WASHINGTON, DC—Recent data from Sweden has suggested that smoking is a risk factor for rheumatoid arthritis (RA) associated with anti-cyclic citrullinated peptide (CCP) antibodies.1  However, an analysis of data from three large North American RA cohorts reported at the American College of Rheumatology 2006 meeting presents a more complicated picture.

"We suggest that in North America the presence of other environmental risk factors obscures the effect of smoking on risk for anti-CCP-positive RA." —Patricia Irigoyen, MD.
"In three large North American RA cohorts, smoking appeared to increase the risk of anti-CCP-positive RA, but had no clear interaction with the shared epitope. We think it is likely that other environmental factors in North America are contributing to anti-CCP production in this group," reported Patricia Irigoyen, MD, of North Shore/Long Island Jewish Medical Center in Manhasset, New York.2

Researchers hoped to sort out interactions between smoking, other RA factors

The researchers examined the association between smoking, shared epitope and anti-CCP antibodies in a case- only analysis of three large North American RA cohorts. This analysis included data for 2479 Caucasian patients with RA who were included in three cohorts: the Study of New Onset RA (SONORA) cohort, a recently published inception cohort (Fries, Wolfe, Apple, et al. Arthritis Rheum. 2002.46:2320-2329), and the North American RA Consortium (NARAC) family collection. For each subject the researchers measured anti-CCP and rheumatoid factor (RF) levels, and performed HLA-DRB1 typing. Demographic data included age, sex, ethnicity, and smoking status.

Patients from the NARAC cohort (n = 1105) had longstanding, severe RA, with a mean disease duration of 14 years. Smoking data were limited to ever/never smoked. Both the SONORA and inception cohorts were prospective studies of new-onset RA. Smoking information was ever/never for SONORA patients and pack-years of smoking for the inception cohort.

Dr. Irigoyen said that the first question asked was whether having ever smoked was associated with anti-CCP positivity in RA patients. "Overall, it looked like yes for the NARAC and inception cohorts, but not for SONORA," she said. Odds ratios (ORs) for anti-CCP were 1.38 for NARAC, 1.57 for the inception cohort, and 1.02 for the SONORA cohort.

Dr. Irigoyen said that in the SONORA cohort, smoking was actually protective against anti-CCP in patients who had two HLA-DR shared epitopes.

Meta-analysis of all three groups showed that the OR for anti-CCP antibodies was 1.14 for RA patients with no shared epitope, 1.35 for RA patients with one shared epitope (P <.001), and 1.20 for RA patients with two shared epitope alleles (P = NS). The investigators then did multiple logistic regression analyses in each group to evaluate the effect of smoking when controlling for other factors known to affect anti-CCP status, such as age and gender. "We got mixed results," Dr. Irigoyen said. "In all the cohorts, the shared epitope was strongly associated with anti-CCP production, but in smokers it was not significant in any of the three groups."

The investigators found that smoking does appear, overall, to show an independent effect on anti-CCP when controlling for the shared epitope. They observed "only weak evidence for an interaction between smoking and the shared epitope in anti-CCP positive RA," Dr. Irigoyen said. However, she cautioned that these cohorts were from studies that were not originally designed to address this question, and incomplete smoking data for two of the three cohorts were included. In the inception cohort, which had the most complete smoking data, there was a suggestion of an interaction between smoking and shared epitope, but when that group was analyzed according to heavy versus light smokers, there was not much difference.

"Our conclusion is that in these three cohorts of patients with RA smoking does not appear to be associated with the presence of anti-CCP antibodies, and there is no evidence of interaction between smoking and the shared epitope. Throughout all data sets we continued to see a strong association between the shared epitope alleles and the presence of CCP. These data are not consistent with recently published findings from Sweden showing that smoking is predominantly a risk factor for anti-CCP positive RA," Dr. Irigoyen concluded. "We suggest that in North America the presence of other environmental risk factors obscures the effect of smoking on risk for anti-CCP positive RA."


E-mail any comments to .

References

1. Klareskog L, Stolt P, Lundberg K, et all. A new model for an etiology of rheumatoid arthritis: Smoking may trigger HLA-DR (shared epitope)-restricted immune reactions to autoantigens modified by citrullination. Arthritis Rheum. 2006;54:38-46.
2. Irigoyen P, Lee H-S, Vlietinck R, et al. Lack of association between smoking and anti CCP antibodies in three large North American rheumatoid arthritis cohorts. Presented at: American College of Rheumatology Meeting; November 15, 2006; Washington, DC. Abstract 2088.