PAM50 assay and the three-gene model for identifying the major and clinically relevant molecular subtypes of breast cancer

Abstract

It has recently been proposed that a three-gene model (SCMGENE) that measures ESR1, ERBB2, and AURKA identifies the major breast cancer intrinsic subtypes and provides robust discrimination for clinical use in a manner very similar to a 50-gene subtype predictor (PAM50). However, the clinical relevance of both predictors was not fully explored, which is needed given that a ~30 % discordance rate between these two predictors was observed. Using the same datasets and subtype calls provided by Haibe-Kains and colleagues, we compared the SCMGENE assignments and the research-based PAM50 assignments in terms of their ability to (1) predict patient outcome, (2) predict pathological complete response (pCR) after anthracycline/taxane-based chemotherapy, and (3) capture the main biological diversity displayed by all genes from a microarray. In terms of survival predictions, both assays provided independent prognostic information from each other and beyond the data provided by standard clinical-pathological variables; however, the amount of prognostic information was found to be significantly greater with the PAM50 assay than the SCMGENE assay. In terms of chemotherapy response, the PAM50 assay was the only assay to provide independent predictive information of pCR in multivariate models. Finally, compared to the SCMGENE predictor, the PAM50 assay explained a significantly greater amount of gene expression diversity as captured by the two main principal components of the breast cancer microarray data. Our results show that classification of the major and clinically relevant molecular subtypes of breast cancer are best captured using larger gene panels.

Figures

Fig. 1

Distant metastasis-free survival likelihood ratio statistics of subtypes defined by the PAM50 or the SCMGENE predictors, after accounting for clinical–pathological variables (age at diagnosis, nodal status, treatment and tumor size). Models were first conditioned on one predictor and the clinical–pathological variables, and then the significance of the other was tested. (A–B) Entire combined dataset (n = 2,008), (C–D) subset of patients that did not receive adjuvant systemic therapy (n = 994), (E–F) subset of patients with HR+ tumors that received adjuvant tamoxifen-only (n = 491). Similar results are obtained if a term for dataset is included in the model

Fig. 2

PC1 and PC2 loading plots of 3,316 samples using 18 Affymetrix and Agilent-based datasets taken from Haibe-Kains et al. [6]. Samples colored based on the a SCMGENE calls, or b PAM50 subtype calls. PC1 and PC2 R2 values obtained from simple linear regression models are shown. Only datasets with >50 % and <90 % ER+ tumors were included in this analysis. Blue Luminal A or ER+/HER2−/Low Proliferative, light blue Luminal B or ER+/HER2−/High Proliferative, pink HER2-enriched or HER2+, red Basal-like or ER−/HER2−, green normal-like, black normal breast samples (only present in the UNC337 dataset [29]). For the UNC337 dataset, we colored samples based on the subtype calls obtained after median centering as shown in Supplemental Fig. 1