Magnetic resonance imaging-based lymph node radiomics for predicting the metastasis of evaluable lymph nodes in rectal cancer

ARTICLE HIGHLIGHTS

Research background

Lymph node (LN) staging in rectal cancer (RC) affects treatment decisions and patient prognosis. For radiologists, the traditional preoperative assessment of LN metastasis (LNM) using magnetic resonance imaging (MRI) poses a challenge.

Research motivation

The accuracy of assessing LNM based on MRI remains limited. A meta-analysis demonstrated a sensitivity of approximately 77% and specificity of approximately 71% when using MRI to diagnose metastasis in evaluable LNs. Therefore, there is a risk of diagnostic insufficiency and overdiagnosis.

Research objectives

To explore the value of a nomogram model that combines Conventional MRI and radiomics features from the LNs of RC in assessing the preoperative metastasis of evaluable LNs.

Research methods

A total of 270 LNs (158 LNM and 112 metastatic) were included and randomly allocated to training set (111 nonmetastatic and 78 metastatic) and validation set (47 nonmetastatic and 34 metastatic) at a 7:3 ratio. Radiomic features were extracted from T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI) images of individual LN. The least absolute shrinkage and selection operator regression analysis was used for feature selection. Multivariate logistic regression analysis was used to develop the Rad-score and nomogram model. Receiver operating characteristic curves were constructed to evaluate the diagnostic performance of the models for predicting LNM. The performance of the nomogram was assessed using decision curve analysis (DCA).

Research results

The nomogram model outperformed conventional MRI and single radiomics models in evaluating LNM. In the training set, the nomogram model achieved an area under the curve (AUC) of 0.92, which was significantly higher than the AUCs of 0.82 (P < 0.001) and 0.89 (P < 0.001) of the conventional MRI and radiomics models, respectively. In the validation set, the nomogram model achieved an AUC of 0.91, significantly surpassing 0.80 (P < 0.001) and 0.86 (P < 0.001), respectively.

Research conclusions

The nomogram model showed the best performance in predicting metastasis of evaluable LNs.

Research perspectives

Radiomics holds great promise for transforming medical practice, especially for patients with RC. However, before its widespread adoption, challenges regarding sample size, model design, and robust multicenter validation sets must be addressed. To validate the proposed model externally, future prospective multicenter studies with larger sample sizes are crucial.