Evaluating Risk of Tuberculosis Infection and Disease among Adolescents who are HIV-exposed Uninfected in Botswana

Abstract

This Botswana-based study evaluated for prevalence of tuberculosis (TB) infection and disease among 50 adolescents, including 25 HIV-exposed uninfected (HEU) and 25 HIV-unexposed uninfected (HUU). We found that 12% of adolescents HEU had a positive QuantiFERON-TB Gold Plus test while no adolescent HUU tested positive. This exploratory study found that adolescents HEU may be at increased risk for TB infection and disease, with larger studies needed.

INTRODUCTION

Each year, 1.8 million adolescents and young adults are diagnosed with TB disease globally.¹ Adolescents are at high risk of progression from TB infection to disease, due to various biological factors, including age and sex, and social factors.² Adolescents with TB disease disproportionately contribute to community TB transmission, compared to other age groups, and are at risk for poor outcomes and adverse lifelong sequelae.¹ Despite this, gaps persist in adolescent TB services.¹

In Botswana, a country with high HIV and tuberculosis (TB) prevalence, adolescent TB infection and disease prevalence has not been well-described.³ Importantly, adolescents HIV-exposed uninfected (HEU) may be a key risk group for TB infection, progression to disease, and transmission, given increased risk of TB exposure in households affected by HIV, in addition to biological and socioenvironmental risk factors. We describe prevalence of TB infection and disease by HIV exposure status among adolescents participating in a sub-study of the Botswana-based FLOURISH study.

METHODS

Study Population

The study was conducted at Botswana Harvard Health Partnership (BHP) in Gaborone, Botswana from February to October 2023. Participants from the FLOURISH Study [R61HD103099] were recruited. The FLOURISH study recruited mother-child dyads who had participated in previous BHP birth cohort studies.⁴ Therefore, each adolescent’s gestational HIV exposure status was known. Adolescents provided assent with parental consent to participate in the TB sub-study.

Eligibility Criteria

Eligibility for adolescent participants in the TB sub-study included current participation in the FLOURISH study, a citizen of Botswana, a negative HIV test at time of FLOURISH enrollment, and assent for TB infection testing. The age for eligibility was 10–17 years, which was based on the FLOURISH study, which enrolled adolescent participants less than 18 years old.

Data Collection

Study procedures included an enrollment visit, with data collection on demographics, history of TB infection/disease, and TB household contacts within the last year. For evaluation of TB disease, TB symptom screening assessed for cough, fever, night sweats, and/or weight loss ≥ two weeks, following World Health Organization (WHO) and Botswana guidelines.^5,6 For evaluation of TB infection, interferon gamma release assay (IGRA) testing was performed at enrollment using the QuantiFERON-TB Gold Plus (QFT-Plus) test, according to the manufacturer’s instructions (Cellestis Ltd., Victoria, Australia). Test results were interpreted using the manufacturer’s algorithm, classifying outcomes as positive, negative, or indeterminate. If a result was indeterminate, the participant returned for a study visit, with blood again drawn to perform repeat QFT-Plus testing. Participants with a positive TB symptom screen, a household contact with TB within the last year, and/or positive QFT-Plus testing were referred to local government health clinics for TB evaluation. For referred adolescents, a follow up visit occurred 2 months after enrollment to assess TB referral outcomes, including diagnostic testing performed and treatment provided.

Statistical Analysis

Adolescent characteristics were summarized using medians and interquartile ranges (IQR) for continuous variables and counts with percentages for categorical variables. TB infection/disease prevalence was reported by gestational HIV exposure status. A Fisher’s exact test was used to compare TB prevalence and binary demographic/clinical factors of interest between exposure groups, with a p-value of less than 0.05 considered statistically significant. Analyses were completed in R Version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria).

Ethical Considerations

The Botswana Health Research Development Committee, Massachusetts General Hospital, and Boston Children’s Hospital provided Institution Review Board study approval. Weill Cornell Medicine approved de-identified data analysis.

RESULTS

From February to August 2023, 50 adolescents were enrolled in the TB sub-study, 50% of whom were HEU (Table 1). Median age was 11 years (IQR 11, 12). Age did not differ by HIV gestational exposure groups. Males accounted for 40% of adolescents HEU and 52% of adolescents HUU (p=0.57). No participant had a previous reported history of TB infection or disease. None reported a household TB contact in the last year.

Table 1:

Adolescent Participant Characteristics

Characteristics	Adolescents HEU (n=25)	Adolescents HUU (n=25)
Age, median [IQR], years	11 [11,15]	11 [11,11]
Male, n (%)	10 (40%)	13 (52%)
Adolescent Antiretroviral Gestational Exposure, n (%)
a) PI-based maternal ART regimen	15 (60%)
b) NNRTI-based maternal ART regimen	6 (24%)	n/a
c) Zidovudine	4 (16%)
BMI, median, kg/m2	17.5 [15.4, 18.6]	16.0 [14.7, 17.7]
Number of persons in household, median [IQR]	7 [5, 8]	6 [3, 9]
Positive TB symptom screen, n (%)	3 (12%)	1 (4%)
a) Cough ≥ 2 weeks	1	1
b) Cough ≥ 2 weeks, weight loss	1
c) Cough ≥ 2 weeks, fever, night sweats, weight loss	1
Household contact with TB, n (%)	0 (0%)	0 (0%)
Positive QFT-Plus testing, n (%)	3 (12%)	0 (0%)
Clinical TB diagnosis	1 (4%)	0 (0%)

Abbreviations: HEU -HIV-exposed uninfected; HUU - HIV-unexposed uninfected; PI- protease inhibitor; ART- antiretroviral treatment; NNRTI- non-nucleoside reverse transcriptase inhibitor; IQR- interquartile range; BMI - body mass index; QFT-Plus -QuantiFERON-TB Gold Plus

Four adolescents reported TB symptoms (3 adolescents HEU, 1 HUU) (Table 1). Three adolescents HEU (12%), including two females and one male, had a positive QFT-Plus test, compared to none of the adolescents HUU (0%) (p=0.23). One adolescent HUU with an initial indeterminate QFT-Plus result had negative result upon repeat testing.

Follow up visits were conducted with six participants, who were referred to government health clinics for further evaluation in the setting of positive TB symptom screening and/or QFT-Plus testing. All six participants reported that they attended a local clinic for TB evaluation. There was variation in TB diagnostic testing performed at the government clinics as reported by participants, with challenges with sputum testing and transport of specimens to the national lab. One adolescent with a positive TB symptom screen and QFT-Plus test was clinically diagnosed with TB and initiated on TB treatment. Of the remaining five adolescents with symptoms or positive QFT-Plus, none were started on TB treatment or TB preventative treatment (TPT).

DISCUSSION

In this Botswana-based pilot study, there were differences observed in prevalence of TB infection and disease in adolescents by HIV exposure status. While this finding was not statistically significant, there was a trend of increased prevalence of TB infection/disease for adolescents HEU (12%) compared to adolescents HUU (0%). Of note, one adolescent HEU had both a positive TB symptom screen and QFT-Plus test and was clinically diagnosed with TB disease. Based on this preliminary data, adolescents HEU may be a high-risk group to prioritize for TB screening and prevention strategies.

Studies from high prevalence HIV/TB countries have examined the prevalence of TB infection using IGRA and tuberculin skin test (TST) testing in adolescents, without specifically stratifying by HIV status, as well-documented HIV-exposure status in adolescence is rare. Our study used prospectively collected maternal data and HIV testing results to distinguish adolescents HEU from HUU with high certainty per the DECIPHER criteria,⁷ based upon the original birth cohort study in which the adolescents had participated. Prior studies have reported mixed findings on risk of TB infection using IGRA and TST testing in young children HEU, but this has not been examined in adolescents HEU.⁸ Given that over 25% of adolescents in Botswana currently are HEU, further studies are warranted.

There may be biological factors and epidemiologic exposures that explain increased prevalence of TB infection in adolescents HEU. Several studies have shown that there was greater immune dysfunction and higher levels of inflammation in infancy and childhood among individuals HEU who are now in their teens.⁹ Whether these immunological differences persist into adolescence has not been well studied. Other exposures that may differ by HIV exposure status include higher risk of exposure to TB or differences in socioeconomic factors in households affected by HIV.

For adolescents referred to clinics for TB evaluation, there were barriers to TB diagnosis and prevention. Diagnostic barriers included lack of sputum production and lack of availability of transport for specimens at the clinic. For adolescents with positive QFT-Plus testing without TB disease, TPT was not offered, as this is not standard of care in Botswana for adolescents without HIV. Currently Botswana guidelines recommend TPT for people living with HIV and children under five years who are contacts of bacteriologically confirmed individuals with TB.^6,10 WHO guidelines recommend TPT for individuals with risk factors who have positive TB infection testing, after an evaluation for TB disease.⁵ Future quantitative and qualitative studies are needed to understand whether adolescents HEU are a high risk group for TB, who would benefit from targeted screening and preventative strategies, including TPT and novel TB vaccines.

This study had several strengths, but it also had limitations. The ability to perform QFT-Plus testing in this study is a strength, given the strong laboratory infrastructure. However, we recognize that this was an exploratory study with a small sample size, which was not powered to detect statistically significant differences. We were unable to evaluate for other biological or socioenvironmental factors associated with adolescents HEU in this setting. Additionally, the median age of participants was 11 years, which reflects early adolescence and may underestimate prevalence of TB infection in older adolescents. We also recognize that timing of TB infection in this cohort is unknown, as seroconversion may have occurred prior to adolescence.

CONCLUSION

In this small exploratory study, there appeared to be a higher prevalence of TB infection and disease among adolescents HEU compared to adolescents HUU. Larger studies are needed to determine prevalence and risk factors for TB infection in adolescents by HIV exposure status to guide TB screening and prevention interventions for this population.