Dosimetric Study of Current Treatment Options for Radiotherapy in Retinoblastoma

International Journal of

Radiation Oncology biology

physics

www.redjournal.org

Clinical Investigation: Central Nervous System

Dosimetric Study of Current Treatment Options for Radiotherapy in Retinoblastoma Eman Eldebawy, M.D.,*,z William Parker, M.Sc., F.C.C.P.M.,y Wamied Abdel Rahman, Ph.D., F.C.C.P.M.,y and Carolyn R. Freeman, M.B.B.S., F.R.C.P.C.* *Departments of Radiation Oncology, and yMedical Physics, McGill University Health Centre, Montreal, Quebec, Canada; and zDepartment of Radiation Oncology, Children’s Cancer Hospital, Cairo, Egypt Received May 7, 2010, and in revised form Jul 6, 2011. Accepted for publication Jul 27, 2011

Summary The purpose of this article is to compare current radiation treatment options in the treatment of pediatric retinoblastoma. Three previously treated patients are replanned with 10 modern radiotherapy techniques including basic 3D conformal planning and electron radiotherapy to intensity modulated and stereotactic techniques. A dose volume analysis of the treatment plans is performed, and plan evaluation metrics used for comparison. We found that arc based IMRT techniques provided the most conformal plans while affording the greatest critical structure and normal tissue sparing.

Purpose: To determine the best treatment technique for patients with retinoblastoma requiring radiotherapy to the whole eye. Methods and Materials: Treatment plans for 3 patients with retinoblastoma were developed using 10 radiotherapy techniques including electron beams, photon beam wedge pair (WP), photon beam three-dimensional conformal radiotherapy (3D-CRT), fixed gantry intensitymodulated radiotherapy (IMRT), photon volumetric arc therapy (VMAT), fractionated stereotactic radiotherapy, and helical tomotherapy (HT). Dose-volume analyses were carried out for each technique. Results: All techniques provided similar target coverage; conformity was highest for VMAT, nine-field (9F) IMRT, and HT (conformity index [CI] Z 1.3) and lowest for the WP and two electron techniques (CI Z 1.8). The electron techniques had the highest planning target volume dose gradient (131% of maximum dose received [Dmax]), and the CRT techniques had the lowest (103% Dmax) gradient. The volume receiving at least 20 Gy (V20Gy) for the ipsilateral bony orbit was lowest for the VMAT and HT techniques (56%) and highest for the CRT techniques (90%). Generally, the electron beam techniques were superior in terms of brain sparing and delivered approximately one-third of the integral dose of the photon techniques. Conclusions: Inverse planned image-guided radiotherapy delivered using HT or VMAT gives better conformity index, improved orbital bone and brain sparing, and a lower integral dose than other techniques. Ó 2012 Elsevier Inc. Keywords: IMRT, Pediatric, Radiotherapy, Retinoblastoma, Treatment planning

Reprint requests to: William Parker, M.Sc., F.C.C.P.M., Department of Medical Physics L5-112, McGill University Health Centre, 1650 Cedar Avenue, Montreal, Quebec, Canada H3G 1A4. Tel: (514) 934-8052; Fax: (514) 934-8229; E-mail: [email protected] Int J Radiation Oncol Biol Phys, Vol. 82, No. 3, pp. e501ee505, 2012 0360-3016/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ijrobp.2011.07.024

Conflict of interest: none. Supplementary material for www.redjournal.org.

this

article

can

be

found

at

e502 Eldebawy et al.

International Journal of Radiation Oncology  Biology  Physics

Introduction

IMRT fields (9F) with 6-MV beam using beams at 0 , 40 , 80 , 120 , 160 , 200 , 240 , 280 , and 320 and a sliding window delivery technique; (8) two VMAT delivery beams with 6 MV, using continuous 360 arc IMRT delivery; (9) stereotactic radiotherapy with five noncoplanar arcs (FSRT), using the following table and gantry parameters (table, gantry start angle, gantry stop angle in degrees), 90 to 30 e130 , 315 to 230 e330 , 0 to 230 e330 , 60 to 30 e130 , and 10 to 120 e150 , respectively, with blocking directly to the PTV; and (10) HT delivery with 1-cm collimator, 0.28 pitch, and unrestricted gantry angles. Field arrangements for the 10 techniques are shown in Figure 1. Dose distributions were calculated with the (1) Eclipse (Varian Medical Systems, Palo Alto, CA) treatment planning system (using the pencil beam algorithm module) for the WP, CRT 6F, and CRT 9F techniques; (2) Eclipse treatment planning system (electron Monte Carlo module) for en face anterior and oblique electron techniques; (3) Eclipse treatment planning system (IMRT module) for the IMRT 6F and IMRT 9F techniques; (4) Eclipse treatment planning system (Rapid Arc module) for the VMAT techniques; (5) Brain Lab (Brain Scan) treatment planning system for the FSRT technique; and (6) Tomotherapy treatment planning system for the HT technique. Plans were designed to minimize the dose delivered to the organs at risk without compromising target coverage. For direct comparison of the treatment plans, all dose volume data were exported into a spread sheet. The prescribed dose was 45 Gy in 25 fractions, and the prescription method depended on the planning technique. For the WP and 3D-CRT 6F and 9F techniques and the FSRT technique, dose was prescribed so that 100% of the PTV was covered by 95% of the prescription dose, i.e., a maximum dose of 100% (D100%) Z 95%. For the two electron beam techniques, the dose was prescribed to the 90% isodose volume. For the VMAT, HT, and two IMRT techniques, the plan was prepared such that 95% of the PTV received the prescription dose, i.e., V45Gy Z95%. Plans were compared using dose-volume parameters obtained from dose-volume histograms (DVH) calculated for each plan. Target volume coverage was evaluated by comparing the volume of the PTV receiving 95% (V95%) with that receiving 99% (V99%) of the prescription dose, as well as the Dmax received. Conformity indices (CI), commonly used in radiosurgical applications, were used to assess the volume of normal tissue irradiated by the prescription dose, and although spatial information was not accounted for, these indices reflect the dose conformality to the target. The CI95% is defined as the absolute volume of tissue receiving at least 95% of the prescription dose relative to that of the PTV and was used to evaluate techniques that were prescribed such that 95% of the dose covered the target volume. The CI100% (the ratio of the volume of tissue receiving the prescription dose to the PTV volume) was used to evaluate the other techniques: V95% CI95% Z VPTV and V100% CI100% Z VPTV For organs at risk, other DVH metrics, also obtained from DVH analysis, were used. Volume indices such as the volume of the structure receiving at least 20 Gy (V20Gy), V15Gy, V10Gy, V5Gy, and V2Gy were extracted from DVH data. Metrics such as Dmax and mean dose received by the organ were also used when relevant. Integral dose is a measure of the energy (typically, in units of joules [J]) deposited in the patient with radiation. For this study, the integral

Retinoblastoma is a radiosensitive tumor that typically affects children less than 4 years of age. It is the most common intraocular malignant tumor in early childhood and the second most common tumor in all age groups after choroidal melanoma (1). Retinoblastoma affects approximately 1 child in 20,000. It is seen in hereditary (40%) and nonhereditary (60%) forms (2). Management of retinoblastoma has changed considerably over recent years as a result of the availability of local treatment options such as photocoagulation (3e6), cryotherapy (7, 8), and plaque therapy (9e13) and new approaches involving various combinations of chemotherapy agents (14e17). External beam radiotherapy, although an established and extremely effective treatment for retinoblastoma, now is often regarded as a last resort because of the risk of late side effects, most especially the risk of second cancers. All the new radiotherapy (RT) delivery techniques including intensity-modulated RT (IMRT) and fractionated stereotactic RT (FSRT) and new technologies such as volumetric arc therapy (VMAT) and helical tomotherapy (HT) and new modalities such as protons provide better sparing of normal structures than previously possible. This study focused on the evaluation of the various radiotherapy techniques that could be used for treatment of retinoblastoma when the target volume was the whole globe. Conventional photon beam RT techniques including anterior and lateral wedged pairs (WP), multifield three-dimensional conformal RT (3D-CRT), and electron beam therapy, as well as dynamic multileaf IMRT, VMAT, HT, and FSRT, were compared. Dose-volume parameters and statistics were used to assess the ability of each technique to minimize the dose to normal structures while meeting target dose requirements.

Methods and Materials Three children with retinoblastoma, each of whom required treatment to one eye, were anesthetized and immobilized in the supine position by using a thermoplastic mask system and scanned using an AcQSim (Philips, Cleveland, OH) computed tomography (CT) simulator with 3-mm-thick slices. The clinical target volume (CTV) was defined as the globe of the eye and proximal 0.5 cm of the optic nerve. The planning target volume (PTV) was defined as the CTV plus a uniform 3D 3-mm margin. Organs at risk were delineated on the CT datasets, including the ipsilateral and contralateral bony orbits, the whole brain, the contralateral eye and lens, the lacrimal glands, and the optic chiasm. Ten treatment techniques were planned for comparison purposes, as follows: (1) anterior and lateral WP photon technique using beams at 0 and 330 with 60 wedges and conformal multileaf collimator (MLC) blocking (7 mm to the PTV to account for penumbra); (2) 3D-CRT with six 6-MV beams (CRT 6F) using beams at 0 , 40 , 120 , 160 , 200 , and 320 and conformal MLC blocking (7 mm to the PTV); (3) 3D-CRT with nine 6-MV beams (CRT 9F), using beams at 0 , 40 , 80 , 120 , 160 , 200 , 240 , 280 , and 320 , and conformal MLC blocking (7 mm to the PTV); (4) en face anterior 9-MeV electron technique using a single direct beam at 0 and a conformal alloy cutout for blocking (7 mm to PTV); (5) oblique 9 MeV electron technique using a single direct beam at 340 and a conformal alloy cutout for blocking (7 mm to PTV); (6) six dynamic delivery IMRT fields (6F) with 6-MV beam using beams at 0 , 40 , 120 , 160 , 200 , and 320 and a sliding window delivery technique; (7) nine dynamic delivery

Volume 82  Number 3  2012 dose was calculated as the product of the mean dose in Gy to the external or surface contour and the mass of the external contour in kilograms. For simplicity, the mass of the external contour was taken as the product of its volume and a tissue density of 1 g/cm3.

Radiotherapy for retinoblastoma e503

Results Target volume coverage For the eight photon techniques, 100% of the PTV was covered by at least 95% of the prescription dose, i.e., V95% Z 100%. For the two electron techniques, 95% of the prescription dose covered 95% and 97% of the PTV for the en face technique and the electron oblique technique, respectively. Supplementary Table E1 gives values of V95%, V99%, and Dmax and CI95% or CI100% averaged over the three cases for the 10 different techniques. As expected for the WP, CRT 6F, and CRT 9F techniques, average V99% values were 79%, 72%, and 81%, respectively, while the average V99% value was 90% for the other seven techniques. The PTV Dmax for the WP, CRT 6F, and CRT 9F techniques was lower than that for the other seven techniques, having an average value of 103% of the prescription dose. The Dmax for the other five photon-based techniques (VMAT, HT, IMRT 6F, IMRT 9F, and FSRT) was slightly higher, having a value approximately 110% of the prescription dose. It was much higher for the two electron techniques, having a value more than 125% of the prescription dose. The CI95% ranged from 1.3 to 1.8 and was lowest for the VMAT and IMRT 9F techniques and highest for the WP and two electron techniques. The CI100%, for obvious reasons, shows a strong correlation with the prescription method. For example, the prescription for the WP, CRT 9F, and the CRT 6F techniques was made to the isocenter of the beam arrangements, and the CI100% for these three techniques was less than 1.0.

Orbital bone The mean volume of ipsilateral bony orbit receiving more than 20 Gy, a suggested threshold for bone growth inhibition, averaged over the three cases, was 56% for the VMAT and HT techniques, 63% for the FSRT technique, 85% for the WP technique, approximately 69% for the two electron techniques, approximately 77% for the two IMRT techniques, and approximately 89% for the two CRT techniques. The V20Gy for the ipsilateral bony orbit was lowest for the VMAT and HT techniques and highest for the CRT techniques. Supplementary Table E2 gives average values of V15Gy, V20Gy, and V25Gy for the orbital bone for the 10 techniques.

Brain Generally, the electron beam techniques and FSRT were superior in terms of brain sparing compared to the other seven techniques. The brain volumes receiving 5, 10, and 15 Gy were greatest with the WP technique. Supplementary Table E3 gives the average values of V5Gy, V10Gy, and V15Gy in units of cm3 for the brain for the 10 techniques.

Dose to unspecified tissues and integral dose Fig. Beam arrangements and dose distributions for the 10 treatment techniques are compared. The orange-red color represents the prescription dose (45 Gy) or higher, green represents approximately 25 Gy, and blue represents 5 Gy.

The volumes of unspecified tissue receiving doses greater than 2 Gy, 5 Gy, and 25 Gy (V2Gy, V5Gy, and V25Gy) are listed in supplementary Table E4, as well as the mean and integral doses averaged for the 3 patients. The tissue volume receiving at least 2 Gy was greatest for the FSRT and VMAT techniques and least for

e504 Eldebawy et al.

International Journal of Radiation Oncology  Biology  Physics

the two electron techniques. The average V5Gy was considerably lower for the two electron techniques than for the eight photonbased techniques. When we compared the photon-based techniques with each another, the V5 Gy values for the VMAT, HT, and FSRT and the two IMRT techniques were slightly lower than those for the 3D-CRT and WP techniques. The mean integral dose for the eight photon-based techniques ranged between 7 J and 9 J and was much lower for the two electron-based techniques, having a value of approximately 3 J.

both techniques, the mean relative volume of the ipsilateral bony orbit treated to a dose above 20 Gy was 56%. Moreover, this was achieved despite a larger CTV in our study that included the proximal 0.5 cm of the optic nerve. There is no consensus in the literature as to whether it is necessary to include the proximal part of the optic nerve in the CTV or not, but if the proximal part of the optic nerve is not considered to be at risk and not included in the target volume, the dose to the bony orbit would be reduced even further using these new treatment techniques. Treatment room imaging technologies such as EPIDs (electronic portal imaging device), CBCT (cone-beam computed tomography), MVCT (mega-voltage computed tomography), and CT-on-rails allow, on a daily basis, accurate patient setup as well as tracking of the target or adjacent bony landmarks. With the use of these technologies and rigid immobilization, it is possible to use a smaller PTV for planning (34, 35). For our 3 patients, FSRT plans using a margin for the PTV of 1 mm resulted in further sparing of the ipsilateral bony orbit: the mean relative V20Gy of the ipsilateral bony orbit was reduced to 44% (approximately 1.5 cm3 less than those of the VMAT and HT techniques). Treatment with a smaller PTV margin is feasible using newer treatment units such as the Novalis TX treatment unit (Varian Medical Systems, Palo Alto, CA; and BrainLab, Munich, Germany) using the (1) BrainLab stereotactic thermoplastic mask system for patient immobilization; (2) a vacuum contact lens for immobilization of the globe; (3) CBCT and kV images for patient setup; and (4) ExacTrac for online tracking of the bony landmarks surrounding the PTV volume or possibly of implanted fiducials. Frameless stereotactic radiosurgery using a “zero” PTV margin (36, 37) may even be an option if eye immobilization and precision can be assured in this way, resulting in further reduction in dose to the ipsilateral bony orbit. Although the effects of radiation to a portion of the brain during treatment for retinoblastoma are not well documented, this exposure is another important consideration in choice of technique. In our study, FSRT provided the greatest sparing of brain, and further reductions can be expected by using further reduced PTV margins. A major concern with respect to the use of IMRT in children is the increased exposure of tissues outside the target to low-dose radiation that may be associated with an increased risk of secondary malignancy, particularly in this vulnerable patient population. Retrospective studies (38, 39) of patients with heritable and nonheritable retinoblastoma show that secondary malignancies occur both within and outside the treatment fields, with a suggested threshold of 5 Gy for infield sarcoma induction. Electron techniques, which are undesirable because of dose inhomogeneity within the target, result in the lowest volume of tissue receiving 5 Gy and above. However, there were no significant differences in doses to nonspecified tissues between the eight photon-based techniques. Although margin reduction will result in further reductions in low-dose exposure with such techniques, it is the use of protons that holds the greatest potential in this regard, the size and location of the target making this type of disease the perfect candidate for conformal proton beam irradiation (40, 41). For now, proton therapy is not as widely available as the other techniques discussed in this article, and further discussion would be beyond the scope of this work.

Other normal tissues For the contralateral optic nerve, lens, and other eye structures, as well as the optic chiasm, the received doses were well below their respective tolerance doses. The dose to the optic chiasm was minimal for most techniques, with the exception of the WP photon technique, for which the mean dose was 23 Gy. Supplementary Table E5 lists the mean doses received by all of these structures.

Discussion Use of systemic chemotherapy in conjunction with focal therapies such as cryotherapy, laser therapy, transpupillary thermocoagulation, and radioactive plaque brachytherapy has changed the role of external beam radiotherapy, which is now most commonly used as a treatment of last resort for ocular salvage in patients with extensive uncontrolled disease. Patients are typically very young (less than 3 years of age) and more likely to have a germ-line Rb mutation, placing them at risk for not only significant bony changes as they grow but also for radiation-induced secondary malignancies. The CTV for patients in this situation is typically the entire globe, minus the anterior chamber. No attempt is made to spare the lens because of the risk of underdosage of a portion of the retina and vitreous that may be associated with a higher risk of recurrence (18e26) but also because of improvements in cataract surgery (27, 28). For the PTV dose coverage, no significant differences were observed for the photon-based delivery techniques, while the two electron techniques produced poorer dose coverage as well as significantly higher maximum doses. Inhomogeneity within the target volume of this magnitude (>125%), which would result in a dose/fraction to part of the target volume that would put the patient at increased risk of late effects (e.g., retinopathy and optic neuropathy), is clearly undesirable. Inhomogeneities of smaller magnitude, typically, for example, with FSRT techniques, may be acceptable. Generally, therefore, photon techniques are preferred, and doses to critical structures have higher priority than PTV coverage in evaluation of the different techniques. Growth inhibition due to radiation therapy correlates with dose delivered, and sparing of a significant volume of the bony orbit dose beyond a threshold dose may obviate or at least reduce orbital bone growth retardation. The dose threshold for bone growth inhibition is not precisely known but may be as low as 20 Gy (29e32). Krasin et al. (33) published results comparing four techniques used in treatment of retinoblastoma. The mean relative volumes of the ipsilateral bony orbit treated above 20 Gy in that study were 60%, 78%, 91%, and 89% for IMRT, conformal, anterior-lateral photon, and en face electron techniques, respectively, for a prescribed dose of 45 Gy given in daily fractions of 1.8 Gy. In our study, doses to the bony orbit were lower than those reported by Krasin et al. for the VMAT and HT techniques. For

Conclusions Inverse planned image-guided radiotherapy delivered using HT or VMAT gives a better CI, improved orbital bone and brain sparing,

Volume 82  Number 3  2012 and lower integral dose thanother techniques. With the use of sophisticated external beam radiation therapy delivery techniques such as IMRT, VMAT, HT, and FSRT, it is reasonable to anticipate a reduction in the risk of late effects in patients with retinoblastoma. Further improvements can be expected if the PTV margin can be safely reduced through the use of improved immobilization and treatment verification techniques.

References 1. Kanski JJ. Tumours of the eye in clinical ophthalmology. 3rd ed. Oxford, UK: Butterworth Heinemann; 1994. p. 222e226. 2. Munier FL, Balmer A, van Melle G, Gailloud C. Radial asymmetry in the topography of retinoblastoma. Clues to the cell of origin. Ophthalmic Genet 1994;15:101e106. 3. Hopping W, Meyer-Schwickerath G. Light coagulation treatment in retinoblastoma. In: Boniuk M, editor. Ocular and adnexal tumors: New and controversial aspects. St. Louis: CV Mosby; 1964. p. 192e196. 4. Hopping W, Bunke-Schmidt A. Light coagulation and cryotherapy. In: Blodi FC, editor. Retinoblastoma. New York: Churchill Livingstone; 1985. p. 95e110. 5. Hamel P, Heon E, Gallie BL, et al. Focal therapy in the management of retinoblastoma: When to start and when to stop. J Am Assoc Pediatr Ophthalmol Strabismus 2000;4:334e337. 6. Shields CL, Shields JA, Kiratli H, et al. Treatment of retinoblastoma with indirect ophthalmoscope laser photocoagulation. J Pediatr Ophthalmol Strabismus 1995;32:317e322. 7. Abramson DH, Ellsworth RM, Rozakis GW. Cryotherapy for retinoblastoma. Arch Ophthalmol 1982;100:1253e1256. 8. Shields JA, Parsons H, Shields CL, et al. The role of cryotherapy in the management of retinoblastoma. Am J Ophthalmol 1989;108:260e264. 9. Amendola BE, Markoe AM, Augsburger JJ, et al. Analysis of treatment results in 36 children with retinoblastoma treated by scleral plaque irradiation. Int J Radiat Oncol Biol Phys 1989;17:63e70. 10. Shields JA, Giblin ME, Shields CL, et al. Episcleral plaque radiotherapy for retinoblastoma. Ophthalmology 1989;96:530e537. 11. Shields CL, Shields JA, De Potter P, et al. Plaque radiotherapy for retinoblastoma. Int J Radiat Oncol Biol Phys 1989;17:63e70. 12. Shields JA, Shields CL, De Potter P, et al. Plaque radiotherapy for residual or recurrent retinoblastoma in 91 cases. J Pediatr Ophthalmol Strabismus 1994;31:242e245. 13. Shields CL, Shields JA, Cater J, et al. Plaque radiotherapy for retinoblastoma: Long-term tumor control and treatment complications in 208 tumors. Ophthalmology 2001;108:2116e2121. 14. Shields CL, De Potter P, Himelstein BP, et al. Chemoreduction in the initial management of intraocular retinoblastoma. Arch Ophthalmol 1996;114:1330e1338. 15. Friedman DL, Himelstein B, Shields CL, et al. Chemoreduction and local ophthalmic therapy for intraocular retinoblastoma. J Clin Oncol 2000;18:12e17. 16. Rodriguez-Galindo C, Wilson MW, Haik BG, et al. Treatment of intraocular retinoblastoma with vincristine and carboplatin. J Clin Oncol 2003;21:2019e2025. 17. Shields CL, Meadows AT, Leahey AM, et al. Continuing challenges in the management of retinoblastoma with chemotherapy. Retina 2004; 24:849e862. 18. Merchant T, Gould C, Hilton N, et al. Ocular preservation after 36 Gy external beam radiation therapy for retinoblastoma. J Pediatr Hematol Oncol 2002;24:246e249. 19. Fontanesi J, Pratt CB, Kun LE, et al. Treatment outcome and doseresponse relationship in infants younger than 1 year treated for retinoblastoma with primary irradiation. Med Pediatr Oncol 1996;26:297e304. 20. Hungerford JL, Toma NMG, Plowman PN, et al. External beam radiotherapy for retinoblastoma: Whole eye technique. Br J Ophthalmol 1995;79:109e111.

Radiotherapy for retinoblastoma e505 21. Blach LE, McCormick B, Abramson DH. External beam radiation therapy and retinoblastoma: Long-term results in the comparison of two techniques. Int J Radiat Oncol Biol Phys 1996;35:45e51. 22. Schipper J, Tan KEWP, van Peperzeel HA. Treatment of retinoblastoma by precision megavoltage radiation therapy. Radiother Oncol 1985;3:117e132. 23. Schvartzman E, Chantada G, Fandino A, et al. Results of a stage-based protocol for the treatment of retinoblastoma. J Clin Oncol 1996;14: 1532e1536. 24. Pradhan DG, Sandridge AL, Mullaney P, et al. Radiation therapy for retinoblastoma: A retrospective review of 120 patients. Int J Radiat Oncol Biol Phys 1997;39:3e13. 25. Egbert PR, Donaldson SS, Moazed K, et al. Visual results and ocular complications following radiotherapy for retinoblastoma. Arch Ophthalmol 1978;96:1826e1830. 26. Scott TG, Feuer WJ, Van Quill K, et al. External beam radiotherapy in retinoblastoma. Arch Ophthalmol 1999;117:766e770. 27. Portellos M, Buckley EG. Cataract surgery and intraocular lens implantation in patients with retinoblastoma. Arch Ophthalmol 1998; 116:449e452. 28. Honavar SG, Shields CL, Shields JA, et al. Intraocular surgery after treatment of retinoblastoma. Arch Ophthalmol 2001;119: 1613e1621. 29. Donaldson SS, Torrey M, Link MP, et al. A multidisciplinary study investigating radiotherapy in Ewing’s sarcoma: End results of POG 8346. Int J Radiat Oncol Biol Phys 1998;42:125e135. 30. Silber JH, Littman PS, Meadows AT. Stature loss following skeletal irradiation for childhood cancer. J Clin Oncol 1990;8:304e312. 31. Eifel PJ, Donaldson SS, Thomas PRM. Response of growing bone to irradiation: A proposed late effects scoring system. Int J Radiat Oncol Biol Phys 1995;31:1301e1307. 32. Willman KY, Cox RS, Donaldson SS. Radiation induced height impairment in pediatric Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1994;28:85e92. 33. Krasin MJ, Crawford BT, Zhu Y, et al. Intensity-modulated radiation therapy for children with intraocular retinoblastoma: Potential sparing of the bone orbit. Clin Oncol 2004;16:215e222. 34. Beltran C, Herman MG, Davis BJ. Planning target margin calculations for prostate radiotherapy based on intrafraction and interfraction motion using four localization methods. Int J Radiat Oncol Biol Phys 2008;70:289e295. 35. Schaly B, Bauman GS, Song W, Battista JJ, Van DJ. Dosimetric impact of image guided 3D conformal radiation therapy of prostate cancer. Phys Med Biol 2005;50:3083e3101. 36. Ramakrishna N, Rosca F, Friesen S, Tezcanli E, Zygmanszki P, Hacker F. Clinical comparison of patient setup and intrafraction motion using frame-based radiosurgery versus a frameless image guided radiosurgery. Radiother Oncol 2010;95:109e115. 37. Wiersma RD, Wen Z, Sadinski M, Farrey K, Yenice KM. Development of a frameless stereotactic radiosurgery system based on real time 6D position monitoring and adaptive head motion compensation. Phys Med Biol 2010;55(2):389e401. 38. Moll AC, Imhof SM, Schouten-Van Meeteren AY, et al. Second primary tumors in hereditary retinoblastoma: A register based study, 1945e1997: Is there an age effect on radiation-related risk? Ophthalmology 2001;108:1109e1114. 39. Wong FL, Boice JD Jr., Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA 1997;278: 1262e1267. 40. Krengli M, Hug EB, Adams JA, et al. Proton radiation therapy for retinoblastoma: Comparison of various intraocular tumor locations and beam arrangements. Int J Radiat Oncol Biol Phys 2005;61(2): 583e593. 41. Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: How do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63(2):362e372.