Proceedings of the 49th Annual ASTRO Meeting S575

appeared to be lower than expected with conventional regimens. The safety and efficacy of this regimen should be tested in largerprospective trials.

Author Disclosure: D. Mansur, None; T. Druley, None; S. Shenoy, None; R. Hayashi, None; J. Zhang, None; K. Trinkaus, None;E. Klein, None.

2673 Impact of Total Body Irradiation Technique on Interstitial Pneumonitis and Clinical Outcomes of 623

Patients Undergoing Hematopoietic Cell Transplantation for ALL: 25 Years at the University of Minnesota

M. B. Tomblyn, T. E. DeFor, M. R. Tomblyn, M. L. MacMillan, D. J. Weisdorf, P. D. Higgins, K. E. Dusenbery

University of Minnesota, Minneapolis, MN

Purpose/Objective(s): Total body irradiation (TBI) is a standard part of the preparative regimen for myeloablative hematopoieticcell transplantation (HCT) for acute lymphoblastic leukemia (ALL). Anterior-Posterior (AP) and opposed lateral (Lat) techniqueshave been used, as have both fractionated and single-fraction schedules. Interstitial pneumonitis is a relatively common and some-times fatal complication of TBI.

Materials/Methods: We reviewed 623 consecutive patients (63% male, median age 13 years) undergoing HCT for ALL between1980 and 2005 at the University of Minnesota to evaluate the impact of TBI technique on the rates of Grade 4+ interstitial pneu-monitis (IP) and clinical endpoints including overall survival (OS), progression free survival (PFS), relapse, treatment-related mor-tality (TRM), and Grades II-IV acute and chronic graft versus host disease (GVHD). Donor types were autologous (34%), related(45%), unrelated (16%) or cord (11%). Disease status at transplant was CR1 (24%), CR2 (50%), CR3 (18%) and relapse (8%).

Results: Median follow-up among survivors was 8.3 years (1.0–22.9). Overall, Grade 4+ IP was noted in 7% of patients by 100days and 9% by one year. On multivariate analysis, there was no difference between AP and Lat techniques for IP, OS, PFS, re-lapse, or GVHD. However, there was a significantly higher TRM at one year for the AP technique (RR 1.7, p = 0.03). FractionatingTBI resulted in superior OS (RR 0.7, p = 0.04) and PFS (RR 0.7, p = 0.02) at both one and five years. TRM at one year was sig-nificantly higher with single-fraction TBI (RR 1.7, p = 0.03). IP occurred in the absence of an isolated infectious agent in 57% of thecases. The most frequent infectious cause was CMV. However, neither recipient nor donor CMV serostatus was predictive ofdevelopment of IP.

Conclusions: Fractionated TBI schedules result in improvement in OS, PFS and TRM while use of opposed lateral beams results inless TRM at one year compared to an anterior-posterior technique. Most cases of IP in our series were non-infectious, with CMVthe most common viral agent isolated.

Author Disclosure: M.B. Tomblyn, None; T.E. DeFor, None; M.R. Tomblyn, None; M.L. MacMillan, None; D.J. Weisdorf, None;P.D. Higgins, None; K.E. Dusenbery, None.

2674 Proton Radiotherapy for Ependymoma: Initial Clinical Outcomes and Dose Comparisons for Intensity

Modulated Radiation With Photons, Proton Radiation, and Intensity Modulated Proton Therapy

T. I. Yock, S. M. MacDonald, S. Safai, J. A. Wolfgang, B. Fullerton, A. Trofimov, T. R. Bortfeld, N. J. Tarbell

Massachusetts General Hospital, Boston, MA

Purpose/Objective(s): To report preliminary clinical outcomes for fifteen pediatric patients treated with proton beam radiation forependymoma. To compare the dose distributions of IMRT, 3D proton radiation, and intensity modulated proton radiation therapy(IMPT) for representative patients, one with infratentorial ependymoma and one with supratentorial ependymoma.

Materials/Methods: All children with ependymoma confined to the supratentorial or infratentorial brain treated at the Francis H.Burr Proton Facility and Harvard Cyclotron between 2002 and 2006 were included in this study. IMRT, proton, and IMPT planswere generated with similar clinical constraints for representative infratentorial and supratentorial ependymoma cases. Tumor andnormal tissue dose volume histograms were calculated and compared.

Results: Fifteen patients were treated with proton radiotherapy. Nine had a GTR and 6 had a STR; 11 were infratentorial and 4 weresupratentorial tumors. At a median follow-up of 21.6 months, 2 patients (both with local residual disease at RT treatment) had localfailures and one patient recurred in the spine; all other patients remain disease free. The crude rates of local failure and distant failurewere 13% and 6%, respectively. Two of the three failures were anaplastic ependymomas and the third had an area of focal ana-plasia. Comparable tumor volume coverage was achieved with IMRT, proton therapy, and IMPT. Substantial normal tissue sparingwas seen with the proton therapy plan as compared to IMRT. The use of IMPT will allow for additional sparing of some criticalstructures.

Conclusions: Preliminary disease control for proton radiotherapy compares favorably to the literature. Dosimetric comparisonsdemonstrate the advantage of proton radiation as compared to IMRT in the treatment of ependymoma. Further sparing of normalstructures appears possible with IMPT. Superior dose distributions were accomplished with fewer beam angles with the use ofprotons and IMPT.

S576 I. J. Radiation Oncology d Biology d Physics Volume 69, Number 3, Supplement, 2007

Author Disclosure: T.I. Yock, None; S.M. MacDonald, None; S. Safai, None; J.A. Wolfgang, None; B. Fullerton, None;A. Trofimov, None; T.R. Bortfeld, None; N.J. Tarbell, None.

2675 Successful Treatment of Pediatric Desmoids Using Radiation as a Component of Multimodality Therapy

S. Jabbari, D. Andolino, V. K. Weinberg, J. Law, W. M. Wara, K. K. Matthay, R. Goldsby, R. J. O’Donnell, D. A. Haas-Kogan

University of California San Francisco, San Francisco, CA

Purpose/Objective(s): Aggressive fibromatosis (desmoid) is a locally aggressive tumor. Recent studies suggest that pediatric des-moids are less responsive to external beam radiation therapy (EBRT) than their adult counterparts. Current studies are assessing therole of systemic therapy, and early results suggest that local therapy may be needed for postoperative residual disease. We sought toevaluate the role of radiation therapy (RT) as a component of multimodality therapy for pediatric desmoids.

Patients/Methods: Twenty-four patients with desmoids diagnosed between June 1987 and October 2005 were identified. Fivepatients without follow-up were excluded, leaving 19 evaluable patients. The median age at start of treatment was 13.1 years (range2–21) and 47% were males. Primary therapy consisted of resection alone (9 patients), resection plus EBRT (5), resection plus che-motherapy (CT; 2), EBRT alone (1) and CT alone (2). EBRT dose for all patients was between 50–60 Gy.

Results: The median overall follow-up time from start of treatment is 83.6 months (range 16–162). Following primary therapy, fivepatients achieved a complete response (CR) and remain NED for periods ranging from 31 to 102 months. Primary therapy for thesefive patients consisted of resection alone with negative surgical margins in two patients and resection with adjuvant EBRT (all withGTR and positive margins) in three patients. Of the 14 patients who recurred, only one patient had achieved a GTR with negativemargins at the time of primary therapy. Thirteen patients were treated with RT, five as a component of primary therapy, seven at thetime of recurrence, and one at the time of initial therapy as well as at the time of recurrence. Eight patients received systemic agents,three as a component of primary therapy, four at the time of recurrence, and one at the time of initial therapy as well as at the time ofrecurrence. Systemic therapy consisted of Tamoxifen alone in one patient and of chemotherapy with or without Tamoxifen or Glee-vec in the rest. Of these eight patients receiving systemic agents, one partial response was achieved following vincristine and meth-otrexate. The median times from start of therapy to first and second recurrences were 22.0 and 60.0 months, respectively. Of the 14patients who recurred, six patients had salvage therapy that did not include RT, and none is NED at last follow-up. In contrast, eightpatients received RT as a component of their final salvage therapy and six of these are NED at last follow-up. The radiation for fiveof these six successfully salvaged patients consisted of high-dose focal RT in the form of brachytherapy (2), radiosurgery (1), orintraoperative radiation therapy (2; IORT). At last follow-up, no patient has died, although toxicities of therapy have occurred. Onepatient who is NED 102 months after GTR and EBRT for a mandibular desmoid, developed papillary carcinoma of the thyroid 80months following RT. Chronic neuropathy developed in one patient treated with brachytherapy and EBRT (resulting in completefoot-drop) and in one patient treated with IORT.

Conclusions: Local control is difficult to achieve in pediatric patients with desmoid tumors. In the setting in which negative sur-gical margins cannot be achieved, RT plays a key role in achieving NED status. Even after multiple recurrences, successful salvageis achievable, particularly when high-dose focal therapy is incorporated.

Author Disclosure: S. Jabbari, None; D. Andolino, None; V.K. Weinberg, None; J. Law, None; W.M. Wara, None; K.K. Matthay,None; R. Goldsby, None; R.J. O’Donnell, None; D.A. Haas-Kogan, None.

2676 High Dose Rate Intraoperative Radiation Therapy for Locally Recurrent Neuroblastoma

D. J. Perry, M. P. LaQuaglia, K. Chan, K. A. Goodman, S. L. Wolden

Memorial Sloan-Kettering Cancer Center, New York, NY

Purpose/Objective(s): Intraoperative high dose rate brachytherapy (IORT) is used as part of multimodality management of locallyrecurrent/persistent neuroblastoma in an attempt to improve local control, while limiting toxicity. We report the experience ofIORT for neuroblastoma at a single institution.