Development of personalized therapies for deadly women’s cancers


Two recent research reports may point the way to developing more effective, personalized therapies for two deadly women’s cancers for which their are currently few treatment options–triple-negative breast cancer and ovarian cancer. The approach followed in both reports is to use gene expression analysis to stratify each of the two diseases into subtypes. Researchers can then use gene expression and order aspects of the biology of each subtype to design subtype-specific targeted therapies, whether single drugs or drug combinations. If the drugs (whether approved or experimental) already exist, they can be tested in clinical trials, stratified by subtype. If no appropriate drugs exist, researchers can discover the drugs based on subtype-appropriate drug targets.

Triple-negative (TN) breast cancer refers to breast cancers that are negative for expression of estrogen receptor (ER), progesterone receptor (PR), and HER2. [HER2 is the target of trastuzumab (Roche/Genentech's Herceptin) and lapatinib (GlaxoSmithKline's Tykerb/Tyverb)]. Lacking all three receptors, it cannot be treated with standard receptor-targeting breast cancer therapeutics (e.g., tamoxifen, aromatase inhibitors, trastuzumab) but must be treated with cytotoxic chemotherapy. TN breast cancer is generally more aggressive than other types of breast cancer, and even treatment with aggressive chemotherapy regimens typically results in early relapse and metastasis.

TN breast cancers constitute approximately 25 percent of breast cancers. They are diagnosed most often in younger women, those who have recently given birth, women with BRCA1 mutations, and African-American and Hispanic women.

There is a Triple Negative Breast Cancer Foundation, which was founded in 2006 in honor of a mother in her mid-thirties who died of the disease.

Ovarian cancer, the ninth most common cancer in women, caused nearly 14,000 deaths in the U.S. in 2010. In its earliest stages, its symptoms are usually very subtle and mimic other, less serious diseases. As a result, it is usually detected at later stages in which treatment is more difficult and gives poorer outcomes. The 2001 five-year survival rate was 47%, up from 38% in the mid-1970s. This compared to an overall survival rate for cancer of 68% in 2001, up from 50% in the mid-1970s.

Treatment usually involves surgery and chemotherapy, and sometimes radiotherapy. Surgery (preferably by a gynecological oncologist) may be sufficient for earlier-stage tumors that are well-differentiated and confined to the ovary. In this early-stage disease (which represents about 19% of women presenting with ovarian cancer), the five-year survival rate is 92.7%. However, about 75% of women presenting with ovarian cancer already have stage III or stage IV disease, in which the cancer has spread beyond the ovaries. Then the prognosis is much poorer, and the vast majority of patients will have a recurrence.

The triple-negative breast cancer study

The TN breast cancer study was carried out by researchers at the Vanderbilt-Ingram Cancer Center (Vanderbilt University, Nashville, TN), and published in the 1 July 2011 issue of the Journal of Clinical Investigation. In this study, the researchers analyzed gene expression profiles from 21 publicly available breast cancer data sets, and identified  587 cases of TN breast cancer (by non-expression of mRNAs that encode ER, PR, and HER2). Using cluster analysis, they identified six TN breast cancer subtypes:

  • two basal-like subtypes (BL1 and BL2),
  • an immunomodulatory (IM) subtype (i.e., expressing genes involved in immune cell processes)
  • a mesenchymal (M) subtype
  • a mesenchymal stem–like (MSL) subtype
  • a luminal androgen receptor (LAR) subtype.

Using gene expression analysis, the researchers identified TN breast cancer model cell lines that were representative of each of these subtypes. On the basis of their analysis, the researchers predicted “driver” signaling pathways, and targeted them pharmacologically as a proof-of-principle that analysis of gene expression signatures of cancer subtypes can inform selection of therapies.

BL1 and BL2 subtypes had higher expression of genes involved in the cell cycle and response to DNA damage, and model cell lines preferentially responded to cisplatin. M and MSL subtypes were enriched for expression of genes involved in the epithelial-mesenchymal transition (EMT), and growth factor-related pathways in model cell lines responded to the PI3K/mTOR inhibitor BEZ235 (Novartis, now in Phase 1 and 2 for solid tumors) and to the ABL/SRC inhibitor dasatinib [Bristol-Myers Squibb's Sprycel, currently approved for treatment of chronic myelogenous leukemia (CML) and Philadelphia chromosome-positive acute lymphoblastic leukemia (ALL), and under investigation for treatment of solid tumors). The LAR subtype was characterized by androgen receptor (AR) signaling, and included patients with decreased progression-free survival. LAR model cell lines were uniquely sensitive to the AR antagonist bicalutamide (AstraZeneca's Casodex/Cosudex, currently approved for the treatment of prostate cancer and hirsutism, and under investigation for treatment of androgen receptor-positive, ER negative, PR negative breast cancer).

The researchers plan to use the TN breast cancer subtype-specific model cell lines for further molecular characterization, to identify new components of the “driver” signaling pathways for each subtype. These pathways can be targeted in further drug discovery efforts. The subtype-specific cell lines can also be used in preclinical studies with targeted agents, and in identification of subtype-specific biomarkers that can potentially be used in stratifying TN breast cancer patients so that they might be treated with the best agents for their disease.

The ovarian cancer study

The ovarian cancer study was carried out by the Cancer Genome Atlas Research Network [a consortium of academic researchers jointly funded and managed by the National Cancer Institute (NCI) and the National Human Genome Research Institute (NHGRI)], and published in the 30 June 2011 issue of Nature. In this study, the researchers analyzed mRNA expression, microRNA expression, promoter methylation and DNA copy number in 489 high-grade serous ovarian adenocarcinomas, as well as the DNA sequences of exons from coding genes in 316 of these tumors. Serous adenocarcinoma is the most prevalent form of ovarian cancer, accounting for about 85 percent of all ovarian cancer deaths.

The researchers found that nearly all of the high-grade serous ovarian cancers (HGS-OvCa) studied had mutations in the TP53 gene, which encodes the p53 tumor suppressor protein. On the basis of their gene expression (mRNA) signatures, the researchers divided the population of HGS-OvCa into four subtypes:

  • an immunoreactive subtype (i.e., expressing genes involved in immune cell processes)
  • a differentiated subtype (high expression of markers of differentiated female reproductive tract epithelia)
  • a proliferative subtype (high expression of markers of cell proliferation)
  • a mesenchymal subtype (high expression of HOX genes and of markers of mesenchymal-derived cells)

The researchers also determined subtypes on the basis of microRNA expression and promoter methylation. microRNA subtype 1 overlapped the mRNA proliferative subtype and miRNA subtype 2 overlapped the mRNA mesenchymal subtype. Patients with miRNA subtype 1 tumors survived significantly longer that those with tumors of other microRNA subtypes.

Although the researchers found no significant difference in survival between the four transcriptional subtypes, they did identify a 193-gene expression signature that was predictive of overall survival. 108 genes were correlated with poor survival and 85 were correlated with good survival.

The researchers identified cancer-associated pathways in the HGS-OvCA population; this is equivalent to the prediction of “driver” signaling pathways in the TN breast cancer study. They found that 20% of the HGS-OvCA samples had germline or somatic mutations in BRCA1 or BRCA2, and that 11% lost BRCA1 expression through DNA hypermethylation. As we discussed in an earlier article on this blog, BRCA1- or BRCA2-negative tumor cells cannot repair their DNA via homologous recombination. They are dependent on an alternative pathway of DNA repair, which involves the enzyme poly(ADP) ribose polymerase (PARP). These tumors are thus sensitive to a class of drugs known as PARP inhibitors, such as KuDOS/AstraZenaca’s olaparib. There are now six PARP inhibitors, including olaparib, in clinical development.

The researchers found genetic alterations in several other genes involved in homologous recombination. Altogether, defects in homologous recombination may be present in approximately half of HGS-OvCa cases, and these tumors may be sensitive to PARP inhibitors. This provides a rationale for clinical trials of PARP inhibitors in women with ovarian cancers with defects in homologous recombination-related genes.

Olaparib and other PARP inhibitors are in clinical trials in women with advanced with BRCA-1 or -2 mutations and with other defects in homologous recombination. As discussed in the 2011 ASCO meeting, early Phase 2 results indicate that olaparib gives dramatic improvements in progression-free survival in these women. (See this article.) In these studies, in addition to tumors with genetic defects in homologous recombination, olaparib or another PARP inhibitor, Abbott’s ABT-888, appears to give improved progression-free survival in women who have previously been treated with chemotherapy drugs that result in DNA damage. This suggests that oncologists may be able to use a “one-two punch”, consisting of a DNA-damaging drug [such as the alkylating agent temozolomide [Merck's Temodar]) followed by a PARP inhibitor, to treat advanced ovarian cancer.

In addition to BRCA-1 and BRCA-2 mutations and other genetic alterations that result in defects in homologous recombination, the HGS-OvCa population exhibited genetic changes that would result in deregulation of several other cancer related pathways. These pathways included the RB1 (67% of cases), RAS/PI3K (45% of cases), and NOTCH (22% of cases) pathways, as well as the FOXM1 transcription factor network (87% of cases). All of these pathways represent opportunities for target identification and drug discovery. FOXM1 (Forkhead box protein M1) was named the Molecule of the Year for 2010 by the International Society for Molecular and Cell Biology and Biotechnology Protocols and Research (ISMCBBPR) because of “its growing potential as a target for cancer therapies.” FOXM1 overexpression results in destabilization of the cell cycle, which can lead to a malignant phenotype.

The researchers also identified 22 genes that were frequently amplified or overexpressed in HGS-OvCA tumors (other than genes that are involved in homologous recombination). Inhibitors (including approved and experimental compounds) already exist for the products of these genes, and researchers might assess these compounds in HGS-OvCa cases in which target genes are amplified.

Can Verastem develop new therapeutics for triple negative breast cancer?

The private biotechnology company Verastem (Cambridge, MA) focuses on discovery and development of drugs to target cancer stem cells. The company was founded in 2010, and is based on a strategy for screening for compounds that specifically target cancer stem cells. This strategy, published in the journal Cell in 2009, was developed by Drs. Robert Weinberg (MIT Whtehead Institute), Eric Lander (Broad Institute of MIT and Harvard University), and Piyush Gupta (MIT and Broad Institute) and their colleagues. Drs. Weinberg, Lander, and Gupta are on the Scientific Advisory Board of Verastem.

On July 14, 2011, Verstem announced that it had raised $32 million in a Series B financing. Verastem had previously raised $16 million from a group led by former Christoph Westphal’s Longwood Founders Fund. Dr. Westphal (formerly of Sirtris) is now Chairman of Verastem.

Cancer stem cells are best known in acute myeloid leukemia (AML), but their existence in other cancers (especially solid tumors) is controversial. The cancer stem cell hypothesis asserts that a small subpopulations of cells in a leukemia or solid tumor have characteristics that resemble normal adult stem cells, such as self renewal, the ability to give rise to all the cell types found in the leukemia or cancer, and stem cell markers. The hypothesis further asserts that most cancer treatments fail to knock out cancer stem cells, which can repopulate a tumor cell population, resulting in treatment relapses. Cancer stem cell researchers therefore propose developing cancer stem-cell specific therapeutics that can be used to eliminate these cells, which can block these relapses.

Whether cancer stem cells are involved in the pathobiology of solid tumors or not, the biology of the putative cancer stem cell phenotype can be important in certain subtypes of cancer. Cancer stem cells are characterized by the epithelial-mesenchymal transition (EMT), and in the Cell paper the researchers screened for compounds that specifically targeted breast cancer cells that had been experimentally induced into an EMT, and which as a result exhibited an increased resistance to standard chemotherapy drugs.   They identified the compound salinomycin as a drug that specifically targeted these cells, as well as putative cancer stem cells from patients.

As discussed earlier in this article, TN breast cancer includes two subtypes that have gene expression signatures related to the EMT: the mesenchymal (M) subtype and the mesenchymal stem–like (MSL) subtype. One or both of these subtypes might be sensitive to compounds that specifically target putative breast cancer stem cells. This may be true whether the cancer stem cell hypothesis applies to TN breast cancer or not. Verastem recognizes this, and is thus focusing on TN breast cancer as its first therapeutic target. The Vanderbilt TN breast cancer study suggests that trials of any “cancer stem cell-specific” therapeutics for TN breast cancer should be guided by subtype-specific biomarkers.

Hope for treatment of TN breast cancer and advanced ovarian cancer

Researchers and oncologists have made great strides in increasing the percentage of breast cancers that are treatable or even curable in recent years. For example, prior to the FDA approval of trastuzumab in 1998, HER2 positive breast cancer carried a grim prognosis. But the advent of trastuzumab (and later, lapatinib) has had a major impact on treatment of this once uniformly deadly type of breast cancer.

We hope that the new, personalized medicine-based approach to TN breast cancer and advanced serous ovarian adenocarcinoma will also result in successful new therapeutic strategies for these deadly women’s cancers.


As the producers of this blog, and as consultants to the biotechnology and pharmaceutical industry, Haberman Associates would like to hear from you. If you are in a biotech or pharmaceutical company, and would like a 15-20-minute, no-obligation telephone discussion of issues raised by this or other blog articles, or of other issues that are important to  your company, please click here. We also welcome your comments on this or any other article on this blog.

Leave a Reply