COST-EFFECTIVENESS OF ENDOSCOPIC SURVEILLANCE OF PRECANCEROUS GASTRIC LESIONS IN THE UNITED STATES

Purpose: Endoscopic surveillance of patients with precancerous gastric lesions has the potential to improve survival by early detection and treatment. While there are evidence-based guidelines that recommend endoscopic surveillance for secondary prevention of cancer with Barrett’s esophagus, no analogous guidelines exist for patients with precursor gastric lesions. Our objective was to estimate the effectiveness and cost-effectiveness of endoscopic surveillance of precancerous lesions.

Methods: We developed a state-transition model simulating the natural history of intestinal type gastric cancer and calibrated it to age-specific prevalence of precancerous lesions and cancer in the US. For a cohort of 50-year old men with a diagnosis of dysplasia or intestinal metaplasia, we estimated life expectancy (LE), lifetime gastric cancer risk and costs associated with no surveillance and with endoscopic surveillance every 1 to 5 years. Incremental cost-effectiveness ratios (ICER) were expressed as 2007 US$ per year of life saved (YLS). Upon detection, dysplasia or asymptomatic gastric cancer were treated with endoscopic mucosal resection and for incomplete resections, surgery. Based on clinical studies, we conservatively estimated that treatment reduced the risk of subsequent gastric cancer by 50%. Assumptions related to biopsy sensitivity, disease progression, treatment efficacy and costs were explored using sensitivity analysis.

Results: In the absence of endoscopic surveillance, the lifetime gastric cancer risk following dysplasia or intestinal metaplasia was 12.0% and 0.6%, respectively. For dysplasia, the discounted per-person lifetime cost was $3480 and LE was 17.91 years with no surveillance; endoscopic surveillance every 5 years reduced lifetime cancer risk by 10.1% with an ICER of $24,600/YLS. The most effective strategy under $50,000/YLS was surveillance every 3 years. Surveillance on an annual basis exceeded $100,000/YLS. Results were most sensitive to the effectiveness of treatment in preventing the progression to cancer; for example, if treatment reduced gastric cancer risk by 80%, the ICER were reduced by nearly 5-fold. In contrast to patients with dysplasia, endoscopic surveillance every 5 years for intestinal metaplasia patients had an ICER of >$1,400,000/LYS compared to no surveillance.

Conclusions: Routine endoscopic surveillance every 3 to 5 years for patients with gastric dysplasia is promising for secondary prevention of invasive cancer, and has a cost-effectiveness ratio that would be considered attractive in the U.S.  Endoscopic surveillance of less advanced metaplastic lesions does not appear to be cost-effective.