Abstract
264 Background: There is great interest in delivering personalized phase I clinical trials for patients. The value of this strategy will be determined by the increased benefit of the testing against the testing’s associated increased costs. We developed and employed a rapid method for determining this value in a cohort of metastatic melanoma patients who participated in phase I clinical trials at M.D. Anderson Cancer Center. Methods: The clinical parameter of benefit used was time to treatment failure (TTF) and the charges were those accrued by patients during their participation on the trial. Charge data were extracted from the MD Anderson business database. Clinical and charge data were employed in a simulation to compare two strategies: testing all patients' tumor for genetic aberrations or not. Results: In the subcohort of patients (N = 103) that were reviewed for this analysis, patients who underwent molecular testing experienced an increase in TTF of 2 months (4.2 versus 2, p=0.1) compared to those not tested, but this was not the primary concern of this study. Rather, the charges and clinical data allowed for a calculation of the value of the testing. The incremental cost-effectiveness ratio (ICER) of genetic sequencing was $49,103 per failure-free month. Ninety-four percent of the this incremental cost was due to the cost of staying on the phase I trial, with only 6% due to the cost of the genetic testing. If the cost of the genetic test was zero, the incremental ICER would still over $46,000 per failure-free month. Probabilistic sensitivity analysis demonstrated that the untested strategy was predicted to be the most cost-effective strategy with a willingness-to-pay (WTP) under $52,000 per failure-free month. The sequencing of tumors becomes 85% likely to be the most acceptable strategy by a WTP threshold of $360,000 per failure free month. Conclusions: These data suggest that making molecular testing cost effective requires reducing the costs of clinical research. This methodology is a way to make a determination of value, as well as attributable costs, which we believe will have to be made for each individual clinical setting.