Volume 90 � Number 1S � Supplement 2014 Poster Viewing Abstracts S321

clear that this high a dose is necessary to control small lesions, and con-

cerns over neurotoxicity/radionecrosis have led many institutions,

including ours, to routinely use lower SRS doses for metastases � 20 mm.

The outcomes of this approach are unclear.

Materials/Methods: We examined patients with brain metastases up to 20

mm in maximal diameter who were treated with Linac-based SRS to a

dose of 18 or 20 Gy at our institution from 1998 to 2012. Local progression

was defined as an increase in the diameter of the contrast-enhancing lesion

of at least 25% on any follow-up MRI or any clinical progression or

toxicity requiring surgical intervention. Patients were not censored for

receipt of subsequent SRS to other lesions. Local control (LC) and overall

survival (OS) were estimated with the Kaplan-Meier method, and Cox

regression analysis was performed to determine factors associated with

LC.

Results: Ninety-two patients (92 lesions) were identified, with median age

of 58 (range, 30-99), median KPS of 90, and median time from cancer

diagnosis to brain metastasis diagnosis of 23 months. The most common

histologies were breast (35%), lung (28%), melanoma (18%), and renal

cell carcinoma (9%). Extracranial disease was progressive in 43% of pa-

tients. Median brain metastasis diameter was 9 mm (range, 2-20 mm). SRS

was delivered as salvage treatment in 43% of cases. 86% of patients

received SRS alone, while 14% received SRS plus whole brain radiation

therapy (WBRT). Median SRS normalization was 77% (range, 69-87%).

SRS dose was 18 Gy in 43% of cases and 20 Gy in 57% of cases. Median

radiographic follow-up and OS after SRS were 5.0 months and 10.0

months, respectively. The 6-month LC rates were 75% and 86% for 18 and

20 Gy, respectively. LC at 12 months was 65% for both 18 and 20 Gy.

Younger age (P Z .02) and progressive extracranial disease (P Z .03)

were associated with improved LC, whereas tumor size, histology, receipt

of WBRT, and dose of 18 vs 20 Gy were not associated with LC. Two

patients (2%) underwent surgery for pathologically-confirmed

radionecrosis.

Conclusions: LC after SRS using doses of 18 or 20 Gy for small brain

metastases� 20 mm was similar to that reported in the literature for higher

doses, and had a lower risk of radionecrosis requiring surgery compared to

other series. The finding that LC was higher among patients with pro-

gressive systemic disease reflects shorter brain MRI follow-up among

these patients, and highlights the competing risk of death among many

patients with brain metastases and the importance of considering both

efficacy and toxicity in SRS dose selection.

Author Disclosure: J. Burkeen: None. B.M. Alexander: None. M.C.

Horvath: None. L.W. Christianson: None. M.A. Dyer: None. I.F. Dunn:

None. A.J. Golby: None. M.D. Johnson: None. E.B. Claus: None. E.A.

Chiocca: None. E.Q. Lee: None. N.U. Lin: None. K.J. Marcus: None. S.

Friesen: None. E.G. Mannarino: None. M. Wagar: None. F.L. Hacker:

None. N.D. Arvold: None.

2262The Effect of Energy Index on Brain Metastases Local Control WithStereotactic RadiosurgeryA. Jani, A. Yaeh, T. Nanda, Y. Qureshi, S. Saad, W. Feng, S. Isaacson,

M. Sisti, J. Bruce, G. McKhann, J. Lesser, A. Lassman, and T.J. Wang;

Columbia University Medical Center, New York, NY

Purpose/Objectives: The integral dose to a volume is the area under the

curve of a differential dose-absolute volume histogram. The energy index

is a measure of dose homogeneity within a target volume. It is calculated

by the integral dose divided by the product of prescription dose and tumor

volume. We hypothesize that a higher energy index is associated with

greater local control for brain metastases (BM) treated by stereotactic

radiosurgery (SRS).

Materials/Objective(s): We reviewed all patients treated with SRS from

November 2011 to December 2013 at our institution for one or multiple

BM. Data regarding the prescription dose, prescription isodose line,

minimum dose, mean dose, integral dose, tumor volume, and energy index

were collected. Tumor response was assessed by reviewing follow-up brain

imaging studies and classified according to the Response Evaluation

Criteria in Solid Tumors (RECIST). Local control was calculated using the

Kaplan-Meier method. Dosimetric prognostic factors for local control

were assessed using Cox proportional hazards regression analysis. Uni-

variate and multivariate analyses were performed for local failure. Statis-

tical analysis was performed using SPSS, version 20.

Results: Eighty-eight patients were treated with SRS for BM. Twenty

patients underwent multiple SRS procedures. Sixty-three patients had

follow-up imaging available for our review. Two hundred and thirty nine

individual tumors were identified in this cohort of patients. The median

follow-up was 6 months (range 1-26 months). The median prescription

dose was 20 Gy (range 13-20 Gy). The mean integral dose was 22.2 mJ.

The mean energy index was 1.66 mJ/cm3/Gy. The mean pre-treatment

tumor volume was 0.831 cm3. The mean post-treatment tumor volume was

0.456 cm3. The local control rates at 6-, 12-, and 18-months were 95.7%,

84.3%, and 81.3%, respectively. Of the 239 individual tumors, 95 achieved

a complete response (39.7%), 101 achieved a partial response (42.3%), 23

showed stable disease (9.6%), and 20 showed progression of disease

(8.4%). On univariate analysis, prescription dose was the only variable

significant for local failure: HR 0.798 (95% CI Z 0.640-0.995, p Z0.045). On multivariate analysis, none of the above variables were statis-

tically significant for local failure, including the integral dose and energy

index.

Conclusions: Stereotactic radiosurgery for BM results in excellent local

control. A one-year 84.3% local control rate was achieved and was inde-

pendent of the tumor integral dose and energy index. Our study population

was enriched with small tumors and further analysis is needed to determine

if the same conclusion is true for tumors of all sizes.

Author Disclosure: A. Jani: None. A. Yaeh: None. T. Nanda: None. Y.

Qureshi: None. S. Saad: None. W. Feng: None. S. Isaacson: None. M.

Sisti: None. J. Bruce: None. G. McKhann: None. J. Lesser: None. A.

Lassman: None. T.J. Wang: None.

2263Adult Medulloblastoma: Does Craniospinal Dose Matter?A.J. Song,1 A. Kumar,2 E.S. Murphy,2 T. Tekautz,2 M. Ahluwalia,2

D. Peereboom,2 G. Stevens,2 J.H. Suh,2 and S.T. Chao2; 1Case Western

Reserve University, Cleveland, OH, 2Cleveland Clinic, Cleveland, OH

Purpose/Objective(s): We investigated outcomes of medulloblastoma in

adults (age � 18 years), given that these patients may not do as well

compared to the pediatric patient population.

Materials/Methods: Adult medulloblastoma patients diagnosed between

1992 and 2011 were identified through an IRB-approved database. Patients

received definitive treatment with surgery, radiation therapy (RT), and/or

chemotherapy. JMP 11 was used for statistical analyses.

Results: Thirty patients (median age 30, range 18-44, 15 males and 15

females) were identified with median follow-up of 28 months. Tumor

location was central in 17 (57%) and lateral in 13 patients (43%). Gross

total resection (GTR) was performed in 23 patients (79%). Staging was M0

(90%), M1 (3.3%), and M3 (6.7%). The 5-year overall survival (OS) and

progression-free survival (PFS) was 70% and 45%, respectively for all

patients, and 74% and 54%, respectively, for 21 M0 with GTR patients.

One patient with incomplete information was excluded from further

analysis. Thirteen patients (43%) developed recurrence with median time

of 26 months. RT data available for 20 patients, and described as two

groups: low dose (<36 Gy; 6 patients) and high dose(�36 Gy, 14 patients)

CSI, with total tumor bed dose 54-57.6 Gy. Median dose/fraction in both

groups was 1.8 Gy. Radiation-induced meningioma was observed in 1

patient with high dose CSI. Freedom from progression (FFP) was not

affected by CSI dose (pZ 0.213). FFP was improved by GTR (pZ 0.029)

and worsened by VP shunt placement (p Z 0.023) as well as older age (p

Z 0.030). OS was decreased if planned chemotherapy was not completed

(p Z 0.017). Tumor location (p Z 0.025), time to complete RT (p Z0.005), and time between surgery and RT (p<0.001) affected PFS. For M0

with GTR patients, [average risk (AR)], RT data was available for 16

patients. FFP was also not affected by CSI dosing in AR group (p Z0.4652). OS and PFS in this subgroup appeared to be affected by tumor

International Journal of Radiation Oncology � Biology � PhysicsS322

location (p Z 0.025 for OS, p Z 0.009 for PFS), KPS at initial

RT consult (p Z 0.026, p Z 0.019), use of adjuvant chemotherapy

(p Z 0.012, p Z 0.0213), and postponement of RT during treatment

(p Z 0.006, p Z 0.0073). Gender appeared to affect PFS, with females

doing better (p Z 0.008). Not completing planned chemotherapy

approached significance for OS (p Z 0.057).

Conclusions: Adult medulloblastoma patients treated with radiation ther-

apy did not have different rates of recurrence with the use of high or low

dose CSI. GTR, age, and VP shunt may impact recurrence. AR adult

medulloblastoma may be treated with low dose CSI with the use of

adjuvant chemotherapy. If patient is anticipated not to be able to complete

chemotherapy, high dose CSI may be favored for AR.

Author Disclosure: A.J. Song: None. A. Kumar: A. Employee; Cleveland

Clinic. E.S. Murphy: A. Employee; Cleveland Clinic. T. Tekautz: A.

Employee; Cleveland Clinic. M. Ahluwalia: A. Employee; Cleveland

Clinic. D. Peereboom: A. Employee; Cleveland Clinic. G. Stevens: A.

Employee; Cleveland Clinic. J.H. Suh: A. Employee; Cleveland Clinic.

S.T. Chao: A. Employee; Cleveland Clinic.

2264Use of CT Cisternograms in Target Definition for TrigeminalNeuralgia RadiosurgeryS.A. Althaqfi; The Ottawa Hospital Cancer Center, Ottawa, ON, Canada

Purpose/Objective(s): Radiosurgery (RS) is an effective treatment for

trigeminal neuralgia (TG). RS provides effective pain relief in 60 to 80 %

of pts. For RS planning it is critical to identify the trigeminal nerve to

ensure safe and effective RS. We describe the use of CT Cisternograms

(CS) to help define RS target in patients with contraindication to MRI or

where the nerve is difficult to visualize because of patient and anatomic

factors.

Materials/Methods: Three patients with TG underwent CT cisternogram

as preparation for radiosurgery planning. Two of the patients had received

prior RS for TG. 1 patient had a pacemaker as contraindication to MRI. In

the other 2 cases the TG nerve was difficult to visualize on MRI because of

disrupted anatomy from prior micro-vascular decompression surgery. CS

procedure was performed by a neuro-radiologist. 10 cc of Isovue-300

contrast was injected into the thecal sac under fluoroscopic guidance at L3-

L4 level. Patient is maintained in a Trendelenburg position and transferred

to the diagnostic CT scanner. One mm axial images were obtained through

the brain. CS was imported into the RS planning system and fused with the

CT simulation scan for targeting.

Results: There were no peri-procedural or post-procedural complications

following CS. There was excellent visualization of CSF spaces and cranial

nerves involving the basal cisterns. The Trigeminal Nerve was easily

visualized in the cisternal portion by 3 RS physicians. The Nerve was

targeted adjacent to the root entry zone. For 2 patients being retreated 50

Gy was prescribed to the 80 % isodose line. The 3rd pt was treated with 60

Gy prescribed to 80 isodose line. All patients had complete pain relief with

no complications to date from RS.

Conclusions: CT Cisternogram is a safe and effective technique to visu-

alize the cisternal portion of the trigeminal nerve for RS. CS can be used in

pts who have contra-indications to MRI or for cases in which MRI is

insufficient for identifying the trigeminal nerve.

Author Disclosure: S.A. Althaqfi: None.

2265Outcomes From First 3 Years of Frameless StereotacticRadiosurgery in Treating Brain MetastasesA. Taggar, H. Lau, G. Lim, R. Nordal, R. Khan, K. Breitman, A. Hudson,

D. Spencer, and J. Voroney; Tom Baker Cancer Centre, Calgary, AB,

Canada

Purpose/Objective(s): Stereotactic radiosurgery (SRS) is an important

tool in management of brain metastases. Frameless SRS uses image

guidance to deliver high dose radiation to the target volume, while being

minimally invasive. Frameless techniques have improved patient comfort

and allow flexibility in scheduling radiosurgery delivery. Previous case

series have low patient numbers or short follow-up. In this study we report

clinical outcomes and our center’s experience with SRS in treating newly

diagnosed brain metastases.

Materials/Methods: We reviewed electronic charts of all patients un-

dergoing SRS at the institution between June 2010 and September 2013.

A total of 235 lesions in 120 patients were treated in 130 SRS sessions.

Median age of patients undergoing SRS was 58.5 years (range 26-81).

Median Karnofsky Performance Status (KPS) was 80 (range 60-100).

Patients had non-small cell lung cancer (51), breast cancer (21), mela-

noma (20), renal cell carcinoma (10), colorectal cancer (7) and others

(13). Number of lesions treated in a single session ranged from one to

four.

Results: Median follow-up was 12.7 months (range 0-60). Overall median

survival from the time of brain mets diagnosis was 18.2 months (95% CI

Z 14-22 months) and 10.4 months from time of SRS (95% CI Z 7-13

months). Seventy four (62%) patients received whole brain radiation

therapy (WBRT) prior to SRS. Median survival from the time of brain

mets diagnosis was 22.3 months for patients treated with WBRT plus SRS

versus 9.3 months for patients treated with SRS (95% CI Z 14-30 months

and 1-17 months respectively). Overall, fifty percent of patients received

palliative chemotherapy after SRS (55% if KPS �80; 33% if KPS �70; no

difference if prior WBRT). Median survival for patient who received

chemotherapy was 22.3 months versus 12.3 months who did not. We

identified local failure rate (within SRS site) of 11%, whereas 38% of

patients had distant intracranial relapse; there was no difference between

those who did or did not receive WBRT prior to SRS. Radionecrosis was

identified in 8% of treated lesions.

Conclusions: This is so far the largest frameless SRS series. Our results are

comparable if not better than previously reported. Frameless SRS is an

effective, non-invasive option for patients with less than four metastatic

lesions as well as patients with residual or growing lesions post WBRT.

Patients who underwent chemotherapy post SRS had better overall survival

regardless of WBRT prior to SRS or number of lesions treated with SRS.

Author Disclosure: A. Taggar: None. H. Lau: None. G. Lim: None. R.

Nordal: None. R. Khan: None. K. Breitman: None. A. Hudson: None.

D. Spencer: None. J. Voroney: None.

2266Stereotactic Radiation Therapy for Brain Metastases From AdvancedNon-Small Cell Lung Cancer With or Without Endothelial GrowthFactor Receptor (EGFR) Mutations: Influence of GefitinibTreatmentT. Murai,1 K. Yamada,2 F. Oshita,2 K. Sato,3 K. Tatewaki,4 N. Yokota,5

S. Ohta,6 H. Iwata,7 M. Iwabuchi,8 S. Sadao,9 and Y. Shibamoto8; 1Nagoya

City university, Nagoya, Japan, 2Department of Thoracic Oncology,

Kanagawa Cancer Center Hospital, Yokohama, Japan, 3Japanese Red

Cross Medical Center, Tokyo, Japan, 4Yokohama CyberKnife Center,

Yokohama, Japan, 5Suzukake Central Hospital, Hamamatsu, Japan,6Yokohama CyberKnife Center, Yokohama, Japan, 7Nagoya Proton

Therapy Center, Nagoya City West Medical Center, Nagoya, Japan,8Nagoya City University Graduate School of Medical Sciences, Nagoya,

Japan, 9Department of Public Health, Nagoya City University Graduate

School of Medical Sciences, Nagoya, Japan

Purpose/Objective(s): Gefitinib is a tyrosine kinase inhibitor of EGFR

that have shown efficacy against non-small cell lung cancer (NSCLC) with

EGFR mutations. This agent can also pass through the blood-brain barrier

and is expected to be effective against brain metastases (BM). The purpose

of this study is to examine the relationship between gefitinib treatment and

outcome of stereotactic radiation therapy (SRT) against BM.

Materials/Methods: Patients with BM from NSCLC were treated with

SRT (22 Gy/1 fx or 35 Gy/3 fx) as a first-line therapy for BM between

2007 and 2011. After the treatment, patients were evaluated at 2- to 3-

month intervals using contrast-enhanced magnetic resonance imaging.

When new BM developed, they were treated again with SRT or whole

brain radiation therapy. Regarding systemic therapy, our policy was as