BACKGROUND:
Molecular testing of cancers is quickly becoming standard of care using diverse approaches, either academic or commercial in origin. Some oncologists remain apprehensive about the clinical utility of molecular profiling, based on the degree to which information can be used in a treatment decision, and whether it leads to selection of more expensive treatments that may not be
accessible.
OBJECTIVES:
The aim of this study is to examine the decision impact of a multiplatform tumor profiling service, Caris Molecular Intelligence (CMI), and evaluate CMI-guided treatment costs compared to prior and planned treatments in prospective and retrospective clinical studies.
METHODS:
In 5 physician-led clinical studies, the treatment decision prior to receipt of the CMI report was captured (n=137 patients). A systematic review of treatment data from 10 clinical studies of CMI (n=385 patients) allowed a comparison of planned versus actual (n=137) and prior versus actual (n=229) treatment costs. Costing information was taken from the British National Formulary (BNF) giving a treatment cost per cycle per pa,ent. Decision impact (n=232) and treatment cost per cycle (n=292) were also compared from studies of next generation sequencing (NGS)-only approaches.
RESULTS: Decision impact was changed in 88% of CMI-profiled cases compared to 29% of NGS-only approaches. The CMI-guided treatment cost per cycle was £995 in 385 treated patients. Planned treatment costs were comparable to actual treatment costs (£979 versus £945; p=0.7123) and prior treatment costs were also not significantly different to profiling-guided treatments (£892 versus £850; p=0.6319). NGS-only guided treatments cost £2,795 per cycle per patient.
CONCLUSIONS:
Treatment costs guided by a multiplatform-profiling platform were comparable to planned and prior treatment and do not cause a cost explosion, as the majority of treatments used were conventional chemotherapies. NGS-only approaches rely on more expensive targeted therapies and higher treatment cost per cycle per patient.