AUTHOR=Wang Jiayang , Peng Jin , Luo Honglei , Song Yaqi 

TITLE=Development and internal validation of a nomogram for predicting survival of nonoperative EGFR-positive locally advanced elderly esophageal cancers

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.1097907

DOI=10.3389/fonc.2023.1097907

ISSN=2234-943X

ABSTRACT=Purpose: To develop and validate a prediction model for nonoperative EGFR-positive locally advanced elderly esophageal cancer (LAEEC). Methods: Eighty EGFR-positive LAEEC patients were selected. All patients received radiotherapy, and 41 cases received ectinib concurrent systemic therapy. A nomogram was established by univariate and multivariate cox analysis. AUC values and ROC curves at different time points, time dependent consistency index (C-index), calibration curves, and clinical decision curves were used to assess the efficacy of the model. Bootstrap resampling and out of bag (OOB) cross-validation were utilized to verify the robustness of the model. Survival analysis based on death risk, staging and ECOG score was performed. Results: Univariate and multivariate cox analyses showed that drug, stage and ECOG score were independent prognostic factors of LAEEC patients. The AUCs of model-based prediction scoring (PS) to OS at 1, 2 and 3 years was 0.852, 0.827 and 0.792, respectively. Calibration curves showed that the predicted mortality was consistent with the actual mortality. Time-dependent AUC of the model was higher than 0.75, and internal cross-validation calibration curves showed that predicted mortality was in good agreement with actual mortality. Clinical decision curves showed that the model had good net clinical benefit within a threshold probability between 0.2 and 0.8. Model-based risk stratification analysis showed that the model was better at discriminating patients’ risk of death. Further subgroup analyses showed that erlotinib significantly improved patients survival in the patients with stage III and ECOG 1. (HR 0.122, P<0.001). Conclusions: Our nomogram model can predict overall survival in LAEEC, and benefit of Icotinib was only found in clinical stage III, good ECOG Score population.