Predictive Value of Quantitative Parameters in Evaluating Radiotherapy Efficacy for Hepatocellular Carcinoma Based on IQon Spectral CT: A Prospective Study

Author(s): Jin Ying Lan, Jin Han Yang, Dong Wu Chen, Jin Yuan Liao

Background: Currently, there is a lack of reliable biomarkers for the early prediction of therapeutic response to radiotherapy in patients with hepatocellular carcinoma (HCC). Conventional imaging modalities are limited in their ability to detect early treatment-induced changes. IQon Spectral computed tomography (CT), as an advanced imaging technique, enables the acquisition of quantitative parameters, including iodine concentration and spectral curve slope, thereby providing a novel, noninvasive method for assessing tumor angiogenesis and metabolic activity. This study aims to prospectively evaluate the clinical utility of these quantitative parameters in predicting radiotherapy outcomes in HCC.

Method: This prospective study enrolled 30 patients with HCC who met the predefined inclusion and exclusion criteria were prospectively enrolled between July 2022 and June 2024. All patients received two cycles of radiotherapy and underwent IQon Spectral CT scans both before and after treatment. Based on the modified Response Evaluation Criteria in Solid Tumors (mRECIST), patients were classified into an effective group (complete response [CR] + partial response [PR]) and an ineffective group (stable disease [SD] + progressive disease [PD]). Changes in CT-derived quantitative parameters before and after radiotherapy were compared and analyzed, a predictive model for therapeutic efficacy was constructed, and receiver operating characteristic (ROC) curve analysis was performed to evaluate diagnostic performance.

Results: All 30 patients were diagnosed with HCC at clinical stages II– IV. Among them, 3 achieved CR, 19 achieved PR, 5 had SD, and 3 had PD. To evaluate the therapeutic efficacy of radiotherapy in HCC, three parameters—spectral curve slope (KAP), iodine content (ICAP), and normalized iodine concentration (NICAP)—were integrated during the arterial phase before treatment. Based on these parameters, five prediction models were constructed: bivariate models, a trivariate model, and a joint prediction model. The area under the curve (AUC) values were as follows: KAP_ICAP model, 0.642 (95% CI: 0.410–0.874); KAP_NICAP model, 0.886 (95% CI: 0.759–1); ICAP_NICAP model, 0.898 (95% CI: 0.778–1); and KAP_ICAP_NICAP trivariate model, 0.915 (95% CI: 0.814–1), with a sensitivity of 86.4% and specificity of 87.5%. DeLong test results indicated that the predictive performance of the KAP_ICAP_ NICAP model was significantly superior to that of the KAP_ICAP model (Z = -2.396, P = 0.017), but no statistically significant differences were observed when compared with other bivariate models (P>0.05). The joint prediction model also achieved an AUC of 0.915 (95% CI: 0.810–1), with a sensitivity of 90.9%, specificity of 87.5%, positive predictive value of 95.2%, and negative predictive value of 77.8%. The DeLong test revealed no significant difference between the joint prediction model and the trivariate model (P = 0.999), yet both outperformed the KAP_ICAP model significantly (Z = -2.362, P = 0.018). In conclusion, models incorporating the NICAP parameter demonstrate high accuracy in predicting radiotherapy efficacy for HCC. Both the trivariate model and the ensemble learningbased joint prediction model exhibit excellent predictive performance and can serve as reliable quantitative tools for clinical assessment of treatment outcomes.

Conclusion: Quantitative parameters derived from IQon Spectral CT scan sensitively reflect early radiological changes associated with radiotherapy in HCC. The findings confirm that pretreatment IQon Spectral CT parameters, especially NICAP, can noninvasively and sensitively predict early treatment response to radiotherapy in HCC, offering a reliable quantitative imaging tool to support individualized therapeutic decision-making. The broader clinical applicability of this approach requires further validation through multicenter studies.