Although the clinical utility of gene expression signatures for breast cancer is established, access to these prognostic tools is limited to specific patient profiles. The purpose of our study was to identify a set of gene transcripts associated with the likelihood of recurrence in HR+/HER2- breast cancer patients with the highly efficient TempO-Seq platform.
By targeted expression analysis of FFPE lysate samples from the NHS Greater Glasgow and Clyde biorepository a machine learning approach idendified a subset of transcripts that were prognostic for recurrence. A Cox proportional hazard model for 10-year recurrence was subsequently trained using these transcripts with HR+ samples that lacked overexpression of the HER2 receptor (n = 211) and tested in samples independent of training (n = 50).
The transcript panel analysed on the TempO-Seq platform provided prognostically significant information for breast cancer recurrence in our sample set (LRχ2 = 78.88, p < 0.001; HR = 2.41). These data indicated that patients could be grouped into low- and high-risk cohorts, with significantly different Kaplan-Meier curves for recurrence free survival (p < 0.001) and breast cancer-associated mortality (p = 0.013). Additionally, the prognostic information provided by the TempO-Seq-derived signature was independent of the UK-PREDICT tool (V2), suggesting complementary applications for the technology.
Our novel prognostic signature and testing method based on TempO-Seq profiling was developed using whole transcriptome analysis of a heterogeneously treated cohort in an adjuvant setting. This approach overcomes many of the technical limitations (e.g., does not require extraction) and cost barriers typically encountered with traditional molecular approaches such as qPCR, RNA-Seq, or molecular barcoding. The high throughput test can rapidly generate results from large numbers of patient samples for efficient turnaround times. More widely, the study demonstrates that TempO-Seq technology can be used for accelerated diagnostic panel development and clinical research. The described approach to estimate the risk of breast cancer recurrence can be further validated in prospectively designed trials.
Legal entity responsible for the study
BioClavis Ltd. Teaching and Learning Centre Queen Elizabeth University Hospital, 1345 Govan Rd, Glasgow G51 4TF, United Kingdom.
Has not received any funding.
All authors have declared no conflicts of interest.