Most studies have used genomic profiles to dissect the heterogeneity of breast cancer (BC) to improve clinical outcome, however, very few studies pay attention to protein profiles. Thus, we aimed to explore the prognostic value of reverse phase protein array (RPPA)-based proteins and integrated them into a nomogram with clinical variables to predict outcomes in non-metastatic BC after surgery.
From the Cancer Genome Atlas (TCGA) database, we identified 767 non-metastatic BC patients with RPPA-based proteins data and clinical features, and randomly classified them into the discovery (n = 460) and validation (n = 307) cohorts. A least absolute shrinkage and selection operator (LASSO) Cox’s proportional hazard model was used to build the RPPA-based proteomic signature. A nomogram was constructed combing the proteomic signature and some predictive clinical variables. And the predictive accuracy of the predictive models was assessed with Brier score and Harrell’s concordance index (c-index).
Using the LASSO Cox model, we built a RPPA-based proteomic signature (RPSBC) for BC based on 15 proteins: ATM, B-Raf, Bim, CD49b, Heregulin, Ku80, Lck, P-Cadherin, PDK1, PRAS40pT246, RbpS807_S811, SETD2, SrcpY527, Tuberin, eEF2K. To develop a more quantitative model to predict recurrence, a nomogram (nomoRCBC) combining RPSBC and clinical features (age, T stage, N stage, menopause status, and histological type) was constructed. Using recurrence scores calculated in the nomoRCBC, patients were classified into high-risk or low-risk groups. Between these groups, recurrence-free survival was significantly different in the discovery and validation (hazard ratio[HR]: 2.95,95%CI 1.51-5.77; p = 0.002) cohorts. In addition, the nomoRCBC had better prognostic validity than the TNM stage and other published predictive models. Moreover, this nomoRCBC could be used to predict which patients might benefit from adjuvant chemotherapy after surgery for breast cancer.
Our nomoRCBC is a reliable prognostic and predictive tool for survival in patients with non-metastatic BC, which may facilitate personalized adjuvant chemotherapy.
The results published here are in whole based upon data generated by The Cancer Genome Atlas pilot project established by the NCI and NHGRI. Information about TCGA and the investigators and institutes who constitute the TCGA research network can be found at http://cancergenome.nih.gov.
The authors.
Grants from the National Natural Science Foundation of China (Nos. 81472386, 81272340, and 81572901), the National High Technology Research and Development Program of China (863 Program) (No. 2012AA02A501), and the Science and Technology Planning Project of Guangdong Province, China (Grant Nos. 2014B020212017 and 2014A020209024).
All authors have declared no conflicts of interest.