Periodontal Disease and Risk of Non Hodgkin Lymphoma (NHL) in the Health Professionals Follow-up Study (HPFS)

Introduction: Periodontal disease has been associated with several chronic diseases, including cancer. In an earlier prospective cohort analysis within the HPFS, we observed a 31% higher risk of NHL among participants who reported severe periodontitis at baseline (Michaud et al. Lancet Oncology 2008). Herein, we examined this novel association with an additional 8 years of follow-up, updated periodontal disease status, and analyzed association by subtype of NHL.

Methods: The HPFS is a prospective cohort of male health professionals, aged 45-65 years, recruited throughout the US in 1986 using mail questionnaires. Periodontal status was reported at baseline and biennially thereafter. Incident cases of NHL diagnosed between 1986-2012 are included in this analysis; subtypes of NHL were determined through medical records. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox proportional hazard models adjusting for age, race, number of teeth, body mass index, diabetes, smoking and geographic region. An 8-year lag analysis was performed evaluating periodontal status 8-years prior to diagnosis (including follow-up years 1994-2012, n=514 cases).

Results: During 16 years of follow-up, 876 cases of NHL were ascertained. Subtypes with sufficient numbers for analysis included chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) (n=281), diffuse large B-cell lymphoma (DLBCL) (n=83) and follicular lymphoma (FL) (n=91). Significant associations for periodontal disease at baseline were noted for total NHL, which includes B- and T-cell subtypes (HR = 1.25, 95% CI: 1.05-1.49) and CLL/SLL (HR = 1.37, 95% CI: 1.01-1.86) (Table). With updated periodontal status, similar associations were observed for periodontal disease and risk of total NHL (HR = 1.30, 95% CI: 1.11-1.52) and CLL/SLL (HR = 1.40, 95% CI: 1.06-1.83). In the baseline and updated analyses, associations for DLBCL were similar to those for all NHLs, however, no association was noted for FL. Associations remained elevated in an 8-year lag analysis for total NHL (HR = 1.23, 95% CI = 1.00-1.52), and for CLL/SLL (HR = 1.48, 95% CI = 1.04-2.09).

Conclusion: These data show that the association for periodontitis and NHL has persisted over time in the HPFS. Furthermore, the long latency period between periodontal disease status and NHL diagnosis suggests that the association is not a consequence of the cancer. In addition, the association with periodontal disease appeared strongest for CLL/SLL and DLBCL. Mechanistic and prospective study of this association are warranted.