This study examined the extent to which genetic variability modifies Transcutaneous Electrical Nerve Stimulation (TENS) effectiveness in osteoarthritic knee pain. Seventy-five participants with knee osteoarthritis were randomly assigned to either: (a) High-frequency TENS, (b) Low-frequency TENS or (c) Transient Placebo TENS. Pain measures were collected pre- and post-treatment. Participants were genotyped on genes implicated in central or peripheral pain pathways: NGFB, NTRK1, EDNRA, EDNRB, EDN1, OPRM1, TAC1, TACR1, BDNF, BDKRB1, 5HTT, COMT, ESR2, IL6 and IL1B. Genetic association using linear regression modelling was performed separately for the transient placebo TENS subjects, and within the High-frequency TENS+Low-frequency TENS participants, including TENS level as a covariate. In the placebo group, SNPs rs165599 (COMT) was significantly associated with an increased heat pain threshold (β=-1.87; p=.003) and rs6827096 (EDNRA) with an increased resting pain (β=2.68; p=.001). Within the treatment groups, TENS effectiveness was reduced by the SNP rs6537485 (EDNRA) minor allele in relationship to mechanical sensation (β=184.13; p=5.5E-9). Individuals with the COMT rs4680 minor allele reported lowered pain at rest after TENS (β=-42.30; p=.001), with a higher magnitude of pain reduction (28 unit difference) in the low-frequency TENS group compared to the high-frequency TENS group (β=28.37; p=.0004). EDNRA and COMT are implicated in osteoarthritic knee pain and provide a basis for tailoring TENS interventions according to individual characteristics. Findings from this study demonstrate that genetic variation within the COMT and EDNRA genes influences the effectiveness of TENS, a non-pharmacologic pain-reduction intervention, in the context of osteoarthritic knee pain. Evidence such as this may contribute to risk models that provide a clinically useful tool for personalizing TENS interventions according to individual characteristics in order to best control pain and maximize functional status.