Table 1.
Characteristics of randomized controlled trials investigating the effect of the BCG vaccine on T1DM.
| Study | Country | Participants | Intervention | BCG type | Outcome measures | Main findings |
|---|---|---|---|---|---|---|
| Pozzilli et al. 1997 [20] | Italy | 72 patients within 4 weeks of IDDM; mean age, 14.5 years | Intervention (n = 36): one dose of BCG with NCT 25 mg/kg; control (n = 36): NCT 25 mg/kg | Freeze-dried BCG vaccine (Berna Institute, base) | HbA1c, C-peptide, insulin dosage, and clinical remission; 1-year follow-up | No significant differences in insulin dosages or C-peptide (ng/ml) (0.8/0.9 vs. 0.93/1.0 preintervention/postintervention) and HbA1c (%) (9.7/6.9 vs. 9.8/6.4 preintervention/postintervention) level between BCG and control groups. BCG did not induce long-term clinical remission in IDDM patients. |
|
| ||||||
| Elliott et al. 1998 [21] | Canada | 26 patients within 1 year of IDDM; mean age, 13.1 years | Intervention (n = 13): one dose of BCG; control (n = 13): placebo | BCG vaccine (Connaught, Toronto, Ontario, Canada) | HbA1c and fasting/stimulated C-peptide, and insulin dosage; 1.5-year follow-up | HbA1c, mean basal (ng/ml) (0.81/0.33 vs. 0.90/0.51 preintervention/postintervention), and stimulated (ng/ml) (1.35/0.51 vs. 1.74/0.84 preintervention/postintervention) C-peptide level, and insulin dosage did not differ significantly between BCG and control groups (p = 0.10–1.00). |
|
| ||||||
| Allen et al. 1999 [22] | U.S. | 94 children with a diagnosis of new-onset IDDM; mean age, 10.5 years | Intervention (n = 47): one dose of BCG; control (n = 47): placebo | TICE BCG (PerImmune, Rockville, MD) | HbA1c, fasting/stimulated glucose and C-peptide, insulin dosage, and clinical remission; 2-year follow-up | No differences in HbA1c (%) (8.8/8.4 vs. 9.2/8.5 preintervention/postintervention), insulin requirement, and basal or stimulated C-peptide level between both groups. The trend toward a more rapid decrease in C-peptide level (p = 0.11) in the BCG group (p < 0.05). |
|
| ||||||
| Kühtreiber et al. 2018 [23] | U.S. | 6 subjects with long-term diabetes; mean age, 42 years | Intervention (n = 3): two doses of BCG administered 4 weeks apart; control (n = 3): placebo | Lyophilized BCG (TheraCysH, Sanofi-Pasteur, Toronto, Ontario, Canada) | HbA1c and stimulated C-peptide; 8-year follow-up | After year 3, BCG lowered HbA1c (%) (7.36/6.65 vs. 7.1/7.22 preintervention/postintervention between intervention and control groups) to near normal levels for the next 5 years (p < 0.001 at year 8). No significant return of C-peptide level in the BCG group |
BCG: Bacillus Calmette–Guerin; IDDM: insulin-dependent diabetes mellitus; NCT: nicotinamide; HbA1c: glycated hemoglobin.