Preclinical Studies Yield New Pathway Clues in Medulloblastoma

Rakesh K. Jain, PhD

Director, Edwin L. Steele Laboratory for Tumor Biology

Massachusetts General Hospital

Boston, MA

Since the mid-1970s, Rakesh K. Jain, PhD, has pursued an interest in the role that tumor microenvironment plays in drug delivery and efficacy. His work has led to numerous advances in the field, including imaging technologies, innovative mathematical models, and use of antiangiogenic agents.

Jain, who is director of the Edwin L. Steele Laboratory for Tumor Biology at Massachusetts General Hospital and the Andrew Werk Cook Professor of Tumor Biology at Harvard Medical School, recently discussed a rewarding new finding that has grown from his research with OncologyLive.

Through studies with mice, Jain and his colleagues have gained insights into the role of the stroma surrounding pediatric brain tumors classified as medulloblastomas. They have also proposed a means of fighting the disease with existing experimental agents that target a specific molecular pathway that has a critical role in the tumor-stroma interaction. Their findings were published in February in the journal Cell.

The results, which could potentially improve outcomes for young children with medulloblastoma, “are among the most gratifying and compelling in my scientific career,” Jain said.

Q: Why is it important to find new therapies for children with medulloblastoma?A:Medulloblastoma is the most common malignant pediatric brain tumor, comprising 20% of all brain tumors in children. The average age at diagnosis is young, at 5 years, but infants to teenagers can be afflicted. Current treatment includes a multimodality approach of surgery followed by radiation and chemotherapy. Although survival rates have been high with these treatment modalities, they come with significant side effects that are especially pronounced the younger a patient is at the time of treatment. These serious side effects include impaired neurocopsychiatric abilities that manifest over time, infertility, impaired growth, adverse effects on hearing and hormonal function, as well as significant risk of second malignancy in the future. Children with this disease are in desperate need of more effective treatments that don’t cause such potentially devastating morbidities.

Q: What did your study entail?A:We studied the role of placental growth factor (PlGF) in medulloblastoma and examined it as a potential target for treatment. PlGF does not appear to be necessary for a child’s development, so inhibiting it wouldn’t cause the same kinds of morbidities that current treatments do. In mice, we demonstrated that PlGF is, in fact, necessary for the growth of medulloblastoma, and analyzed how it is generated and where it is expressed in this disease. We then obstructed the PlGF pathway using antibodies against PlGF or its receptor, neuropilin 1 (NRP1), in these mice and showed that the treatments led to impressive tumor regression, tumor-growth delay, and improved survival.

Q: Please describe the pathway you identified and the treatments that were used.A:Like any solid tumor, medulloblastoma recruits a supporting stroma from the host tissue. Our data suggest that the protein sonic hedgehog (Shh)—secreted by the cancer cells—stimulates PlGF production by granule cells in the recruited stroma. In turn, PlGF binds to its receptor, NRP1, on cancer cells. Although widely considered a nonsignaling receptor, NRP1 conveys pro-survival signals to medulloblastoma cells upon PlGF stimulation via activation of the mitogen- activated protein kinase (MAPK) pathway. Thus, the PlGF/NRP1 axis may be an excellent target for therapies against this disease. In our investigation, we used several anti-PlGF antibodies: TB403 (from Roche) and C9.V2 (from Genentech), both of which block mouse and human PlGF; 5D11D4 (from Roche/ThromboGenics), which is specific to mouse PlGF; and an anti-NRP1 antibody (from Genentech). Previous studies showed that targeting Shh in Shh-mutated medulloblastoma resulted in promising initial responses. However, those initial responses were followed by resistance. We believe that using PlGF or NRP1 antibodies in these patients will yield better results because they block the tumor-stroma interaction and because they may work across different medulloblastoma subtypes, irrespective of Shh pathway mutation status.

Q: Did the treatments in your study cure the mice of medulloblastoma?A:Although tumors didn’t completely disappear with these treatments, they regressed dramatically and the mice carried on without symptoms. It is conceivable, though, that a subset of cells is not dependent on the PlGF/NRP1 axis for survival and could cause recurrence of medulloblastoma. Therefore, anti-PlGF or anti-NRP1 therapies would likely be used in combination with conventional regimens, at least at first.

Q: What is the next step that should be taken to follow up on this research?

A:We are currently planning clinical trials in children with medulloblastoma and other pediatric tumors. Since anti-PlGF and anti-NRP1 antibodies have already been tested in adult patients, the time is ripe for these trials.

Major Molecular Subtypes of Medulloblastoma

Subtype 1—Wnt Tumors

• Shows Wnt-signaling gene expression signature and beta-catenin nuclear staining
• Usually classified as classic medulloblastoma tumors
• 6q loss, CTNNB1 mutations, occasional MYC overexpression
• Typically not metastasized at diagnosis
• Primarily observed in older children, adolescents, and adults

Subtype 2—Sonic Hedgehog (Shh) Tumors

• Characterized by chromosome 9q deletions
• Desmoplastic/nodular histology
• Mutations in genes in the Shh pathway, including PTCH1, PTCH2, SMO, SUFU
• Prognosis appears to be negatively affected by chromosome 17p loss, chromosome 3q gain, chromothripsis, TP53 amplification, and the finding of large cell/anaplastic histology
• Primarily observed in children younger than 3 years, as well as older adolescents and adults

Subtype 3—Variety of Mutations

• Either classic or large cell/anaplastic histology
• Mutations vary, including the presence of i17q and MYC amplification
• Frequently metastasized at the time of diagnosis
• Least favorable outcomes among molecular subtypes
• Occur throughout childhood and may occur in infants, with males outnumbering females in a 2:1 ratio

Subtype 4—Several Mutations

• Either classic or large cell/anaplastic histology
• CDK6 amplification, MYCN amplification, and possibly an i17q abnormality
• Prognosis affected by chromosome 17p loss
• Metastasis at diagnosis common, but not as frequent as in subtype 3
• Occur throughout infancy and childhood, and into adulthood

Source: Cellular and molecular classification of CNS embryonal tumors. National Cancer Institute website. http://www.cancer. gov/cancertopics/pdq/treatment/childCNSembryonal/ healthprofessional/Page2#Section_459. Updated August 1, 2013. Accessed August 14, 2013.

Snuderl M, Batista A, Kirkpatrick ND, et al. Targeting placental growth factor/neuropilin 1 pathway inhibits growth and spread of medulloblastoma. Cell. 2013;152(5):1065-1076.