Table 2. Median Time to Cancer Diagnosis, Number of Incident Cancer Diagnoses per 10 000 Person-Years at Risk, and Adjusted HR After at Least 90 Days and Within 2 Years of Treatment Exposure by Drug Category.
| Biologic or targeted synthetic DMARD | No. (%) (N = 27 661) | Time to cancer diagnosis, median (IQR), d | Crude No. (%) of cancer diagnoses | No. of cancer diagnoses/10 000 person-years (95% CI) | Adjusted HR (95% CI)a |
|---|---|---|---|---|---|
| TNFi | 20 586 (74) | 245 (152-437) | 162 (0.8) | 78 (66-91) | 1.00 [Reference] |
| Rituximab | 1182 (4) | 202 (157-303) | 14 (1.2) | 171 (94-285) | 1.91 (1.17-3.14) |
| IL-6i | 1068 (4) | 230 (199-299) | 8 (0.8) | 88 (38-173) | 1.04 (0.57-1.92) |
| Abatacept | 2255 (8) | 289 (150-373) | 30 (1.3) | 142 (96-201) | 1.47 (1.03-2.11) |
| JAKi | 2570 (9) | 276 (169-351) | 22 (0.9) | 94 (59-143) | 1.36 (0.94-1.96) |
Abbreviations: DMARD, disease-modifying antirheumatic drug; HR, hazard ratio; IL-6i, interleukin 6 inhibitor; JAKi, Janus kinase inhibitor; RA, rheumatoid arthritis; TNFi, tumor necrosis factor inhibitor.
Models adjusted for individual covariates: demographics (age, sex, geographic region, and year of initiating biologic), proxies of RA severity (glucocorticoid and conventional synthetic disease-modifying antirheumatic drug prescription fills and number of drug switches), clinical characteristics (number of days from RA diagnosis to biologic initiation, tobacco use, frailty status,16,17 and Charlson comorbidity score18), and health care utilization within 12 months prior to starting treatment (hospital admissions, emergency department visits, outpatient visits, opioid or nonsteroidal anti-inflammatory drug prescription fills, and days of follow-up).