Dual-phase dual-energy computed tomography (DECT) in assessing recurrence-associated histopathological features of
Ruilai Hou1, Zhichao Dou2, Yuanyuan Liu1
1State Key Laboratory of Oral Diseases & National Center for Stomatology & National Clinical Research Center for Oral Diseases & Department of Oral Medical Imaging, West China Hospital of Stomatology, Sichuan University, Chengdu, China.
Background:
Parotid pleomorphic adenoma (PA) is the most common benign salivary gland tumor but carries a risk of postoperative recurrence. Predicting its recurrence-associated histopathological features (stroma-rich subtype and capsular invasion) preoperatively could inform surgical planning. This study aimed to evaluate the efficacy of quantitative parameters derived from preoperative dual-phase dual-energy computed tomography (DECT) in predicting these two critical features.
Methods:
In this retrospective study, 108 patients with pathologically confirmed PA who underwent preoperative dual-phase enhanced DECT were enrolled. Quantitative DECT parameters including normalized iodine concentration (NIC), normalized effective atomic number (nZeff), normalized spectral attenuation slope (nλHU), and difference in Hounsfield units (∆HU) were measured for both early- and late-phase scan. Histopathological analysis confirmed stromal richness and capsular invasion. Following univariable screening, distinct analytical strategies were employed for the two outcomes. For stromal richness, the discriminatory performance of significant individual DECT parameters was compared using receiver operating characteristic (ROC) analysis. For capsular invasion, least absolute shrinkage and selection operator (LASSO) regression was used for variable selection to build a multivariable logistic model, which was subsequently evaluated for discrimination and calibration.
Results:
Stroma-rich group (35%, 38/108) showed significantly lower early-phase DECT parameters (nZeff, NIC, nλHU, ∆HU) than non-stroma-rich PAs (P<0.033). For predicting the stroma-rich subtype, the early-phase ∆HU demonstrated the highest area under the curve (AUC) of 0.804, with a sensitivity of 84.2% and a specificity of 71.4% at the optimal cut-off. Other parameters, such as late-phase ∆HU (AUC =0.801) and early-phase nλHU (AUC =0.762), also showed significant predictive value (all P<0.001). For predicting capsular invasion, LASSO regression selected age, symptom duration, tumor shape, and early-phase ∆HU into the final model. This model achieved an AUC of 0.774 [95% confidence interval (CI): 0.698-0.850], with good calibration (Hosmer-Lemeshow test, P=0.168), yielding a sensitivity of 56.9% and a specificity of 88.4%.
Conclusions:
Dual-phase DECT provides valuable quantitative parameters for preoperatively predicting the stroma-rich subtype and capsular invasion in PA. The early-phase ∆HU is a promising single-parameter predictor for stromal richness, while a combined clinical-DECT model effectively predicts capsular invasion. This approach provides a preoperative tool for assessing recurrence risk, aiding in personalized surgical planning.
