[Skip to Navigation]
Sign In
Comment & Response
May 16, 2019

Changes to Model Assumptions of the Cost-effectiveness of Durvalumab Therapy for Non-Small Cell Lung Cancer—In Reply

Author Affiliations
  • 1Institute for Technology Assessment, Massachusetts General Hospital, Boston
  • 2Massachusetts General Hospital Cancer Center, Boston
  • 3Harvard Medical School, Boston, Massachusetts
JAMA Oncol. 2019;5(7):1066-1067. doi:10.1001/jamaoncol.2019.1100

In Reply We have carefully considered the suggestions proposed by Giri and Huntington regarding modest changes to our model’s assumptions. Their first comment suggests that patients treated with durvalumab consolidation therapy should receive not only chemotherapy after progression to metastatic disease1 but also PD-1 (programmed cell death 1)/PD-L1 (programmed cell death 1 ligand 1) checkpoint inhibitor monotherapy. This suggestion is based on updated trial2 results published after the acceptance of our modeling article that noted 19.5% of patients who progressed while or after receiving durvalumab received some form of subsequent immunotherapy. We believe that this information would not improve our model’s reflection of current clinical practice. Whether patients who have received a PD-1/PD-L1 checkpoint inhibitor in the adjuvant setting would derive benefit from subsequent treatment with another single-agent PD-1/PD-L1 checkpoint inhibitor is still unknown. This treatment strategy is not currently recommended as a standard of care for most patients because data regarding its efficacy are lacking.3 Additionally, patients who subsequently received immunotherapy in the updated analysis were primarily treated with nivolumab monotherapy,2 which is clearly not an agreed on standard course for first-line treatment of metastatic disease.3 For these reasons, we would argue that using the subsequent therapy data from the trial2 would disadvantage our model’s ability to portray current clinical practice. We believe that the assumption to treat postprogression with chemotherapy in patients who received durvalumab provides the most accurate prediction for first-line treatment in this setting at present. However, for good measure, we tested our model under Giri and Huntington’s assumption, and our conclusion that durvalumab consolidation therapy would be cost-effective held, with an incremental cost-effectiveness ratio (ICER) of $79 609 per quality-adjusted life year.

Add or change institution