The development of a nomogram model for predicting left recurrent laryngeal nerve lymph node metastasis in esophageal cancer based on radiomics and clinical factors

Abstract

Background: The lymph node dissection for esophageal cancer is controversial. Some prediction models of lymph node metastasis (LNM) use the short diameter of lymph nodes measured by computed tomography (CT) examination as a predictor, but the size of that for judging metastasis is still controversial. However, radiomics can extract some features in tumors that cannot be obtained by naked eyes, which may have a higher value in predicting LNM. In this study, a nomogram was developed based on radiomics and clinical factors to predict left recurrent laryngeal nerve lymph node (RLNN) metastasis in patients with esophageal squamous cell carcinoma (ESCC).Methods: There were 350 patients included in this retrospective study. And the postoperative pathological results determined whether there was left RLNN metastasis. A univariate analysis was conducted of the clinical data. The least absolute shrinkage and selection operator regression analysis was conducted to filter the radiomics features extracted from CT images. The multivariate logistic regression equation was used to construct a nomogram. The area under the curve (AUC) was used to evaluate the predictive ability. Due to the small sample size, we chose to perform internal validation after the model was established by 10-fold cross-validation, Harrell's concordance index (C-index), bootstrap validation and calibration.Results: Ultimately, 3 indicators were screened out; that is, tumor location, surface volume ratio, and runlength non-uniformity. We then constructed the nomogram using these 3 indicators. The model had good accuracy and calibration performance. It has an AUC of 0.903 (95% confidence interval: 0.861-0.945), a sensitivity of 0.873, and a specificity of 0.756. Ten-fold cross-validation showed that the sensitivity and specificity of the training set were 88.08% and 75.81%, and the validation set had a sensitivity of 85.08% and a specificity of 75.49%. The Brier score was 0.074, and C-index was 0.904, which indicated good consistency between the actual and predicted results. Conclusions: A nomogram constructed based on radiomics features and clinical factors can be used to predict the metastasis of left RLNN in patients with ESCC in a non-invasive way, which provided a reference for clinicians to formulate individualized lymph node dissection plans.

CT-based radiomics for predicting lymph node metastasis in esophageal cancer: a systematic review and meta-analysis

“…The eighth edition of the AJCC’s International Staging Standard for Esophageal Cancer introduced a clinical staging system based on preoperative imaging; it incorporated the count of lymph node metastases in postoperative staging ( 8 , 9 ). While pathological findings continue to be the gold standard for diagnosing LNM, lymph node biopsy represents an invasive procedure with a non-trivial incidence of complications ( 10 ). Therefore, accurate evaluation of lymph node status using non-invasive imaging methods is imperative for making informed treatment decisions and essential for precise prognostication.…”