A 18FDG Uptake Gene Signature Predicts Prognosis After Radiotherapy In Breast Cancer

Abstract

Herein, we aimed to identify a predictive gene signature associated with FDG uptake for patients with breast cancer after having received post-operative radiotherapy by leveraging survival data in public gene expression data sets. By mining RNA expression profiling of samples, we developed a gene expression-based signature using gene set variation analysis (GSVA) algorithm. We performed survival analysis on Sjöström dataset and NKI dataset. The Sjöström dataset consisting of 172 patients with gene expression data was used to validate this signature, among which 118 patients having received radiotherapy. Furthermore, we investigated this signature in NKI cohort of 319 patients. All the patients in this cohort have received RT First, we identified a gene expression signature associated with FDG-uptake (SUVmax). Patients were divided into SUVmax-high and SUVmax-low groups among 172 patients from Sjöström dataset, survival analysis showed that there was no difference between SUVmax-high and SUVmax-low groups. However, among 118 patients after radiotherapy, patients with SUVmax-high score had better progression-free survival (PFS) and overall survival (OS) compared with SUVmax-low group. To further confirm the predictive value of this signature in RT-treated patients, we investigated FDG-uptake signature in NKI cohort of 319 patients with breast cancer after breast conserving therapy. Multivariate Cox regression analysis demonstrated that gene signature was an independent prognostic factor for patients after radiotherapy. Our results suggest that the patients classified as FDG uptake-high group may benefit most from radiotherapy. Further clinical validation is necessary to validate these findings.