Brain tumours are the commonest solid cancer in children and the second commonest cause of paediatric cancer overall, although they remain the leading cause of cancer-related deaths. Leukaemias account for 30% of paediatric cancers and brain tumours about 22%. About 3750 new cases of childhood brain tumours are diagnosed in the US annually. If the incidence in South Africa is similar, we would expect at least 500 new cases annually if adjusted for population size, and likely more given the greater proportion of children in our population.
No, there are many different types and they are highly variable; most commonly astrocytomas (ranging from very benign to highly malignant), medulloblastomas/ PNETs, ependymomas, craniopharyngiomas, and germ cell tumours. Some tumours are benign, and curable by complete, or sometimes even by near-complete surgical resection (eg. juvenile pilocytic astrocytomas, subependymal giant cell astrocytomas); others are highly malignant and may not be curable even with a gross total macroscopic resection (eg. atypical teratoid rhabdoid tumours, glioblastoma). Moreover, tumours occur in different regions in the brain and so pose varying risks to the neurological condition of the child and suitability for treatment. The location of the tumour affects whether the tumour can be removed with an operation. Even when tumours are treatable, a location in critical regions of the brain or brainstem may render the risk of surgery unacceptable for a reasonable quality of life.
Many tumours in childhood are treatable. Survival rates depend on the type of tumour and its location. Treatment usually begins with surgery, and the extent of removal of the tumour remains an important determinant of prognosis for most tumours. As a general rule, with some exceptions for benign tumours, both chemotherapy and radiotherapy currently play important adjuvant roles in continuing treatment. Chemotherapy has substantially changed outcome for leukaemias, but less so for brain tumours, in part because of the blood brain barrier restricts the delivery of certain therapies. Still, some tumours respond very well to some of the newer chemotherapeutic regimes, especially tumours such as optic pathway/ hypothalamic gliomas, for which surgery is often limited. Radiotherapy is essential in many forms of childhood brain tumours to treat residual or recurrent disease, but is limited by the age of the patient and potential injury to the developing brain. Specialized forms of radiotherapy are employed depending on the tumour type.
Awareness of paediatric brain tumours in the public sphere in South Africa is low. Little information exists to guide parents of children with newly diagnosed brain tumours regarding optimal methods of treatment and resources available. Because there is little co-ordination of health services for children with brain tumours, with wide variation existing between institutions in terms of expertise and resources, children are often treated inadequately. Ideally, children should be treated at a unit that has specialized experience with brain tumours in children across disciplines of surgery, chemotherapy, radiotherapy, histopathology and rehabilitation. Adequate imaging of brain tumours in children requires a dedicated environment as magnetic resonance scans in children are suboptimal, if not impossible, without adequate and safe sedation or general anaesthesia. The results of surgery to remove the tumour depend on surgeon’s training, expertise and infrastructure supporting the operation. Very few surgeons in the country operate on childhood brain tumours frequently enough to achieve and maintain the required expertise. Moreover, the experience of the anaesthetist and the availability of dedicated paediatric intensive care are essential to secure optimal outcomes.
Adjuvant treatment depends on an accurate histopathological diagnosis. Because the definitive histological diagnosis of paediatric brain tumours can be notoriously difficult, and because this has a substantial impact on decisions on how best to manage the child, specific experience in paediatric neuro-pathology is essential. Radiotherapy is commonly used to treat residual tumour post-resection, prevent recurrence, and sometimes palliate. Decision-making, planning, mark-up, and delivery for paediatric brain tumours require experience and infrastructure, without which, results are suboptimal. Optimal sedation or anesthesia is frequently required for radiotherapy, and neuro-cognitive outcome is directly related to the technique used, sp expertise is essential. Radiotherapeutic advances in image fusion techniques, 3D conformal radiation therapy, intensity modulated radiotherapy, stereotactic radiosurgery, and proton beam therapy also have improved the safety and effectiveness of radiotherapy but again are not widely available. While most of the chemotherapy regimes use well-established agents, there have been advances with drug delivery such as intra-Omayya interferon for craniopharyingomas. While intensive approaches such as high dose therapy with autologous stem cell rescue have not yielded dramatic improvements in the outcomes of high risk brain tumours, there is a potential role for biological small molecules and monoclonal antibodies, as well as for anti-angiogenic and metronomic strategies.
Increasingly, children receive chemotherapy for brain tumours such as medulloblastomas, ependymomas, and midline astrocytomas. The decision-making, regimens, infrastructure and support in this are critical to its success. Knowledge of the drugs used, in patient management, and follow up protocols are better in the hands of oncologists with specific experience with paediatric brain tumours. Ideally, all of these children should be followed-up in a multidisciplinary clinic, given the risk of recurrence and the potential for long term neurological, endocrine and neuropsychological sequelae; however, this is rarely available. Recent advances in treatment modalities have been promising for childhood brain tumours but are not widely accessible.
One of the most exciting areas in paediatric neuro-oncology has been in molecular biology research. Classification of brain tumours has improved; this relies now not only on the microscopic appearance of tumour cells but also increasingly on immunohistochemical markers, cytogenetic analysis and molecular profiling using genomics and epigenomics. Current work with medulloblastomas illustrates this principle: it is increasingly clear that the entity of medulloblastomas comprises at least 4 distinct subtypes that have epidemiological correlates and prognostic value. It is expected that identification of these subtypes will be required for entry into clinical trials and will eventually guide targeted therapeutic decisions in the future.