Childhood use of antimicrobials and risk of Hodgkin lymphoma: a Danish register-based cohort study

The peculiar bimodal age distribution of Hodgkin lymphoma (HL), together with other epidemiological findings, inspired the so-called "late infection hypothesis" in the 1970s. Under this model, HL in young adults is caused by delayed infection with a relatively common agent, with HL risk increasing with age at infection. We time-dependently tallied prescriptions filled, for a broad spectrum of antimicrobials, at age 0 to 9 years for all Danish HL patients diagnosed in 1997 to 2015 at age 10 to 25 years (n = 296) and up to 10 controls for each of these, individually matched for sex and birthdate (n = 2688). Antimicrobial use was taken as a proxy for general infectious disease pressure. Analyses were also stratified by the 2 main histological subtypes: nodular sclerosis HL (NSHL) (n = 206) and mixed cellularity HL (MCHL) (n = 47). We compared antimicrobial use at ages 0 to 9 years between cases and comparators using stratified Cox regressions with repeated follow-up for a next prescription, to produce hazard ratios (HRs) of antimicrobial use according to (future) HL status. Reverse causation was mitigated by disregarding risk time <2 years before HL (pseudo)diagnosis. Analyses were adjusted for number of older and younger siblings. NSHL patients had received statistically significantly fewer antimicrobials than comparators early in life (HR0-2 years, 0.79; 95% confidence interval, 0.66-0.95), whereas patients with MCHL had received statistically significantly more antimicrobials than comparators throughout the first 10 years of life (HR0-9 years, 1.53; 95% confidence interval, 1.33-1.76). The late infection hypothesis was supported in NSHL, whereas immune dysfunction seemed more prominent in MCHL etiology.