Background/Purpose:  Smoking and high body mass index (BMI) are risk factors for RA. Both are associated with increased systemic inflammation, with elevated C-reactive protein, interleukin (IL)-6, and tumor necrosis alpha (TNFα). We and others have previously reported that circulating plasma levels of IL-6 and TNFRII (a proxy for TNFα) are elevated years prior to RA onset. We hypothesized that there may be modification of the relationship between plasma levels of IL6 and TNFRII and the future risk of RA according to an individual’s BMI and smoking history.    

Methods: We conducted a nested case-control study using stored plasma samples from the Nurses’ Health Study (NHS) and Nurses’ Health Study II (NHSII), prospective cohort studies of U.S. female nurses. Cases were women diagnosed with RA at least 3 months after blood draw. We identified 3 controls for each case randomly chosen from subjects with stored blood, matched on year of birth, race, menopausal status/post-menopausal hormone use, time of day, blood draw fasting status and menstrual cycle timing. Smoking (pack-years) and BMI were obtained from the questionnaire cycle prior to blood collection. IL6 and TNFRII were measured by ELISA. Multivariable logistic regression models were used to calculate odds ratios (OR) and 95% CIs for RA risk associated with a 1 unit increase in log-TNFRII or log-IL6, adjusting for matching factors, alcohol intake, and menstrual regularity.  We tested for multiplicative and additive interactions between cytokine levels, and smoking (never or < 10 vs. ≥ 10 pack years) and BMI <25 (normal weight) vs. ≥25 kg/m2 (overweight).

Results:  Among 62,439 women followed in the NHS cohorts with stored blood, incident RA was confirmed in 196 subjects, who were matched to 588 healthy participants. Mean age at RA diagnosis was 60.1 years (10.0 SD), 54% RF or anti-CCP seropositive. Mean duration until RA diagnosis was 7.6 years (4.5 SD). A significant association between IL6 and future RA risk was found (Table). The OR for TNFRII was elevated but non-significant. We observed a significant multiplicative interaction between TNFRII and BMI (p=0.02).  The relationship between plasma TNFRII cytokine elevation and future RA risk was stronger among women with higher BMI than those of normal weight. We did not detect any significant additive or multiplicative interactions between IL6 and BMI, or IL6 or TNFRII and smoking.

Conclusion: In this large prospective cohort study of women with blood collected up to 14 years prior to the diagnosis of RA, a multiplicative interaction was found between high BMI and elevated TNFRII level  in increasing the risk of RA. This suggests that BMI and TNFα elevation may belong to the same biologic pathway in RA pathogenesis.