**2. BCG the Current TB Vaccine in Use**

Despite its variable efficacy against respiratory forms of TB, BCG remains the only marketed vaccine in use against TB, with more than 90% coverage in countries with high TB incidence [5–7] (Figure 1). BCG is an attenuated vaccine derived from *Mycobacterium bovis*, the etiologic agent of TB in cattle [8]. BCG was first introduced in clinic almost a hundred years ago, when in 1921 it was administered orally to a baby whose mother had died of TB the day after her birth. The baby showed no adverse effects to BCG vaccination and, more importantly, did not develop TB. Between 1921 and 1926 more than 50,000 children were vaccinated. Mortality among vaccinated children was 1.8%, compared to a mortality greater than 25% in unvaccinated children, showing its effectiveness in reducing infant mortality, not only due to TB if not due to other respiratory diseases [9]. Today we know that the main cause of attenuation of BCG is due to the loss of Region of Difference 1 (RD1) associated with the loss of the virulence factor of the secreted immunodominant antigen of 6 kDa (ESAT-6) [8].

**Figure 1.** Bacillus Calmette-Guérin (BCG) vaccination coverage by country. Data from the World Health Organization about the BCG coverage in each country [10,11]. A total of 113 countries reported coverage of at least 90% [2].

In 1976, WHO established the Expanded Program on Immunization (EPI) to ensure universal access of mothers and infants/children to routinely recommended infant/childhood vaccines. Initially, there were six vaccine-preventable diseases included in the EPI: TB, poliomyelitis, diphtheria, tetanus, pertussis and measles. Intradermal vaccination with BCG at birth has been included in the WHO EPI, resulting in more than four billion vaccines administered worldwide to date and approximately 200 million doses given each year. Depending on strain and manufacturer of BCG, concentration of live bacteria in vaccine vials can range between 50,000 to three million per dose [12,13]. BCG vaccination is

recommended as part of national childhood immunization programmes according to a country's TB epidemiology. In 2018, BCG vaccination was reported for 153 countries and 113 of these countries reported at least 90% BCG coverage [2,10,11] (Figure 1).

Today, BCG vaccination is recommended by the WHO in all newborn infants in countries where TB incidence is high. In countries where the incidence of TB is not high, BCG is recommended if the child is continually exposed to TB patient who does not respond to treatment and patient's separation is not possible, or when the child is continually exposed to a patient who has infectious pulmonary TB caused by MDR or XDR strains [14]. The recommendation of BCG vaccination for adults in endemic areas with high exposure to resistant TB remains controversial. Considering the potential risks of anti-TB treatment failure and complications related to BCG vaccination of immunocompromised individuals, administration of BCG could be recommended in unvaccinated individuals, tuberculin-negative or interferon gamma release assays (IGRA)-negative individuals exposed to MDR TB. Thus, trials evaluating protective efficacy of BCG in the context of exposure to MDR TB in adults are needed [14].

Commercial BCG is not a unique product as different formulations exist in terms of BCG strain, composition and/or dosage. Currently, six BCG strains are used worldwide in international immunization programs: BCG Pasteur 1173 P2, BCG Danish1331, BCG Glaxo 107, BCG Tokyo 172-1, BCG Russia-I and BCG Brazil [15]. These different BCG strains in use [16] exhibit different characteristics of attenuation and protection in animal models [17]. The genomic analysis of BCG sub-strains shows multiple differences, including other deletions other than RD1 that contribute to phenotypic variations between them [17], with clear attenuation differences but without being shown to contribute to differences in their efficacy [5]. A recent systematic head-to-head comparison study of BCG vaccine formulations demonstrated marked variations in content of viable mycobacteria which correlated with age-specific induction of cytokines in vitro [13]. According to the same study, these differences in viability possibly contribute to an observed formulation-dependent activation of innate and adaptive immune responses that could account for the variable protection observed in clinic [13].
