Abstract P3-09-03: Single nucletotide polymorphisms of aromatase gene (CYP19A1) and toxicity of adjuvant aromatase inhibitors: A translational, prospective study

Abstract

Abstract Background: Extending adjuvant endocrine treatment (ET) with aromatase inhibitors (AI) to 7-10 years decreases the risk of relapse in hormone receptor-positive (HR+) breast cancer (BC). However, such benefit comes at the price of higher incidence of skeletal and cardiovascular (CV) events. Biomarkers predicting such toxicities might help clinicians in tailoring adjuvant ET to patient’s needs. We conducted a prospective study to assess whether SNPs in the gene encoding for the aromatase enzyme (CYP19A1) affect the risk of skeletal and CV events in HR+ early BC patients enrolled in the GIM4 trial. Methods: The GIM4 trial randomized HR+ BC postmenopausal patients who had been already treated with 2-3 years of adjuvant tamoxifen to either 3-2 years or 5 years of adjuvant letrozole. Four SNPs of CYP19A1 were evaluated: rs10046, rs4646, rs479292 and rs727479. SNPs were genotyped through PCR on DNA obtained from patients’ peripheral blood samples. Skeletal and CV events were assessed from randomization in the GIM4 trial to last follow-up, disease recurrence or death. Skeletal events were defined as the onset of osteoporosis or bone fractures. CV events were defined as the onset of thrombosis, embolism, stroke, myocardial infarction, hearth failure, or arrythmia. Univariate and multivariate logistic regressions were performed to evaluate the association between SNPs and skeletal and CV events. Bonferroni correction for multiplicity was used for univariate SNPs association tests, with a corrected alpha of 0.012. Only associations with an alpha level <0.012 were considered significant and included in multivariate models.Multivariate models’ covariates included: treatment arm (5 vs 3-2 years letrozole), (neo)adjuvant chemotherapy, previous hormone replacement therapy, age, BMI, previous biphosphonates, hypercholesterolemia and hypertension at baseline. SNPs were tested for Hardy–Weinberg equilibrium (HWE) and linkage disequilibrium (LD). Interaction between SNPs of interest and treatment arm was also investigated. Results: Of 2,056 patients enrolled in the GIM4 trial, 647 entered this translational study. All SNPs were in HWE. The CT and TT variants of rs10046 were less associated to skeletal events compared to variant CC, suggesting a dominant effect of the minor T allele (odds ratio [OR]: 0.47,. 95% confidence interval [CI]: 0.28-0.79, p=0.005). The association was independent from the other covariates (adjusted OR [aOR]: 0.43, p=0.002). Patients with rs10046-TT had no significant differences in the odd of skeletal events according to treatment arm. Conversely, in patients with rs10046-CC, skeletal events were more likely with 5-year letrozole (interaction p=0.043). Compared to rs10046-CC, rs10046-TT/CT was also less associated with CV events at both univariate (OR: 0.46 95% CI: 0.27-077, p=0.003) and multivariate analysis (aOR: 0.41, p<0.001).The rs727479—GG variant was associated with more CV events compared to the GT and TT variants (OR: 2.30, 95%CI: 1.27-4.15, p=0.009), suggesting a recessive effect of the minor G allele. The association remained significant at multivariate analysis (aOR: 2.30, p=0.010). No significant interaction was observed among treatment arms, rs10046 or rs272479 variants (interaction p=0.070 and 0.061). Rs10046-CC and rs727479-GG variants were in high LD (D’=0.92, haplotype-CG frequency: 39%). Hence, the association between genotype rs10046-CC/rs727479-GG and CV events was investigated. The association was significant at the multivariate analysis (aOR: 2.79, p=0.002). With such genotype, the odd of CV events was significantly increased with the 5-year letrozole (interaction p=0.046). Conclusions: SNPs of CYP19A1 could be useful biomarkers of long-term toxicity in HR+ BC patients who are candidates for adjuvant AI. Citation Format: Benedetta Conte, Chiara Molinelli, Giancarlo Bisagni, Antonio Durando, Giovanni Sanna, Stefania Gori, Ornella Garrone, Stefano Tamberi, Sabino De Placido, Francesco Schettini, Antonio Pazzola, Riccardo Ponzone, Filippo Montemurro, Gianluigi Lunardi, Rosario Notaro, Anna Turletti, Claudia Bighin, Francesca Poggio, Giulia Buzzatti, Matteo Lambertini, Luca Boni, Lucia Del Mastro. Single nucletotide polymorphisms of aromatase gene (CYP19A1) and toxicity of adjuvant aromatase inhibitors: A translational, prospective study [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-09-03.