Tuberculosis (TB) is the 2nd most deadly infectious disease, and its implication for health and economy is increasingly recognized in different countries. In this PhD study, we characterized Mycobacterium tuberculosis complex (MTBc) isolates collected from TB patients in Ethiopia and Niger, two countries in East and West Africa that carry a high burden of TB. In addition, we tested the efficiency of current diagnostic practices for differentiation of MTBc lineages (Ls) and exclusion of drug-resistant TB. In Ethiopia, five MTBc lineages (L1–L4 and L7) were identified from newly recruited pulmonary and lymph node TB patients, with predominance of East-African-Indian (L3) and Euro-American (L4) strains, but a lower than expected prevalence of Ethiopia L7 isolates. The genotypes were similarly distributed between pulmonary and lymph node TB, and all lineages were equally isolated on culture medium and recognized as MTBc by the rapid MPT64 assay. Additionally, analysis of rifampicin-/multidrug-resistant (RR/MDR) TB isolates in Northwest Ethiopia showed four MTBc lineages (L1, L3, L4, and L7), with an expected high rate of the most common rpoB_S450L and katG_S315T mutations causing MDR, but a low proportion of gyrA and rrs mutations causing resistance to second-line drugs. Characterization of RR/MDR-TB isolates from Niger revealed the six major MTBc lineages, L1-L6, with a predominance of the L4 “Ghana” and “Cameroon” families. We noticed a significantly faster conversion to smear-negative microscopy results during treatment and a more likely favorable outcome for patients infected by “Cameroon” isolates (log-rank p < 0.01). Other risk factors such as gender, age, HIV status, body mass index, and treatment regimen were not associated with sputum smear- and culture- conversion. As a supplemental study, we incorporated retrospective rapid molecular testing (MTBDRsl) results from the STREAM stage 1 central database to assess its negative predictive value (NPV) directly from sputum samples compared to phenotypic drug-susceptibility testing. We found 9 (2.3%) fluoroquinolones (FQ)- and 5 (1.3%) second-line injectables (SLI)- discordant phenotypic/genotypic drug-susceptibility testing results. The NPV of MTBDRsl for exclusion of FQ resistance was 99.2% and for SLI resistance 100%. In conclusion, our studies provided great insights into the population structure of the MTBc in Ethiopia and Niger, demonstrating that treatment response varied by lineage. Finally, MTBDRsl is suitable for excluding resistance to second-line agents from clinical samples at the start of RR/MDR-TB treatment, including testing of paucibacillary specimens.
|Qualification||Doctor of Philosophy|
|Place of Publication||Antwerp, Belgium|
|Publication status||Published - 2022|