Combined Spectral CT and Perfusion CT for Noninvasive Prediction of EGFR Mutation Status in Non-Small Cell Lung Cancer

Objective The determination of epidermal growth factor receptor (EGFR) mutation status is crucial for the treatment and prognosis of non-small cell lung cancer (NSCLC). This study seeks to investigate the predictive capabilities of spectral CT and perfusion CT in identifying EGFR mutations. Methods A retrospective analysis was performed on 107 patients diagnosed with lung cancer who underwent energy spectrum combined perfusion scanning prior to treatment at Shanghai Chest Hospital between March 2023 and January 2025. The patients were categorized into a mutation-positive group (27 cases) and a mutation-negative group (80 cases) based on their EGFR mutation status. The spectral CT and perfusion CT parameters of both groups were analyzed for differential diagnosis, and diagnostic models were developed. The dataset was divided into training and testing groups in an 8:2 ratio to evaluate the diagnostic efficacy of the comprehensive model. Results In the analysis of energy spectrum parameters, it was observed that the 40 keV CT value, 60 keV CT value, energy spectrum curve slope (K), iodine concentration (IC), normalized iodine concentration (NIC), and effective atomic number (Zeff) were significantly elevated in patients within the mutation group compared to the non-mutation group during both the arterial and venous phases (P < 0.05). Regarding perfusion parameters, the surface permeability (PS) was notably higher in the mutant group than in the non-mutant group (P = 0.001). A comparative evaluation of model performance revealed that the venous-phase combined model—developed by integrating spectral and perfusion parameters from the venous phase—demonstrated superior diagnostic accuracy for predicting EGFR mutation status (AUC = 0.837 in the training cohort; AUC = 0.848 in the test cohort), surpassing the arterial-phase combined model, which was based on arterial-phase spectral and perfusion parameters (AUC = 0.813 in the training cohort; AUC = 0.804 in the test cohort). Conclusion The integration of quantitative parameters derived from energy spectrum venous phase computed tomography (CT) and perfusion metrics demonstrates significant diagnostic utility in predicting the epidermal growth factor receptor (EGFR) mutation status in non-small cell lung cancer (NSCLC).