Publication | Closed Access
Genetic associations with taxane-induced neuropathy by a genome-wide association study (GWAS) in E5103.
44
Citations
0
References
2011
Year
NeurogenomicsGeneticsCancer ManagementGenetic EpidemiologyPathologyPharmacotherapyDisease Gene IdentificationPre-clinical PharmacologyGenetic AssociationsNeuro-oncologyTaxane-induced NeuropathyGenome-wide Association StudyMetronomic TherapyClinical TrialsNeurologyRadiation OncologyCancer ResearchMolecular OncologyNeurogeneticsSix SnpsMedicineStandard Adjuvant ChemotherapyNeurodegenerationSeveral SnpsCancer TreatmentPharmacologyNeurodegenerative DiseasesPrognostic BiomarkersCancer RiskCancer EpidemiologyNeuroscienceOncology
1000 Background: Neuropathy, one of the most common toxicities associated with taxane therapy, may be severe, function-limiting, and sometimes irreversible. Established predictors for increased risk include advanced age, diabetes, and type/dose/schedule of taxane. No established biomarkers have been identified to predict patients at greatest risk. Methods: E5103 is a randomized phase III trial comparing standard adjuvant chemotherapy for patients with early stage breast cancer with the same chemotherapy plus concurrent bevacizumab or concurrent and sequential bevacizumab. All 3 arms include weekly paclitaxel for 12 weeks. A GWAS was performed using the Infinium Human Omni1 array on 2204 patients. The primary hypothesis was to identify SNPs associated with time to report of first grade 2−4 neuropathy. SNP data underwent rigorous review to assess both SNP and sample quality. Analyses were performed using Cox regression model including established clinical trial covariates. Bonferroni corrections for multiple comparisons were made. Results: Interim toxicity data from E5103 demonstrated that 613 patients had grade 2−4 neuropathy and 1591 did not. Significant clinical predictors of neuropathy included age (12.9% increase with each 10yrs; p=0.004) and African American race (HR=2.1; p=4.5 ×10−11). Six SNPs with MAF>5% were associated with time to neuropathy (p<5×10−7). These SNPs resided in two genes: RWDD3 and TECTA. A missense SNP in RWDD3 was associated with likelihood of neuropathy at 15 months: 27% for patients with homozygous wild-type, 40% for heterozygous, 60% homozygous variant (allele dose-effect: HR=1.5; p=8.5 ×10−8). A TECTA SNP was associated with likelihood of neuropathy at 15 months: 29% for homozygous wild-type, 32% for heterozygous, and 57% for homozygous variant (recessive-effect: HR=2.1; p=3.2 ×10−7). Multiple other SNPs with MAF<5% were also associated with neuropathy (p<5×10−7). Analysis in the African American subgroup is underway. Conclusions: Using a genome wide approach, we have found several SNPs associated with time to neuropathy in patients undergoing paclitaxel therapy. This is the first time a genetic predictive biomarker has been reported for taxane-induced neuropathy.