Individualized Prediction of Survival by a 10-Long Non-coding RNA-Based Prognostic Model for Patients With Breast Cancer

• Main Authors: Xuemei Yang, Juan Li, Yifan Wang, Peilong Li, Yinghui Zhao, Weili Duan, Abakundana Nsenga Ariston Gabriel, Yingjie Chen, Haiting Mao, Yunshan Wang, Lutao Du, Chuanxin Wang
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2020-10-01
• Series: Frontiers in Oncology
• Subjects: breast cancer; long non-coding RNA; prognosis; signature; nomogram
• Online Access: https://www.frontiersin.org/articles/10.3389/fonc.2020.515421/full

Deregulations of long non-coding RNAs (lncRNAs) have been implicated in the progression of breast cancer (BC). However, the prognostic values of those lncRNAs in BC remain elusive. This study aimed at constructing a lncRNA-based prognostic model to improve the clinical management of BC. Systematic investigation of lncRNA expression profiles and clinical data from The Cancer Genome Atlas (TCGA) database were utilized to establish a 10-lncRNA signature. The prognostic signature efficiently discriminated patients with significantly different prognosis regardless of intrinsic molecular subtypes and tumor–node–metastasis (TNM) stage. A combined model was constructed by multivariate Cox proportional hazards regression (CPHR) analysis, which combined the lncRNA-based signature with certain clinical risk factors (TNM stage, age, and human epidermal growth factor receptor 2 status). This model predicted a survival probability that closely corresponds to the actual survival probability. With respect to the entire set, the time-dependent receiver-operating characteristic curves revealed that the area under the curve of this model was the highest than any of the clinical risk factors. Moreover, functional enrichment analysis indicated that the molecular signature was mainly involved in DNA replication, which was firmly related to BC tumorigenesis. Consistent with the discovery, the knockdown of LHX1-DT, one of the 10 prognostic lncRNAs, attenuated the proliferation of BC cells in vitro and in vivo. Taken together, our study constructed a novel 10-lncRNA signature for prediction prognosis, and the signature-based model could provide new insight into accurate management of BC patients.