METHODS: In this multi-centre quadruple-blind, parallel assignment randomised control trial, 495 high-risk group adults (aged 18-60 years) were randomised into BCG and placebo arms and followed up for 9 months from the date of vaccination. The primary outcome was the difference in the incidence of COVID-19 infection at the end of 9 months. Secondary outcomes included the difference in the incidence of severe COVID-19 infections, hospitalisation rates, intensive care unit stay, oxygen requirement and mortality at the end of 9 months. The primary analysis was done on an intention-to-treat basis, while safety analysis was done per protocol.

RESULTS: There was no significant difference in the incidence rates of cartridge-based nucleic acid amplification test (CB-NAAT) positive COVID-19 infection [odds ratio (OR) 1.08, 95% confidence interval (CI) 0.54-2.14] in the two groups, but the BCG arm showed a statistically significant decrease in clinically diagnosed (symptomatic) probable COVID-19 infections (OR 0.38, 95% CI 0.20-0.72). Compared with the BCG arm, significantly more patients developed severe COVID-19 pneumonia (CB-NAAT positive) and required hospitalisation and oxygen in the placebo arm (six versus none; p = 0.03). One patient belonging to the placebo arm required intensive care unit (ICU) stay and died. BCG had a protective efficacy of 62% (95% CI 28-80%) for likely symptomatic COVID-19 infection.

CONCLUSIONS: BCG is protective in reducing the incidence of acute respiratory illness (probable symptomatic COVID-19 infection) and severity of the disease, including hospitalisation, in patients belonging to the high-risk group of COVID-19 infection, and the antibody response persists for quite a long time. A multi-centre study with a larger sample size will help to confirm the findings in this study.

CLINICAL TRIALS REGISTRY: Clinical Trials Registry India (CTRI/2020/07/026668).