Clinical Scorecard: A Radiomics Nomogram Utilizing CT Imaging for Preoperative Assessment of Lymphovascular Invasion in Colorectal Cancer
At a Glance
Category
Detail
Condition
Colorectal cancer (CRC) with lymphovascular invasion (LVI)
Key Mechanisms
Tumor invasion into lymphatic and blood vessels assessed via CT-based radiomics features including tumor volume, diameter, invasion depth, lymph node size, CEA level, NLR, and intratumoral heterogeneity
Target Population
Patients with pathologically confirmed colorectal cancer undergoing preoperative contrast-enhanced abdominal CT
Care Setting
Preoperative clinical evaluation in oncology and surgical settings
Key Highlights
Developed a CT-based radiomics nomogram integrating Rad-score, clinical, and laboratory factors for preoperative LVI prediction in CRC.
Nomogram demonstrated good predictive accuracy with AUCs of 0.776 (training) and 0.722 (validation).
Model calibration and decision curve analysis confirmed reliability and clinical net benefit for individualized treatment planning.
Guideline-Based Recommendations
Diagnosis
Use postoperative histopathology as the definitive diagnosis for LVI.
Employ preoperative contrast-enhanced abdominal CT imaging with radiomics feature extraction to non-invasively predict LVI status.
Management
Incorporate nomogram predictions to guide individualized clinical strategies and postoperative adjuvant therapy decisions.
Consider tumor volume, maximum tumor diameter, invasion depth, lymph node short-axis diameter, CEA, and NLR as key factors influencing LVI risk.
Monitoring & Follow-up
Monitor clinical and laboratory parameters including Carcinoembryonic Antigen (CEA) levels and Neutrophil-to-Lymphocyte Ratio (NLR) as part of risk assessment.
Use radiomics-based imaging features to assess tumor heterogeneity and progression preoperatively.
Risks
Recognize that LVI presence indicates higher invasive and metastatic potential, impacting prognosis and treatment outcomes.
Be aware of limitations in sensitivity of other imaging modalities (MRI, PET-CT) and the advantages of CT-based radiomics in routine assessment.
Patient & Prescribing Data
360 colorectal cancer patients undergoing preoperative CT imaging and subsequent surgery
The nomogram aids in stratifying patients by LVI risk preoperatively, facilitating tailored surgical and adjuvant treatment decisions to improve prognosis.
Clinical Best Practices
Ensure high-quality contrast-enhanced abdominal CT imaging in the portal venous phase within 1–2 weeks before surgery for optimal radiomics feature extraction.
Apply LASSO regression and multivariate logistic analysis to select robust predictors for LVI in model development.
Validate predictive models with independent patient cohorts to confirm discrimination, calibration, and clinical utility.
Integrate radiomics features with clinical and laboratory data for comprehensive risk assessment.
Use decision curve analysis to evaluate the net clinical benefit of predictive models across risk thresholds.
