Conformal radiotherapy for lung cancer: different delineation of the gross tumor volume (GTV) by radiologists and radiation oncologists.

Abstract

Delineation of the gross tumor volume (GTV) and organs at risk constitutes one of the most important phases of conformal radiotherapy (CRT) procedures. In the absence of a clear redefinition of the GTV, for a given pathology, complemented by detailed contouring procedures, the GTV are likely to be estimated rather arbitrarily with the risk of tumor underdosage or detriment to the surrounding healthy tissues. The objective of this study was to compare the delineation of the GTV of intrathoracic tumors by radiologists and radiation oncologists with experience in the field in various centers. The computed tomography images of ten patients with nonoperated non-small cell lung cancer (NSCLC) eligible for CRT were reviewed. Nine radiologists and eight radiation oncologists working in five different centers, classified as either 'junior' or 'senior' according to their professional experience, had to delineate the GTV (primary tumor and involved lymph nodes) with predefined visualization parameters. A dedicated software was used to compare the delineated volumes in terms of intersection and union volumes and to calculate the 'concordance index' for each patient and each subgroup of physicians. Significant differences between physicians and between centers were observed. Compared to radiation oncologists, radiologists tended to delineate smaller volumes and encountered fewer difficulties to delineate 'difficult' cases. Junior physicians, regardless of their specialty, also tended to delineate smaller and more homogeneous volumes than senior physicians, especially for 'difficult' cases. Major discordances were observed between the radiation oncologists' and the radiologists' delineations, indicating that this step needs to be improved. A better training of radiation oncologists in thoracic imaging and collaboration between radiation oncologists and radiologists should decrease this variability. New imaging techniques (image fusion, positron emission tomography, magnetic resonance imaging spectroscopy, etc.) may also provide a useful contribution to this difficult delineation.