TNF Inhibitors Do Not Appear to Increase Risk for Lymphoma

Alice Goodman

November 16, 2012

WASHINGTON — In patients with rheumatoid arthritis (RA), no association was found between treatment with a tumor necrosis factor (TNF) inhibitor and increased risk for lymphoma, according to a large study presented here at ACR 2012.

"There is no evidence that anti-TNF therapy increases the risk of lymphoma over the background risk associated with RA, but further follow-up is needed to establish if the picture changes with time," said Louise K. Mercer, PhD, from the arthritis research UK epidemiology unit at The University of Manchester in the United Kingdom.

This finding confirms results found previously in 3 large observational studies. The first study found 4 lymphomas in 1152 patients taking TNF inhibitors. The second found 95 lymphomas in 19,591 patients, 55% of whom were on TNF inhibitors. The third found 26 lymphomas in 6604 patients treated with TNF inhibitors, Dr. Mercer explained.

Strengthens Existing Body of Evidence

Dr. Mercer and colleagues conducted a prospective study comparing 2 cohorts of patients with active RA enrolled in the British Society for Rheumatology Biologics Register from 2001 to 2009. Patients were followed until 2010.

At baseline, 3465 patients were receiving nonbiologic disease modifying anti-rheumatic drugs (DMARDs) and 11,987 were receiving TNF inhibitors. Patients in the DMARD group were older and were mostly male; those in the TNF inhibitor group had a longer disease duration and higher Health Assessment Questionnaires scores at baseline. Patients were followed with physician questionnaires, patient questionnaires, and diaries, and the National Health Service Cancer and Death Registries were used to confirm the diagnosis of cancer in these patients.

The study was based on 13,186 patient-years of follow-up for DMARDs and 66,353 patient-years of follow-up for TNF inhibitors. In the DMARD group, 20 lymphomas were identified; in the TNF inhibitor group, 64 lymphomas were identified (rates of 152/100,000 person-years and 96/1000,000 person-years, respectively).

After adjustment for confounding factors such as age, sex, disease characteristics, use of steroids/cyclophosphamide, and smoking, no difference in risk for lymphoma was found between patients treated with a TNF inhibitor and those who were not.

RA itself is associated with a risk for lymphoma that is 130/100,000 person-years, Dr. Mercer told meeting attendees.

Five cases of Hodgkin's lymphoma were identified in the DMARD group and 9 were identified in the TNF inhibitor group; 16 cases of non-Hodgkin's lymphoma were identified in the DMARD group and 55 were identified in the TNF inhibitor group.

The strengths of the study are the size, the detailed patient data, the confirmation of cancer cases, and the use of a propensity model to adjust for confounding factors.

Because of the lag in reporting from the national cancer registry, the complete follow-up data are not available in real time, Dr. Mercer explained. Further follow-up is required, she added.

Primary Lymphoma Risk Related to Disease, Not Drug

"The message is that the primary risk for lymphoma is disease-related, not drug-related," said Eric Ruderman, MD, from the Northwestern University Feinberg School of Medicine in Chicago, Illinois.

Nevertheless, the risk for lymphoma is included in the labeling of all TNF inhibitors, he noted. "In retrospect, all the data suggest that the increased risk is almost completely attributable to disease, not to the drug. Rheumatologists and patients should be comfortable that the majority of the risk is disease-related," Dr. Ruderman said.

Dr. Mercer has disclosed no relevant financial relationships. Dr. Ruderman reports receiving grants for clinical research from Abbott, Amgen, Bristol-Myers Squibb, and Biogen Idec; and serving as an advisor or consultant for Abbott, Amgen, Bristol-Myers Squibb, and Biogen Idec, and Genentech.

ACR 2012: Abstract 1593. Presented November 12, 2012.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....