Mycobacterium tuberculosis (M.tb.) is the main agent of tuberculosis (TB); the bacterium is airborne and transmittable between individuals with TB disease, it is even transmittable during the initial period of treatment. In 2019 the World Health Organization (WHO) reported 10 million TB cases, of whom 12% (<15 years) were children, and 1.2 million deaths due to TB, with most TB deaths occurring in low- and middle-income countries [1]. Approximately 1.7 billion people are infected with M.tb. [1] and hence at risk of developing active TB disease. Although the burden of TB disease was falling globally, that is, in all WHO regions and in most countries [1], COVID -19 pandemic has pushed back the global tuberculosis control progress resulting by 20% drop in diagnosis and treatment worldwide [2]. The risk of M.tb. infection remains much higher and is greater than the risk of developing active TB disease, with an estimated 5%–15% lifetime risk of progression from latent TB infection (LTBI) to TB disease. The largest TB risk is among children, newly infected individuals, those with a compromised immune system (e.g. due to HIV, malnutrition, diabetes, or ageing) [3], and among those exposed to immunomodulating treatment [4].

An improved and safer preventive TB vaccine is urgently needed to combat the disease alongside current tools for TB prevention, diagnostics, and treatment. The 100-year-old Bacille Calmette-Guérin (BCG) vaccine, unfortunately, seems to have had a minimal effect on global TB incidence despite its widespread use [5]. While BCG is effective in preventing severe forms of TB in infants and young children it has shown limited protection against pulmonary TB in adolescents and adults, the populations primarily responsible for transmitting the disease [6]. A vaccine targeting adolescents and young adults not yet infected could prevent infection and potentially progression to disease in those infected maximizing TB control [7–9] thereby interrupting transmission.

LTBI incidence rate have been demonstrated to be high among adolescents in South Africa [10–13], and interferon-gamma (IFN-γ) release assays (IGRAs) conversion has been shown to be a clear risk factor for TB disease in this age group [14].

Adolescents are therefore a suitable population for prevention-of-infection (POI) trials [15, 16]. In Mwanza, Tanzania, a region of interest for TB vaccine studies the LTBI rates are largely unknown for this target population despite the region reporting a high number of cases of TB disease yearly [17]. Although previously reported LTBI prevalence data among adults in the region suggest a 4% annual conversion rate [18], a more accurate LTBI incidence needs to be determined among the adolescents in the same region to appropriately inform both the international and local community to design and power TB vaccine trials with infection endpoints.

LTBI is defined by detection of cell-mediated immune response towards M.tb. antigens either by in vivo in skin tests or in vitro using IGRA [19]. Early Secretory Antigenic Target-6 (ESAT-6) is a component of the H56 fusion protein used in the H56: IC31 vaccine [20], a novel TB subunit vaccine, but ESAT-6 is also one of the antigens used to stimulate IFN-γ production in the QuantiFERON-TB Gold In-Tube (QFT-GIT) IGRA assay [21]. Hence, successful vaccination with H56 leads to false-positive QFT which disqualifies ESAT-6-based vaccines for POI trials using QFT to detect M.tb. infection. To address this issue we have developed an ESAT-6 free IGRA, where ESAT-6 is substituted by fragments of Rv3615c and other M.tb. specific and highly recognized antigens [21]. The aim of this study was to determine if Tanzanian adolescents would be a good population for POI trial by assessing the prevalence and annual risk of tuberculosis infection (ARTI) of TB determined by QFT assays in two groups of 12-and 16-year-old adolescents in Mwanza, and also to evaluate the ESAT-6 free IGRA.

A total of 193 adolescent were enrolled from after being determined to be eligible for this study. Of those enrolled, 96 were in the 12 year old cohort (5 participants in the 12 year old cohort were 11 years old), and 97 were in the 16 year old cohort (12 participants in the 16 year old cohort were 15 years old). There was a higher proportion of females than males in the 12 year old cohort as compared to the 16 year old cohort where the proportion of females and males was similar Table 1. About 65% of the participants reported staying with their biological parents. Seven and 9% of the 12 year old and the 16 year cohort, respectively reported exposure to pulmonary TB within 12 months of the time of enrolment. Further demographic data is presented in Table 1.

Overall estimates of LTBI prevalence were 19.2% [95%CI, 14.1; 25.2] and 18.6% [95%CI, 13.6; 24.6] as measured by QFT-GIT IGRA and ESAT-6 Free IGRA, respectively. The 16-years-old cohort had a higher LTBI prevalence (23.7% [95%CI, 16.1; 32.9]) than the 12-year-old cohort (14.6% [95%CI, 8.6; 22.7]) as measured by QFT-GIT IGRA; a similar trend was observed as measured by ESAT-6 Free IGRA, with 24.7% [95%CI, 16.9; 34.0] among the 16-years-olds compared to 12.5% [95%CI, 7.0; 20.3] among the 12-year-olds. The LTBI prevalence by gender across both cohorts is summarised in Table 2, and the estimation of LTBI prevalence for ages 13, 14, and 15 is presented in Fig 1.

The ARTI among participants as measured by QFT-GIT IGRA was 1.25% in the 12-year-old cohort and 1.63% in the 16 year old cohort. For the ESAT-6 Free IGRA it was 1.06% in the 12 year old cohort and 1.71% in the 16 year old cohort Table 3. The extrapolated ARTI for 13-, 14-, and 15 year olds was 2.7%, 2.8%, and 2.9% respectively for QFT-GIT IGRA. For the ESAT-6 Free IGRA the estimates were 3.6%, 3.7%, and 3.9% for 13-, 14-, and 15 year olds respectively.

Among all enrolled participants, the force of infection was 2.7% [95%CI, 0; 6.2] in the 12 year old cohort and 3.0% [95%CI, 0; 6.9] in the 16-year-old cohort by QFT-GIT IGRA, and it was 3.5% [95%CI, 0; 7.5] in the 12-year-old cohort and 4.7% [95%CI, 0; 8.7] in the 16-year-old cohort by ESAT-6 Free IGRA. The estimation of force of infection at ages 13, 14, and 15 is shown in Fig 2. Of all enrolled participants, 97.4% had concordant results for QFT-GIT IGRA and ESAT-6 Free IGRA (p = 0.65) (Table 4).

We conducted an epidemiological study in Mwanza, Tanzania to estimate the rate of M.tb.infection to facilitate the design of POI studies in this region. We evaluated the prevalence of LTBI using a commercially available IGRA, QFT-GIT, as its utility in settings with a high TB burden and a high risk of infection has been demonstrated. [12, 13]. We compared the performance of QFT-GIT to the ESAT-6 free IGRA assay to explore the force of infection, ARTI, and prevalence of LTBI.

We found that there is a high prevalence of LTBI and corresponding ARTI suggesting ongoing transmission in this population. There was a high concordance between the QFT-GIT and the novel ESAT-6 free IGRA assays in the estimates of M.tb. infection which was similar both to a study by Ruhwald M. et al., where the agreement between ESAT-6 free IGRA using IFN-γ and QFT was high (~80%) [21], and to a study by Nemes E. et al., who reported a significant correlation with 91% concordance between the tests [23]. The concordance of the ESAT-6 free IGRA [21] with QFT suggests it might be an alternative to the regular IFN-γ release assay in TB vaccine trials where antigens in the vaccine are similar to those in the QFT assay. Although IGRA assays are the currently accepted tests to assess the efficacy of TB vaccine candidates and to estimate LTBI, the diagnosis of LTBI remains challenging, as there is no gold standard for LTBI determination [26, 27]. The high incidence of M.tb. infection in this population, as estimated by both tests, was observed to increase with age. These findings are similar to those reported in the age and sex model study done in Zambian and South African communities [28] and by Nduba V. et al. in western Kenya [29]. Our findings, plus those reported elsewhere [10–13, 30, 31], indicate that M.tb. transmission is more intense among adolescents than among young children. The high ARTI and corresponding force of infection observed by age in this study also confirm this trend. In this study, the ARTI, an epidemiological index that measures the extent of M.tb. transmission in the community was observed to range from 1.3% to 2.9%, similar to the force of TB infection estimates (proportion of susceptible individuals that have become infected with M.tb. in specified period) that ranges from 2.6% to 3.0%. The observed ARTI and force-of-infection estimates support a higher transmission of M.tb. infection in this region than was reported before [32–34] and are consistent with findings in South Africa, where an ARTI of 2%–4% was reported among adolescents [35].

A high prevalence of LTBI among adolescents in this population was not a surprise. Our previous study among adults (18 years old and above) within the same community population reported a high rate of LTBI among household contacts of TB patients as compared to neighbourhood controls [18]. The history of TB contact, the mixed socio-economic markers, and high social density in this population can explain the high rate of M.tb. infection observed.

Given that the prevalence of LTBI among adolescents in Mwanza region was unknown, despite the fact that the region reports the highest rate of TB disease and number of TB case notifications yearly [17], our findings provide good estimates of force of infection. Such results will help local and international investigators planning to conduct POI studies within the region. Our findings will both help to accurately determine the sample size of such trials and assist the TB control programmes in measuring the effectiveness of the programme implemented. In addition, these data suggest that a vaccine targeting adolescents and young adults could have a role in preventing M.tb. infection and thus avoid latent infection before they reach adulthood.

Although the study has limited numbers it does provide a valuable method to determine prevalence and estimated ARTI and force of infection in other regions or countries.

We assessed the prevalence and ARTI among adolescents aged 12 and 16 years using two different IGRAs, the QFT-GIT and a novel ESAT-6 free assays, and found a high prevalence of LTBI and estimates of ARTI and force of infection. There was a clear indication that adolescents are a relevant population to assess the effect of new TB vaccine candidates targeting uninfected individuals as our results indicate ongoing transmission between 12 and 16 years of age. This data indicates that adolescents should be considered in global TB vaccine research efforts. Moreover, QFT-GIT and ESAT-6 free assays were reasonably similar in estimating LTBI indicating that the ESAT-6 free assay may be a good alternative if an investigational vaccine contains antigens presented in the QFT-GIT assay.