DURHAM, NC—Increased personalization of cancer therapy has transformed oncology over the past decade as a growing number of targeted therapies have entered the market. To benefit from these new options, patients must have the specific tumor mutations the drugs address. The need to match patients to particular therapies based on characteristics not just of the patient, but of the tumor, has driven wider adoption of tumor profiling.
Two common tumor profiling tools are targeted gene panel testing (TGPT), which looks for a few common mutations in specific genes, and multigene panel sequencing (MGPS), which evaluates samples for the presence of a much broader range of variants across a comprehensive set of genes.
In a number of cancers, targeted therapies have resulted in dramatically improved outcomes for eligible patients. Does tumor genomic profiling make sense for all patients? At what cost?
Researchers at the Durham VAMC and Duke University developed a decision analytic model to evaluate the benefit in a hypothetical cohort of veterans with metastatic lung adenocarcinoma considering treatment to address those questions. Their results were presented during the 2020 American Society of Clinical Oncologists Annual Meeting held May 29-31.
In short, the researchers found that for this population of cancer patients, tumor genomic profiling was not cost effective compared to not conducting tumor profiling. Profiling did enable optimization of anticancer therapy and improved quality-adjusted survival, however.
For the analysis, the team included OncoKB genes with levels of evidence 1 and 2 for guiding therapy. The three profiling strategies examined were targeted testing for ALK, EGFR, ROS1 variants; multigene sequencing for ALK, BRAF, EGFR, HER2, MET, NTRK1, NTRK2, NTRK3, RET, ROS1; and no profiling.
They assumed that 95% of patients who had actionable mutations would receive targeted therapies. The non-targeted therapy options included chemotherapy and immunotherapy, together or separately, as well as no anticancer therapy.
The model used progression-free survival estimates from randomized trials, drug cost information from the VA and Medicare, data on non-drug cancer-related management costs from other published research, and mutation prevalence data from the VHA National Precision Oncology Program and the cBIOPortal for Cancer Genomics databases. The team discounted both costs and quality-adjusted life years (QALY) by 3% per year.
Analyses included base-case scenario, one-way sensitivity, and probabilistic sensitivity using 1,000 Monte Carlo simulations.
The analyses found that the cost per quality-adjusted life year gained was $309,399 (95% CI: $280,371-$343,161) for targeted gene panel testing and $324,707 (95% CI: $296,086 -$359,778) for multigene panel sequencing compared to not profiling the tumors. Multigene panel sequencing produced the most quality-adjusted life years.
Key factors in the analyses were quality of life, costs of the targeted therapy alectinib, and non-drug cancer costs in patients who received targeted therapies. In every one of the 1,000 simulations run, the cost-effectiveness ratios for both profiling approaches exceeded the $150,000 per QALY threshold.
- Poonen P, Dong O, Winski D, Reed SD, Vashistha V, Bates JS, Kelley MJ, Voora D. “Cost-effectiveness of genomic profiling in veterans with metastatic lung adenocarcinoma.” Abstract 7075. Poster 347. 2020 ASCO Annual Meeting. May 29-31, 2020.