Infant brain tumor trials: Beyond feasibility

The management of infants and young children with embryonal tumors of the central nervous system (CNS) remains a major challenge. Early cooperative trials conducted 4 decades ago were designed with the aim to delay the use of craniospinal radiotherapy until the age of 3 years. These trials showed a high rate of progression, usually early during the course of chemotherapy, and as a consequence, a high proportion of these patients received salvage radiotherapy for a 3-year overall survival (OS) ranging between 20% and 50% depending on the type of embryonal tumor. Subsequent trials have explored various concepts, including the use of more intensive chemotherapy and high-dose chemotherapy with autologous bone marrow or stem cell transplant, the administration of intrathecal or intraventricular chemotherapy, the addition of focal irradiation or reduced-dose craniospinal irradiation (CSI) as well as the addition of maintenance chemotherapy. Although progress has been achieved, the outcome of children with embryonal tumors of the CNS remains—with the exception of children with desmoplastic medulloblastoma (MB)—globally unsatisfactory and new treatment modalities are urgently needed. Leary et al report the results of a pilot study, PBTC-026, that tested the feasibility of adding non-cytotoxic agents (vorinostat and isotretinoin) to a sequential high-dose chemotherapy backbone for young children with CNS embryonal tumors.¹ The rationale for this combination was essentially based on data supporting the activity and the synergism of the 2 compounds in preclinical models of MB.

Prospective collection of tumor samples allowed the investigation of the prognostic value of histological and molecular markers. Over the decade during which the study accrued 31 patients and outcome data matured, advances in the knowledge of molecular and genetic alterations involved in these tumors have revolutionized their risk stratification as well as their management. As stated by the authors, this information was not available at the time of study design, and the study was not powered to assess the prognostic value of molecular subgrouping for the MB subgroup. However, similar to other recent studies, Leary et al identify differences in outcomes according to the molecular MB subgrouping, and this work further supports the abandonment of the obsolete statement that all MB in younger children are high risk. With intensive chemotherapy regimens, the large majority of young children with sonic hedgehog (SHH) MB, which constitutes about 50% of MB subtype in this age group, will be long-term survivors. The 80% ± 25% survival rate in the SHH MB subgroup is in keeping with recent series that reported a PFS (progression-free survival) 86.2%-100%, a figure that compares very favorably to the survival in older children with average-risk MB treated with adjuvant chemotherapy and CSI.^2–6 The role of focal radiotherapy in this context is questionable for this group of infants and young children, considering the excellent outcome observed with radiation sparing approaches.

Furthermore, as described in the PBTC-026 series, and recently in a preliminary report on COG ACNS0334, the outcome of patients’ SHH remains excellent even in the presence of metastatic disease.³ Considering the almost complete overlap of SHH subgroup with the group of patients with nodular desmoplastic histology, Dhall et al have reported a 5-year event-free survival (EFS) of 82% ± 12% for the metastatic ND MB patients treated on Headstart III, suggesting that the negative prognostic value of metastatic status could be overcome by the use of high dose chemotherapy (HDC) in this context.⁷ Outcome data on metastatic SHH MB treated with conventional chemotherapy and serial intraventricular injections of methotrexate with the HIT-SKK protocol or with the Headstart IV HDC regimen are awaited to further confirm the lack of prognostic value of dissemination in SHH MB in young children treated with intensive treatment strategies.

By contrast, group 3 MB, the second most important group of MB in young children, remains a challenging group of tumors still warranting their classification into the high-risk category. As with the PBTC-026 strategies, the survival of patients diagnosed with this molecular subgroup in the HIT-2000 clinical trial or treated with HDC using the 99703 backbone remains unsatisfactory ranging from 36% to 49% at 5 years.^2,4 The final report of the COG ACNS0334 is awaited with great attention to confirm the survival benefit observed for high-risk group 3 MB patients (metastatic and/or incompletely resected) who received high-dose methotrexate prior to HDC consolidation, indicating a 5-year OS of 80% compared to 40% for those who did not.³

For CNS embryonal tumors other than MB, the PBTC-026 study further highlights the critical need of molecular investigations to provide accurate pathological diagnosis, as currently incorporated in the new 2021 WHO classification of CNS tumors for tumors, such as embryonal tumor with multilayered rosettes (ETMR) or infant high grade gliomas. The diagnosis may not alter the treatment path for tumors like ETMR in the absence of a clear consensus on treatment, but more specific prognostic information can be conveyed to the family. ETMR remains a very aggressive CNS tumor of infancy, despite encouraging recent data reported in a subset of patients treated with adjuvant HDC and radiotherapy with a 2-year EFS of 66%.⁸

For the first time in this context, PBTC-026 demonstrated the feasibility of adding non-cytotoxic agents to an intensive chemotherapy backbone, and most patients were able to complete induction therapy within the prespecified feasibility window. However, will there be support to incorporate vorinostat and isotretinoin into the next generation of clinical trial of CNS embryonal tumors? This raises the question of the role of biological agents in this disease. The recent publication of the SJYC07 trial for infants and young children with MB did not provide convincing evidence of the benefit of adding erlotinib to a maintenance chemotherapy backbone.⁹ Despite preclinical evidence that supports the use of vorinostat in combination with isotretinoin in MB, the PBTC-026 was not designed to primarily answer the efficacy question of that combination. The early closure for futility of the isotretinoin randomization in the ACNS0332 for older children with high-risk MB somehow altered the enthusiasm for this agent for younger children.¹⁰

The attempt to improve local control with focal radiation, used in the PBTC-026, does not seem necessary for young children with SHH MB, in light of their excellent survival in the absence of irradiation in other recent clinical trials.^2–4,7 By contrast, this treatment modality appears insufficient for group 3 MB in these young patients given their disseminated pattern of relapse described by the authors and others.^2,11

While PBTC-026 trial is providing precious information on the histological classification and molecular characterization of CNS embryonal tumors, given their rarity in young children, this work also exemplifies the critical need for larger cooperation in these orphan conditions. No progress can be achieved with a limited cohort of patients accrued over a prolonged period of time anymore. There is an urgent need to collaborate internationally at a larger scale, as done in other tumor groups such as for infant acute lymphoblastic leukemia, to be able to appropriately address therapeutic questions with larger cohorts and allow the evaluation of new promising agents with rapid turnaround. Ongoing discussions are currently being held between North American and European groups to join forces on common treatment strategies and to undertake pooled data analysis to refine risk stratification.

Funding

No funding.