Biologic treatment of rheumatoid arthritis patients has spurred an increase in melanoma and other skin cancers but not malignancies other than those of the skin, according to an observational study of 13,001 patients, reported Dr. Frederick Wolfe, a rheumatologist at the National Data Bank for Rheumatic Diseases, and the University of Kansas, Wichita, and his colleague, Kaleb Michaud, Ph.D., of the University of Nebraska, Omaha
Extrapolating from previous data in immunosuppressed transplantation patients, some have theorized that immunosuppression from biologics would increase the risk of cancer database (Arthr. Rheum. 2007;56:2886-95).
The hypothesis gained some credence when a meta-analysis found a 3.3-fold increased risk of malignancy in general among transplantation recipients who were treated with infliximab or adalimumab, compared with those who received nonbiologic therapy (JAMA 2006;295:2275-85).
Findings from studies in patients taking biologic agents for rheumatoid arthritis conflict, however.
An observational study using the Swedish inpatient registry found cancer risks "largely similar" between 4,160 patients treated with a tumor necrosis factor (TNF) antagonist, compared with 53,067 other rheumatoid arthritis patients not treated with such an agent (Ann. Rheum. Dis. 2005;64:1421-6).
To address this discrepancy, Dr. Wolfe and Dr. Michaud studied 13,001 rheumatoid arthritis patients who were included in the National Data Bank for Rheumatic Diseases.
In an assessment of the national rheumatoid arthritis database sample only, risk of melanoma (odds ratio, 2.3) and nonmelanotic skin cancer (OR, 1.5) increased among those ever treated with a biologic, compared with the others.
These associations were consistent across the different agents in the biologics class. A total of 4,277 patients (33%) received infliximab; 3,011 received etanercept (23%); 763 received adalimumab (6%); and 319 received anakinra (3%) in the study.
Cancer incidence was based on self-reports from semiannual questionnaires.
Researchers compared their specific cancer rates with a comparison population from the Surveillance, Epidemiology, and End Results (SEER) database.
The overall cancer rate did not differ between rheumatoid arthritis patients and SEER database participants (standardized incidence ratio [SIR] = 1.0). "This result is substantially different from the OR of 3.3 noted by Bongartz et al. [above] in their meta-analysis of clinical trials," the authors wrote.
The mean duration of any type of biologic therapy was 3 years, which the authors of the current study cited as a possible limitation. However, Dr. Wolfe and Dr. Michaud wrote, "true associations are usually seen within this time frame, since post-transplantation studies have shown increased risk after the first year of treatment."
Melanoma (SIR, 1.7) and lymphoma (SIR, 1.7) occurred more often in the rheumatoid arthritis database versus the SEER database participants. Rates of lung cancer and bladder cancer were not statistically different between groups. Some cancer rates were lower among rheumatoid arthritis patients, including breast cancer (SIR, 0.8) and colon cancer (SIR, 0.5), compared with the comparison group.
"In summary, biologic therapy is associated with increased risk for skin cancers, but not for solid tumors or lymphoproliferative malignancies."
During the period when these data were collected, the National Data Bank for Rheumatic Diseases received funding from Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Pfizer, and Wyeth-Australia. Centocor reviewed the completed manuscript.