S51 Evaluation of treatment approaches for hospitalized Covid-19 patients

Abstract

BackgroundCOVID-19 has driven innovation leading to emergency use authorization of treatments that address its urgent healthcare needs. However, physicians, payers and healthcare systems are challenged to select among these novel treatments for both effectiveness and value. Reliance on change in relative, rather than absolute, risk often makes discrimination of treatment effects between medications impractical, with potentially misleading conclusions. Number Needed to Treat (NNT), the reciprocal of the Absolute Risk Reduction, can be a valuable alternative in assessing benefit:risk. The objective of the current analysis was to calculate NNT for reported endpoints of COVID-19 treatments.MethodsClinical information was captured from published literature and pre-prints from investigations of COVID-19 treatments. NNTs were calculated for reported endpoints. Outpatient treatments to reduce viral load included neutralizing antibody ‘cocktails’ REGN-COV21 and bamlanivimab/etesevimab.2 Inpatient treatments included the anti-viral: remdesivir 3,4;and immune modulators: baricitinib5, and lenzilumab.6ResultsREGN-COV2 reduced the number of medically attended visits with NNT of 33.3. The NNT for hospitalization or death was 20 for bamlanivimab/etesevimab. NNTs for 28-day mortality with inpatient treatment were 37 for baricitinib, 26.3 for remdesivir alone, and 22.7 for lenzilumab. Additional analyses of lenzilumab resulted in NNT of 14.7 when combined with remdesivir and corticosteroids, 15.4 when combined with remdesivir, and 13.9 in patients with baseline CRP<150mg/L and age<85 years. The NNT for lenzilumab to prevent survival without ventilation (SWOV) was 15.4 which, decreased to 6.4 in patients with baseline CRP<150mg/L and age<85 years. The number needed to prevent a serious adverse event was 20 for baricitinib, 15 for remdesivir, and 20.4 for lenzilumab.ConclusionThe NNT for COVID-19 treatments varied with setting, endpoint, and mechanism. The NNT for lenzilumab improved with refinement of concomitant medications and patient phenotype. NNT provides a simple measure for comparative analyses that helps inform clinical decision-making and resource allocation.ReferencesWeinreich, et al. N Engl J Med 2021;384:238–51.Gottlieb, et al. JAMA 2021;325:632–44.Garibaldi, et al. JAMA NetwOpen 2021;4:e213071.Beigel, et al. N Engl J Med 2020;383:1813–26.Kalil, et al. N Engl J Med 2020.Temesgen, et al. medRxiv 2021.