Poster Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2017

Utility values across lines of therapy in immuno-oncology treatments: an example from advanced melanoma (#220)

Dominic Tilden 1 , Will Sierakowski 1 , Suzi Cottrell 1 , Hansoo Kim 2
  1. Thema, Sydney
  2. BMS, Melbourne


Cost utility analyses of oncology treatments are most commonly performed using partitioned survival models, applying health state utilities to progression-free and progressive disease and relative to a specific line of therapy. The objective of this study was to assess utility values across treatments and lines of treatment using data in advanced melanoma for two immuno-oncology agents, nivolumab and ipilimumab.


Utility values from 1st line (1L) and 2nd line (2L) advanced melanoma populations treated with nivolumab and ipilimumab were extracted from three randomised controlled clinical trials: CheckMate-067 (1L nivolumab and 1L ipilimumab), CheckMate-037 (2L nivolumab) and MDX010-20 (2L ipilimumab). Visual assessment of QoL over time as well as comparisons of baseline and change from baseline values were performed using summary statistics.


Baseline values for 1L and 2L were similar for nivolumab (0.80 vs 0.75, p=0.001) and ipilimumab (0.79 vs 0.81, p=0.123). Across all lines of therapy nivolumab use resulted in improvements in utility whilst patients remained progression free. Ipilimumab treatment regimens showed initial declines in utilities in the first 3 months followed by improvements over the remainder of time on treatment.

The change in utility from baseline to 12 months was similar for 1L and 2L nivolumab (0.050 v 0.047, p=0.930). Both these changes were greater than that observed for 1L ipilimumab at 12 months (0.035, p=0.533 v 1L nivolumab and p=0.767 v 2L nivolumab). 2L ipilimumab results were limited to short term follow-up, however utility values were comparable to 1L ipilimumab at 5 months (-0.023 v 0.014, p=0.091).


The quality of life of patients on immune-based therapies appears to be independent of therapy line. Furthermore, economic modelling in an immune-oncology setting should reflect that quality of life looks to be a function of time on treatment.