SAN FRANCISCO – A recent study has found that testing non-small cell lung cancer patients with multi-gene next-generation sequencing panels is moderately cost effective compared to testing for single markers.
However, only a small percentage of patients where multi-gene testing uncovered an actionable mutation were able to receive targeted treatment.
The study, funded by the Personalized Medicine Coalition, was published last month in the Journal of Clinical Oncology.
Stakeholders have debated the benefits and drawbacks of NGS-based testing for cancer patients, in particular the extent to which such tests should be covered by payors, with advocates arguing that such tests can identify a wider range of mutations that can suggest potential targeted therapies, while others think that testing does not have broad utility across the range of cancers.
Lead author Lotte Steuten, who is now head of consulting and vice president at the Office of Health Economics in the UK, said that a goal of the study, which she began while at the Fred Hutchinson Cancer Center in Seattle, was to add some data to the debate, and to look at how such testing impacted patient outcomes as well as costs.
Thus far, such cost-effectiveness studies have been limited and predominantly performed in small patient cohorts. For instance, a study by Intermountain Healthcare in 72 patients found that those who received NGS testing and molecularly guided treatment had better outcomes than those who received standard of care, without increased costs of care.
Researchers from MD Anderson Cancer Center have also found that molecular profiling can lead to better outcomes in cancer patients, reporting that when patients with actionable mutations were treated with matched targeted therapies, they had better responses and survived longer than patients who were treated with unmatched therapies.
Steuten said that the researchers wanted to add to these other studies. The goal was to try and "understand the economic value of" next-generation sequencing tests. "To what extent can we demonstrate the value beyond just the cost of the new, multi-gene tests versus the old single-marker tests," she said.
To conduct the study, the researchers tapped data from Flatiron Health's database, which includes patients seen at a variety of cancer centers in the US. They evaluated data from patients with advanced non-small cell lung cancer who had been diagnosed between 2011 and 2016. Patients were divided into groups according to whether they received single-marker or multi-marker testing, defined as an NGS test that included 30 or more genes, and then the researchers analyzed the percentage of patients from each group that received targeted therapy as well as the survival rates and total costs for each group.
Overall, the researchers analyzed 5,688 patients, 875 of whom received multi-marker testing. They found that of patients who received NGS testing, 30 percent had an actionable mutation and 21 percent received targeted therapy. Of patients who received single-marker testing, 23 percent had an actionable mutation and 19 percent received targetable therapy. Survival was slightly higher in patients who received multi-marker testing, at 1.20 life years versus 1.14 life years. Total lifetime costs were $8,814 higher per patient who received multi-marker testing, resulting in a cost-effectiveness ratio of $148,478 per life-year gained.
In general, Steuten said, interventions in the range of $50,000 to $200,000 per life year gained are deemed acceptable costs, so the study found that multi-gene testing was in this acceptable cost-effective range, but that it wasn't "wildly cost effective," she said.
Notably, however, the researchers found a discrepancy between the number of patients for whom a multi-marker test found an actionable mutation and the percentage of those patients who actually received a targeted therapy.
"Only a small fraction actually receive that targeted treatment," Steuten said. "That reduces the value of testing, because if you don't act on the results, then the patients don't get the benefit."
And indeed, the researchers found better outcomes among the patients who actually received targeted therapy, regardless of whether or not they had NGS testing, with mean overall survival at 2.13 years versus 1.73 years for patients who did not receive targeted therapy.
That discrepancy between the number of patients for whom multi-gene testing found an actionable mutation and those who actually received targeted treatment also had an impact on cost effectiveness. If all those patients had received targeted treatment, the researchers estimated that the cost-effectiveness ratio would have been $110,000 per life year gained.
Steuten said that one avenue for further research is to try and identify the reasons patients aren't receiving targeted treatment even when a test identifies an actionable mutation. Sometimes the patient is too sick, but there are also other barriers to treatment access, including financial barriers and geographical ones, she said. Sometimes a payor will not cover a specific treatment or the out-of-pocket cost may be too high for the patient. Other times, a test indicates eligibility for a clinical trial but the patient is unable to travel. She added that while this study was not designed to identify the specific causes for these patients, doing so would be an important next step.
Her group has also been conducting a similar cost-effectiveness study for multi-gene testing in melanoma, which she said would be likely published later this year. The results were similar to the lung cancer study in that the researchers found that multi-gene testing was moderately cost-effective. Steuten added that for melanoma, testing seems to be slightly more cost-effective that for lung cancer, and that a higher proportion of patients who have actionable mutations ultimately receive the indicated targeted treatment. "We didn't see such a huge drop-off in patients struggling to get access to targeted therapies," she said.
Another issue that Steuten's group did not study was how testing at different stages of cancer progression would impact cost effectiveness. In the lung cancer study, the researchers evaluated only patients with advanced cancer. And, as Steuten noted, sometimes even if the multi-gene test identified an actionable target, patients were too sick to start on the targeted treatment. In addition, patients with advanced cancer have already been treated with chemotherapy and other drugs. "We know that targeted treatments are slightly less effective in patients who have already been heavily treated," she said. However, from an economic perspective, testing patients at earlier stages may lead to more patients receiving the much more expensive targeted agents — sometimes six to 10 times more expensive than standard chemotherapy — potentially even patients who would benefit from the standard of care. "That should be investigated," she said, and it should be evaluated whether testing earlier is "something worth doing that leads to better outcomes and is something that the health system can afford."