NEW YORK – Precision medicine advocates are encouraging the healthcare community to consider the clinical utility of genomic profiling for cancer patients and healthcare systems in broader terms and move beyond more limited and outdated definitions.
In a commentary in JCO Precision Oncology last month, Daryl Pritchard from the advocacy organization Personalized Medicine Coalition, Clifford Goodman from the UnitedHealthcare-owned consulting firm Lewin Group, and Lincoln Nadauld from Intermountain Healthcare wrote that when those in the field currently consider the clinical utility of genetic testing in cancer, they only look at the benefits and risks associated with positive and negative test results.
"These definitions of clinical utility are inadequate to recognize the wider scope of benefits that accrue from more comprehensive genomic tests," they wrote. Based on a roundtable discussion with leaders from across the healthcare system, Nadauld and colleagues penned a new definition of clinical utility and published recommendations that they believe will allow stakeholders to more fully capture the impact of precision oncology interventions in patient care.
In their new definition, the authors consider the net benefit to cancer patients and health systems stemming from improved patient outcomes from screening, prevention, treatment, clinical trial enrollment, doctor-patient shared decision making, and reduced care disparities. "The utility of genomic profiling depends on its ability to provide information that is used to guide patients more efficiently to safer and more effective prevention and treatment strategies and its ability to improve the practical knowledgebase for health system decision making," the authors said in the paper.
Since its publication online on Jan. 27, the commentary has already climbed to the top of JCO Precision Oncology's "most read" list, which perhaps reflects the field's widespread interest in overcoming existing reimbursement and other barriers to multigene profiling and making it accessible to more cancer patients.
"We've heard from payors, providers, and industry partners that they are thrilled this definition has come out," said Nadauld, chief of precision health at Intermountain, a healthcare system serving Utah, Idaho, and Nevada where advanced cancer patients receive genomic profiling as a routine part of their care. "They've all felt like this is long overdue."
Nadauld, Goodman, and Pritchard argue in their commentary that providers, health systems, and payors rely on different and narrow definitions of clinical utility when deciding whether to order, invest in, or reimburse multigene panel testing. One widely cited 2006 definition published in Genetics in Medicine, for example, defines clinical utility as "the ability of a screening or diagnostic test to prevent or ameliorate adverse health outcomes."
While the authors of the latest paper don't dispute that this is one way genetic tests can benefit patients, they see other positives, especially since genetic testing in cancer care looks very different now compared to 2006, when single-gene tests were the lay of the land as opposed to multigene next-generation sequencing panels and other advanced tests.
"Previous definitions of clinical utility of genetic testing had mostly been developed before the advent of multiplex testing," said Pritchard, PMC's senior VP of science policy. "In the commentary, we focus on genomic tests [including] whole-genome, exome, or multigene tumor profiling acquired through a sequencing platform."
Nadauld explained that the experts included in the roundtable considered "multigene" tests to mean those that assess several hundred genes. For example, the in-house NGS panel that Intermountain uses to profile solid tumors, TheraMap, includes over 500 genes. Commercially available NGS panels like Foundation Medicine's FoundationOne CDx gauge over 320 genes and several gene signatures, like tumor mutational burden.
When evaluating the impact of multigene tests like these, the authors recommend healthcare stakeholders consider the tests' ability to improve clinical outcomes; the benefits to health systems and the populations they serve in efficiently processing samples for testing and compiling outcomes data from those tested; the extent to which test results inform prevention and precision treatment decisions, including access to clinical trials; and whether test results are being acted on with input from patients.
These criteria, Pritchard said, "should be factored into a health system's policies and processes if they want to realize the full utility of personalized medicine technologies."
The roundtable that PMC convened in late 2019 to craft the expanded clinical utility definition brought together providers, payors, patients, clinical guideline developers, and laboratory directors. "And [then], we just unpacked this entire situation," Nadauld said. The payor presence was particularly crucial, he noted, since reimbursement remains one of the biggest barriers to accessing precision oncology.
Roundtable participants representing the payor perspective included Vincent Nelson at Centene, who at the time was Blue Cross Blue Shield's VP of medical affairs. One of the paper's authors, Goodman, works for the Lewin Group, which is a consulting business that is "payor adjacent" in that it shares the same parent company, UnitedHealth Group, as the insurer by the same name.
"If [payors] would pay for [multigene tests] more regularly, our providers would order them more regularly, and then patients would get more access to clinical trials and have all the better outcomes that we know accompany these types of tests," Nadauld said. "The lag in the payor community is our number one barrier."
Although the Centers for Medicare & Medicaid Services in 2018 agreed to cover US Food and Drug Administration-approved multigene panels for advanced cancer patients, commercial payor coverage is still inconsistent. Commercial payors have different coverage policies for large panels according to patients' cancer type and treatment setting, the evidence underlying specific platforms, and internal pricing arrangements with specific test providers.
Broadly speaking, commercial payors, traditionally focused on the bottom line, want to see evidence that tests match patients to treatments that work, which in turn can help payors avoid spending on treatments that don't work. But the results of multigene panels that gauge hundreds of genes, with varying levels of clinical evidence, can point to FDA-approved treatments, off-label drugs, or investigational treatments in clinical trials. When testing points to more experimental options, there may be less evidence the treatments will benefit patients, making payors more reluctant to pay for it.
When testing doesn't point to a clear FDA-approved therapy, even doctors are challenged to figure out what course to take for patients. "A lot of physicians … look at a long list of gene variants and throw up their arms," Nadauld said. "If they get a really standard variant like a BRAF mutation or an EGFR mutation, they know what to do with that. But they don't know what to do with the more nuanced variants [like] a HER2 amplification outside of breast cancer or a CDK4/6 variant."
Common situations like this create a negative loop, in which payors don't universally cover comprehensive tests because they fear unnecessary spending, oncologists don't order the tests fearing coverage denials, and as a result, the clinical utility evidence on these tests aren't generated, which would allow labs and healthcare systems to make a case for broader coverage with payors.
Based on his experience implementing precision oncology at Intermountain, Nadauld is convinced that if more cancer centers and health systems make a commitment to providing genetic testing for more cancer patients based on a broader definition of clinical utility, it could break this negative loop. Intermountain began offering comprehensive genomic profiling to all advanced cancer patients in 2014 and has since grown the breadth of testing options and resources for patients and doctors.
"At Intermountain, for the first year or so that we were doing comprehensive genomic profiling on every [advanced cancer] patient, we had heavy reliance on molecular tumor boards [to help with interpretation]," he recalled. "But after some time, our providers became increasingly comfortable doing their interpretations."
Nadauld's team at Intermountain has also tracked the clinical utility and cost-effectiveness of its broad genomic testing strategy. In a published paper several years ago, they showed that patients who had genomic testing to inform treatment had better progression-free survival than those who received standard treatments. This analysis also found that the groups had comparable weekly care charges reflecting fewer toxicities, emergency room visits, and hospital admissions in the precision care setting. This was despite the total cost of care — factoring in cost of treatment, associated radiology and lab work, toxicity, and sequencing — being higher in the precision care arm.
The findings weren't a ringing endorsement of precision oncology's clinical utility using traditional outcomes and cost-effectiveness parameters. But to Nadauld and other advocates it demonstrated the need to consider clinical utility in broader terms. With this latest commentary, the PMC is hoping other stakeholders in the field will embrace an expanded definition in the hopes that it will not only improve test access for patients but convince cancer centers to invest in the infrastructure — databases, sample procurement procedures, molecular tumor boards, genetic counselors, and nurse navigators — that early precision oncology adopters like Intermountain have implemented and are relying on to make the case for better coverage with payors.
Though PMC believes the payor perspective was adequately represented in its roundtable, it remains unclear whether this broader conceptualization of clinical utility will be embraced by the payor community. Nelson, who worked for BCBS when he participated in the roundtable, and paper co-author Goodman from the Lewin Group, both declined to provide comments for this article, saying they would need to seek permission from their employers and that they didn't want to speak on behalf of the payor industry.
But PMC's Pritchard maintained that since payors were involved in the discussions and in drafting the commentary, this affirms their support of the expanded definition. "Payors contributed to the development of the definition [and] knowing that this was developed with payor perspectives included may bolster its recognition with other payors," he said. "All in all, we are just trying to make the point that it could and should be used by all different stakeholders, including payors, as they consider policies and perspectives."
Recognizing that provider knowledge and payor coverage gaps are hindering precision oncology uptake, the authors included a series of recommendations for implementing multigene tumor profiling. One recommendation is to "use continuously updated decision support mechanisms to help clinicians determine when to order genomic testing and how to interpret the test results," and another is to "provide consistent evidence-based coverage and value-based reimbursement of genomic testing for appropriate cancer cases."
In terms of the clinical decision support mechanisms that can help doctors make sense of test reports, Nadauld highlighted molecular tumor boards, in addition to guidelines and treatment pathways. Knowledgebases that provide evidence-based ratings on the actionability of genomic findings, including the European Society for Medical Oncology's Scale for Clinical Actionability, or ESCAT, the Memorial Sloan Kettering Oncology Knowledgebase, and ClinGen are also useful. According to Pritchard, these resources help validate individual biomarkers and foster community-wide acceptance of genetic test results.
In a study presented during the 2021 San Antonio Breast Cancer Symposium, Fabrice André, head of research at the Institute Gustave Roussy in France, spoke to the value of "ranking" genomic alterations in molecular testing reports. His findings showed that metastatic breast cancer patients treated according to genomic alterations scored in the top two ESCAT tiers had improved survival outcomes, but those treated with alterations in the lesser-validated tiers experienced better outcomes on standard chemo.
As this study shows, results from multigene panels can have varying levels of utility on an individual level, depending on whether a patient has a tumor alteration with strong evidence of actionability. Patients' ability to benefit from multigene testing can also vary based on the type of cancer they have and where they receive care.
Nadauld and colleagues proposed the four elements of clinical utility — improved outcomes, health system benefit, clinical trial eligibility, and shared decision making — recognizing that large and small health systems, providers, payors, and patients will embrace precision oncology with different aims. For example, a community hospital system may place the most value on improving clinical outcomes and benefiting the larger health system, whereas patients and caregivers may value shared decision making alongside improved outcomes most. Payors, meanwhile, may prioritize the individual patient outcome improvements in the short term, and in the long run, with a transition to value-based care, prioritize the population-level benefit, too.
"These kinds of genomic tests really enable value-based healthcare, because the value accrual is spread across the entire spectrum of healthcare," Nadauld said. "A single biomarker test for a single drug is all about fee-for-service." Going forward, he believes that a gradual transition from fee-for-service to value-based payment models will lead to broader coverage for multigene panels.
Transitioning to value-based care in oncology will take years. In the meantime, Nadauld said, the number one thing that the field can do today is to "ensure that all patients with advanced or metastatic cancer are getting comprehensive genomic profiling as early in their care as possible."
"That will do the most for patients and the most for educating providers," he said. "It will send a message to payors that this is how providers are practicing."