OBJECTIVES:

To conduct an SLR of health economic evidence (cost-effectiveness models, cost and resource use, utility inputs) in order to inform an economic model of approved anaplastic lymphoma kinase (ALK) tyrosine kinase inhibitors (TKIs) for the treatment of ALK+ non-small-cell lung cancer (NSCLC).

METHODS:

Systematic searches were conducted of MEDLINE, EMBASE, EconLit, Database of Abstracts and Review of Effects and Cochrane Library databases, alongside manual web searches. Using pre-defined eligibility criteria, abstract and full-text screening for inclusion was conducted by two independent reviewers. Quality appraisal checklists from the Methods for the Development of NICE Public Health Guidance (third addition) and Picot et al. (2015) were used for quality assessment of studies. Searches were designed to identify inputs to inform an economic model for health technology appraisal.

RESULTS:

A total of 5,468 abstracts were retrieved. After full-text screening, 44 economic evaluations (all directly related to ALK+ NSCLC), 148 cost and resource studies (20 ALK+ specific), and 120 utility studies (16 ALK+ NSCLC specific) were included for data extraction. Fourteen of the 44 economic articles used partitioned-survival models, which was the expected approach for the economic model. The incremental cost-effectiveness ratio expressed per quality-adjusted life-year gained was the most reported outcome from the economic evidence. Utility values were measured using both the EQ-5D-3L and EQ-5D-5L. Treatment specific, heath state specific and time-to-death utilities were all identified. From cost studies a range of costs were identified including drug acquisition, administration, adverse event, terminal care, total cycle, total treatment, and indirect costs. Four studies reported ALK+ NSCLC specific resource use.

CONCLUSIONS:

The SLR confirmed that the partitioned-survival model is the most widely used, with the cost/QALY as the most reported outcome. Utilities were available by treatment, health state and time-to-death. Identified data gaps were resource use in ALK+ NSCLC and change in utility after progression.