By Chris Berrie
BARCELONA, SPAIN -- June 15, 2007 -- The use of tumor necrosis factor (TNF) antagonists is not associated with any significant further increase in lymphoma occurrence for patients with rheumatoid arthritis (RA), both compared with the general population and with patients not treated with TNF antagonists, a population-based study shows.
This study was presented here by coinvestigator Johan Askling, MD, PhD, associate professor of epidemiology, clinical epidemiology unit and rheumatology unit, department of medicine, Karolonska University Hospital and Institute, Stockholm, Sweden, on June 14th at the Annual European Congress of Rheumatology (EULAR).
Patients with RA are known to have an increased risk of lymphoma occurrence, and a growing concern relates to the possibility that the use of TNF antagonists could increase their already elevated lymphoma risk. It is also not known if accumulated TNF antagonist exposure will have increased this risk.
To assess the risk of lymphoma associated with TNF antagonist treatment in patients with RA, Dr. Askling and colleagues conducted an analysis of various national health and census registers.
The cohort of patients with RA that was alive in 1998 (n = 66,471) was derived from 51,588 on the Swedish Inpatients Register, 45,903 on the Swedish Outpatients Register, and 5,529 on the Early RA Register. This RA cohort was matched to a population comparator from the Swedish Population Census Register (n = 240,181). Then through consultation with the Swedish Biologics Register (ARTIS), the RA cohort was seen to contain 6,304 patients who had started TNF antagonist treatment.
These subject numbers were converted to person-years, with the time spent without TNF antagonist exposure in the TNF-antagonist-treated RA group included with the non-TNF-antagonist-treated RA group (RA comparator), giving 19,997 and 322,713 person-years, respectively, with 1,486,848 person-years for the population comparator.
When matched to the Swedish Cancer Register from 1998 to 2005, these datasets indicated 21 lymphomas for TNF-antagonist-treated patients with RA, 296 for the RA comparator, and 738 for the population comparator.
Overall, this indicated relative risks for lymphoma occurrence in the TNF-antagonist-treated RA patients of 3.22 (95% confidence interval [CI], 2.05-5.04) for the population comparator, and 1.37 (95% CI, 0.85-2.21) for the RA comparator.
The relative risks of lymphoma occurrence in the TNF-antagonist-treated RA group were also compared with the RA and population comparators, to reveal potential risk disparities within the TNF-antagonist-treated RA group.
Indeed, the only significant risk ratio of 1.86 (95% CI, 1.12-3.10) seen for the RA comparator for the 1998-to-2001 group as the year of first treatment start thus more than accounted for the entire initial increase in the relative risk between the TNF-antagonist-treated RA group and the RA comparator. Interestingly, this would also indicate here a more recent decrease in relative risk for TNF antagonist treatment.
Relative risks were noted not to have varied significantly according to the type of TNF antagonist used.
Therefore, this analysis demonstrates that although there has been concern expressed relating to a potential increase in the already elevated lymphoma risk in patients with RA, in this population there appears to be no cause for concern, the researchers concluded.
Considering the significant relative risk identified here from early TNF antagonist use, as Dr. Askling indicated, "You shouldn't just stop safety monitoring after a year or two -- you should go on, because if you look what happens here, if safety monitoring had stopped in 2003, the conclusion would have been that yes, TNF antagonists increase the risk, whereas now we see that it probably doesn't, and it will probably be the reverse."
[Presentation title: RA Patients Exposed to TNF Antagonists Are Not at Increased Risk of Malignant Lymphomas Compared to Other RA Patients. Abstract THU124]
E-Mail this DGDispatch to a colleague
