Secretome proteomics reveals candidate non-invasive biomarkers of BRCA1 deficiency in breast cancer

// Marc Warmoes 1, * , Siu W. Lam 1, * , Petra van der Groep 2, 3 , Janneke E. Jaspers 4, 5 , Yvonne H.C.M. Smolders 2 , Leon de Boer 1 , Thang V. Pham 1 , Sander R. Piersma 1 , Sven Rottenberg 5, 6 , Epie Boven 1 , Jos Jonkers 4 , Paul J. van Diest 2 , Connie R. Jimenez 1 1 Oncoproteomics Laboratory, Department of Medical Oncology, VU University Medical Center, Amsterdam, The Netherlands 2 Department of Pathology, University Medical Center Utrecht, Utrecht, The Netherlands 3 Department of Internal Medicine, University Medical Center Utrecht, Utrecht, The Netherlands 4 Division of Molecular Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands 5 Division of Molecular Pathology, The Netherlands Cancer Institute, Amsterdam, The Netherlands 6 Institute of Animal Pathology, Vetsuisse Faculty, University of Bern, Bern, Switzerland * These authors contributed equally to this work Correspondence to: Connie R. Jimenez, email: c.jimenez@vumc.nl Keywords: proteomics, biomarkers, BRCA1, breast cancer Received: May 11, 2016      Accepted: August 13, 2016      Published: August 23, 2016 ABSTRACT Breast cancer arising in female BRCA1 mutation carriers is characterized by an aggressive phenotype and early age of onset. We performed tandem mass spectrometry-based proteomics of secretomes and exosome-like extracellular vesicles from BRCA1 -deficient and BRCA1 -proficient murine breast tumor models to identify extracellular protein biomarkers, which can be used as an adjunct to current diagnostic modalities in patients with BRCA1 -deficient breast cancer. We identified 2,107 proteins, of which 215 were highly enriched in the BRCA1 -deficient secretome. We demonstrated that BRCA1 -deficient secretome proteins could cluster most human BRCA1 - and BRCA2-related breast carcinomas at the transcriptome level. Topoisomerase I (TOP1) and P-cadherin (CDH3) expression was investigated by immunohistochemistry on tissue microarrays of a large panel of 253 human breast carcinomas with and without BRCA1/2 mutations. We showed that expression of TOP1 and CDH3 was significantly increased in human BRCA1 -related breast carcinomas relative to sporadic cases ( p = 0.002 and p < 0.001, respectively). Multiple logistic regression showed that TOP1 (adjusted odds ratio [OR] 3.75; 95% confidence interval [95% CI], 1.85 - 7.71, p < 0.001) as well as CDH3 positivity (adjusted OR 2.45; 95% CI, 1.08 - 5.49, p = 0.032) were associated with BRCA1/2 -related breast carcinomas after adjustment for triple-negative phenotype and age. In conclusion, proteome profiling of secretome using murine breast tumor models is a powerful strategy to identify non-invasive candidate biomarkers of BRCA1 -deficient breast cancer. We demonstrate that TOP1 and CDH3 are closely associated to BRCA1 -deficient breast cancer. These data merit further investigation for early detection of tumors arising in BRCA1 mutation carriers.