Higher BCG-induced trained immunity prevalence predicts protection from COVID-19: Implications for ongoing BCG trials

Endeavors to identify potentially protective variables for COVID-19 impact on certain populations have remained a priority. Multiple attempts have been made to attribute the reduced COVID-19 impact on populations to their Bacillus-Calmette-Guérin (BCG) vaccination coverage ignoring the fact that the effect of childhood BCG vaccination wanes within 5 years while most of the COVID-19 cases and deaths have occurred in aged with comorbidities. Since the supposed protection being investigated could come from heterologous 'trained immunity' (TI) conferred by exposure to Mycobacterium spp. (i.e., environmental and BCG), it is argued that the estimates of the prevalence of TI in populations currently available as latent tuberculosis infection (LTBI) prevalence would be a better variable to evaluate such assertions. Indeed, when we analyze the European populations (24), and erstwhile East and West Germany populations completely disregarding their BCG vaccination coverage, the populations with higher TI prevalence consistently display reduced COVID-19 impact as compared to their lower TI prevalence neighbors. The TI estimates of the populations not the BCG coverage per se, negatively correlated with pandemic phase-matched COVID-19 incidences (r(24): -0.79 to -0.57; p-value < .004), mortality (r(24): -0.63 to -0.45; p-value < .03), and interim case fatality rates (i-CFR) data. To decisively arrive at dependable conclusions about the potential protective benefit gained from BCG vaccination in COVID-19, the ongoing or planned randomized controlled trials should consciously consider including measures of TI as: (a) all individuals immunized do not respond equally, (b) small study groups from higher background TI could fail to indicate any protective effect.