Risk of non-Hodgkin lymphoma in relation to tricyclic antidepressant use

Abstract

Purpose: We investigated the relationship between use of tricyclic antidepressants (TCA) and risk of the malignancies that comprise non-Hodgkin lymphoma (NHL). Previous studies provided some evidence of an association, but did not assess the risk of specific subtypes of NHL, which have been shown to be etiologically diverse. Methods: We conducted a population-based case-control study among members of Group Health (GH), an integrated healthcare delivery system in Washington State. Cases included GH members diagnosed with NHL between 1980-2011 at age 25+ years with no record of a prior cancer or of certain autoimmune conditions, who had been enrolled for 2+ years at diagnosis. Each case was matched to eight GH enrollees on age, sex, and length of prior enrollment at GH; these controls were cancer-free as of the date of the case's diagnosis. Information on prior TCA use, including dose, duration, recency, and type, was ascertained from automated pharmacy data. We used conditional logistic regression to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for NHL, overall and for common subtypes, for various patterns of TCA use. Results: We identified 2,768 cases and 22,127 matched controls. We did not observe an appreciably higher risk of NHL among persons who had filled 2+ TCA prescriptions prior to the date of the case's diagnosis compared to those who had filled none (OR: 1.1; 95% CI: 1.0-1.2). Overall risk of NHL was associated to at most a small degree with longer-term use (OR: 1.2; 95% CI: 1.0-1.4 for 10+ prescriptions), high-dose use (OR: 1.1; 95% CI: 0.8-1.5 for 50+ mg or equivalent), or use that began more than 5 years prior to the reference date (OR: 1.0; 95% CI: 0.9-1.2). TCA use was generally not associated with most major NHL subtypes, though longer-term use was associated with increased risk of chronic lymphocytic leukemia/small lymphocytic lymphoma (OR: 1.5; 95% CI: 1.1-2.0). Conclusions: We found little evidence that TCA use increases risk of NHL, overall or for specific common subtypes of NHL.