S454 I. J. Radiation Oncology d Biology d Physics Volume 75, Number 3, Supplement, 2009

a pre-treatment PET documenting absence of nodal or distant metastases. Patient data was reviewed for local control (LC), relapse-free survival (RFS), overall survival (OS), and toxicity.

Results: Median follow-up was 17.3 months (range, 6.2–51.48). Median age was 76 years (range, 49–88), and 14 (51.8%) werefemale. Median KPS was 80 (range, 60–90), and baseline FEV1 was 1.27 L (range, 0.38–2.41). Twenty-two patients were clin-ically staged T1N0, and 5 were T2N0. Median tumor size and SUV prior to SBRT was 1.6 cm (range, 0.8–4.1) and 5.1 (range,2.2–20), respectively. Thirteen patients underwent at least one biopsy attempt. The remaining 14 patients were felt to be at highrisk for biopsy secondary to their pulmonary status. Thirteen patients had a previous history of definitively treated invasivecancer more than 2 years prior to treatment. Eighteen patients were treated with 60 Gy in 3 fx, and 9 received 50 Gy in 5fx. Two-year LC was 100% and RFS was 86%. One patient developed an isolated liver metastasis, and another failed in themediastinum. Two-year OS was 65.4%. No patient died of lung cancer. Seven patients died from progressive pulmonary dis-ease, and one from a CHF exacerbation. Toxicity included one Grade 1 and two Grade 2 chest wall toxicities, and one clinicalGrade 1 pneumonitis.

Conclusions: SBRT for well characterized radiographically diagnosed early stage lung cancer results in high rates of local control,and minimal toxicity. Death in this population is overwhelmingly related to underlying cardiopulmonary disease.

Author Disclosure: C. Robinson, None; K. Stephans, None; C. Reddy, None; T. Djemil, None; G. Videtic, None.

2606 RapidArc for Hippocampus-avoiding Prophylactic Cranial Irradiation (PCI)

B. J. Slotman, E. A. P. van der Hoorn, W. F. Verbakel, S. Senan, F. J. Lagerwaard

VU University Medical Center, Amsterdam HV, The Netherlands

Purpose/Objective(s): PCI has a definite role in the treatment of patients with limited and extended stage SCLC who respond tochemotherapy, and is being evaluated in selected patients with NSCLC. Whole brain irradiation, however, can lead to depletion ofthe radiosensitive stem cell compartment within the hippocampus. Long-term survivors after PCI may develop neuropsychologicaldeficits, which are related to hippocampus-dependent functions such as short and long-term memory and processing of spatial in-formation. Since the risk of brain metastases in this region is low, conformal avoidance of the hippocampus during PCI may beclinically relevant. RapidArc (Varian medical system) is a novel treatment technique for performing volumetric intensity modu-lated radiotherapy, during one or two gantry rotations, using variable MLC positions, dose rate and rotation speed. We investigatedthe feasibility of RapidArc for planning PCI with conformal avoidance of the hippocampi and report on the treatment in the first 3patients.

Materials/Methods: Both hippocampi were contoured on MRI scans by an experienced neuroradiologist. The mean volume was 3cc. The main objective was to reduce the dose to the hippocampi as much as possible, without jeopardizing the homogeneity of thedose distribution to the rest of the brain. PCI was delivered using a fractionation scheme of 10 x 2.5 Gy. RapidArc (v. 8.2.22) planswere generated and dose calculation was performed using Varian Eclipse (v. 8.6.3). The dose distribution, delivered using a VarianTrilogy linear accelerator, was measured using Gafchromic EBT films. Measured and calculated dose distributions were comparedusing 2D gamma evaluation with limits of 2 mm and 3% of the dose.

Results: The mean hippocampus dose in all three patients could be reduced to 12.5 – 14.1 Gy, i.e., 50–56% of the total dose. Thisequals RapidArc was able to generate homogenous dose distributions to the rest of the brain, with 97% of normal brain receiving $95% of the prescribed dose, and a dose maximum of 116%. Any dose less than 95% of prescribed dose was located within the dosefalloff around the hippocampi. There was an excellent agreement between the calculated and measured doses (Gamma value 0.34;less than 1.1% of the pixels with a gamma .1).

Conclusions: Hippocampus-sparing PCI can be delivered using RapidArc without compromising the dose to the rest of the brain.A prospective study on neuropsychological functioning after hippocampus-sparing PCI in NSCLC patients has been initiated.

Author Disclosure: B.J. Slotman, None; E.A.P. van der Hoorn, None; W.F. Verbakel, None; S. Senan, None; F.J. Lagerwaard,None.

2607 Should Extrapulmonary Small Cell Cancer be Managed Like Small Cell Lung Cancer?

S. M. Brennan, D. L. Gregory, A. Stillie, A. Herschtal, M. MacManus, D. L. Ball

Peter MacCallum Cancer Centre, Melbourne, Australia

Purpose/Objective(s): To determine if extrapulmonary small cell carcinomas (EPSCC) should be managed using protocols sim-ilar to those for small cell lung cancer (SCLC).

Materials/Methods: Treatment strategies, survival, patterns of failure and prognostic factors for patients with EPSCC were an-alyzed retrospectively at a large Cancer Center. SCLC was excluded by thoracic CT (75%) or chest radiography (25%).

Results: Of 120 eligible patients, 70% had limited disease (LD). Treatment modalities included chemotherapy (n = 82, 68%), ra-diotherapy (RT, n = 80, 67%) and surgery (n = 41, 34%). Median survival for patients with LD and extensive disease (ED) was 1.4years and 0.7 years, respectively. Gynecological (n = 31) and gastrointestinal (n = 28) were the most common primary sites.Gynecological and head and neck primary sites had better 1 year survival than other sites (p = 0.019 and 0.005, respectively). Brainmetastasis was the site of first distant failure in 4.1% vs. 35% for soft tissue metastases. Lifetime risk of brain metastasis was 13%.Radical RT (p = 0.004), LD (p = 0.028), prophylactic cranial irradiation (PCI) (p = 0.022) were positive prognostic factors andweight loss (p \ 0.001) was a negative prognostic factor on multivariate analysis.

Conclusions: Patients with EPSCC usually experienced short survival, often with early distant metastasis. Although PCI was as-sociated with improved overall survival, brain metastasis was less frequent than in SCLC and therefore the potential benefit of PCIis less than in SCLC. Radical chemo radiotherapy is associated with better outcomes and should be delivered where feasible.

Author Disclosure: S.M. Brennan, None; D.L. Gregory, None; A. Stillie, None; A. Herschtal, None; M. MacManus, None; D.L.Ball, None.