Application value of CT-based extracellular volume fraction in assessing microsatellite status in colorectal cancer

To investigate the application value of computed tomography (CT)-derived extracellular volume fraction (ECV) for predicting microsatellite instability (MSI) status in colorectal cancer (CRC).

We enrolled a total of 121 consecutive patients with CRC who underwent curative resection. Prior to surgery, each patient underwent a contrast-enhanced CT scan for ECV quantification. MSI status was determined by immunohistochemistry or polymerase chain reaction. Univariate and multivariate techniques of logistic regression analysis were conducted to find the independent predictors. Three predictive models were developed and compared: ECV-only, clinical, and combined. Model performance was assessed using receiver operating characteristic curves, DeLong test, calibration curves, and decision curve analysis.

The mean ECV% of the MSI-high group (n = 28) was significantly higher than that of the MSS group (n = 84) (P<0.001). Significant difference was noted among MSI-H, MSI-L (n = 9), and MSS groups (F = 8.37; P<0.001). Based on multivariate analysis, tumor diameter ≥5 cm, N0 stage and higher ECV% were independent predictors. As per the results, the combination model had an AUC of 0.81 (95% CI: 0.72-0.90), which was higher than the other models but the differences were statistically not significant (P = 0.051 and P = 0.135). The model demonstrated solid calibration and positive net clinical benefit between 10% and 80% threshold probabilities.

CT derived ECV is significantly associated with MSI status in CRC. A non-invasive preoperative model for predicting microsatellite instability that combines ECV, tumor diameter and N stage has a promising diagnostic accuracy.