Association between Bacillus Calmette-Guerin vaccination and lymphoma: a population-based birth cohort study

Background Most risk factors for lymphoma identified so far relate to immunosuppression, but its aetiology remains unclear. Objectives We investigated whether Bacillus Calmette-Guerin (BCG) vaccination is associated with lymphoma, overall and separately for Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). Methods A cohort of 400 611 subjects born in the province of Quebec, Canada, between 1970 and 1974 was used. Information on BCG vaccination was extracted from the Quebec BCG Vaccination Registry. Lymphomas cases were individuals who had >= 2 health encounters, medical visits or hospitalizations, for lymphoma within 2 months or who were identified through the Quebec Tumor Registry. Cox proportional hazard regression was used to estimate hazard ratios (HRs) and 95% confidence interval (CI), adjusting for potential confounders. Results A total of 178 335 (46.0%) subjects were BCG-vaccinated, and 1478 (0.38%) cases of lymphomas were ascertained. Amongst them, 922 were identified as NHL and 421 as HL. After adjustment, no association was observed between BCG vaccination and either lymphoma (any type) (HR = 1.03, 95% CI: 0.96-1.11) or NHL (HR = 0.99, 95% CI: 0.86-1.13). For HL, nonproportional hazards were observed. Before the age of 18, the risk of HL was elevated amongst vaccinated individuals (HR = 2.26, 95% CI: 1.39-3.69). However after 18 years of age, no association was found (HR = 0.93, 95% CI: 0.75-1.15). Conclusion Bacillus Calmette-Guerin vaccination may increase the risk of HL before 18 years of age, but residual confounding cannot entirely be excluded. Given the benefits of BCG vaccination, these results need to be reproduced in other populations before firm conclusions can be drawn.