key: cord-0881818-crg4b8v4
authors: Shields, Adrian M; Anantharachagan, Ariharan; Arumugakani, Gururaj; Baker, Kenneth; Bahal, Sameer; Baxendale, Helen; Bermingham, William; Bhole, Malini; Boules, Evon; Bright, Philip; Chopra, Charu; Cliffe, Lucy; Cleave, Betsy; Dempster, John; Devlin, Lisa; Dhalla, Fatima; Diwakar, Lavanya; Drewe, Elizabeth; Duncan, Christopher; Dziadzio, Magdalena; Elcombe, Suzanne; Elkhalifa, Shuayb; Gennery, Andrew; Ghanta, Harichandrana; Goddard, Sarah; Grigoriadou, Sofia; Hackett, Scott; Hayman, Grant; Herriot, Richard; Herwadkar, Archana; Huissoon, Aarnoud; Jain, Rashmi; Jolles, Stephen; Johnston, Sarah; Khan, Sujoy; Laffan, James; Lane, Peter; Leeman, Lucy; Lowe, David M; Mahabir, Shanti; Mac Lochlainn, Dylan James; McDermott, Elizabeth; Misbah, Siraj; Moghaddas, Fiona; Morsi, Hadeil; Murng, Sai; Noorani, Sadia; O’Brien, Rachael; Patel, Smita; Price, Arthur; Rahman, Tasneem; Seneviratne, Suranjith; Shrimpton, Anna; Stroud, Catherine; Thomas, Moira; Townsend, Katie; Vaitla, Prashantha; Verma, Nisha; Williams, Anthony; Burns, Siobhan O; Savic, Sinisa; Richter, Alex G
title: Outcomes following SARS-CoV-2 infection in patients with primary and secondary immunodeficiency in the United Kingdom
date: 2022-01-31
journal: Clin Exp Immunol
DOI: 10.1093/cei/uxac008
sha: 9dab964463539c3d41303ebd9964628569d2072c
doc_id: 881818
cord_uid: crg4b8v4

In March 2020, the United Kingdom Primary Immunodeficiency Network (UKPIN) established a registry of cases to collate the outcomes of individuals with PID and SID following SARS-CoV-2 infection and treatment. A total of 310 cases of SARS-CoV-2 infection in individuals with PID or SID have now been reported in the UK. The overall mortality within the cohort was 17.7% (n=55/310). Individuals with CVID demonstrated an infection fatality rate (IFR) of 18.3% (n=17/93), individuals with PID receiving IgRT had an IFR of 16.3% (n=26/159) and individuals with SID, an IFR of 27.2% (n=25/92). Individuals with PID and SID, had higher inpatient mortality and died at a younger age than the general population. Increasing age, low pre-SARS-CoV-2 infection lymphocyte count and the presence of common co-morbidities increased the risk of mortality in PID. Access to specific COVID-19 treatments in this cohort was limited: only 22.9% (n=33/144) of patients admitted to hospital received dexamethasone, remdesivir, an anti-SARS-CoV-2 antibody-based therapeutic (e.g. REGN-COV2 or convalescent plasma) or tocilizumab as a monotherapy or in combination. Dexamethasone, remdesivir and anti-SARS-CoV-2 antibody-based therapeutics appeared efficacious in PID and SID. Compared to the general population, individuals with PID or SID are at high risk of mortality following SARS-CoV-2 infection. Increasing age, low baseline lymphocyte count and the presence of co-morbidities are additional risk factors for poor outcome in this cohort.

The risk of morbidity and mortality following SARS-CoV-2 infection in patients with primary immunodeficiency (PID) and secondary immunodeficiency (SID) remains unclear. National epidemiological studies tend to consider immunocompromised individuals as a homogenous group (e.g., HIV-1 infection, solid organ transplant, immunosuppressive medications) and are unable to inform our understanding of outcome in individuals with rare diseases (1, 2) .

Instead, retrospective case series have been used to develop our understanding of the risk in patients with PID and SID (3) (4) (5) (6) (7) (8) (9) . While the results of such studies must be interpreted with caution, a consistent pattern has emerged showing that adult patients with PID and SID are at an increased risk of morbidity and mortality from COVID-19 compared to the general population, and that an increased prevalence of chronic co-morbidities, potentially driven by underlying immunodeficiency, partially contributes to that risk. The risk of severe morbidity and mortality from COVID-19 in children with PID or SID remains less clear with data from the UK showing no increased risk in this group (7), although mortality has been reported elsewhere (8) .

In order to more comprehensively understand the risk of morbidity and mortality from COVID-19 in patients with PID and SID, the United Kingdom Primary Immunodeficiency Network (UK PIN) has systematically collated outcomes from COVID-19 in patients with PID and SID throughout the pandemic. This national effort has enabled a large cohort of individuals with PID and SID to be developed, and their outcomes understood in comparison to well curated national statistics.

Herein, we report morbidity and mortality from COVID-19 in a cohort of 310 individuals with PID or SID from the United Kingdom; we build upon our previous work (5) by providing revised estimates of mortality in clinically important PID and SID subgroups, insight into independent risk factors for mortality and the efficacy of targeted COVID-19 treatments in this cohort.

The United Kingdom Primary Immunodeficiency Network (UK PIN) has systematically collected data from its members on the outcomes of SARS-CoV-2 infection in patients under the care of Clinical Immunology teams across the United Kingdom since March 2020.

In March 2020, data collection proformas were sent to all UK PIN affiliated pediatric and adult immunologists and centres by email. During the first wave of the UK pandemic (March 2020 -July 2020), data collected included age, sex, ethnicity, body mass index, pre-infection lymphocyte count (taken from a full blood count at the last immunology outpatient appointment prior to SARS-CoV-2 infection), prior immunological treatments [e.g. immunoglobulin replacement, immunosuppression (any biologic M a n u s c r i p t 6 immunosuppressive within 1 month of SARS-CoV-2 infection, any anti-CD20 depleting agents within 6months of SARS-CoV-2 infection, or any daily oral immunosuppression including steroids), and antibiotic prophylaxis], existing chronic comorbidities, SARS-CoV-2 PCR result, whether individuals were hospitalized, and whether individuals survived or died. The outcomes from the first 100 cases in this case series have been published previously (5) . During the subsequent waves of the UK pandemic (September 2020 -present), additional data collected included whether an individual received any targeted COVID-19 treatments during their inpatient stay and whether an individual seroconverted following natural infection.

In the United Kingdom, care for hospitalised patients with COVID-19 was protocolised at a national level.

Only treatments with demonstrable efficacy in randomised control trials (e.g. RECOVERY) were approved for mainstream use. Patients' access to dexamethasone, remdesivir, tocilizumab and antibody-based therapeutics was either through enrolment and randomisation within the RECOVERY study, or through protocolised care as laid out by the National Institute for Health and Care Excellence (NICE) guidelines (https://www.nice.org.uk/guidance/ng191/chapter/Recommendations). The published literature was also reviewed to capture any further UK cases of COVID-19 in immunodeficiency patients that had been published elsewhere: this literature search revealed two additional cases (10, 11) which have been included.

Data were collated according to 2019 IUIS classification of inborn errors of immunity and further analysis undertaken on three subgroups: i) individuals with common variable immunodeficiency (including six patients with monogenic CVID-like disease secondary to haploinsufficiency of NFκB1, NFκB2, BACH-2 or CTLA-4), collectively referred to as CVID-phenotype, ii) all individuals with primary immunodeficiency receiving immunoglobulin replacement (IgRT) including CVID but excluding those who had received definitive treatment with allogenic stem cell transplantation or gene therapy (i.e. those with a clinically significant antibody deficiency regardless of proximal diagnosis) and iii) individuals with secondary immunodeficiency.

Data were analyzed using GraphPad Prism 9.0 (GraphPad Prism Software, San Diego, Calif). Differences between the distributions of continuous variables were evaluated using the 2-tailed Mann-Whitney U test. Differences between categorical variables were evaluated using the 2-tailed Chi-square test. Odds ratios were calculated using the Baptista-Pike method. Infection fatality ratios and case fatality ratios are defined according to the World Health Organisation (https://www.who.int/newsroom/commentaries/detail/estimating-mortality-from-covid-19). Briefly, case fatality ratio (CFR) is defined as the total number of deaths from SARS-CoV-2 divided by the total number of PCR proven SARS-CoV-2 infections. Infection fatality ratio (IFR) is defined as the total number of deaths from SARS-CoV-2 divided by the total number of suspected cases regardless of whether they are proven by molecular diagnostics. Both IFR and CFR are presented to more accurately capture mild COVID-19 cases occurring early in the pandemic (February -May 2020), a period where UK community transmission was high, but the availability of molecular diagnostics for non-hospitalised cases low. All deaths reported herein occurred in individuals with PCR proven COVID-19. UK national statistics are sourced from the UK Department of Health and Social Care (1/1/2020 -29/3/2021). Detailed methodology on how these statistics are collated is available on the UK Coronavirus Dashboard (https://coronavirus.data.gov.uk/). When analyzing pre-SARS-CoV-2 infection lymphocyte counts in the SID cohort, patients with chronic lymphocytic leukemia were excluded. M a n u s c r i p t 7 Multiple logistic regression models were constructed using survival as the outcome variable. Age and baseline lymphocyte count were included as continuous variables; sex, receipt of prophylactic antibiotics, receipt of immune suppression, and the presence of different comorbidities were included as categorical variables. Odds ratios for continuous variables are expressed as the change in odds of survival per unit increase of that continuous variable within the model. Odds ratios for categorical variables are expressed as the odds ratio associated with mortality if that comorbidity is present.

A total of 310 cases of SARS-CoV-2 infection in patients with PID or SID were recorded between March 2020 and July 2021 from 27 UK PIN affiliated immunology departments representing all four nations of the United Kingdom. Cases included 218 individuals with primary immunodeficiency, 6 of whom had received allogenic stem cell transplantation or gene therapy for their underlying condition, 92 individuals with secondary immunodeficiency. In addition, 13 individuals with C1 inhibitor deficiency and 3 with autoinflammatory diseases were also reported.

In patients with primary or secondary immunodeficiency, 45.8% (n=142/310) of patients were hospitalised and the overall IFR in this cohort was 17.7% (n=55/310). 86.5% (n=268/310) had SARS-CoV-2 confirmed by PCR or rapid antigen testing; the remainder had either an illness consistent with COVID-19 when the availability of molecular testing for SARS-CoV-2 was limited and had the diagnosis made clinically or radiologically. Hospitalisation rates, IFR and CFR data for specific immunodeficiencies are presented in Table 1 .

To understand risk factors associated with mortality from COVID-19 in patients with immunodeficiency, the following sub-groups of individuals were considered: individuals with common variable immunodeficiency including those with monogenic CVID-like disease, collectively referred to as CVIDphenotype (n=93), individuals with PID receiving immunoglobulin replacement therapy (n=159), and individuals with secondary immunodeficiency (n=92).

Ninety-three cases of SARS-CoV-2 infection in patients with a CVID-phenotype were analysed; 87 individuals had genetically undifferentiated CVID and 6 had a monogenic CVID-like disease ( Table 1) . The median age of this cohort was 48 years (IQR 30.3-57.0), 57.0% (n=53/93) were female and 51.6% (n=48/93) were receiving antibiotic prophylaxis. IFR in this cohort was 18.3% (n=17/93) and increased CFR were observed in all age brackets over 20 years in comparison to national statistics ( Table 2) . Median age of death from COVID-19 in individuals with CVID was 63.0 years. M a n u s c r i p t 8 Univariate analysis demonstrated that increasing age, lower baseline lymphocyte count, receipt of prophylactic antibiotics, and the presence of chronic lung disease, cardiovascular disease, chronic liver disease, diabetes mellitus and chronic gastrointestinal disease were all associated with a significantly increased risk of mortality from COVID-19 in these individuals (Figure 1, Supplementary Table 1 ). 64.0% of individuals who died had a pre-COVID lymphocyte count less than 1.0x10 9 /L compared to 21.8% who survived (p<0.0001). The use of current immunosuppression was significantly over-represented in patients with a lymphocyte count less than 1.0x10 9 /L (36.0% vs 13.3%, p=0.02), as was the prevalence of granulomatous-lymphocytic interstitial lung disease (GL-ILD) (28.0% vs 10.0%, p=0.04), organ-specific autoimmunity (48.0% vs 18.0%, p=0.005) and splenomegaly (24.0% vs 8.0% p=0.05) in comparison to individuals with CVID with a lymphocyte count greater than 1.0x10 9 /L. In multivariate analysis, a low pre-SARS-CoV-2 infection lymphocyte count and receipt of prophylactic antibiotics were independently associated with an increased risk of mortality from COVID-19 in these individuals (Supplementary Table  S2 ). Six patients who survived COVID-19 had serological responses to natural infection measured; 100% were found to be positive (5 spike glycoprotein antibody positive, 1 nucleocapsid antibody positive).

One-hundred and fifty-nine cases of SARS-CoV-2 infection in individuals with PID receiving IgRT have been recorded in this case series. The median age of these individuals was 40 years (IQR 28.0 -56.3), 44.7% (n=71/159) were female and in addition to IgRT, 55.3% (n=88/159) were receiving prophylactic antibiotics. 41.5% (n=66/159) were hospitalised as a result of SARS-CoV-2 infection, a further 2 patients attended the emergency department but were not admitted to hospital and 1 patient was admitted for a flare of immune thrombocytopenia purpura, temporally associated with SARS-CoV-2 infection. IFR in this subgroup was 16.3% (n=26/159) and an increased CFR were observed in all age brackets above 20-29 compared to UK national statistics ( Table 2) . Median age of death from COVID-19 in this cohort was 57.0 years compared to the UK national average of 83.0 years.

In univariate analysis, patients with PID receiving IgRT who died of COVID-19 were significantly older, had lower pre-SARS-CoV-2 infection lymphocyte counts ( Figure 1A ) and had a significantly higher prevalence of chronic lung disease (specifically bronchiectasis), cardiovascular disease, chronic liver disease and diabetes mellitus than those who survived ( Table 3) . 61.5% of individuals who died had a preinfection lymphocyte count of less than 1.0x10 9 /L compared to 21.8% who survived (p<0.0001). The use of current immunosuppression was significantly over-represented in patients with a lymphocyte count less than 1.0x10 9 /L (23.8% vs 10.7%, p=0.04) in this subgroup, as was the prevalence of GL-ILD, organ specific autoimmunity and splenomegaly. In multivariate analysis, a higher pre-infection lymphocyte count was independently associated with survival but chronic liver disease and diabetes mellitus were independently associated with mortality ( Table 4) .

M a n u s c r i p t 9

Twenty-six patients with X-linked agammaglobulianemia were included ( Table 1 ). The cohort of XLA patients were, on average, younger than other individuals with antibody deficiency (median age 29.5 years) and IFR was lower at 7.7% (n=2/26). Of the two individuals who succumbed to COVID-19, one had received lung transplantation for respiratory complications of their immunodeficiency (11) ; the other was reported to suffer from nodular regenerative hyperplasia but had no pre-existing structural lung disease and died despite receiving treatment with dexamethasone, remdesivir, tocilizumab and convalescent plasma.

Ninety-two patients with SID, under the care of a clinical immunologist were included in this case series ( Table 1) . This cohort was, on average, older than the PID cohort with a median age of 63.0 years and outcomes were worse, with an IFR of 27.2% (n=25/92). Compared to UK national statistics, higher IFR were observed in all age brackets above 40 years ( Table 2) . Haematological malignancy and its treatment was the most common cause of secondary immunodeficiency in this cohort accounting for 56.5% (n=52/92) of cases: non-Hodgkin's lymphoma (n=27), chronic lymphocytic leukaemia (n=11) and plasma cell dyscrasias were the most common underlying diseases in this group. 55.8% (n=29/52) of these patients were hospitalised with and IFR of 23.1% (n=12/52).

Individuals whose secondary immunodeficiency arose from underlying rheumatological disease were younger, more likely to be receiving immune suppression at the time of infection and had a higher IFR of 38.1% (n=8/21). In univariate analysis, patients with SID who died from COVID-19, had significantly lower pre-infection lymphocyte counts than those who survived (Figure 1, Table 5 ). 52.6% of individuals who died had pre-SARS-CoV-2 lymphocyte count less than 1.0x10 9 /L compared to 26.6% who survived (p=0.03). We also observed that in the SID cohort, patients receiving daily prednisolone as immunosuppression were far more likely to die than those not receiving daily prednisolone (mortality 44.0% vs. 7.46%, p<0.0001). However, multivariate analysis did not show a significant independent effect of any of these variables (Supplementary Figure 3) . Seroconversion following natural infection was lower in SID than in CVID; 14 patients were assessed and 42.8% (n=6/14) were found to be seropositive following their illness compared to 100% (n=6/6) in CVID.

Data on the outcomes of 13 patients with C1 inhibitor deficiency were submitted, all of whom survived ( Table 1) ; the median age of patients with C1 inhibitor deficiency was 45 years (IQR: 29.5-51.5) and 46.1% were female. Only one patient, a man in his 20s with other pre-existing comorbidities, required hospital treatment specifically for COVID-19; another required hospitalisation for renal disease. Of note, additional information was submitted for 2 patients implicating SARS-CoV-2 infection as a precipitating factor for concurrent flares of hereditary angioedema, as has been previously reported (12, 13) . M a n u s c r i p t 10 Acute treatments for 144 patients in this cohort required hospital treatment for COVID-19, with an overall inpatient survival of 61.8% (n=89/144). 22.9% of hospitalised patients (n=33/144) received a targeted COVID-19 treatment within the RECOVERY trial, on a compassionate use basis, or as standard of care following adoption of treatment proven to be efficacious in the RECOVERY trial ( Table 6 ). In this cohort, 20 patients received dexamethasone, 26 patients received remdesivir and 10 patients received anti-SARS-CoV-2 antibodybased therapies (e.g. REGEN-COV2, convalescent plasma) either as a monotherapy or in combination: the survival rates of individual patients treated with dexamethasone and/or remdesivir and/or antibody-based treatments exceeded 75%. Five patients received tocilizumab as a monotherapy or in combination: survival rates in this cohort were 20%.

The United Kingdom has recorded amongst the highest number of cases (8,918 per 100,000 population) and deaths (193.9 per 100,000 population) from COVID-19 in the world (14). Despite public health measures to minimise the exposure of clinically extremely vulnerable individuals to SARS-CoV-2 (15), the pandemic has disproportionately affected patients with PID and SID. In this large nationwide study, we demonstrate an increased risk of mortality from COVID-19 in individuals with PID requiring antibody replacement therapy and individuals with CVID in all age groups above 20 years old compared to the UK general population. In patients with SID, an increased risk of mortality emerges beyond 40 years of age. Inpatient mortality in patients with immunodeficiency was higher than the general population (38.2% vs 26.0%) (1) . In comparison to the general population, where the median age of death from COVID-19 was 83.0 years, immunodeficiency patients died, on average, up to 26 years younger than the general population.

The risk factors for poor outcome from COVID-19 in patients with primary immunodeficiencies closely mirror those in the general population (16); increasing age and common comorbidities were all associated with an increased risk of death in univariate analysis; chronic liver disease and diabetes mellitus (PID on IgRT) and receipt of prophylactic antibiotics (CVID) were independently associated with death in multivariate analysis of these cohorts. In SID, no additional independent risk factors were observed, but the overall CFR of 31.6% compared to 2.95% in the UK general population illustrates the vulnerability of individuals requiring long-term immunological support following treatment for other diseases. Furthermore, the CFR of 38.1% for patients with rheumatological disease in this study is markedly higher than the estimated CFR of 5.6% for unselected rheumatoid arthritis patients demonstrating the subgroup of patients with SID are at especially high risk of poor outcome (17) . Our data also highlights that those individuals with secondary immunodeficiency who continue to receive treatment with any daily dose of prednisolone also appear at increased risk of mortality. Reassuringly and concordant with other UK studies in immunocompromised children (7), we found no increased risk of mortality amongst in the 18 paediatric patients within this study, although 27.7% were hospitalised. M a n u s c r i p t 11 One striking observation from this study is that pre-existing lymphopenia is independently associated with mortality in all three groups with immunodeficiency. Previous studies have demonstrated that pre-existing lymphopenia is independently associated with an increased risk of developing pneumonia, skin infections, urinary tract infections, sepsis and endocarditis in the Danish general population (18) and an increased risk of mortality from pneumonia in the UK general population (19) . Peripheral CD4 and CD8 T cell lymphopenia and dysregulated T cell responses have also been associated with severe disease during acute COVID-19, although some studies suggest this observation is secondary to lymphocyte redistribution during acute illness (20, 21) . In our study, pre-existing lymphopenia in patients with CVID and PID receiving IgRT was associated with GL-ILD, organ-specific autoimmunity and splenomegaly and the use of immunosuppressive drugs. The relative contributions of these variables to the pathogenesis of the observed lymphopenia remain unclear; characterising the nature of pre-existing lymphopenia, its composition by lymphocyte subsets analysis, and its impact on functional immunity with respect to outcomes from infectious disease is an important research priority in both the general population and patients with immune deficiency.

There is a paucity of evidence regarding the efficacy of treatments for COVID-19 in patients with PID and SID (22) . In this case series, compared to an overall inpatient survival following COVID-19 of 61.8%, survival was improved in individuals who received dexamethasone (75.0%), remdesivir (84.6%) or antibody-based treatments (80.0%) as a monotherapy or in any combination. In contrast, only 20% of individuals receiving tocilizumab survived. These data require cautious interpretation; specific data on the timing of these pharmacological interventions in relation to disease onset and COVID-19 severity was not gathered as part of this study and the number of treated individuals is small. However, it is possible, or even likely, that modulation of the immune response in a patient with immunodeficiency is different to the general population. For example, although convalescent plasma demonstrated no benefit in healthy individuals during acute severe COVID-19 infection (23), antibody-based treatments appear effective herein and in case reports of antibody deficiency patients where ex vivo studies have confirmed viral neutralization (24) . Furthermore, polyclonal immunoglobulin replacement is the standard of care in the prevention of chronic sinopulmonary infection in individuals with antibody deficiency and hyperimmune serum may be used for post-exposure prophylaxis in seronegative individuals against certain infectious diseases, a concept supported by early data in anti-SARS-CoV-2 monoclonal antibody trials (25, 26) . Similarly, inhibition of the IL-6 axis in individuals already suffering from an existing immunodeficiency, may have unintended consequences despite success in randomised control trials in previously healthy individuals (27) .

There are significant differences in the estimated mortality rates from COVID-19 in the different national and international registry-based studies published to date (3-9) and many estimates of the case fatality ratio in patients with inborn errors of immunity do not exceed those of the general population (28) . Comparison of national statistics is challenging; differences in the response to and course of the pandemic worldwide will impact mortality statistics. The age-stratified CFR reported in this study are broadly concordant with other studies: the risk of death in paediatric patients is very low regardless of their underlying immunodeficiency but a significantly increased risk of morbidity and mortality does emerge with increasing age, becoming pronounced above the age of 40 years in comparison to the general population. The high overall CFR observed in this study may arise because the UK PIN cohort is, on average, older M a n u s c r i p t 12 than the cohorts reported in other studies which are relatively enriched with paediatric patients (3) (4) (5) (6) (7) (8) (9) . Difference in access to rational therapeutics to treat severe COVID-19 may also contribute to the observed differences.

The major strength of this study is its size and comprehensive representation amongst UK PIN affiliated centres across the United Kingdom reducing the potential for bias seen in smaller case series and facilitating the enrolment of large numbers of patients with rare disease. The work confirms the increased risk of mortality from COVID-19 we observed in our original case series (5) and builds upon it by providing revised estimates of the magnitude of that risk in relation to the UK general population and the determinants of that risk in clinically relevant disease subgroups. However, this study remains a clinicianreported registry and we are unable to guarantee that all SARS-CoV-2 infections in patients with PID or SID have been captured by this study. Furthermore, bias may exist within the SID cohort described herein; by definition, these individuals have been referred to a Clinical Immunologist for immunological assessment which may enrich for more severe phenotypes of SID. Cases of SARS-CoV-2 in paediatric patients remain underrepresented by this study, but similar outcomes in a larger cohort of UK patients have been reported elsewhere (7) . Owing to the retrospective nature of this study, we are unable to characterize the nature or longevity of serological responses of individuals with PID or SID following natural infection; however, the COV-AD study, a national UK study studying the cellular and humoral response to SARS-CoV-2 natural infection and vaccination is in progress and will be able to inform upon these important immunological questions.

In summary, our study highlights the burden of morbidity and mortality in individuals with PID and SID following infection with SARS-CoV-2 and elucidates independent risk factors associated with poor outcome. The impact of the COVID-19 on individuals with PID and SID cannot be underestimated; these data must inform public health policy, including the urgent provision of anti-SARS-CoV-2 antibody-based therapies this population, to minimise the risk of poor outcome during future waves of the evolving pandemic. M a n u s c r i p t M a n u s c r i p t 23 Age-stratified hospitalisation rates, CFR and IFR are presented and compared to UK national statistics for the general population. UK national data is sourced from UK Department of Health and Social Care statistics between 1/1/2020 and 29/3/2021; ** age not provided for one patient with CVID described in Table 1 who died, $ excludes one patient admitted with immune thrombocytopenic purpura immediately following SARS-CoV-2 infection, * excludes 2 patients attending emergency department but not admitted to hospital. All reported COVID-19 deaths occurred in patients with PCR-proven SARS-CoV-2 infection during their hospital admission. 

Features of 20 133 UK patients in hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: prospective observational cohort study

Underlying conditions and risk of hospitalisation, ICU admission and mortality among those with COVID-19 in Ireland: A national surveillance study

Coronavirus disease 2019 in patients with inborn errors of immunity: An international study

Minor Clinical Impact of COVID-19 Pandemic on Patients With Primary Immunodeficiency in Israel

COVID-19 in patients with primary and secondary immunodeficiency: The United Kingdom experience

COVID-19 AND PRIMARY IMMUNODEFICIENCY: ONE-YEAR EXPERIENCE

Immunocompromised children and young people are at no increased risk of severe COVID-19

Impact of SARS-CoV-2 Pandemic on Patients with Primary Immunodeficiency

Outcome of SARS-CoV-2 Infection in 121 Patients with Inborn Errors of Immunity: A Cross-Sectional Study

Treatment of COVID-19 with remdesivir in the absence of humoral immunity: a case report

Possible COVID-19 reinfection in a patient with X-linked agammaglobulinaemia

COVID-19 as a trigger of acute attacks in people with hereditary angioedema

COVID-19 affecting hereditary angioedema patients with and without C1 inhibitor deficiency

Coronavirus (COVID-19) in the UK

Guidance on protecting people who are clinically extremely vulnerable from COVID-19

OpenSAFELY: factors associated with COVID-19 death in 17 million patients

Risk of COVID-19 in Rheumatoid Arthritis: A National Veterans Affairs Matched Cohort Study in At-Risk Individuals. Arthritis Rheumatol

Lymphopenia and risk of infection and infection-related death in 98,344 individuals from a prospective Danish populationbased study

Association of prior lymphopenia with mortality in pneumonia: a cohort study in UK primary care

Immunology of COVID-19: Current State of the Science

Profound dysregulation of T cell homeostasis and function in patients with severe COVID-19

Persistent SARS-CoV-2 infection: the urgent need for access to treatment and trials

Convalescent plasma in patients admitted to hospital with COVID-19 (RECOVERY): a randomised controlled, open-label, platform trial

Resolution of Persistent COVID-19 After Convalescent Plasma in a Patient with B Cell Aplasia

REGN-COV2, a Neutralizing Antibody Cocktail, in Outpatients with Covid-19

Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial

Coronavirus disease 2019 in patients with inborn errors of immunity: lessons learned

A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t 33