ESTRO 37 Abstract book

S1144

ESTRO 37

5 Fudación Instituto Valenciano de Oncología, Radiation Therapy Department, Valencia, Spain Purpose or Objective The iodine content quantification parameter can be generated by Dual Energy Computed Tomography (DECT). In lung cancer, the iodine contrast absorbed by a tumour (I abs ) is related to the angiogenesis process and, therefore, to the tumour aggressiveness and overall survival. In this study, the relationship of I abs to the survival of lung cancer patients treated with radiation therapy is assessed. Material and Methods A total of 38 lung cancer and metastatic patients treated with radiation therapy who underwent initial diagnostic DECT following the suspicion of lung cancer were prospectively studied. 9 patients were treated with hypofractionated schemes (40-70 Gy in 5-10 fractions) and 29 with conventional fractionation (60-70 Gy in 30-35 fractions). The lesions were adenocarcinoma (55.3%), metastasis (13.2%), squamous cell carcinoma (21.1%), and small cell lung cancer (10.5%). The CT examinations were performed on a Discovery CT750 HD scanner (GE Healthcare, USA). The patients were injected with 1.35 ml/kg of body weight of an iodinated contrast material (Iopamidol, 300 mg/ml; Bracco, Italy). The iodine measurement study (quantified in mg/cm 3 ) was reconstructed with workstation ADW4.6 (GE Healthcare, USA), and the lesion volumes were semi- automatically segmented using the Dexus® lung nodule function. Minimum, mean, maximum, and standard deviation I abs values were registered as (I abs ) minimum , (I abs ) mean , (I abs ) maximum , and (I abs )σ respectively. The cohort was stratified into two groups according to threshold values above or below the percentile 25%, 50%, and 75% for each parameter. Different outcomes were assessed with a Mantel-Cox test in SPSS 20 (IBM, USA). Statistical significance was set at p < 0.05. A Cox regression was used for the clinical data evaluation of survival. In addition to (I abs ) minimum , (I abs ) mean , (I abs ) maximum , and (I abs )σ, several dummy variables were included: T2 and T3 vs. T1, N2 and N3 vs. N1 and N0, conventional vs. hypofractionated radiotherapy, and squamous cell carcinoma, metastasis, and small cell lung cancer vs. adenocarcinoma. Results The mean survivorship of the cohort was 780 ± 60 days (95% level of confidence: 663-1371 days)

confidence level (C.L.) for the Cox regression of these covariables.

Conclusion (I abs

) in the lung lesion quantified by DECT reflects probably a process of angiogenesis and is related to patient survivorship. EP-2081 Contrast medium extraction method using raw-data based electron density with Dual energy-CT D. Kawahara 1 , T. Sodai 2 , O. Shuichi 3 , Y. Kazushi 4 , H. Toru 5 , F. Chikako 4 , S. Akito 3 , N. Takeo 4 , O. Yoshimi 4 , N. Yasushi 3 1 Hiroshima University, Graduate School of Biomedical & Health Sciences, Hiroshima, Japan 2 University of Tokyo, Department of Nuclear Engineering and Management- School of Engineering, Tokyo, Japan 3 Hiroshima University, Department of Radiation Oncology- Institute of Biomedical & Health Sciences, Hiroshima, Japan 4 Hiroshima University Hospital, Division of Clinical Support, Hiroshima, Japan 5 Hiroshima University, Departments of Diagnostic Radiology and Radiology, Hiroshima, Japan Purpose or Objective Dual-energy CT (DECT) obtains various data of scanned materials such as raw-data based electron density (rED) and CT value and effective atomic number, which enables the tissue segmentation. In our work, we created a contrast medium extraction method using rED and CT number from DECT for the automatic delineation of the contrast region. Material and Methods A CT-ED phantom which contains tissue-equivalent inserts and an acrylic phantom which contains inserts with the various concentrations (0-130 mg/mL), of iodinated contrast medium were scanned by DECT. CT scans were performed at tube voltages of 80/135 kV and 100/135 kV. Contrast medium was separated from other tissue materials by using the threshold of CT number and the difference of the gradient in the rED-CT correlation, which was analyzed in an in-house software. The accuracy of the contrast medium extraction was evaluated by measuring the diameter in terms of the full width at half maximum (FWHM), and the ratio of the volume (ROV).

Results The CT number of the bone materials and the contrast medium was obviously greater than 0 HU. CT number of only bone materials and contrast medium obtained by 80 kV and 100 kV is higher than that by 135 kV. The gradient of the bone materials was greater than that of the contrast medium. The ratio of the measurement to the actual diameter in FWHM and the ROV was 0.98-1.00 at more than 2 mg/mL. At tube voltage of 100/135 kV, the ratio of the measurement to the actual diameter in FWHM and the ROV was 0.98-1.00 at more than 3 mg/mL.

(I abs ) mean lung lesion for a threshold set at percentile 25% (I abs ) threshold = 21.68 mg/cm 3 was the only parameter that distinguished significantly (p = 0.012) sub-cohorts according to their global survivorship. (I abs ) mean , (I abs ) maximum and squamous cell carcinoma were significant in the Cox regression model. Table 1 shows the values of β, significance, exp(β), and its 95%