Total joint replacement (TJR) is by far the most effective therapy for end-stage OA patients. Most of patients achieve joint pain reduction and function improvement following to TJR, however up to 22% of them either do not improve or deteriorate after surgery. The aim of this study was to identify genetic variants to be associated with poor outcome of TJR in primary OA patients by a genome-wide association approach (GWAS). Study participants were primary OA patients from the Newfoundland Osteoarthritis Study (NFOAS) that comprised total knee or hip replacement and recruited before 2016 in St. John's, NL. DNA samples were extracted from patients' blood. Study participants completed their pre-operation and 3.99±1.38 years post-surgery outcome assessment using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). DNA samples were genotyped using the genome-wide Illumina HumanOmni2.58 genotyping microarray containing 2.4 million SNPs. Pre-association quality control filtering was conducted for the raw genotyping data using PLINK 1.7 program, and genotype imputation was performed using the IMPUTE2 algorithm with multiple population reference data from 1000 Genome Project. The imputed data with ∼3.1 million variants was used to test the association with non-responders to TJR using the additive genetic model. Eighty three primary OA patients (44 responders and 39 non-responders) were included in the analysis. Association analysis detected three chromosomal regions on chr5, 7, and 8 to be significantly associated with non-responding to pain. The top SNPs at these loci are intergenic variants that include SNP (rs17118094, p=4.4×10-5) on chr5. This SNP is adjacent to SGCD gene that plays an important role in muscular strength and maintenance. Another associated SNP (rs71572810, p=4.7×10-5) is nearby IMMP2L gene on chr7. This gene is reported to be associated with behavioral abnormalities. Finally, SNP (rs6992938, p=5.8×10-5) on chr8 is located downstream of TRPA1 gene that is known to have a central role in the pain response to endogenous inflammatory mediators. Three loci were also found to be significantly associated with non-responding to function. The lead variant in the locus on chr1 is an intergenic SNP (rs9729377, p=1.7×10-5) falling between CTBS and MCOLN2 genes. CTBS gene is associated with TNF-α, a cytokine that stimulate the inflammation acute phase reaction, and MCOLN2 gene plays a role in the chemokine secretion and macrophage migration in the innate immune response. Other top SNPs in loci on chr2 and 10 harbor CCDC93, INSIG2, and KLF6 genes that are associated with heel bone mineral density, hypercholesterolemia, obesity and BMI. To our knowledge, this project is the first study that investigated the association between genetic factors and TJR non-responders. Our results demonstrated that genes related to muscle strength, behavioral trait, pain response, and inflammation play a significant role in poor outcome of TJR, warranting further investigation.