Verschlimmbesserung: Craniospinal Radiotherapy Is Essential in WNT Medulloblastoma Patients

In this issue of Clinical Cancer Research, Cohen and colleagues (1) present the outcome of a prospective pilot study that eliminated the use of craniospinal irradiation (CSI) in the treatment of Wingless (WNT) medulloblastoma. The study was closed to accrual early due to early (3 and 6 months following completion of therapy) recurrent disease in the first two patients. Of the 6 patients enrolled on the study five patients received irradiation at the time of relapse (n = 3) or at the end of planned chemotherapy (n = 2). Parents of the last patient opted for high dose chemotherapy with stem cell rescue. However, the patient developed myelodysplastic syndrome and needed a haploidentical allogenic transplant. Another study that attempted to treat nonmetastatic WNT medulloblastoma with just focal irradiation therapy and adjuvant chemotherapy also had to be closed early due to recurrent disease in the neuraxis (2). The patients were salvaged by administering high dose (36Gy) CSI, but this salvage strategy defeated the purpose of reduction in long-term toxicity in this population.

Using advanced molecular techniques medulloblastoma is defined as a disease that is composed of four distinct molecular diseases, WNT, Sonic Hedgehog (SHH), Group 3 and Group 4, each with distinct cell of origin, molecular features, and clinical outcomes. WNT medulloblastoma stood out as a subgroup that occurred in approximately 15% of all newly diagnosed patients (3). Distinctive clinical and molecular features in this subgroup include a median age of 10.4 years, female preponderance (F:M ratio – 2:1); classic histology, presence of monosomy 6, nuclear β-catenin on immunohistochemistry and mutations in the CTNNB1 gene, which encodes for β-catenin. These molecular features facilitate identification of these tumors in a timely manner for prospective clinical trials. After the initial reports from prospective clinical trials that have reported an excellent outcome (4, 5), subsequent clinical trials have confirmed the excellent outcome for WNT medulloblastoma treated with surgery, CSI and adjuvant chemotherapy (6, 7).

However, cure often comes at a cost with medulloblastoma survivors often having neurocognitive sequelae that impedes higher education, social functioning, and financial independence (8, 9). Hence, investigators have attempted to modify curative therapy to maintain cure and reduce late effects. Initial studies reported curing medulloblastoma with 36 Gy CSI alone. Subsequent studies reduced the dose of CSI to 23.4 Gy with the addition of adjuvant chemotherapy for patients that had the primary tumor gross totally resected or <1.5 cm² and no metastatic disease (defined as standard risk medulloblastoma), without affecting efficacy (10). However, a recent study that randomized standard risk medulloblastoma patients 3 to 7 years of age between 23.4 Gy and 18 Gy CSI, demonstrated inferior outcome for the latter cohort of patients (7), highlighting the importance of CSI dose in curing medulloblastoma. In this study, the randomization did not account for the molecular subgrouping of medulloblastoma and only seven WNT medulloblastoma patients were randomized to receive 18 Gy CSI. This study revealed that no statistically significant difference in survival was observed between this group and those patients that received 23.4Gy CSI (7).

Building on the excellent results for WNT medulloblastoma treated with 23.4 Gy and adjuvant chemotherapy three prospective trials are testing disease control with 18 Gy CSI (International Society of Paediatric Oncology SIOP-PNET 5 (NCT02066220); Children's Oncology Group, ACNS 1422 (NCT02724579) and 15 Gy CSI (St Jude Children's Research Hospital SJMB12 – (NCT01878617) and adjuvant chemotherapy (Table 1). These studies have completed or are close to finishing accrual and the results are eagerly awaited. Though investigators focus on reduction of CSI dose to reduce long-term sequalae of therapy there have been several refinements in therapy that have led to reduction of toxicity. Proton beam therapy is getting more accessible across the globe and early reports document lesser impact on central nervous system toxicity as compared with photon beam therapy (11). Reduction of the radiation boost dose from the entire posterior fossa to a 0.5 mm margin surrounding the tumor bed has spared the temporal lobes and cochlea from additional radiation exposure. Delivery of radiotherapy using parallel opposed fields to sophisticated 3-dimensional conformal radiotherapy or intensity modulated radiotherapy has spared radiation to critical structures of the brain. Improvements in surgical technique has facilitated safe resection of the tumor and optimized MRI imaging has facilitated accurate documentation of subtle metastatic disease and thus appropriate risk-stratification (12). Finally, reduction in dose and duration of chemotherapy being tailored to clinical and biological risk features can prevent ototoxicity and neurotoxicity that also impact cognitive outcomes. It is hard to quantify the impact that each of these advances and modifications of therapy will have on long-term sequelae but taken together there will be a preservation of cognitive abilities.

The Cohen manuscript demonstrates that CSI is an essential component of curative therapy for WNT medulloblastoma in addition to surgery and adjuvant chemotherapy (Fig. 1). Excellent outcomes for this subgroup of patients is a combination of the biology of the tumor and the delivery of adequate therapy. The refinement in therapy as enumerated above will preserve neurocognition as compared with the prior generation of patients. Fortunately, WNT patients are older so the lower dose of CSI in this age group may spare these patients the neurocognitive decline seen in younger children treated with higher doses of CSI. Pending results of the current studies, further dose reduction of CSI may not be possible without a reduction in cure rate. Prior to planning the next generation of studies, investigators need to document the neurocognitive outcomes of patients treated in the current era and not historical controls.

No disclosures were reported.

This work was supported by Stan Perron Charitable Foundation (Stan Perron Charitable Trust; to N.G. Gottardo) and American Lebanese Syrian Associated Charities (ALSAC; to A. Gajjar).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.