2609 - Dosimetric Comparison between Photon and Proton Stereotactic Radiosurgery for Cerebral Arteriovenous Malformations: A Single Institution Experience

W. Sperduto¹, M. J. Flakus², S. A. Vora¹, E. L. Clouser¹, B. R. Bendok³, P. R. Lara¹, C. A. Stefan¹, and M. Bues¹; ¹Department of Radiation Oncology, Mayo Clinic, Phoenix, AZ, ²Mayo Clinic School of Health Sciences Medical Physics, Phoenix, AZ, ³Department of Neurosurgery, Mayo Clinic, Phoenix, AZ

Purpose/Objective(s): Stereotactic radiosurgery (SRS) is a well-established treatment for inoperable cerebral arteriovenous malformations (AVMs). Months to years following SRS, patients may develop radionecrosis – a local inflammatory reaction due to tissue death commonly associated with edema. Post-SRS symptomatic radionecrosis is a complication associated with AVM location and V12 (brain tissue volume receiving = 12 Gy). Due to the physical properties of protons, proton SRS may potentially reduce V12 and therefore the risk of symptomatic radionecrosis. With the development of proton SRS, there is growing interest in its utility to treat cerebral AVMs. The aim of this study was to quantify potential dosimetric advantages of proton-based SRS compared to conventional photon-based SRS in our first seven patients with cerebral AVMs. Materials/

Methods: We performed a retrospective review of proton SRS plans for seven adult patients with cerebral AVMs treated at our institution. Proton SRS was delivered via brass aperture enhanced spot scanning using multifield optimization. The cohort included two men and five women, and three patients received staged SRS due to location and volume. Five-field volumetric modulated arc therapy plans were generated for dosimetric comparison; the photon plan was normalized to match the clinical target volume prescription coverage of the delivered proton plan. To evaluate the quality of SRS plans, the Radiation Therapy Oncology Group conformity and homogeneity indices (CI and HI) were calculated. The optimal value for CI and HI is 1.0. To evaluate normal tissue sparing, the V12 of the total brain was reported.

Results: Patient, AVM, and treatment characteristics, and dosimetric parameters were recorded (Table). The 26-year-old patient already completed both SRS stages. CIs were comparable between protons and photons. All but one HI was improved with protons. V12 was lower for proton SRS for 75% of the cases. Interestingly, only for the two instances where photons achieved lower V12s were the AVMs superficially located near the skull and a 4 cm water-equivalent thickness range shifter was employed, which increased the spot size and lateral penumbra.
Conclusion: Proton SRS yielded an improved plan in terms of V12 in six of eight cases, including small volumes. There was, on average, a 14.68% improvement in V12 as compared to photon-based SRS. These seven adult patients will be followed with brain magnetic resonance imaging to assess for development of radionecrosis.

Age	Volume (cc)	Location	Dose (Gy)	Proton			Photon
				CI	HI	V12 (cc)	CI	HI	V12 (cc)
26	19.7	Temporoparietal	18	0.75 / 0.68	1.09 / 1.07	17.5 / 21.15	0.75 / 0.68	1.12 / 1.11	19.34 / 20.16
59	7.1	Temporooccipital	20	1.23	1.15	19.29	1.15	1.39	17.61
34	9.1	Thalami / Midbrain	20	1.29	1.18	21.96	1.32	1.35	30.60
51	0.4	Midbrain	20	1.58	1.17	2.60	1.28	1.12	2.93
23	8.3	Temporal	20	1.22	1.17	21.16	1.23	1.27	35.11
36	30.6	Frontoparietal	17	1.21	1.12	31.29	1.41	1.38	40.75
64	2	Cerebellar	20	1.35	1.10	7.35	1.06	1.23	9.18