Background/Purpose: There is concern about the use of tumor necrosis factor inhibitors (TNFi) in patients with early cancer given their immunosuppressant effects. Few studies have examined cancer outcomes in patients with rheumatoid arthritis (RA) and cancer, and they have generally included long-term survivors. Therefore, the safety of TNFi in patients with recent cancer is unknown. We aimed to examine the survival of patients with RA recently diagnosed with colorectal (CRC), lung, or prostate cancer, who received TNFi after cancer diagnosis.

Methods: We conducted a retrospective cohort study of patients with RA diagnosed with CRC, lung, or prostate cancer (stage 1-3) identified in the Surveillance, Epidemiology, and End Results Program (SEER)–Medicare linked database (2008-2019). Use of disease modifying antirheumatic drugs (DMARD) and glucocorticoids after cancer diagnosis was identified from claims and prescription files. The outcomes of interest were overall survival (OS) and disease-specific survival (DSS) defined as the time from cancer diagnosis to death from all causes or death from cancer diagnosis, respectively. We compared patients who received TNFi with those who received conventional cDMARD, or no DMARD during the first three years after cancer diagnosis. We estimated a propensity score for use of TNFi through a logistic regression model. We conducted multivariable Cox proportional hazard regression analysis, controlling for various covariates and for the propensity score. We performed landmark analysis to avoid immortal time bias. We also performed landmark analysis with instrumental variables (IV). Finally, we also modelled TNFi use as a time-varying covariate.

Results: We identified 514 individuals with CRC, 864 with lung cancer, and 603 with prostate cancer. Eighty (15.6%), 102 (11.8%), and 120 (19.9%) individuals received TNFi in the CRC, lung cancer, and prostate cancer cohorts, respectively. In the CRC cohort, no significant differences in OS or DSS was observed after multivariable adjustment in any year of the landmark analysis between patients treated with TNFi and patients treated with csDMARD only; for OS the hazard ratios (HR) for the first, second and third year were: 0.72, 95% confidence interval (CI) 0.43-1.21; 0.74 95% CI 0.40-1.38; 0.61 95% CI 0.30-1.25, respectively. For the lung cancer cohort, OS was improved in the third-year landmark analysis in patients treated with TNFi compared to the csDMARD only group (HR 0.58, 95%CI 0.34–0.96). No difference was observed in the first- or second-year analysis. Improved DSS was observed in the first-year analysis (HR 0.40 95% CI 0.17-0.96). For the prostate cancer cohort, no difference was observed in the first-, second-, or third-year landmark analysis between the TNFi groups and the csDMARD groups. Similar results were observed when using IV or time-varying models. No association of TNFi dosage (< 6 or ≥6 months and < 12 or ≥12 months) with OS or DSS was observed. Glucocorticoid therapy showed worse OS than no glucocorticoids in the three cancer cohorts in the first-year landmark analysis.

Conclusion: Treatment with TNFi during the first three years after early colorectal, lung and prostate cancer diagnosis was not associated with worse survival.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/survival-in-patients-with-rheumatoid-arthritis-and-early-stage-colorectal-lung-or-prostate-cancer-receiving-tumor-necrosis-factor-alpha-inhibitors/