Dingxie Liu1,2. 1. Bluewater Biotech LLC, PO Box 1010, New Providence, NJ, 07974, USA. dliu@bluewater-biotech.com. 2. Division of Endocrinology, Diabetes & Metabolism, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, 21287, USA. dliu@bluewater-biotech.com.
Abstract
PURPOSE: The development of multi-gene signatures has led to improvements in identification of breast cancer patients at high risk of recurrence. The prognostic power of commercially available gene signatures is mostly restricted to estrogen receptor (ER)-positive breast cancer. On the contrary, immune-related gene signatures predict prognosis only in ER-negative breast cancer. This study aimed to develop a better prognostic signature for breast cancer. METHODS: The expressions of long non-coding RNA (lncRNA) genes from 30 independent microarray datasets with a total of 4813 samples were analyzed. A prognostic lncRNA signature was developed based on likelihood-ratio Cox regression analysis. Survival analysis was used to compare the prognostic efficiencies of our signature and 10 previously reported prognostic gene signatures. RESULTS: Cox regression analysis on 30 independent datasets showed that the 6-lncRNA signature identified in this study performed as well as five commercially available signatures in recurrence prediction for ER-positive breast cancer. In ER-negative breast cancer, this lncRNA signature was as prognostic as three immune-related gene signatures. Moreover, our lncRNA signature also demonstrated a good capacity to predict recurrence risk for triple-negative breast cancer. Function analysis showed that several lncRNAs in this signature were probably involved in cell proliferation and immune processes. CONCLUSIONS: A six-LncRNA signature was identified that is prognostic for ER-positive, ER-negative, and triple-negative breast cancers and thus deserves further validation in prospective studies.
PURPOSE: The development of multi-gene signatures has led to improvements in identification of breast cancerpatients at high risk of recurrence. The prognostic power of commercially available gene signatures is mostly restricted to estrogen receptor (ER)-positive breast cancer. On the contrary, immune-related gene signatures predict prognosis only in ER-negative breast cancer. This study aimed to develop a better prognostic signature for breast cancer. METHODS: The expressions of long non-coding RNA (lncRNA) genes from 30 independent microarray datasets with a total of 4813 samples were analyzed. A prognostic lncRNA signature was developed based on likelihood-ratio Cox regression analysis. Survival analysis was used to compare the prognostic efficiencies of our signature and 10 previously reported prognostic gene signatures. RESULTS: Cox regression analysis on 30 independent datasets showed that the 6-lncRNA signature identified in this study performed as well as five commercially available signatures in recurrence prediction for ER-positive breast cancer. In ER-negative breast cancer, this lncRNA signature was as prognostic as three immune-related gene signatures. Moreover, our lncRNA signature also demonstrated a good capacity to predict recurrence risk for triple-negative breast cancer. Function analysis showed that several lncRNAs in this signature were probably involved in cell proliferation and immune processes. CONCLUSIONS: A six-LncRNA signature was identified that is prognostic for ER-positive, ER-negative, and triple-negative breast cancers and thus deserves further validation in prospective studies.
Entities:
Keywords:
Breast cancer; Cancer recurrence; Estrogen receptor; Gene signature; Long non-coding RNA; Triple-negative breast cancer