46P - Metabolomic analysis as a tool to identify breast cancer (BC) cell lines resistant to palbociclib (PD)

Date 04 May 2017
Event IMPAKT 2017
Session Welcome reception and Poster Walk
Topics Breast Cancer
Presenter Martina Bonechi
Authors M. Bonechi1, C. Guarducci1, G. Meoni2, L. Tenori3, C. Biagioni4, R. Schiff5, C.K. Osborne5, C. Luchinat6, L. Malorni7, I. Migliaccio1
  • 1Sandro Pitigliani Translational Research Unit, Hospital of Prato, Azienda USL Toscana Centro, 59100 - Prato/IT
  • 2Magnetic Resonance Center (cerm), University of Florence, 50019 - Sesto Fiorentino (FI)/IT
  • 3Department Of Experimental And Clinical Medicine, University of Florence, 50134 - Florence/IT
  • 4Sandro Pitigliani Medical Oncology Department, Hospital of Prato, Azienda USL Toscana Centro, 59100 - Prato/IT
  • 5Lester And Sue Smith Breast Center, Dan L Duncan Comprehensive Cancer Center, Baylor College of Medicine, 77030 - Houston/US
  • 6Magnetic Resonance Center (cerm) And Department Of Chemistry, University of Florence, 50019 - Sesto Fiorentino (FI)/IT
  • 7Sandro Pitigliani Translational Research Unit And Sandro Pitigliani Medical Oncology Department, Hospital of Prato, Azienda USL Toscana Centro, 59100 - Prato/IT

Abstract

Body

Introduction: To date, no biomarker has been validated to select patients for treatment with PD. We hypothesize that PD might induce metabolic changes in BC cell lines at the time of resistance. We have previously shown that nuclear magnetic resonance (NMR) spectroscopy discriminates HR+/HER2neg BC cell lines with acquired resistance to PD (PDR) from their sensitive counterpart (PDS). In this study we extended our analyses to HR+/HER2+ and HR+ endocrine resistant BC models.

Material and methods: A panel of PDR HR+ BC models (MCF7, T47D, ZR75-1, BT474, MDAMB361 and two MCF7 endocrine resistant derivatives) has been established in our lab by chronically exposing cells to escalating doses of PD. Whole-cell lysates and conditioned cell culture media from five replicates of each of the PDS and PDR models and from HR+/HER2neg PDS cells treated with PD for 3 days (D3) were analyzed by NMR. Principal Component Analysis (PCA) was used as first exploratory analysis and as dimension reduction technique. Canonical Analysis was used to discriminate different groups. K-nearest neighbors method (k=3) was used to classify the samples. Analysis of NMR-based metabolites was performed.

Results: Individual cell lines were correctly identified by an unsupervised PCA analysis of NMR spectra on both cell lysates and conditioned media. Using a supervised approach and the related cross-validation analysis, PDS cells were discriminated from PDR with an accuracy of 90% and 81% in cell lysates and conditioned media, respectively. D3 cells occupied a distinct position in the metabolic space between PDS and PDR groups. A comparison of NMR-based metabolites in cell lysates showed that glycerophosphocholine (GPC) levels were significantly higher in PDR compared to PDS cell lines (adjusted p-value 0.0064). Interestingly, D3 cells showed even higher GPC levels compared to PDR (adjusted p-value 0.0002).

Conclusions: In this study we confirm that a discrimination of PDR from PDS cell lines based on NMR spectra is possible. Perturbations of GPC metabolism might be implicated in the response to PD and the acquisition of PDR phenotype. Further studies are warranted to assess the role of metabolomics in patients receiving PD.

Clinical trial identification