Medulloblastoma and other primitive neuro-ectodermal tumors (PNET) are the most prevalent malignant tumors arising in the central nervous system, representing together more than 25 percent of all childhood central nervous system (CNS) tumors. The peak onset of these tumors is < 6 years of age; accordingly, not only is it difficult to cure young children of these tumors, but the “price” of cure using the traditional first-line modality of CNS irradiation commonly results in irreversible and substantial developmental, intellectual and physical problems that actually increase over the ensuing years due to premature onset of cerebrovascular disease.
In addition, the risk of developing irradiation-induced second malignant brain cancers is heightened in these young children; recent reports suggest that for children over 3 years of age with otherwise “standard-risk” medulloblastoma, despite a 75 percent 10-year disease-free survival, the risk of dying from a second cancer more than 5 years post-diagnosis is greater than the risk of dying from medulloblastoma recurrence.
The concern for serious side effects in young children with brain cancer treated with radiotherapy had already been recognized by the early to mid-1980s; a series of prospective clinical trials in both North America and Europe were developed in the 1980s and 1990s, mainly for children less than 3 years of age, attempting to use conventional chemotherapy to either delay or avoid CNS irradiation. Such trials demonstrated that long-term survival without irradiation was indeed achievable, albeit in a minority of the children. Clinical factors that contributed to prolonged survival were extent of surgical resection of the primary tumor and absence of metastatic disease at initial diagnosis.
Significant improvement in outcome for young children with localized medulloblastoma was reported from the German HIT clinical trial; the improvement was associated with the introduction of both intravenous high-dose methotrexate as well as intraventricular methotrexate. In addition to the previously recognized favorable prognostic clinical features, they reported that the 50 percent of children who harbored the subtype of medulloblastoma recognized histologically as desmoplastic/nodular was associated with an especially favorable survival (>80%).
Head Start 4
A different approach was initiated in the early 1990s under the aegis of the “Head Start” Consortium of institutions. In recognition that most young children with primary malignant CNS tumors developed tumor recurrence within 6 to 12 months of initial diagnosis, a strategy of a total of 6 months of intensive chemotherapy was developed for children less than 6 years of age with all primary CNS malignant tumors, including five cycles of intensive induction chemotherapy (vincristine, cisplatin, cyclophosphamide, and etoposide), followed by a single consolidation cycle of marrow-ablative chemotherapy (thiotepa, carboplatin and etoposide), followed by autologous hematopoietic cell rescue–and no irradiation.
Three sequential Head Start clinical trials have since been conducted, completed and largely published. These studies have demonstrated the best outcomes yet reported for young children with desmoplastic/nodular medulloblastoma, anaplastic medulloblastoma, and for young children with disseminated medulloblastoma. The strategy of consolidating initial remission with marrow-ablative chemotherapy has now been incorporated into clinical trials in both North America and Europe for young children with medulloblastoma, CNS PNET and CNS atypical teratoid/rhabdoid tumors (Pediatric Blood & Cancer 2014;61:907-912 and Pediatric Neurology 2015;53:31-46). In addition, long-term follow-up studies of surviving children have demonstrated remarkable preservation of intellectual, developmental and physical functioning in those children avoiding CNS irradiation.
The Head Start 4 (HS4) prospective clinical trial recently opened at the lead institutions of Nationwide Children's Hospital, Columbus, and the Children's Hospital of Los Angeles. Over the next six months, an additional 40 to 50 institutions throughout the U.S., Canada, Australia, New Zealand, the United Kingdom, and Brazil will begin participation in the HS4 trial. This trial will be restricted to children with newly diagnosed, previously untreated medulloblastoma and CNS PNET. Children less than 6 years of age will be eligible and treated without irradiation unless they have residual confirmed tumor at the conclusion of consolidation. Children with CNS PNET and disseminated or unresectable residual medulloblastoma will be eligible and will receive reduced doses of craniospinal irradiation following completion of consolidation.
The HS4 trial will use a PCR-based gene expression assay (MBL-31) developed at Children's Hospital Los Angeles to classify medulloblastoma tumors into four molecular subgroups with prognostic significance, which in turn will be used to tailor treatment. The first primary study question is whether outcome for higher-risk children defined molecularly (Wnt or Shh subtypes of medulloblastoma, known to be low risk, whereas groups 3 and 4 medulloblastoma are known to be higher-risk) and children with CNS PNET can be improved by dose-intensifying consolidation with three sequential cycles of marrow-ablative chemotherapy followed by autologous hematopoietic cell rescue compared, in a randomized fashion, with the “standard” single-cycle consolidation therapy.
Additional objectives include:
- Identifying molecular profiles that predict successful avoidance of irradiation in these children;
- Documentation of improved outcomes for molecularly defined lower risk children (Shh or Wnt medulloblastoma) with just three induction cycles of chemotherapy followed by a single cycle of consolidation—no more than 12 weeks of treatment in total;
- Confirmation of improved outcomes with less morbidity for children through dose-compressed induction chemotherapy; and
- Use of second surgery to resect residual tumor, leading to earlier achievement of complete response (by the end of three induction cycles), proceeding then to consolidation.
In addition, intensive monitoring of hearing with state-of-the-art technologies and correlating imaging changes with neuropsychological function testing in follow-up will permit early interventions with a goal of reducing long-term morbidities of treatment.
We anticipate that the HS4 clinical trial will not only afford the current best available treatment for cure and quality of survival for these young children, but will set a new benchmark for future studies using not only molecular profiling for risk-stratifying, but also for identifying and implementing targeted therapies that can be incorporated within the current HS4 strategy that may afford yet further improvements in outcome concomitant with further reduced morbidities of such therapy.