[Skip to Navigation]
Comment & Response
May 8, 2019

The Importance of Outcome and Precise Evaluation in Economic Analysis of Cancer Drugs

Author Affiliations
  • 1Department of Medicine, James Graham Brown Cancer Center, University of Louisville, Louisville, Kentucky
  • 2Roswell Park Compreshensive Cancer Center, Buffalo, New York
  • 3Department of Surgical Oncology, University of Texas MD Anderson Cancer Center, Houston
JAMA Dermatol. 2019;155(7):862-863. doi:10.1001/jamadermatol.2019.0594

To the Editor The recent article by Almutairi et al1 concluded that the cost of combining talimogene laherparepvec (T-VEC) with ipilimumab therapy in patients with advanced melanoma to gain 1 progression-free life-year was $1.6 million.1 Their conclusion was based on an economic assessment of the results of a phase 2 exploratory study designed to test the hypothesis that ipilimumab would augment systemic antitumor responses triggered by T-VEC and yield an improved objective response rate.2,3 Unfortunately, this economic assessment was substantially flawed by multiple erroneous assumptions made by the authors and subsequently supported by an Editorial written by DiMagno and Emanuel.4 First, this phase 2 trial was a proof-of-principle study, which was not intended to be label enabling and was not powered to evaluate progression-free survival (PFS). As such, the authors’ use of PFS as the main clinical outcome to conduct their economic assessment invalidates all of their results and conclusions. Second, the economic assessment overestimates the number of T-VEC injections: the published median duration of treatment was 21.1 weeks for T-VEC in the phase 2 trial, whereas the authors’ economic modeling assumed T-VEC dosing occurred to the median PFS time of 35.7 weeks (the median number of T-VEC doses was only 10).2 Third, the economic assessment assumed that the maximum dose of T-VEC (4 mL) was given to each patient, which was in fact rarely the case. Based on these second and third points, the authors vastly overestimated both the treatment duration and dose of T-VEC, which in turn led to a significant overestimation of the true cost of the treatment. Fourth, the presumption that a 1-year improvement in PFS from immunotherapies only leads to 1 year of life ignores the durability of immune-mediated responses on overall survival (an outcome that was not and could not be assessed in this phase 2 trial).5 We strongly believe that cost-effectiveness evaluations in oncology should account for the impact of therapeutics on overall survival. To conclude, the economic evaluation conducted by Almutairi et al1 was flawed as a result of the inappropriate use of the PFS outcome and the overestimation of the expenses related to the study agent. We recommend that future economic assessments of cancer drugs by these authors and others rely on appropriately powered clinical trial outcomes and accurate calculations of drug dosing and associated expenses.