Using BCG Vaccine to Protect Health Care Workers in the COVID-19 Pandemic

BACKGROUND Health-care workers face an elevated risk of exposure to - and infection with - SARS-CoV-2. It is imperative to ensure the safety, health and fitness of hospital personnel in order to safeguard continuous patient care. Strategies to prevent COVID-19 or to mitigate its clinical consequences are urgently needed.

Bacillus Calmette-Guérin (BCG) was developed as a vaccine against tuberculosis, but our group has shown that it can protect against death from other infections, it has what we have called non-specific effects (NSEs).[3] In clinical studies, BCG vaccination was associated with less child mortality, mainly as a result of reduced neonatal sepsis and respiratory infections. In a WHO-commissioned meta-analysis, BCG was associated with 42% (95%CI: 24-55%) lower child mortality.

NSEs of BCG are not limited to children. An Indonesian trial has shown that consecutive BCG vaccination for 3 months reduced the incidence of acute upper tract respiratory infections by 80% (95%CI=22-95%).

It has been recently demonstrated that the non-specific beneficial effects of BCG vaccination are due to epigenetic and metabolic reprogramming of innate immune cells, a process termed 'trained immunity'. E.g. among humans receiving yellow fever vaccine virus, those who had received BCG had - compared to placebo treated subjects - lower viral load, and improved anti-viral responses.

We hypothesize that BCG vaccination may induce (partial) protection against susceptibility to and/or severity of COVID-19. This study will evaluate the efficacy of BCG to improve the clinical course of COVID-19 and to prevent absenteeism.

Given the immediate threat of the COVID-19 epidemic the trial has been designed as a pragmatic study with a highly feasible primary endpoint, that can be continuously measured. We have therefore chosen unplanned absenteeism from work as primary outcome, in line with an ongoing clinical trial of BCG in the Netherlands.

OBJECTIVES Primary objective: To reduce absenteeism among HCW with direct patient contacts during the COVID-19 epidemic. Secondary objective: To reduce the number of HCW that are infected with SARS-CoV-2 during the COVID-19 epidemic and to reduce the number of hospital admissions amongst HCW with direct patient contacts during the COVID-19 epidemic.

HYPOTHESIS BCG vaccination of HCW will reduce absenteeism by 20% over a period of 6 months.

METHODS

Study design and follow-up:

A multi-center randomized placebo-controlled trial.

STUDY POPULATION As the COVID-19 epidemic has already started in Denmark, our intention is to start as soon as possible. Recruitment of study participants will take place at participating hospitals, at departments were COVID-19 patients are treated and, thus, the risk of exposure to SARS-CoV-2 is high.

Participating hospitals Hospital PI Odense University Hospital Isik Somuncu Johansen Sygehus Lillebælt, Kolding Poul-Erik Kofoed Hvidovre Hospital Thomas Lars Benfield Nordsjællands Hospital Ellen Løkkegaard Aarhus University Hospital Christian Wejse Herning Hospital Lars Skov Dalgaard Sygehus Sønderjylland Christian B. Mogensen

TREATMENT OF SUBJECTS Participants will be randomized 1:1 to receiving one intradermal BCG vaccine or placebo. Participants that are randomized in the active arm will receive a BCG vaccine (BCG-Denmark, AJ Vaccines). Placebo will be 0.1 ml sterile 0.9 % NaCl, which has a similar color as the resuspended BCG vaccine.

STUDY PROCEDURES Day 0: Participants fill in a short questionnaire regarding baseline characteristics. A blood sample will be drawn for subsequent testing for SARS-CoV-2 antibodies. Participants will be randomised to BCG or placebo in a 1:1 ratio. Randomisation will be done centrally using the REDCap tool with stratification per hospital and by age groups (+/- 45 years of age) in blocks of 8.

Participants will be blinded to treatment. The physicians administering the BCG vaccine or placebo will not be blinded. In case of serious adverse events, the participant can be unblinded after consultation with the coordinating PI or sponsor.

The end of the trial is defined as whichever comes latest: the last patient's last registration in the online data collection, or 180 days.

Day 0 till end of trial (weekly): Short electronic questionnaire regarding work absenteeism, symptoms and side effects through REDCap. In case a participant reports acute respiratory symptoms indicating COVID-19, he or she will be tested depending on test availability.

End of trial: Participants are asked to fill in a final questionnaire. A blood sample will be drawn for subsequent testing for SARS-CoV-2 antibodies.

STATISTICAL ANALYSIS All analyses will be performed from the intention-to-treat principle. The primary endpoint will be analysed as counts per week (i.e. multiple observations per subject) using a Bayesian negative binomial regression. Similar regression models will be used for secondary endpoints: Number of days of unplanned absenteeism because of documented COVID; number of days of self-reported fever. Documented COVID, death for any reason, death due to documented COVID, cumulative incidence of Intensive Care Admission for any reason, cumulative incidence of Intensive Care Admission due to documented COVID, cumulative incidence of Hospital Admission for any reason, cumulative incidence of Hospital Admission due to documented COVID will be analysed in Cox proportional hazards models.

Participants will be followed until a new vaccine is given (but data will also be analyzed after receipt of a new vaccine type, to study potential interactions).

When applicable (i.e. in the event that one or more participants have died during the follow-up period) a competing events analysis will be performed in addition.

Interim analysis Every month, an interim analysis will be conducted by the independent statistician of the trial. The parallel Dutch trial will enable combined analysis that will strengthen the conclusions and increase the power to analyse the secondary outcomes and more rare outcomes as intensive hospitalization.

ETHICAL CONSIDERATIONS The study has ben approved by the Ethics Committee and by the Danish Medicines Agency.