WICHITA, Kansas—The final report of data presented in preliminary form at the 2006 American College of Rheumatology (ACR) meeting^1,2 confirms initial conclusions that treatment with TNF inhibitors is associated with increased risk for melanomas and other skin cancers but not for other neoplasms. Importantly, the analysis appears to clear these drugs of causing lymphomas, solid tumors, or other lymphoproliferative malignancies. The report, by Frederick Wolfe, MD, with the National Data Bank for Rheumatic Diseases and the University of Kansas School of Medicine, in Wichita, Kansas, and Kaleb Michaud, PhD, with the University of Nebraska Medical Center, in Omaha, Nebraska, is published in the current issue of Arthritis & Rheumatism.¹

The authors offer some caveats, however. "Although our data do not show associations between malignancy and biologic therapy, except for skin cancers, the mean and median exposure to biologics was only 3.0 years," notes Dr. Michaud. "It is possible that with increasing time of follow-up or exposure, the associations between malignancy and biologic therapy would become stronger. However, true associations are regularly seen within this time frame."

Observational study data show much lower risk than reported in prior meta-analysis

The investigators studied cancers in 13,001 rheumatoid arthritis (RA) patients over about 49,000 patient-years of observation from 1998-2005. Of these patients, 5257 had been treated with some type of biologic, 4277 with infliximab, 3011 with etanercept, 763 with adalimumab, and 319 with anakinra. Mean treatment duration was 3 years for any therapy, 2.9 for infliximab, 2.7 for etanercept, 1.2  for adalimumab, and 1.6 for anakinra.

The investigators studied incident cases of cancer among the subjects participating in a study of RA outcomes. Dr. Wolfe said that cancer reports were validated by hospital, physician, and death records, except for nonmelanomatous skin cancers, which were based on self-report and follow-up interview. Biologic therapy use was classified as ever or never, by duration of individual biologic use, and by duration of total biologic use. The US National Cancer Institute Surveillance, Epidemiology, and End-Results (SEER) database was used to estimate general population cancer rates for comparative purposes. The cancer types examined included breast, colon, lung, lymphoma, melanoma, and nonmelanomatous skin cancer. Biologic therapy was associated with increased rates of melanoma and other skin cancers (primarily basal cell) but not any other cancers, including lung cancer and lymphoma.

Collectively and individually, TNF-inhibitor therapy was linked to an increased risk of skin cancers. The odds ratio for developing melanoma was 2.3. Biologic therapy had no impact on any other type of cancer. The overall risk for all malignancies was 1.0—a result substantially different from the overall risk of 3.3 noted in a meta-analysis of clinical trials of biologic treatment of RA, according to Dr. Wolfe.

"Our conclusions are that melanoma and nonmelanoma skin cancer rates are increased in RA patients treated with biologics and that there is no increase in solid tumors. Therefore, at least two large observational studies are in substantial disagreement with the meta-analysis,³" Dr. Wolfe pointed out at the 2006 ACR meeting.

"This report confirms the results of European studies and shows an overall increase in lymphoma and lung cancer and a decrease in breast and colon cancer in persons with RA, as well as non-increase when all cancers are considered simultaneously," Dr. Wolfe said. "There was no association between either infliximab or etanercept and all cancers, [including] lung cancers and lymphomas. Rates of melanomas and skin cancers were approximately equal for infliximab and etanercept." He cautioned that longer follow-up may be needed for definitive answers regarding the association of TNF inhibitors with cancer types other than melanomas and other skin cancers.

Why do observational studies and meta-analyses of RCTs show different outcomes?

Dr. Wolfe said that the conflicting outcomes of the observational studies and the meta-analysis of randomized clinical trial data are hard to reconcile.

"I posed that question to [a doctor from the FDA]," Dr. Wolfe said, acknowledging his own puzzlement. "I said, 'Look, we study thousands and thousands of patients, and we don't see an effect, but the clinical trial studies do show an effect. How do you explain that?'" One possibility suggested was that anti-TNF therapy might unmask ongoing cancer processes that would show up in relatively short-term clinical trials data, but fade into the background over time and not be apparent in longitudinal studies of RA patients treated with TNF inhibitors or other biologics as compared with patients who were treated with other drugs.

References

1. Wolfe F, Michaud K. Biologic treatment of rheumatoid arthritis and the risk of malignancy. Analysis from a large US observational study. Arthritis Rheum. 2007;56:2886-2895.
2. Wolfe F, Michaud K. The association of new cases of cancer with biologic therapy. Presented at: American College of Rheumatology Meeting; November 11–15, 2006; Washington, DC. Abstract #1321.
3. Bongartz T, Sutton AJ, Sweeting MJ, et al. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies. Systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA. 2006;295:2275-2285.