PHYSICS CONTRIBUTION TREATMENT PLANNING STUDY TO DETERMINE POTENTIAL BENEFIT OF INTENSITY-MODULATED RADIOTHERAPY VERSUS CONFORMAL RADIOTHERAPY FOR UNRESECTABLE HEPATIC MALIGNANCIES CYNTHIA L. ECCLES, B.SC.,* JEAN-PIERRE BISSONNETTE,PH.D.,* TIM CRAIG,PH.D.,* MOJGAN TAREMI, M.D.,* XIA WU,PH.D., AND LAURA A. DAWSON, M.D.* * Radiation Medicine Program and Department of Radiation Oncology, Princess Margaret Hospital, University of Toronto, Toronto, ON, Canada Purpose: To compare intensity-modulated radiotherapy (IMRT) with conformal RT (CRT) for hypofractionated isotoxicity liver RTand explore dose escalation using IMRT for the same/improved nominal risk of liver toxicity in a treatment planning study. Methods and Materials: A total of 26 CRT plans were evaluated. Prescription doses (24–54 Gy within six fractions) were individualized on the basis of the effective liver volume irradiated maintaining #5% risk of radiation- induced liver disease. The dose constraints included bowel (0.5 cm 3 ) and stomach (0.5 cm 3 ) to #30 Gy, spinal cord to #25 Gy, and planning target volume (PTV) to #140% of the prescribed dose. Two groups were evaluated: (1) PTV overlapping or directly adjacent to serial functioning normal tissues (n = 14), and (2) the liver as the dose- limiting normal tissue (n = 12). IMRT plans using direct machine parameter optimization maintained the CRT plan beam arrangements, an estimated radiation-induced liver disease risk of 5%, and underwent dose escalation, if all normal tissue constraints were maintained. Results: IMRT improved PTV coverage in 19 of 26 plans (73%). Dose escalation was feasible in 9 cases by an av- erage of 3.8 Gy (range, 0.6–13.2) in six fractions. Three of seven plans without improved PTV coverage had small gross tumor volumes (#105 cm 3 ) already receiving 54 Gy, the maximal prescription dose allowed. In the remaining cases, the PTV range was 9.6–689 cm 3 ; two had overlapped organs at risk; and one had four targets. IMRT did not improve these plans owing to poor target coverage (n = 2) and nonliver (n = 2) dose limits. Conclusion: Direct machine parameter optimization IMRT improved PTV coverage while maintaining normal tis- sue tolerances in most CRT liver plans. Dose escalation was possible in a minority of patients. Ó 2008 Elsevier Inc. Liver radiotherapy, Intensity-modulated radiotherapy, IMRT, Conformal radiotherapy, generalized equivalent uniform dose, gEUD. INTRODUCTION For patients with unresectable primary and metastatic liver cancer refractory to treatment, including chemotherapy and radiofrequency ablation, focal liver radiotherapy (RT) can be used as a treatment option (1–5). A dose–response relationship has been observed, with greater doses associ- ated with improved outcome (6–8). However, dose escala- tion for large liver cancers is challenging, because radiation-induced liver disease (RILD) can occur when the whole liver dose is >30 Gy in 2-Gy fractions (9, 10). Addi- tionally, safe partial liver RT can be limited by the proximity of visceral organs at risk (OARs) such as the stomach and duodenum. Technological advancements such as three-dimensional treatment planning, image-guided RT and breathing motion management have facilitated the safe use of highly conformal RT (CRT) in patients with focal liver cancer. Conformal, forward-planned, segmentally modulated treatment plans with beam arrangements to avoid OARs and minimize path lengths through uninvolved liver can achieve reduced integral liver doses and allow greater doses to be delivered to the planning target volume (PTV). Beam modulation from manual beam segmentation allows the dose gradients Reprint requests to: Laura A. Dawson, M.D., Department of Ra- diating Oncology, Princess Margaret Hospital, 610 University Ave., Toronto, ON M5G 2M9 Canada; Tel: (416) 946-2124; Fax: (416) 946-5666; E-mail: laura.dawson@rmp.uhn.on.ca Presented in part the European Society of Therapeutic Radiology and Oncology, Leipzig, Germany, October 2006. Patients evaluated in this manuscript were treated on clinical pro- tocols funded in part by the National Cancer Institute of Canada (grant 018207), the Canadian Cancer Society, and Elekta Oncology Systems. Conflict of interest: none. Acknowledgments—The authors gratefully acknowledge Colleen Dickie Euler and Maria Hawkins for their contributions to this research. Received April 28, 2008, and in revised form June 12, 2008. Accepted for publication June 15, 2008. 582 Int. J. Radiation Oncology Biol. Phys., Vol. 72, No. 2, pp. 582–588, 2008 Copyright Ó 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/08/$–see front matter doi:10.1016/j.ijrobp.2008.06.1496