External validation of the FAMISH predicting score for early gastric cancer with endoscopic submucosal dissection.

Abstract

To externally validate the robustness of the FAMISH predicting score designed to estimate the risk of metachronous gastric lesions (MGLs) after endoscopic submucosal dissection (ESD) for early gastric cancer (EGC). This multicenter, retrospective study included 618 patients with EGC who underwent curative ESD at two tertiary referral teaching hospitals between January 2014 and December 2019. FAMISH score was a composite indicator of age, sex, family history, corpus intestinal metaplasia, synchronous lesions, and H. pylori infection. Discrimination, calibration, and risk stratification of these scores were assessed. Associations between MGL characteristics and FAMISH scores were also explored. After a median follow-up period of 60 months, 83 of 618 patients (13.4%) developed MGL. The discrimination ability according to the area under the curve was 0.708 (95% CI, 0.645-0.772) for predicting the 5-year MGL. The calibration results showed good consistency between the predicted and actual MGL (Hosmer-Lemeshow, P > 0.05). In terms of risk stratification, the 5-year MGL rates were 4.1% (95% CI, 1.6%-6.5%), 10.8% (95% CI, 7.2%-14.3%), and 32.1% (95% CI, 20.9%-41.7%) in the low-, intermediate-, and high-risk groups, respectively (P < 0.001). For patients with MGL, the curative resection rate of ESD was significantly higher in the low- and intermediate-risk groups than in the high-risk group (100% vs. 80%, P = 0.037). The FAMISH predicting score was externally validated and can be generalized to an independent patient population. This adjuvant tool can assist in individual clinical decision-making.