Lung Tumor Characterization by CT Imaging

Images of the lung tumors were taken using a 64 multislice CT (Light Speed VCT, GE Medical Systems, Milwaukee, WI,) and 16 multislice CT (Aquilion LB, Toshiba Medical Systems, Tokyo, Japan). Raw CT data were reconstructed into an axial CT image with 2.5 mm slice thicknesses according to the JCOG0201 definition.^{9 (link)} The CT image was displayed with a window level of −600 Hounsfield unit (HU) and a window width of 2000 HU as the lung image. Based on the JCOG0201 study, ground-glass opacity was defined as an area of a slight, homogenous increase in density that did not obscure any underlying vascular markings, and was considered to be tumor.^{9 (link)} The solid tumor component was defined as an area of increased opacification that completely obscured any underlying vascular markings. The diameters of the whole tumor and solid tumor component were measured by thin-slice CT using the SYNAPSE VINCENT software program (Fujifilm Medical, Tokyo, Japan). Peripheral and central types were distinguished to determine whether the tumor was located in the “no-fly zone”—2 cm around the proximal airway, as defined in the Radiation Therapy Oncology Group 0236 trial.^{10 (link)} Measurements of tumor diameter and tumor location were determined from the CT data without accompanying clinical data by two different radiologists.