doi:10.1016/j.ijrobp.2006.03.050 CLINICAL INVESTIGATION Brain A PATHOLOGY-BASED SUBSTRATE FOR TARGET DEFINITION IN RADIOSURGERY OF BRAIN METASTASES BRIGITTA G. BAUMERT, M.D., PH.D.,* ISABELLE RUTTEN, M.D.,* † CARY DEHING-OBERIJE, M.SC.,* ALBERT TWIJNSTRA, M.D., PH.D., ‡ MIRANDA J. M. DIRX,PH.D., § RIA M. T. L. DEBOUGNOUX-HUPPERTZ,* PHILIPPE LAMBIN, M.D., PH.D.,* AND BELA KUBAT, M.D., PH.D. ¶ *Department of Radiation Oncology (MAASTRO), GROW, University Hospital Maastricht, Maastricht, The Netherlands; † Department of Radiotherapy, University Hospital (C.H.U.), Liege, Belgium; ‡ Department of Neurology, University Hospital Maastricht, Maastricht, The Netherlands; § Comprehensive Cancer Centre, Maastricht, The Netherlands;  Netherlands Forensic Institute (NFI), The Hague, The Netherlands; and ¶ Department of Pathology, Atrium Medical Center, Heerlen, The Netherlands Purpose: To investigate the need of a margin other than for accuracy reasons in stereotactic radiosurgery (SRS) of brain metastases by means of histopathology. Methods and Materials: Evaluation of 45 patients from two pathology departments having had brain metastases and an autopsy of the brain. Growth patterns were reviewed with a focus on infiltration beyond the metastases boundary and made visible with immunohistochemical staining: the metastasis itself with tumor-specific markers, surrounding normal brain tissue with a glial marker, and a possible capsule with a soft tissue marker. Measurements were corrected by a tissue-shrinkage correction factor taken from literature. Outcomes param- eters for infiltration were mean and maximum depths of infiltration and number of measured infiltration sites. Results: In 48 of 76 metastases, an infiltration was present. The largest group of metastases was lung cancer. Small-cell lung cancer (SCLC) and melanoma showed a maximum depth of infiltration of >1 mm, and other histologies <1 mm. For non–small-cell lung cancer (NSCLC), melanoma, and sarcoma, the highest number of infiltrative sites were observed (median, 2; range, 1– 8). SCLC showed significantly larger infiltrative growth, compared with other diagnostic groups. In NSCLC, the highest percentage of infiltration was present (70%). Conclusions: Infiltrative growth beyond the border of the brain metastasis was demonstrated in 63% of the cases evaluated. Infiltrative growth, therefore, has an impact in defining the clinical target volume for SRS of brain metastases, and a margin of 1 mm should be added to the visible lesion. © 2006 Elsevier Inc. Neuropathology, Radiosurgery, Clinical target volume, Target volume, Brain metastases. INTRODUCTION Stereotactic radiosurgery (SRS) of brain metastases is a high-dose single-fraction irradiation to a limited target vol- ume to eradicate the tumor cells and preserve surrounding normal tissue. The accurate definition of the lesion itself (gross tumor volume [GTV]) and the clinical target volume (CTV) is crucial. The CTV is defined as the lesion itself plus the inclusion of a possible infiltration beyond the visible tumor margin on imaging. There is no consensus about the definition or necessity of a CTV in SRS of brain metastases. Many centers include the contrast-enhancing tumor only (1, 2), and some add a margin of 1–2 mm under the hypothesis of possible tumor cells outside of the contrast- enhancing lesion (3–5). To date, there is one clinical com- parison published of local tumor control using different definitions for the CTV for SRS (6). These authors observed a significant improvement of local control if a 1-mm margin was added (6). The growth pattern of metastases can vary according to histology. A systematic study of cerebral metastases, focus- ing on tumor delineation in 66 patients, reported on 26 cases of anaplastic small-cell carcinomas, on one case of adeno- Reprint requests to: Brigitta Baumert, M.D., Ph.D., Department of Radiation Oncology, MAASTRO Clinic, Dr. Tanslaan 12, 6229 ET Maastricht, The Netherlands. Tel: (+31) 88-55-66666; Fax: (+31) 88-55-66667; E-mail: brigitta.baumert@maastro.nl Presented at the 23rd ESTRO (European Society for Therapeu- tic Radiology and Oncology) Meeting, Amsterdam, The Nether- lands, October 24 –28, 2004. Acknowledgments—We gratefully acknowledge the pathologists of the Pathology Departments of the University Hospital Maas- tricht, Dr. Robert Jan Van Suylen and Dr. Marius Nap of the Atrium Medical Center, Heerlen, The Netherlands, for their sup- port and for providing histopathologic material without which this study would not have been possible. Received Jan 8, 2006, and in revised form March 20, 2006. Accepted for publication March 23, 2006. Int. J. Radiation Oncology Biol. Phys., Vol. 66, No. 1, pp. 187–194, 2006 Copyright © 2006 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/06/$–see front matter 187