The goal of this project is to develop and validate novel biological reagents against glioblastomas (GBMs), which are one of the most common and the most aggressive kind of brain tumors. GBMs are highly invasive in brain tissue and their dispersion is exacerbated by cytotoxic and antiangiogenic therapies, resulting in a much worse clinical picture when the tumor recurs. Novel therapeutics addressing the biology of GBM should be able to disrupt invasion or vascularization and sensitize tumor cells, thus causing tumor disruption in multiple fronts when combined with conventional therapies. This project will focus on the protein fibulin-3 identified by our group, which is secreted by GBM cells and has unique localization and mechanisms in GBM but is absent from normal brain. Fibulin-3 is a novel activator of the Notch pathway and genetic targeting of this protein disrupts Notch signaling in the tumor and reduces GBM growth and resistance. We have produced a function-blocking antibody against fibulin-3 ("mAb428.2") and in this project we will use it as a starting point to generate and validate a family of anti-fibulin-3 recombinant reagents. In Specific Aim 1, we propose to validate safety and efficacy of mAb428.2 against intracranial GBMs to advance it towards investigational new drug status. We hypothesize that anti-fibulin-3 will reduce tumor growth and vascularization with no significant peripheral toxicity or local off-target effects. We will study mAb428.2 specificity, cross-reactivity, biodistribution, safety and anti-tumor efficacy following FDA guidelines, aiming at completing pre-IND validation. In Specific Aim 2, we propose to develop and validate an anti-fibulin-3 single-chain immunotoxin as cytotoxic agent for GBM. We have already generated a single-chain variable fragment (scFv) derived from mAb428.2 and hypothesize that the smaller size of this fragment will improve access and toxicity against GBM. We will conjugate the scFv with Pseudomonas exotoxin and will validate this novel immunotoxin for specificity, safety and efficacy against intracranial GBM. Finally, using fibulin-3 scFv, in Specific Aim 3 we propose to generate and test a chimeric antigen receptor (CAR) against fibulin-3. We hypothesize that a fibulin-3-binding CAR expressed in T lymphocytes will be robustly activated in the tumor microenvironment and will result in strong anti-tumor response even in presence of tumor cell heterogeneity. This is a high-risk/high-reward concept for future immunotargeting of the tumor stroma, and the goal of this aim will be to establish a proof of feasibility and demonstrate that such CAR can be generated and specifically activated by fibulin-3. Successful completion of this project will result in a family f lead biological agents targeting the GBM microenvironment, with different levels of preclinical validation. These results will be used to advance the most promising candidate(s) for submission as investigational new drug(s) for GBM.