The familial risk of rheumatoid arthritis (RA) cannot be explained fully by established genetic and environmental RA risk factors, so family history remains an independent risk factor for RA, researchers reported online in Arthritis and Rheumatism.

The results imply that “family history of RA will remain an important clinical risk factor for RA that cannot be replaced by genotyping of known risk loci or collection of specific risk factors,” wrote Xia Jiang, MD, at Karolinska Institutet in Stockholm, and colleagues. “In particular, for ACPA [anti-citrullinated protein antibody]-negative RA, it seems that we have only found factors responsible for a very minor part of the familial aggregation.”

The researchers linked national Swedish population registers to the Epidemiological Investigation of Rheumatoid Arthritis (EIRA) study, a population-based, case-control study comprised of incident RA cases 18 to 70 years old and matched controls. First-degree relatives of cases and controls were identified through the Swedish Multi-Generation register.

The intent was to assess the extent to which RA familial risk could be explained by established genetic and nongenetic RA risk factors. Medical evidence has identified genetic (shared epitope [SE] and 76 single nucleotide polymorphisms [SNPs]) and several environmental risks for RA.

The 2,844 cases had serology (ACPA) measured through ELISA, using a cutoff of 25 U/mL as positive. There were 1,828 ACPA-positive cases and 1,016 ACPA-negative cases and 4,072 controls in the analysis. First-degree relatives of cases had serology (rheumatoid factor [RF]) determined based on their history of diagnoses of seropositive or seronegative RA in the Patient Register. Information on RF was also available for EIRA cases. Blood from EIRA participants was analyzed, and HLA-DRB1 was genotyped. HLA-DRB1*01, *04, and *10 were defined as SE alleles.

The authors used multiple imputation when information was missing for covariates and to accommodate the potential for selective participation among controls.

Cases were more likely to be smokers, to have one or more biological offspring, to be exposed to silica, and to carry copies of SE. Controls were more likely to be ever drinkers and were more frequent consumers of fatty fish.

Some 11.9% of ACPA-positive cases, 6.5% of ACPA-negative cases, and 3.5% of controls had at least one first-degree relative with RA. The crude familial odds ratio (OR) for RA was 3.05 (95% CI, 2.48-3.76), a ratio that did not change appreciably when adjusting for environmental risk factors individually.

The OR for familial risk declined moderately when adjusted for genetic risk factors. When adjusted for SE, the OR was 2.69 (95% CI 2.17-3.33), when adjusted for 76 SNPs, the OR was 2.55 (95% CI 2.03-3.20), and when adjusted for both, the OR was 2.44 (95% CI 1.93-3.07).

When adjusting for all genetic and nongenetic factors, the OR for familial risk was only slightly lower than ORs adjusted only for genetic factors.

A stronger influence from familial aggregation was found in ACPA-positive RA (crude OR: 4.10) than in ACPA-negative RA (crude OR: 1.61). Adjustment for environmental factors did not influence the familial risk of ACPA-positive RA, but adjustment for genetic risk factors did reduce the OR to 3.72 when adjusted for SE, to 3.46 when adjusted for 76 SNPs, and to 3.35 when adjusted for both.

In contrast, the familial risk of ACPA-negative RA decreased substantially when adjusting for genetic or nongenetic factors, whether or not the family history was with RF-positive RA (crude OR: 1.61; adjusted ORs range: 1.51-1.65) or RF-negative RA (crude OR: 2.28; adjusted ORs range: 2.15-2.32).

“Further studies trying to find genetic risk factors of RA or identify disease subtypes with a stronger familial risk are warranted, especially for ACPA-negative disease,” the authors declare.

A limitation of the study was the use of imputed values in the analysis where values were missing, but estimates from a complete case analysis supported the results from the imputed analysis, the authors note.

“Indeed, there is a lot we don’t yet understand regarding who gets this disease and why,” said Brian Mandell, MD, PhD of Cleveland Clinic. “Because particularly seronegative RA may be a syndrome, it is not surprising that we can’t identify risk factors which can identify a large majority of patients. The importance of family history suggests that as yet undefined genetic risks, in a polygenic manner, contribute to the likelihood of getting the disease.”

The authors declare no conflicts of interest.

###

Primary source: Arthritis and Rheumatism
Source reference: Jiang X, et al “To which extent may the familial risk of rheumatoid arthritis be explained by established RA risk factors?” Arth Rheum 2014; DOI: 10.1002/art.38927.