Mycobacterium tuberculosis Beijing Lineage and Risk for Tuberculosis in Child Household Contacts, Peru

Few studies have prospectively compared the relative transmissibility and propensity to cause disease of Mycobacterium tuberculosis Beijing strains with other human-adapted strains of the M. tuberculosis complex. We assessed the effect of Beijing strains on the risk for M. tuberculosis infection and disease progression in 9,151 household contacts of 2,223 culture-positive pulmonary tuberculosis (TB) patients in Lima, Peru. Child contacts exposed to Beijing strains were more likely than child contacts exposed to non-Beijing strains to be infected at baseline, by 12 months of follow-up, and during follow-up. We noted an increased but nonsignificant tendency for child contacts to develop TB. Beijing strains were not associated with TB in adult contacts. These findings suggest that Beijing strains are more transmissible in children than are non-Beijing strains.

The study design and methods were previously described in detail (22). In brief, we conducted a prospective cohort study of household contacts of pulmonary TB patients in Lima, Peru. The study area comprised 20 districts inhabited by ≈3.3 million residents living in urban areas and in periurban, informal shantytown settlements.

Mycobacterium tuberculosis Beijing Lineage and Risk for Tuberculosis in Child Household Contacts, Peru
During September 2009-August 2012, we identified patients >15 years of age who had received a diagnosis of clinically presumptive pulmonary TB at any of 106 participating health centers. We confirmed the microbiological status of their disease with either a positive sputum smear or mycobacterial culture. We also recorded the index patient's sociodemographic data; baseline smear status; presence or absence of cavitary disease; tobacco and alcohol use; HIV status; time from symptom onset until initiation of treatment; and, for patients, with drug-resistant TB, the time from TB diagnosis until start of an effective treatment (i.e., a drug regimen deemed appropriate for the index patient's drug-resistant profile).
Within 2 weeks after index patient diagnosis, we enrolled all consenting household contacts. We assessed baseline M. tuberculosis infection status with the tuberculin skin test (TST) in household contacts with no history of a positive TST or TB. Household contacts who had signs or symptoms of TB underwent clinical evaluation and, if indicated, initiated treatment under Peru's National Tuberculosis Program guidelines (23). We offered HIV testing to all study participants. In accordance with Peru's National Tuberculosis Program guidelines, isoniazid preventive therapy (IPT) was offered to household contacts <19 years of age and persons with specified concurrent conditions. At the time of household contact enrollment, we collected age, sex, sociodemographic, and occupational information; height and weight; alcohol and tobacco use information; HIV status; self-reported diabetes mellitus; history and presence of M. bovis BCG vaccination scars; use of IPT; history of TB; and housing features. We repeated the TST 6 and 12 months after initial evaluation among household contacts who had previously tested negative; 2, 6, and 12 months after initial evaluation, we reevaluated them for pulmonary and extrapulmonary TB. We also accessed the medical records of health centers in the catchment areas to identify household contacts whose TB was diagnosed at a health center during the follow-up period.

Whole-Genome Sequencing
Among 3,027 index patients with culture-positive isolates at baseline, 2,143 had isolates that also underwent whole-genome sequencing (WGS) using the Illumina HiSeq 4000 platform (Illumina, https:// www.illumina.com) with a read length of 100-150 bp and >50-fold coverage. The raw sequence data were trimmed using the sickle package (24) and mapped to the H37Rv reference genome using the BWA-MEM algorithm (25). We used a coverage-based approach, specifically SAMtools (default settings) (26) and pilon (27), to identify single-nucleotide polymorphisms across the whole genome. We assigned a call as missing if the valid depth of coverage at a specific site was <8 reads, if the mean read-mapping quality at the site did not reach 9, or if none of the alternative alleles accounted for >85% of the valid coverage.

Lineage Identification of M. tuberculosis Strains
Among index patients for whom WGS data were available, we identified M. tuberculosis lineages and sublineages on the basis of a previously published barcode that differentiates the groups on the basis of singlenucleotide polymorphisms (28). For index patients without available WGS data, we genotyped their M. tuberculosis isolates using 24-locus mycobacterial interspersed repetitive units-variable-number tandem repeats (MIRU-VNTR) (29). We used the best-match algorithm from the MIRU-VNTRplus online database (http://www.miru-vntrplus.org) to identify the lineages of the typed samples by cross-referencing them with reference lineages stored on the database (30).

Outcomes
We considered 4 distinct outcomes: M. tuberculosis infection at baseline, time to TST conversion during the 12 months of follow-up, M. tuberculosis infection by

Statistical Analysis
We made an a priori decision to analyze child household contacts (<15 years) and adult household contacts (>15 years) separately for several reasons. First, we considered child contacts to be a more representative age group for household-based TB transmission than their more mobile, socially active adult counterparts. Second, several previous studies suggest that Beijing strains more commonly affect children than adults (18)(19)(20). We used complete data analyses in all multivariate adjustment models and conducted all statistical analyses in R version 3.5.1 (https://www.r-project.org).   (7) *Numbers might not add to total because of missing data.

TB at Baseline and by 12 Months of Follow-Up
We used a modified Poisson generalized estimating equation to measure the association between exposure to index patients infected with Beijing strains and the likelihood of infection at baseline and by 12-months of follow-up. To account for correlation within households, we specified an exchangeable working correlation structure for observations within the same household and obtained empirical SE estimates for robust inference. We constructed univariate and multivariate models to compare the effect of exposure to Beijing and non-Beijing strains on the risk for TB infection at baseline and at 12 months. In the multivariate models, we included covariates from index patients (age, drug resistance profile, and HIV status) and their household contacts (age, diabetes status, BCG vaccination status, socioeconomic status, and nutritional status). We conducted sensitivity analyses in which we excluded household contacts who reported a history of TB, a previous positive TST result within 6 months after study enrollment, or a history of receiving TB treatment.

Time to TST Conversion and to Incident Disease
We assessed the time to TST conversion among household contacts who were uninfected at baseline by defining the date of infection as the midpoint between the date of enrollment and the date of a first positive TST result or TB diagnosis. We excluded from analysis household contacts who remained TST negative at the date of their last TST result. For the time to TB analysis, we excluded household contacts with co-prevalent TB and excluded from analysis household contacts in whom TB had not been diagnosed by the time of their death or at the end of the study. We first plotted Kaplan-Meier survival curves to visually evaluate the association between the index patient's M. tuberculosis genotype and the 2 time-to-event outcomes. We then used Cox frailty proportional hazards model to evaluate risk factors for time to TST conversion and time to incident disease while accounting for clustering within households. We constructed univariate and multivariate models to compare the effect of exposure to Beijing and non-Beijing strains on the risk for TST conversion and incident disease during follow-up. In the multivariate models, we included covariates from index patients (age, drug resistance profile, and HIV status) and their household contacts (age, diabetes status, BCG vaccination status, socioeconomic status, and nutritional status). We further adjusted for the use of IPT and TB history in the multivariate model of the time to incident disease analyses. We verified the proportional hazards assumptions for each covariate by including an interaction term for the covariate and time and stratifying by covariates for which the assumption did not hold. We conducted 2 sensitivity analyses, 1 in which we defined TST conversion using an increment of >6-mm induration size in repeat TST measurements during follow-up and 1 in which we restricted the analysis to household contacts in whom active TB developed during follow-up and who received their diagnoses within >30 days after index patient enrollment.

Protective Efficacy of BCG Vaccination
We also considered the possibility that index patient M. tuberculosis lineage might modify the protective

M. tuberculosis Infection at Baseline and After 12 Months of Follow-up
Child contacts exposed to Beijing strains were more likely than those exposed to non-Beijing strains to be infected at baseline (risk ratio  (Tables 3, 4). In the sensitivity analyses, we obtained almost identical results when we excluded household contacts who reported a history of TB, a previous positive TST result within 6 months after study enrollment, or a history of TB treatment (Appendix Tables 5, 6).

TST Conversion
The TST status of children exposed to Beijing strains were more likely than that of children exposed to other strains to convert from negative to positive during 12 months of follow-up in both the univariate and   Figures 1, 2). In the sensitivity analyses, we defined TST conversion by an increment of >6 mm in TST induration size during follow-up, and the results changed by <5% (Appendix Table 7).

Progression to Active TB
After we adjusted for potential confounders, we found the HR for TB among child contacts exposed to an index patient with Beijing strains compared with non-Beijing strains to be 1.45 (95% CI 0.77-2.7). We found no evidence that exposure to Beijing strains affected the risk for incident TB in adult contacts (aHR 1.06 [95% CI 0.64-1.77]) (Table 6; Figures 3, 4). The results persisted when we restricted the analysis to household contacts in whom TB was diagnosed within >30 days after index patient diagnosis (Appendix Table 8).

Effect of Index Patient M. tuberculosis Lineage on the Protective Efficacy of BCG Vaccine
Although they lacked statistical significance, the aRRs for TB outcomes were higher for BCG-vaccinated children exposed to Beijing strains than for non-BCGvaccinated children exposed to Beijing strains (aRR This finding raises the possibility that BCG vaccination might be less efficacious in children exposed to Beijing strains than in those exposed to other strains     (Table 7).

Discussion
In our study, children exposed to strains of M. tuberculosis Beijing sublineage were more likely than were those exposed to strains of other lineages to become infected. In addition, children exposed to Beijing strains were more likely than children exposed to other strains to progress to TB, although the relative risk of disease progression was not statistically significant (type 1 error rate of 0.05). We did not observe these effects among adult contacts.
Our results are consistent with those from previous studies that assessed the relative transmissibility of the Beijing strain through cohort-based studies or molecular epidemiology. In a study in South Africa, Marais et al. found that children <5 years of age who had household exposure to Beijing strains were 1.5 times more likely than those exposed to other strains to be TST-positive at baseline (13). In a prospective cohort study of household contacts in The Gambia, de Jong et al. found that Beijing lineage-exposed contacts were 7 times more likely than those exposed to the M. africanum lineage to progress to TB after 2 years of follow-up but were equally likely to be TST-positive at baseline or convert at 3 months (12). In the De Jong et al. study, the proportion of TST-positive household contacts at baseline and the incidence of TST conversion and disease did not differ significantly between Beijing and other strains within M. tuberculosis sensu strictu.
Although some molecular epidemiologic studies have found that Beijing strains are more likely than other strains to form genotypic clusters, other studies have not supported this conclusion. Many of these studies assumed that patients within a chain of recent M. tuberculosis transmission will share molecular fingerprints (32)(33)(34)(35)(36) and that clustering of genotypes is therefore a proxy for transmissibility and disease progression. In China, Yang et al. reported that patients infected with Beijing strains were 1.56 times more likely than those infected with other lineages to share molecular fingerprints with other patients (15). Similar findings have been reported from Lima; in a study by Imawoto et al., 80% of Beijing strains were in clusters, although clustering among other strains was not reported (14). In Vietnam, Holt et al. reported that, although the Beijing strains accounted for 58% of the total study sample, they constituted 70% of all clustered strains (21). In Papua New Guinea, Bainomugisa et al. found that 82% of isolates with Beijing strains were found in clusters, compared with 43% of isolates with strains from other lineages (37). In China, Yang et al. reported that the proportions of clustering were 34% for Beijing and 18% for non-Beijing strains (38). However, in Canada, Langlois-Klassen et al. found that Beijing was less likely to be found in clusters than other strains (21% vs. 37%) (16), and in the Netherlands, Nebenzahl-Guimaraes et al. showed that Beijing strains were no more transmissible than strains of other lineages, after adjustment for host factors using an approach that controlled for the propensity of a strain to propagate (17).
Some animal studies have provided evidence to support the hypothesis that Beijing strains are more likely than other lineages to cause disease. Mice experimentally infected with Beijing strains not only died earlier and had higher death rates but also had more lung tissue damage than controls (39)(40)(41)(42). Some in vitro studies of macrophages have also found that the Beijing strain can downregulate the expression of pathogen recognition receptors and major histocompatibility complex class II, modify the secretion of inflammatory cytokines, and induce the necrosis of host immune cells (43)(44)(45)(46).
We considered possible explanations for the difference in the effect of the Beijing strain in children and adults. Given their mixing patterns, adults may be more likely than children to be infected within the community rather than in the household. If the household contacts in the cohort reported here had been infected by someone other than the household index patient, strain-specific exposure status might have been misclassified. Such misclassification would have been more likely in adults and would have biased the results for this group toward the null of no effect. However, another possibility is that Beijing strains might lead to earlier disease progression in younger persons with newly acquired M. tuberculosis. Several studies from Vietnam report that Beijing strains make up the highest proportion of incident cases in persons <20 years of age and that the prevalence of Beijing strains declines with increasing age (18,19). In Iran, Erie et al. similarly showed that 27% of patients <20 years of age were infected by Beijing strains and that the prevalence of Beijing strains among patients >20 years of age was 13% (20). The increasing prevalence of Beijing strains in children may be related to use of BCG vaccine. This explanation would be consistent with our finding that that the protective efficacy of BCG vaccine against Beijing strains was reduced in children but not in adults. Possible explanations for these observations include a decrease in the protective efficacy of BCG vaccination with increasing age, a shift in the administered BCG strain in Peru's recent history, or a reduction in the immunogenicity of BCG vaccine over time (47)(48)(49).
Our study had several notable limitations. First, within a high-transmission setting such as Lima, children still could have been infected outside the household. Misclassification of the lineage exposure status in children also would have led to an underestimation of the effect. Second, discrepancies between 24-locus MIRU-VNTR genotyping and WGS have been noted previously (50), and some lineages assigned on the basis of MIRU-VNTR could have been inaccurate, again leading to a misclassification error that could have underestimated the true effect of Beijing strains.
In conclusion, we found that exposure to household index patients infected with Beijing strains was associated with increased risk for TST conversion and disease in children <15 years of age but not in adults. These findings raise the possibility that genotypic variation in M. tuberculosis may have important phenotypic effects that should be further studied. In particular, it will be essential to determine whether BCG vaccination provides less protection against Beijing strains than strains from other lineages and whether the efficacy of any newly developed vaccines varies by M. tuberculosis genoype.