Clinicopathologic features of superficial esophageal squamous cell carcinoma in different infiltrative growth pattern

Superficial esophageal squamous cell carcinoma (SESCC) is defined as neoplastic lesions limited to the mucosa or submucosa regardless of the nodal status. The infiltrative growth pattern (INF) has been implicated in tumor aggressiveness and prognosis in various cancers, but the application research of INF in SESCC is still unclear. We aimed to investigate the association between INF types and clinicopathological features in SESCC.

We retrospectively analyzed 368 SESCC patients who underwent endoscopic submucosal dissection and precisely defined INF classification from 2014 to 2023. INF was classified as INFa/b/c per Japanese Esophageal Society guidelines. Clinicopathological characteristics were compared across INF types using univariate analysis. Significant variables from univariate analysis were included in multivariate logistic regression to identify independent predictors of INF types.

Univariate analysis revealed that the INF of tumor was associated with tumor size, gross morphology, intraepithelial papillary capillary loop (IPCL), infiltration depth, lymphovascular invasion, and vertical positive margins (All P < 0.05). Multivariate logistic regression demonstrated that tumor size (β=0.28, P=0.004; OR=1.32, 95%CI:1.09-1.59), IPCL (β=0.81, P=0.004; OR=2.24, 95%CI:1.30-3.85), and infiltration depth (β=0.81, P=0.017; OR=2.24, 95%CI:1.15-4.35) were significantly correlated with INFb, while lymphovascular invasion (β=8.77, P=0.007; OR=6456.93, 95%CI:10.96-3803785.49) as an independent risk factor for INFc.

Increased tumor size, presence of IPCL type B2, and depressed gross morphology were more indicative of INFc-type SESCC. Compared with INFa and INFb, INFc type SESCC has deeper infiltration depth and is more likely to have lymphovascular invasion and positive postoperative resection margins. Therefore, careful endoscopic visualization of tumor size, IPCL, and gross morphology can improve the prediction of INF and tumor status, facilitating informed preoperative selection of surgical approach and subsequent postoperative treatments.