomized placebo-controlled studies and the risk of cardiac dysrhythmias in severe COVID-19 patients with organ dysfunction. To date, corticosteroids and antiinterleukin-6 receptor Fab fragments are extensively used due to their demonstrated contribution to limit mortality and severity of COVID-19; however, their use has resulted in marked increase in bacterial and mycological infections, especially in mechanically ventilated patients. Various other drugs such as colchicine and ivermectin are prescribed and the extent of their toxicity in COVID-19 patients is still worthy assessing. Finally, dietary supplements and vitamins including vitamin D and high-dose vitamin C have also been used. Despite an expected excellent tolerance and an interesting rationale to support their anti-COVID-19 activities, their preventive or therapeutic effectiveness has not been confirmed in randomized trials.

The role of clinical toxicologists and poison control centers remains essential to identify drug-related toxicities. Evaluation of anti-COVID-19 drug safety should rely on single or case series and reported side effects in clinical trials.Unapproved treatments should systematically require attention of health authorities.

Understanding vaccine-induced immune thrombocytopenia and thrombosis *B. Hunt

Vaccine-induced immune thrombocytopenia and thrombosis (VITT) is a rare but serious complication seen post COVID-19 vaccination, most notably in the UK after the Astra Zeneca vaccines, occurring in 1 per 50,000 vaccines in the under 50s and about 1 per 100,000 in the over 50s.

VITT causes thromboses in both the arterial and venous beds, generally targeting unusual sites, with cerebral venous sinus thromboses accounting for 50% of the thromboses, and splanchnic vein thrombosis accounting for another 30%. Thrombosis is associated with thrombocytopenia, often a low fibrinogen and very high D-dimers. The syndrome has many similarities to heparin-induced thrombocytopenia and thrombosis and indeed the same antibody to Platelet Factor 4 is found in VITT and is considered to be the pathogenic cause. This is an immune response and therefore no identifying risk factors have been yet found other than younger age.

The presentation and evolving management will be discussed.

Defining Molecular Initiating Events in the Adverse Outcome Pathway Framework for Risk Assessment

A Next-Generation Risk Assessment Case Study for Coumarin in Cosmetic Products

Structural Alerts and Random Forest Models in a Consensus Approach for Receptor Binding Molecular Initiating Events

Neural Network Activation Similarity: A New Measure to Assist Decision Making in Chemical Toxicology

Using Transition State Modeling to Predict Mutagenicity for Michael Acceptors