Submucosal Invasive Depth Predicts Lymph Node Metastasis and Poor Prognosis in Submucosal Invasive Esophageal Squamous Cell Carcinoma

Tomohiro Kadota, MD, PhD; Tomonori Yano, MD; Takeo Fujita, MD, PhD; Hiroyuki Daiko, MD, PhD; Satoshi Fujii, MD, PhD


Am J Clin Pathol. 2017;148(5):416-426. 

In This Article

Abstract and Introduction


Objectives. Lymph node metastasis (LNM) in submucosal invasive esophageal squamous cell carcinoma (SM-ESCC) is a prognostic factor. The aim of this study was to identify a histopathologic predictor of LNM in SM-ESCC.

Methods. In total, 108 patients who underwent an esophagectomy and lymph node dissection without preoperative therapy and who were pathologically diagnosed with SM-ESCC were enrolled in this study. Relationships between several clinicopathologic factors and LNM were examined.

Results. A multivariate analysis revealed that a tumor size of 35 mm or more (P = .0025), submucosal invasive depth (SID) of 2,000 μm or more (P = .013), and lymphatic infiltration (P < .0001) were significant independent predictors of LNM. In addition, there were significant differences in recurrence-free survival curves between patients with SID less than 2,000 μm or not (P = .029) and tumor size less than 35 mm or not (P = .049).

Conclusions. This study suggests that SID may predict not only LNM but also poor prognosis.


Esophageal squamous cell carcinoma (ESCC) prognosis is known to be dismal among gastrointestinal cancers.[1,2] This is partly because ESCC with lymph node metastasis (LNM) frequently recurs, even after curative resection.[3] LNM is reportedly observed in about 20% to 50% of superficial ESCCs, which is characterized by submucosal invasion.[1,4–9] Recently, the incidental detection of superficial ESCC has increased as advances in diagnostic technology using narrow-band imaging have been implemented.

LNM is known to be a significant prognostic factor for all disease stages.[10–12] Survival rates of patients with pT1 to pT2 tumors without LNM who were confirmed using pathologic examination after surgical resection accompanied with three-field lymph node dissection were significantly longer than those of patients with pT1 to pT2 tumors with LNM.[13]

Previous studies have reported that several clinicopathologic factors, including macroscopic type (0-I, superficial and protruding type; 0-III, superficial and excavated type),[7] tumor size,[7,14] depth of tumor invasion (pSM2 or pSM3),[6,8,15,16] tumor differentiation, infiltrative growth pattern (INF),[17] and lymphovascular infiltration,[1,5,7,8,14–21] are sensitive and specific predictors of LNM. However, even a combination of these risk factors for LNM described above has not enabled the prediction of LNM with complete accuracy.

Some studies have reported that the submucosal layer subclassification of tumor invasive depth into three groups (SM1-SM3) can be used as a significant predictor of LNM;[6,8,15,16] however, the SM1 to SM3 subclassification is not consistent with the actual submucosal invasive depth (SID) among cases and does not reflect the absolute submucosal invasive distance. Moreover, the relationship between LNM and SID that was measured as a vertical distance from the muscularis mucosae (MM) line to the deepest portion of the invasive carcinoma cell has not been previously investigated.

The aim of the present study was to explore sensitive and specific risk factors for the accurate prediction of LNM in patients with ESCC with submucosal invasion and examine whether these risk factors had an impact on prognosis. The understanding of risk factors for LNM paves the way for selecting those patients with submucosal invasive ESCC who would need additional therapy, which may lead to improving patient prognosis in the near future.