Comparison of PET-CT images with the histopathological picture of a resectable primary tumor for delineating GTV in nonsmall cell lung cancer

Objective This study was designed to investigate the feasibility of a method derived from fused positron emission tomography-computed tomography (PET-CT) images for the delineation of gross tumor volume (GTV) in nonsmall cell lung cancer (NSCLC) and to explore the correlation between PET-CT and histopathological findings. Methods Thirty-seven patients were enrolled in this study; all patients were evaluated by PET-CT and underwent surgery within 1 week after the scan. The radiation oncologist, together with the radiologist, first delineated the GTV-based CT and then with an experienced nuclear medicine physician contoured the same GTV using the distinctive visual 'halo' in fused PET-CT images. The characteristics of the halo phenomenon were analyzed, including the standardized uptake value (SUV). The excised surgical specimens were labeled and the maximum diameter of the tumor in the right-left axis of the tissue blocks was measured on consecutive histopathology slides; the histopathological slice images were scanned using a digital pathology scanner after staining with hematoxylin and eosin. Results The mean SUV of the external margins of the halos was 2.41 +/- 0.73 (range 1.4-4.1); the histopathological type and T-stage significantly influenced the SUV of the external margin of the halos (P=0.004 and 0.027). The correlation coefficients of maximum diameter in the right-left axis and in the anterior-posterior axis between fused PET-CT images and histopathology were 0.935 and 0.943, respectively; the values between CT imaging and histopathological examination were 0.82 and 0.763. Conclusion There is a correlation between GTV based on the 'halo' of fused PET-CT images and the excised surgical specimen of primary tumor. It may be feasible to use the 'halo' characteristics in fused PET-CT images to delineate GTV in NSCLCs, but its reliability should be further investigated to establish an accurate spatial imaging-pathology correlation for primary tumors delineation in NSCLCs. Nucl Med Commun 32:479-485 (C) 2011 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins.