Systematic Evaluation of Rheumatoid Arthritis Risk by Integrating Lifestyle Factors and Genetic Risk Scores

Abstract

Background: Effective identification of high-risk rheumatoid arthritis (RA) individuals is still a challenge. Whether the combined effects of multiple previously reported genetic loci together with lifestyle factors can improve the prediction of RA risk remains unclear.

Methods: Based on previously reported results and a large-scale Biobank dataset, we constructed a polygenic risk score (PRS) for RA to evaluate the combined effects of the previously identified genetic loci in both case-control and prospective cohorts. We then evaluated the relationships between several lifestyles and RA risk and determined healthy lifestyles. Then, the joint effects of healthy lifestyles and genetic risk on RA risk were evaluated.

Results: We found a positive association between PRS and RA risk (OR = 1.407, 95% confidence interval (CI) = 1.354~1.463; HR = 1.316, 95% CI = 1.257~1.377). Compared with the low genetic risk group, the group with intermediate or high genetic risk had a higher risk (OR = 1.347, 95% CI = 1.213~1.496; HR = 1.246, 95% CI = 1.108~1.400) (OR = 2.169, 95% CI = 1.946~2.417; HR = 1.762, 95% CI = 1.557~1.995). After adjusting for covariates, we found protective effects of three lifestyles (no current smoking, regular physical activity, and moderate body mass index) on RA risk and defined them as healthy lifestyles. Compared with the individuals with low genetic risks and favorable lifestyles, those with high genetic risks and unfavorable lifestyles had as high as OR of 4.637 (95%CI = 3.767~5.708) and HR of 3.532 (95%CI = 2.799~4.458).

Conclusions: In conclusion, the integration of PRS and lifestyles can improve the prediction of RA risk. High RA risk can be alleviated by adopting healthy lifestyles but aggravated by adopting unfavorable lifestyles.

Keywords: epidemiology; genetic factor; healthy lifestyle; polygenic risk score; rheumatoid arthritis.

Figure 1

(A) Flowchart for filtering participants from the UK Biobank cohort; (B) Study design and workflow of our research.

Figure 2

The distributions of PRSs among participants with and without RA. (A) Distribution of PRS among participants with and without RA in all participants; (B) Restricted cubic spline models for the relationship between PRS and RA risk in case–control analysis; (C) Restricted cubic spline models for the relationship between PRS and RA risk in prospective analysis; (D) Distribution of PRS _non-MHC among participants with and without RA in all participants; (E) Restricted cubic spline models for the relationship between non-MHC PRS and RA risk in case-control analysis; (F) Restricted cubic spline models for the relationship between non-MHC PRS and RA risk in prospective analysis; Associations in B, C, E, and F were adjusted for age, sex, genotyped batch, assessment center, TDI and the first 10 principal components of ancestry. RA, rheumatoid arthritis; PRS, polygenic risk score; MHC, major histocompatibility complex; OR, odds ratio; HR, hazard ratio.

Figure 3

(A) Standardized rates of RA events in low (bottom quartile), intermediate (quartiles 2 to 3), and high (top quartile) genetic risk groups in the UKB cohort. (B) Standardized rates of RA events in favorable, intermediate, and unfavorable lifestyle groups in the UKB cohort (C) Standardized rates of RA events in participates with different genetic risk score groups and different lifestyle groups in the UKB cohort. Associations were adjusted for age, sex, genotyped batch, assessment center, TDI, and the first 10 principal components of ancestry. RA, rheumatoid arthritis; PRS, polygenic risk score; MHC, major histocompatibility complex; OR, odds ratio; HR, hazard ratio. Definition of healthy lifestyle indicators: include 0 to 1 healthy lifestyle indicator for unfavorable lifestyle, 2 for moderate lifestyle, and all 3 for favorable lifestyle. The genetic risk groups were evaluated by PRS: low genetic risk (bottom quartile of PRS), intermediate genetic risk (quartiles 2 to 3), and high genetic risk (top quartile).

Figure 4

RA risks in the subgroups stratified by genetic PRSs and lifestyles in the UKB cohort (versus participants with favorable lifestyles in the low genetic risk group) (A) Case-control study; (B) Cohort study. Definition of healthy lifestyle indicators: include 0 to 1 healthy lifestyle indicator for unfavorable lifestyle, 2 for moderate lifestyle, and all 3 for favorable lifestyle. The genetic risk groups were evaluated by PRS: low genetic risk (bottom quartile of PRS), intermediate genetic risk (quartiles 2 to 3), and high genetic risk (top quartile).