Comprehensive multigene mutation spectra of breast cancer patients from Northeast China obtained using the Ion Torrent sequencing platform.

Abstract

The objectives of the present study were to obtain the multigene mutation spectra of female breast cancer patients in Northeast China, to explore the correlation between mutations and clinicopathological characteristics, and to identify genetic mutations that correlate with the prognosis and survival of breast cancer patients. An Ion Torrent sequencing platform was used to detect mutations, including 31 known gene mutations associated with breast cancer, in 621 specimens from 286 breast cancer patients. A total of 286 patients were enrolled in this study. Eleven harmful/pathogenic gene mutations were found in 54.2% (155/286) of the patients, and 179 somatic nonsynonymous mutations were detected. Approximately 5.6% (16/286) of the patients carried two or more gene mutations. Among the 11 pathogenic gene mutations, those in PIK3CA were the most common and were detected in 65.4% (117/179) of the patients; TP53 gene mutations were the second most common and were detected in 20.7% (37/179) of the patients. Additional mutations were found in AKT (14/179; 7.8%) and PTEN (4/179; 2.2%), and mutations in the remaining 7 genes were each detected in approximately 0.6% (1/179) of the patients. Excluding 6 cases of breast ductal carcinoma in situ, the remaining 280 breast cancer cases were divided into four groups by molecular subtype, and the mutation frequencies of the 11 breast cancer‑associated genes differed among the four groups. Furthermore, these 280 breast cancer cases were divided into two clinically relevant therapeutic groups: the HR+/HER2‑ and triple‑negative groups. The triple‑negative group had a high frequency of TP53 mutations (21.8%) and a low frequency of PIK3CA mutations (21.8%), whereas the HR+/HER2‑ group harbored TP53 mutations at a low frequency (10.1%) and PIK3CA mutations at a high frequency (50.0%). Cancerous, paracancerous, and normal tissues were collected from 72 patients and subjected to next‑generation sequencing. The types and frequencies of somatic nonsynonymous mutations differed among the three studied tissue types, reflecting the genetic heterogeneity of different tissues from the same individual. In addition, tissues from 70 patients (excluding 2 patients with ductal carcinoma in situ) were divided into four groups according to molecular subtype, and the gene mutation frequencies in cancerous, paracancerous, and normal tissues differed among the four groups. After normalization, gene mutations were detected at a higher rate in cancerous tissues than in paracancerous and normal tissues in all groups, except for the HER2‑positive group (which had a small sample size). In addition, Cox multivariate analyses of clinicopathological data, gene sequencing results, and 5‑year survival rates of the 286 patients showed that gene mutations in the PTEN‑PI3K/AKT signaling pathway were independently associated with a poor prognosis (P<0.05). In conclusion, mutations in the PTEN‑PI3K/AKT signaling pathway may be valuable in the prediction of the prognosis and survival of breast cancer patients.