Related Experiment Video
Updated: Jun 22, 2025
Performing Data Mining And Integrative Analysis Of Biomarker in Breast Cancer Using Multiple Publicly Accessible Databases
Published on: May 17, 2019
Effect of cross-platform gene-expression, computational methods on breast cancer subtyping in PALOMA-2 and PALLET
Maggie Chon U Cheang1, Mothaffar Rimawi2, Stephen Johnston3
1The Institute of Cancer Research, Sutton, UK. maggie.cheang@icr.ac.uk.
Abstract:
Intrinsic breast cancer molecular subtyping (IBCMS) provides significant prognostic information for patients with breast cancer and helps determine treatment. This study compared IBCMS methods on various gene-expression platforms in PALOMA-2 and PALLET trials. PALOMA-2 tumor samples were profiled using EdgeSeq and nanostring and subtyped with AIMS, PAM50, and research-use-only (ruo)Prosigna. PALLET tumor biopsies were profiled using mRNA sequencing and subtyped with AIMS and PAM50. In PALOMA-2 (n = 222), a 54% agreement was observed between results from AIMS and gold-standard ruoProsigna, with AIMS assigning 67% basal-like to HER2-enriched. In PALLET (n = 224), a 69% agreement was observed between results from PAM50 and AIMS. Different IBCMS methods may lead to different results and could misguide treatment selection; hence, a standardized clinical PAM50 assay and computational approach should be used.Trial number: NCT01740427.
