Functional proteomics outlines the complexity of breast cancer molecular subtypes

Abstract

Breast cancer is a heterogeneous disease comprising a variety of entities with various genetic backgrounds. Estrogen receptor-positive, human epidermal growth factor receptor 2-negative tumors typically have a favorable outcome; however, some patients eventually relapse, which suggests some heterogeneity within this category. In the present study, we used proteomics and miRNA profiling techniques to characterize a set of 102 either estrogen receptor-positive (ER+)/progesterone receptor-positive (PR+) or triple-negative formalin-fixed, paraffin-embedded breast tumors. Protein expression-based probabilistic graphical models and flux balance analyses revealed that some ER+/PR+ samples had a protein expression profile similar to that of triple-negative samples and had a clinical outcome similar to those with triple-negative disease. This probabilistic graphical model-based classification had prognostic value in patients with luminal A breast cancer. This prognostic information was independent of that provided by standard genomic tests for breast cancer, such as MammaPrint, OncoType Dx and the 8-gene Score.

Figure 1

ER-true/TN-like subtype definition and characterization. Left panel: Hierarchical clustering analysis from 224 proteins identified by SAM analysis between ER+ and TNBC tumors with FDR < 5%. Right panel: Kaplan-Meier analysis showing survival for ER-true, TN-like and TNBC tumors (n = 51, 21 and 26, respectively; p = 0.17).

Figure 2

Protein- and miRNA-based probabilistic graphical model. Probabilistic graphical model showing protein (squares) and miRNA (circles) mean expression in each sample type. Color range from -2-fold change (green) to 2-fold change (red). White means no change between groups. ER-true subtype is compared with TN-like subtype and vice versa. TNBC type is compared with all ER+ tumors.

Figure 3

Tumor growth rate predicted by flux balance analysis. FBA results for ER-true, TN-like and TNBC tumors (n = 51, 21 and 26, respectively; *p < 0.05).

Figure 4

SRM validation of new subtypes. Kaplan-Meier analysis showing survival rates for ER-true and TN-like tumors on the basis of SRM data (n = 17 and 29, respectively).

Figure 5

Prognostic value of ER-true/TN-like subtype within breast cancer molecular subtypes. Kaplan-Meier analysis showing ER-true and TN-like tumor survival rates in luminal (A) (left panel: ER-true n = 262, TN-like n = 101) and luminal (B) (right panel: ER-true n = 59, TN-like n = 164) subtypes.

Figure 6

ER-true/TN-like subtype and prognostic signatures. Kaplan-Meier analysis showing survival rates of risk groups defined by prognostic gene signatures and ER-true/TN-like subtypes. (A) 70-gene Signature: Low risk = 586; High risk = 349; p < 0.0001; HR = 3.24 (2.73–4.85). (B) 70-gene Signature and ER-true/TN-like subtypes: Low risk/ER-true = 449; High risk/ER-true = 154; Low risk/TN-like = 137; High risk/TN-like = 195; p < 0.0001. (C) Recurrence Score: Low risk = 472; Intermediate risk = 195; High risk = 268; p < 0.0001. (D) Recurrence Score and ER-true/TN-like subtypes: Low risk/ER-true = 358; Intermediate risk/ER-true = 120; High risk/ER-true = 268; Low risk/TN-like = 125; Intermediate risk/TN-like = 108; High risk/TN-like = 143; p < 0.0001. (E) 8-gene Score: Low risk = 610; High risk = 325; p < 0.0001; HR = 2.61 (2.19–3.94). (F) 8-gene Score and ER-true/TN-like subtypes: Low risk/ER-true = 445; High risk/ER-true = 158; Low risk/TN-like = 165; High risk/TN-like = 167; p < 0.0001.