Radiomics Model for Assessing Lymph Node Status in cN0 Patients withHNSCC

CT-based Radiomics Predicts Occult LNM and Uncovers Immune Microenvironment of Head and Neck Cancer

Occult lymph node metastasis (LNM) remains one of the most critical and challenging aspects of managing head and neck squamous cell carcinoma (HNSCC). Defined as the presence of metastatic disease in lymph nodes that are clinically undetectable through routine imaging or physical examination, occult LNM has profound implications for treatment planning, prognosis, and overall patient management. In HNSCC, accurate detection and prediction of occult LNM are crucial as they significantly influence decisions regarding the extent of neck dissection, the need for adjuvant therapies, and the overall therapeutic strategy. Undiagnosed or underestimated LNM can result in inadequate treatment, increasing the risk of locoregional recurrence and poor survival outcomes.

Study Overview

• Status: Completed
• Conditions: HNSCC
• Intervention / Treatment: Diagnostic test: AI

Detailed Description

Occult lymph node metastasis (LNM) remains one of the most critical and challenging aspects of managing head and neck squamous cell carcinoma (HNSCC). Defined as the presence of metastatic disease in lymph nodes that are clinically undetectable through routine imaging or physical examination, occult LNM has profound implications for treatment planning, prognosis, and overall patient management. In HNSCC, accurate detection and prediction of occult LNM are crucial as they significantly influence decisions regarding the extent of neck dissection, the need for adjuvant therapies, and the overall therapeutic strategy. Undiagnosed or underestimated LNM can result in inadequate treatment, increasing the risk of locoregional recurrence and poor survival outcomes.

The complex biology of HNSCC adds to the challenge of predicting occult LNM. These tumors are often characterized by substantial heterogeneity in their microenvironment, comprising a mix of tumor cells, immune infiltrates, stromal components, and vasculature. This heterogeneity plays a pivotal role in determining the metastatic potential of the primary tumor and its interaction with the surrounding lymphatic system. Traditional imaging modalities such as CT, MRI, and PET/CT have limitations in accurately identifying microscopic metastases, leading to the ongoing search for more sensitive and specific predictive tools.

Recent advances in radiomics have opened new avenues for addressing this challenge. Radiomics, an emerging field that extracts high-dimensional data from medical imaging, allows for the quantitative analysis of tumor characteristics beyond what is visible to the naked eye. By converting imaging data into a rich repository of features that reflect tumor phenotype, radiomics has the potential to identify subtle patterns associated with metastatic behavior.

Accurate prediction of occult LNM also holds critical prognostic value. Patients with undetected LNM often face a worse prognosis due to delayed or insufficient treatment. Conversely, unnecessary prophylactic neck dissection in patients without metastasis can lead to overtreatment, increased surgical morbidity, and diminished quality of life. Therefore, predictive models that can stratify patients based on their risk of occult LNM are essential for personalizing treatment plans, reducing unnecessary interventions, and improving patient outcomes.

In this context, the integration of radiomics with multi-omics data, including transcriptomics and single-cell RNA sequencing, represents a transformative approach. This integrative strategy not only enhances the predictive power of radiomics models but also provides a window into the biological processes underlying tumor behavior. By linking imaging-derived features to molecular and cellular pathways, such approaches can help bridge the gap between imaging phenotypes and the complex biology of metastasis.

In summary, occult LNM poses a formidable challenge in the clinical management of HNSCC, with significant implications for treatment and prognosis. The advent of advanced radiomics techniques, particularly habitat radiomics, offers a promising avenue for improving the accuracy of LNM prediction. By unraveling the interplay between tumor heterogeneity, microenvironmental dynamics, and metastatic potential, these approaches pave the way for more precise and personalized management of HNSCC patients.

• Study Type: Observational
• Enrollment (Actual): 700

Contacts and Locations

Study Locations

• China
  • Chongqing, China
    • The First Affiliated Hospital of Chongqing Medical University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Probability Sample

Study Population

All patients with pathologically confirmed laryngeal carcinoma between January 2016 and December 2024 from three hospitals were collected and were divided into different test groups.

Description

Inclusion Criteria:

1. Availability of complete clinical data;
2. Diagnosis of laryngeal squamous cell carcinoma confirmed by surgery or biopsy;
3. CT contrast-enhanced examination performed within two weeks prior to surgery.
4. All patients underwent neck lymph node dissection surgery.

Exclusion Criteria:

1. Patients who received other treatments before surgery;
2. CT images with significant artifacts;
3. Patients with tumor recurrence.

Study Plan

How is the study designed?

Number of groups / cohorts

3

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Training set; The training set comprised approximately 500 cN0 patients diagnosed with head and neck squamous cell carcinoma (HNSCC), including approximately 150 patients with lymph node metastasis and approximately 350 patients without metastasis. All patients underwent preoperative contrast-enhanced CT scans.	Diagnostic test: AI; Using artificial intelligence models to distinguish between patients with lymph node metastasis and those without lymph node metastasis.
internal test set; The internal validation set included approximately 150 patients, randomly selected from the training cohort. This set was used for model evaluation and tuning.	Diagnostic test: AI; Using artificial intelligence models to distinguish between patients with lymph node metastasis and those without lymph node metastasis.
external test set; The external validation set consisted of approximately 200 patients with HNSCC. These patients were enrolled from other centers, and their data included preoperative contrast-enhanced CT images. This independent dataset was used to assess the generalizability of the radiomics model.	Diagnostic test: AI; Using artificial intelligence models to distinguish between patients with lymph node metastasis and those without lymph node metastasis.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
AUC	AUC (Area Under the Curve) is a performance metric used in classification tasks to evaluate the ability of a model to distinguish between classes. Specifically, it measures the area under the Receiver Operating Characteristic (ROC) curve, which plots the true positive rate (sensitivity) against the false positive rate (1-specificity) at various threshold settings.	The prediction results can be obtained immediately after the model completes processing.

Collaborators and Investigators

• Sponsor: First Affiliated Hospital of Chongqing Medical University
• Collaborators: Zigong The Fourth People's Hospital

Study record dates

Study Major Dates

• Study Start (Actual): November 27, 2024
• Primary Completion (Actual): April 15, 2025
• Study Completion (Actual): April 15, 2025

Study Registration Dates

• First Submitted: December 27, 2024
• First Submitted That Met QC Criteria: December 27, 2024
• First Posted (Actual): January 3, 2025

Study Record Updates

• Last Update Posted (Actual): May 28, 2025
• Last Update Submitted That Met QC Criteria: May 27, 2025
• Last Verified: May 1, 2025

More Information

Terms related to this study

• Keywords: Radiogenomics, Occult lymph node metastasis
• Other Study ID Numbers: 2024-Chenxin

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No