Brain tumours are the second most common type of tumour in children. Surgical approaches to some of these brain tumours—namely, hypothalamic chiasmatic low-grade glioma (HCLGG), diffuse intrinsic pontine glioma (DIPG), and high-grade glioma (HGG)—are controversial. High-quality studies of surgical approaches for these tumours are lacking. In a recently published multidisciplinary consensus statement, Walker and colleagues attempt to address this limitation.1 Their statement on surgical approaches to low-grade and high-grade astrocytomas and pontine glioma was devised following a meeting of 92 experts in paediatric neuro-oncology at the Consensus Conference on Paediatric Neurosurgery (CPN) held in Paris, France, on 11–12 February 2011.1

The researchers adopted the Delphi method for reaching consensus, seeking more than 70% agreement from more than 60% of participants. After two rounds of surveys, 27 statements meeting consensus criteria were reported (Box 1 and Supplementary Figure 1 online).1 In regard to HCLGG, primary surgical resection was not recommended, and intervention (including biopsy, observation or treatment) should be aimed at limiting the risk of visual damage.1 In regard to DIPG, observation was deemed to be the standard of care, and biopsy was only recommended in the setting of clinical trials.1 For HGG, biopsy was the preferred approach unless extensive resection could be safely achieved without causing an iatrogenic deficit.1

HCLGGs are rare tumours that are typically seen in children less than 10 years of age, and are associated with neurofibromatosis type 1 in 30–50% of cases.2 Symptoms vary and include vision loss, strabismus, proptosis and endocrine abnormalities.2 The majority of these tumours are low-grade gliomas and, historically, biopsy and surgical interventions have not been recommended.2 The CPN consensus statement emphasized that the aim of therapy is to gain time by controlling tumour progression and preserving neurological function.1 Surgery should not be performed routinely, and the decision to pursue biopsy, chemotherapy and/or radiation therapy should be made by a multidisciplinary paediatric neuro-oncology team.

Surgical treatment of HCLGGs is controversial...

Surgical treatment of HCLGGs is controversial, as randomized controlled trials of such an approach are lacking. Radiation has historically been a primary treatment option and surgery has been avoided.2,3 Recent years have seen an increasing trend towards immediate treatment with chemotherapy in children with HCLGG, as radiation is associated with long-term adverse effects, including neurocognitive decline, endocrine abnormalities, and secondary malignancies.2,3 Surgery for these tumours is rarely advocated, but can have an important role in some situations—for example, in helping to preserve neurological function in certain situations including enlarging tumours that might cause visual impairment, tumours causing hydrocephalus, and tumours with a considerable mass effect.4 Moreover, some tumours in the hypothalamic chiasmatic region may not be low-grade gliomas, so tissue samples could be needed for pathology characterization, as well as to determine molecular characteristics for chemotherapeutic targeting.4 It is important, therefore, that surgery still be considered a potential option for these tumours on a case by case basis.

DIPGs typically occur in children aged 5–10 years, and represent 75–80% of paediatric brainstem tumours.5 The median survival for children with these tumours is less than 1 year, and survival has not significantly changed in the past few decades.5 Treatment usually entails radiation with or without chemotherapy.1,5 Surgery for these tumours is rarely advocated, and the CPN's consensus statements recommends that biopsy should only be pursued for atypical DIPG and for typical DIPG as part of a clinical trial, for which detailed information on the molecular characteristics of the tumour can inform interpretation of trial results.1 Biopsy has, however, been routinely performed with minimal morbidity in Europe since 2003, to confirm pathology and guide selection of adjuvant treatments.6 Additionally, because these tumours are biologically heterogeneous, increasing efforts are being made to molecularly characterize the tumour tissue and identify potential molecular targets.5,6 Surgery should, therefore, still be considered—particularly in tumours that are exophytic, contrast-enhancing, or located in the medulla or midbrain, as these tumours have increased chance of being a nonmalignant infiltrative astrocytoma.6 Tissue sampling might also be needed in future to guide chemotherapeutic regimens. These surgical interventions should, however, be performed by an experienced neurosurgeon with the help of an experienced neuromonitoring team.

HGGs include WHO grade 3 (anaplastic astrocytoma) and grade 4 (glioblastoma multiforme) tumours, and account for less than 10% of paediatric brain tumours.7 Although paediatric HGGs are grossly similar to such tumours in adults, they have distinctive biological characteristics at the molecular level.8 Paediatric HGGs more commonly occur de novo, and have a higher frequency of positive IDH1 status compared with adult HGG. Paediatric HGGs also have distinctive chromosomal abnormalities including chromosomal 1q gain.8

Whether extensive resection prolongs survival in paediatric patients with HGG is unclear, owing to a lack of large-scale studies such as have been conducted in adult populations.2,7,9 The CPN reached consensus that surgical biopsy of HGG is needed, and extensive resection should only be pursued if this approach can be safely achieved and is agreed upon by a multidisciplinary panel.1 As in adult patients, extensive surgical resection without causing an iatrogenic deficit can prolong survival, delay recurrence times, and enable more-comprehensive diagnosis and better biological characterization.7,9 Moreover, with development of improved neuroimaging to detail functional tracts, tumours in less accessible brain locations might also be amenable to surgical resection.7,9 Surgery can, therefore, have an important role as first-line therapy for children with HGG.

In light of uncertainty and controversy over optimal management of paediatric patients with HCLGG, DIPG and HGG, the consensus statement from Walker et al.1 represents a welcome addition to the clinical community (Box 1 and Supplementary Figure 1 online). A key limitation of the Delphi method, however, is that it assumes that recommendations are safe when most experts on the panel are in agreement. As such, these consensus statements, although important, only provide a general guideline for surgical approaches to these tumours.

Importantly, each patient should be considered individually. Some patients may have tumours that are more amenable to surgical resection, such as HCLGG causing obstructive hydrocephalus, DIPG with a dorsal exophytic component, or HGG in accessible brain regions. Additionally, the surgical approach considered for each patient should be based on the skill set of the surgeon and the experience of the institution, possibly including guidance from a multidisciplinary team.10 Moreover, molecular characterization of these tumours is becoming increasingly important for optimization of clinical trial design and treatment. Needle biopsies might not always provide sufficient tissue to enable such tumour characterization, and surgical resection should, therefore, still be considered in the treatment armamentarium for children with HCLGG, DIPG and HGG.