John Wiley & Sons, Inc.
Tumor necrosis factor blockers may not cause cancer after all
Study finds no increased risk of lymphoma or tumors associated with anti-TNF therapy over methotrexate use among rheumatoid arthritis patients
Patients with rheumatoid arthritis (RA), an autoimmune, inflammatory disease marked by progressive joint and organ damage, face a high risk of developing cancer. Their
vulnerability, especially to lymphoma and leukemia, may be due to the nature of RA. Disease-modifying antirheumatic drugs (DMARDs) – including tumor necrosis factor alpha (TNFá) antagonists, a type
of biologic DMARD – have also been implicated. TNF blockers, which work by attaching to and impeding chemical messengers behind inflammation, have had a significant impact on the treatment of RA.
They have also been linked to lymphoma among users. In fact, reports of lymphoma prompted the Food and Drug Administration to mandate adding a cancer risk warning to the labels of three TNF blockers:
etanercept (Enbrel), infliximab (Remicade), and adalimumab (Humira).
Motivated by persistent concerns and inconclusive studies, researchers at Harvard Medical School's Brigham and Women's Hospital set out to investigate the
association between treatment with TNF blockers and occurrence of cancer in a large sample of patients with RA. Their results, featured in the September 2006 issue of Arthritis & Rheumatism (http://www.interscience.wiley.com/journal/arthritis), indicate that biologic DMARD therapy poses no greater risk for cancer than therapy
with a standard prescription DMARD, methotrexate (MTX).
The researchers focused on 7,830 RA patients, age 65 and older, drawn from healthcare databases in New Jersey, Pennsylvania, and British Columbia, Canada. Subjects included
1,152 biologic DMARD users and 7,306 MTX users. Of those treated with a biologic DMARD, 64 percent had used etanercept, 33 percent had used infliximab, and the remaining 2 percent had used anakinra
(Kineret). 55 percent had previously used MTX and 39 percent were receiving MTX when they began anti-TFN therapy.
Based on medical utilization information, the researchers measured every subject's general health and RA-related characteristics during the 6-month period before
exposure to either a biologic DMARD or MTX. In general, biologic DMARD users had more severe RA than MTX users. No subject had a diagnosis of any cancer before starting their prescribed drug treatment.
Each subject was followed after beginning use of either a biologic DMARD or MTX, with attention to diagnosis of cancer followed by any course or combination of palliative care, biopsy, radiation,
chemotherapy, or surgery. Using time-varying propensity scores to adjust for a wide range of possible factors and stratified proportional hazards regression, researchers estimated the effects of
biological DMARDs on cancer.
From their rigorous definition of cancer, researchers identified a total of 11 cancers of the blood and lymphatic systems and 46 cancerous tumors during 2,940 person-years
of biologic DMARD use, and 58 cancers of the blood and lymphatic systems and 558 tumors during 30,300 person-years of MTX use. Comparing biologic DMARD users with MTX users, the hazard ratio was 1.37 for
blood-related cancers and 0.91 for solid tumors. The overall difference in estimated cancer risk was less than 5 percent between RA patients treated with a TNF blocker and those treated with a standard
"We found no significant increase in the risk of cancers in biologic DMARD users," notes the study's lead authors, Soko Setoguchi, M.D. Dr.P.H and Sebastian
Schneeweiss, M.D, Sc.D. "Our data indicate that it is unlikely that RA patients who have received biologic agents have a much greater risk of lymphoproliterative disorders, hematologic malignancies,
and solid tumors compared with MTX users." Drs, Setoguchi and Schneeweiss, however, acknowledges the challenge of studying the effects of a relatively novel therapy on rare forms of blood cancer, as
well as the need for ongoing studies into the risks and benefits associated with anti-TNF therapy.