|
Miles to Go:
Perspectives on Medulloblastoma Management
by
Massimiliano De Bortoli, M.D.,
Murali
Chintagumpala, M.D. and
John Y. H. Kim, M.D., Ph.D.
Medulloblastoma is the most common primary malignant brain tumor of childhood. Decades of clinical studies have resulted in improved outcomes. A generation ago saw
three-year survival rates of less than 50
percent, which have risen to greater than 80 percent today. Although medulloblastoma is both radiosensitive and chemosensitive,
it continues to pose significant challenges for the future
evolution of management. This article reviews the prognostic
factors, current management approaches, and emerging strategies
to prevent or minimize long-term complications.
Brain tumors are the most common solid tumors of childhood, accounting for approximately 3.3 diagnoses per 100,000 children and contribute disproportionately to the morbidity and mortality of pediatric cancer. Medulloblastoma (MB) comprises up to 20
percent of pediatric brain tumors, the most common malignant type. MB has yielded modestly to improvements in neurosurgical techniques, radiation and chemotherapy. Aggressive multi-modality therapy fails to cure many children, and survivors of MB suffer profound complications such as growth impairment and neurocognitive delays as treatment sequelae. Improvement of clinical outcomes in this devastating cancer remains a priority research goal [http://prg.nci.nih.gov/brain/finalreport.html].
The mainstay of clinical management remains maximally feasible surgical resection. Standard postoperative therapy for medulloblastoma includes radiation therapy (RT) and chemotherapy. Because of its propensity for subarachnoid and leptomeningeal dissemination, craniospinal irradiation (CSI) is widely accepted as requisite for maximizing cures. However, the avoidance of CSI in infants due to their developmental sensitivity and the optimal dose of CSI remains an unresolved issue. The optimal schedule and dosing of combination chemotherapy agents continue to fuel debate.
Unfortunately, curative management may be accompanied by devastating long-term treatment effects, including neuropsychologic and neuroendocrine sequelae
and cognitive dysfunction. Therefore, current treatment
challenges include the identification of prognostic factors that
may identify patients who require less intensive therapy as well
as definition of the optimal dosage and timing of RT and chemotherapy.
Imaging. Magnetic resonance imaging (MRI) is the modality of choice for diagnostic and serial non-invasive assessment. In the initial assessment for metastatic disease, contrast-enhanced diagnostic scans typically include spine MRI to complement cytological examination of cerebrospinal fluid (obtained by postoperative lumbar puncture). Definitive diagnosis will continue to depend upon thorough examination of tumor specimens by experienced neuropathologists.
Additional MR techniques such as diffusion/perfusion-weighting, diffusion tensor imaging and spectroscopy can assist in the preoperative definition of the differential diagnosis, especially in following treatment response.
Prognostic factors. Current treatment for medulloblastoma depends upon the stage of the disease, patient age at diagnosis, and the extent of postoperative residual tumor. Patients with higher disease stage, younger than 3 years
old, and residual tumor greater than 1.5 cm2 have a poorer prognosis. Historically, classification of medulloblastoma patients into standard risk (also called favorable risk, low risk or average
risk) and high risk categories has been based on the Chang staging system, which includes factors such as primary tumor extent (T) and sites of metastatic disease (M). According to this system, metastatic disease is high risk, and most nonmetastatic disease is average risk.
Neuroaxis dissemination (M stage) is based on neuroimaging and cerebrospinal fluid (CSF) cytology. Ideally, diagnostic spine MRI should be obtained before surgery. When preoperative spinal imaging is not possible, subsequent imaging studies for staging should not be performed for a minimum of 10 to 14 days after surgery in order to avoid over-interpretation of postoperative changes. In addition, cytologic studies of CSF should be performed following definitive spinal imaging and at least 10 to 14 days after surgery. CSF should be obtained from the lumbar space, because of higher sensitivity compared to ventricular CSF cytology.
Age younger than 3 years has consistently been demonstrated to be a poor prognostic factor. Recent evidence indicates that histologic subtype also contributes to outcome in that the large cell and anaplastic variants confers additional risk. The inverse association between outcome and the nodular/desmoplastic variant has not been consistently reproducible.
Few would question the necessity of an experienced multidisciplinary team approach to the clinical management of medulloblastoma patients, the complications of treatment and its long-term sequelae.
The team should include specialists in pediatric neurosurgery,
radiation oncology, pediatric neuro-oncology, oncology nursing,
pediatric neurology, pediatric endocrinology, pediatric
ophthalmology, neuropsychology, physical therapy/occupational
therapy, and last (but certainly not least) psychosocial support
and advocacy.
Surgical resection constitutes the basis for obtaining definitive diagnosis and remains the mainstay of treatment. The benefits of increasing experience with advances in neurosurgical techniques, pediatric anesthesia and critical care cannot be overstated. Undoubtedly, the extent of residual tumor depends upon its precise location, perhaps reflecting underlying biological differences, and varies with local surgical practice.
Radiation therapy (RT) occupies a critical role in the management of medulloblastoma, not only in the local treatment of residual or bulk disease, but also in the regional control of potentially disseminated tumor. Maximally tolerated RT to the primary tumor bed and posterior fossa
is typically delivered as conventional doses to 54 Gray (Gy) or
5400 centiGray (cGy).
Craniospinal irradiation therapy
Craniospinal irradiation (CSI) therapy remains the primary adjuvant therapy in the postsurgical management of medulloblastoma for children younger than 3 years.
Typically, CSI involves delivery of 36 Gy to the entire neuroaxis and a local “boost” to a cumulative dose of 54 Gy to the posterior fossa, which remains the preferred radiation dose for high-risk medulloblastoma.
Because radiation may adversely impact quality of life due to long-term cognitive, neuropsychologic, or neuroendocrine deficits, an obvious question was whether cure rates could be maintained if conventional radiation doses were reduced. Preliminary data suggest that a combination of reduced-dose radiation (23.4 Gy) following surgery and adjuvant chemotherapy during and after RT may result in an improved outcome and is widely considered the standard of care for patients with standard-risk medulloblastoma. Further dose reduction of RT (18 Gy) plus chemotherapy is currently being studied prospectively by the Children’s Oncology Group in the Phase III clinical trial for standard-risk medulloblastoma in attempt to improve overall outcome while simultaneously minimizing long-term treatment-related toxicities.
Conformal techniques
New techniques for delivering RT such as intensity modulated RT (IMRT) make it possible to “conform” the radiation dose to the contour of the tumor volume despite its irregular shape and sparing surrounding normal tissues. A putative advantage of conformal radiotherapy in the management of medulloblastoma is that the delivery of a decreased dose to the cochlea may reduce the potential for ototoxicity, which is of particular importance in children who are receiving adjuvant chemotherapy with cisplatin. There is no strict consensus on conformal RT to the posterior fossa. The use of conformal delivery techniques may confound the assessment of long-term treatment sequelae in the cooperative group setting. This observation is illustrated by recent reports of the analyses of doses to surrounding normal tissue using different conformal radiotherapy techniques for the posterior fossa boost. It is critical to develop a consensus for the optimal conformal technique for use in the management of medulloblastoma.
Hyperfractionation (increasing fraction number with reduced dose per fraction) can increase the tolerability of increased total radiation dose, but several trials have concluded that the lack of an apparent survival benefit does not support the use of this treatment modality in childhood medulloblastoma. Similar challenges have arisen with the advent of proton beam therapy, which is available in only a few centers. The theoretical advantage in CSI for the use of protons over photons lies in the limited dose to non-target tissues, which is under active study.
Neo-Adjuvant (pre-irradiation) chemotherapy
–
Recent studies evaluating the role of chemotherapy in high-risk patients have assessed the use of neoadjuvant (pre-irradiation) compared to post-radiation chemotherapy in high-risk patients are limited. A Pediatric Oncology Group study (POG 9031) failed to show a difference in outcome between high-risk patients who received pre- vs. post-RT chemotherapy with cisplatin, vincristine, and etoposide. However, patients who received cisplatin after radiation therapy had increased ototoxicity.
The German Society of Pediatric Hematology and Oncology conducted a randomized trial (HIT ‘91) confounded by the fact that the treatment arms are not equivalent. However, it was evident that neoadjuvant therapy in this study was associated with increased myelotoxicity, which resulted in delays in initiating RT and also interruptions during RT during subsequent radiation therapy. As a result, there were more treatment interruptions and an overall increased treatment time.
Although applied in select situations including minimally myelosuppressive chemotherapy concurrently with RT, standard approaches have shifted away from pre-irradiation chemotherapy to avoid delaying definitive regional treatment with radiation.
Adjuvant (post-irradiation) chemotherapy
–
Given the encouraging results of early clinical trials,
and to limit the impact of craniospinal irradiation on
neurodevelopmental outcome, contemporary trials for
standard-risk patients have employed reduced-dose CSI plus
adjuvant chemotherapy in an attempt to maintain cure rates with
decreased toxicity.
The prospective randomized trial by Children’s Oncology Group
for standard-risk medulloblastoma (COG-A9961) compared a
cyclophosphamide-based regimen (cytoxan, cisplatin, and
vincristine) to a CCNU-based regimen (CCNU, cisplatin, and
vincristine) after reduced-dose CSI (23.4 Gy), resulting in
similar five-year overall survival rates of 81percent (±
2.1percent) and 86 percent (± 9 percent), respectively.
Dose-Intensified chemotherapy with stem cell support
–
As a
result, recent trials for both standard and high-risk patients
are evaluating the feasibility and efficacy of high-dose
chemotherapy with stem cell rescue following risk-adapted
neuroaxis radiation.
The long-term outcomes of this compressed protocol are being
analyzed and the follow-up study is currently accruing patients
at our institutions and others.
The prognosis for children less than 3 years of age with medulloblastoma is
considerably worse than that of older children. Factors
responsible for the inferior outcome in these patients include:
increased incidence of relapse and neuroaxis dissemination. Although CSI constitutes effective therapy, the unacceptably high incidence of adverse neuropsychologic sequelae in young children and infants has led to current treatment strategies that rely on primary postoperative combination chemotherapy regimens intended to delay or obviate irradiation to diminishing the age-associated risks of developmental delay and intellectual deficits in these young patients.
Although clinical studies have demonstrated that postsurgical chemotherapy in infants is feasible, the primary goal of using such therapy (to substantially delay the need for radiation) has been achieved in only a minority of patients, because most patients had either local or disseminated disease recurrence within 6 to 9 months after the initiation of therapy. Infants with disease progression during initial chemotherapy can be cured with salvage radiation. Not surprisingly, infants thus treated experience substantial declines in cognitive function, and the majority experience neuroendocrine problems.
The German Society of Pediatric Hematology and Oncology applied such an approach for infants and young children with medulloblastoma. They recently reported encouraging results with post-operative chemotherapy alone in those patients less than three years of age with non-metastatic disease. Of note, treatment included three cycles of intravenous chemotherapy (cyclophosphamide, vincristine, methotrexate, carboplatin, and etoposide) and intraventricular methotrexate (Rutkowski et al.
N Engl J Med 2005;352:978-86.). Concerns have been raised, however, regarding the known neurotoxicity of methotrexate especially in this young age group.
In attempt to improve the overall survival and therapeutic outcome for infants with medulloblastoma, the Pediatric Brain Tumor Consortium initiated a study that uses a new therapeutic strategy incorporating intrathecal therapy plus the early introduction of limited-field radiation therapy in an attempt to control leptomeningeal disease without the toxicities associated with neuroaxis radiation. In this trial, infants receive a brief period (approximately 20 weeks) of postoperative systemic and intrathecal chemotherapy, followed by conformal radiotherapy to the site of the original tumor, followed by an additional 20 weeks of systemic chemotherapy. The primary study endpoint is the assessment of the feasibility of administering postoperative intrathecal chemotherapy to infants with newly diagnosed medulloblastoma. In addition, an estimate of the subsequent progression-free survival and pattern of failure associated with the use of this treatment strategy in infants with initially local disease (stage M0) at diagnosis will be made.
Ongoing COG clinical trials include a pilot Phase I/II study of intensive chemotherapy with peripheral stem cell support for infants with malignant brain tumors including medulloblastoma (ACNS-99703). The induction post-operative chemotherapy, based on CCG-9921A, consists of cyclophosphamide, etoposide, vincristine and cisplatin, with stem cell harvests; followed by three consolidation cycles of carboplatin and dose escalated thiotepa with stem cell support. Another approach in young children (>8months but <3 years) highlighted in a Phase III COG study (9932) is the application of second look surgery and conformal RT (limited to posterior fossa and primary tumor site) with systemic chemotherapy.
Discussion of biologically active prognostic markers is beyond the scope of this article. Briefly, increasingly sophisticated and detailed studies of tumor biology and genomic analyses of medulloblastoma specimens in conjunction with clinical studies have yielded cytogenetic profiles and gene expression patterns associated with biological features and outcomes. A growing catalog of significant cytogenetic lesions, cell surface markers and intracellular factors has been identified, including ErbB2/Her1, TrkC, and Shh signaling pathways. Suffice it to mention that they are under active investigation for their mechanisms of action. They are being analyzed prospectively and several are targets of biologically based agents in clinical trials.
A wide variety of clinical trials are available to children with progressive or recurrent medulloblastoma. These range from Phase I trials of novel agents or combinations: oxaliplatin, VNP40101M (Cloretazine), temozolomide with irinotecan, and vincristine. Others test biologically based agents that inhibit cell signaling by the epidermal growth factor receptor (e.g. GW572016 (Lapatinib), OSI-774 (Tarceva) with temozolomide), interfere with tumor cell invasion (EMD121974 (Cilengitide)), or disrupt tumor cell gene expression (valproic acid). Advances in tumor biology and developmental neurobiology have identified potential targets for the application of novel technologies and offer the hope of therapeutic approaches that are more effective and less toxic than conventional treatments.
Improvements have been made in the management of childhood medulloblastoma during the past several decades. However, despite this progress, continued investigation is required to develop effective treatment strategies for high-risk patients and to minimize the potentially devastating long-term neurocognitive and neuroendocrine sequelae that are associated with current treatment regimens. An area of ongoing investigation and focus is the optimal dosage and delivery method for radiation therapy. As studies of adjuvant therapy in infants have demonstrated, disease control without radiation therapy is not feasible for the majority of patients. Therefore, new management strategies that combine chemotherapy with reduced-dose or limited-field irradiation are being evaluated. In addition, the optimal chemotherapy combinations and dosages need to be further defined. Studies to evaluate the feasibility of high-dose chemotherapy following neuroaxis radiation are in progress. Finally, ongoing and future prospective studies evaluating the biological characteristics of the tumor will be instrumental in further defining treatment risk categories. The ultimate goal is to develop risk-based management strategies that incorporate appropriate combinations of chemotherapy with irradiation in order to maximize the cure rate for all patients while reducing long-term sequelae of therapy.
Massimiliano De Bortoli, M.D., is
a postdoctoral associate at
Texas Children's
Cancer Center. A pediatric hematologist/oncologist, Dr.
Bortoli trained at the University of Padua, Italy. His areas of
research interest include the genomic analysis, intracellular
signaling pathways and growth regulation of pediatric solid
tumors.
John Y. H. Kim, M.D., Ph.D., is
a pediatric oncologist/hematologist and member of Texas Children’s Cancer Center's
Brain Tumor Program and an assistant professor of pediatrics at
Baylor College of Medicine. Dr.
Kim's research focuses on medulloblastoma from the perspectives
of brain tumor biology and developmental neurobiology. His lab
investigates mechanisms that mediate normal neurogenesis and
also influence tumorigenesis. His research applies a wide range
of methods that span genomic, molecular genetic, biochemical and
cellular biological approaches.
Drs. Stacey Berg and Susan Blaney of the Texas Children’s Cancer Center and Baylor College of Medicine, upon whose published review article this update was based.
Publications of particular interest, are reviewed in:
Chintagumpala M, Berg S, Blaney SM. (2001). Treatment controversies in medulloblastoma. Curr Opin Oncol. 13: 154-9.
Translation:
http://world.altavista.com/ (BabelFish Translation: text can be entered and translated)
http://www.nlm.nih.gov/medlineplus/ency/article/000768.htm (Medline Plus Article, U.S. National Library of Medicine and National Institutes of Health)
http://www.cancer.gov/cancertopics/pdq/treatment/childbrain/patient/ (PDQ® Article, U.S. National Cancer Institute, National Institutes of Health)
http://www.abta.org/ (The American Brain Tumor Association)
http://www.cancer.org/ (The American Cancer Society)
http://www.childhoodbraintumor.org (The Childhood Brain Tumor Foundation)
http://www.cbtf.org/ (The Children’s Brain Tumor Foundation)
http://www.pbtfus.org/ (The Pediatric Brain Tumor Foundation of the U.S.)
http://www.tbts.org/ (The Brain Tumor Society)
http://www.childrensoncologygroup.org/ (Children’s Oncology Group)
http://www.cancer.gov/clinical_trials/ (Clinical Trials search engine, U.S. National Cancer Institute, National Institutes of Health)
http://www.soc-neuro-onc.org/
(Society for Neuro-Oncology)
http://www.siop.nl/ (Société Internationale d’Oncologie Pédiatrique)
http://www.survivorshipguidelines.org/ (Clinical Guidelines for Long-term Survivor Care)
Back to top
|
