Treatment outcomes in people with diabetes and multidrug-resistant tuberculosis

Longitudinal glucose data from Stage 2 of the STREAM trial enabled a preliminary assessment of factors associated with glucose control in individuals with multidrug-resistant tuberculosis (MDR-TB). This post-hoc analysis included participants from both Stage 1 and Stage 2 of STREAM, using the modified intention-to-treat (mITT) and safety populations. The mITT group comprised participants who were randomized and culture-positive for M. tuberculosis at baseline, excluding those whose isolates were later found to be rifampicin-susceptible or resistant to both fluoroquinolones and aminoglycosides. The safety population included all participants who received at least one dose of the trial drug. Participants were categorized into the diabetes mellitus (DM) group based on self-report, random blood glucose ≥200 mg/dL at baseline, or reported use of diabetes medications. Those with undetermined DM status were excluded. Of 1,012 total participants, 896 were included in the analysis; 84 (9.4%) had DM.

The DM group differed significantly from the non-DM group in several baseline characteristics. They had a markedly higher mean random blood glucose at baseline (263.2 mg/dL vs. 95.5 mg/dL, p<0.001), were generally older (p<0.001), and had higher BMI (p<0.001), with nearly twice the proportion classified as overweight. Men were more likely to have diabetes than women (p=0.049). In terms of trial outcomes, the proportion of participants with an unfavourable outcome was similar between DM and non-DM groups across both trial stages. Additionally, the incidence of failure or recurrence (FoR) was comparable between the DM (8.3%) and non-DM (5.9%) groups, with an adjusted hazard ratio of 1.38 (95% CI: 0.51–3.73), indicating no statistically significant difference.

However, serious adverse events (SAEs), excluding deaths, were significantly more common in the DM group (41% vs. 22%, p<0.001). Notably, 20.6% of SAEs in the DM group were related to surgical/medical procedures or endocrine issues, compared to under 3% in the non-DM group. There were also differences in SAEs by treatment regimen: while rates were similar in the long regimen (32% vs. 31%), the DM group experienced higher SAE rates in the six-month (62% vs. 17%, p<0.001), oral (29% vs. 15%, p=0.123), and short (44% vs. 23%, p=0.002) regimens. These findings suggest that while diabetes did not significantly affect efficacy outcomes, it was associated with a substantially higher risk of adverse events, highlighting the importance of tailored clinical management in this subgroup.

Source: Gurumurthy, M., Gopalan, N., Patel, L., Davis, A., Srinivasalu, V.A., Rajaram, S., Goodall, R., Bronson, G. and STREAM Trial Collaboration, 2025. Treatment outcomes in people with diabetes and multidrug-resistant tuberculosis (MDR TB) enrolled in the STREAM clinical trial. PLOS Global Public Health, 5(4), p.e0004259.