OBJECTIVES : Conventionally, cancer classification was based primarily on tumor type and histology but recently, the focus is shifting towards specific genomic alterations. Entrectinib targets tumors driven by ROS1-fusion proteins in NSCLC and it is a new advanced option for the untested share of ROS+ NSCLC patients that would otherwise be treated with pemetrexed-based chemotherapy.

METHODS : Main clinical drivers are PFS and OS curves, estimated by parametric extrapolation of observed data from the pooled entrectinib trials; given the lack of head-to-head studies, a Matched Adjusted Indirect Comparison has been carried out to estimate hazard ratios vs. pemetrexed-based chemotherapy. The model assumes that after failure, 20% of patients treated with chemotherapy would be tested for mutation and start TKI. Utilities for PFS and OS states are based on STARTRK-2 EQ5D-3L data, valued with Italian tariffs. Economic inputs concern drug cost, administration (if injectable), adverse events management, and supportive care; unit costs are based on national literature and current tariffs, from NHS perspective.

RESULTS : Lifetime model results estimated that entrectinib, when compared to chemotherapy, induces a survival gain of 3.8 years, of which 30% in progression-free phase, corresponding to a total of 2.6 QALYs. The associated cost increase of almost 215K €, due to life expectancy improvement, doubling of treatment duration and higher acquisition costs, leads to an ICER of 57 K€/LY and to an ICUR of 82 K€/QALY.

CONCLUSIONS : State-of-the-art indirect comparison methods have been applied to currently available clinical data, indicating an expected relevant clinical benefit and acceptable ICER and ICUR associated to entrectinib use, especially considering the approximation of ROS1-NSCLC to a rare disease. These results should be interpreted cautiously, given the intrinsic uncertainty of indirect comparative effectiveness estimates, and will need to be confirmed by more robust follow up data, and possibly by prospective Real World Evidence.