Risk of lymph node metastasis in T1 esophageal adenocarcinoma: a meta-analysis.

Patients with early (T1) esophageal adenocarcinoma (EAC) are increasingly having definitive local therapy endoscopically. Endoscopic resection is not able to pathologically stage or treat lymph node metastasis (LNM). Accurate identification of patients having nodal metastasis is critical to select endoscopic therapy over surgery. This study aimed to define the risk of LNM in T1 EAC. A meta-analysis of studies of patients who underwent surgery and lymphadenectomy with assessment of LNM was performed according to PRISMA. Main outcome was probability of LNM in T1a and T1b disease. Secondary outcomes were risk factors for LNM and rate of LNM in submucosal T1b (SM1, SM2, and SM3) disease. Registered with PROSPERO (CRD42022341794). Twenty cohort studies involving 2264 patients with T1 EAC met inclusion criteria: T1a (857 patients) with 36 (4.2%) node positive and T1b (1407 patients) with 327 (23.2%) node positive. Subgroup analysis of T1b lesions was available in 10 studies (405 patients). Node positivity for SM1, SM2, and SM3 was 16.3%, 16.2%, and 29.4%, respectively. T1 substage (odds ratio [OR] 7.72, 95% confidence interval [CI] 4.45-13.38, P < 0.01), tumor differentiation (OR 2.82, 95% CI 2.06-3.87, P < 0.01), and lymphovascular invasion (OR 13.65, 95% CI 6.06-30.73, P < 0.01) were associated with LNM. T1a disease demonstrated a 4.2% nodal metastasis rate and T1b disease a rate of 23.2%. Endoscopic therapy should be reserved for T1a disease and perhaps select T1b disease, which has a moderately high rate of nodal metastasis. There were inadequate data to stratify T1b SM disease into 'low-risk' and 'high-risk' based on tumor differentiation and lymphovascular invasion.