Juan Ruiz1, Chi-Fang Wu2, Hui Zhao2, Sharon Giordano2, Suja Rajan3 and Maria Suarez-Almazor4, 1The University of Texas MD Anderson Cancer Center, Houston, TX, 2The University of Texas, MD Anderson Cancer Center, Houston, TX, 3University of Texas School of Public Health, Houston, TX, 4MD Anderson Cancer Center, Houston, TX
Background/Purpose: Biologic and targeted synthetic disease modifying antirheumatic drugs (bDMARD and tsDMARD) are immunosuppressant, and there have been concerns that they might impact tumor immunity in cancer patients who receive these therapies for rheumatoid arthritis (RA). However, recent data, primarily on bDMARD, has not shown deleterious effects on cancer outcomes. The purpose of this study was to describe time trends in the utilization of bDMARD and tsDMARD in patients with RA who develop breast cancer (BC).
Methods: We performed a retrospective cohort study of patients 30 years or older with RA and BC diagnosed from 2008 to March 2021 identified in the Optum Clinformatics claims data, using International Classification of Diseases-9 and 10 (ICD-9, ICD-10) diagnostic codes. In order to include early stages of BC, we only included patients who underwent mastectomy or lumpectomy within 30 days before or 270 days after the BC diagnosis; RA was defined as having at least two RA claims in the year prior to BC diagnosis. The outcome of interest was use of bDMARD/tsDMARD starting 3 months after cancer diagnosis, defined as time to first drug claim, censoring at last follow-up or death. We used a Cox proportional hazard multivariable model to evaluate covariates including age, year of diagnosis, insurance, race, comorbidity and use of bDMARD/tsDMARD in the 3 months prior to the BC diagnosis.
Results: 1,707 patients diagnosed with BC and RA were identified. In the first five years after BC, 264 (15.5%) patients used bDMARD (98.5%) or tsDMARD (1.5%). The most common bDMARD used were the tumor necrosis factor inhibitors (TNFi) (76.4%) followed by rituximab (15.4%), abatacept (8.1%), and IL-6 inhibitors (6.2%). Among the patients who received bDMARD/ tsDMARD, 210 (79.5%) received them initially during the first year after BC diagnosis. The median time from BC diagnosis to first bDMARDs/tsDMARDs claims was 5.6 months for new users. In the Cox model, the largest predictor of bDMARD/tsDMARD use after BC was prior use of any of these agents before the BC diagnosis; hazard ratio (HR) 22.19 95% confidence interval (CI) 16.41-30.00. Compared to years of BC diagnosis 2008-2009 (reference value), years 2012 -2015 had significantly lower HR for bDMARD/tsDMARD use (HR=0.52 95%CI 0.32-0.86 and HR=0.43 95%CI 0.24-0.75 respectively) but no significant differences were observed from 2016 onwards. Having one additional comorbidity versus none was also associated with bDMARD/tsDMARD use (HR=1.55 95%CI 1.15-2.08). No significant associations were observed with age, race, type of insurance or geographical region.
Conclusion: The use of bDMARD/tsDMARD in patients with RA and recently diagnosed early stage BC has not increased in the past decade, despite data suggesting that some bDMARD do not adversely impact cancer progression in patients with early, treated, non-metastatic disease.
Disclosures: J. Ruiz, None; C. Wu, None; H. Zhao, None; S. Giordano, None; S. Rajan, None; M. Suarez-Almazor, Eli Lilly, Pfizer, Celgene, ChemoCentryx, Gilead.
