Erika L. Swanson

Comparison of Three-Dimensional (3D) Conformal Proton Radiotherapy (RT), 3D Conformal Photon RT, and Intensity-Modulated RT for Retroperitoneal and Intra-Abdominal Sarcomas

PURPOSE To compare three-dimensional conformal proton radiotherapy (3DCPT), intensity-modulated photon radiotherapy (IMRT), and 3D conformal photon radiotherapy (3DCRT) to predict the optimal RT technique for retroperitoneal sarcomas. METHODS AND MATERIALS 3DCRT, IMRT, and 3DCPT plans were created for treating eight patients with retroperitoneal or intra-abdominal sarcomas. The clinical target volume (CTV) included the gross tumor plus a 2-cm margin, limited by bone and intact fascial planes. For photon plans, the planning target volume (PTV) included a uniform expansion of 5 mm. For the proton plans, the PTV was nonuniform and beam-specific. The prescription dose was 50.4 Gy/Cobalt gray equivalent CGE. Plans were normalized so that >95% of the CTV received 100% of the dose. RESULTS The CTV was covered adequately by all techniques. The median conformity index was 0.69 for 3DCPT, 0.75 for IMRT, and 0.51 for 3DCRT. The median inhomogeneity coefficient was 0.062 for 3DCPT, 0.066 for IMRT, and 0.073 for 3DCRT. The bowel median volume receiving 15 Gy (V15) was 16.4% for 3DCPT, 52.2% for IMRT, and 66.1% for 3DCRT. The bowel median V45 was 6.3% for 3DCPT, 4.7% for IMRT, and 15.6% for 3DCRT. The median ipsilateral mean kidney dose was 22.5 CGE for 3DCPT, 34.1 Gy for IMRT, and 37.8 Gy for 3DCRT. The median contralateral mean kidney dose was 0 CGE for 3DCPT, 6.4 Gy for IMRT, and 11 Gy for 3DCRT. The median contralateral kidney V5 was 0% for 3DCPT, 49.9% for IMRT, and 99.7% for 3DCRT. Regardless of technique, the median mean liver dose was <30 Gy, and the median cord V50 was 0%. The median integral dose was 126 J for 3DCPT, 400 J for IMRT, and 432 J for 3DCRT. CONCLUSIONS IMRT and 3DCPT result in plans that are more conformal and homogenous than 3DCRT. Based on Quantitative Analysis of Normal Tissue Effects in Clinic benchmarks, the dosimetric advantage of proton therapy may be less gastrointestinal and genitourinary toxicity.

Second tumors in pediatric patients treated with radiotherapy to the central nervous system.

ObjectiveTo determine the rate of second tumors in pediatric patients treated with radiotherapy to the central nervous system (CNS) with long-term follow-up. MethodsWe retrospectively reviewed the charts of 370 consecutive pediatric patients with solid tumors and leukemia treated at the University of Florida from 1963 to 2006 with curative CNS radiotherapy. The median age was 8.1 years (range, 0.2 to 19.0 y). One hundred seventy-two (47%), 79 (21%), and 119 (32%) patients received focal, whole-brain, and craniospinal irradiation, respectively. Variables analyzed for prognostic value included primary tumor histology, patient age at primary treatment, volume of tissue irradiated, dose to the tumor bed, treatment with chemotherapy, and location of the primary tumor. ResultsEighteen second tumors were diagnosed in 16 patients. The actuarial incidences of second tumors were 3%, 8%, and 24% at 10, 20, and 30 years of follow-up, respectively. On univariate analysis, no single variable was found to be predictive of second tumor incidence. The most common second tumor after radiation for a primary solid CNS tumor was meningioma (63%), for which successful salvage was common (89%). Second gliomas were most common among patients treated for leukemia and were uniformly fatal. The most common cause of death among 5-year survivors was late relapse of their primary tumor. ConclusionsThe risk of second tumors after CNS radiation is significant and does not plateau with long-term follow-up. Most second tumors after radiotherapy for solid CNS tumors are meningiomas that can be successfully salvaged.

MULTIMODALITY LOCAL THERAPY FOR RETROPERITONEAL SARCOMA

PURPOSE Soft-tissue sarcomas of the retroperitoneum are rare tumors comprising less than 1% of all malignancies. Although surgery continues as the mainstay of treatment, the large size of these tumors coupled with their proximity to critical structures make resection with wide margins difficult to achieve. The role and timing of radiotherapy are controversial. This study updates our institutional experience using multimodality local therapy for resectable retroperitoneal sarcoma and identifies prognostic factors impacting disease control and survival. METHODS AND MATERIALS Between 1974 and 2007, 58 patients with nonmetastatic retroperitoneal sarcoma were treated with surgery and radiation at the University of Florida. The median age at radiotherapy was 57 years old (range, 18-80 years). Forty-two patients received preoperative radiotherapy and 16 received postoperative radiotherapy. Nineteen patients received 1.8 Gy once daily and 39 patients received 1.2 Gy twice daily. Variables analyzed for prognostic value included age, grade, kidney involvement, histology, de novo versus recurrent presentation, tumor diameter, margin status, radiotherapy sequencing (preoperative vs. postoperative), total radiation dose, fractionation scheme, and treatment era. RESULTS The 5-year overall survival, cause-specific survival, and local control rates were 49%, 58%, and 62%, respectively. Nearly two-thirds of disease failures involved a component of local progression. On multivariate analysis, only margin status was significantly associated with improved 5-year local control (85%, negative margins; 63%, microscopic positive margins; 0%, gross positive margins; p < 0.0001) and 5-year overall survival (64%, negative margins; 56%, microscopic positive margins; 13%, gross positive margins; p = 0.0012). Thirty-one Grade 3 or greater toxicities were observed in 22 patients, including two treatment-related deaths (3%). CONCLUSION For retroperitoneal sarcoma, local control remains a challenge and combined-modality therapy may be associated with significant acute and late morbidity. Our patterns of failure data suggest that improvements in local control may translate into a survival benefit.

Favorable Outcomes of Pediatric Patients Treated With Radiotherapy to the Central Nervous System Who Develop Radiation-Induced Meningiomas

PURPOSE To report the outcome of patients treated at the University of Florida who developed meningiomas after radiation to the central nervous system (CNS) for childhood cancer. METHODS AND MATERIALS We retrospectively identified 10 patients aged ≤19 years who received radiotherapy to sites in the craniospinal axis and subsequently developed a meningioma. We report the histology of the radiation-induced meningioma, treatment received, and ultimate outcome among this cohort of patients. RESULTS Meningioma was diagnosed at a median of 23.5 years after completion of the primary radiation. Fifty percent of second meningiomas were World Health Organization Grade 2 (atypical) or higher. All cases were managed with a single modality: resection alone (n = 7), fractionated radiotherapy (n = 2), and stereotactic radiosurgery (n = 1). The actuarial event-free survival and overall survival rate at 5 years after treatment for a radiation-induced meningioma was 89%. Three patients who underwent resection for retreatment experienced a Grade 3 toxicity. CONCLUSIONS Radiation-induced meningiomas after treatment of pediatric CNS tumors are effectively managed with single-modality therapy. Such late-effect data inform the overall therapeutic ratio and support the continued role of selective irradiation in managing pediatric CNS malignancies.

Radiotherapy for basal cell carcinoma of the medial canthus region

To report outcome for patients treated with radiotherapy (RT) for basal cell carcinoma of the medial canthus.

Analysis of Dose at the Site of Second Tumor Formation After Radiotherapy to the Central Nervous System

PURPOSE Second tumors are an uncommon complication of multimodality treatment of childhood cancer. The present analysis attempted to correlate the dose received as a component of primary treatment and the site of the eventual development of a second tumor. METHODS AND MATERIALS We retrospectively identified 16 patients who had received radiotherapy to sites in the craniospinal axis and subsequently developed a second tumor. We compared the historical fields and port films of the primary treatment with the modern imaging of the second tumor locations. We classified the location of the second tumors as follows: in the boost field; marginal to the boost field, but in a whole-brain field; in a whole-brain field; marginal to the whole brain/primary treatment field; and distant to the field. We divided the dose received into 3 broad categories: high dose (>45 Gy), moderate dose (20-36 Gy), and low dose (<20 Gy). RESULTS The most common location of the second tumor was in the whole brain field (57%) and in the moderate-dose range (81%). CONCLUSIONS Our data contradict previous publications that suggested that most second tumors develop in tissues that receive a low radiation dose. Almost all the second tumors in our series occurred in tissue within a target volume in the cranium that had received a moderate dose (20-36 Gy). These findings suggest that a major decrease in the brain volume that receives a moderate radiation dose is the only way to substantially decrease the second tumor rate after central nervous system radiotherapy.

SU‐GG‐T‐562: Dosimetric Comparison of 3D Conformal Proton Therapy and IMRT for Retroperitoneal‐Sarcoma Treatments

Purpose: To determine the difference in dose to target and critical structures when treating retroperitoneal sarcoma with 3D conformal proton therapy or IMRT, and to evaluate the sensitivity to setup errors for both techniques. Method and Materials: 3D conformal proton,IMRT, and 3D conformal photon plans were made for 8 patients. GTV volumes varied between 277 and 3482cc. A uniform 2cm GTV to CTV margin was applied, followed by a 0.5cm CTV to PTV margin. Laterally the proton plans were conformed to the PTV, while a distal margin equal to 1.5% of the range plus 1mm was applied to the CTV. Range‐compensator smearing of 0.5cm was used. For proton plans 2 to 4 beams were used; 5 to 8 for photon plans. Prescribed dose is 50.4CGE. To evaluate the effect of setup uncertainty on target coverage and dose to critical structures, proton and IMRT plans were re‐calculated for 5mm shifts along the cardinal axes. Pinnacle (Philips) treatment planning system was used to generate photon plans; Eclipse (Varian) to generate proton plans. Results: The bowel V15Gy is lower in the proton plan (19.9% versus 51.7% averaged over all patients), while V45Gy is near identical (6.9% versus 6.3%). Mean dose to liver is 5.8 and 13.2Gy for proton and IMRT respectively. When not involved in the CTV, the V20Gy of the ipsilateral kidney is significantly lower in proton plans (by up to 45%). Both techniques avoid the contralateral kidney giving comparable V20Gy (0.2% and 2.3%). Conclusion: Conformity of the dose to the target is comparable for both techniques. Critical‐structure volumes receiving high doses are comparable, while the volume receiving low dose is reduced significantly in proton plans. IMRT plans are less sensitive to variations in setup.