Cost-Effectiveness Analysis of Stereotactic Ablative Radiation Therapy Versus Single-Fraction External Beam Radiation in Painful Vertebral Body Metastases

Abstract

Purpose/Objective(s): Stereotactic radiosurgery (SRS) with or without whole brain radiation therapy (WBRT) is currently a widely accepted method of treatment for brain metastases. Studies on WBRT in combination with SRS have shown conflicting results. To aid in evidence synthesis and interpretation, we created a decision model to estimate the incremental cost-effectiveness of adding WBRT to SRS. Materials/Methods: We created a decision tree informed by three randomized trials of patients with brain metastases. WBRT + SRS vs SRS alone was the decision modeled (i.e., the decision node), followed by potentially occurring events (i.e., chance nodes) that included outcomes such as cognitive decline, recurrence, and salvage. The model had a one year time horizon. We produced summary estimates weighting the result from each trial by the inverse of its variance, in accord with standard methods. Costs used were from the healthcare perspective and were determined from Medicare 2012 reimbursement rates. Utilities and utility decrements (preference-weighted measures of health-related quality of life scaled from 0 to 1) were derived from published EQ-5D scores for chronic conditions. One-way sensitivity analyses were conducted to determine robustness of the decision analysis model. Results: Compared to SRS alone, WBRT + SRS had a higher average cost ($28,050 vs $29,550) and a lower average health-related quality of life (utility 0.50 vs 0.41), yielding an incremental cost-effectiveness ratio of $15,970 per quality-adjusted life year. Therefore, SRS alone dominated the WBRT + SRS strategy because it was both cheaper and delivered a higher quality of life. Sensitivity analyses revealed that altering assumptions regarding the following inputs could change the preferred decision: utility decrement after WBRT and utility decrement after recurrence. As the utility decrement from treatment with WBRT decreased (below 0.016), WBRT + SRS became the dominant strategy. Similarly, as the utility decrement from recurrence increased (above 0.440), WBRT + SRS became the dominant strategy. Conclusions: Our preliminary results demonstrate that SRS alone may be a more cost-effective method of treatment for brain metastases, likely due to smaller utility decrements. This should be further explored with multiway sensitivity analyses, threshold analyses, and larger randomized trials. Recommendations to patients should be on a case by case basis, taking into account patient preferences and ability to follow-up after treatment. For patients who decline WBRT, SRS alone may be a reasonable treatment modality for brain metastases. Author Disclosure: C. Min: None. H.T. Gold: None. A. Narayana: None. S.C. Formenti: None.