Cost-effectiveness analysis based on neurocognitive function of a randomized study with stereotactic radiosurgery (SRS) alone versus SRS plus whole-brain radiation therapy (WBRT) for patients with one to three brain metastases.

Abstract

e16528 Background: This paper illustrates the incremental cost-effectiveness ratio (ICER) of SRS alone versus SRS plus WBRT for brain metastases with neurocognitive scores as the primary outcome. Methods: A decision analysis model, including subsequent treatment for recurrences, was developed and populated with data from patients randomized to either SRS alone (SA) versus SRS plus WBRT (SW), using costs from the healthcare institution perspective and outcomes based the raw function score and standardized score of the Hopkins Verbal Learning Test- Revised (HVLT-R) Total Recall variable at the last study visit. Tornado diagrams and one-way sensitivity analyses were conducted to determine robustness of the model. Results: Fifty-eight patients were included in the final analysis, with 31 patients in the SA arm and 27 patients in the SW arm. The average cost of the SA arm is $119,000 versus $74,000 for the SW arm. The average HVLT-R raw score for the SA arm was 23.28 points versus 20.94 points for the SW arm. The ICER for total cost was $19,528 per unit of increase in raw score. The average standardized HVLT-R score was - 0.876 for the SA arm and -1.335 for the SW arm. The ICER for the total costs per unit of change in the standardized score was $99,691. Tornado diagrams revealed that the following variable had the highest impact on the ICER in both analyses: probability of being alive with SA in recursive partitioning analysis (RPA) class 2 and being treated for recurrence. Assuming the probability of being alive after treatment for recurrence is 1.0, a one-way sensitivity analysis demonstrated the ICER favoring SRS alone over SRS plus WBRT would $5,000 per unit raw score and $22,000 per unit of standardized score. Conclusions: Compared to other interventions in the $50,000 to $100,000 per QALY cost- effectiveness range, from an economic perspective the application of SA, with subsequent management of recurrences, is shown to be a reasonable treatment modality for brain metastases, in terms of cost per unit of change in neurocognitive function. No significant financial relationships to disclose.