Estrogen Receptor–Negative Breast Cancer: New Insights into Subclassification and Targeting

Commentary on Speers et al., p. 6327

In this issue of Clinical Cancer Research, Speers and colleagues (1) provide new insight into estrogen receptor (ER)-negative breast cancer. Breast cancer is not one disease, but many. Physicians have long realized that some breast cancers are responsive to hormonal manipulation; we now know that these cancers express the estrogen and/or progesterone receptors. In addition, in the mid-1980s, it became clear that about 25% of breast cancer overexpresses the Her2/Neu receptor tyrosine kinase as a result of a specific genomic amplification event (2). Analysis of gene expression microarray data recapitulated these clinical categories, and added two nuances. ER-positive Her2/Neu nonamplified breast cancers could be subdivided by their pattern of gene expression into two groups, the relatively indolent luminal A tumors, and the more aggressive luminal B tumors. Furthermore, transcriptional array profiling showed that, at least at first glance, the breast cancers that do not express hormone receptors or harbor Her2/Neu amplification, the so-called basal-like or triple negative breast cancers, seem to be a relatively homogeneous group (3).

The classification of breast cancers in this manner is not an idle academic pursuit; significant therapeutic implications are conveyed by these distinctions. Hormonal therapy targeting the ER signal transduction pathway is a mainstay of the treatment of ER-positive tumors. Her2/Neu amplified tumors can be treated with therapy targeting the Her2/Neu molecule, and the humanized monoclonal antibody trastuzumab has made a significant impact on the treatment of this subtype of breast cancer (4–6). At the present time, patients with triple negative tumors do not have the benefit of targeted therapy (see Fig. 1). We know that a fraction of triple negative tumors are responsive to standard chemotherapies (7, 8); however, at the moment we lack tools to predict which triple negative tumors will be chemotherapy sensitive, and even if we did have such tools, there is no established therapy to offer to those patients who are not chemotherapy sensitive.

Speers and colleagues (1) make interesting contributions to both the classification issue and the therapy issue. The authors use transcriptional microarray data to look for kinases that were differentially expressed in ER-positive compared with ER-negative breast tumors. Of 779 known or putative human kinases and kinase-interacting proteins, 86 were differentially expressed in these two sets of breast cancers at the P < 0.05 level. 52 of these 86 genes were expressed at a level at least twofold higher than in the ER-negative breast cancer group and were chosen for further study. The authors then used the expression of these 52 kinases to subclassify ER-negative tumors and cell lines; this produced four distinct subtypes of ER-negative breast cancer, designated by the authors as the map kinase cluster, the immunomodulatory cluster, the S6 kinase cluster, and the cell cycle checkpoint cluster, on the basis of the kinases highly expressed within each group. ER-negative cell lines also fell into the same four groups. Because the authors chose to focus on ER-negative breast cancer without regard to Her2/Neu amplification, ER-negative Her2/Neu amplified tumors were included in this study. Most of these tumors fell within the immune group. The map kinase group consisted almost entirely of luminal B tumors or Her2/Neu amplified tumors; the reasons for the presence of luminal B tumors among ER-negative breast cancers in this study are unclear (the vast majority are ER positive), but may relate to the thresholds used to determine ER negativity in this study. The S6 kinase group and the cell cycle checkpoint group seemed to subdivide the triple negative tumors into two distinct groups.

What potential clinical utility might offer this subdivision of ER-negative breast tumors? Speers and colleagues show that metastasis-free survival and overall survival differ among tumors divided into the groups defined above. Moreover, they suggest that some of these kinases, e.g., EPHB4, LIMK2, DAPK1, YES1, RYK, VRK2, PTK7, which are differentially expressed in ER-negative breast cancer, may serve as “druggable targets” for the treatment of ER-negative breast cancer, and in particular, for the treatment of basal-like triple negative breast cancer.

Kinases are key regulators of signaling pathways that control many cellular functions including proliferation, migration, and survival. Perturbation of kinase signaling by mutations, genomic amplification, or other genetic alterations is a frequent event central to tumorigenesis in many cancers. Fortunately, kinases have proven to be good druggable targets. For example, imatinib (also called Gleevec or STI571, Novartis) is the first FDA-approved drug of this new class of agents that directly block the signal of a kinase; Imatinib significantly benefits patients with chronic myelogenous leukemia (CML) featuring BCR-ABL fusions, and also patients with gastrointestinal stromal tumor (GIST). Since then, the development of molecularly targeted drugs that inhibit the action or activity of oncogenic kinases or their signaling pathways has been one of the most exciting and productive areas in cancer therapy. Gefitinib (Iressa, AstraZeneca) and erlotinib (Tarceva, Genentech/OSI/Roche) have been developed for the treatment of patients with lung cancer bearing mutations and/or amplification of the epidermal growth factor receptor (EGFR) tyrosine kinase. Trastuzumab (Herceptin, Genentech/Roche) and lapatinib (Tykerb, GlaxoSmithKline) have proven to be effective treatments for patients with Her2 positive breast cancer. Unfortunately, the 20% of breast cancer patients who suffer from triple negative breast cancer lacking expression of ER and Her2 do not have the advantage of benefiting from modern targeted therapies, and thus identification of a “druggable” target could have tremendous clinical impact on triple negative breast cancer therapy.

Interestingly, the current study finds that a number of the kinases that are highly expressed in ER-negative breast cancer also seem to be important for the growth of triple negative breast cancer cells (MDA-MB-468 and MDA-MB-231) but not for ER-positive breast cancer cells (MCF7 and T47D) in culture. However, this finding is only an early step toward targeted drug discovery, and many questions remain to be answered. For example, what are the mechanisms and biological consequences underlying and accompanying increased expression of these kinases in ER-negative breast cancer? Does elevated expression of these kinases actually render the tumors more sensitive to the relevant inhibitors? The authors show that siRNA-mediated inhibition of various kinases inhibit growth in cell culture; a critical issue for the future will be to see if small molecule kinase inhibitors that can be developed into drugs also have similar effect. To date, most kinase inhibitors have shown maximal clinical benefit in tumors in which the gene encoding the relevant kinase is either amplified, translocated, or harbors a gain-of-function mutation. Whether any of these changes are present in any of these kinases in ER-negative breast cancer is largely unknown. Recent studies have revealed that a high percentage of triple negative breast cancer have lost the tumor suppressor phosphatase and tensin homolog (PTEN; ref. 9). Loss of PTEN is perhaps the most common mechanism of activation of the phosphoinositide 3-kinase (PI3K) pathway. Notably, the importance of one of the PI3K isoforms, PIK3CB as a “driver” for cancer cell growth in the absence of PTEN has been recently shown (10, 11), making PI3KCB a promising target in PTEN-null tumors. Intriguingly, PIK3CB is also one of the kinases identified by Speers and colleagues to be highly expressed in triple negative breast cancer. It would be worthwhile to explore the correlation between PTEN loss and high expression of PIK3CB in triple negative breast cancer. Elucidation of the roles of these and many other kinase signaling pathways in triple negative breast cancer should provide important insights into molecular mechanisms of signaling action and help identify prime candidates for therapeutic intervention.

J.J. Zhao, consultant, Novartis Pharmaceuticals.