The cost-effectiveness of radiofrequency ablation for Barrett's esophagus with low-grade dysplasia: results from a randomized controlled trial (SURF trial)

BACKGROUND AND AIMS: The Surveillance versus Radiofrequency Ablation (SURF) trial randomized 136 patients with Barrett's esophagus (BE) containing low-grade dysplasia (LGD), to receive radiofrequency ablation (ablation, n = 68) or endoscopic surveillance (control, n = 68). Ablation reduced the risk of neoplastic progression to high-grade dysplasia and esophageal adenocarcinoma (EAC) by 25% over 3 years (1.5% for ablation vs 26.5% for control). We performed a cost-effectiveness analysis from a provider perspective alongside this trial.

METHODS: Patients were followed for 3 years to quantify their use of health care services, including therapeutic and surveillance endoscopies, treatment of adverse events, and medication. Costs for treatment of progression were analyzed separately. Incremental cost-effectiveness ratios (ICER) were calculated by dividing the difference in costs (excluding and including the downstream costs for treatment of progression) by the difference in prevented events of progression. Bootstrap analysis (1000 samples) was used to construct 95% confidence intervals (CIs).

RESULTS: Patients who underwent ablation generated mean costs of U.S.$13,503 during the trial versus $2236 for controls (difference $11,267; 95% CI, $9996-$12,378), with an ICER per prevented event of progression of $45,066. Including the costs for treatment of progression, ablation patients generated mean costs of $13,523 versus $4,930 for controls (difference $8593; 95% CI, $6881-$10,153) with an ICER of $34,373. Based on the various ICER estimates derived from the bootstrap analysis, one can be reasonably certain (>75%) that ablation is efficient at a willingness to pay of $51,664 per prevented event of progression or $40,915 including downstream costs of progression.

CONCLUSIONS: Ablation for patients with confirmed BE-LGD is more effective and more expensive than endoscopic surveillance in reducing the risk of progression to high-grade dysplasia/EAC. The increase in costs of ablation can be justified to avoid a serious event such as neoplastic progression. At a willingness to pay of $40,915 per prevented event of progression, one can be reasonably certain that ablation is efficient. (www.trialregister.nl number: NTR 1198.).