Inr J Hodmr on Oncology Em Phys Vol 8, pp lb I l- lb23 036CL3016/ 82/ 091617471603.00/ 0 Pnnted m the U S A All rights reserved Copyright 0 1982 Pergamon Press Ltd. ??zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Technical Innovations and Notes zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA ADVANTAGES AND LIMITATIONS OF COMPUTED TOMOGRAPHY SCANS FOR TREATMENT PLANNING OF LUNG CANCER JOAQUIN zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA G. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA MIRA, M.D., * JANET L. POTTER, M.D., PH.D.,? GARY D. FULLERTON, PH.D.* AND JOAN EZEKIEL, M.D.8 Department of Radiology, University of Texas Health Science Center at San Antonio, San Antonio. TX Forty-five Chest computed tomography (CT) scans performed on patients with lung carcinoma (LC) were evaluated in an attempt to understand the pattern of intrathoracic tumor spread and the advantages and limitations this technique offers for treatment planning when compared to planning done by conventional X rays. The following findings can help treatment planning. (1) When regular X rays do not show tumor location (i.e., hemithorax opacification), CT scan will show it in 68% of patients. If regular X rays show a well localized mass, unsuspected tumor extensions were disclosed in 78 % of these patients. Hence, CT scans should be done in all LC patients prior to treatment planning; (2) Mediastinal masses frequently spread anteriorly toward the sternum and posteriorly around the vertebral bodies toward the cord and costal pleura. This should be considered for radiotherapy boost techniques; 13) Lung masses spread in one third of cases toward the lateral costal pleura. Thus, the usual l-2cm of safety margin around the LC are not sufficient in some cases; (4) Tumor size can appear much smaller in regular X rays than in CT scans. Hence, CT scans are necessary for accurate staging and evaluation of tumor response. Some CT scan limitations are: (1) Atelectasis blends with tumor in approximately half of the patients, thus obscuring tumor boundaries.; (2) CT numbers and contrast enhancement did not help to differentiate between these two structures; and (3) Limited definition of CT scan prevents investigation of suspected microscopic spread around tumor masses. Computed tomography scans, Lung carcinoma. INTRODUCTION The design of proper radiotherapy ports for unresectable lung carcinoma (LC) can be a very frustrating task. Concomitant atelectasis, infiltration and pneumonic pro- cesses that frequently surround these neoplasms can make the tumor location and boundaries impossible to delineate by conventional chest X rays. In our experience, tomograms are usually not helpful in these circumstances. We have already described the high frequency of relapses outside the radiotherapy field in small cell lung carci- noma,14 indicating perhaps, the inability by the radiother- apist to know actual tumor boundaries and to design proper radiotherapy ports. Hence, when a whole body computed tomography (CT) scanner became available to us in 1978, we embarked in this study hoping the new technique would solve these problems. At that time some articles had already discussed the possible advantages of CT scan for radiotherapy planning,‘~“~20 although no information was available about its application for LC. Since then a few articles have described how CT scans can show lung tumor extensions unsuspected by conven- tional diagnostic methods, and the frequent need to change the treatment planning done according to regular X rays because of these new findings.h~‘O.‘x~” In our work we have tried to understand the parameters and tumor characteristics that would influence these unsuspected tumor extensions. We have classified the patterns of intrathoracic tumor spread and attempted to clarify the reasons why conventional diagnostic X rays fail to delineate tumor properly. ‘We will point out not only possible advantages of CT scanning but also the limitations we have encountered. METHODS AND MATERIALS We reviewed 45 chest CT scans performed from June 1978 to September 1981, in patients sent to us for initial radiotherapy evaluation. All patients had biopsy proven carcinoma of the lung except for one patient with adeno- *Associate Professor, Division of Radiation Oncology. Klinical Assistant Professor, Division of Diagnostic Radiolo- gy. *Assistant Professor, Division of Physics. $Clinical Associate Professor, Division of Diagnostic Radiol- ogy. This investigation was supported in part by Grants NIH- R-R0564, NIH 789-l4-OSOand NIH 5RlOCA22433 Reprint requests to: Joaquin G. Mira. M.D., Department of Radiology, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio. TX 78284. This work was presented at the 15th International Congress of Radiology. Brussels. Belgium, June I98 I. Accepted for publication 21 April 1082. 1617