Benefits and Harms of Gastric Cancer Screening and Prevention in Japan: A Microsimulation Modeling Analysis

Abstract

Background: In 2016, gastric cancer is the second leading cause of cancer mortality worldwide. Globally, Japan and Korea are the only two countries implementing national screening program for gastric cancer. South Korea launched a biennial national -endoscopic screening program in 1999, which demonstrated a 47% reduction in gastric cancer mortality compared with no screening. In 2015, endoscopic screening was recommended by the Japanese Guideline for Gastric Cancer Screening. Currently, Japan's screening policy recommends endoscopic screening every 2 to 3 years for individuals aged 50 or above. Aim: This study aims to quantify the potential impacts of Japan's current national endoscopic screening policy and other alternative endoscopic screening options, and to identify a set of efficient endoscopic screening strategies for the current population of Japan. Methods: A microsimulation model of intestinal-type noncardia gastric adenocarcinoma was developed to simulate a virtual population with similar risk profile and life expectancy to the actual population of Japan. The model was calibrated and validated to the nationally-representative data from vital statistics and population-based cancer registries. The baseline scenario was modeled to project the current trend of gastric cancer. Variations among age of initiation (40, 45, 50, and 55), age of termination (75, 80, and 85), and screening interval (2, 5, 10, and 15 years) resulted in 39 unique endoscopic screening strategies for evaluation. In addition, the following three existing screening scenarios were included for analysis: the current screening program in Japan (screening every two or three years from age 50 years, with no termination) and Korea (biennial screening from age 40 years, with no termination). In total, 42 unique endoscopic screening strategies were simulated. Full adherence for all screening scenarios were assumed. Results: In comparison with baseline scenario, endoscopic screening strategies provided 43 to 203 life-years gained per 1000 individuals. The number of lifetime endoscopies per 1000 individuals ranged from 2281 to 28,635. The complications caused by screening were from 6 to 77 per 1000 individuals. The lifetime number of false-positive results ranged from 381 to 4941 per 1000 individuals. Scenarios simulating the current national endoscopic screening program in Japan and Korea fell below the efficient frontier. Under full adherence assumption, the most advantageous endoscopic screening strategy among evaluated scenarios is screening from age 40 to 80 years every 5 year. Conclusion: In this microsimulation modeling study with full adherence assumption, the findings suggest that the current national endoscopic screening program for gastric cancer in Japan might not be the most efficient screening option under benefit-to-harm consideration. The finding underpins the need for further investigation to bring up an efficient and cost-effective endoscopic screening program in Japan.