Skip to main content
Premium Trial:

Request an Annual Quote

Proteogenomic Analysis Uncovers New Breast Cancer Subtypes, Suggests Treatment Options

NEW YORK – A new proteogenomic analysis of breast cancer samples points to tumor subtypes that may be vulnerable to immune checkpoint or other therapies. 

By combining tumor DNA and RNA sequencing with mass spectrometry-based proteomics, researchers from the Clinical Proteomic Tumor Analysis Consortium aimed to deeply profile tumors to identify modifications that could suggest potential treatment approaches.

In their new study, the researchers analyzed more than 100 breast cancer tumors, examining not only their genomic landscape, but also their proteome, phosphoproteome, and acetylproteomes. As they reported on Wednesday in Cell, they used this data to cluster the samples into four groups that partially overlapped with their PAM50 gene signature-based classifications, but with some differences. Additionally, the tumor profiles revealed potential vulnerabilities of different subtypes to various treatment approaches.

"We offer the first comprehensive report of the [breast cancer] acetylome; present testable hypotheses regarding therapeutic vulnerabilities, cancer biology, and advancement of diagnostic standards; and provide an extensive resource to stimulate further discovery," senior author Michael Gillette from the Broad Institute and his colleagues wrote in their paper.

The researchers noted that their previous effort to conduct a proteogenomic analysis of breast cancer was stymied by how the samples were collected. For this study, they prospectively collected 122 primary tumor samples under a strict protocol to prevent tissue ischemia and the loss of post-translational modifications. The samples were cryo-pulverized and DNA, RNA, and protein extracted from the homogenized samples for analysis.

A multi-omic-based clustering analysis separated the samples into four groups. Two clusters broadly reflected the tumors' PAM50 classifications of luminal A-inclusive and basal-inclusive, but the two other clusters did not align with PAM50 designations. The luminal B-inclusive cluster, for instance, included all but one PAM50 LumB case in addition to a subset of PAM50 LumA samples. The HER2-inclusive cluster, meanwhile, was heterogeneous and included mostly HER2-enriched samples, but also tumors from other PAM50 subtypes. This suggests that the PAM50 classification scheme may overlook distinctions between LumA and LumB tumors, the researchers noted.

The analysis further highlighted the different pathways that go awry in the different tumor subtypes. For instance, the researchers uncovered increased acetylation of TCA cycle and beta-oxidation proteins within their Basal-I cluster and of glucose metabolism and interleukin-1 signaling proteins in their LumB-I clusters. At the same time, they noted increased activation of glycolysis and serine synthesis pathways in the Basal-I cluster.

Moreover, the analysis indicated that some tumors might be susceptible to certain therapies. For instance, it uncovered subsets of LumA and LumB tumors with APOBEC-mediated mutagenesis or single-strand DNA repair defects that could possibly respond to immune checkpoint inhibitor treatment.

The researchers further suggested that a proteogenomic analysis of the tumor suppressor Rb might serve as a predictive marker of ER+ breast cancer. They also noted that Rb protein levels were generally correlated with response to the CDK4/5 inhibitor palbociclib, no matter the Rb1 genotype. A proteogenomic analysis of Rb could then indicate whether CDK4/5 inhibitor treatment might be effective.

Additionally, using their proteogenomic approach, the researchers teased out some tumors that were determined via immunohistochemistry to be ERBB2-positive, but that were actually cases of pseudo-ERBB2 positivity, where anti-ERBB2 treatment might not be effective. For instance, one case of pseudo-ERBB2 appeared to be due to the amplification and overexpression of TOP2A.

"Our results underscore the potential of proteogenomics for clinical investigation of breast cancer through more accurate annotation of targetable pathways and biological features of this remarkably heterogeneous malignancy," the researchers wrote.