This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wolden, S. L. Articles by Leibel, S. A. Search for Related Content PubMed PubMed Citation Articles by Wolden, S. L. Articles by Leibel, S. A. Social Bookmarking                            What's this?

Patterns of Failure Using a Conformal Radiation Therapy Tumor Bed Boost for Medulloblastoma

From the Departments of Radiation Oncology, Pediatrics, and Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY.

Address reprint requests to Suzanne L. Wolden, MD, Department of Radiation Oncology, 1275 York Ave, Box 22, New York, NY 10021; email: woldens{at}mskcc.org.

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Purpose: To assess the patterns of failure for patients with medulloblastoma receiving a conformal tumor bed boost rather than a boost to the entire posterior fossa.

Patients and Methods: From 1994 to 2002, 32 consecutive patients with newly diagnosed medulloblastoma treated at Memorial Sloan-Kettering Cancer Center (New York, NY) received a conformal boost to the tumor bed in conjunction with craniospinal radiation therapy. Twenty-eight patients also received chemotherapy. The median age was 9 years (range, 3 to 34 years), and the male to female ratio was 3:1. Twenty-seven patients had standard-risk disease, and five patients had high-risk disease. Craniospinal doses ranged from 23.4 to 39.6 Gy, and total tumor bed doses ranged from 54 to 59.4 Gy.

Results: With a median follow-up of 56 months, six patients have relapsed; five relapsed outside of the posterior fossa, and one failed within the posterior fossa, outside of the high-dose boost volume. Five-year actuarial disease-free and overall survival rates were 84% and 85%, respectively. Freedom from posterior fossa failure was 100% and 86% at 5 and 10 years, respectively. Freedom from distant failure was 84% at 5 years, with a trend for improvement when full-dose craniospinal radiation (36 to 39.6 Gy) was used compared with a reduced dose (23.4 Gy) of radiation (100% v 63%, respectively; P = .06). No other predictive variables were identified.

Conclusion: Conformal treatment to the tumor bed allows for significant sparing of critical structures. The posterior fossa failure rate in this series is similar to that reported when the entire posterior fossa is treated. This approach should be investigated further in a phase III trial.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
MEDULLOBLASTOMA IS the most common malignant brain tumor in children. The traditional treatment for this disease has consisted of surgery followed by craniospinal radiation therapy plus a high-dose boost to the entire posterior fossa, with or without the addition of chemotherapy. The posterior fossa is approximated from standard bony landmarks identified on plain cranial radiographic simulation films and generally treated with opposed lateral fields. This approach, in conjunction with cisplatin-based chemotherapy, has resulted in 5-year disease-free survival rates approaching 80% for standard-risk patients.¹ However, survivors often manifest numerous long-term complications related to cranial radiation therapy, including neurocognitive dysfunction, pituitary deficiency, and hearing loss.^2–4

There have been significant technologic advances in radiation oncology over the past decade. Most notable among these advances is the development of conformal techniques that allow treatment planning based on three-dimensional reconstructions of individual patient anatomy using axial computed tomography (CT) or magnetic resonance imaging (MRI). With these capabilities, the paradigm of delivering high-dose radiation therapy to the entire posterior fossa for medulloblastoma has been challenged. Since 1994, we have limited the high-dose boost for patients with medulloblastoma to residual tumor and the postoperative resection cavity plus a margin using three-dimensional conformal techniques. This report presents the outcome and patterns of failure for patients treated with this approach.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Between May, 1994, and September, 2002, 32 consecutive patients with newly diagnosed medulloblastoma received definitive treatment consisting of craniospinal radiation therapy plus a conformal boost to the tumor bed. Informed consent for radiation was obtained for all patients. Some but not all of the patients were enrolled onto an institutional review board–approved protocol. The median patient age was 9 years, with a range of 3 to 34 years. Twenty-four patients were 18 years old at diagnosis. There were 24 male and eight female patients.

All patients underwent routine staging, including MRI of the brain and spine plus a lumbar puncture to assess CSF cytology. Twenty-seven patients had standard-risk (M0) disease, defined as less than 1.5 mL of residual tumor and no evidence of spread beyond the primary tumor. Five patients had high-risk disease based on positive CSF cytology (M1, n = 1), intracranial (solitary supratentorial) metastases (M2, n = 2), spinal metastasis (M3, n = 1), and greater than 1.5 mL of residual tumor (n = 1).

After resection of the primary tumor, four patients were treated with radiation alone, and 28 patients received radiation plus chemotherapy. The dose of craniospinal radiation therapy depended on disease stage and treatment protocol. Patients with standard-risk disease were treated with full-dose craniospinal radiation (36 Gy) earlier in the study, before reduced-dose radiation with chemotherapy became an accepted standard. Craniospinal radiation doses were as follows: 23.4 Gy (n = 21), 36 Gy (n = 10), and 39.6 Gy (n = 1). The cumulative dose to the tumor bed, including the contribution of the craniospinal treatment, ranged from 54 to 59.4 Gy, with a median of 55.8 Gy. Six patients received a boost dose of more than 55.8 Gy (all received 59.4 Gy) as part of the protocol offered at that time. There was no reduction in boost volume except to limit dose to the cervical spinal cord.

Twenty-eight patients received chemotherapy as part of initial management. All patients received weekly chemotherapy during radiation therapy (vincristine, n = 27; carboplatin, n = 1). Twenty-seven patients received adjuvant chemotherapy starting approximately 6 weeks after completion of radiation therapy. All patients received platinum agents; 24 patients received vincristine, cisplatin, and either lomustine or cyclophosphamide.¹

Treatment Planning
All patients were immobilized in the prone position for simulation and daily treatments using customized molds to support and restrict their head and body motion. Patients who were too young to cooperate with strict immobilization were simulated and treated with anesthesia. The craniospinal portion of therapy was planned and delivered with traditional techniques using posterior field(s) for the spine, which were carefully matched to lateral fields to treat the entire brain. During the same simulation procedure and in the same position, an isocenter was chosen in the approximate center of the anticipated planning target volume (PTV) for the boost treatment. The patient was then tattooed so that this isocenter could be recreated for treatment with triangulation techniques.

Treatment planning for all patients included a CT scan of the brain in the treatment position using each patient’s customized immobilization device. In the mid-1990s, a CT scan with contrast was performed separate from the simulation procedure but with radiopaque markers on the skin to indicate the triangulation points for the isocenter. The CT images were then transferred to a treatment planning system. Since 1998, the entire simulation procedure has been performed on a dedicated CT simulator. Since 1999, diagnostic MRI scans have been fused with the CT simulation images to enhance the accuracy of target localization for all patients.

Once the images have been obtained, the patient’s relevant normal anatomy, including the orbits, optic nerves, optic chiasm, pituitary gland, brainstem, spinal cord, parotid glands, and cochlea, are contoured in three dimensions. Using pre- and postoperative MRI images for reference in addition to the treatment planning CT scan, a gross tumor volume (GTV) was outlined. The GTV consisted of any residual tumor plus the entire resection cavity. A margin of 1 to 2 cm was then added to the GTV in all dimensions to create a clinical target volume (CTV). The CTV is intended to include the visually identified target plus a margin to account for subclinical or microscopic extension of disease. For this reason, the CTV did not extend beyond the bony confines of the posterior fossa. The range in margin size was based on the judgment of the treating physician. An additional margin of 0.5 cm, to account for variation in daily set-up reproducibility, was then added to the CTV to create the PTV. The PTV served as the final target for conformal planning. The three patients who had metastatic deposits outside of the posterior fossa received a separate conformal boost to those sites.

Beam orientations were individually selected for each patient to include the PTV and exclude the critical normal structures to the extent possible. Noncoplanar beam angles were found to be ideal and were used in many cases. The number of beams used for each patient ranged from two to six beams, with a median of five beams. The algorithm for treatment planning evolved over the study period from the use of forward planning with differential beam weighting, custom shaped blocks, and wedges (n = 20) to the current method of intensity-modulated radiation therapy (IMRT) with dynamic multileaf collimators (n = 12).⁵ In either case, numerous treatment plans were generated for each patient, along with dose-volume histograms, to quantify doses in all structures of interest. The dose-volume histogram analyses allowed an objective comparison of plans before the optimal plan was selected. Plans were selected based on the completeness of coverage and homogeneity within the PTV as well as avoidance of normal structures, such as the hypothalamic-pituitary axis, temporal lobes, and cochlea. The prescription boost dose was prescribed to cover the entire PTV. This is in contrast to the technique of prescribing to midplane, when traditional lateral fields are used to treat the entire posterior fossa.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
With a median follow-up of 56 months (range, 2 to 101 months), six patients have suffered a recurrence or progression of medulloblastoma. Four patients experienced relapses that were diffuse leptomeningeal failures with extensive disease throughout the craniospinal axis without a significant posterior fossa component. A fifth patient experienced recurrence with two foci in the bilateral frontal horns and no disease in the posterior fossa. The sixth patient experienced an isolated failure within the posterior fossa, outside of the tumor bed boost volume. Three of the six patients who relapsed are currently receiving salvage therapy, including the patient with a posterior fossa failure; the other three patients have died from disease progression. The actuarial disease-free survival rate was 84% at 5 years and 65% at 10 years, whereas the overall survival rate was 85% at both 5 and 10 years.

Four of the five patients who relapsed outside of the posterior fossa were patients with standard-risk disease at 12, 17, 36, and 61 months after initiation of therapy. They had completed all prescribed therapy, including concurrent vincristine with 23.4 Gy to the craniospinal axis plus a boost to the primary site followed by adjuvant chemotherapy. The fifth patient who experienced distant failure was a 3-year-old patient with high-risk disease based on a small solitary nodule at the cervicomedullary junction. He had completed 23.4 Gy to the craniospinal axis with a boost to 55.8 Gy and concurrent vincristine chemotherapy. Against medical advice, the patient’s parents refused full-dose craniospinal radiation and all adjuvant chemotherapy because of concerns about toxicity at his young age. He was found to have leptomeningeal spread 5 months after the initiation of radiation.

The actuarial freedom from distant failure was 84% at 5 years and 78% at 10 years (Fig 1). There was a trend for improved freedom from distant failure among patients receiving full-dose craniospinal radiation (36 to 39.6 Gy) compared with patients receiving reduced-dose (23.4 Gy) radiation (100% v 63% at 10 years, respectively; P = .06). All but one of the patients who received reduced-dose radiation had standard-risk disease and received concurrent as well as adjuvant chemotherapy. When this patient is removed from the analysis, the difference is slightly less significant (100% for full-dose radiation v 66% for reduced-dose radiation, P = .08). No other patient or treatment variables were found to relate to the risk of distant failure.

View larger version (10K): [in this window] [in a new window]   Fig 1. Actuarial freedom from failure outside of the posterior fossa.

A recent representative case was selected to demonstrate the dosimetric differences between lateral whole posterior fossa fields and a conformal IMRT boost to the tumor bed. Figure 2A to Figure 2D illustrates the target volumes used for conformal therapy and the distribution of doses with each technique. Quantitative comparisons of doses to the PTV and critical normal tissues for this example case are outlined in Table 1.

View larger version (136K): [in this window] [in a new window]   Fig 2. (A and B) Dose distributions for a sample case treating the posterior fossa with lateral fields and (C and D) treatment to the tumor bed using conformal treatment with intensity-modulated radiation therapy. The gross tumor volume (GTV) represents the resection cavity and any residual tumor and is shown as the innermost green line. The clinical target volume (CTV) is the light blue line, where a 1- to 2-cm margin has been added to the GTV. The planning target volume is the dark blue line, derived by adding a 0.5-cm margin to the CTV. The isodose lines are labeled and show the areas covered by each percentage of the prescribed dose. The cochlea (inner ears) are outlined in pink. The high cervical spinal cord would routinely be blocked after a dose of 45 to 50 Gy on lateral fields.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

Preliminary results of a reduced-volume boost at our institution were published by Merchant et al.⁹ That report included 10 patients with newly diagnosed medulloblastoma treated between 1994 and 1996. At the time of that analysis, there had been no local failures. More recently, Douglas et al¹⁰ presented an abstract reporting no local failures in 31 medulloblastoma patients treated with a conformal reduced-volume boost, but the median follow-up in that series was only 23 months.

A number of modern institutional and cooperative group studies using full posterior fossa radiation boosts have reported detailed patterns of failure analyses. In these series, the rate of isolated failure in the posterior fossa ranges from 3% to 17%.^1,7,8,11,12 The studies varied with regard to overall treatment strategy and length of follow-up. The length of follow-up is the most likely explanation for the variation in posterior fossa failure rates. These rates are comparable with the 10-year actuarial posterior fossa failure rate of 14% in our series. Only the study by Fukunaga-Johnson et al¹³ analyzed the specific location of failure within the posterior fossa. They found that only 10% of failures were limited to the posterior fossa (7% within the tumor bed and 3% outside of the tumor bed).

There have been no cases of isolated failure within the tumor bed in our series. Patients in this report have been observed for a median of 56 months, yet the single posterior fossa failure occurred 84 months after initiation of therapy. The long latency to this recurrence raises the concern that additional late failures could occur. Therefore, extended follow-up may be necessary to establish the safety of a reduced-volume conformal boost.

On the other hand, the benefits of a reduced-volume conformal tumor bed boost for medulloblastoma include potentially significant sparing of critical normal tissues, including the inner ears, pituitary gland, temporal lobes, and parotid glands. This is illustrated in the example used for Table 1 and Figure 2. Although the outcome data regarding the long-term clinical impact of these dose reductions for normal tissues are just beginning to mature, it is reasonable to assume that a lower dose of radiation will correlate with a lower risk of complications.

Even when the entire posterior fossa is included in the boost volume, there is a benefit to using conformal therapy to reduce exposure to normal tissues. Two detailed dosimetric analyses have shown that three-dimensional conformal techniques, compared with lateral fields, significantly lowered the radiation dose to the inner ears.^13,14 The study by Paulino et al¹⁴ also showed a benefit in reducing dose to the pituitary gland and supratentorial brain at the expense of increased exposure to other normal tissues such as the thyroid gland. In an accompanying editorial by Tarbell et al,¹⁵ a case is made for further reduction in normal tissue exposure when proton beams are used to treat the entire posterior fossa. Huang et al⁴ have been the first to publish lower rates of ototoxicity when the dose to the inner ear is reduced by using IMRT to boost the entire posterior fossa.

A previous report from our institution included a dosimetric analysis demonstrating that normal tissue sparing is a function not only of beam orientation but also of target volume.⁹ As the volume of the PTV decreased, there was a significant decrease in doses of radiation to surrounding critical structures. This is a well-established principal in radiation oncology. If it is deemed acceptable to treat less than the whole posterior fossa, the question of optimal target volume must then be established. The criteria for designating a GTV, CTV, and PTV in this series were fairly conservative, allowing a 1- to 2-cm margin to account for subclinical disease extension beyond residual tumor and the resection cavity. In several cases where the resection cavity was large and located in the midline, the PTV nearly filled the posterior fossa. The solitary case of posterior fossa failure in our series would not have been included in the boost volume with an expanded margin because it was in the opposite cerebellar hemisphere.

With our increasing ability to fuse diagnostic studies, such as MRI and positron emission tomography images, with planning CT scans, we may be able to further refine target volume delineation. Additional advances in immobilization techniques and verification may also allow incremental reductions in the margin used to account for errors in set-up reproducibility. Our current institutional medulloblastoma protocol mandates MRI fusion for treatment planning. The margin for subclinical extension (GTV to CTV expansion) is limited to 1 cm beyond residual disease and the resection cavity.

The potential benefits in long-term quality of life for survivors of medulloblastoma by using a conformal reduced-volume boost are compelling. A randomized prospective trial comparing this strategy to the traditional posterior fossa boost is warranted to ascertain whether there is an increased risk of posterior fossa failure. Such a study is currently under development by the Children’s Oncology Group.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Packer RJ, Goldwein J, Nicholson HS, et al: Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: A Children’s Cancer Group study. J Clin Oncol 17:2127–2136, 1999[Abstract/Free Full Text]

2. Palmer SL, Goloubeva O, Reddick WE, et al: Patterns of intellectual development among survivors of pediatric medulloblastoma: A longitudinal analysis. J Clin Oncol 19:2302–2308, 2001[Abstract/Free Full Text]

3. Adan L, Sainte-Rose C, Souberbielle JC, et al: Adult height after growth hormone (GH) treatment for GH deficiency due to cranial irradiation. Med Pediatr Oncol 34:14–19, 2000[CrossRef][Medline]

4. Huang E, The BS, Strother DR, et al: Intensity-modulated radiation therapy for pediatric medulloblastoma: Early report on the reduction of ototoxicity. Int J Radiat Oncol Biol Phys 52:599–605, 2002[CrossRef][Medline]

5. Leibel SA, Fuks Z, Zelefsky MJ, et al: Intensity modulated radiation therapy. Cancer J 8:164–176, 2002[Medline]

6. Halperin EC: Impact of radiation technique upon the outcome of treatment for medulloblastoma. Int J Radiat Oncol Biol Phys 36:233–239, 1996[Medline]

7. Fukunaga-Johnson N, Lee JH, Sandler HM, et al: Patterns of failure following treatment for medulloblastoma: Is it necessary to treat the entire posterior fossa? Int J Radiat Oncol Biol Phys 42:143–146, 1998[CrossRef][Medline]

8. Miralbell R, Bleher A, Huguinin P, et al: Pediatric medulloblastoma: Radiation treatment technique and patterns of failure. Int J Radiat Oncol Biol Phys 37:523–529, 1997[CrossRef][Medline]

9. Merchant TE, Happersett L, Finlay JL, et al: Preliminary results of conformal radiation therapy for medulloblastoma. Neuro-oncol 1:177–187, 1999[Abstract]

10. Douglas J, Barker J, Geyer R, et al: Concurrent chemotherapy and low-dose craniospinal irradiation followed by conformal posterior fossa tumor bed boost for average risk medulloblastoma: Efficacy and patterns of failure. Int J Radiat Oncol Biol Phys 54:149, 2002 (abstr 255)

11. Thomas PR, Deutsch M, Kepner JL, et al: Low-stage medulloblastoma: Final analysis of trial comparing standard-dose with reduced-dose neuraxis irradiation. J Clin Oncol 18:3004–3011, 2000[Abstract/Free Full Text]

12. Merchant TE, Wang MH, Haida T, et al: Medulloblastoma: Long-term results for patients treated with definitive radiation therapy during the computed tomography era. Int J Radiat Oncol Biol Phys 36:29–35, 1996[CrossRef][Medline]

13. Fukunaga-Johnson N, Sandler HM, Marsh R, et al: The use of 3D conformal radiotherapy (3D CRT) to spare the cochlea in patients with medulloblastoma. Int J Radiat Oncol Biol Phys 41:77–82, 1998[CrossRef][Medline]

14. Paulino AC, Narayana A, Mohideen MN, et al: Posterior fossa boost in medulloblastoma: An analysis of dose to surrounding structures using 3-dimensional (conformal) radiotherapy. Int J Radiat Oncol Biol Phys 46:281–286, 2000[CrossRef][Medline]

15. Tarbell NJ, Smith AR, Adams J, et al: The challenge of conformal radiotherapy in the curative treatment of medulloblastoma. Int J Radiat Oncol Biol Phys 46:265–266, 2000[CrossRef][Medline]

Submitted November 27, 2002; accepted June 2, 2003.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				C. Carrie, J. Grill, D. Figarella-Branger, V. Bernier, L. Padovani, J. L. Habrand, M. Benhassel, M. Mege, M. Mahe, P. Quetin, et al. Online Quality Control, Hyperfractionated Radiotherapy Alone and Reduced Boost Volume for Standard Risk Medulloblastoma: Long-Term Results of MSFOP 98 J. Clin. Oncol., April 10, 2009; 27(11): 1879 - 1883. [Abstract] [Full Text] [PDF]

				P. P. Connell and S. Hellman Advances in Radiotherapy and Implications for the Next Century: A Historical Perspective Cancer Res., January 15, 2009; 69(2): 383 - 392. [Full Text] [PDF]

				B Pizer and S Clifford Medulloblastoma: new insights into biology and treatment Arch. Dis. Child. Ed. Pract., October 1, 2008; 93(5): 137 - 144. [Full Text] [PDF]

				S. L. Palmer, W. E. Reddick, and A. Gajjar Understanding the Cognitive Impact on Children Who are Treated for Medulloblastoma J. Pediatr. Psychol., October 1, 2007; 32(9): 1040 - 1049. [Abstract] [Full Text] [PDF]

				G Bauman, S Yartsev, T Coad, B Fisher, and T Kron Helical tomotherapy for craniospinal radiation Br. J. Radiol., June 1, 2005; 78(930): 548 - 552. [Abstract] [Full Text] [PDF]

				D. G. Kirsch and N. J. Tarbell Conformal Radiation Therapy for Childhood CNS Tumors Oncologist, July 1, 2004; 9(4): 442 - 450. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wolden, S. L. Articles by Leibel, S. A. Search for Related Content PubMed PubMed Citation Articles by Wolden, S. L. Articles by Leibel, S. A. Social Bookmarking                            What's this?