Optimization of Helicobacter pylori-naïve gastric cancer screening using a mathematical simulation model

Abstract

Since Helicobacter pylori (H. pylori)-associated gastric cancer is expected to decrease in Japan, the current population-based gastric cancer screening may have decreased effectiveness and increased disadvantages for the examinees. Tracking outcomes such as cancer-specific mortality for each screening strategy is labor-intensive. Therefore, mathematical simulation models are often used to compare strategy effectiveness instead of real-world cohort data. In a simulation, evaluation of the effectiveness of gastric cancer screening must be multifaceted; however, the evaluation methods that can be implemented vary depending on the type of simulation model. Microsimulation models are excellent at tracking changes at the individual level, simulating the developmental progression of gastric cancer in each individual, and even identifying the characteristics of gastric cancer detected by applying screening strategies. In contrast, macroscopic methods, including Markov models, are superior in identifying the cost-effectiveness of a screening strategy in the entire population. The optimal screening strategy can be determined by combining these various simulation models to predict the effectiveness of strategies. Going forward, gastric cancer screening should be conducted to effectively identify H. pylori-naïve gastric cancer while targeting the population with a decreased prevalence of H. pylori-associated gastric cancer, and the current screening strategy should be modified based on the results of various simulations.