CYP1A2--a novel genetic marker for early aromatase inhibitor response in the treatment of breast cancer patients

Abstract

Background: Endocrine resistance is a major obstacle to optimal treatment effect in breast cancer. Some genetic markers have been proposed to predict response to aromatase inhibitors (AIs) but the data is insufficient. The aim of the study was to find new genetic treatment predictive markers of AIs.

Methods: The ongoing population-based BC-blood study in Lund, Sweden includes women with primary breast cancer. This paper is based on AI-treated patients with estrogen receptor positive tumors who underwent breast cancer surgery in 2002-2008. First, an exploratory analysis of 1931 SNPs in 227 genes involved in absorption, distribution, metabolism, and elimination of multiple medications, using DMET™ chips, was conducted in a subset of the cohort with last follow-up in December 31st 2011 (13 cases, 11 controls). Second, selected SNPs from the first analysis were re-analyzed concerning risk for early breast cancer events in the extended cohort of 201 AI-treated with last follow-up in June 30th 2014. Clinical data were obtained from medical records and population registries.

Results: Only CYP1A2 rs762551 C-allele was significantly associated with increased risk for early events in the 24 patients (P = 0.0007) and in the extended cohort, adjusted Hazard ratio (HR) 2.22 (95% CI 1.03-4.80). However, the main prognostic impact was found within five years, adjusted HR 7.88 (95% CI 2.13-29.19). The impact of the CYP1A2 rs762551 C-allele was modified by a functional polymorphism in the regulator gene AhR Arg554Lys (G > A). Compared to patients who were homozygous for the major allele in both genes (CYP1A2 A/A and AhR G/G), a 9-fold risk for early events was found in patients who had at least one minor allele in both genes, adjusted HR 8.95 (95% CI 2.55-31.35), whereas patients with at least one minor allele in either but not both genes had a 3-fold risk for early events, adjusted HR 2.81 (95% CI 1.07-7.33). The impact of CYP1A2 rs762551 C-allele was also modified by the CYP19A1 rs4646 C/C, adjusted HR 3.39 (95% CI 1.60-7.16) for this combination. This association was strongest within the first five years, adjusted HR 10.42 (95% CI 3.45-31.51).

Conclusion: CYP1A2 rs762551 was identified as a new potential predictive marker for early breast cancer events in AI-treated breast cancer patients. Moreover, combined genotypes of CYP1A2 rs762551 and CYP19A1 rs4646 or AhR Arg554Lys could further improve prediction of early AI-treatment response. If confirmed, these results may provide a way to more personalized medicine.

Keywords: AhR; Aromatase inhibitor; Breast cancer; CYP19A1; CYP1A2; Polymorphisms; Treatment response.

Fig. 1

Flowchart illustrating patients included and excluded in the different analyses

Fig. 2

a-d Kaplan-Meier estimates of event-free survival in relation to CYP1A2 and CYP19A1 genotypes in AI-treated breast cancer patients with ER+ tumors are illustrated. LogRank P-values are presented for the entire follow-up time. In Fig. 2 a, b, and d, 5-year adjusted HRs are also presented. a CYP1A2 rs762551. The main association between CYP1A2 rs762551 any C-allele and early events was observed within 5 years of inclusion. b CYP19A1 rs4646. No significant association between CYP19A1 rs4646 and early events was observed. c Combinations of CYP1A2 rs762551 and CYP19A1 rs4646 genotype. Patients with any C-allele of CYP1A2 rs762551 and C/C genotype of rs4646 had a worse prognosis compared to patients with the three other genotype combinations. d CYP1A2 rs762551 any C-allele and rs4646 C/C. A combined variable of any C-allele of CYP1A2 rs762551 and C/C genotype of rs4646 was created and patients with this combination had a worse prognosis compared to patients with any other genotype. The main association was observed within 5 years of inclusion

Fig. 3

a Combinations of CYP1A2 rs762551 and Ahr Arg554Lys genotype. Patients with any C-allele of CYP1A2 rs762551 and any A-allele of AhR Arg554Lys had a worse prognosis followed by patients with at least one minor allele in either but not both genes and the lowest risk was seen in patients with the CYP1A2 rs762551 A/A genotype combined with the AhR Arg554Lys G/G genotype. b Three combinations of CYP1A2 rs762551 and Ahr Arg554Lys genotype. Patients who had a CYP1A2 A/A genotype and AhR any A-allele or CYP1A2 any C-allele and AhR G/G genotype were combined to one moderate risk group as the curves were similar in these groups. Patients with CYP1A2 rs762551 any C-allele and AhR Arg554Lys any A-allele had the highest risk for early events, followed by the combined group, compared to patients with CYP1A2 rs762551 A/A genotype and AhR Arg554Lys G/G genotype. Please note that there are fewer patients with longer follow-up times as this is an on-going cohort