Targeted therapies for medulloblastoma: understanding the mechanisms of drug resistance and exploring new peptide-based therapeutics

Medulloblastoma, the most common malignant pediatric brain tumor, is currently treated with nonspecific cytotoxic therapies including surgery, radiation, and aggressive chemotherapy. Despite enhancements in treatment strategies, the majority of medulloblastoma cancer patients are left with significant neurological, intellectual and physical disabilities secondary to the effects of these nonspecific cytotoxic therapies on the developing brain. Therefore, additional therapeutic possibilities need to be explored to minimize adverse effects and increase response rate in medulloblastoma patients. A major challenge of brain tumor surgery is removing all of the tumor tissues while preserving as much healthy brain tissue as possible. The extent of surgical resection can modulate 5-year survival potential by 30- 50%, far greater than the impact of any drug - yet there has never been a clinical trial focused on improving surgical outcomes. With the goal of improving detection of cancer foci during surgical resection, our lab developed NP-chitosan-Chlorotoxin-Cy5.5 nanoparticle, also known as tumor paint nanoparticle. In the first part of my thesis, I have shown simple near-infrared fluorescence based methodology to assess the pharmacokinetic properties of these nanoparticles. These data obtained can be incorporated into early nanoparticle synthesis decisions as well as more detailed documentation required for the FDA approval for human clinical trials. As medulloblastoma exhibits marked intertumor heterogeneity, with at least four distinct molecular variants, patients may require separate therapeutic strategies. The development of a targeted inhibitor of a pathway that is implicated in malignancy, together with identification of patient populations most likely to respond to these therapies may offer more effective therapeutic option than standard treatment strategies available in clinics. The second part of my dissertation work involves investigating the efficacy and pharmacodynamic effects of saridegib (IPI-926), one of the Sonic hedgehog (Shh) pathway inhibitors identified, in Shh-driven mouse model of medulloblastoma. Although saridegib has shown effectiveness in reducing and eliminating tumor cells in mice initially, drug resistance inevitably occurred. Understanding the mechanisms of resistance to targeted therapy and developing strategies to overcome them are critical for achieving long-term efficacy of targeted therapy in patients. Therefore, I further investigated mechanisms adopted by medulloblastoma cells during cancer progression. I found that overexpression of the Pgp drug transporter contributes to drug resistance in Shh-driven medulloblastoma and the combination of saridegib with Pgp inhibitor partially reversed the drug resistance. Our data provide a strong rationale for the use of a concurrent combination of Shh pathway inhibitors with Pgp inhibitors for the treatment of medulloblastoma and suggest that this approach may delay the development of resistance to Shh pathway inhibitors. The last part of my dissertation work involved engineering cysteine rich peptides toward the goal of producing Pgp-insensitive therapies for medulloblastoma. Toward this goal, I used a nature-derived knottin peptide as a scaffold and computationally designed variants that could specifically bind to glypican-2, which is overexpressed in medulloblastoma. My aim was to express only a fragment of midkine, a known ligand of glypican-2, that would bind to glypican-2 but not crosslink with growth factor signaling pathways, thereby inhibiting tumor growth. Our ultimate goal is to generate cysteine knot peptide variants capable of targeting and killing brain tumor initiating cells in medulloblastoma patients. This work may provide significant clinical benefit to human medulloblastoma patients, and may serve as an important step towards integrating knottin drugs into mainstream therapy