Background/Purpose: Rheumatoid arthritis (RA) develops insidiously, often over many years. Given that diet and obesity have been associated with increased RA risk, we investigated whether metabolomic profiling of plasma could identify novel biomarkers associated with risk of developing RA among women.

Methods: Incident RA cases with plasma samples drawn prior to disease onset in the Nurses’ Health Study (NHS) and Nurses’ Health Study II (NHSII) cohorts were matched to controls at a 1:2 ratio based on age, race, menopausal status, post-menopausal hormone use, as well as blood-draw variables: fasting status, time of day, and date of blood draw. Untargeted liquid chromatography tandem mass-spectrometry molecular profiling of plasma samples drawn prior to RA onset from incident cases and matched controls participating in NHS and NHSII were collected. The final sample after quality control filtering consisted of a total of 256 pre-RA cases and 511 matched controls, measured across 437 unique, known metabolites. Conditional logistic regression was used to assess the association between individual metabolites and risk of RA, with adjustment for BMI and smoking. Subgroup analyses were also performed among case-control matched groups with 1) seropositive RA cases, and 2) with RA cases whose plasma was drawn between 1-5 years prior to diagnosis. Multiple comparison adjustments were made using the number of effective tests.

Results: Top metabolites associated with increased risk of incident RA included C18:1 lysophospholipid (OR: 1.23, 95%CI: 1.05-1.45), and C22:0 lysophosphatydlserine isomer (OR: 1.22, 95%CI: 1.03-1.43). 4-acetamidobutanoic acid (OR: 0.78, 95%CI: 0.66-0.93) and N-acetyltryptophan (OR: 0.83, 95%CI: 0.70-0.98), as well as C5 (OR: 0.84, 95%CI: 0.71-0.995) and C5:1 carnitines (OR: 0.82, 95%CI: 0.69-0.96) were shown to have a protective effect against incident RA. Although these metabolites did not survive adjustment for multiple comparisons, they were found to be marginally significant and shared consistent directions of effect in the seropositive-only RA analyses. In the 1-5 year period prior to RA diagnosis, homoarginine was identified as a significant risk factor for incident RA in both the full group (OR: 1.77, 95%CI: 1.24-3.53) and in the seropositive-only (OR: 3.01, 95%CI: 1.30-6.96) analyses.

Conclusion: We have identified several metabolic markers of incident RA. Further replication in an independent cohort is ongoing to confirm our findings.