Purpose: Patients with unresectable wild-type KRAS metastatic colorectal cancer benefit from fluoropyrimidines (FP), oxaliplatin (O), irinotecan (I), bevacizumab (Bev), and epithelial growth factor receptor inhibitors (EGFRI). The most cost-effective regimen remains unclear.
Methods: A Markov model was constructed to examine the costs and outcomes of three treatment strategies: strategy A (reference strategy): EGFRI monotherapy in third line ([3L]; ie, first-line [1L]: Bev + FOLFIRI [FP + I] or FOLFOX [FP + O]; second line [2L]: FOLFIRI/FOLFOX; 3L: EGFRI); strategy B: EGFRI and I in 3L (ie, 1L: Bev + FOLFIRI/FOLFOX; 2L: FOLFIRI/FOLFOX; 3L: EGFRI + I); and strategy C: EGFRI in 1L (ie, 1L: EGFRI + FOLFIRI/FOLFOX; 2L: Bev + FOLFIRI/FOLFOX; 3L: best supportive care). Efficacy data of the treatments were obtained from the literature. Health system resource use information was derived from chart review and the literature. Using Euro-QOL 5 Dimensions, utilities were obtained by surveying medical oncologists and costs from the Ontario Ministry of Health and the literature. The perspective of the Canadian public health care system was used over a 5-year time horizon with a 5% discount in 2012 Canadian dollars.
Results: All three strategies had similar efficacy, but strategy C was most expensive. The incremental cost-effectiveness ratios (ICERs) for strategies B and C compared with A were 119,623 and 3,176,591, respectively. The model was primarily driven by the acquisition cost of the drugs. Strategy B was most cost effective when the willingness-to-pay threshold was > $130,000 per quality-adjusted life-year. Sensitivity analysis showed that strategy C would be cost-effective only when the progression-free survival of EGFRI is better than Bev in 1L with hazard ratio < 0.23 at willingness-to-pay of $150,000 per quality-adjusted life-year.
Conclusion: First-line use of EGFRI in metastatic colorectal cancer is not cost effective at its current pricing relative to Bev.
Copyright © 2016 by American Society of Clinical Oncology.