ABSTRACT
Background: Cancer patients are at increased risk of severe COVID-19. As COVID-19 presentation and outcomes are heterogeneous in cancer patients, decision-making tools for hospital admission, severity prediction and increased monitoring for early intervention are critical. Objective: To identify features of COVID-19 in cancer patients predicting severe disease and build a decision-support online tool; COVID-19 Risk in Oncology Evaluation Tool (CORONET) Method: Data was obtained for consecutive patients with active cancer with laboratory confirmed COVID-19 presenting in 12 hospitals throughout the United Kingdom (UK). Univariable logistic regression was performed on pre-specified features to assess their association with admission ([≥]24 hours inpatient), oxygen requirement and death. Multivariable logistic regression and random forest models (RFM) were compared with patients randomly split into training and validation sets. Cost function determined cut-offs were defined for admission/death using RFM. Performance was assessed by sensitivity, specificity and Brier scores (BS). The CORONET model was then assessed in the entire cohort to build the online CORONET tool. Results: Training and validation sets comprised 234 and 66 patients respectively with median age 69 (range 19-93), 54% males, 46% females, 71% vs 29% had solid and haematological cancers. The RFM, selected for further development, demonstrated superior performance over logistic regression with AUROC predicting admission (0.85 vs. 0.78) and death (0.76 vs. 0.72). C-reactive protein was the most important feature predicting COVID-19 severity. CORONET cut-offs for admission and mortality of 1.05 and 1.8 were established. In the training set, admission prediction sensitivity and specificity were 94.5% and 44.3% with BS 0.118; mortality sensitivity and specificity were 78.5% and 57.2% with BS 0.364. In the validation set, admission sensitivity and specificity were 90.7% and 42.9% with BS 0.148; mortality sensitivity and specificity were 92.3% and 45.8% with BS 0.442. In the entire cohort, the CORONET decision support tool recommended admission of 99% of patients requiring oxygen and of 99% of patients who died. Conclusions: CORONET, a decision support tool validated in hospitals throughout the UK showed promise in aiding decisions regarding admission and predicting COVID-19 severity in patients with cancer presenting to hospital. Future work will validate and refine in further datasets.
Subject(s)
Neoplasms , Death , Kyasanur Forest Disease , COVID-19ABSTRACT
Background.There is conflicting evidence about how HIV infection influences COVID-19. We compared the presentation characteristics and outcomes of people with and without HIV hospitalised with COVID-19 at 207 centres across the United Kingdom. Methods.We analysed data from people with laboratory confirmed or highly likely COVID-19 enrolled into the ISARIC CCP-UK study. The primary endpoint was day-28 mortality after presentation. We used Kaplan-Meier methods and Cox regression to describe the association with HIV status after adjustment for sex, ethnicity, age, indeterminate/probable hospital acquisition of COVID-19 (definite hospital acquisition excluded), presentation date, and presence/absence of ten comorbidities. We additionally adjusted for disease severity at presentation as defined by hypoxia/oxygen therapy. Findings.Among 47,539 patients, 115 (0{middle dot}24%) had confirmed HIV-positive status and 103/115 (89{middle dot}6%) had a record of antiretroviral therapy. At presentation, relative to the HIV-negative group, HIV-positive people were younger (median 55 versus 74 years; p<0{middle dot}001), had a higher prevalence of obesity and moderate/severe liver disease, higher lymphocyte counts and C-reactive protein, and more systemic symptoms. The cumulative incidence of day-28 mortality was 25{middle dot}2% in the HIV-positive group versus 32{middle dot}1% in the HIV-negative group (p=0{middle dot}12); however, stratification for age revealed a higher mortality among HIV-positive people aged below 60 years. The effect of HIV-positive status was confirmed in adjusted analyses (adjusted hazard ratio [HR] 1{middle dot}49, 95% confidence interval [CI] 0{middle dot}99-2{middle dot}25; p=0{middle dot}06). Following additional adjustment for disease severity at presentation, mortality was higher in HIV-positive people (adjusted HR 1{middle dot}63; 95% CI 1{middle dot}07-2{middle dot}48; p=0{middle dot}02). In the HIV-positive group, mortality was more common among those who were slightly older and among people with obesity and diabetes with complications. Interpretation.HIV-positive status may be associated with an increased risk of day-28 mortality following a COVID-19 related hospitalisation. Funding.NIHR, MRC, Wellcome Trust, Department for International Development, Bill and Melinda Gates Foundation. Study registrationISRCTN66726260 RESEARCH IN CONTEXTO_ST_ABSEvidence before this studyC_ST_ABSWe searched PubMed for articles in all languages containing the words "COVID*", "coronavirus", "SARS CoV-2" AND "HIV". After screening on 23rd July 2020, we found 51 articles reporting outcomes of COVID-19 in HIV-positive people. Of these, 2 were systematic reviews, 24 were single case reports or case series of under 10 participants, and 12 were larger case series or retrospective cohorts without matched controls. There were two cohort studies that matched HIV-positive people diagnosed with COVID-19 to the general population attending for HIV care in the same area, and three studies that matched HIV-positive people diagnosed with COVID-19 to HIV-negative controls. Some of the evidence from the United States and Europe to date suggests that people with HIV experience a similar disease course and outcomes of COVID-19 compared to the general population. However, many of the studies are limited by small sample size, lack of comparator group and lack of adjustment for potential confounding. In contrast, preliminary results from a cohort study of over 20,000 participants in South Africa indicate that HIV-positive status more than doubles the risk of COVID-19 related mortality. Currently, the evidence from the United Kingdom is limited to two case series comprising a total of 21 patients. Added value of this studyThis study analysed data collected from 207 sites across the United Kingdom as part of ISARIC CCP, the largest prospective cohort of patients hospitalised with COVID-19, to evaluate the association between HIV-positive status and day-28 mortality. The study has the benefit of a relatively large number of participants with HIV (n=115, almost all receiving antiretroviral therapy) and importantly, the ability to direct compare their presenting characteristics and outcomes to those of 47,424 HIV-negative controls within the same dataset. This includes the ability to assess the influence of gender, ethnicity and age, as well as the effect of key comorbidities including chronic cardiac, pulmonary, renal and haematological disease, diabetes, obesity, chronic neurological disorder, dementia, liver disease, and malignancy. Unlike some of the other evidence to date, but in line with the data from South Africa, this study indicates that HIV-positive status may increase the risk of mortality with COVID-19 compared to the general population, with an effect that was especially evident among people with HIV aged below 60 years and was independent of gender or ethnicity. Although we detected an association between mortality among people with HIV and occurrence of obesity and diabetes with complication, the effect of HIV-positive status persisted after adjusting for comorbidities. Implications of all the available evidencePeople with HIV may be at increased risk of severe outcomes from COVID-19 compared to the general population. Ongoing data collection is needed to confirm this association. Linkage of hospital outcome data to the HIV history will be paramount to establishing the determinants of the increased risk. COVID-19 related hospitalisation should pursue systematic recording of HIV status to ensure optimal management and gathering of evidence.