Pathwork Diagnostics last week released a health-economic analysis that concluded its Tumor of Origin test can increase overall survival of patients with metastatic cancer of unknown origin at a cost-effectiveness level well below the generally accepted threshold for new technologies.
Working with collaborators at Cedar Associates, Stanford University, Memorial University of Newfoundland, and the University of Washington, the company created a model combining the data from a retrospective observational study of 107 patients tested by Pathwork's TOO test with information from clinical trials and databases.
The model showed that the use of the test to guide treatment regimens should result in longer survival for CUP patients at a cost per quality-adjusted life year of around $47,000 — less than half the generally accepted threshold of $100,000 for cost-effectiveness in the US, according to Pathwork.
The researchers presented the results in a poster at the American Association for Cancer Research and International Association for the Study of Lung Cancer joint conference last week.
Shawn Becker, Pathwork's vice president of marketing and reimbursement, told PGx Reporter that the results demonstrated that the TOO test had a clear impact on outcome, with a projected gain of 3.6 months overall survival for those tested. With costs per year falling below $50,000, the test also showed itself to be "strongly cost-effective," he said.
According to Becker, the study was spurred by interest from both payors and doctors in seeing harder evidence of the test's economic efficiency.
"When [doctors] can't identify where the cancer came from using traditional techniques, they add technologies such as the Pathwork Tissue of Origin test," Becker said. "Naturally in today's [cost-conscious] environment, [payors and healthcare providers are] raising the question of how cost effective [this] testing is versus the usual care."
The study "set out to answer that question," and to show that Pathwork's test can compete in an increasingly cost-focused environment, he said. "It's really good to supplement more intuitive arguments [about the value of this testing] with some real data, and rigorous modeling."
Additionally, Becker said, the company believes that clear cost-effectiveness modeling can only help Pathwork if and when it seeks to offer the test outside the US.
"Though Pathwork is currently focused on the US market, many international markets are even more focused on this cost-effectiveness data for their payor decisions, so this helps us set the stage for international expansion as well," he said.
Pathwork received US Food and Drug Administration clearance in 2008 for its TOO test for use with frozen tissue samples (PGx Reporter 8/06/08) and in 2010 for use with formalin-fixed, paraffin-embedded samples. The test is based on a custom microarray manufactured by Affymetrix and compares the expression of more than 2,000 genes in the patient’s tumor to expression patterns of 15 known tumor types, representing 58 morphologies and 90 percent of all solid tumors.
In the cost-benefit analysis, the research team used data from an observational study tracking treatment changes made for patients by physicians who received TOO test results. Pathwork recorded changes in planned chemotherapy, surgery, radiation therapy, blood tests, imaging, and referral to hospice care before and after TOO testing.
The observational study found that after TOO testing, 23 percent more patients received a chemotherapy regimen consistent with guidelines for the final tumor-site diagnosis and 15 percent fewer patients received non-guideline-consistent chemotherapy regimens.
In the cost-effectiveness study, Pathwork and its collaborating researchers projected the effect of these changes in chemotherapy on survival based on data from the National Comprehensive Cancer Network, they reported in their poster.
They based calculations of drug and administration costs on average doses reported in NCCN guidelines and used Centers for Medicare and Medicaid Services fee schedules to obtain other unit costs, they wrote, estimating changes in overall survival, costs, and cost per quality-adjusted life year using "bootstrap methods."
Based on its modeling, Pathwork found that "overall survival was projected to increase 3.6 months from 15.9 to 19.5 months," Becker said. Adjusted for quality of life, the increase was 2.7 months.
Based on the overall increase in survival, "average third-party payor costs would increase by an average of $10,000 per patient, but if you look at the overall impact of survival relative to increase in cost, the cost per quality-adjusted life year [gained] was $46,858," Becker added.
"In the field of cost-effectiveness studies in developed countries, about $100,000 is the benchmark below which a new technology is found to be cost effective," he said. "So, the fact that ours was below $50,000 demonstrates that it is strongly cost effective."
Encouragingly, Becker said, the company also found that the actual survival increases measured in the observational study closely matched the 3.6 month increase projected by the model.
"It was interesting that two different methods — one modeled and one measured directly — came out so similar," he said. "That gives me confidence that the test [really] can be associated with in increased survival."
While Becker stressed that Pathwork considers the results of this first cost-effectiveness analysis to be robust, the company is also considering conducting prospective analysis in this regard.
"We are also looking at a lot of questions on what other ways we can demonstrate clinical utility of the test," he said.
Medicare began covering the TOO test in July 2011, and SelectHealth, a private health plan that is part of Utah's Intermountain Healthcare system, also pays for the test, according to Becker.
He said that Pathwork is "working very actively" to use this new study, as well as previous peer-reviewed publications, to help negotiate more formal coverage policies.
Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at mashford [at] genomeweb [.] com.