Molecular tumor testing transformative, yet 'under-optimized'

In 2010, results of the phase 2 BATTLE study demonstrated that choice of cancer therapy could be individualized based upon a patient’s tumor characteristics.

Edward S. Kim, MD, MBA, FACP, FASCO, physician-in-chief for City of Hope Orange County, HemOnc Today Editorial Board Member and lead author of BATTLE, and fellow researchers described that study of patients with advanced non-small cell lung cancer as “a substantial step toward realizing personalized lung cancer therapy.”

Eleven years later, molecular testing has expanded its reach in oncology, accompanied by an array of molecularly driven treatments across cancer types.

“It all started with discovery of a small number of genes and accelerated into patient care because of its clinical significance,” Kim told HemOnc Today. “Numerous studies have now shown that we can achieve better effectiveness and improved safety by taking this approach and matching treatment to a specific biomarker abnormality.”

However, in order to identify an appropriate targeted treatment for patients, they need to undergo genetic sequencing of their tumors. Real-world data indicate use of molecular testing and follow-through is lower than what experts had hoped.

“Annual statistics that show a general increase in molecular testing rates but lower-than-expected use of testing results is disheartening,” Kim said. “In our field, where we treat patients with cancer and have for decades expressed concern about not having effective enough therapies, we now have an opportunity to implement cutting-edge therapies that have improved survival and have better safety profiles. It is our duty to ensure that we are not only ordering comprehensive genomic testing but also waiting and utilizing the results to best treat our patients’ cancers.”

HemOnc Today spoke with oncologists about how often molecular testing leads to treatment changes, the degree of underuse and challenges that may be driving it, and ongoing initiatives that aim to improve uptake of testing and implementation of the results.

Importance of molecular testing

Molecular testing is critical for treatment selection and can help avoid chemotherapy for certain patients for whom targeted therapy or immunotherapy may provide more benefit.

Identifying genetic abnormalities and targets has become “prolific” in oncology research, with multiple cancer types benefiting from next-generation sequencing, according to Martin E. Gutierrez, MD, director of drug discovery and chief of thoracic oncology at Hackensack Meridian Health.

“In lung cancer, for example, multiple genetic abnormalities have been identified with actionable mutations to different agents,” Gutierrez told HemOnc Today. “Identification of those targets can make a significant difference in response rates and OS for patients.”

According to Gutierrez, multiple studies have shown that objective response rates from targeted therapy for NSCLC are “two to three times better” than from chemotherapy. In one such study in ALK-positive NSCLC, Shaw and colleagues reported a response rate of 65% with crizotinib (Xalkori, Pfizer), an ALK tyrosine kinase inhibitor, compared with 20% with chemotherapy.

“We know that patients with lung cancer with genetic driver mutations are less likely to respond to immunotherapy, and their response rates and survival with chemotherapy only would be suboptimal by today’s standards,” Gutierrez said. “Also, if we treat patients with the appropriate targeted therapy, it leads to significantly improved outcomes, with fewer hospitalizations and ED visits. That makes a big difference.”

Although the degree of utility depends on the cancer setting, it is “almost impossible” to select therapy for colorectal cancer, for instance, without molecular testing, Christopher Lieu, MD, deputy associate director for clinical research at University of Colorado Cancer Center, told HemOnc Today.

“There are situations in the refractory setting where there are no opportunities for standard chemotherapy because the patient has gone through all standard-of-care options,” he added. “In this setting, molecular testing can help lead to the selection of a clinical trial.”

Molecular testing is becoming more useful as the identification of targets and matching treatments has increased, according to Pauline Funchain, MD, director of the genomics and melanoma programs at Cleveland Clinic Taussig Cancer Institute and a HemOnc Today Next Gen Innovator.

Funchain and colleagues conducted a retrospective review of 600 patients with incurable solid tumor malignancies who underwent tumor genomic profiling at Cleveland Clinic from 2013 to 2016. The results, published in 2018 in JCO Precision Oncology, showed 51.7% of patients were recommended for targeted therapy based on their results. Median OS for patients who went on to receive such therapy was 18 months, compared with 14.4 months for patients who received nongenomics-driven treatment and 5.5 months for patients who received no treatment (P < .001).

The researchers updated their data, reviewing 782 patients who underwent next-generation sequencing between November 2019 and January 2021, in an analysis presented during this year’s virtual ASCO Annual Meeting. The results showed the proportion of patients for whom targeted therapy was recommended increased from 51.7% in their prior study to 73.5%.

“Molecular testing hasn’t just evolved; the word I would use is exploded,” Funchain said. “Yes, the indications have increased, and we have more drugs than we did before. But, importantly for the practicing clinician, molecular testing also has exploded in terms of what’s available. There’s anything from immunohistochemistry, a basic test with fast results, to multigene panels and everything in between.”

Molecular testing also can help patients save costs by not paying for treatments that will not work.

Gutierrez and colleagues conducted a retrospective review of data pooled from electronic health records of 1,497 patients with metastatic colorectal cancer treated across 23 U.S. academic and community practices. Researchers estimated the proportion of patients who would have been considered candidates for anti-EGFR therapies had all patients undergone complete biomarker testing.

Results, published in 2019 in JCO Precision Oncology, showed only 210 patients would have tested wild-type for KRAS, NRAS and BRAF and therefore qualified for anti-EGFR therapy. Yet, the actual testing rates showed 895 patients had no documented mutation, meaning 685 patients may wrongly have been offered anti-EGFR therapy.

Consequently, many patients may have experienced severe infusion reactions, which occur in 1% to 3% of patients treated with anti-EGFR therapies, and increased costs.

“These therapies are expensive relative to testing costs,” the researchers wrote. “Assuming a conservative cost of $6,500 for comprehensive next-generation sequencing and $6,000 per week for cetuximab [Erbitux, Eli Lilly], 32 out of the 685 undergenotyped patients receiving inappropriate therapy for 1 year would cover the costs for testing of all 1,497 patients in this study.”

Underuse

Despite the benefits of molecular testing, experts with whom HemOnc Today spoke agreed that there is a definite need for improvement in its use across cancer types.

Guidelines from National Comprehensive Cancer Network recommend testing for 10 specific biomarkers before initiating treatment for advanced NSCLC, with a specific focus on EGFR in exons 19, 20 and 21. Other guidelines call for testing of six biomarkers in colon cancer, including KRAS/NRAS and BRAF mutations, as well as HER2 amplifications and microsatellite instability (MSI) high/mismatch repair (MMR) status among those with metastatic colorectal cancer.

“The NCCN guidelines are great as they include just about every tumor type and they do a wonderful job of staying up to date in regard to the latest data on which molecular alterations should be guiding therapy,” Lieu said, adding that ASCO also is working on guidelines that are expected this year.

Despite such guidelines, Gutierrez and colleagues found that up to 60% of patients do not undergo the recommended biomarker testing for advanced colorectal cancer.

In their 2019 study, they specifically sought to describe patterns of guideline-aligned biomarker testing among patients with pathologically confirmed metastatic colorectal cancer.

Results showed rates for guideline-based biomarker testing of 41% for RAS, 43% for BRAF and 51% for MSI/MMR deficiency, rates researchers described as “suboptimal.”

Underuse of molecular testing is ubiquitous across cancer types.

For instance, only 18% of 282 patients receiving treatment across 15 community clinics for previously untreated, nonsquamous cell metastatic NSCLC underwent testing for all guideline-recommended biomarkers, according to data from Leighl and colleagues.

In a study of California and Georgia SEER data published in 2019 in Journal of Clinical Oncology, Kurian and colleagues reported that 24.1% of 77,085 patients with breast cancer and 30.9% of 6,001 patients with ovarian cancer underwent genetic testing. That rate among women with ovarian cancer is particularly “inadequate,” the researchers wrote, as guidelines have existed for more than 10 years recommending that all such women undergo testing.

The problem is that not everybody has the time to figure out the nuances of molecular testing, and so it is “under-optimized rather than underused,” Funchain said.

“The good news is that oncologists are aware of the major indications for molecular testing, such as EGFR for lung cancer and BRAF for melanoma — most of us in oncology know what to order,” she said. “Part of the underutilization stems from the difficulty in practitioners knowing how to get the fastest test. For example, with BRAF, many are using molecular sequencing, but this takes 2 weeks to turnaround. Instead, an immunohistochemistry assay can be performed, with the turnaround time of a couple of days. It will only catch BRAF V600E mutations, but that would represent the vast majority of BRAF-mutated cancers. Picking the right test reduces the time to make a treatment decision.”

Time also is a problem when it comes to interpreting results.

“There are others who order the big panel test for all patients because they do not want to have to think about which test to use, but beyond the major indications, they do not have the time to figure out what to do with a gene they have never heard of before,” Funchain added. “These tests are a wonderful source of incidental germline findings. Molecular tests can identify a certain mutation that may indicate hereditary breast cancer that tumor-only sequencing may not necessarily catch. These tests have so much information in them, and we can get much more out of them, but that is difficult to do without the time to do it.”

Certain patient groups may be less likely to receive testing.

During last year’s American Association for Cancer Research Conference on The Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved, Iorgulescu and colleagues found that molecular testing occurred less frequently among older patients, those in historically underrepresented groups, and among uninsured or Medicaid-insured patients.

Underuse of molecular testing was more likely among those of older age (70-79 years vs. 60-69 years, adjusted OR [aOR] = 0.83; 95% CI, 0.77-0.89), men (aOR = 0.94; 95% CI, 0.9-0.99) and Blacks compared with whites (aOR = 0.87; 95% CI, 0.82-0.94).

In addition, compared with patients who were privately insured, those who were uninsured (aOR = 0.78; 95% CI, 0.7-0.86), Medicaid-insured (aOR = 0.87; 95% CI, 0.8-0.94) and Medicare-insured (aOR = 0.87; 95% CI, 0.81-0.93) were less likely to undergo testing, as well as those diagnosed at a community (aOR = 0.6; 95% CI, 0.56-0.66) or comprehensive community cancer program (aOR = 0.76; 95% CI, 0.72-0.8) compared with an academic or NCI comprehensive cancer program.

“Equity of care is a hot topic across medicine and the entire world,” Lieu said. “We want to ensure that molecular testing is available to all patients and there is certainly room for improvement. We know that there are disparities and looking at these disparities in terms of access will be critically important in the future.”

Kim agreed.

“Whether it’s biomarker testing, care or clinical trial eligibility, we absolutely need to create a standard of care that is inclusive,” he said.

Challenges on ‘many different levels’

In addition to the time it takes to order and interpret molecular testing, community-based oncologists also are presented with challenges related to costs and test reimbursement, coordination of sample handling and access.

“Molecular testing is expensive,” Funchain said. “On average, these tests cost about $6,000 per test, which is scary for patients to take on financially. Physicians probably hesitate, as well, to some degree, to order these tests unless they truly need them, and some may not even know how to fully utilize the test.”

Access to molecular testing can differ depending upon location.

“Most oncologists have access to testing, but barriers could be that the laboratory they have access to may only have a limited type of test and turnaround times can be longer than they would desire,” Kim said. “However, it is important to have this information and I strongly encourage our clinicians to put forth the effort to get the testing done for patients, as well as wait for the results.”

Another difficulty with testing is that it requires a large enough sample or biopsy of the tumor material, Kim added.

“In the absence of adequate tumor tissue, we do have liquid-based assays that measure biomarkers of the tumor, but the preference is to test the tissue so we can also obtain a histological diagnosis,” he said.

The logistics behind ordering tests is another challenge, Funchain said.

“Some clinicians are not ordering from their own laboratory, rather they are ordering directly from the company that makes them, and this is yet another relationship to manage and figure out,” she said. “This all becomes just one more very complex thing to add onto an already time-crunched oncologist’s plate.”

Gutierrez agreed.

“Molecular testing is a multifactorial activity that can be accompanied by many access issues — from access to health care, insurance, transportation and so on — that are affected on many different levels,” he said.

Clinicians also may be challenged to keep up with a field that is “evolving at an unprecedented speed,” Gutierrez said.

“We had only one molecular abnormality in lung cancer back in the early 2000s that we were able to target,” he said. “Now, we have at least 10 to 13 targets that can be identified, and this keeps changing.”

The stages at which patients should undergo testing also have evolved.

“Before, we only recommend genetic testing for stage IV disease, but we now recommend testing to be done in stage I disease and beyond,” Gutierrez said. “This evolution alone makes it difficult for physicians and practices to keep up. The integration of electronic health records is yet another roadblock because this is adding even more steps to the process.”

Improving access, clinician support

As the underuse of molecular testing in oncology has become apparent — even as the importance of molecular testing has grown — initiatives and campaigns aim to raise awareness and increase rates of testing and follow-through.

For instance, the Comprehensive Genomic Profiling Coalition is a collaboration among molecular diagnostic companies that aims to improve awareness of molecular testing for advanced cancer.

Also, the “No One is Missed” campaign, launched by the LUNGevity Foundation, seeks to ensure all patients with advanced nonsquamous NSCLC undergo comprehensive biomarker testing.

Likewise, the “Clear Your View” campaign is a collaboration between Guardant Health and patient advocacy groups that encourages clinicians to “stop, test and wait” to obtain biomarker testing results before starting first-line treatment. Last year, the initiative focused on testing in advanced NSCLC and this year, the campaign is focused on advanced colorectal cancer.

Increasing awareness about the use of liquid biopsy is one way to increase access to molecular testing, according to Gutierrez.

“One of the steps we have been taking is to increase awareness about circulating tumor DNA [ctDNA],” he said. “Genomic testing usually can be done on tissue, but it also can be accessible through ctDNA. Tissue is important but, if it is not available, ctDNA can establish the diagnosis for most patients.”

Although these campaigns and initiatives are important, increased testing must be accompanied by an increase in education and support for clinicians, Funchain said.

“I am hesitant to increase testing without increasing support for clinicians because these are complicated tests. There are a lot of technical points about each type of test that have nuances in terms of interpretation with no support system currently for that,” she said. “There are a couple of centers that have genomic tumor boards, but not every center does. There are medical science liaisons for testing companies that are available, but it is patchwork support and, honestly, oncology has become so much busier and there are so many tests to keep track of that our average oncologist does not have time to go out and get that support that is out there. There is a need for support that gets to our oncologists rather than the other way around.”

Lieu said continuing medical education is important because a “one-size-fits-all” strategy for treating cancer is rapidly going away.

“I am always amazed by our general oncologists who do an incredible job of staying up to date in terms of the latest advances for all tumor types,” he said. “The more information we can get out there, the better. We should also focus on the patient side, because if patients are aware, it can increase the likelihood of testing.”

Some cancer centers are seeking to overcome the challenges associated with molecular testing with an in-house comprehensive panel, Kim said.

“For example, at City of Hope, we are taking the lessons we have learned firsthand during the past 20 years as we build our new Orange County Campus and we are proactively implementing standardized molecular testing,” he said. “We have a molecular tumor board setup, and our in-house testing platform can implement precision medicine oncology into regional sites, as well as across our other campuses.

“This is important for patients and how we treat them, and we will ensure that there is access for all communities so that we can provide the best care to all of our patients,” Kim added.

Ultimately, improving access to molecular testing is within the goal of providing the most up-to-date care for patients, Lieu said.

“When thinking about where we were just 2 decades ago — in terms of molecular testing and offering targeted therapies — we did not even have immunotherapy at that time,” he said. “How great it is that we are having this discussion now and that we can offer our patients better and more target-directed or immunotherapy-directed treatments. It really does show the advancements in oncology during the past several years and how quickly this field is moving, which is great because our patients need better and more personalized therapy options.”

• References:
• Gutierrez ME, et al. JCO Precis Oncol. 2019;doi:10.1200/PO.19.00274.
• Iorgulescu JB, et al. Abstract PO-091. Presented at: American Association for Cancer Research Virtual Meeting: The Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; Oct. 2-4, 2020.
• Kim ES, et al. Cancer Discov. 2011;doi:10.1158/2159-8274.CD-10-0010.
• Kurian AW, et al. J Clin Oncol. 2019;doi:10.1200/JCO.18.01854.
• Leighl NB, et al. Clin Cancer Res. 2019;doi:10.1158/1078-0432.CCR-19-0624.
• Sadaps M, et al. Abstract 3128. Presented at: ASCO Annual Meeting (virtual meeting); June 4-8, 2021.
• Sadaps M, et al. JCO Precis Oncol. 2018;doi:10.1200/PO.18.00186.
• Shaw AT, et al. N Engl J Med. 2013;doi:10.1056/NEJMoa1214886.
• Stadler ZK, et al. J Clin Oncol. 2021;doi:10.1200/JCO.20.03661.

• For more information:
• Pauline Funchain, MD, can be reached at funchap@ccf.org.
• Martin E. Gutierrez, MD, can be reached at martin.gutierrez@hackensackmeridian.org.
• Edward S. Kim, MD, MBA, FACP, FASCO, can be reached at City of Hope Orange County, 15161 Alton Parkway, Irvine, CA 92602.
• Christopher Lieu, MD, can be reached at christopher.lieu@cuanschutz.edu.

Click here to read the Point/Counter to this Cover Story.