Prognostic Value of Pathological Grade in pT1a Esophageal Cancer after Endoscopic Submucosal Dissection: A Retrospective Study (EC-GX-04)

Abstract

Purpose: The purpose of this study was to evaluate the association between pathological grade and survival outcomes in patients with pT1a esophageal cancer treated with endoscopic submucosal dissection.

Methods: We identified patients with pT1aN0M0 esophageal cancer diagnosed between 2010 and 2021 from the Surveillance, Epidemiology, and End Results (SEER) database. Patients were categorized by pathological grade. Overall survival (OS) and cancer-specific survival (CSS) were compared with pairwise comparisons.

Results: A total of 1,085 patients were included, 114 (10.5%) were classified as G1, 224 (20.6%) as G2, 53 (4.9%) as G3, and 694 (64.0%) as Gx. In the adenocarcinoma cohort, 5-year CSS rates were 93.7%, 91.7%, 92.6%, and 93.8% for G1, G2, G3, and Gx, respectively; corresponding 5-year OS rates were 73.4%, 69.8%, 77.9%, and 74.9%. In the squamous cell carcinoma cohort, 5-year CSS rates were 100.0%, 89.1%, 85.7%, and 86.4%; 5-year OS rates were 100.0%, 73.3%, 57.1%, and 35.1%. No significant differences in OS or CSS were observed across grade groups for either histologic subtype on univariable or multivariable analyses. Pairwise comparisons revealed no statistically significant differences among grade groups.

Conclusion: Pathological grade is not associated with survival outcomes in patients with pT1a esophageal cancer treated with endoscopic submucosal dissection, regardless of histologic subtype.

Graphical abstract

Keywords

esophageal cancer / tumor grade / pT1a / endoscopic submucosal dissection / survival

Introduction

Esophageal cancer ranks as the seventh most common malignancy and the sixth leading cause of cancer-related mortality worldwide, with an estimated 544,000 deaths in 2020[1]. Advances in screening and early detection have led to a growing proportion of patients diagnosed at early disease stages[2]. For patients with T1a disease, endoscopic submucosal dissection has emerged as the standard treatment[3], offering superior en bloc resection rates, precise histopathological assessment, and lower morbidity compared with esophagectomy[4]. Five-year overall survival (OS) rates following endoscopic submucosal dissection exceed 90%[5].

Despite these favorable outcomes, a subset of patients experience recurrence, underscoring the need for refined risk stratification. Pathological grade represents a potential prognostic factor in this context. The 8th edition staging system incorporates grade into the pT1a disease, stratifying adenocarcinoma into three subgroups (IA, IB, and IC) and squamous cell carcinoma into two subgroups (IA and IB)[6–8]. However, emerging evidence challenges these stratifications, with several studies reporting comparable survival outcomes across grade-defined subgroups, suggesting that the incremental prognostic value of grade in pT1a disease may be limited[9–11].

Given this uncertainty, the present study aims to evaluate the association between pathological grade and survival outcomes in patients with pT1a esophageal cancer treated with endoscopic submucosal dissection using large-scale, population-based data. By addressing this knowledge gap, we seek to inform more precise risk stratification and post-treatment surveillance strategies for this patient cohort.

Materials and methods

Patient cohort and data source

This retrospective cohort study analyzed de-identified data from the Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database: Incidence-SEER Research Data, 22 Registries, Nov 2023 Sub (2010–2021), National Cancer Institute, Surveillance Research Program, released April 2024, based on the November 2023 submission‌. Patients with histopathologically confirmed primary esophageal cancer were identified using International Classification of Diseases for Oncology, Third Edition codes. Inclusion criteria were: (1) diagnosis of esophageal adenocarcinoma (codes: 8140–8389) or esophageal squamous cell carcinoma (codes: 8050–8089); (2) treatment with endoscopic submucosal dissection (SEER Site-Specific Surgery Codes 10–14, 20–27); and (3) pathological stage pT1a disease. Exclusion criteria were: (1) receipt of chemotherapy or radiotherapy; and (2) treatment with esophagectomy. The patient selection algorithm was consistent with our previous methodology[12–14].

Clinicopathological variables extracted included age at diagnosis, sex, race, primary tumor site, pathological grade, and histologic subtype. Pathological grade was categorized according to the classification: G1 (well-differentiated), G2 (moderately-differentiated), G3 (poorly-differentiated/undifferentiated), and Gx (grade unknown).

Outcomes

The primary endpoint was OS, defined as the interval from diagnosis to death from any cause. The secondary endpoint was cancer-specific survival (CSS), defined as the interval from diagnosis to death attributable to esophageal cancer. Patients alive at the last follow-up were censored.

Statistical analysis

Continuous variables, which deviated from a normal distribution, were summarized as medians with interquartile ranges. Categorical variables were compared using the Pearson’s chi-square test, with Fisher’s exact test applied when expected frequencies were ≤ 5.

Survival probabilities were estimated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. Given the exploratory nature of the subgroup analyses, a two-sided P < 0.05 was considered statistically significant without adjustment for multiple comparisons, except where specified. For the primary analysis of overall and cancer-specific survival across grade groups, a global log-rank test was performed. When significant, post-hoc pairwise comparisons were conducted using a Bonferroni correction to control for type I error.

All statistical analyses were performed using SPSS version 26.0 (IBM Corp, Armonk, NY, USA) and R software (version 4.4.0, R Foundation for Statistical Computing, Vienna, Austria).

Results

Patient characteristics

Of 1,085 patients included, 114 (10.5%) were classified as G1, 224 (20.6%) as G2, 53 (4.9%) as G3, and 694 (64.0%) as Gx (Figure 1). Baseline characteristics of adenocarcinoma patients and squamous cell carcinoma patients are detailed in Tables 1 and 2. Median follow-up was 41 months (interquartile range, 19–75 months).

Survival outcomes in esophageal adenocarcinoma cohort

The 5-year CSS rates were 93.7% (G1), 91.7% (G2), 92.6% (G3), and 93.8% (Gx) (Figure 2A). Compared with G1, no significant differences were observed in CSS for G2 (hazard ratio [HR] = 1.42, 95% confidence interval [CI]: 0.59–3.41; P = 0.433), G3 (HR = 2.11, 95% CI: 0.71–6.29; P = 0.118), or Gx (HR = 1.30, 95% CI: 0.57–2.98; P = 0.584). Pairwise comparisons across all grade groups did not reveal any statistically significant differences. Multivariable analysis confirmed that pathological grade was not an independent prognostic factor for CSS (Figure 2B).

The 5-year OS rates were 73.4% (G1), 69.8% (G2), 77.9% (G3), and 74.9% (Gx) (Figure 3A). OS did not differ significantly between G1 and G2 (HR = 1.17, 95% CI: 0.80–1.73; P = 0.400), G3 (HR = 1.40, 95% CI: 0.82–2.39; P = 0.150), or Gx (HR = 1.16, 95% CI: 0.81–1.66; P = 0.459). Pairwise comparisons revealed no significant differences among grade groups. Multivariable analysis confirmed that grade was not an independent predictor of OS (Figure 3B).

Survival outcomes in esophageal squamous cell carcinoma cohort

The 5-year CSS rates were 100.0%, 89.1%, 85.7%, and 86.4% for the G1, G2, G3, and Gx groups, respectively (Figure 4A). Compared to G1, CSS was not significantly different for G2 (HR = 1.83, 95% CI: 0.19–17.66; P = 0.504), G3 (HR = 1.36, 95% CI: 0.08–22.66; P = 0.964), and Gx (HR = 2.29, 95% CI: 0.19–27.45; P = 0.449). Pairwise comparisons revealed no significant differences among grade groups. Multivariable analysis confirmed that grade was not an independent prognostic factor for CSS (Figure 4B).

The 5-year OS rates were 100.0%, 73.3%, 57.1%, and 35.1% for the G1, G2, G3, and Gx groups, respectively (Figure 5A). Compared to G1, the differences did not reach statistical significance for G2 (HR = 3.83, 95% CI: 0.47–30.96; P = 0.175), G3 (HR = 5.35, 95% CI: 0.59–48.48; P = 0.175), or Gx (HR = 7.71, 95% CI: 0.90–65.96; P = 0.062). No significant differences were identified in pairwise comparisons. Multivariable analysis further confirmed that grade was not an independent prognostic factor for OS (Figure 5B).

Discussion

This population-based cohort study found that pathological grade was not associated with OS or CSS in patients with pT1a esophageal cancer treated with endoscopic submucosal dissection. These findings were consistent across univariable and multivariable analyses, with no significant differences observed in pairwise comparisons among grade groups for either adenocarcinoma or squamous cell carcinoma.

The absence of a prognostic association between grade and survival warrants careful consideration. First, patients with pT1a disease represent a highly selected population with uniformly favorable prognosis. In our cohort, 5-year CSS rates exceeded 90% across all grade groups for adenocarcinoma and ranged from 85.7% to 100% for squamous cell carcinoma, aligning with contemporary series reporting excellent outcomes following endoscopic resection for superficial esophageal cancer[3]. This uniformly favorable prognosis may attenuate the discriminative ability of any single pathological feature, including grade.

Second, the biological behavior of pT1a disease may differ fundamentally from that of more advanced esophageal cancer. In invasive disease, poorly differentiated histology is associated with increased lymph node metastasis risk and worse outcomes[15–19]. However, for lesions confined to the epithelium or lamina propria, the risk of nodal involvement is exceedingly low (0.0%–3.3%) irrespective of grade, potentially rendering grade less clinically relevant at this early stage. Supporting this hypothesis, even the G3 subgroup, conventionally considered high-risk, demonstrated excellent survival outcomes, with 5-year CSS of 92.6% for adenocarcinoma and 85.7% for squamous cell carcinoma.

Third, the distinct biology of adenocarcinoma and squamous cell carcinoma may influence the prognostic utility of grade. Some studies suggest that grade carries differential prognostic weight across histologic subtypes, with squamous cell carcinoma potentially exhibiting greater grade-dependent variability in metastatic potential[20–23]. In our analysis, we observed a numerical trend toward worse OS in squamous cell carcinoma patients with higher-grade tumors, although these differences did not reach statistical significance, likely due to limited sample size in the G3 subgroup. This pattern suggests that the lack of statistical significance may reflect insufficient power rather than true equivalence, particularly for squamous cell carcinoma.

Our findings have several clinical implications. First, pathological grade alone should not guide post-endoscopic submucosal dissection management decisions for pT1a disease. Instead, risk stratification should emphasize two established factors: depth of tumor invasion and lymphovascular invasion. The risk of lymph node metastasis is low for lesions confined to the epithelium or lamina propria (0.0%–3.3%)[24], but increases substantially for those involving the muscularis mucosa (5.6%–21.4%)[25–29]. Accordingly, annual endoscopic surveillance is recommended for lesions confined to the epithelium or lamina propria, whereas additional treatment should be considered for muscularis mucosa lesions with lymphovascular invasion. Second, for patients with Gx tumors, the absence of a prognostic signal provides reassurance that missing grade information does not preclude favorable outcomes when other favorable characteristics are present. This observation is particularly relevant given that grade is frequently unreported or indeterminate in clinical practice.

Several limitations warrant consideration. First, the Surveillance, Epidemiology, and End Results database lacks information on key pathological variables, including lymphovascular invasion, depth of mucosal invasion (epithelium versus lamina propria versus muscularis mucosa), and margin status. These factors are known to influence outcomes in pT1a disease and may confound the observed association between grade and survival.

Second, the statistical power for the subgroup analysis of esophageal squamous cell carcinoma was suboptimal. The sample sizes within each grade category were small, resulting in estimated hazard ratios with exceedingly wide 95% confidence intervals, indicating substantial statistical uncertainty. The absence of statistically significant associations should not be construed as evidence that pathological grade lacks prognostic value. Notably, the progressive decline in 5‑year OS from 100% (G1) to 73.3% (G2), 57.1% (G3), and 35.1% (Gx) suggests a potential dose‑response relationship between grade and survival, although the limited number of events precluded formal statistical verification. Consequently, the prognostic significance of pathological grade in squamous cell carcinoma remains an open question and warrants confirmation in larger, adequately powered cohorts.

Third, the median follow‑up of 41 months limits reliable 5‑year survival estimation. Accurate estimation requires sufficient observation time to capture events; shorter follow‑up means few patients reach 5 years, and many contribute censored data, rendering the survival function unstable at the right tail. Thus, 5‑year OS and CSS estimates, particularly in subgroups with few events, carry wider confidence intervals and greater uncertainty. Moreover, if grade exerts a delayed effect on prognosis, our study lacks power to detect it.

In conclusion, using large-scale population-based data, we found no significant independent association between pathological grade and survival in patients with pT1a esophageal cancer treated with endoscopic submucosal dissection. While our findings suggest that grade alone may have limited prognostic utility in this highly selected, early-stage population, we cannot exclude the possibility that grade provides complementary prognostic information when integrated with other clinicopathological risk factors, particularly lymphovascular invasion and precise depth of mucosal invasion, which are not captured in SEER database. These results should be interpreted with caution, and prospective studies with comprehensive pathological characterization are warranted to definitively establish the role of grade in risk stratification for pT1a disease.

References

[1] Siegel RL, Kratzer TB, Giaquinto AN, Sung H, Jemal A. Cancer statistics, 2025. CA Cancer J Clin. 2025;75(1):10-45.

[2] Phan E, Ohashi S, Singh AP, et al. Screening for esophageal cancer. J Gastroenterol Hepatol. 2026;41(1):104-116.

[3] Kitagawa Y, Ishihara R, Ishikawa H, et al. Esophageal cancer practice guidelines 2022 edited by the Japan esophageal society: part 1. Esophagus. 2023;20(3):343-372.

[4] Yang H, Wang F, Hallemeier CL, Lerut T, Fu J. Oesophageal cancer. Lancet. 2024;404(10466):1991-2005.

[5] Yeh JH, Huang RY, Lee CT, et al. Long-term outcomes of endoscopic submucosal dissection and comparison to surgery for superficial esophageal squamous cancer: a systematic review and meta-analysis. Ther Adv Gastroenterol. 2020;13:1756284820964316.

[6] Donohoe CL, Phillips AW. Cancer of the esophagus and esophagogastric junction: an 8th edition staging primer. J Thorac Dis. 2017;9(3):E282-E284.

[7] Rice TW, Gress DM, Patil DT, Hofstetter WL, Kelsen DP, Blackstone EH. Cancer of the esophagus and esophagogastric junction-Major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(4):304-317.

[8] Rice TW, Ishwaran H, Hofstetter WL, et al. Recommendations for pathologic staging (pTNM) of cancer of the esophagus and esophagogastric junction for the 8th edition AJCC/UICC staging manuals. Dis Esophagus. 2016;29(8):897-905.

[9] Ancona E, Rampado S, Cassaro M, et al. Prediction of lymph node status in superficial esophageal carcinoma. Ann Surg Oncol. 2008;15(11):3278-3288.

[10] Lin Y, Wang SF, Liang HW, Liu Y, Huang W, Pan XB. Additional chemoradiotherapy following endoscopic submucosal dissection in patients with esophageal squamous cell carcinoma: a narrative review. Front Oncol. 2025;15:1527634.

[11] Mo R, Chen C, Pan L, Yu A, Wang D, Wang T. Is the new distribution of early esophageal adenocarcinoma stages improving the prognostic prediction of the 8th edition of the TNM staging system for esophageal cancer? J Thorac Dis. 2018;10(9):5192-5198.

[12] Wang RZ, Liang HW, Liu Y, Huang W, Pan XB. Chemoradiotherapy versus radiotherapy in patients with stage T1N0M0 esophageal cancer: a retrospective cohort study. J Gastroenterol Hepatol. 2025;40(11):2705-2713.

[13] Lin Y, Ma CX, Liang HW, Huang W, Pan XB. Endoscopic submucosal dissection followed by chemoradiotherapy versus radiotherapy in patients with stage T1bN0M0 esophageal cancer. J Gastrointest Cancer. 2025;56(1):150.

[14] Liu D, Liang HW, Liu Y, Huang W, Pan XB. Causes of death in locally advanced esophageal cancer undergoing neoadjuvant chemotherapy and neoadjuvant chemoradiotherapy: a retrospective cohort study. Dis Esophagus. 2025;38(2):doaf017.

[15] Fiocca R, Mastracci L, Lugaresi M, et al. The prognostic impact of histology in esophageal and esophago‑gastric junction adenocarcinoma. Cancers (Basel). 2021;13(20):5211.

[16] Lagarde SM, ten Kate FJ, Reitsma JB, Busch OR, van Lanschot JJ. Prognostic factors in adenocarcinoma of the esophagus or gastroesophageal junction. J Clin Oncol. 2006;24(26):4347-4355.

[17] Lin MQ, Li YP, Wu SG, et al. Differences in esophageal cancer characteristics and survival between Chinese and Caucasian patients in the SEER database. Onco Targets Ther. 2016;9:6435-6444.

[18] Xie L, Zhang Y, Niu X, et al. A nomogram for predicting cancer-specific survival in patients with locally advanced unresectable esophageal cancer: development and validation study. Front Immunol. 2025;16:1524439.

[19] Bassiri A, Boutros CS, Pennacchio C, et al. Risk factors associated with poor prognosis after endoscopic treatment of clinical T1a esophageal cancer. Surg Endosc. 2025;39(9):6010-6017.

[20] Liu H, Meng J. Comparison of LNM and survival in T1 stage esophageal cancer patients based on histological classification: a large population-based study. Medicine (Baltimore). 2022;101(51):e32143.

[21] Mariette C, Finzi L, Piessen G, Van Seuningen I, Triboulet JP. Esophageal carcinoma: prognostic differences between squamous cell carcinoma and adenocarcinoma. World J Surg. 2005;29(1):39-45.

[22] Marom G. Esophageal cancer staging. Thorac Surg Clin. 2022;32(4):437-445.

[23] Rice TW, Rusch VW, Ishwaran H, Blackstone EH; Worldwide Esophageal Cancer Collaboration. Cancer of the esophagus and esophagogastric junction: data-driven staging for the seventh edition of the American Joint Committee on Cancer/International Union Against Cancer Cancer Staging Manuals. Cancer. 2010;116(16):3763-3773.

[24] Kitagawa Y, Uno T, Oyama T, et al. Esophageal cancer practice guidelines 2017 edited by the Japan Esophageal Society: part 1. Esophagus. 2019;16(1):1-24.

[25] Akutsu Y, Uesato M, Shuto K, et al. The overall prevalence of metastasis in T1 esophageal squamous cell carcinoma: a retrospective analysis of 295 patients. Ann Surg. 2013;257(6):1032-1038.

[26] Hisano O, Nonoshita T, Hirata H, et al. Additional radiotherapy following endoscopic submucosal dissection for T1a-MM/T1b-SM esophageal squamous cell carcinoma improves locoregional control. Radiat Oncol. 2018;13(1):14.

[27] Takahashi K, Hashimoto S, Mizuno KI, et al. Management decision based on lymphovascular involvement leads to favorable outcomes after endoscopic treatment of esophageal squamous cell carcinoma. Endoscopy. 2018;50(7):662-670.

[28] Yamashina T, Ishihara R, Nagai K, et al. Long-term outcome and metastatic risk after endoscopic resection of superficial esophageal squamous cell carcinoma. Am J Gastroenterol. 2013;108(4):544-551.

[29] Yoshii T, Ohkawa S, Tamai S, Kameda Y. Clinical outcome of endoscopic mucosal resection for esophageal squamous cell cancer invading muscularis mucosa and submucosal layer. Dis Esophagus. 2013;26(5):496-502.

Rights & permissions

The Author(s) 2026. This article is published by Higher Education Press at journal.hep.com.cn.