Therapy response assessment of non-small cell lung cancer using dual-energy computed tomography iodine map

OBJECTIVE: To investigate feasibility of the quantitative parameters of dual-energy computed tomography (DECT) to assess therapy response in advanced non-small cell lung cancer (NSCLC) compared with the traditional enhanced CT parameters based on the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines. METHODS: Forty-five patients with unresectable locally advanced NSCLC who underwent DECT before and after chemotherapy or concurrent chemoradiotherapy (cCRT) were prospectively enrolled. By comparing baseline studies with follow-up, patients were divided into two groups according to RECIST guidelines as follows: disease control (DC, including partial response and stable disease) and progressive disease (PD). The diameter (D), attenuation, iodine concentration and normalized iodine concentration of arterial and venous phases (ICA, ICv, NICA, NICv) and the percentage of these changes pre- and post-therapy were measured and calculated. The Pearson correlation was used to analyze correlation between various quantitative parameters. The receiver operating characteristic (ROC) curves were used to evaluate accuracy of therapy response prediction. RESULTS: The change percentages of Attenuation (Delta-Attenuation- A and Delta-Attenuation-V), IC (Delta ICA and Delta ICV) and NIC (Delta NICA and Delta NICV) pre- and post-therapy correlate with the change percentage of D (Delta D). Among these, Delta ICA strongly correlates with Delta D (r = 0.793, P < 0.001). The areas under ROC curves generated using Delta-Attenuation-A, Delta ICA, and Delta NICA are 0.796, 0.900, and 0.880 with the corresponding cutoff value of 9.096, -15.692, and -4.7569, respectively, which are significantly different (P < 0.001). CONCLUSIONS: The quantitative parameters of DECT iodine map, especially iodine concentration, in arterial phase provides a new quantitative image marker to predict therapy response of patients diagnosed with advanced NSCLC.