RADIATION THERAPY

doi:10.1093/neuonc/not187

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-002. PREDICTING VOXEL OUTCOMES IN PATIENTS WITH GBM TREATED WITH ANTI-ANGIOGENIC AGENTS USING INTEGRATED DOSIMETRY, METABOLIC AND PHYSIOLOGIC IMAGING

Mekhail Anwar ¹, Janine Lupo ¹, Annette Molinaro ¹, Jennifer Clarke ¹, Nicholas Butowski ¹, Michael Prados ¹, Susan Chang ¹, Daphne HaasKogan ¹, Sarah Nelson ¹

Abstract

Treatment of GBM with anti-angiogenic agents (AA) is changing the paradigm of patterns of recurrence, with tumor often recurring in a non-enhancing, diffusely invasive phenotype. Despite this, delineation of voxels at risk for progression for radiotherapy (RT) is purely a geometric expansion of anatomic (T1,T2) MRI. We hypothesize that by integrating spectroscopic, diffusion- and perfusion-weighted parameters that can identify voxel-level subclinical disease at diagnosis with planned dosimetry, the likelihood of voxel progression at radiographic recurrence (RR) can be predicted. Twenty-four patients with GBM treated with 60 Gy RT, temozolamide, and an AA (enzastaurin (N = 11) or bevacizumab (N = 13)), with dosimetry and imaging at baseline and RR were analyzed. Pre-treatment diffusion (ADC,FA), perfusion (rCBV,%Recovery), and spectroscopy (choline-to-NAA index:CNI) of voxels within the lesion (3654 voxels, 274 cc) or normal brain (19664 voxels, 1475 cc) were calculated and a multinomial logistic regression model (MLR) integrating dosimetry and distance from tumor was generated to predict voxel outcome (stable, improved or progressed). 90% of voxels recurred in the non-enhancing lesion (23% >2 cm from the baseline lesion). Decreasing FA and distance, increasing ADC and CNI, and dose were highly significant in predicting voxel progression. This is consistent with subclinical involvement showing decreased structure, increased diffusion and metabolism, and proximity to the baseline lesion. For voxels within the baseline lesion that improved at RR, both univariate and multivariate analysis showed decreased significance for FA, ADC, and CNI, but increased significance for rCBV and %Recov. Dose remained highly significant. MLR models predicting progression correctly identified 84% of voxels [range 47-95] compared to 65% [29–87] when using 2 cm outside the lesion as a cutoff. MLR correctly identified 68% [13–95] of voxels within the lesion that improved. Integrating physiologic and metabolic MRI with dosimetry can identify voxels at risk for progression and improvement, allowing for risk-adapted voxel-level RT planning.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-003. INTRACTABLE SEIZURES SECONDARY TO MENINGIOANGIOMATOSIS CONTROLLED WITH SINGLE FRACTION STEREOTACTIC RADIOSURGERY

Jonathan Ashman ¹, Joseph Drazkowski ¹, Richard Zimmerman ¹, Thomas Lidner ¹, Caterina Giannini ², Alyx Porter ¹, Naresh Patel ¹

Abstract

BACKGROUND: Meningioangiomatosis (MA) is a rare, benign lesion associated with intractable seizures in young adults. Surgery is the primary diagnostic and therapeutic intervention. An atypical case is presented in which only subtotal resection could be achieved, and long-term seizure control was obtained after single fraction stereotactic radiosurgery (SRS). METHODS: The electronic medical record was reviewed retrospectively including imaging, surgical reports, pathology slides, and radiation treatment planning. SRS was delivered using a stereotactic head-frame and a linear accelerator-based technique with 5mm multileaf collimators. RESULTS: A 17 year old male presented in 2006 with a 2-year history of intractable epilepsy. Magnetic resonance imaging (MRI) demonstrated an unusual lesion in the left temporal lobe with homogeneous gadolinium enhancement and decreased signal on T2-weighted images. Diffuse calcification of the lesion was appreciated by computed tomography (CT). Only partial resection could be achieved because of dense adhesions to multiple middle cerebral artery vessels. Histopathologically, the lesion demonstrated perivascular meningothelial proliferation consistent with MA. The lesion remained radiographically stable for 4 years, but seizures continued despite medical management. Single-fraction SRS was performed in December, 2010 to a dose of 2200 cGy prescribed to the 77% isodose line. MRI imaging at 3 months post-SRS demonstrated reduction in the lesion size, which then remained stable on subsequent imaging. Seizure activity abated approximately 1 month post-SRS, and the patient has now remained seizure-free up to the present day. DISCUSSION: To our knowledge, this is the first case reporting SRS as a successful treatment of seizures induced by MA. SRS has been used previously to treat mesial temporal lobe epilepsy (MTLE). Thus far, no SRS-associated toxicity has been identified in the current case, but delayed radiation necrosis has been reported after SRS for MTLE. Therefore, although this patient has remained seizure-free for over 2 years, continued follow-up is warranted.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-004. PREDICTING THE INTELLECTUAL OUTCOME IN CHILDREN AND ADOLESCENT WITH BRAIN TUMORS TREATED WITH MODERN RADIATION TECHNIQUES: A POPULATION-BASED STUDY

I Atean ^1,², N Shin ³, A Toltz ⁴, C Laude ³, C Freeman ³, J Seuntjens ⁴, D Roberge ¹

Abstract

INTRODUCTION: Brain cancer is uncommon but devastating in young patients. The deleterious cognitive effects of radiation are well documented. Proton radiotherapy promises reduced late effects through better tissue sparing. This has been modeled for selected cases, but not in wider, less biased, populations. Without data from trials comparing radiotherapy techniques, formulae have been proposed correlating radiation dosimetry to cognitive outcomes. METHODS AND MATERIALS: In a province-wide study, all adolescent-young-adult patients treated with photon radiotherapy in Quebec during a 6-month period of 2010 were identified. Of this 114 patient cohort, 19 were treated as children for CNS tumors. After excluding radiosurgery patients, it was possible to replan 10 IMRT cases for intensity modulated proton therapy (IMPT). These patients, aged 2–17, were treated for varied benign and malignant tumors (medulloblastoma, glial and germ cell tumors) to 50.4–59.4 Gy. IMRT and IMPT dose-volume data were applied in 4 published models. Statistical significance was tested using paired t-tests. RESULTS: Compared to photons, IMPT planning predicted modest and variable improvements in the intermediate dose-volumes associated with cognition: mean reduction in whole brain V35-65 of 2% (range -8 to 10, p = 0.02), mean reduction in supratentorial brain V35-65 of 2% (range -1 to 6, p = 0.04), mean reduction in temporal lobe V40 of 4% (range -40 to 38, p = 0.4). All models predicted that IMPT would lessen the impact of treatment on IQ by an average of 2-4 points at 5 years. Unrealistically, post-treatment IQ improvement was often predicted especially for older children. CONCLUSION: Our results suggest that increasing use of proton therapy will likely improve cognitive outcomes in young brain tumor patients. Current models may not reliably predict the magnitude of this improvement underlining the need to collect more data, particularly in patients treated as adolescents and with a wider range of diagnoses and radiation techniques.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-005. WIDE-FIELD RE-IRRADIATION WITH BEVACIZUMAB IS A FEASIBLE SALVAGE OPTION FOR PATIENTS WITH REFRACTORY HIGH GRADE GLIOMA

Michael Back ^1,², Marina Kastelan ¹, Linxin Guo ¹, Helen Wheeler ^1,²

Abstract

AIM: Clinical studies of re-irradiation (ReRT) for relapsed high grade glioma (HGG) have generally reported use of small volume ReRT techniques such as stereotactic radiosurgery in selected patients with isolated focal relapse. This study reports the outcome with Wide-Field Re-irradiation (WFReRT) to manage the more common scenario of extensive diffuse relapse of HGG. METHODS: All HGG patients managed with an overlapping second course of RT for symptomatic refractory progression of HGG between June 2010-December 2012 were included. All patients had WFReRT with IMRT and specific RT dosimetry factors were analysed including including the target volume treated (PTV), cumulative RT dose maximum (Dmax) and volume of brain receiving cumulative 80Gy (V80) and 90 Gy (V90). Timing of RT in relation to initiation of bevacizumab (Bev) use was initially at time of progression on Bev; and then later in recent experience at time of Bev commencement. Overall survival post WFReRT was calculated in months using Kaplan-Meier in SPPS v19 software. RESULTS: 15 patients with HGG received WFReRT with initial pathology glioblastoma in 11 patients and anaplastic glioma in 4 patients. All had evidence of high grade relapse with bulky contrast enhancing disease at time of WFReRT. Bev was used in 12 patients, with ReRT starting post Bev failure in 8 and with Bev in 4 patients. WFReRT dose was 35-40Gy in 15 fractions; median PTV was 75cm3 (12–313 cm³); median Dmax was 95Gy (67–112Gy); and median V80 and V90 were 38.1 cm³ and 28.2 cm³. Median survival post WFReRT was 43 weeks (95%CI:22-64]. One patient treated without Bev required craniotomy for radiation necrosis at 6 weeks post RT. No other significant morbidity was reported. Survival in relation to use of BEV was not significantly different in this small patient cohort. CONCLUSION: WFReRT combined with Bevacizumab is a feasible salvage treatment option for diffuse refractory HGG.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-006. TEMOFRAC, AN INNOVATIVE AND NOVEL IRRADIATION SCHEDULE, CONCURRENT 3-TIMES DAILY ULTRAFRACTIONATED RADIATION THERAPY AND TEMOZOLOMIDE FOR NEWLY INOPERABLE GLIOBLASTOMA

Patrick Beauchesne ¹, Guillaume Faure ², Georges Noel ³, Thierry Schmitt ⁴, Laurent Martin ⁵, Eric Jadaud ⁶, Charlotte Carnin ¹

Abstract

BACKGROUND: Ultrafractionation radiation therapy consists in irradiating cells or tumours several times daily, delivering low doses at which hyperradiosensitivity occur. We recently reported the efficiency of ultrafractionation radiotherapy regimen in newly inoperable glioblastoma. We are now conducting a phase II clinical trial to determine the effect of a concurrent ultrafractionation regimen and temozolomide for inoperable glioblastoma patients. METHODS: A prospective, multicenter, phase II study has opened for accrual in February 2008. Patients over 18 years of age who are able to give informed consent and have histologically proven, newly inoperable diagnosed and supratentorial glioblastoma are eligible. Three doses of 0.75Gy spaced by at least four hours are delivered daily, five days a week for six consecutive weeks for a total of 67.5Gy, and concomitant chemotherapy consisted of temozolomide given at dose of 75mg/m², 7 days per week during the ultrafractionated radiotherapy. After a 4-week break, chemotherapy is resumed up to 6 cycles of adjuvant temozolomide every 28 days, according to the standard 5-day regimen. Tolerance and toxicity is the primary endpoints; survival and progression- free survival are secondary endpoints. RESULTS: To date 42 patients have been enrolled in this study, 30 men and 12 women, median age 58, median Karnofsky performance status was 80. The concomitant ultrafractionated radiotherapy – temozolomide was been well tolerated; no acute grade 3 and/or 4 CNS toxicity has been observed. Complete response was reported in 4 patients, and stabilisation or minor responses in 12 patients. Median survival from initial diagnosis was 15 months, and overall survival at 24 months was 26 %.  An analysis of the status of MGMT was performed. CONCLUSIONS: TEMOFRAC regimen is safe and well tolerated. It may prolong the survival of patients with glioblastoma. Updated definitive results will be presented at the meeting. 

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-007. PALLIATION OF SACRAL SPINE METASTASES WITH STEREOTACTIC BODY RADIATION THERAPY: AN EVALUATION OF LOCAL CONTROL AND TOXICITY

J Bowers ¹, N Bennion ¹, H Lomas IV ¹, K Spencer ¹, M Richardson ¹, W McAllister ³, J Sheehan ⁴, D Schlesinger ⁴, R Kersh ^1,²

Abstract

OBJECTIVES: There is significant evidence that Stereotactic Body Radiation Therapy (SBRT) for spinal metastases provides excellent palliation in well selected patients. The authors examine a series of patients with sacral spine metastasis treated with SBRT with attention to local control and toxicity. METHODS: Using our institutional database we collected data from 253 cases of spinal radiation for metastases from August 2007 to May 2012. We identified 18 patients with 19 metastatic lesions to the sacrum followed for a mean of 12.8 months. Ten lesions (53%) treated with radiotherapy primarily and 9 lesions (47%) previously received conventional external beam radiotherapy. Primary indication for treatment was pain; other indications were neurologic complaints and oligo-metastasis, with each accounting for 16, 6 and 3 respectively. Median prescribed dose was 24Gy ranging from 35Gy to 50Gy. Cone beam CT was used for image guidance and treatments were delivered using IMRT. Immobilization was attained using BodyFIX system. Patient follow-up involved history and physical with PET, CT and/or MRI imaging to determine local control. Toxicity analyzed using CTCAEv4. RESULTS: After treatment, local control was 94.7%. Pain improved in 84% of patients, with 16% reporting stable pain, and no patients reporting increased pain. Neurological symptoms stabilized in 5 patients while 1 reported progressive symptoms. Toxicity was minimal with only 2 patients experiencing grade 1 acute pain. No reported cases of acute radiation dermatitis, sacral fracture, progressive fractures, GI symptoms, rectal symptoms or plexopathy. In our sub-group analysis, reirradiated population had a nearly identical response to therapy as our initial SBRT group. CONCLUSION: SBRT offers excellent local control with minimal toxicity for patents with spinal metastasis to the sacrum. Patients tolerated the treatment regimen well with no high grade toxicity. Responses to therapy and toxicity profile of previously irradiated lesions were similar to those receiving first time treatment.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-008. CRANIO-SPINAL IRRADIATION FOR MEDULLOBLASTOMA: DOSIMETRIC COMPARISON OF PROTON AND PHOTON MODALITIES TO REDUCE THE LATE DEVELOPMENT OF DIABETES MELLITUS

Jeffrey Brower ^1,², Shae Gans ¹, William Hartsell ¹, Stewart Goldman ^1,³, John Han-Chih Chang ¹, Nasiruddin Mohammed ¹, Moe Siddiqui ¹, Vinai Gondi ^1,²

Abstract

BACKGROUND: Induction of diabetes mellitus (DM) has never previously been identified as a long-term toxicity of cranio-spinal irradiation (CSI) for medulloblastoma. A recent study assessing survivors of various childhood malignancies demonstrated that radiotherapy dose >10 Gy to the pancreatic tail predicted an 11-fold increased risk of DM, likely attributable to the radiosensivity of pancreatic islet cells. Based on these findings, we sought to delineate the pancreatic dose utilizing conventional photon therapy (3DCRT) and the differential capacity to spare the pancreas using modern helical tomotherapy (HT) and proton beam therapy (PBT). METHODS: Five average-risk medulloblastoma patients who received CSI to 23.4 CGE using PBT at a single institution were identified. 3DCRT plans were generated using opposed lateral cranial fields and posterior spinal fields. Pancreas and pancreatic tail were delineated on the planning CT scan, and multiple HT plans were developed for each patient. HT plans delivering the lowest dose to pancreas and pancreatic tail without compromising target coverage were selected. RESULTS: All three modalities covered the planned target volume with the 95% isodose line. As compared to 3DCRT and HT, PBT was associated with significantly reduced pancreatic volume receiving ≥ 10 Gy and lower mean dose to the pancreas and pancreatic tail (p < 0.001 for all comparisons). For all patients, maximum dose to the pancreatic tail exceeded 20 Gy with 3DCRT, but remained below 10 Gy with PBT. Maximum dose to the pancreatic tail using HT exceeded 10 Gy in 1 of 5 patients. CONCLUSION: These dosimetric findings demonstrate that radiotherapy-related induction of DM may be an underappreciated toxicity of conventional photon CSI. Compared to HT, PBT provides superior pancreatic sparing due to its distal dose fall-off. Thus, PBT for CSI may be associated with the lowest risk of DM in pediatric medulloblastoma survivors.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-009. OUTCOMES IN STEREOTACTIC BODY RADIOTHERAPY FOR SPINAL TUMORS: AN UPDATED MD ANDERSON EXPERIENCE WITH PATTERNS OF FAILURE ANALYSIS

Eva Christensen ¹, Slade Klawikowski ¹, Amit Garg ¹, Mary McAleer ¹, Laurence Rhines ¹, James Yang ¹, Paul Brown ¹, Eric Chang ^1,², Stephen Settle ¹, Amol Ghia ¹

Abstract

PURPOSE: To investigate outcomes from all patients treated with stereotactic body radiotherapy (SBRT) for spinal tumors at our institution. MATERIALS/METHODS: In all, 368 consecutive patients with 446 individual spinal tumors received SBRT from 2002 - 2012. Near simultaneous CT-guided SBRT was delivered in 1-5 fractions. The most common one (n = 200), three (n = 191), and five fraction (n = 51) doses were 24 Gy (n = 93, range 12–24 Gy), 27 Gy (n = 186, range 24–27 Gy) and 30 Gy (n = 46, range 20–30 Gy), respectively. Single fraction treatment was delivered in radiation naïve patients while multifraction regimens were usually reserved for re-irradiation cases. Spinal imaging reports were reviewed for progression of disease (POD). POD was categorized as: local- POD at the treated site; marginal- POD adjacent to the treated site; field-associated epidural- epidural POD at the level of the treated site. RESULTS: Actuarial 1-year, 3-year and 5-year overall survival (OS) were 67.0%, 33.4%, and 16.8% (median 22.7 months). At 1, 3 and 5 years the actuarial local control and marginal control were 81.8%, 64.8% and 58.3%, and 89.6%, 78.4%, and 75.6%. Median tumor control was 80.6 months. For the 143 tumors that progressed, the median time to tumor progression was 12.6 months. OS was significantly greater with 1 or 5 fractions compared to 3 fractions (median 25.7, 23.5 and, 19.9 months, p = 0.02). Regional, but not local, failure free survival was significantly greater with the use of 1 fraction compared to 3 or 5 fractions (RFS at 1 year 93.3% vs. 85.8% and 89.5%, p = 0.0138). Of the 112 sites of in-field POD, 83 (74.1%) had field-associated epidural POD. CONCLUSION: While stereotactic body radiotherapy for spinal tumors provides effective local control, many failures occur in the epidural space. Current studies assessing adverse events and dose distributions might reveal opportunities to improve tumor control.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-010. OUTCOMES IN PATIENTS WITH PAPILLARY TUMORS OF THE PINEAL REGION: SINGLE INSTITUTIONAL EXPERIENCE

Mark Edson ¹, Gregory N Fuller ¹, Pamela Allen ¹, Jing Li ¹

Abstract

BACKGROUND: Papillary tumor of the pineal region (PTPR) is a rare neuroectodermal tumor first described in 2003 and formally codified by the WHO in 2007. Limited reports suggest surgical resection is the mainstay of treatment; however, the role of multimodality therapy is not well defined. We evaluated our institutional experience in the treatment of PTPR. METHODS: Nine patients with pathologically confirmed PTPR diagnosed between 1999 and 2013 were retrospectively reviewed. RESULTS: The median age at diagnosis was 37 (range 25–56). After a median follow-up of 44.8 months (range 4.1-148.1), seven of nine patients were still living. All patients underwent maximum safe surgical resection. Gross total resection (GTR) was achieved in three cases (33%). Whereas five (55%) patients received adjuvant radiotherapy (RT), three (33%) received RT for recurrence, and one (11%) was not treated with RT. In the adjuvant setting, RT techniques consisted of stereotactic radiosurgery or proton RT to the surgical bed, craniospinal irradiation with a tumor bed boost, or whole ventricular RT, followed by a tumor bed boost. Only one of five patients who received adjuvant RT demonstrated progressive disease. Overall and progression free survival at 12 and 24 months were 100% and 83.3%, and 85.7% and 71.4%, respectively (median OS 83.6 mo, PFS 34.6 mo). Of the four patients with disease progression one died; one had previously undergone GTR and was doing well two years after salvage RT; one never received RT; and one received salvage RT for local and subsequently leptomenigeal recurrences. The other death was secondary to suicide. CONCLUSIONS: Our institutional experience confirms a recent multicenter retrospective series showing high risk of recurrence for PTPR. However, our findings suggest RT administered in an adjuvant setting or at time of recurrence appears to provide durable local control.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-011. TUMOR LOCATION PREDICTS FOR LOCAL CONTROL OF BRAIN METASTASES FROM NON-SMALL CELL LUNG CANCER TREATED WITH STEREOTACTIC RADIOSURGERY

Adam Garsa ¹, Shahed Badiyan ¹, Joseph Simpson ¹, Joshua Dowling ², Keith Rich ², Michael Chicoine ², Eric Leuthardt ², Albert Kim ², Clifford Robinson ¹

Abstract

PURPOSE: To evaluate the local control rate of brain metastases from non-small cell lung cancer (NSCLC) treated with stereotactic radiosurgery (SRS), and to determine predictors of local progression. METHODS: Between June 1998 and May 2011, 401 lesions in 229 patients were treated with Gamma Knife single-fraction SRS, and their medical records were retrospectively reviewed. Eligible patients had at least one follow-up brain magnetic resonance imaging (MRI). Brainstem metastases were excluded from analysis. Local progression was defined as any increase in lesion size after SRS. Tumor control was defined as a decrease in size or no change in size after SRS. Tumor location was categorized as supratentorial or cerebellar. Local control was estimated using the Kaplan-Meier method. Cox proportional hazards models were used for univariate and multivariate analysis. A p-value < 0.05 was considered statistically significant. RESULTS: The median age was 60 years (range 27–84). There were 66 cerebellar metastases (16%) and 335 supratentorial metastases (84%). 42% of patients received prior whole brain irradiation. The median prescription dose was 20Gy (range 14–24Gy). The median prescription isodose was 50% (range 45–90%). The median overall survival from time of SRS was 10.5 months (range 1.3–115). The estimated local control (LC) at 12 months was 72.6%. The estimated LC at 12 months for cerebellar lesions was 58.9%, compared to 75.6% for supratentorial lesions (log-rank p = 0.0235). On univariate analysis, cerebellar location, tumor volume, dose, and conformity index (CI) were significant predictors of local progression. On multivariate analysis, cerebellar location (hazard ratio (HR) 1.851 (95% confidence interval (CI) 1.126-3.042) p = 0.0152), tumor volume (cm³) (HR 1.06 (95% CI 1.005-1.119), p = 0.0331), and CI (HR 0.685 (95% CI 0.471-.994), p = 0.0466) remained significant predictors of local progression. CONCLUSIONS: Cerebellar tumor location, larger tumor volume, and lower conformity index were significant independent predictors of local progression following SRS for brain metastases from NSCLC.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-012. HYPOFRACTIONATED RE-IRRADIATION OF RECURRENT ENHANCING GLIOMAS WITH 35 Gy: CAN 20 Gy BE DELIVERED TO SURROUNDING T2 EDEMA SAFELY?

Beant Gill ¹, Dawn Peskorski ¹, Ron Lalonde ¹, M Saiful Huq ¹, John Flickinger ¹

Abstract

OBJECTIVE: High grade gliomas frequently recur within the field of prior irradiation. Published series of hypofractionated re-irradiation of gliomas, as well as a current RTOG protocol, neglect targeting surrounding T2 edema that can harbor microscopic disease to limit potential toxicity. We investigated the dosimetric impact of 20 Gy dose-shaping to T2 edema. METHODS: Eleven consecutive patients with recurrent high grade gliomas based on radiographic and clinical findings were included. All received full dose (59.4-60 Gy) involved field radiotherapy previously. Hypofractionated re-irradiation was delivered with 35 Gy in 10 fractions to enhancing disease (PTV35) with intensity modulated radiotherapy (IMRT). We created plans with and without a simultaneous secondary prescription of 20 Gy in 10 fractions to surrounding T2 edema (PTV20) for dosimetric comparison. Additionally, we calculated brain V20 (volumes receiving 20 Gy) for comparison. RESULTS: Median PTV35 and PTV20 values were 41.5 cc (range, 6.6–87.0) and 81.9 cc (range, 27.1–261.7), respectively. The mean ratio between PTV20 (surrounding T2 signal) and PTV35 (enhancement) was 2.39. The mean incidental V20 without T2 dose-shaping was 162.6 cc, which was similar to the mean targeted total 20 Gy volume (PTV35 + PTV20) of 129.3 cc, but resulted in a slightly larger mean shaped-V20 of 279.1 cc. Mean incidental PTV20 coverage was 44%, which improved with shaping to 99.2%. Dose-shaping increased brain V20 by a mean factor of 1.4 (range, 0.8-6.6). CONCLUSION: Secondary 20 Gy T2 dose-shaping with IMRT can be done with a relatively modest increase in V20 for most patients over what would be expected from targeting enhancing tumor alone to 35 Gy. Whether or not 20 Gy to surrounding T2 regions can improve tumor control and how much it might increase toxicity seems reasonable to investigate.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-013. A REVIEW OF TOTAL RADIATION DOSAGES WITH WHOLE BRAIN RADIATION AND RADIOSURGERY IN PATIENTS TREATED FOR METASTATIC CANCER TO THE BRAIN

Andrew Graff ⁴, Patrick Clerkin ², Hayden Smith ^2,³, Robert Isaak ¹

Abstract

INTRODUCTION: Limited data is available on whether patients treated for metastatic cancer to the brain with both whole brain radiation (WBRT) and stereotactic radiosurgery (SRS) receive dosages within published normal tissue limits. METHODS: A retrospective observational study was conducted on patients receiving both WBRT and SRS from 2009-2012. Four critical organ structures (brainstem, optic nerve, optic chiasm, cochlea) were contoured using Eclipse^™ treatment planning system. Total radiation dosages included entire treatment course (WBRT plus subsequent SRS) and were converted to bioequivalent dose (BED) using linear quadratic model and to isoeffective dose display (EQD2). Mean total dosages and standard deviations (SD) were calculated and compared to published dose constraints for SRS (Timmerman, 2008) and WBRT (RTOG Protocol 0615 and 0825) using one-sample t-tests. RESULTS: Twenty metastatic patients were seen during study period. Mean volume BED doses for brainstem were 90.2 (SD:28.5)Gray(Gy). Mean point BED doses were 79.5 (SD:21.2)G), 68.3 (SD:18.3)Gy, 70.4 (SD:10.3)Gy, and 70.1 (SD:9.0)Gy for brainstem, optic nerve, optic chiasm, cochlea, respectively. Mean total EQD2 doses were 54.1 (SD:17.1)Gy, 41.0 (SD:11.0)Gy, 42.2 (SD:6.2)Gy, and 42.1 (SD:5.4)Gy for brainstem, optic nerve, optic chiasm, cochlea, respectively. All mean BED doses were significantly higher (p-value < 0.0001) than SRS constraints and all EQD2 doses(except brainstem) were significantly lower(p-value < 0.0001) than WBRT constraints. CONCLUSION: Cumulative radiation dosages delivered to target organs in patients receiving both SRS and WBRT exceeded normal tissue constraints for SRS, but not WBRT. Exceeded threshold values can be partly explained by current dose constraints for SRS not taking into account patients treated with WBRT. Many patients in the study received multiple repeat treatments with SRS. Time between treatments were not measured, which could have resulted in tissue repair. Further research is needed on dose tolerances over extended periods of time, along with a better understanding of long-term outcomes. When treating patients with WBRT and SRS, previous radiation dosages should be taken into account and total culmulative doses to critical structures should be prospectively monitored.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-014. RE-IRRADIATION PLUS BEVACIZUMAB TREATMENT IN PATIENTS WITH RECURRENT HIGH-GRADE GLIOMAS

Jeffrey Dinh ¹, David Grosshans ¹, Pamela Allen ¹, John de Groot ¹, Susan McGovern ¹, Mary McAleer ¹, Mark Gilbert ¹, Paul Brown ¹, Anita Mahajan ¹

Abstract

BACKGROUND: Bevacizumab (BEV) is increasingly utilized in the treatment of recurrent high-grade gliomas. However, the utility of BEV in the setting of re-irradiation has not been fully evaluated. We examined the timing of BEV treatment in relation to re-irradiation for recurrent high-grade gliomas. METHODS: We retrospectively reviewed the medical records of patients with recurrent high-grade gliomas, treated between 2007 and 2012, with a second course of radiation therapy. We compared overall survival (OS) and progression free survival (PFS) as calculated from the time recurrent high-grade glioma was diagnosed. RESULTS: A total of 50 patients were identified with a median follow-up of 10 months (range 2.2 to 70). 12 patients received no BEV, 18 received concurrent BEV with re-irradiation, 16 BEV after re-irradiation, and 4 before re-irradiation. Median PFS was 14.4, 3.6, 5.8, and 3.5 months and median OS 25.2, 6.3, 9.9, and 5.7 months, respectively. There were no significant differences in age, sex, or number of surgical resections between groups. However, patients not receiving BEV were more likely to have received either concurrent or adjuvant temozolomide with re-irradiation (p = 0.002, Fisher's Exact Test) or to have been first diagnosed with low-grade tumors. No patients that received BEV prior to re-irradiation experienced overt radiation necrosis, whereas other groups experienced a 30-33% incidence of necrosis. CONCLUSION: In this small retrospective study, patients receiving no BEV had superior survival. However, selection bias, temozolomide use and histology may account for this finding. Among patients receiving BEV, BEV administered following re-irradiation yielded the best PFS and OS. However, BEV with re-irradiation did not appear to improve PFS or OS compared with historical data with single agent BEV or re-irradiation alone. BEV did not appear to lower incidence of radiation necrosis, except when given prior re-irradiation.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-015. PROSPECTIVE LONGITUDINAL ASSESSMENT OF SENSORI-NEURAL HEARING LOSS WITH HYPERFRACTIONATED RADIATION THERAPY ALONE IN PATIENTS WITH AVERAGE-RISK MEDULLOBLASTOMA

Tejpal Gupta ¹, Sarthak Mohanty ², Sadhana Kannan ¹, Rakesh Jalali ²

Abstract

PURPOSE: To report on sensori-neural hearing loss (SNHL) with hyperfractionated radiation therapy (HFRT) without upfront platinum-based chemotherapy in average-risk medulloblastoma. METHODS: Hearing thresholds were assessed by ear-specific standard-frequency pure-tone audiograms. Audiometric assessments were done serially longitudinally at baseline, between 6-12 months after HFRT, and annually thereafter. Ototoxicity was graded according to Brock's pediatric ototxicity grading criteria. Longitudinal changes in audiometric thresholds following HFRT were tested for significance by the linear mixed model. RESULTS: A total of 63 audiograms (median of 3 per patient) were included. Five of 20 (25%) children had communicatively and developmentally significant SNHL (Brock's grade 2 or worse) at baseline even before HFRT. On follow-up, new onset Brock's grade 2 or worse ototoxicity was documented in 6 previously normal ears. Eleven patients had preserved hearing in both ears on last-audiometric follow-up. Compared to baseline testing, post-HFRT audiometry at 2-3 years showed modest decline in hearing threshold across all frequencies. Age at diagnosis and gender were not significantly associated with hearing impairment. Tumors that extended more towards one side expectedly showed significant worsening in the ipsilateral ear. There was a differential impact of treatment on right and left ears with the right ear (and not the left ear) showing significantly worse hearing thresholds in the low-to-intermediate speech frequency range over time. CONCLUSION: The use of HFRT for craniospinal irradiation and conformal tumor bed boost without upfront platinum-based chemotherapy in children with average-risk medulloblastoma results in preserved hearing in a large proportion of patients in the audible speech range.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-016. RECURRENCE PATTERNS IN ANAPLASTIC ASTROCYTOMA: A RETROSPECTIVE STUDY OF PATIENTS TREATED AT THE MAYO CLINIC

John Hardie ¹, Nadia Laack ¹, Sani Kizilbash ¹, Jan Buckner ¹, Caterina Giannini ¹, Joon Uhm ¹, Ian Parney ¹, Robert Jenkins ¹, Paul Decker ¹, Jesse Voss ¹

Abstract

OBJECTIVES/PURPOSE: Determine the patterns of relapse and recurrence for patients with Anaplastic Astrocytoma (AA) and evaluate tumor and treatment characteristics that may be associated with recurrence locations. MATERIALS/METHODS: This retrospective chart review study examined recurrence patterns in patients treated for AA at Mayo Clinic between 2001 and 2012. Of 184 such patients, 86 had both a documented recurrence and sufficient imaging available to analyze the timing and location of the recurrence. Recurrences were classified as “central”, “marginal”, “distant ipsilateral”, “multiple”, or “contralateral” with respect to the surgical bed and radiation treatment fields. For this analysis, marginal recurrence was determined qualitatively with some but not all of the recurrent disease overlapping the radiation field (T2 flair + 1 cm). Correlation of recurrence location with the extent of surgical resection, the presence or absence of IDH mutations, and the use of temozolomide were performed, as was a survival analysis stratified by recurrence location. RESULTS: Of the 86 available patients, 38 recurred centrally, 32 marginally, 4 ipsilaterally but distant from the treatment fields, 6 contralaterally, and 6 had recurrence at multiple locations between imaging studies. Median survival for patients with marginal recurrence was 51 months, compared to 22 months for central recurrence (p < 0.02). Correlation analysis was done with the central and marginal recurrences. Marginal recurrence was significantly correlated with the presence of IDH mutations (p < 0.01). Neither extent of surgery (biopsy alone vs subtotal resection vs gross total resection) nor the use of temozolomide therapy was correlated with recurrence location. CONCLUSIONS: In this retrospective analysis we found an association between pattern of failure and survival outcomes in AA. Marginal recurrence was associated with both longer median survival and with the presence of IDH mutations. Subsequent analysis will focus on dosimetric and tumor imaging correlates with pattern of failure.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-017. THE UTILITY OF BORON NEUTRON CAPTURE THERAPY USING REACTOR NEUTRON FOR IRRADIATED PATIENTS WITH RECURRENT MALIGNANT GLIOMA

Ryo Hiramatsu ¹, Shinji Kawabata ¹, Motomasa Furuse ¹, Shin-Ichi Niyatake ¹, Toshihiko Kuroiwa ¹, Minoru Suzuki ², Koji Ono ²

Abstract

INTRODUCTION: The prognosis for patients with recurrent malignant gliomas (MGs) is poor and few therapies have been found to be efficacious. Boron neutron capture therapy (BNCT) has been developed in the hope of achieving a breakthrough in recurrent MGs treatment. BNCT is based upon the nuclear reaction that occur when non-radioactive boron-10 (10B) is irradiated with low energy neutrons to produce high energy a particles (10B[n,alpha] 7Li). We have applied this BNCT to malignant brain tumors. Here we introduce the survival benefit of BNCT for recurrent MGs, with special reference to new Recursive Partitioning Analysis (RPA) classification recurrent MGs as advocated by New Approaches to Brain Tumor Therapy (NABTT) CNS Consortium. PATIENTS AND METHODS: From January 2002 to December 2012, we treated a total of 34 cases of recurrent MGs using BNCT. All cases had been treated by radiotherapy mainly by fractionated external beam X-ray irradiation (29/34 cases; 85%) prior to BNCT. Median age was 50.0 with Male:Female of 22:12. KPS at relapse was ranged 50 to 100 (median; 80) and tumor volume was 32.0 (1.2–70.2) mL. After BNCT, patients were followed by MRI and 18F-BPA (amino acid) - positron emission tomography (PET). Also overall survival was evaluated with special reference to RPA classes advocated by NABTT as above. RESULTS: The median overall survival of BNCT-treated recurrent gliomas was 11.2 (8.9–12.4) months, while that of total NABTT cases (N = 333) was 7.0 (6.2–8.0) months. The median overall survival of NABTT RPA classes 3 + 7 and BNCT RPA class 3 + 7 were 4.4 (3.6–5.4) and 11.0 (7.8–11.6), respectively. CONCLUSIONS: BNCT showed good survival benefit especially for the poorest prognostic group (RPA class 3 + 7). Further clinical trial is underway by accelerator based neutron capture therapy instead of the nuclear reactor as a neutron source since last year in Japan.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-018. REIRRADIATION OF PRIMARY CNS GLIOMAS: A SINGLE INSTITUTION EXPERIENCE

Corey Hobbs ¹, Laura Vallow ¹, Jennifer Peterson ¹, Kurt Jaeckle ¹, Michael Heckman ¹, Rawal Bhupendra ¹

Abstract

PURPOSE: Primary central nervous system gliomas are initially treated with a combined modality approach. Despite intensive therapy at initial diagnosis, there is still a propensity for recurrence. Recurrences often progress and treatment options are limited due to cumulative toxicities. A growing body of literature suggests that tumor reirradiation is a feasible treatment option with an overall survival of approximately 6–20 months for selected patients. Our study attempts to add to this growing body of evidence. MATERIALS/METHODS: Between 1997 and 2012, 24 recurrent gliomas were treated at our institution with reirradiation. Retrospective chart review was performed to determine survival following treatment. Age, time between radiation treatments, gender, initial grade and histology, chemotherapy administration, conventional vs hypofractionated RT, and pretreatment ECOG grade were evaluated for any effect on primary outcomes. The cause of death and potential radiation toxicities were also recorded. RESULTS: With a median follow up of 5.4 months (0.6–55.7), median overall survival was 13.5 months and median time to progression was 5 months. Odds of progression were decreased with concurrent or adjuvant chemotherapy. Odds of progression were increased with astrocytoma compared to oligodendroglioma or mixed histology. There was a significantly increased risk of death for patients with a high grade tumor at initial diagnosis and for those who received salvage chemotherapy for recurrence prior to radiation. CONCLUSION: This data suggest that a second course of radiation therapy for recurrent CNS glioma is feasible in highly selected patients. Survival was similar to other published data for CNS glioma reirradiation. Patients who were candidates for concurrent or adjuvant chemotherapy had decreased risk of progression, which may be attributable to either the added benefit of chemotherapy or the superior ECOG score in these patients deemed appropriate to receive chemotherapy. Of 4 patients unable to complete radiation, 3 had an ECOG grade ≥ 2.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-019. FETAL RADIATION MONITORING AND DOSE MINIMIZATION DURING RADIATION THERAPY FOR GLIOBLASTOMA IN PREGNANCY

David Horowitz ¹, Cheng-Shie Wuu ¹, Wenzheng Feng ¹, Daphnie Drassinower ¹, Anita Lasala ¹, Andrew Lassman ¹, Tony Wang ¹

Abstract

BACKGROUND: To examine the fetal dose from irradiation of glioblastoma during pregnancy using intensity modulated radiation therapy (IMRT), and describe fetal dose minimization using mobile shielding devices. METHODS: Radiation dosing was prompted by the case of a pregnant woman with glioblastoma treated during the third trimester of gestation with 60 Gy of radiation delivered via a 6 MV photon IMRT plan. Fetal dose without shielding was estimated using an anthropomorphic phantom with ion chamber and diode measurements. Clinical fetal dose with shielding was determined during the course of radiation therapy with optically stimulated lumiescent dosimeters and ion chamber. RESULTS: Clinical target volume (CTV) and planning target volume (PTV) coverage was 100% and 98% receiving 95% of the prescription dose, respectively. Normal tissue tolerances were kept below quantitative analysis of normal tissue effects in the clinic (QUANTEC) recommendations. Without shielding, anthropomorphic phantom measurements showed a cumulative fetal dose of 0.024 Gy. In vivo measurements with shielding in place demonstrated a cumulative fetal dose of 0.016 Gy. The use of shielding resulted in a 33% decrease in dose delivered to the fetus. The fetal dose estimated without shielding was 0.04% and with shielding was 0.026% of the target dose, similar to the radiation dose from 1-2 CT scans of the chest. CONCLUSIONS: Using modern treatment techniques, brain irradiation can be delivered to pregnant patients with very low measured doses to the fetus, without compromising target coverage or normal tissue dose constraints. Fetal dose can further be reduced with the use of shielding devices, in keeping with the principle of as low as reasonably achievable.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-020. STEREOTACTIC PROTON THERAPY FOR PEDIATRIC SUPRATENTORIAL EPENDYMOMA: EARLY CLINICAL OUTCOMES

Daniel Indelicato ¹, Ronny Rotondo ¹, Julie Bradley ¹, Eric Sandler ¹, Philipp Aldana ¹, Nancy Mendenhall ¹, Robert Marcus ¹

Abstract

PURPOSE/OBJECTIVES: By limiting exposure of non-target normal tissues to radiation, proton therapy (PT) may reduce side effects in children with supratentorial ependymoma. The purpose of this study was to report acute toxicity and early disease control in patients treated with stereotactic PT. MATERIALS/METHODS: From 2008 to 2013, 28 patients with a median age of 5.4 years old (range, 1–15 years) were treated for non-metastatic supratentorial ependymoma. Four and 24 patients had WHO grade II and III tumors, respectively. 21/24 patients underwent a gross total or near total resection prior to PT. Eight patients received chemotherapy before PT, mainly vincristine/cyclophosphamide/platinum-based regimens. Daily stereotactic image guidance was based on titanium fiducial markers and bone landmarks. The median total dose was 59.4 CGE (range, 54-59.4 CGE). The median follow-up is 1.9 years (range, 0.1-5 years). RESULTS: All patients are alive and free of disease. Two of 3 patients, both < 3 years old, who underwent an initial subtotal resection were able to undergo a second look gross total resection within 6 months following 54 CGE. Among the entire cohort, no grade 2+ acute toxicity was observed and no patients required steroids. Late toxicity attributable to RT includes 2 patients with ipsilateral hearing loss, 2 patients with chronic otitis media, and 1 patient requiring growth hormone replacement. We have not observed vision decline, radiation vasculopathy, or brain necrosis. On post-treatent MRI, asymptomatic T2 prolongation was more common in those patients who had received chemotherapy. CONCLUSIONS: Stereotactic PT allows highly conformal radiation delivery without compromising early tumor control or increasing toxicity compared to conventional radiotherapy. To date, this is the largest cohort of children with supratentorial ependymoma treated with PT and provides the foundation for the long-term outcome studies necessary to accurately characterize the therapeutic ratio of PT.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-021. PATTERNS OF TUMOR RESPONSE AFTER STEREOTACTIC RADIOSURGERY FOR VESTIBULAR SCHWANNOMA

Rafi Kabarriti ¹, Waleed F Mourad ¹, Diana Molinares Mejía ¹, Jonathan Glanzman ¹, Shyamal Patel ¹, Rebekah Young ¹, Michael Bernstein ¹, Linda Hong ¹, Jana Fox ¹, Patrick LaSala ¹, Shalom Kalnicki ¹, Madhur Garg ¹

Abstract

BACKGROUND: Stereotactic radiosurgery (SRS) is commonly used for the treatment of vestibular schwannoma. In this study, we evaluate the control rates and MRI changes after treatment with Linac based SRS. MATERIAL AND METHODS: 27 consecutive patients treated with 12 to 14 Gy using LINAC based SRS at a single institution from 2007-2013 were analyzed. Baseline and post-treatment tumor volumes were assessed on T1 weighted contrast enhanced magnetic resonance imaging. Local control and changes in tumor volume were calculated. RESULTS: Median follow up time was 35.6 months (range 1.5–61.9) and median treatment tumor volume was 1.1 cc (range: 0.1–8.7 cc). Sixty three, 33, and 4% of patients had reduction, stable, and progression in the tumor volume at their last post treatment MRI, respectively. Median SRS treatment dose was 12 Gy (range: 8–14) treated to 100% of the PTV. Median Dmin was 99.6% (range: 60.9-100), and median Dmax was 124.2% (range: 108.7-148.5%). Median number of beams was 12 (range: 8–15). 42% developed an early transient increase in their post treatment tumor volume occurring 3 to 7 months after SRS. Additionally, 19% had late transient increase in their post treatment volumes occurring 26–37 months post SRS. 7% had both an early and late transient increase in their tumor volumes. Tumor volume, treatment dose, and Dmax did not correlate with MRI changes. CONCLUSIONS: Early and late transient increases in tumor volume occur in a significant number of patients with vestibular schwannoma treated with SRS. It is important to continue to monitor theses patients with follow up imaging to avoid mistaking them for treatment failures.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-022. DIFFUSION TENSOR IMAGING CONFIRMS EFFICACY OF RE-IRRADIATION IN PEDIATRIC PATIENTS WITH DIFFUSE INTRINSIC PONTINE GLIOMA

Soumen Khatua ¹, Ping Hou ¹, Johannes Wolff ², Jackson Hamilton ¹, Wafik Zaky ¹, Anita Mahajan ¹, Leena Ketonen ¹

Abstract

INTRODUCTION: The prognosis of diffuse intrinsic pontine glioma (DIPG) remains dismal with a median survival of 1 year. Radiation is the only therapeutic modality that has consistently shown transient clinical and radiographic improvement in these children. Re-irradiation to progressive DIPG is feasible and has shown to improve symptoms. Use of quantitative diffuse tensor imaging (DTI) estimation has enabled better understanding of the relationship between the tumor and the fiber tracts anatomy. This study evaluates changes in DTI of the pons in patients with progressive DIPG who receive re-irradiation with clinical correlation. METHODS: Apparent diffusion coefficient (ADC) and fractional anisotropy (FA) were analyzed pre and post-re-irradiation and correlated with clinical features in six patients with progressive DIPG who received re-irradiation therapy at our institution. RESULTS: DTI analysis showed statistically significant differences in FA values from areas of tumor progression versus improvement after re-irradiation in these six patients (p = 0.004). These changes were concordant with the 3D tractography analysis showing improved FA values after re-irradiation noted both in the transverse pontine fibers and corticospinal tracts. Changes in neuroimaging profile ranged from decrease in the size of the tumor to resolving enhancement and T2 FLAIR changes which correlated with clinical improvement in all the six patients. CONCLUSION: Serial changes in the FA and ADC values using DTI could provide clinically more correlative information in patients with progressive DIPG, who receive re-irradiation. This study for the first time shows the potential and efficacy of complementing these techniques to routine neuroimaging, to better assess anatomical and clinical response in these children.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-023. GAMMA KNIFE RADIOSURGERY FOR CYSTIC BRAIN METASTASES

Se-Hyuk Kim ¹, Sang Ryul Lee ¹, Ji ¹, Young Oh ¹

Abstract

Stereotactic radiosurgery has shown beneficial therapeutic effects for patients with metastatic brain tumors. However, cystic brain metastases are regarded as unsuitable for radiosurgery because of the large volume or unresponsiveness to the radiation. We retrospectively reviewed the clinical, radiological and dosimetry data of 37 cystic brain metastases of 28 patients, who were treated with Gamma Knife radiosurgery (GKRS). Eight large cystic metastases were treated with Ommaya reservoir insertion before GKRS to decrease the target volume. The most common primary tumor site was the lung (n =16). The mean lesion volume was 4.8 cc (range, 0.3-15.8) at the time of GKRS, and the mean prescription dose of 16.6 Gy (range, 13-22) was delivered to the mean 50.8% (range, 50-65) isodose line. The median follow-up duration was 8 months (range, 1-37.8) and the overall median survival time was 25.2 ± 14.3 months. Controlled primary tumor was a significant factor related to survival. The median survival time of the patients with controlled and uncontrolled primary tumor was 33.2 ± 7.6 and 9.0± 1.7 months, respectively. The actuarial local tumor control rate at 6 and 12 months was 92.9 and 82.2 %, respectively. Among various factors, margin dose (>15 Gy) was significantly related to local tumor control. Local tumor control rates of lesions treated with margin doses ≤15 and >15 Gy were 64.2% and 90.9% at 12 months. The tumor volume, cyst volume, and cyst/tumor volume ratio, and additional whole brain radiation therapy did not influence local tumor control. According to our results, we suggest that stereotactic radiosurgery should be considered for the treatment of cystic brain metastases when large tumor volume can be reduced by surgical drainage before radiosurgery, especially for patients with controlled primary tumor.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-024. PROSPECTIVE LONG-TERM NEUROENDOCRINE ASSESSMENT IN CHILDREN AND YOUNG ADULTS WITH PROGRESSIVE BENIGN/LOW-GRADE BRAIN TUMORS TREATED WITH CONVENTIONAL OR STEREOTACTIC CONFORMAL RADIOTHERAPY (SCRT) - RESULTS OF A RANDOMISED CONTROLLED TRIAL

Uday Krishna ¹, Nalini Shah ², Rima Pathak ¹, Tejpal Gupta ¹, Anurag Lila ², Padma Menon ², Atul Goel ², Rakesh Jalali ¹

Abstract

BACKGROUND: We report prospective detailed neuroendocrine outcome data in children and young adults with progressive benign/low-grade brain tumors randomised to either Conventional (Conv) or Stereotactic Conformal Radiotherapy (SCRT). METHODS: Two hundred patients (132 males, 68 females; median age-13 years) were enrolled in the trial out of which, 105 and 95 respectively were randomised to receive SCRT and Conv radiotherapy (RT) to a dose of 54Gy in 30 fractions. Pre-RT endocrine evaluation was performed to assess involvement axes of GH (anthropometry, serum peak GH & IGF-1), Gonadotrophs (SMR, Serum FSH, LH & Testosterone), Thyroid (Thyroid Function Test) and Adrenal (Basal Cortisol). Serial evaluations were performed post RT at 6-months, 2-yrs, 3-yrs, 4-yrs and 5-yrs to assess involvement of each of the above axes. RESULT: Most common histology was craniopharyngioma (40%), followed by astrocytomas and ependymoma (40%) and low-grade gliomas of the optic pathway/hypothalamus (20%). One hundred and twenty patients had tumors located ≤2cm from hypothalamic-pituitary axis (HPA). At baseline pre-RT evaluation, 56% in SCRT & 69% patients in Conv arm had dysfunction in atleast one hormone axis (p = 0.2). Endocrine function in all axes in SCRT and Conv RT arms was maintained in 27% and 13% at 3 years (p = 0.2) and 24% and 11% (p = 0.04) at 5 years, respectively. At 5 years, worsening of existing endocrine dysfunction with involvement of more than one axis was found to be lower in patients in SCRT arm compared to Conv RT arm (39% Vs 59%, p = 0.06). In patients with tumor >2cm from HPA, no endocrine dysfunction was seen in 36% Vs 23% (p = 0.16). CONCLUSION: We demonstrate Level I evidence through this randomised controlled trial that young patients with progressive low grade/benign brain tumors treated with high-precision SCRT, have significantly lower endocrine dysfunction at 5 year follow up and in fewer axes compared to conventional radiotherapy.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-025. VERY LOW DOSE STEREOTACTIC RADIOSURGERY FOR VESTIBULAR SCHWANNOMA

Rohan Lall ¹, Rishi Lall ¹, Timothy Smith ¹, Andrew Schumacher ¹, Addason McCaslin ¹, John Kalapurakal ¹, James Chandler ¹

Abstract

INTRODUCTION: While microsurgical resection of vestibular schwannomas yields high rates of disease control, operative morbidity has limited its appeal. In smaller volume lesions, stereotactic radiosurgery has gained favor and has been shown to effectively control tumor growth. Recently, practitioners have employed smaller treatment doses, typically between 12-14 gray, to limit hearing loss and radiation injury to adjacent neurologic structures. Published findings in the literature have been favorable. We have employed even lower doses at our institution, less than 12 gray. We describe here our series, utilizing the lowest doses of stereotactic radiosurgery for vestibular schwannomas thus far reported. METHODS: We examined the records of all patients at our institution that underwent stereotactic radiosurgery between 2004 and 2011, with dosing less than 12 gray to the 50% isodose line. 58 patients were included in the analysis. Serial radiographic and clinical outcomes were monitored. RESULTS: Median treatment dose was 11.0 gray. Median follow up was 37 months. Overall median progression free survival was 86.0 months. Progression free survival at 1, 2, 3, and 5 years was: 95%, 95%, 73%, and 56% respectively. Hearing preservation rates at 1, 2, 3, and 5 years were 94%, 94%, 88%, and 74% respectively. 0 patients (0.0%) developed post-treatment facial palsy. 3 patients (5.2%) developed dizziness/vertigo. 2 patients (3.4%) developed trigeminal neuralgia. 2 patients (3.4%) required subsequent surgical resection, and 0 patients (0.0%) required repeat radiosurgery. CONCLUSIONS: Treatment with sub 12-gray doses of radiosurgery in vestibular schwannoma appears to produce higher rates of hearing preservation than using higher doses. Cranial neuropathy and vertigo rates are very low. Progression free survival appears to be comparable to published series that utilize higher doses of radiation. Very low dose stereotactic radiosurgery provides effective treatment for vestibular schwannoma with excellent hearing preservation and low side effects.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-026. RETREATMENT RADIATION AS SALVAGE THERAPY FOR PROGRESSIVE GLIOBLASTOMA AFTER BEVACIZUMAB FAILURE

William Magnuson ¹, H Ian Robins ¹, Pranshu Mohindra ¹, Steven Howard ¹

Abstract

BACKGROUND: Outcomes after bevacizumab (BEV) failure for recurrent glioblastoma multiforme (GBM) are poor. Our analysis of 13 phase II trials (n = 913) revealed a median overall survival (OS) of 3.7 months after BEV failure with no discernible activity of salvage chemotherapy for this patient population. Thus, the optimal treatment for disease progression after BEV has yet to be elucidated. This study evaluated the efficacy of reirradiation as a treatment modality for recurrent GBM after BEV failure. METHODS: An IRB approved retrospective (2/2008- 5/2013) analysis was performed of 23 patients with recurrent GBM (after standard radiotherapy/temozolomide) treated with BEV (10 mg/kg) every 2 weeks until progression (median age 53 years; median KPS 80; median progression free survival on BEV 3.7 months). Within 7-14 days of BEV failure patients received re-irradiation to median dose of 54 Gy in 27 fractions using pulse-reduced dose rate (PRDR) [6.67 cGy/min] technique (Int J Radiat Oncol Biol Phys, 79(3):835-41, 2010). The median planning target volume was (424) cm³. At the start of reirradiation BEV (10 mg/kg) was given every 4 weeks for 1 to 2 additional cycles. RESULTS: The median OS after BEV failure was 6.7 months. The 3 and 6-month OS after BEV failure was (91%) and (65 %), respectively. Reirradiation was well tolerated with no Grade 3-4 events. CONCLUSIONS: Re-irradiation after BEV failure in patients with recurrent GBM appears to improve overall survival compared to historical controls. Furthermore, PRDR-reirradiation may be better tolerated than the continuation of systemic therapies. A prospective trial of progressive GBM is in progress at our institution comparing BEV followed by PRDR-reirradiation (at the time of BEV failure) versus reirradiation with BEV (at time of first BEV initiation). Support: Kathleen Reader Neuro-oncology Fund.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-027. CLASS SOLUTION IMRT PLANNING: BLINDED COMPARISON TO RTOG 0539

Anita Mahajan ¹, Denise Manfredi ², C Leland Rogers ³, Matthew Palmer ¹, Evangeline Hillebrandt ¹, Stephen Bilton ¹, Greg Robinson ⁴, Kyle Velasco ⁴, Minesh Mehta ⁵

Abstract

BACKGROUND: Intensity modulated radiotherapy (IMRT) allows highly conformal radiation delivery to complex target volumes. We have developed a class solution (CS) approach to optimize IMRT planning to increase efficiency, and possibly generate plan quality. In this study we compared individually-generated IMRT plans from RTOG 0539 to CS plans, created for the same cohort. We used multi-criteria plan quality metrics for plan assessment and comparison. METHODS: RTOG 0539 DICOM data including the planning scan and ROI's were electronically imported to generate CS plans. The planners were blinded to the original RTOG 0539 study plans, but utilized the pre-specified protocol specific requirements and original contours for CS planning. The resulting plan data were analyzed using two approaches: 1) semi-quantitative physician assessment to score the plan as acceptable per protocol criteria, or not, and 2) "hands-off" quantitative assessment using the Quality Reports [EMR] approach by an uninvolved third party. RESULTS: A total of 25 planning CT scans and associated ROI's have been imported. The CS plans were generated by 3 dosimetrists in a very efficient manner. By semi-quantitative physician assessment of the plans, based on pre-specified planning requirements, 86/86 (100%) CS plans were found to be within protocol guidelines. The multi-criteria metrics assessment is on-going and will be presented in full. CONCLUSIONS: IMRT planning is highly user dependent, and resource-intensive. A class solution approach appears feasible in terms of this highly challenging radiotherapeutic target, given its highly non-geometric shape distribution. Protocol guideline-consistent plans are achievable with consistent efficiency and timeliness. Semi-quantitative physician assessment demonstrated that 100% of the plans met protocol planning requirements, and a multi-criteria plan quality metric evaluation allow an unbiased customizable method for plan evaluation is ongoing. The highly efficient time commitment represents an unprecendented reduction in dosimetry time investment for such complex planning processes.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-028. LONG TERM FOLLOW-UP AFTER STEREOTACTIC RADIOSURGERY FOR THE TREATMENT OF MENINGIOMAS

John McGregor ¹, John Grecula ¹, Mario Ammirati ¹, Christopher Pelloski ¹, Lanchun Lu ¹, Nilendu Gupta ¹, Susan Bell ¹

Abstract

Complete surgical resection of intracranial meningiomas including the dura of origin and infiltrated bone remains the treatment of choice. Factors such as residual disease, advanced tumor grade, location within the cranial vault, or patient comorbidities may limit the success of surgery alone. A retrospective review of outcomes was performed at our institution from 1999–2012 of 81 patients with 89 presumed or confirmed meningiomas. The median age was 56 (range 17–86). There were 19 males and 62 females. The median marginal dose was 16.0 Gy (range 10 -31 Gy) to an averaged 50.8% isodose line (range 47–75%). 39 patients had undergone surgery previous to treatment (48%). 12 patients had undergone previous radiation therapy prior to SRS (15%). The patients were followed clinically and radiographically at 1, 3, 5 and ≥10 years after SRS. The majority (n = 58) of tumors (65%) were located in the falx/parasagital or cerebral convexities while the others (n = 31) involved the skull base (35%). Tumor histology when available included 27 WHO grade I, 17 WHO grade II and 1 WHO grade III meningiomas. 44 patients had no definitive pathologic analysis. In reviewing all 89 tumors, 7 lesions required further treatment during the follow up period (8%), either further radiation or surgery. Three of these progressive lesions were tumors at the skull base, and 4 were convexity/parasagital in location. Four of these tumors were known WHO grade ll. Two were WHO grade l, and one did not have confirmatory pathology. Of the 7 lesions that progressed, 3 had tumor volumes < 4 cm³, 2 lesions were between 4–13 cm³, and 1 lesion was > 13 cm³. Radiosurgery is a reasonable adjunct to management of patients with primary, residual or recurrent meningiomas and presumed meningiomas. The control rates are > 90%.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-029. PROGNOSTIC VALUE OF FET-PET IMAGING IN REIRRADIATION OF HIGH-GRADE GLIOMA: RESULTS OF A PROSPECTIVE TRIAL

Soren Moller ¹, Ian Law ², Per Munck af Rosenschold ¹, Junia Costa ^1,², Hans Skovgaard Poulsen ^3,⁴, Svend Aage Engelholm ¹

Abstract

PURPOSE: To evaluate the potential of FET-PET imaging as an early outcome marker for reirradiation of relapsed high-grade glioma in a prospective study. Data from an interim analysis of a prospective study are presented. MATERIAL AND METHODS: A phase I trial examining different dose levels for reirradiation of high-grade glioma is ongoing. Two of four dose level groups (3.5 Gy × 10 and 3.5 Gy × 10 + 7 Gy boost) have completed accrual and the results are reported. Target delineation and follow-up was carried out using MRI (magnetic resonance imaging) T1w + contrast and FET-PET (18F-flouroethyl-tyrosine positron emission tomography). FET-PET scans were acquired at baseline, during radiotherapy (at median dose =31.5 Gy) and 4 weeks post-treatment. The images were compared with regards to tumor volume (tumor-to-normal cut-off = 1.6) and maximal activity (Tmax/brain). Changes were quantified relative to baseline value. Treatment was evaluated using the RANO criteria. Hazard ratios (HR) were estimated using univariate Cox regression. RESULTS: Twenty-one patients have currently been included (16 glioblastoma (GBM)/5 anaplastic glioma) between December 2011–March 2013. All patients had received prior irradiation and 57% had received bevacizumab previously. Median follow-up was 9.3 months (range: 1.7–16.6 months). Currently, 17 patients have experienced disease progression and 9 have died. Baseline tumor volume defined by FET-PET, but not MRI, was prognostic for overall survival (OS) ( HR= 2.05 per increase of 10 cm³, p = 0.003) and progression-free survival (PFS) (HR = 1.40, p = 0.027) for all patients and for GBM when analysed separately (OS: HR = 1.78, p = 0.016 PFS: HR = 1.33, p = 0.099). Changes in tumor volume or Tmax/brain during treatment and 4 weeks post-treatment did not correlate with outcome. CONCLUSION: Baseline tumor volume defined by FET-PET, but not MRI, was strongly prognostic for OS and PFS. Relative changes in tumor volume or Tmax/brain during- and after reirradiation were not found to correlate with clinical outcome.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-030. INTRACRANIAL CALCIFICATIONS IN PEDIATRIC BRAIN TUMOR PATIENTS AFTER RECEIVING PROTON BEAM THERAPY

April Morrison ¹, Branko Cuglievan ¹, Ziad Khatib ¹

Abstract

Proton radiation therapy is emerging as an effective way to treat brain tumors in children. It provides high-dose treatments that are confined to the tumor area and minimizes the radiation to surrounding tissues. It is this precise control as well as fewer side effects that renders proton beam therapy more advantageous than conventional photon radiotherapy. We retrospectively reviewed 17 cases of brain tumors in pediatric patients who underwent proton beam radiation therapy. MRI and/or CT of head/brain that were performed after proton beam therapy had ended were examined.Evidence of calcified microangiopathy or overt calcification as read by the radiologist was noted and recorded. The patients reviewed were between the ages of 2 and 25 years. Seven of these seventeen patients developed variable degrees of intracranial calcification within months to years after proton beam radiation, however only one was symptomatic. The patients who developed calcification had Medulloblastoma (2 cases), Embryonal Tumor with Abundant Neuropil and True Rosettes (ETANTR)-1case, Primitive Neuroectodermal Tumor (PNET) 1 case, Anaplastic Aastrocytoma 1 case and Atypical Teratoid Rhabdoid Tumor (ATRT) 2 cases.No patients had any evidence of calcification prior to treatment. Six had asymptomatic calcified microangiopathy, punctate microcalcification or dystrophic calcification. One symptomatic patient with ATRT had significant calcifications in the midbrain, pons, medulla and periventricular white matter as well as basal ganglia 6 months after completion of proton therapy. Some of these calcifications measured up to 1 cm in diameter and were missed on MRI.CT scan showed large calcifications more clearly. Calcifications seen as calcified microangiopathy, mineralization and mineralizing angiopathy are seen in 7/17 pediatric brain tumor patients after receiving varying amounts of proton beam radiation. Patients with more extensive calcification had a greater degree of clinical manifestations. Further investigation into this phenomenon should be done to determine if the incidence of microangiopathy is similar to conventional radiation techniques.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-031. THE INCIDENCE OF BRAIN METASTASES IN CORRELATION TO DOMINANT HEMISPHERE AND DIFFERENT BRAIN LOBES

Waleed F Mourad ^1,², Rafi Kabarriti ^1,², Rebekah Young ^1,², Tamara Santiago ^1,², Dukagjin M Blakaj ^1,², Mary Welch ^1,², Jerome Graber ^1,², Shyamal Patel ^1,², Linda X Hong ^1,², Asif Patel ^1,², Adesh Tandon ^1,², Michael B Bernstein ^1,², Rania A Shourbaji ^1,², Jonathan Glanzman ^1,², Merritt D Kinon ^1,², Jana L Fox ^1,², Patrick Lasala ^1,², Shalom Kalnicki ^1,², Madhur K Garg ^1,²

Abstract

PURPOSE: Blood flow to the brain varies according to the dominance and functionality of each brain lobe. We hypothesized that this greater blood flow would lead to an increase in seeding of metastases to areas within the dominant lobe when compared to the nondominant lobe. METHODS: This is a single institution retrospective study of 190 patients who underwent SRS for 382 metastatic brain lesions from 2007-2012. All pts underwent single fraction LINAC-based SRS via 6 MV photon using fixed beams (68%) or dynamic arcs (32%) prescribed to 100% of the PTV. The median prescribed dose and PTV were 21Gy (15-25) and 2cc (0.1-32). The median Dmax (dose to isocenter) and Dmin were 26.5 and 21Gy, respectively. All patients had an ECOG performance status ≤2. The cohort consisted of Hispanics (39%), Caucasians (31%), African Americans (29%), and Asians (1%); 50% of the cohort was male. Histologies included lung (52%), breast (20%), renal (14%), and other malignancies (14%). RESULTS: The median follow-up for the entire cohort was 5.5 months (ms); however, the median follow-up intervals for survivors vs. expired pts were 12.8 ms (4-94.4ms) vs. 5 ms (1-60). The incidence of brain metastases in the different brain lobes were: 43%, 24%, 12%, 11.5%, and 9.5%, for frontal lobe, parietal lobe, cerebellum, temporal lobe, and occipital lobe, respectively. The incidence of brain metastases in the dominant hemisphere was 97% (p = 0.01). The LC for the entire cohort was 88%. The actuarial overall survival (OS) was 26%. No patients experienced RTOG grade ≥3 toxicities. Breast cancer pts had a longer median actuarial OS (9 ms) vs. other histologies (5 ms). Chi-square analysis showed that metastectomy prior to SRS was a statistically significant predictor for better LC (p < 0.0001). CONCLUSION: There is increased seeding of brain metastases to the dominant brain hemisphere with higher blood flow.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-032. DEFERRED RADIATION IS AN APPROPRIATE OPTION FOR PATIENTS PRESENTING FOR TREATMENT OF NON-FUNCTIONING PITUITARY MACROADENOMAS WITH RESIDUAL POST SURGICAL TUMOR

Sarah Nicholas ¹, Roberto Salvatori ¹, Michael Lim ¹, Kristin Redmond ¹, Alfredo Quinones ¹, Gary Gallia ¹, Daniele Rigamonti ¹, Lawrence Kleinberg ¹

Abstract

OBJECTIVE: To review the outcomes of patients treated with surgical resection in whom radiation therapy was delayed until the time of progression. METHODS: 267 patients with pituitary adenoma were identified from a pathology database between 1999-2005 who were operated at Johns Hopkins Medical Center. 82 of these patients had a non-fucntional macroadenomas (NFPM), and at least 2 years of follow up. These patients were managed with delayed radiation or repeated surgery at the time of recurrence or regrowth of the lesion. Baseline characteristics were defined as present at time of presentation, or noted <12 months from resection. Outcomes, including radiographic progression, further treatment and development of new endocrine and visual field abnormalities were collected. RESULTS: At baseline, 61% patients had visual field defects, 7.3% with cranial nerve II-VI abnormalities, and 66% with endocrine abnormalities. 93.9% of patients underwent transsphenoidal resection and 6.1% craniotomy. Of 78 patients with postoperative MRI, 91% had residual tumor reported, and 47% had a ≥ 1 cm remnant. 4 patients had persistent postoperative diabetes insipidus present at last follow up. During a follow up that ranged from 25 to 183 months (median, 96 months), 52.5% of patients were found to have radiographic evidence of growth. 41.25% required additional treatment. As a result of disease progression or additional therapy, 21.25% of patients developed new or progressive visual field deficits and 17.5% developed new endocrinopathies. 26.8% of patients had resolving or normal visual fields at last follow up. Of 72 patients with endocrine evaluation at last follow up, 15% had resolution of some or all endocrinopathies. CONCLUSION: Close monitoring and observation with delayed radiation therapy is an appropriate option for patients presenting with residual post-operative NFPM's. Physicians must weigh the risk of requiring subsequent resection against those of immediate radiation for individuals.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-033. STEREOTACTIC RADIOSURGERY FOR BRAIN METASTASES FROM LUNG CANCER: THE IMPACT OF POST-SRS SYSTEMIC THERAPY

Shyamal Patel ^1,², Waleed Mourad ^1,², Rebekah Young ^1,², Rafi Kabarriti ^1,², Tamara Santiago ^1,², Jonathan Glanzman ^1,², Michael Bernstein ^1,², Asif Patel ^1,², Ravi Yaparpalvi ^1,², Linda Hong ^1,², Jana Fox ^1,², Patrick LaSala ^1,², Shalom Kalnicki ^1,², Madhur Garg ^1,²

Abstract

PURPOSE: To report our experience with stereotactic radiosurgery (SRS) for brain metastases from lung cancer and assess the prognostic significance of clinical predictors including post-SRS systemic therapy (PSST). METHODS: In this single-institution retrospective study, we analyzed 97 patients with lung cancer and ECOG PS ≤ 2 who underwent SRS for 183 brain metastases between 2007 and 2013. Patients underwent LINAC-based SRS with a median prescribed dose and planning target volume of 21Gy and 1.4cc, respectively. Patients underwent metastectomy (24%) or whole brain radiation (WBRT) (12%) prior to SRS while 50% received PSST. Histologies included adenocarcinoma (60%), poorly differentiated carcinoma (16%), squamous cell carcinoma (12%), and small cell carcinoma (7%). Median age was 65 (41-85). Dates of primary diagnosis, first CNS metastases, and PSST initiation were recorded. For analysis, the SRS-to-PSST interval (SPI) was divided into ≤30 days and >30 days. RESULTS: At a median follow-up of 5 months (1-59), the median survival was 6 months and the actuarial overall survival at 3, 6, 12, 24, and 36 months was 77%, 50%, 31%, 13%, and 5%, respectively. Cox regression analysis revealed PSST to be a positive predictor for survival (HR 0.58, 95% CI 0.36-0.95, p = 0.030); age, gender, metastectomy or WBRT prior to SRS, and the primary-to-CNS disease diagnosis interval were not significant. In patients receiving PSST, predictors for improved survival were SPI >30 days (HR 0.77, 95% CI 0.61-0.96, p = 0.020) and longer PSST duration (HR 0.92, 95% CI 0.85-1.00, p = 0.040). Log-rank test revealed no survival difference between histologies. CONCLUSION: Our results suggest the prognostic significance of PSST. While a longer duration of PSST was found to be beneficial, delaying its initiation to >30 days was also prognostic. This latter finding was potentially influenced by pre-SRS systemic therapy, systemic disease control, or neurotoxicity. Further investigation is warranted to define the optimal SPI.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-034. PROTON MR SPECTROSCOPY (MRS) AND NEURO-PSYCHOLOGICAL DYSFUNCTION FOLLOWING CRANIAL IRRADIATION FOR PEDIATRIC MALIGNANCIES: A PROSPECTIVE STUDY

Kristin Redmond ¹, Omar Mian ¹, Mahaveer Degaonkar ¹, Haris Sair ¹, Stephanie Terezakis ¹, Lawrence Kleinberg ¹, Todd McNutt ¹, Moody Wharam ¹, Mark Mahone ¹, Alena Horska ¹

Abstract

PURPOSE: Mechanisms underlying neuro-psychological sequelae of cranial RT remain uncertain. This prospective study examines the relationship between neuro-metabolites using MRS, RT dose, and neuro-psychological deficits. METHODS: 12 patients and 35 age/sex matched controls included. MRS and cognitive assessment conducted at baseline and 6, 15, 27 months following RT and compared to controls. Neuro-psychological battery included: motor dexterity, visual attention/processing speed, visual working memory, visual perception, verbal learning, verbal working memory, single word vocabulary. Thalamus, caudate, putamen, hippocampus, temporal lobes and genu/splenium of corpus callosum contoured. Concentrations of N-acetylaspartate (NAA) and choline (Cho) in these regions expressed as ratios NAA/Cho and Cho/Cr (creatine). Linear mixed effects (LME) regression models examined associations among metabolite ratios, RT doses to contoured structures and performance on testing, and compared patients with controls. Mean age at RT: 12 years (range 8.6–16.1). Mean prescription: 50 Gy (range 18-54). RESULTS: Overall LME analyses revealed no group differences in Cho/Cr but lower NAA/Cho in RT group based on region (p = 0.04). No group differences in regional NAA/Cho detected at baseline, but there was reduced NAA/Cho in splenium (p = 0.01) and thalamus (p = 0.038) 6 months following RT. For patients, a negative association between RT dose and NAA/Cho detected (p = 0.014). Following RT, reduced regional NAA/Cho ratio in temporal lobe predicted poorer verbal learning (p = 0.006) and visual perception (p = 0.049); caudate poorer motor dexterity (p = 0.028); genu poorer visual working memory (p = 0.03); and hippocampus poorer visual perception (trend p = 0.058). Increased mean RT dose to caudate associated with poorer verbal memory (p = 0.043); and genu (p = 0.043) poorer verbal memory (p = 0.043) and visual attention/processing speed (p = 0.036). CONCLUSION: This study suggests significant changes in NAA/Cho following cranial RT, and significant relationships between NAA/Cho, mean dose, and performance on neuro-psychological testing. Since NAA is a neuronal marker, these observations suggest regional impairment in neuronal integrity and function following RT.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-035. JUDICIOUS USE OF RADIOSURGERY (SRS) MAY CHANGE THE ULTIMATE PATTERNS OF FAILURE IN PATIENTS WITH BRAIN METASTASIS FROM MELANOMA

Uzma Rezvi ¹, Edward Melian ¹, Murat Surucu ¹, Ibrahim Mescioglu ², Vikram Prabhu ¹, Joseph Clark ¹, Douglas Anderson ¹

Abstract

INTRODUCTION: In melanoma brain metastases the metastases contribute to death up to 95% of the time (Sampson, 1998). We assess the impact of SRS on cause of death. METHOD: Thirty sequential patients (68 lesions) had SRS. Median age was 58. Overall survival (OS) from diagnosis and from SRS, number of lesions at first SRS, local control (LC) and various prognostic factors were assessed. RESULTS: At initial SRS, 15 patients were treated at 1 site, 9 at 2, 5 at 3, and 1 at 4. Five patients were treated twice and 1 thrice. SRS controlled 56/68 lesions until death. Dose and tumor volume was not correlated with LC. Fourteen patients were controlled intracranially while 13 were not (3 unknown) at death. With 3-4 lesions at SRS 72% had uncontrolled intracranial disease at death while with 2 or less lesions 34% had uncontrolled intracranial disease (p < 0.05). Median survival from diagnosis was 41.0 months and from SRS was 5.4 months. Cause of death was systemic in 18 patients, brain and systemic in 2, brain in 6, and unknown in 2. Time to SRS, age, RPA or KPS class, IL-2, any systemic therapy didn't correlate with OS from SRS or cause of death. Primary tumor site was correlated with OS after SRS (2.5 months trunk, vs. 6.4 non-trunk p < 0.05), but not with cause of death. Mean survival was 5.8 months for patients receiving whole-brain irradiation (WBRT) and 4.4 months not receiving WBRT (p < 0.05). In patients with stable systemic disease at SRS, cause of death was brain only in 63% while progressive disease cause of death was systemic only in 88% (p < 0.05). CONCLUSION: SRS controlled the irradiated lesion for most patients perhaps shifting the failure pattern. WBRT added a small benefit to intracranial control and OS, greatest for those with controlled systemic disease and/or 3+ lesions.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-036. CLINICAL AND THERAPEUTIC IMPLICATIONS OF SOLITARY VERSUS SINGLE BRAIN METASTASIS

Jared Robbins ¹, Raphael Yechieli ¹, Samuel Ryu ¹

Abstract

INTRODUCTION: True solitary brain metastasis (SolBM) as defined by one brain metastasis without any evidence of malignancy outside the brain is a relatively rare presentation compared to a single brain metastasis (SingBM) with active primary or systemic disease. Little evidence is available directly comparing outcomes between SolBM and SingBM. METHODS: A review was performed of 103 patients with one brain metastasis all treated with surgical resection followed by radiosurgery to the cavity without whole brain radiation therapy from July 1996 to April 2011. Thirty-seven were solitary, while sixty-six were single. Kaplan-Meier and log-rank test methods were used to evaluate outcomes with Cox regression analysis to determine significant associations. RESULTS: The median follow-up for all patients was 14.5 months. Median follow-up for living patients was 56 months (25 months to 15 years). SolBM patients had higher KPS (p = 0.003) and longer intervals from cancer diagnosis to brain metastasis (p < 0.001). Overall survival (OS) was significantly better for SolMB compared to SingMB (median OS 27.3 months vs. 9.5 months; 1-year 84% vs. 40%; 2-year 65% vs. 17%; 5-year 32% vs. 5%; and 10-year 13% vs. 2% (p < 0.001)). SolBM had lower rates of new intracranial metastasis with 1-year and 2-year rates of 24% and 47% compared to 49% and 77% for SingBM (p = 0.002), and required less additional cranial radiation treatments with 1-year and 2-year rates of 40% and 57% compared to 58% and 73% for single metastasis (p = 0.014). On multivariate analysis, SolBM was associated with longer survival (HR 0.494, p = 0.027), and a trend for fewer new intracranial metastasis (HR 0.571, p = 0.070). CONCLUSIONS: Patients with SolBM have long-term survival potential, fewer new intracranial metastases, and require less additional brain therapies after the initial treatment. Since long-term cure is possible, aggressive treatment selected to reduce local recurrence and late complication is important.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-037. STEREOTACTIC BRACHYTHERAPY USING IODINE-125 AS A TREATMENT OPTION FOR PRIMARY AND RECURRENT ANAPLASTIC GLIOMAS WHO°III

Maximilian I Ruge ¹, Bogdana Suchorska ², Christina Hamisch ¹, Kerstin Mahnkopf ⁴, Ralf Lehrke ³, Harald Treuer ¹, Volker Sturm ¹, Juergen Voges ⁴

Abstract

OBJECTIVE: The current study evaluates the effect of stereotactic brachytherapy (SBT) with iodine-125 seeds in both newly diagnosed anaplastic glioma (AG) including astrocytoma, oligoastrocytoma, oligodendroglioma not suitable for resection as well as recurrent tumors after chemo/or radiotherapy. METHODS: Out of 223 patients harboring an AG treated according to a prospective protocol, 183 patients were selected who received SBT in case of grade WHO°III de-novo/residual tumor (n = 95) or in case of tumor recurrence after previous therapy (n = 88). Progression free survival (PFS), overall survival (OS), radiological and clinical outcome were assesed. For identification of prognostic factors among clinical and treatment parameters, univariate and multivariate regression analyses were performed. RESULTS: Median follow-up time was 38 months, median OS was 32.7 months in the de-novo group, while patients in the recurrence group survived 58.5 months. Median PFS was 22.4 months in the de-novo group and 36.1 months in the recurrence group. Tumors with an oligo-component were more often seen (p = 0.03) in the recurrence group while the number of eloquently located tumors was higher (p < 0.001) in the de-novo group. Cohort-specific multivariate analyses revealed following variables to have an impact on PFS in the de-novo group: age (p= 0.03), KPS at SBT (p = 0.02), tumor volume (p = 0.005), eloquent tumor location (p = 0.01) and histology subtype (astrocytic versus oligo) (p = 0.05). Only eloquent tumor location remained significant (p = 0.02) for OS. In the recurrence group, Cox regression has shown histology subtype to be of significance for PFS (p= 0.006) and OS (p = 0.05). Transient and permanent morbidity was low (∼1%) in both groups. CONCLUSION: The standard treatment for AG consists of surgery followed by radiotherapy or chemotherapy. For patients unable to undergo surgery due to eloquent tumor location or general condition and in case of tumor recurrence after multimodal treatment, SBT offers a reasonable option which does not foreclose additional therapeutic interventions.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-038. SURGICAL RESECTION OF EPIDURAL DISEASE IMPROVES LOCAL CONTROL FOLLOWING POST-OPERATIVE SPINE STEREOTACTIC BODY RADIOTHERAPY (SBRT)

Arjun Sahgal ¹, Ameen Al-Omair ¹, Laura Masucci ¹, Laurence Masson-Cote ¹, Eshetu Atenafu ¹, Daniel Letourneau ¹, Eugene Yu ¹, Raja Rampersaud ¹, Stephen Lewis ¹, Albert Yee ¹, Isabelle Thibault ¹, Michael Fehlings ¹

Abstract

BACKGROUND: Spine SBRT is increasingly being applied to the post-operative spine metastases patient. Our aim was to identify clinical and dosimetric predictors of local control and survival. METHODS: 80 patients treated between October 2008 and February 2012 with post-operative SBRT were identified from our prospective database and retrospectively reviewed. RESULTS: The median follow-up was 8.3 months. Thirty-five patients (44%) were treated with 18-26 Gy in 1 or 2 fractions, and 45 patients (56%) with 18-40 Gy in 3 to 5 fractions. Twenty-one local failures (26%) were observed, and the one-year local control (LC) and overall survival (OS) rates were 84% and 64%, respectively. The median time to local failure was 6.9 months. The most common site of failure was within the epidural space (15/21, 71%). We identified systemic therapy post-SBRT as the only significant predictor of OS (p = 0.03). Multivariate proportional hazards analysis identified treatment with 18-26 Gy/1 or 2 fractions, and a post-operative epidural disease grade of 0 or 1 (0 is no epidural disease, 1 is epidural disease that compresses dura only) as significant predictors of LC. Subset analysis for only those patients (n = 48/80) with high grade pre-operative epidural disease (cord deformed) indicated significantly greater LC rates when surgically down-graded to a 0/1 vs. 2 (p = 0.0009). CONCLUSIONS: Post-operative SBRT with high total doses ranging from 18-26 Gy delivered in 1-2 fractions predicted for superior LC, as did the post-operative epidural grade.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-039. MOLECULAR PROFILING OF GBM PATIENTS DEVELOPED PSEUDOPROGRESSION AFTER CHEMORADIATION TREATMENT

Wenyin Shi ¹, Joshua Palmer ¹, Jianliang Li ², Lawrence Kenyon ¹, Jon Glass ¹, Lyndon Kim ¹, Maria Werner-wasik ¹, David Andrews ¹

Abstract

INTRODUCTION: Pseudoprogression (psPD) is now recognized following radiotherapy with concurrent temozolomide (RT/TMZ) for glioblastoma multiforme (GBM). The purpose of this study was to explore biomarker expression profile of GBM patients with psPD. MATERIALS/METHODS: This study included 29 patients with newly diagnosed GBM treated between January 2012 and April 2013 at Thomas Jefferson University. All patients received standard radiation treatment to 60 Gy with concurrent temozolomide. All patients had pre-treatment MRI and 1-m post treatment MRI. All the patients were assessed in multidisciplinary clinic. The psPD was defined per Revised Assessment in Neuro-Oncology (RANO) criteria. Molecular profile through IHC testing, microarray analysis of RNA expression, FISH and mutational analysis was performed by Caris Life Sciences, Inc. Quantitative and qualitative analysis was performed using defined thresholds based on published evidence. RESULT: A total of 12 patients (41%) developed psPD after chemoradiation treatment. MGMT methylation was less frequent in patients with psPD as compared to those do not develop psPD, 25% vs 58%, respective. TOPO1 expression was more frequent in patients with psPD, 50% vs 29%. TS was found to uniformed expressed in patients with psPD (100%), while only expressed 52% of patients without psPD. PI3KCA mutation was more frequent in patients developed psPD, though the incidence is still low, 16%. No PI3KCA mutation was found in patients without psPD. The expression and mutation rate of other genes examined were similar between patients with and without psPD. CONCLUSIONS: Our findings demonstrate different gene expression profile of GBM patients with pseudoprogression. The observed gene expression profile will be confirmed with a validation data set. This may help identifying patients with pseudoprogression, and thus direct more appropriate treatment.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-040. RE-IRRADIATION WITH HYPO-FRACTIONATED STEREOTACTIC ROBOTIC RADIOSURGERY FOR SALVAGE IN ADULT PATIENTS WITH BRAINSTEM GLIOMA

Sridhar Susheela ¹, Swaroop Revannasiddaiah ^1,², Sandeep Muzumder ¹, Govindarajan Mallarajapatna ¹, Ajaikumar Basavalingaiah ¹, Monica Gupta ¹, Kumar Kallur ¹, Madhusudhan Hassan ¹, Ramesh Bilimagga ¹

Abstract

PURPOSE: Brainstem-glioma (BSG) is often treated with definitive irradiation. However, subsequent progression and death occur as a rule rather than the exception, after varying periods of control. The outlook of patients with post-irradiation progression is dismal, and most of these patients are treated with supportive care alone. Despite the obvious risks with an area as critical as the brainstem, it is a possibility to encounter situations wherein the patients (themselves or their associates) ask for re-irradiation, with the hope of a few extra months of life. The risk of radiation induced brainstem toxicity may be justifiable under the assumption that the patients may benefit from reirradiation, but still may not live long enough to manifest brainstem toxicity. METHODS: Five adult BSG patients were treated with re-irradiation by the use of a robotic-arm stereotactic radiation therapy (SRT) between September–2009 and July–2012, primarily on the request of the concerned patient parties. Re-irradiation doses ranged from 16 to 25 Gray delivered by robotic arm stereotactic irradiation in 2 to 5 fractions. RESULTS: Four of five patients enjoyed a prolongation of survival in the order of months (3,5, 6 and 14 months), which was very significant given that all patients had severe neurological compromise and poor performance status prior to re-irradiation. One patient however survived 26 months after re-irradiation and thus has lived long enough to manifest late radiation induced brainstem-toxicity. CONCLUSIONS: Despite the obvious risks of brainstem toxicity associated with the use of re-irradiation for BSG, the use of fractionated stereotactic re-irradiation seems to offers prospects of additional periods of local control and augments duration of useful life.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-041. TUMOR VASCULAR DAMAGE AND GLIOMA STEM-LIKE CELL EXPANSION AFTER RADIOTHERAPY IN MALIGNANT GLIOMAS

Kaoru Tamura ¹, Masaru Aoyagi ¹, Noboru Ando ², Takahiro Ogishima ¹, Massaki Yamamoto ¹, Kikuo Ohno ², Taketoshi Maehara ¹

Abstract

PURPOSE: Glioma stem cells could survive radiotherapy that may induce damage on tumor cells as well as tumor blood vessels. We examined malignant gliomas for stem-cell marker expression and tumor vascular damage after radiotherapy. PATIENTS AND METHODS: Tumor samples were collected during surgery from 70 patients with varying grades of gliomas (grade 2 in 12, grade 3 in 16 and grade 4 in 42 patients) prior to any adjuvant treatment. The samples were subjected to immunohistochemistry for MIB-1, factor VIII and stem cell markers. Tumor vascular density and CD133 expression on glioma cells were compared between primary and recurrent tumors in 31 patients after radiotherapy, including high-dose irradiation with additional stereotactic radiosurgery. RESULTS: CD133 expression on glioma cells was confined to de novo glioblastomas but was rarely observed in secondary glioblastomas following radiotherapy. In de novo glioblastomas, the percentage of CD133-positive glioma cells at recurrence (12.2 ± 10.3%) was significantly higher than that from the primary surgery (1.08 ± 1.78%). CD133 and Ki-67 dual-positive glioma cells were significantly increased in recurrent de novo glioblastomas after radiotherapy as compared to those in primary tumors (14.5 ± 6.67% vs. 2.16 ± 2.60%). Vascular density in recurrent tumors was significantly reduced after high-dose irradiation but not after external beam radiation (EBRT) alone (42.3 ± 50.6 vessels/mm vs. 178.7 ± 89.3 vessels/mm). The frequency of CD133-positivie glioma cells after high-dose irradiation appeared greater than that after EBRT alone (17.1 ± 11.3% vs 6.18 ± 4.33%). CONCLUSION: Our results indicate that glioma stem-like cells could survive, change to a proliferative cancer stem cell phenotype, and may invade surrounding brain tissues by co-opting normal blood vessels after high-dose radiation in cases with de novo glioblastomas.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-042. HYPOPITUITARISM FOLLOWING STEREOTACTIC RADIOSURGERY FOR PITUITARY ADENOMAS

Zhiyuan Xu ¹, Mary Lee Vance ¹, David Schlesinger ¹, Jason Sheehan ¹

Abstract

BACKGROUND: Studies of new-onset Gamma Knife stereotactic radiosurgery (SRS) -induced hypopituitarism in large cohort of pituitary adenoma patients with long-term follow-up are lacking. OBJECTIVE: We investigated the outcomes of SRS for pituitary adenoma patients with regard to newly developed hypopituitarism. METHODS: This was a retrospective review of patients treated with SRS at the University of Virginia between 1994 and 2006. A total of 262 patients with a pituitary adenoma treated with SRS were reviewed. Thorough endocrine assessment was performed immediately prior to SRS and in regular follow-up. It consisted of 24-hour urine free cortisol (patients with Cushing's disease), serum ACTH, cortisol, FSH, LH, IGF-1, GH, testosterone (men), PRL, TSH, and free T4. RESULTS: Endocrine remission occurred in 144 of 199 patients with a functioning adenoma. Tumor control rate was 89%. Eighty patients experienced at least one axis of new-onset SRS-induced hypopituitarism. The new hypopituitarism rate was 30% based upon endocrine follow-up ranging from 6 to 150 months; the actuarial rate of new pituitary hormone deficiency was 31.5% at 5 years post-SRS. On univariate and multivariate analyses, variables regarding the increased risk of hypopituitarism included suprasellar extension and higher radiation dose to the tumor margin; there were no correlations among tumor volume, prior transsphenoidal adenomectomy, prior RT, and age at SRS. CONCLUSIONS: SRS provides an effective and safe treatment option for patients with a pituitary adenoma. Higher margin radiation dose to the adenoma and suprasellar extension were two independent predictors of SRS-induced hypopituitarism.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-043. RADIATION-INDUCED BRAIN NECROSIS FOLLOWING PRIMARY STEREOTACTIC RADIOSURGERY: PROGNOSTIC FACTORS AND CLINICAL SEQUELAE

Rebekah Young ^1,², Dukagjin Blakaj ^1,², Merritt Drew Kinon ^1,², WF Mourad ^1,², Patrick A LaSala ^1,², Linda Hong ^1,², Shalom Kalnicki ^1,², Madhur Garg ^1,²

Abstract

PURPOSE: To determine prognostic factors and clinical sequelae of radiation-induced necrosis in patients treated for metastatic brain lesions utilizing LINAC-based stereotactic radiosurgery (SRS). METHODS AND MATERIALS: This single institution study retrospectively reviewed patients treated with LINAC-based SRS between 2007-2010. Records of treated patients (n = 127) were reviewed, and 62 patients (101 lesions) with no retreatment and 2 months or greater follow up were included in this analysis. The follow-up MRI showing necrosis was fused with the original treatment planning scan, and the area of necrosis was contoured. Histology, volumes of normal brain receiving 8, 10 and 12 Gy (V8, V10, V12); V100, D100, Conformality Index (CI), and tumor volumes were evaluated as possible predictive factors for necrosis. RESULTS: 16 patients (23 lesions) had radiographic changes associated with necrosis on MRI. Primary histology of these lesions included renal (10 lesions), lung (5 lesions, both NSCLC and small cell), and breast (8 lesions). Among patients with necrosis, mean V₈, V₁₀ and V₁₂ were: 17.1 ± 18.1, 12.2 cm³ ± 12.5, and 8.9 cm³ ± 9.1, respectively as compared to 8.4 cm³ ± 5.8, 6.0 cm³ ± 4.2, and 4.4 cm³ ± 3.0, respectively in patients without necrosis. This difference was statistically significant. (p = 0.04) The primary histology was found to not be a statistically significant factor in the development of necrosis. Of the 16 patients with necrosis, 5 (31%) were symptomatic (hemiparesis, dysphonia, writing difficulties and gait disturbances). Higher D100 significantly correlated with symptomatic sequelae (p = 0.04). CONCLUSIONS: In patients with metastatic brain lesions treated with primary SRS, higher V₁₂, V₁₀, and V₈ correlated with greater risk for radiographic necrosis, and higher D100 correlated with an increased likelihood for symptomatology. Histology was found to not be a statistically significant factor in the incidence of necrosis. Inclusion of these parameters in SRS planning should be considered.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-044. SALVAGE AND PRIMARY STEREOTACTIC RADIOSURGERY FOR BREAST CANCER BRAIN METASTASES: THE IMPACT OF RECEPTOR STATUS AND TIME TO TREATMENT ON CLINICAL OUTCOMES

Rebekah Young ^1,², WF Mourad ^1,², Shyamal Patel ^1,², Jana Fox ^1,², Patrick A LaSala ^1,², Linda Hong ^1,², Jerome J Graber ^1,², Tamara Santiago ^1,², Shalom Kalnicki ^1,², Madhur Garg ^1,²

Abstract

PURPOSE: To report our experience with stereotactic radiosurgery (SRS) for brain metastases from breast cancer, in both primary and salvage settings, and to evaluate potential prognostic factors. METHODS: This single institution retrospective review analyzed 73 lesions among 35 patients with brain metastases from breast adenocarcinoma who underwent SRS between 2007 and 2012. Patients underwent LINAC-based SRS with a median prescribed dose and planning target volume of 18Gy and 1.5 cc, respectively. Adjuvant treatments included resection and Whole Brain Radiotherapy. Seven patients underwent salvage SRS. Median age was 56 years (35-83). ER and HER-2 receptor status was recorded. Intervals between primary and CNS diagnoses and between the latter and SRS treatment were calculated. Each lesion was followed radiographically for local recurrence (increasing nodularity/size in treatment bed on MRI), and all scans were reviewed for distant recurrences (new lesions within brain outside previously treated area). RESULTS: At a median follow-up of 9.1 months, 17 of 35 patients progressed within the brain. Nine patients expired without evidence of progression. Nine patients have not progressed to date and continue to be followed. Thirty of the 73 lesions (41.1%) progressed, 11 locally and 19 distantly. The overall median time to progression was 9.2 months. Forty-three lesions did not progress. Cox regression analysis revealed significantly less progression among HER-2 receptor positive lesions (HR 0.21, 95% CI 0.06-0.69, p = 0.010) with a median CNS progression-free interval of 10.2 vs. 4.8 months for HER-2 negative lesions. Log-rank tests revealed no differences in progression with ER status or adjuvant treatments. Locally progressive lesions were associated with a longer interval between CNS diagnosis and SRS (medians: 11.8 vs. 4 months). CONCLUSIONS: Our analysis suggests that HER-2 positivity and decreased time to SRS treatment improve clinical outcomes. We will further evaluate these initial time interval analyses and continue to follow our patients.

Neuro Oncol. 2013 Nov;15(Suppl 3):iii178–iii188.

RT-045. TREATMENT OF LARGE BRAIN METASTASES WITH STEREOTACTIC RADIOSURGERY (SRS)

Amy L Zimmerman ², Michael A Vogelbaum ¹, Gene H Barnett ¹, Erin S Murphy ¹, John H Suh ¹, Lilyana Angelov ¹, Chandana A Reddy ¹, Samuel T Chao ¹

Abstract

INTRODUCTION: Large brain metastases (BMet) are not well-controlled with SRS likely due to inadequate dose. In order to minimize local failure (LF), the use of whole brain radiation (WBRT) or surgery in combination with SRS are considered. We report on our series of patients with large BMet (LBMet), >3 cm in diameter. MATERIALS AND METHODS: Our IRB-approved brain tumor database was queried for patients treated with Gamma Knife (GK) for LBMet. Variables reviewed include gender, age, radioresistant histology (defined as renal cell carcinoma, melanoma, colon, or sarcoma), RPA class, and treatment strategy. LF and distant brain recurrence (DBR) rates were calculated using cumulative incidence analysis and overall survival (OS) was calculated using Kaplan-Meier analysis. RESULTS: A total of 161 patients with 175 lesions were entered into the database. Median age was 61 (range: 29-92). RPA Class was 1 in 15 patients, 2 in 120, and 3 in 26. GK alone was the treatment approach in 121, surgery followed by GK (S + GK) to the resection bed in 32, and WBRT + GK in 22. Median GK dose was 15 Gy (10-20). Median tumor diameter was 3.5 cm (3-6.8). 1-year LF was less in patients receiving S + GK (19.3%) and WBRT + GK (12.9%), compared to GK alone (25.7%), although not statistically significant (p = 0.58). 1-year DBR was higher in patients receiving S + GK (57.7%) compared to GK alone (31.4%) or WBRT + GK (17.8%) (p = 0.0031). Median OS was highest in patients receiving S + GK at 11.2 mo, versus 8.8 mo for WBRT + GK and 6.9 mo for GK alone, although statistically not significant (p = 0.1549). Of note, 3 patients had planned GK to the same area twice, separated by 1-2 mo. At a median follow-up time of 4.5 mo, none developed LF. CONCLUSION: The treatment of LBMet with GK needs to be optimized if WBRT is to be omitted. Prospective studies are needed.

PERMALINK

RADIATION THERAPY

RT-002. PREDICTING VOXEL OUTCOMES IN PATIENTS WITH GBM TREATED WITH ANTI-ANGIOGENIC AGENTS USING INTEGRATED DOSIMETRY, METABOLIC AND PHYSIOLOGIC IMAGING

Mekhail Anwar

Janine Lupo

Annette Molinaro

Jennifer Clarke

Nicholas Butowski

Michael Prados

Susan Chang

Daphne HaasKogan

Sarah Nelson

Abstract

RT-003. INTRACTABLE SEIZURES SECONDARY TO MENINGIOANGIOMATOSIS CONTROLLED WITH SINGLE FRACTION STEREOTACTIC RADIOSURGERY

Jonathan Ashman

Joseph Drazkowski

Richard Zimmerman

Thomas Lidner

Caterina Giannini

Alyx Porter

Naresh Patel

Abstract

RT-004. PREDICTING THE INTELLECTUAL OUTCOME IN CHILDREN AND ADOLESCENT WITH BRAIN TUMORS TREATED WITH MODERN RADIATION TECHNIQUES: A POPULATION-BASED STUDY

I Atean

N Shin

A Toltz

C Laude

C Freeman

J Seuntjens

D Roberge

Abstract

RT-005. WIDE-FIELD RE-IRRADIATION WITH BEVACIZUMAB IS A FEASIBLE SALVAGE OPTION FOR PATIENTS WITH REFRACTORY HIGH GRADE GLIOMA

Michael Back

Marina Kastelan

Linxin Guo

Helen Wheeler

Abstract

RT-006. TEMOFRAC, AN INNOVATIVE AND NOVEL IRRADIATION SCHEDULE, CONCURRENT 3-TIMES DAILY ULTRAFRACTIONATED RADIATION THERAPY AND TEMOZOLOMIDE FOR NEWLY INOPERABLE GLIOBLASTOMA

Patrick Beauchesne

Guillaume Faure

Georges Noel

Thierry Schmitt

Laurent Martin

Eric Jadaud

Charlotte Carnin

Abstract

RT-007. PALLIATION OF SACRAL SPINE METASTASES WITH STEREOTACTIC BODY RADIATION THERAPY: AN EVALUATION OF LOCAL CONTROL AND TOXICITY

J Bowers

N Bennion

H Lomas IV

K Spencer

M Richardson

W McAllister

J Sheehan

D Schlesinger

R Kersh

Abstract

RT-008. CRANIO-SPINAL IRRADIATION FOR MEDULLOBLASTOMA: DOSIMETRIC COMPARISON OF PROTON AND PHOTON MODALITIES TO REDUCE THE LATE DEVELOPMENT OF DIABETES MELLITUS

Jeffrey Brower

Shae Gans

William Hartsell

Stewart Goldman

John Han-Chih Chang

Nasiruddin Mohammed

Moe Siddiqui

Vinai Gondi

Abstract

RT-009. OUTCOMES IN STEREOTACTIC BODY RADIOTHERAPY FOR SPINAL TUMORS: AN UPDATED MD ANDERSON EXPERIENCE WITH PATTERNS OF FAILURE ANALYSIS

Eva Christensen

Slade Klawikowski

Amit Garg

Mary McAleer

Laurence Rhines

James Yang

Paul Brown

Eric Chang

Stephen Settle

Amol Ghia

Abstract

RT-010. OUTCOMES IN PATIENTS WITH PAPILLARY TUMORS OF THE PINEAL REGION: SINGLE INSTITUTIONAL EXPERIENCE