Tumor necrosis factor agents do not increase the overall risk of cancer in patients with rheumatoid arthritis but may be associated with an increased risk of lymphomas, an analysis of data from the South Swedish Arthritis Treatment Group register suggests.
The study is the first to question whether use of anti-TNF-α agents increases the risk of lymphoma independently of disease severity.
Pierre Geborek, M.D., and colleagues at Lund (Sweden) University Hospital, identified 757 patients treated with etanercept (Enbrel) or infliximab (Remicade) from the register and 800 conservatively treated patients recruited from an outpatient clinic and private practices (Ann. Rheum. Dis. 2005;64:699–703).
Patients were followed from initiation of anti-TNF treatment (or July 1, 1997, for the comparison group) until death or Dec. 31, 2002.
In the anti-TNF group, there were 16 tumors (5 lymphomas) in 1,603 person-years at risk, compared with 69 tumors (2 lymphomas) in 3,948 person-years in the control group.
The standardized incidence ratios in the anti-TNF group and the control group were 1.1 and 1.4 for all tumors and 11.5 and 1.3 for lymphomas, respectively.
The increased overall tumor incidence in the control group was mainly due to smoking-related lesions, according to the investigators.
The total cancer risk excluding lymphomas was 0.79 among patients treated with anti-TNF agents and 1.39 in the comparison group.
The unadjusted hazard ratio for lymphoma was 4.9 in anti-TNF-treated patients relative to the conventionally treated patients. The hazard ratio was 5.0 after adjusting for differences in baseline Health Assessment Questionnaire scores, which were used as a marker of severity.
The results for overall tumor standardized incidence ratios in the anti-TNF-treated patients, compared with controls, must be interpreted with caution because of the limited number of observations and the relatively short follow-up period, the investigators wrote.
The withholding of anti-TNF treatment in patients with a known previous cancer also may have contributed to the lower incidence of cancer in this group.
In an accompanying editorial, Jarrod Franklin and colleagues at Manchester (England) University noted the study failed to detect a raised incidence of lymphoma in the control group, despite detecting an increased risk of all-site cancers in this population.
This is surprising, they said; given that the results of several studies would suggest such patients would be at an increased risk of lymphoma.
They welcomed the investigation but agreed it is difficult to reach a “robust conclusion” on the question of anti-TNF-α agents, disease severity, and lymphoma risk.
“With increasing recruitment and follow-up of such cohorts, a more definitive answer should be available in the not too distant future,” they wrote.
Eric Ruderman, M.D., a rheumatologist with Northwestern University, Chicago, concurred. “It's important information, but the issue you have to take into consideration is that their denominators were fairly small to look at such a rare event,” he said in an interview. “One or two patients one way or the other may have made a significant difference.”
