Broader NGS Use in Oncology Hinges on Payers, Access, and Education

David Spigel, MD

David R. Spigel, MD, chief scientific officer and director of lung cancer research at Sarah Cannon Research Institute in Nashville, Tennessee, is among a growing number of believers in the power of next-generation sequencing (NGS) of tumor tissue. They say that using up-front, broadpanel genomic tests that include hundreds of genes can save money and, in some cases, improve outcomes compared with other diagnostic approaches, especially in lung cancer but also increasingly in breast, colorectal, skin, and other cancers.

“If you can do that testing in a comprehensive, efficient way, that’s probably the best way to do things. I don’t view this much differently than I would getting a scan to properly stage a patient. How am I going to meet a patient tomorrow in clinic and have all the information I need to chart the best course for that patient’s care? Genomic results are part of that equation,” Spigel said.

NGS testing uses high-throughput, massively parallel sequencing methods to read DNA or RNA more quickly and cheaply than the older Sanger method, which uses primer-extension and chain-termination techniques for sequencing polynucleotides. Use of NGS tests to aid in prognosis, identify targetable alterations, and assign patients to clinical trials is expanding, with recent surveys finding that about 75% of oncologists say they have ordered the tests for some of their patients.^1,2 At several research institutions, broad-panel genomic tests are now part of the standard workup for patients with metastatic disease.

The growth of NGS has been driven by the falling cost of sequencing and the increasing number of targeted drugs. The advent of tumor-agnostic therapies has also contributed to enthusiasm for routine genomic profiling. The number of actionable alterations has increased to the point that switching to up-front NGS testing in advanced non—small cell lung cancer (NSCLC) would be less costly for payers than giving a series of singlegene marker tests, according to the results of a recent study.³ Other analyses find that in a number of clinical scenarios, NGS tests have clinical benefit, are cost-effective, and increase patients’ quality-adjusted life years (QALYs; Table 1).⁴

Yet, the utility of NGS testing is not uniformly accepted, especially given its higher cost relative to other marker tests. Many payers will not cover large multigene panels that are not approved as companion diagnostics or prognostic tools, describing them as experimental, investigational, or not medically necessary. In the community setting, poor access to trials also often limits the usefulness and cost-efficiency of broad gene panels.

As a result, NGS is still used rarely in many clinics. In a survey of ASCO members, 69% of community oncologists in North America reported that they order the tests for a quarter or less of their patients with metastatic cancer, and 19% said they never order them.² Eightysix percent of oncologists encounter barriers to using NGS, with many saying they have difficulty finding trials (52%), getting drug payments (48%), or receiving payment for the genomic test (43%).

NGS Knowledge Gap

Another recent survey found that community oncologists use gene profiling far less frequently than academic oncologists, ordering the tests in 33% of lung cancer cases compared with 74% for their academic counterparts.⁵ The study also found a knowledge gap with regard to tumor profiling. When asked to match molecular alterations with targeted therapies, 69% of respondents answered incorrectly or said they didn’t know. Physicians also continue to struggle to manage the large amounts of data with unclear therapeutic significance that are produced by comprehensive genomic profiling (Table 2⁵, Figure⁵).

Since NGS methods were first developed in the early 2000s, the cost of whole-genome sequencing has plummeted to $1301 per genome, according to National Institutes of Health (NIH) data.6 Commonly used panel tests use the same methods but focus on a smaller set of genes. They also cost more, as the NIH figure does not include all the costs of running a sequencing lab or generating clinically useful reports.

FoundationOne CDx, an FDA-approved panel that detects mutations in 324 genes, has a list price of $5800. The alterations it tests for include indications for 18 FDA-approved therapies, along with tumor mutational burden (TMB), microsatellite instability, and, optionally, PD-L1. A similarly broad, nonapproved laboratory-developed test from Caris Molecular Intelligence costs $6500. The actual sum paid for such tests is often much lower: An analysis of claims data pegged the cost of a broad NGS panel at $2860 for commercial payers, while Medicare would pay $627.50.³

Spigel said NGS tests commonly cost $3000 to $4000, although he is aware of a laboratory that offers a panel for $1800. He said blood-based circulating tumor DNA assays are somewhat more expensive. Several companies also offer smaller-panel NGS-based tests, such as Thermo Fisher Scientific’s Oncomine Dx Target Test for NSCLC, which includes 23 genes and 3 other markers. NGS can also be used to develop polygenic risk scores for breast cancer and other conditions, although scoring tests often use non-NGS methods in practice.

The FDA lists more than 60 DNA- and RNA-based cancer-related diagnostic tests that the Center for Devices and Radiological Health has approved or cleared for marketing, some of which use NGS.7 The NIH’s Genetic Testing Registry lists more than 1100 NGS tests from US labs and providers that include cancerrelated analyses, as well as hundreds more in other countries.8 These include commercially available and in-house tests.

NSCLC is Ideal for Testing

Table 1. NGS Cost-Effectiveness by Strategy, According to Results From Multiple Studies^4,a

NGS is particularly relevant to treatment of NSCLC because numerous drugs are available to target alterations associated with the disease. Nathan A. Pennell, MD, PhD, director of the Lung Cancer Medical Oncology Program at the Cleveland Clinic Taussig Cancer Institute, said treatment guidelines for metastatic NSCLC recommend EGFR, ALK, ROS1, and, in some cases, BRAF V600E testing. He said recent trial results suggest that tests for MET exon 14 skipping mutations and RET rearrangements may be added to the lists in the near future, and he noted that KRAS and HER2 have shown promise as targets. Clinics often test for alterations one at a time using immunohistochemistry, fluorescence in situ hybridization, and polymerase chain reaction-based (PCR) methods, either sequentially in a process of elimination or in a multigene panel that reports on several genes at once. For commercial payers, costs for individual tests range from $406 for BRAF to $3299 for a multigene panel.³ For patients with Medicare, they range from $180.20 for BRAF to $1345.50 for a multigene panel, according to data cited in a study Pennell co-authored.

Pennell’s study used a decision analytic model to calculate testing costs in a hypothetical million-member health plan with 2066 Medicare-insured and 156 commercially insured patients with metastatic NSCLC. It compared 4 testing models: exclusionary KRAS followed by sequential single-gene testing for 9 alterations, sequential, hotspot panel, and up-front NGS. All patients had separate PD-L1 testing. NGS was associated with cost savings for both CMS ($1,393,678; $1,530,869; and $2,140,795 less than exclusionary, sequential testing, and hotspot panels, respectively) and commercial payers ($3809; $127,402; and $250,842 less). NGS also delivered results more quickly than working through a series of smaller tests.

Pennell said other advantages of NGS testing include reduced need for repeat biopsies, which may not be feasible for some patients, and discovery of rare druggable alterations such as TRK fusions. He also noted that panel testing should be able to provide more information without added expense, because sequencing costs are the same whether the result is a report on 5 targets or 350.

As payers and providers try to understand the utility of comprehensive genetic profiling, Pennell said the study’s findings clearly demonstrate its value in NSCLC. “There are arguments within the field about whether we should be doing broad NGS for all cancer patients. There’s a lot of debate about how useful that is, but no one really believes that you should include lung cancer in that argument. This is really the poster child for doing broad genomic testing, because there are so many actionable targets and it’s such a common disease,” he said.

NGS May Improve Outcomes

Table 2. Survey Identifies High Uncertainty About Application of Molecular Profiling Among Community Oncologists^5,a

Figure. Community Versus Academic Use of Molecular Profiling⁵

Other data modeling efforts have found that in some scenarios, up-front NGS testing not only saves money but may also improve outcomes compared with other diagnostic approaches. In one study, testing with 5- to 50-gene tumor panels for advanced NSCLC more than doubled the use of targeted therapy and sharply decreased nontargeted therapy, reducing overall spending.⁹ Patients had fewer adverse events, average progression-free survival increased, and trial enrollment rose dramatically. The potential for costs savings was shown in a 2015 study that compared a 34-gene NGS panel with a single-site BRAF V600 mutation test for patients with metastatic melanoma.¹⁰ In both strategies, patients who tested positive for BRAF received vemurafenib (Zelboraf). Patients who tested negative by single-site test received ipilimumab (Yervoy). Those who were negative by gene panel received imatinib if they carried KIT mutations, received ipilimumab if they carried no actionable mutations, or were enrolled in a clinical trial if they had other actionable mutations. Over a 2-year period, panel testing resulted in reduced total cost of $8943 per patient compared with single-site testing ($120,022 vs $128,965). The use of NGS also resulted in a 0.0174 increase in QALYs per patient (0.721 vs 0.0704). One QALY is equal to a year in good health. Investigators concluded that the NGS strategy could result in an annual savings of $79.6 million and a gain of 155 QALYs if applied to the 8900 patients with newly diagnosed metastatic melanoma. Another study found that gene panel testing increases QALYs and reduces costs for patients with colorectal cancer and polyposis syndromes (CRCP), resulting in an incremental cost-effectiveness ratio (ICER) of $36,500 per QALY compared with standard testing methods.¹¹ That is well below a cost-effectiveness threshold of $100,000 per QALY. In a Canadian review of 55 NGS cost-effectiveness studies, several studies concluded that gene panels to predict disease prognosis in breast cancer had ICERs below $50,000 per QALY.⁴ Some of the other studies found that using NGS panels was either more cost-effective than standard methods or was under ICER thresholds. These included prognostic NGS panels in CRCP, head and neck cancers, leukemia, and lung cancer, and diagnostic panels in head and neck cancers, thyroid cancer, CRCP, and unknown primary cancers.

The Canadian investigators noted a number of limitations to past cost-effectiveness studies of NGS. Most economic evaluations they examined were not conducted alongside clinical trials. Few studies examined the cost-effectiveness of applying NGS to identify targeted therapeutic options. The breast cancer riskrecurrence studies they reviewed often included long-term time horizons, which can involve extrapolation beyond available data and may yield favorable ICER estimates. Privately funded studies yielded lower, more favorable ICER results than publicly funded studies.

In addition, in some studies, NGS has shown little or no benefit to alternative diagnostic methods. A Chinese study of ALK testing for patients with advanced NSCLC found that NGS panel testing was no more cost-effective than multiplex PCR.¹² Patients in the Chinese healthcare system using NGS spent $31,388 and gained 0.780 QALYs, while those using PCR spent $31,362 and gained 0.780 QALYs. The results were highly dependent on financial assistance for drug coverage. When assistance was not available, the ICER for gene-guided treatment strategies was about 6 times higher than the threshold of willingness-to-pay of $32,000 per QALY gained.

Operative Word is Actionable

Even in conditions like advanced breast cancer, where data modeling suggests that NGS testing can lead to longer survival or better quality of life, the lack of approved targeted therapies blunts arguments for the usefulness of up-front NGS. Sara M. Tolaney, MD, MPH, associate director, Susan F. Smith Center for Women’s Cancers, Dana-Farber Cancer Institute, Boston, Massachusetts, noted that there were no FDA-labeled indications based on somatic alterations in breast cancer until May 2019. That’s when the agency approved alpelisib (Piqray) for use in combination with fulvestrant (Faslodex) as a treatment for postmenopausal women, and men, with hormone receptor—positive, HER2-negative, PIK3CA-mutated, advanced, or metastatic breast cancer following progression on or after an endocrine-based regimen.

The utility of broad genomic panels is “very disease dependent,” said Tolaney, who coauthored the study of tumor genomics utilization among ASCO members. “In breast cancer, you’re not getting as much bang for your buck out of NGS testing. You’d be fine if you just did hotspot testing. PIK3CA is the only indication, at least at this time.” Other test findings are only useful for getting patients onto trials, she said.

Tolaney said all Dana-Farber patients with metastatic breast cancer are tested using OncoPanel, an in-house gene assay. She looks for markers such as high TMB and ESR1, ATK, and HER2 mutations that may make her patients eligible for trials. But for clinicians in settings where trial access is limited, it can be much more difficult to find benefits in NGS, she said.

“It’s a challenge, particularly if you’re not at a big academic institution where you have all these resources at your fingertips. A lot of oncologists in the community spend a lot of time trying to find a trial for their patient, and it’s not easy to do. That’s definitely a barrier,” she said.

The 30 clinicians at the Center for Cancer and Blood Disorders in Fort Worth, Texas, use 1- or 2-gene non-NGS tests as needed, but they would prefer to order broad gene panels more often, said Jennifer Sassano, MSN, MHA, RN, OCN, director of clinical operations. “Instead of trying to play the blind person game, you have it all laid out on the table. You have all of that direction up front instead of going, ‘Well, this didn’t work. OK, let’s go to the next one.’ Each of those individual tests runs up a cost,” she said.

Sassano said the center participates in trials, including some that require NGS testing. But when the group’s physicians try to order panels for regular clinical use, they typically are denied preauthorization by payers. “Right now, we’re kind of on the fence with using NGS, just because at least initially we’ve come up against a lot of barriers,” she said.

Yet some payers are more willing to cover broad panel tests, including some Blue Cross plans.^13,14 Last year, CMS expanded Medicare coverage of NGS, allowing payment for FDA-approved or -cleared companion diagnostics for patients with recurrent, relapsed, refractory, metastatic, or advanced cancer. Following complaints that the new policy removed coverage for germline NGS testing for early-stage patients, CMS reopened its national coverage determination and is expected to announce modified rules early next year.

Spigel said Sarah Cannon doctors have been using NGS panels widely for several years and almost never encounter payment problems. To help ensure patients are getting maximum benefit from testing, the group launched a personalized medicine program staffed with genetics specialists who advise physicians on treatment options. Spigel said the effort has also helped boost enrollment in clinical trials.

He said he has heard an argument that NGS testing is like playing the lottery, in that it provides a major benefit on rare occasions, like discovery of a targetable NTRK fusion, which encode TRK kinases, but is otherwise a waste of money.

“I get that analogy if all we had was TRK. At the same time, if it’s my mother who has cancer, I want to know about relevant targets, TRK and beyond, that we have potential studies or therapies for. Or even more so, what we shouldn’t do, from emerging data about maybe who won’t benefit from certain therapies. That would be important information,” he said. “And if the cost is getting down to the cost of a PET scan or cheaper, I don’t really buy that argument any more that it’s too expensive and it’s not going to help you.”

References

1. Freedman AN, Klabunde CN, Wiant K, et al. Use of next-generation sequencing tests to guide cancer treatment: results from a nationally representative survey of oncologists in the United States. JCO Precis Oncol. 2018;2:1-13. doi: 10.1200/PO.18.00169.
2. Barroso-Sousa R, Guo H, Srivastava P, et al. Utilization of tumor genomics in clinical practice: an international survey among ASCO members. Future Oncol. 2019;15(21):2463-2470. doi: 10.2217/fon-2019-0010.
3. Pennell NA, Mutebi A, Zhou ZY, et al. Economic impact of next-generation sequencing versus single-gene testing to detect genomic alterations in metastatic non—small-cell lung cancer using a decision analytic model. JCO Precis Oncol. 2019;3:1-9. doi: 10.1200/PO.18.00356.
4. Weymann D, Pataky R, Regier DA. Economic evaluations of next—generation precision oncology: a critical review. JCO Precis Oncol. 2018;2:1-23. doi: 10.1200/PO.17.00311.
5. Singh BP, Britton SL, Prins P, et al. Molecular profiling (MP) for malignancies: knowledge gaps and variable practice patterns among United States oncologists (onc). Presented at: 2019 American Society of Clinical Oncology Annual Meeting; May 31- June 4, 2019; Chicago, IL. Abstract 10510. meetinglibrary.asco. org/record/173392/abstract.
6. DNA sequencing costs: data. National Human Genome Research Institute website. genome.gov/about-genomics/fact-sheets/ DNA-Sequencing-Costs-Data. Accessed July 23, 2019.
7. Nucleic acid based tests. FDA website. www.fda.gov/medical- devices/vitro-diagnostics/nucleic-acid-based-tests. Accessed July 23, 2019.
8. GTR: Genetic Testing Registry. National Center for Biotechnology Information website. ncbi.nlm.nih.gov/gtr. Accessed July 23, 2019.
9. Sabatini LM, Mathews C, Ptak D, et al. Genomic sequencing procedure microcosting analysis and health economic cost-impact analysis: a report of the Association for Molecular Pathology. J Mol Diagn 2016;18(3):319-328. doi: 10.1016/j. jmoldx.2015.11.010.
10. Li Y, Bare LA, Bender RA, et al. Cost effectiveness of sequencing 34 cancer-associated genes as an aid for treatment selection in patients with metastatic melanoma. Mol Diagn Ther. 2015;19(3):169-177. doi: 10.1007/s40291-015-0140-9.
11. Gallego CJ, Shirts BH, Bennette CS, et al. Next-generation sequencing panels for the diagnosis of colorectal cancer and polyposis syndromes: a cost-effectiveness analysis. J Clin Oncol. 2015;33(18):2084-2091. doi: 10.1200/JCO.2014.59.3665.
12. Lu S, Yu Y, Fu S, Ren H. Cost-effectiveness of ALK testing and first-line crizotinib therapy for non-small-cell lung cancer in China. PLoS One. 2018;13(10):e0205827. doi: 10.1371/journal. pone.0205827.
13. Independence Blue Cross becomes first major insurer to cover next-generation whole genome sequencing for a variety of cancers [press release]. Philadelphia, PA: BlueCross Blue Shield; January 11, 2016. bcbs.com/news/press-releases/independence- blue-cross-becomes-first-major-insurer-cover-next-generation- whole. Accessed July 23, 2019.
14. Blue Cross Blue Shield of Massachusetts. Medical policy: expanded molecular panel testing of cancers to identify targeted therapies. BCBSM website. bluecrossma.com/common/ en_US/medical_policies/790%20Expanded%20Molecular%20 Panel%20Testing%20of%20Cancers%20to%20Identify%20Targeted% 20Therapies%20prn.pdf. Accessed July 23, 2019.