Replication of Genetic Polymorphisms Reported to Be Associated with Taxane-Related Sensory Neuropathy in Patients with Early Breast Cancer Treated with Paclitaxel—Letter

We have with great interest read the pharmacogenetic study by Abraham and colleagues (1) reporting SNPs associated with paclitaxel-induced neuropathy. The identification of markers predictive of sensory neuropathy is an important clinical problem for taxanes, vinca-alkaloids, platinum compounds, bortezomib, and thalidomide, among other anticancer drugs. In this respect, the study by Abraham and colleagues is a remarkably large study investigating 73 SNPs previously associated with taxane-related sensory neuropathy (TRSN) in 1,303 European individuals treated with paclitaxel (1). The authors found significant results for nine SNPs, including EPHA6-rs301927. Two genome-wide association studies (GWAS; refs. 2, 3) suggest EPHA5-rs7349683 as a neuropathy marker (meta-analysis P value of 1.4 × 10⁻⁹), and in our study, other members of the Eph receptor family members were also associated with paclitaxel-induced neuropathy (3).

To follow up our initial results suggesting that ephrin type A receptors are important factors influencing TRSN, we analyzed detailed neuropathy data, recorded cycle by cycle using the NCI-CTCAE, from 146 patients treated with first-line paclitaxel. Patients had either ovarian (72%) or breast cancer; 57 (39%) were prospectively recruited in Spain and 89 patients were from a previously described Danish cohort (4). The study was approved by the corresponding ethical review committees and was carried out in accordance with the Helsinki declaration. We genotyped 4 SNPs in EPHA4, EPHA5, EPHA6, and EPHA8 genes (rs17348202, rs7349683, rs301927, and rs209709, respectively) and 3 SNPs in XKR4, PIK3IP1, and SGCG genes (rs4737264, rs5749248, and rs1753097, respectively), all top signals in our GWAS (3). When tested against TRSN using a cumulative dose analysis, all SNPs in EPHA genes, except for EPHA4-rs17348202 (minor allele frequency = 0.05, indicating low statistical power), were associated with an increased neuropathy risk (Fig. 1). When analyzing the SNPs using maximum neuropathy grade, only EPHA6-rs301927 showed a trend toward increased toxicity (P = 0.069), suggesting that cumulative dose analysis is more sensitive to detect differences in neuropathy. No evidence of association was found for SNPs in other genes.

From a biologic perspective, Eph receptors represent a family of receptor kinases, involved in axon guidance and other neural-related functions, such as neuronal regeneration following nerve injury (5). Thus, this prospective study, together with that from Abraham and colleagues and previous reports, supports an increased TRSN risk for EPHA5-rs7349683 (2, 3), EPHA6-rs301927 (1, 3), and EPHA8-rs209709 (3). Furthermore, because EPHA proteins mediate neural injury repair, these SNPs could act as broad-spectrum neuropathy risk markers relevant for many neurotoxic drugs. Abraham and colleagues performed an exhaustive study of SNPs previously associated with TRSN; however, in view of these results, it would be interesting if the authors could further investigate these potentially clinically relevant markers (e.g., EPHA8-rs209709 and EPHA5-rs7349683 under different genetic models).

See the Response, p. 3094