Watch and Wait Works for Intermediate-Risk Superficial Esophageal Cancer

Review by Dr. Luís Correia Gomes (Portugal)

Study Summary 

Endoscopic resection (ER) is the standard treatment for superficial esophageal squamous cell carcinoma (ESCC). Patients whose tumor invades the muscularis mucosa (pT1a-MM) but without lymphovascular invasion (LVI) fall into an intermediate risk category, with a reported metastatic recurrence rate of 4.3%–5.6%. Whether these patients benefit from additional chemoradiotherapy (CRT) or surgery after ER, versus observation alone, has remained unclear, as prior evidence came mainly from single-center studies with limited long-term follow-up.

This multicenter retrospective study, conducted through the Japan Clinical Oncology Group across 47 institutions, enrolled 540 patients with pT1a-MM ESCC without LVI, treated between 2008 and 2015 (median age, 67 years). After ER, 485 patients (89.8%) were observed, 45 (8.3%) received CRT, and 10 (1.9%) underwent surgery, at the discretion of each institution. Cumulative recurrence rate (CRR), overall survival (OS), and recurrence-free survival (RFS) were compared over a median follow-up of 7.0 years.

Recurrence occurred in 5.7% of patients overall, almost always in lymph nodes. The 5-year CRR was 4.9% for observation, 2.2% for CRT, and 10.0% for surgery, while 5-year OS was 92.7%, 93.3%, and 100%, and 5-year RFS was 89.6%, 91.1%, and 90.0%, respectively. Notably, nearly one in five recurrences (17.9%) occurred more than 5 years after ER, and 57.1% of patients who recurred eventually died of their primary ESCC.

Commentary 

The principal strength of this study is its scale and duration: 540 patients from 47 institutions followed for a median of 7 years is the largest and longest duration of follow up dataset yet assembled for this specific intermediate-risk subgroup, and the use of a competing-risk framework for CRR adds statistical rigor. Its main weakness is that treatment allocation was left to each institution's discretion rather than randomized, so the CRT and surgery groups are small, imbalanced, and almost certainly subject to selection bias, limiting direct comparison of efficacy between strategies.

The findings broadly support observation as a reasonable default for this population, given its low absolute recurrence risk and outcomes comparable to more invasive strategies. However, the fact that 57.1% of patients with recurrence died of their ESCC, and that a substantial share of recurrences emerged beyond the 5-year mark, challenges current surveillance windows and suggests that follow-up protocols may need to extend further, potentially incorporating complementary modalities such as endoscopic ultrasound.

Regarding generalizability, although this cohort was treated almost exclusively by endoscopic submucosal dissection (ESD) in Japan, ESD has become increasingly established in Western high-volume centers with outcomes approaching those of Japanese series. The core finding—that observation is safe for pT1a-MM ESCC without LVI—is therefore likely applicable beyond East Asia, though differences in ESCC incidence, tumor biology, and access to intensive endoscopic surveillance in lower-resource settings should be considered. Going forward, prospective multicenter studies—ideally with standardized surveillance protocols extending beyond 5 years—are needed to confirm these findings and better define which patients, if any, warrant additional treatment after endoscopic resection.

Citation

Kadota T, Kawachi H, Fujii S, et al. Long-term outcome after endoscopic resection for esophageal squamous cell carcinoma invading muscularis mucosa without lymphovascular invasion: a multicenter retrospective study. Gastrointest Endosc 2026;104:286–292.

View past News You Can Use commentaries