The Nuanced Challenge of Identifying Actionable Mutations With Appropriate Treatments

Note: A previous version of this article appeared with our meeting coverage.

A panel of experts discussed the explosion of information and advances in cancer genomics and the accompanying challenges of understanding the data and taking action.

There has been a rapid accumulation of knowledge in genomics and precision medicine in oncology, which not only has led to advancements, but also to a volume of information that can be difficult to master and apply equally to all patients, said Davey B. Daniel, MD, chief medical officer of OneOncology, while introducing his panel on “Cancer Genomics: Advances and Access” during the 2022 meeting of Patient-Centered Oncology Care®, held November 9-10 in Nashville, Tennessee.

Daniel moderated a panel that covered advances in cancer genomics as well as ensuring equitable access. Panelists were:

• Ben Ho Park, MD, PhD, professor of medicine, Vanderbilt University Medical Center, and director, Vanderbilt-Ingram Cancer Center
• Pranil Chandra, DO, FACP, FASCP, chief medical officer of genomic and molecular pathology, PathGroup
• Gabriel A. Bein-Willner, MD, PhD, FCAP, program medical director, MolDX, and chief medical officer, Palmetto GBA.

A few years ago, there was a lot of excitement whenever there was a new FDA indication for a genomic-targeted therapy or a drug companion diagnostic, but Park explained that the realities of the real world have dampened this enthusiasm because the impact of these agents and tools can be variable.

“There’s an explosion right now of new ways to find genomic alterations, and then also the drugs that are there to target them,” he said. “But to really understand which ones are the ones that are game changers, and which ones are the ones that you have to act upon immediately, I think is very nuanced many times.”

Vanderbilt has been convening molecular tumor boards both locally and internationally to evaluate the results they’ve seen and allow for the collection of data. Most physicians are doing N-of-1 experiments, Park said, so shared knowledge reduces waste of resources and time.

The boards and their shared real-world experience also help provide insight into potential off-label uses and making a decision of whether a patient should forego a known standard of care for something more novel with less data.

For the FDA approvals, they have response rates that can range anywhere from 40% to 60%, and other biomarkers could affect the response rates, said Chandra. He was also a proponent of convening molecular tumor boards, which can provide differing perspectives.

He added that the report results of biomarker testing can be very complicated to understand. For instance, human epidermal growth factor receptor 2 (HER2) mutations have clinical significance in lung cancer, but HER2 expression or amplification does not.

“So, there needs to be…a multidisciplinary conversation to assess the biomarkers and really align that with the best care for the patient,” Chandra said.

Bien-Willner is not a stranger to hearing from busy oncologists that the test results are too complicated to understand and need to be made simpler.
“The problem is that if you reduce the complexity of the test, you remove value of the test,” he said. “It’s a complex process. It’s a complex problem.”
He felt that the path Chandra and Park have taken to have experts review the results and help the oncologist understand them and make the right decision is the correct path to take. However, it’s not always clear for the practicing oncologist who that expert should be.

The general process of ordering the test and acting on the results can be riddled with issues, however. Bien-Willner noted that the expectation is that the physician orders the test correctly, receives the results in the electronic medical record (EMR), reads and understands the results, and acts on them.

“There [are] many failures along every single one of those steps,” he said.
On the payer side, insurers want to make sure they’re paying for something that has clinical value. Some biomarkers are clearly actionable with an FDA-approved drug, but there are others that are less understood.

“There are certain circumstances where it’s clear, you need to act on it, and everyone would agree, and then other circumstances where you should only do so in the context of that specific patient and what their other options are and what other treatments they’ve already failed,” Bien-Willner said.

Over the years, payer mentality has moved away from doing one test on only what is needed for the patient, to a preference for running one test that does everything needed and maybe more than what is needed, explained Bien-Willner. That extra information could be useful down the line, and as long as it isn’t harming a patient, there’s no reason not to do a 500-gene panel, he said.

However, that’s a huge amount of data and information. According to Chandra, that data has to be in a discrete format, democratized, and available to the patient, the treating clinician, and the treating oncology practice through the EMR.

“The point is that when there are new FDA approvals, you need to be able to go into that your data set and be able to slice and dice that data and find, ‘OK, we have a KRAS G12C mutation, how many patients in a practice are potentially eligible for therapy?’” he explained.

Vanderbilt does something similar with a program that allows anyone to go back into the data when a new FDA approval comes out to see which patients are now eligible for treatment. For instance, when the FDA approved larotrectinib for solid tumors with NTRK gene fusions, they were able to go back into the data to see how many patients who were still alive were now potentially eligible, Park said.

However, not everything approved is at the same level of importance as that approval, he added. “I think that’s where we’re having problems right now: Just because the drug’s available doesn’t necessarily mean that’s the right drug at the right time for that patient.”