ASSESSING THE EFFICACY OF ENDOSCOPIC GASTRIC CANCER SCREENING BY RACE AND SEX IN THE UNITED STATES

Background:
Gastric cancer (GC) incidence is declining and too low in the US to recommend general population screening. However, the risk of developing GC in the US is substantially higher in ethnic minorities, potentially warranting screening in specific population subgroups. While H.pylori test-and-treat strategies at young ages demonstrate efficacy in preventing non-cardia intestinal GC (NI-GC), the effectiveness diminishes for older age cohorts in whom precancerous lesions may have already developed. For such cohorts, endoscopic screening may serve as a viable alternative, yet its usefulness depends on the benefits vs. harms. This study aims to assess the effectiveness of endoscopic NI-GC screening and whether it differs by sex and race in the US by simulation modeling.

Methods:
Utilizing the established Microsimulation Analysis Screening (MISCAN) framework, we developed a novel sex- and race-specific NI-GC model. The model is calibrated to SEER incidence data and incorporates evidence from a systematic review on GC precursors, and H.pylori prevalence data derived from NHANES and literature. One-time endoscopic screening scenarios were evaluated at ages 50, 55, 60 and 65, assuming a sensitivity of 89% for intestinal metaplasia and 71% for dysplasia and early-stage NI-GC, with management per AGA guidelines: treatment of H.pylori; surveillance of extensive intestinal metaplasia; and, removal of dysplasia. The reduction in cumulative incidence and mortality and the Number Needed to Screen (NNS) to prevent one NI-GC case/death were assessed per race (black/white) and sex (female/male).

Results:
One-time endoscopic screening with subsequent surveillance demonstrated optimal efficacy when initiated at age 50. Screening reduced incidence by 50% in all subgroups (Figure 1), yet the corresponding NNS to prevent one NI-GC case varied substantially by sex & race: 468 vs. 1327 for black & white males; 629 vs. 2332 for black & white females. Similar to incidence, the reduction in mortality was most substantial in black males with substantial variations in the NNS by subgroup (Figure 2). The NNS for the overall US population was 1711 to prevent one NI-GC case and 3506 to prevent a NI-GC death.

Conclusion:
Our modeling shows that endoscopic screening has the potential to reduce NI-GC incidence. Within black men, the NNS metrics of gastric screening are comparable to those of established screening practices such as colorectal cancer screening [PMID: 27133893]. The benefits of screening may also outweigh the harms for other ethnic groups with similar risk. The projected incidence reduction relies on successful dysplasia removal, stressing the importance of endoscopic detection, resection and submucosal dissection skills. Further clinical studies corroborating the reduction in NI-GC incidence from endoscopy screening are crucial for conclusive recommendations.