Cost-Effectiveness of Serum Pepsinogen as a Gastric Cancer Targeted Screening Strategy in the United States

Abstract

Background and aims: Current gastric cancer (GC) screening modalities are invasive and expensive. Noninvasive screening for GC precursors with serum pepsinogen (PG) may improve early detection and prevention. Test characteristics of PG based on US prospective data was recently reported and used to study the cost-effectiveness of PG screening vs no screening in the US.

Methods: A patient-level state transition microsimulation of gastric adenocarcinoma analyzed noninvasive screening vs no screening in a hypothetical cohort of average risk US individuals. Primary outcomes included life expectancy, quality-adjusted life years, total costs, and incremental cost-effectiveness ratios. Secondary outcomes included total GC incidence and mortality. Base-case PG sensitivity and specificity were 34.1% and 94.7%, respectively, with a wide range of PG performance characteristics also examined.

Results: One-time serum PG screening at age 40 was cost-effective compared to no screening with an incremental cost-effectiveness ratio of $4913.29 per quality-adjusted life year. PG screening resulted in 10.9% relative reduction in lifetime GC incidence and 10.8% relative decrease in cumulative GC mortality. Localized stage at diagnosis increased from 30.5% to 33.6% and metastatic stage decreased from 40.8% to 37.4%. Sensitivity analysis showed PG screening was most sensitive to endoscopy costs, chronic atrophic gastritis quality of life, and PG prevalence. PG screening remained cost-effective across a wide range of test values.

Conclusion: PG screening is a cost-effective strategy to improve GC mortality; however, mortality benefit will depend on the test characteristics of the biomarker. Future blood-based screening tests that have better performance characteristics could further improve GC prevention.

Keywords: Cancer Prevention; Cancer Screening; Cost-Effectiveness Analysis; Gastric Cancer; Serum Pepsinogen.

Figure 1

A. Model schematic for No Screening Strategy. This figure depicts the natural history progression of to the Correa cascade, from normal mucosa to gastric cancer. Patients can move from any health state to death, either due to all-cause mortality or to gastric cancer related mortality. B. Model schematic for Pepsinogen Screening Strategy. In this model, patients are offered pepsinogen screening age 40 to detect those at high risk of gastric cancer precursor lesions and offer surveillance for earlier detection of gastric cancer. Patients can move from any health state to death.

Figure 2

The one-way sensitivity analysis revealed that serum PG screening is most sensitive to the cost of EGD, the health state utility of CAG, and PG positive prevalence. All variables were tested but only the most sensitive variables were displayed.

Figure 3

Willingness-to-pay (WTP) vs percent iterations cost-effective.