Radiographic osteoarthritis at three joint sites and FRZB, LRP5, and LRP6 polymorphisms in two population-based cohorts.

OBJECTIVE

To examine the association of genetic variation in key players in the Wnt signaling pathway with aspects of osteoarthritis (OA) in two population-based cohort studies: the Rotterdam Study and the Chingford Study.

METHODS

Radiographic OA (ROA) was defined as a Kellgren/Lawrence score (K/L) score > or = 2 for the knee and hip. Total hip replacement (THR) was scored. Hand OA was defined as presence of ROA (K/L > or = 2) in two out of three hand joint groups [distal interphalangeal (DIPs), proximal interphalangeal (PIPs), first carpometacarpal (CMC1)/trapezio-scaphoid joint (TS)] of each hand. The concentration of urinary C-terminal cross-linked telopeptide of type II collagen (CTX-II) was standardized to the total urine creatinine. Genotypes for the amino acid variants, Arg200Trp and Arg324Gly of Frizzled-Related protein gene (FRZB), Ala1330Val of Low-density lipoprotein receptor-related protein 5 (LRP5) and Ile1062Val of Low-density lipoprotein receptor-related protein 6 (LRP6), were obtained using the Taqman allelic discrimination assay. A meta-analysis was performed for the FRZB Arg324Gly polymorphism and hip- and knee-OA using RevMan version 4.3.

RESULTS

No consistent associations were observed between the FRZB, LRP5 and LRP6 amino acid variants and radiographic hip-, knee-, or hand-OA or THR, in either study population. While power was limited for most studies to date, a meta-analysis of all published studies regarding the FRZB Arg324Gly polymorphism was performed for hip- and knee-OA separately. This showed no significant associations between the Gly324 allele and risk for hip- or knee OA, although there was large heterogeneity between studies for hip OA in females.

CONCLUSION

No association was seen between FRZB, LRP5 and LRP6 variants with radiographic osteoarthritic outcomes in two population-based cohorts. In future studies, increased power and standardization of OA-phenotypes are highly recommended for replication studies and to allow meta-analysis.