Streamlined Genomic Testing Platforms Are Taking on A Bigger Role in Cancer Care

Lee S. Schwartzberg, MD, FACP

Genetic testing has afforded oncologists with the opportunity to identify patients who may have a higher risk of developing cancer at some point in their life, especially for patients with a predisposition for certain cancers and patients who may have family members with cancer. Additionally, genetic testing to identify actionable mutations has allowed patients to receive targeted therapy and have their long-term treatment plan mapped out at diagnosis, according to Lee S. Schwartzberg, MD, FACP.

However, this testing has created a vast amount of information to digest for each patient, and streamlining platforms to pull more relevant information together will be important as the genetic testing field continues to grow, Schwartzberg said.

“Linking the germline genetics with outcomes, like the ancestry data, is becoming increasingly interesting,” Schwartzberg said. “We want to know if patients who have a specific ancestry might have a predisposition for multiple genes interacting to get a cancer, or, more importantly, how the outcome [of] that cancer might differ.”

In an interview with OncLive^®, Schwartzberg, the chief of Medical Oncology and Hematology at the Renown Institute for Cancer, and a professor of clinical medicine at the University of Nevada, discussed barriers to genetic testing in oncology, the role of a molecular tumor board, and the expanding number of platforms available to perform testing.

OncLive^®: What have been some of the barriers to genetic testing in oncology?

Schwartzberg: When discussing genetic testing in oncology, we’re talking about 2 broad themes. The first is germline testing, which has been available for about 25 years and should be considered for patients who typically have family history [of cancer] or meet the criteria for germline testing. These criteria have evolved dramatically over time to multi-panel testing. We usually test multiple genes that we now know increase the susceptibility to cancer. We test individuals with a diagnosis of cancer and potentially their family members if an alteration, a pathogenic variant, or a probable pathogenic variant [is found] in the germline. We also test high-risk individuals when they don’t have cancer.

The barriers include finding the right patients to test. It is easy for oncologists, if you have a patient with a cancer diagnosis, to determine, based on their family history or the type of cancer they have, or both, whether germline testing should be considered. Those guidelines have evolved dramatically over the past few years and include patients that have strong family histories. We have therapeutic drugs for patients with certain germline alterations, namely PARP inhibitors, particularly in patients with breast cancer. That has led to a broader discussion of testing where we believe that the majority of patients with breast cancer should be tested and not miss an opportunity to receive a PARP inhibitor in the adjuvant setting if they have higher-stage disease.

One of the barriers in the space can be insurance in some cases, [when it needs to be determined] whether patients fall into NCCN [testing] guidelines. Making sure that the high-risk patients are identified by their primary care physicians for testing [is necessary]. Many times, oncologists end up doing the testing or arrange with a genetic counselor to do testing. But we can only do the testing if we’re aware of those patients.

There is still a gap in access to patients who may benefit from testing [even though] there may be a strategy of drug therapy, imaging, or more intensive surveillance, which can be an issue. The second issue with germline testing is interpretation of the results. In other words, not finding the patients is one issue, but finding patients to test and misinterpreting the results is another issue. We see this frequently, particularly with people with too much to do, such as primary care doctors. It’s too much to stay [up-to-date] on everything with the vast breath of what they see.

Particularly when patients get a variant of unknown significance in a cancer susceptibility gene, occasionally, those patients will be counseled to undergo prophylactic surgery or increased surveillance. That is not the recommendation for a variant of unknown significance. Understanding the difference between a pathogenic or likely pathogenic variant vs a variant of unknown significance [is important], and we need to get the word out. Primary care doctors, oncologists, and surgeons [need to be educated].

What have become some of the best practices to ensure that potential pitfalls are avoided?

The way to avoid the pitfall is to have a high-risk clinic for patients who might be susceptible to breast and ovarian cancer, for example. Increasingly, we recognize that there are other patients [who may benefit from testing]. The simple thing is to acknowledge that a comprehensive family history is taken. For people who don’t have cancer, that’s where you find the gold. If you have families that have multiple cancers that fall into a pattern of potentially suggesting a hereditary predisposition gene, that’s where you make that recognition and send them to the appropriate person if you’re not equipped to do the testing. That can be for primary care physicians, but it can be for oncologists, as well. Taking that family history is really important.

Do platforms need to be streamlined to consider risk assessment, predictive responses to therapy, and genomic profiling itself?

[It is important to have] electronic platforms that help on the genetic side and hereditary side to complete the family history and clinical decision support that will highlight and identify a patient who might be a good candidate for multi-gene panel testing in the germline. We are also seeing that the awareness of clinical decision support and the integration of results into the electronic record is important for patients who have cancer where we do a comprehensive genomic profile.

For about 10 years, we’ve had the ability to do multi-gene panels that have grown in size, and they now typically run anywhere from several hundred genes to whole exome and whole transcriptome sequencing. This includes both DNA and RNA at one end, and, at the very least now, we have the ability to do genomic profiling of multiple genes, including the actionable genes. That delivers a tremendous amount of information.

A barrier there includes reimbursement. There are large payers who are not yet convinced of the broad-scale benefit of doing comprehensive genomic profiling in advanced cancers across the board. [For example], in many of the guidelines, including non–small cell lung cancer [NSCLC], comprehensive genomic profiling is recommended, as opposed to doing individual tests for the actionable genes. Given the fact that [NSCLC] now has upward of 10 actionable alterations, even in the first-line setting, it’s critical to know that information up front. Payers are still requiring that the sequential single-gene approach be taken. I believe that’s wrong.

Going beyond to other diseases like breast cancer, where you might not make a treatment decision in the first line based on a comprehensive genomic profile, I strongly believe in having that information at hand when starting to plot out the different courses of therapy that a patient may have during their lifetime with advanced breast cancer. It’s good to have that information before you have to make the decision in a situation when a patient progresses. There are also many rare cancers that have specific alterations, and they should be tested, which can have a huge effect on outcome.

I recommend doing comprehensive genomic profiling on patients at their diagnosis of advanced disease. Today, we can even follow them with liquid biopsies on a regular basis, although that’s still in evolution in terms of the most valuable and impactful way to do that [in terms of improving] clinical outcome.

Another barrier is awareness [of knowing] if we should do comprehensive genomic profiling on all patients. Not all oncologists are doing that yet, even in NSCLC. Although it is ironclad that we should do it in all patients, only about 70% of patients with advanced NSCLC, up until the past year, receive comprehensive genomic profiling at diagnosis. That number should be closer to 90%. Although we’ve made steady progress over time, we’re not yet at the optimal level because of some of the other barriers.

Another barrier for genomic profiling is interpreting the results. We get a wealth of information, typically a 30- or 40-page report, when we do comprehensive genomic profiling from a blood or tissue sample. The amount of information is huge. The problem is, no one has the time to sit and read a 30-page report, word for word, so it does get summarized. However, many of the nuances can be lost in the summary. One way to get around that barrier is to have a molecular tumor board with a group of people that have familiarity with comprehensive genomic profiling, including genetic counselors, pathologists, molecular pathologists, clinical oncologists, and imagers, to go through the report.

If you get genes that look like you can do something with in terms of a therapy, it is important to present those in a real-time fashion and get the input of a tumor board, just like we would with a standard case without the genomics or the molecular findings. That can be done in a disease-specific tumor board, although it gets complicated there. Utilizing the molecular tumor board with the most impactful cases presented on a regular basis can be very useful for changing patients to the right therapy, for agreeing with the therapy, and, importantly, for [enrollment in] clinical trials. For directing patients to clinical trials, a molecular tumor board is fantastic. Whether it’s right at the time the patient gets testing or as a clinical decision support tool, every time a patient progresses and changes therapy, oncologists are reminded that this patient has a molecular alteration, and they may be a candidate for these current trials that are available. We are not quite yet at the sophistication of clinical decision support to do that, but we are getting closer. The idea that you can surface an alteration that would prompt the clinician to look for a clinical trial at the time of progression is coming along nicely. It’s a great use of technology to avoid that barrier to best care.

Myriad Genetics presented several testing platforms at the 2022 ASCO Annual Meeting. How could tools like these help expand the field of genomic testing in cancer care?

Myriad Genetics is going in multiple directions to improve care through the combination of genetics, genomics, and developing new tools. The homologous recombination deficiency [HRD] score is something that’s had a lot of attention. In this case, we’re looking at a variety of different genomic alterations, individual genes, and broader genome-wide [factors] such as loss of heterozygosity. In pulling those all together into an HRD score, we use that to make clinical decisions. This is most notably [applied] in ovarian cancer, but it’s starting to extend out into other diseases. Having that information in hand is really going to be critical in the future for making the right decision for therapy for these patients.

Another group of genomic testing includes genomic profiling, genomic expression, or genomic classifiers, which look at the pattern of expression of certain genes, then pulling them together into a model that predicts either prognosis or response to types of therapies. EndoPredict is a good example of that, with good data showing prognosis of patients, low to high, based on EndoPredict score. You can have a patient with a clinically high-risk tumor that has a genomically low-risk tumor, and you would treat that patient differently. That’s what the study gets at: How often do you use that kind of data to make a decision, or how does the test affect your decision making? That is important when you have genomic classifier tests that will tell you information that’s both prognostic and predictive.

[Research is also being done with genetics and ancestry.] For example, Black women with triple-negative breast cancer have a worse outcome. Is any of that due to their ancestry in the sense of inheriting multiple genes? Not genes that are single actors, like BRCA, which has, as an individual gene, influenced the risk and outcomes of breast cancer, but how groups of genes that are inherited over generations might also affect that. That is an area of active discovery and research.

Is there anything coming down the pipeline that you see potentially having an impact on practice patterns?

We are at the dawn of the age of how to best use liquid biopsies. One of the benefits of a liquid biopsy is that it’s simple, minimally invasive, and it can be repeated. As opposed to using tissue, liquid biopsy is a repetitive source of information about how a cancer is acting. The use of liquid biopsy is extremely wide ranging. On one end, we’re at the dawn of using liquid biopsy to do multi-cancer early detection, and the first test just rolled out where a tube of blood may be able to identify early-stage patients with a variety of different cancers. These tests are now out commercially, and much more research will be done to improve the sensitivity and specificity for the accuracy of these tests.

For the first time, we can think about screening patients for cancer beyond their traditional screening technologies. We can do it broadly, although the question remains if we can afford it as a society, and if insurance will pay to screen broad populations. The way it’s going to go, in my opinion, is the higher-risk populations will get access to these tests. There will be more impact in terms of the number of positives that are found, as opposed to the lower-risk groups.

This is exciting. Data that were presented at the 2022 ASCO Annual Meeting about minimal residual disease from liquid biopsy examined the sensitivity of liquid biopsies to pick up DNA or methylation patterns, [similar to] the multi-cancer early detection test. In patients who would be at risk after their initial tumor is removed, who are those patients that are destined to relapse? Can we intervene and do something about it early before they come in with symptoms or abnormal imaging? That’s a fascinating area, and it’s going to be one that’s going to yield a lot of information over the next few years.

We can use liquid biopsies to monitor patients on therapy, to find early relapse, and [to detect] defined patterns of mutations that change over time. It gives us insight into the reasons for resistance. Sometimes, like in NSCLC, we can use liquid biopsies to see what the cause of resistance is and get patients on clinical trials for targeted therapies for new generations of treatments. That is a whole area that is exploding right now.