Summary
A medRxiv preprint reports lower HbA1c in the BCG group of a randomized, placebo-controlled phase 2 trial, with follow-up extending to eight years. The abstract also reports reduced insulin dose and no recovery of C-peptide from negligible baseline levels.
A medRxiv preprint reports that people with type 1 diabetes assigned to BCG had lower HbA1c over long-term follow-up than those assigned to placebo in a randomized, double-blind phase 2 trial. The study’s primary outcome was mean percentage change in HbA1c from baseline; the report describes follow-up of up to eight years.
The reported results
At year 5, HbA1c in the BCG group was reduced from baseline by 7.04%; at year 8, the reported reduction was 12.84%. The placebo group showed no change from baseline at either point. A repeated-measures analysis comparing changes over time between the groups was statistically significant (p=0.0017). These are reported percentage changes, not percentage-point reductions.
The preprint also reports a statistically significant reduction in average insulin dose in the BCG group (P=0.024). Continuous glucose monitoring results favoured BCG for blood sugar levels (P=0.0189). The authors report an improvement of up to 183.7% over baseline in time spent in the study’s normoglycemia range of 70–99 mg/dL (P=0.0391).
C-peptide, a marker of the body’s own insulin production, did not recover from negligible baseline levels. The authors interpret this as evidence that the reported glycemic improvement occurred without restored pancreatic insulin production. They also describe BCG as safe in the trial.
What the findings can tell us
BCG, or Bacillus Calmette–Guérin, is a vaccine against tuberculosis. The study’s rationale draws on “immune training”: lasting changes in aspects of the immune response after exposure to a stimulus. The reported blood-sugar changes and C-peptide results are compatible with the authors’ proposed non-pancreatic pathway, but the abstract does not establish the mechanism responsible.
The results are from a phase 2 clinical trial reported as a preprint, rather than a peer-reviewed publication. The posted abstract does not state how many participants were enrolled or give the BCG dosing schedule, limiting assessment of the study’s scale from this account. The findings are a research result, not by themselves an established basis for changing diabetes care.