A lasso-based model to predict lateral lymph node metastasis in unifocal papillary thyroid carcinoma with central lymph node metastasis

ObjectiveTo screen the risk factors for lateral lymph node metastasis (LLNM) in unifocal papillary thyroid carcinoma (PTC) with central lymph node metastasis (CLNM) and create a corresponding model.MethodsA retrospective analysis of 362 patients from our hospital was performed. All patients were randomized into training and validation groups in a ratio of 7:3. Risk factors were screened using the least absolute shrinkage and selection operator (LASSO) and logistic regression analysis.ResultsThe analysis indicated that upper location, number of CLNM >= 3, rate of CLNM >= 0.172, prelaryngeal LNM, pretracheal LNM, and tall cell variant of papillary thyroid carcinoma (TCV-PTC) are independent risk factors. Visualizing the model with a nomogram, receiver operating characteristic (ROC) curves revealed an area under the curve (AUC) of 0.773 for the training group and 0.779 for the validation group. This confirms the stability and outstanding accuracy of the model. Also, the calibration curves and clinical descision curves reflect strong calibration,offering potential clinical benefits.ConclusionsThe risk factors for LLNM include metastasis to the prelaryngeal lymph nodes, metastasis to the pretracheal lymph nodes, location in the upper level, number of metastases >= 3 in CLNM, TCV-PTC and metastasis rate >= 0.172. A nomogram incorporating these factors exhibits excellent predictive value and stability.