1584P - Germline genetic background contribution to metastatic dissemination in breast cancer extreme phenotype patients

Date 10 October 2016
Event ESMO 2016 Congress
Session Poster display
Topics Pathology/Molecular Biology
Basic Scientific Principles
Presenter Angela Santonja
Citation Annals of Oncology (2016) 27 (6): 545-551. 10.1093/annonc/mdw393
Authors A. Santonja1, B. Pajares2, B. Jiménez-Rodríguez2, C. Fernández-De Sousa2, N. Ribelles2, A. Lluch-Hernandez3, I. Catoira3, E. Perez-Ruiz4, M. Martin5, M. Del Monte-Millan5, A. Gonzalez-Neira6, G. Pita6, M.A. Pujana7, M. Ruiz8, N. Bonifaci7, J. De la Haba9, P. Sanchez-Rovira10, E. Alba2, A. Romero11
  • 1Instituto De Investigacion Biomedica De Malaga (ibima), Hospital Universitario Regional y Virgen de la Victoria, 29010 - Malaga/ES
  • 2Hospital Universitario Regional Y Virgen De La Victoria, Instituto de Investigacion Biomedica de Malaga (IBIMA), 29010 - Malaga/ES
  • 3Hematology And Medical Oncology, Hospital Clinico Universitario de Valencia, 46010 - Valencia/ES
  • 4Medical Oncology, Hospital Costa del Sol, Marbella/ES
  • 5Medical Oncology, Hospital General Universitario Gregorio Marañon, Madrid/ES
  • 6Spanish National Cancer Research Centre (cnio), CNIO- Spanish National Cancer Center, Madrid/ES
  • 7Breast Cancer And Systems Biology Unit, Catalan Institute of Oncology (ICO), Bellvitge Institute for Biomedical Research (IDIBELL), Barcelona/ES
  • 8Medical Oncology, Hospital Universitario Virgen del Rocio, Sevilla/ES
  • 9Medical Oncology, Hospital Provincial de Córdoba, Cordoba/ES
  • 10Medical Oncology, Complejo Hospitalario de Jaen Universidad de Jaen, Jaen/ES
  • 11Medical Oncology, Hospital Clinico San Carlos, Madrid/ES



Previous studies suggest that breast cancer dissemination is not only driven by the tumor somatic alterations but also by the host's genetic background. The main objective of this study is to identify germinal genetic variations predictive of metastatic dissemination in breast cancer patients.


Breast cancer patients with discordant extreme phenotypes were selected for a genome-wide association study (GWAS): low risk patients (10 positive lymph nodes) without relapse. Patients were distributed in a discovery set including controls (low risk without relapse and high risk with relapse) and a validation set. Around 4.3 million SNPs were genotyped with the HumanOmni5-Quad Beadchip (Illumina) from peripheral blood DNA. Data analysis was performed as an individual association analysis using Plink software, GenomeStudio (Illumina) and snpMatrix package from R and a pathways analysis with GSA-SNP (http://gsa.muldas.org).


We successfully genotyped 145 patients, 95 in the discovery set (47 cases and 48 controls) and 50 cases in the validation set. In the individual association analysis, the strongest associations were located in SNPs in/close to genes related with metastatic dissemination. One of the top SNPs was rs28452734 (p = 3.43x10−5, OR = 9.86 in discovery set; p = 2.96x10−3, OR = 4.3 in validation set). This SNP is located near XYTL1, a gene involved in the biosynthesis of glycosaminoglycan chains which form the extracellular matrix. XYLT1 has been previously associated with adhesion, proliferation, angiogenesis and metastasis. In turn, the pathways analysis identified an enrichment of genes from KEGG pathways including ECM-receptor interaction and cell adhesion and in three pathways previously published in association with metastasis 1) PIK3CA multipotent genetic program 2) early-stage metastatic disease or low burden signature 3) exosomal integrins and organ-specific metastasis.


These results suggest that there may be an association between patient's germline genetic background and breast cancer prognosis, independently or in conjunction with intrinsic tumor alterations.

Clinical trial identification

Legal entity responsible for the study

FIMABIS (Fundacion Publica Andaluza para la Investigacion de Malaga en Biomedicina y Salud)


Instituto de Salud Carlos III (Fondos de Investigacion Sanitaria ISCIII-FIS)


All authors have declared no conflicts of interest.