Identification of lymphophilic subtype in resectable esophageal squamous cell carcinoma using genetic signatures from large-scale genomic repertoires

BACKGROUND: Driven by efforts to balance oncologic outcomes and perioperative morbidity, the individualized surgical management of oesophageal cancer (EC) lymph node metastases is evolving away from systemic lymph node dissections. Creating personalized treatment plans could become problematic when no genetic tests are available to detect aggressive tumors before surgery. METHODS: This cohort study was performed on 564 oesophageal squamous cell carcinoma (ESCC) patients across six next-generation sequencing studies. The genomic classifier for nodal metastasis risk prediction was generated by the least absolute shrinkage and selection operator (LASSO) logistic regression analysis with 10-fold cross-validation based on the selected different genes. Receiver operating characteristic (ROC) analysis was used to assess the performance of the classification model. RESULTS: After excluding three patients with missing lymph node status, a total of 561 ESCC patients met the inclusion criteria: 335 (59.7%) with nodal-positive (NP) and 226 (40.3%) with nodal-negative (NN). Feature selection identified 112 mutated genes to predict patients with NP versus NN. The LASSO model identified NP patients with an accuracy of 86.45% and an area under the ROC curve (AUC) of 0.95. CONCLUSIONS: Using molecular profiling to identify lymphophilic ESCC is feasible for creating personalized surgical plans in clinical decision-making.