Apparatus-dependent Normal Tissue Dose For Radiosurgery Of Multiple Brain MetastasesDavid Larson1, LIJUN MA2, Igor Barani3, Arjun Sahgal4, Martina Descovitch5, Cynthia Chuang5, Sandeep Kunwar6, Paula Petti71Fremont, CA United States 2UCSF 3San Francisco, CA United States 4University of Toronto 5 University of California San Francisco 6 Washington Hospital 7Fremont, United States Keywords: dosimetry, brain metastasis, dose planning, gamma knife, radiosurgeryInteractive ManuscriptAsk Questions of this Manuscript: What is the background behind your study?The number of targets N considered clinically appropriate for radiosurgery continues to increase. Leksell Gamma Knife Perfexion( has made it practical to perform radiosurgery without sacrificing target dose conformity, even for large N. For large N it is unknown how normal tissue DVHs calculated with Cyberknife Multiplan 3.5 (CKMP3.5) compare with those calculated with Leksell GammaPlan 8.3 (LGP8.3). What is the purpose of your study?We addressed this issue by dose planning an increasing number of targets with both LGP8.3 and CKMP3.5 and comparing resultant DVHs. Describe your patient group.For the purposes of this study we chose a patient with metastases scattered throughout the cerebral and cerebellar hemispheres who had undergone Perfexion( radiosurgery. Describe what you did.We delineated 12 metastases (volume 0.06-1.0 cc) on the patient’s treatment-day CT scan using LGP8.3. We also delineated the pituitary, pituitary stalk, brainstem, optic nerves, optic chiasm, globes, lens, cochlea, and brain. The CT images, target volumes and normal structure volumes were transferred to CKMP3.5 for the planning comparison. We examined a subset of possible target combinations, with specific targets for N = 1, 3, 6, 9, and 12. For each plan we prescribed 16 Gy to each target and we placed dose constraints on all normal tissues except the brain. We calculated target coverage, target dose conformity, and DVHs for all delineated normal structures. The DVHs for normal brain were computed by subtracting the sum of the planned target volumes from the brain volume. Describe your main findings.All dose constraints were satisfied with both planning systems. However, CKMP3.5 calculations resulted in larger maximum doses to all normal structures, including normal brain. For N=12, for example, with total target volume of 5.4 cc, the normal brain volumes receiving =16 Gy, =12 Gy, =8 Gy, =4 Gy with LGP8.3 (CKMP3.5) were 2.1 (14.1) cc, 8.4 (31.5) cc, 22.4 (75.0) cc, 105.9 (275.4) cc. Describe the main limitation of this study.This is a retrospective study. Describe your main conclusion.Plans created for the Perfexion( with LGP8.3 for large numbers of brain metastases result in substantially lower normal intracranial tissue doses than plans created for the CyberKnife with CKMP3.5. Describe the importance of your findings and how they can be used by others.The number of radiosurgery targets that can be safely treated without exceeding a given normal brain dose tolerance is apparatus-dependent. The number of targets N considered clinically appropriate for radiosurgery continues to increase. Leksell Gamma Knife Perfexion( has made it practical to perform radiosurgery without sacrificing target dose conformity, even for large N. For large N it is unknown how normal tissue DVHs calculated with Cyberknife Multiplan 3.5 (CKMP3.5) compare with those calculated with Leksell GammaPlan 8.3 (LGP8.3). We addressed this issue by dose planning an increasing number of targets with both LGP8.3 and CKMP3.5 and comparing resultant DVHs. For the purposes of this study we chose a patient with metastases scattered throughout the cerebral and cerebellar hemispheres who had undergone Perfexion( radiosurgery. We delineated 12 metastases (volume 0.06-1.0 cc) on the patient’s treatment-day CT scan using LGP8.3. We also delineated the pituitary, pituitary stalk, brainstem, optic nerves, optic chiasm, globes, lens, cochlea, and brain. The CT images, target volumes and normal structure volumes were transferred to CKMP3.5 for the planning comparison. We examined a subset of possible target combinations, with specific targets for N = 1, 3, 6, 9, and 12. For each plan we prescribed 16 Gy to each target and we placed dose constraints on all normal tissues except the brain. We calculated target coverage, target dose conformity, and DVHs for all delineated normal structures. The DVHs for normal brain were computed by subtracting the sum of the planned target volumes from the brain volume. All dose constraints were satisfied with both planning systems. However, CKMP3.5 calculations resulted in larger maximum doses to all normal structures, including normal brain. For N=12, for example, with total target volume of 5.4 cc, the normal brain volumes receiving =16 Gy, =12 Gy, =8 Gy, =4 Gy with LGP8.3 (CKMP3.5) were 2.1 (14.1) cc, 8.4 (31.5) cc, 22.4 (75.0) cc, 105.9 (275.4) cc. This is a retrospective study. Plans created for the Perfexion( with LGP8.3 for large numbers of brain metastases result in substantially lower normal intracranial tissue doses than plans created for the CyberKnife with CKMP3.5. The number of radiosurgery targets that can be safely treated without exceeding a given normal brain dose tolerance is apparatus-dependent. Project Roles:
D. Larson (), L. MA (), I. Barani (), A. Sahgal (), M. Descovitch (), C. Chuang (), S. Kunwar (), P. Petti ()
|
