Biomarker Testing in Metastatic Breast Cancer Management: ‘Essential’

Victoria Stern, MA


October 05, 2021

Identifying biomarkers in metastatic breast cancer (MBC) has become an integral part of choosing treatments and understanding disease progression. The American Society of Clinical Oncology Clinical Practice Guideline, published in 2015, recommends an initial biopsy to confirm estrogen receptor (ER), progesterone receptor (PR), or human epidermal growth factor receptor 2 (HER2) status as well as repeat biopsies to watch for receptor status changes over time.

Kelly McCann, MD, PhD

"Decisions concerning the initiation of systemic therapy or selection of systemic therapy for metastatic breast cancer should be guided by ER, PR, and HER2 status in conjunction with clinical evaluation, judgment, and the patient's goals for care," according to the guideline authors.

Along with tumor subtypes, experts continue to identify a host of other actionable targets that can shape treatment decisions. Medscape reached out to Kelly McCann, MD, PhD, a hematologist and oncologist in the Department of Medicine at the David Geffen School of Medicine, University of California, Los Angeles, to explore the role biomarker testing plays in managing MBC.

Medscape: How important is biomarker testing in guiding MBC treatments? Is there a standard or recommended process?

Dr McCann: Biomarker testing is essential to breast cancer treatment and the development of targeted therapies. Oncologists typically identify a tumor's canonical biomarkers — ER, PR, and HER2 — using immunohistochemistry or fluorescence in situ hybridization (FISH) testing and then try to match the tumor biology to drugs that target that subtype.

For tumors that lack canonical biomarkers — for example, triple-negative breast cancer (TNBC) — I send the tumor tissue for next-generation sequencing at the time of metastatic diagnosis to identify a wider range of potential targets or oncogenic drivers, such as somatic or germline mutations in homologous recombination repair genes ( BRCA1, BRCA2, and PALB2 ) or mutations in the PI3K/AKT/mTOR pathway.

In our attempts to define tumor biology and design a treatment strategy, two additional issues quickly arise. First, tumors are heterogeneous from the start. Second, tumors evolve.

Let's start with how we define or subtype a tumor. Would you walk us through this process?

Defining a breast tumor can be tricky because these cancers often don't fit neatly into predefined categories. Let's take the estrogen receptor. In clinical trials, we need to define the cutoff for what constitutes ER-positive MBC or TNBC. Some trials define ER-positive as 1% or greater, others define it as 10% or greater.

But is a PR- and HER2-negative tumor with 1% or even 5% ER expression really ER-positive in the biological or prognostic sense? Probably not. A tumor with less than 10% ER expression, for instance, will actually behave like a triple-negative tumor. Instead of choosing a regimen targeting the ER-positive cells, I'll lean more toward cytotoxic chemotherapy, the standard treatment for TNBC.

Tumors may have multiple drivers as well. What are some aberrations in addition to the main subtypes?

Tumors also often harbor more than one targetable driver. For instance, PIK3CA gene mutations are present in about 40% of hormone receptor–positive, HER2-negative tumors. Activating mutations in ESR1 develop in anywhere from 10% to 50% of MBCs as a resistance mechanism to estrogen deprivation therapy, conferring estrogen independence to the cells. Activating mutations in ERBB2, which essentially turns HER2 into an active receptor, are found in 2%-4% of breast cancers, including ER-positive, HER2-mutant breast cancers, and are enriched in lobular breast cancers, which are typically ER positive, HER2 negative.

What about tumor evolution, given the growing body of evidence that biomarker status in MBC can change over time?

Patients with MBC often have several active areas of cancer, and these areas will evolve differently. During each line of treatment, some metastases will develop resistance and others won't. For instance, if my patient's liver metastases start to grow, I will change therapy immediately. If, however, a single bone metastasis begins to grow and the liver metastases have responded well, I might consider local therapy — such as radiation — to target that bone metastasis, though this particular approach hasn't been formally studied.

Ultimately, we can expect tumors to change over time as they become more biologically aggressive or resistant to current therapy. The most common biomarker change is probably loss of ER or PR expression, but the frequency of ER, PR, or HER2 biomarker changes is still not well understood.

Resistance mutations can also happen. When, for instance, activating mutations in ESR1 occur, the estrogen receptor becomes independent of estrogen and tumors then develop resistance to endocrine therapies. We see a similar problem arise in metastatic prostate cancer. With chronic testosterone deprivation, eventually the androgen receptor evolves to become independent of testosterone in a stage known as castrate-resistant prostate cancer.

Which biomarkers or combinations of biomarkers can be paired with an approved treatment?

We have a range of treatments targeting ER-positive and HER2-positive MBC in particular. For tumors harboring additional targetable mutations, preliminary data suggest that HER2-targeted tyrosine kinase inhibitors (TKIs), such as tucatinib and neratinib, are effective against activating mutations in ERBB2.

The PI3K inhibitor alpelisib in combination with fulvestrant has been approved for patients with ER-positive, HER2-negative MBC and mutations in PIK3CA. The mTOR inhibitor everolimus plus exemestane is an option for patients with ER-positive, HER2-negative. And for those with activating mutations in ESR1, I switch patients to a selective estrogen receptor degrader, such as fulvestrant.

PARP inhibitors, including olaparib or talazoparib, target metastatic HR-positive disease or TNBC with deleterious germline BRCA1 or BRCA2 mutations. Sacituzumab govitecan has been approved for treating metastatic TNBC and targets the cell surface protein TROP2, expressed in almost 90% of TNBC tumors.

What targets, on the other hand, are less informative for treatment choice?

When we order next-generation sequencing, we also will get a list of possible targets for which there are currently no therapeutic options, but there may be in the future. I find this knowledge is helpful. For example, an activating mutation in KRAS tells me that the cancer has a very strong oncogenic driver that I won't be able to target. I know that activating KRAS mutations in lung cancer and colon cancer portend a poorer prognosis, which helps me to prepare the patient and family.

Atezolizumab in combination with paclitaxel has been FDA-approved for PD-L1 TNBC in the first-line setting, though data show that immune checkpoint inhibitors may be effective even without PD-L1 expression. Although cell surface protein TROP2 has emerged as a target in recent years, its expression is so common in TNBC that confirmatory testing for TROP2 expression is not required to prescribe sacituzumab govitecan.

What factors do you weigh when selecting among the large number of tests available for tumor testing?

We have many biomarker tests available, but the National Comprehensive Cancer Network does not have guidelines for tumor genetics testing in breast cancer. That means insurance does not have to cover the cost, and many companies don't. Ultimately, though, drug companies and some testing companies have an incentive to cover the cost themselves because a companion diagnostic might be linked to their drug — therascreen PIK3CA RGQ PCR kit for alpelisib, for instance.

I tend not to use a companion diagnostic test because I want more information with a wider panel. The tumor tests I often use are FoundationOne CDx, Caris Molecular Intelligence, and Tempus. I use Tempus because their financial aid is very generous and almost all of my patients qualify to be tested for less than $100. For germline genetic testing, Invitae, Myriad, and Color are also options. Invitae and Color are about $250 out of pocket without insurance. Many academic centers have their own gene panels as well. 

How far have we come in identifying biomarkers in MBC?

Targeted treatment for breast cancer has advanced significantly since doing my PhD research in cancer biology about 15 years ago. Of course, targeted therapies for ER-positive and HER2-amplified cancers were available at that point, but many more have been developed. The most significant advance has been the development of efficient and affordable genome sequencing, which has led to these large panels and identification of therapeutic targets. We’ve also expanded our knowledge of genetic predispositions for breast cancer beyond BRCA1 and BRCA2, which not only allows us to preemptively advise patients and their families about cancer risks and recommendations for cancer screening, but also to select a therapy to target a cancer’s DNA repair deficits.

I feel that we are in an exciting discovery phase in oncology. We currently rely on biomarkers to manage MBC and will continue to refine our strategies and develop more effective drug therapies as we identify more oncogenic drivers, tumor-specific proteins, and cancer cell vulnerabilities.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.