KEY TAKEAWAYS
- The study aimed to evaluate the use of radiomics analysis with enhanced CT images for predicting VI in gastric cancer.
- Integrated models with radiomics features and clinical factors effectively predict VI and aid personalized treatment.
Vascular invasion (VI) significantly influences metastasis, recurrence, prognosis, and treatment outcomes in gastric cancer. However, accurately predicting VI before surgery using conventional clinical methods remains difficult.
Zhicheng Chen and the team aimed to investigate the potential of radiomics analysis using preoperative enhanced CT images to predict VI in gastric cancer.
Researchers reviewed records of 194 patients with gastric adenocarcinoma who had enhanced CT scans. They categorized the patients into 2 groups: those with VI (VI, n=43) and those without VI (non-VI, n=151). Radiomics features were pulled from both arterial phase (AP) and portal venous phase (PP) CT images to compute a radiomics score (Rad-score).
They built prediction models using image features, clinical data, or a combination of both and assessed these models with receiver operating characteristics (ROC) curves and decision curve analysis (DCA).
The results indicated that the combined prediction model used the Rad-score from both AP and PP, along with Ki-67 and Lauren classification. In the training group, this model achieved an area under the curve (AUC) of 0.83 (95% CI: 0.76-0.89), with 64.52% sensitivity and 92.45% specificity.
For the validation group, the AUC was 0.80 (95% CI: 0.67-0.89), with 66.67% sensitivity and 88.89% specificity. The DCA suggested that this model might offer more significant clinical benefits than clinical factors alone.
The study concluded that integrated models incorporating enhanced CT radiomics features, Ki-67 and clinical factors exhibit significant predictive capability for VI. Additionally, the radiomics model can potentially optimize personalized treatment and prognosis assessment.
The study received no funding.
Source: https://pubmed.ncbi.nlm.nih.gov/39152398/
Chen Z, Zhang G, Liu Y, et al. (2024). “Radiomics analysis in predicting vascular invasion in gastric cancer based on enhanced CT: a preliminary study.” BMC Cancer. 2024 Aug 16;24(1):1020. doi: 10.1186/s12885-024-12793-7. PMID: 39152398; PMCID: PMC11330039.
