Time-dependent improvement in the clinical outcomes from COVID-19 in cancer patients: An updated analysis of the OnCovid registry

ABSTRACT

Background: Early reports from registry studies demonstrated high vulnerability of cancer patients from COVID-19, with case-fatality rates (CFR) >30% at the onset of the pandemic. With advances in disease management and increased testing capacity, the lethality of COVID-19 in cancer patients may have improved over time. Methods: The OnCovid registry lists European cancer patients consecutively diagnosed with COVID-19 in 35 centres from Jan 2020 to Feb 2021. We analysed clinical characteristics and outcomes stratified in 5 trimesters (Jan-Mar, Apr-Jun, Jul-Sep, Oct-Dec 2020 and Jan-Feb 2021) and studied predictors of mortality across 2 semesters (Jan-Jun 2020 and Jul 2020-Feb 2021). Results: At data cut-off, the 2634 eligible patients demonstrated significant time-dependant improvement in 14-days CFR with trimestral estimates of 29.8%, 20.3%, 12.5%, 17.2% and 14.5% (p<0.0001). Compared to the 2nd semester, patients diagnosed in the Jan-Jun 2020 time period were ≥65 (60.3% vs 56.1%, p=0.031) had ≥2 comorbidities (48.8% vs 42.4%, p=0.001) and non-advanced tumours (46.4% vs 56.1%, p<0.001). COVID-19 was more likely to be complicated in Jan-Jun 2020 (45.4% vs 33.9%, p<0.001), requiring hospitalization (59.8% vs 42.1%, p<0.001) and anti-COVID-19 therapy (61.7% vs 49.7%, p<0.001). The 14-days CFR for the 1st and 2nd semester was 25.6% vs 16.2% (p<0.0001), respectively. After adjusting for gender, age, comorbidities, tumour features, COVID-19 and anti-cancer therapy and COVID-19 complications, patients diagnosed in the 1st semester had an increased risk of death at 14 days (HR 1.68 [95%CI 1.35-2.09]), but not at 3 months (HR 1.10 [95%CI 0.94-1.29]) compared to those from the 2nd semester. Conclusions: We report a time-dependent improvement in the mortality from COVID-19 in European cancer patients. This may be explained by expanding testing capacity, improved healthcare resources and dynamic changes in community transmission over time. These findings are informative for clinical practice and policy making in the context of an unresolved pandemic. Clinical trial identification NCT04393974. Legal entity responsible for the study Imperial College London. Funding: Has not received any funding. Disclosure D.J. Pinato Financial Interests, Personal, Speaker’s Bureau ViiV Healthcare;Financial Interests, Personal, Speaker’s Bureau Bayer;Financial Interests, Personal, Advisory Board EISAI;Financial Interests, Personal, Advisory Board Roche;Financial Interests, Personal, Advisory Board AstraZeneca. All other authors have declared no conflicts of interest.