Analysis of Local Control and Pain Control After Spine Stereotactic Radiosurgery Reveals Inferior Outcomes for Hepatocellular Carcinoma Compared With Other Radioresistant Histologies.

PURPOSE: This study aimed to evaluate the efficacy of stereotactic radiosurgery (SRS) for spinal metastases from hepatocellular carcinoma (HCC) compared with other radioresistant histologies (renal cell carcinoma [RCC], melanoma, and sarcoma) in terms of local control (LC) and pain control. METHODS AND MATERIALS: We performed a retrospective review of patients treated with SRS to the spine for metastatic HCC, RCC, melanoma, and sarcoma between January 2007 and May 2014. Radiographic assessments of LC, overall survival, and patient-reported pain control were analyzed as univariable analyses and with various patient- and treatment-related parameters as multivariable analyses (MVA). RESULTS: Of the 96 patients treated with SRS, 41 patients had radioresistant histologies, including 18 HCC, 1 mixed HCC and cholangiocarcinoma, 15 RCC, 6 melanoma, and 1 leiomyosarcoma. Extraosseous disease was present in 63% of patients (74% in HCC; 55% in non-HCC; P = not significant). Spinal cord compression was present in 29% of patients (32% in HCC; 27% in non-HCC; P = not significant), and 24% of patients had decompressive surgery before SRS (26% in HCC; 23% in non-HCC; P = not significant). With a median follow-up time of 8.7 months, the actuarial 3-, 6-, and 12-month LC rates were 71%, 61%, 41%, respectively, for HCC, and 94%, 94%, and 85%, respectively, for non-HCC. The median time to local failure was 3 months for HCC and 11 months for non-HCC. On MVA, there was a strong trend toward inferior LC with HCC (P = .059). Of the 28 patients with pretreatment pain, pain relief was achieved in 93% of patients, but the 2 patients who did not experience pain relief both had HCC. The actuarial 3-, 6-, and 12-month pain control rates were 68%, 51%, 17%, respectively, for HCC, and 100%, 89%, and 89%, respectively, for non-HCC (P = .023), and remained significant on MVA (P = .034). CONCLUSIONS: Compared with other radioresistant histologies, HCC has inferior LC and pain relief after SRS. Whether HCC may benefit from further dose escalation or combined treatment with new therapies is an area of future research.

Off-target-isocentric approach in non-coplanar Volumetric Modulated Arc Therapy (VMAT) planning for lung SBRT treatments.

PURPOSE: Volumetric Modulated Arc Therapy (VMAT) has emerged as an efficient alternative to traditional three-dimensional (3D) non-coplanar conformal (3D-NC-C) beams for lung cancer stereotactic body radiotherapy (SBRT) because of its superior dosimetric properties and native ease in planning and treatment delivery. However, patient immobilization in lung SBRT often presents challenging geometrical clearance issues in the execution of large (in excess of 180°) non-coplanar arcs. In this study, we present an off-target-isocentric, non-coplanar VMAT (OTI-NC-VMAT) technique that appears to be simple, dosimetrically robust and allows for ample patient/couch-gantry clearance. We compared this technique to a target-isocentric, non-coplanar VMAT (TI-NC-VMAT) technique and the 3D-NC-C beams for dosimetric evaluations. METHODS: Nineteen lung cancer patients previously treated with 3D-NC-C SBRT technique at our institution were selected. For each patient, an OTI-NC-VMAT plan and TI-NC-VMAT plan were created and compared to the original 3D-NC-C treatment plan. All of the plans were created for the same prescription dose of 54 Gy total in 3 fractions, covering 95% of the planning target volume (PTV). Nine to ten non-coplanar beams were used for the 3D technique and three non-coplanar arcs were used in both the TI-NC-VMAT and OTI-NC-VMAT plans, with the couch set at ± 20° and 0°, with each arc rotation in excess of 180°. Progressive Resolution Optimizer (PRO) in Varian Eclipse version 11 was used for all of the treatment planning. Conformity Index (CI), conformity number (CN), gradient index (GI), maximum dose at 2 cm away from the PTV (D2cm), mean lung dose (MLD), V20, V5 and mean target dose (MTD) were analyzed for all of the plans. We also performed statistical analysis to examine differences in the dosimetric indices between 3D and VMAT techniques. RESULTS: Dosimetric indices CI, CN, GI, V20 and MTD values were similar, within 5%, for all three plans: 3D-NC-C, TI-NC-VMAT and OTI-NC-VMAT. However, both types of VMAT plans were dosimetrically superior to 3D conformal plans in organ-at-risk (OAR) sparing; D2cm, MLD, and V5 values were significantly lower at 6-8%, 9-12% and 26-30% in VMAT plans, respectively. The OTI-NC-VMAT plans showed equivalent plan quality to the TI-NC-VMAT plans and exhibited robust freedom from limiting arc rotation due to potential patient/couch-gantry collision. CONCLUSIONS: The OTI-NC-VMAT plans appear dosimetrically equivalent to TI-NC-VMAT plans for lung SBRT, while permitting large angle arc selection, free from obstructional limitations. Both OTI-NC-VMAT and TI-NC-VMAT plans were dosimetrically superior to 3D-NC-C plans in terms of organ-at-risk (OAR) sparing.

Spatial variations of multiple off-axial targets for a single isocenter SRS treatment in Novalis Tx linac system.

PURPOSE: To develop a method to evaluate the positional variations of multiple off-axial targets for a single isocenter stereotactic radiosurgery (SRS) treatment in Novalis Tx linac system.Method and Materials: Five metallic ball bearing (BB) markers were placed sparsely in 3D off-axial locations (non-coplanar) inside a skull phantom as the representatives of multiple targets mimicking multiple brain metastases. The locations of the BB markers were carefully chosen to minimize overlapping of each other in a portal imaging detector plane. The skull phantom was immobilized by a frameless mask and CT scanned with a BrainLab Head&Neck Localizer using a GE Optima multi-detector CT (MDCT) scanner. The CT images were exported to iPlan treatment planning software and a multiple target PTV was drawn by combining all the contours of the BB markers. The margin of the MLC opening was selected as 3 mm expansion outward. Four non-coplanar arc beams were placed to generate a single isocenter SRS plan to treat the PTV. The skull phantom was localized to the treatment position using ExacTrac 6D Patient Positioning system. The four dynamic conformal arc beams were delivered using Novalis Tx system with portal imaging acquisition mode per 10% temporal resolution. The locations of the BB markers were visualized and analyzed with respect to the MLC aperture in the treatment plan similar to the Winston-Lutz (WL) test. RESULTS: All the BB markers were clearly identified inside the MLC openings. The total positional errors for the MLC aperture were 0.61 ± 0.2 mm along the rotational path of the four arcs. CONCLUSION: This study verified that the spatial deviations of multiple off-axial targets for a single isocenter SRS treatment plan is within sub-millimeter range in Novalis Tx linac system. Accompanied with the WL test, this simple test will quality-assure the spatial accuracies of the isocenter as well as the positions of multiple off-axial targets for the SRS treatment using a single isocenter multiple target treatment plan.