Abstract P2-09-11: Metabolite-guided long-term prediction of outcome in tamoxifen treated breast cancer patients

Abstract

Abstract Background The genetic polymorphisms of CYP2D6 determine its enzymatic activity thereby the pharmacokinetic velocity by which tamoxifen (tam) is converted into active metabolites. Combined analyses of tam metabolites compared to CYP2D6 type and long-term survival could allow prediction of tam response and personalization of therapies. The clinical importance of defining such subgroups in the era of the new long-term (10-year) tam treatment paradigm is actualized. Material and Methods From May 1995 to December 1998, patients were included in an observational micro-metastasis study in the Oslo region and treated according to the national guidelines at the time (1). Serum samples were drawn at 3-year follow-up from 356 relapse-free patients, 106 of these were treated with tamoxifen. The median follow up time for breast cancer death was 16.8 years (3.5-19.4). Serum samples were processed using protein precipitation with acetonitrile. An Aquity UPLC system was used to chromatographically separate 10 tam metabolites using a BEH C18 Phenyl column (100 x 2.1 mm, 1.7 μm particle size) that was developed with a water-methanol gradient containing 0.1% formic acid. The LC system was coupled to a Xevo TQ-S tandem mass spectrometer equipped with an atmospheric pressure photoionization source. The method was validated with respect to linearity, imprecision, accuracy, and functional sensitivity according to FDA guidelines. Results The new LC-MS/MS method separated the active Z-isomers of 4OHtam and 4OHNDtam (endoxifen) from its inactive Z'-isomers and E-isomers. Imprecision (intra and inter-day CV %) was within 10 % for target concentrations for all metabolites and accuracies were in the range 95-106%. The method was validated with serum samples from 42 breast cancer patients using 20 mg of tamoxifen. The endoxifen concentrations ranged from 0 to 90 nM, with a median value of 25 nM. The previous observed endoxifen level of 10 nM in poor metabolizers (2) was used as cut-off for the grouping of patients. The nil endoxifen (NE) group (< 0.1nM, n=14) or low-endoxifene (LE) group (0.1-10 nM, n=8) were grouped together. Univariate survival analysis did not show a significant association between breast cancer specific survival and endoxifen levels. (p=0.15; logrank and p=0.18;Breslow). However, for the period beyond 10-years of follow-up the breast cancer survival differed between the high endoxifene (HE) group and the NE+LE groups. For patients surviving the first 10 years the breast cancer specific survival was 94.2% vs. 77.8% for the HE and NE+LE groups respectively (p=0.020, logrank and p=0.017, Breslow, HR=4.5, CI 95=1.1-17.9). In the multivariate analysis endoxifen ≤/> 10 nM remained the only factor in the final model. Discussion We developed a new accurate and precise LC-MS/MS method for the measurement of 10 tamoxifen metabolites. Importantly, the method separates active and inactive isomers of 4OHtam and 4OHNDtam/endoxifen. Despite the low number of patients, we observed a poorer long-term survival beyond 10 years in patients with nil or low serum concentration of endoxifen. A comprehensive analysis is presented addressing the relationship between genotyped based and metabolite based prediction of long-term outcome in tamoxifen treated breast cancer patients. Ref: 1. Wiedswang G, et al. Clin Ca Res; 2004 2. Mürdter TE, et al. Clin Pharmacol Ther; 2011. Citation Format: Helland T, Henne N, Bifulco E, Hustad SS, Kristensen VN, Lash T, Borgen E, Janssen EAM, Lien EA, Naume B, Mellgren G, Søiland H. Metabolite-guided long-term prediction of outcome in tamoxifen treated breast cancer patients [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P2-09-11.