The role of genetic testing to guide therapy for patients with metastatic breast cancer is expanding. There are now several mutations with approved indications for treatment, and more are being studied.
But for the clinician, knowing which tests to order for a particular patient can be complicated. Furthermore, with research in this area evolving and new information continually published, interpreting assay results and deciding on the best way to apply them can also be difficult.
“The two main purposes of genetic profiling are to identify targetable mutations with approved therapies and to identify mutations with matched clinical trial options,” Zahi Mitri, MD, of Oregon Health & Science University in Portland, told MedPage Today. “From the standpoint of identifying targetable mutations with approved therapies, it is important for oncologists to be familiar with the rapidly expanding indications.”
“The role is continuing to change over time,” Stephen Schleicher, MD, of Tennessee Oncology in Nashville, told MedPage Today. “The newest updates are with PD-L1 [programmed death-ligand 1] testing to determine the role of immunotherapy in triple-negative metastatic breast cancer. That has become more complex, because you have different testing criteria and test assays for different indications for immunotherapy.”
For example, he said, if a specific PD-L1 assay, SP142, is greater than 1%, the FDA-approved first-line treatment is atezolizumab (Tecentriq) plus nab-paclitaxel (Abraxane). More recently, the FDA approved a new indication for pembrolizumab (Keytruda) plus chemotherapy if the result of a different type of PD-L1 test, combined positive score (CPS), is greater than 10%.
“So it’s getting complicated,” he said. “Both of these have implications for first-line treatments and they are not overlapping, so ideally patients with triple-negative breast cancer that is metastatic should have both of these. That again points to the challenges of oncologists keeping up with that, plus pathologists knowing that both of those PD-L1 tests with different assays should be done. So that is one example of how it’s changing over time.”
Both Schleicher and Mitri also noted the two other targetable mutations for which testing should be considered: somatic phosphoinositide-3 kinase (PI3K) testing for use of a PI3K inhibitor in hormone-positive metastatic breast cancer, and germline BRCA testing for use of PARP inhibitors in HER2-negative metastatic breast cancer.
Oncologists should also be aware of recent “tumor-agnostic” approvals, Mitri noted: “With tumor-agnostic approval for agents toward MSI, TMB-H [tumor mutational burden-high], and NTRK translocation, genetic panels should include those as well. Beyond standard of care, when a clinical trial is being considered, broader testing can be considered.”
Schleicher added: “So we now have a few changing targetable or immunotherapy indications at the moment in the metastatic setting that all have little nuances, so it’s important to figure out ways to help keep every oncologist and pathologist up to date with that.”
Mitri noted that many of the commercial sequencing services provide annotated reports and list matched potential clinical trials. Even so, interpreting the results and using them to guide decisions may not be a simple task.
For example, an assay might find a new mutation of a known oncogene whose significance is not known, or perhaps a new mutation altogether. Sometimes multiple actionable mutations will be found with no information to help prioritize them.
Fortunately, there are resources to help with interpretation. One publicly available online resource Mitri recommended is OncoKB, a precision oncology knowledge base that includes evidence-based information on individual somatic mutations and structural alterations; OncoKB also integrates clinical, biologic, and therapeutic information with national guidelines and recommendations.
As noted in a review Mitri, Schleicher, and colleagues wrote in the 2020 American Society of Clinical Oncology Educational Book, OncoKB is organized by gene, alteration, tumor type, and clinical implication, and uses a system of different levels of evidence to describe the clinical utility of individual mutations.
For example, level 1 is the strongest evidence, which includes specific gene mutations that are recognized by the FDA with matched FDA-approved therapy. In addition to actionable mutations, OncoKB also classifies mutations that have been shown to confer resistance to specific targeted therapies.
“The benefit of the online knowledge base is that it is being updated regularly to include new clinically significant mutations along with matched therapies,” Mitri said.
He also recommended the European Society for Medical Oncology Scale for Clinical Actionability of Molecular Targets (ESCAT) classification system, which is designed to aid oncologists in prioritizing potential targets for clinical use when interpreting a gene sequencing panel. The level of evidence is designated into tiers on the basis of existing regulatory approvals as well as clinical and preclinical data.
Schleicher said that while resources such as OncoKB and ESCAT can be useful, the oncology community should go even further in providing clinicians with the most current and relevant information. “The problem with a database is you have to know to look into it, which is going to be a select group of providers who are already tuned in to that,” he said. “How do you get system-wide mechanisms to enable providers to make the correct choice at the point of care?”
One option is treatment or diagnostic pathways built into healthcare systems. “So one example is a pathway we are working on at Tennessee Oncology/One Oncology, so that as we are entering orders for, say, a triple-negative breast cancer patient, it has pop-ups that guide us to test for these things,” he explained.
Another possibility would be standardized order sets by disease types across a bigger healthcare system or practice, Schleicher continued. “So, for instance, I know I have a breast cancer panel for anybody with metastatic breast cancer who comes in. A group of thought leaders can then dictate what every provider should be ordering to make sure the right tests get done. That is the challenge: how do you push this type of information to the point of care so that generalist oncologists who aren’t thinking to log on to OncoKB are getting the prompts to be doing this in real time as the evidence comes out?”
Words of Caution
Mitri advised caution when interpreting the results of genomic panels. “Given the limited targeted therapies in breast cancer, it is often tempting to apply therapeutics in an off-label fashion, especially when standard of care is exhausted. While anecdotal evidences exist, it is important to keep in mind that a driver mutation in one particular histology may not be a driver in a different histology,” he said.
For example, although targeting the epidermal growth factor receptor has been successful in treating non-small cell lung cancer, multiple studies have failed to show significant benefit in breast cancer. Furthermore, it is important to keep in mind that not all mutations are equal: More than 100 PIK3CA mutations have been identified, and not all are “gain-of-function” or oncogenic, he explained.
Finally, oncologists should also be aware that clinical cut-offs and definitions for various genetic alterations are often either not uniform or evolve over time, Mitri said. “For example, various cutoffs have been used to define TMB-H, from the most recent 10 or more mutations/Mb to 243 or more mutations/Mb. It is important to be familiar with the definition used for the approved therapy.”
Mitri disclosed financial relationships with Lilly Pfizer/EMD, Serono, Puma Biotechnology, Samsung, and Bioepis, and institutional funding with AstraZeneca, Daiichi Sankyo, Novartis, Radius Health, and Seattle Genetics.
Schleicher disclosed financial relationships with Bayer, Magnolia Innovation, and Rigel.