Stem Cell-Mediated Oncocidal Gene Therapy of Glioblastoma (GBM)
Brain tumors (BTs) are incurable, whether they start in the brain or spread there from other sites. Despite advances in surgical, radiation, pharmacologic, and gene therapies, survival with a BT remains dismal. Current therapies are limited by their inability to reach widely disseminated tumor cells that become dispersed within normal brain structures. Interestingly, the therapeutic property that is needed to overcome this major obstacle to effective treatment of BTs matches well with one of the better accepted attributes of neural stem cells (NSCs): an attraction for sites of pathology in the adult brain, including primary & metastatic cancer. If armed with a proper tumor-killing gene, NSCs (whether administered into the brain or into the bloodstream), that are drawn to cancers, will dramatically reduce tumor burden, and will track after even single migrating tumor cells. The NSCs perform this action without themselves becoming tumorigenic or augmenting the pre-existing tumor, and this can be assured by having NSCs express a suicide gene that can be activated and cause NSCs to die. The tumor homing phenomenon of NSCs was first revealed by researchers on this proposed team and, in fact, the central concepts presented here have since been extended to many other kinds of disease. In this proposal, we will use a number of authentic mouse models of primary BTs to pre-clinically test therapeutic NSCs. Human NSCs (hNSCs) will be derived from 3 distinct sources, with each having been proffered as therapeutic, but never having been compared head-to-head in treating tumors. Each of these hNSCs will be modified using two therapeutic genes: TRAIL, which is a protein that specifically kills tumor cells, but does not harm normal cells and tissues, and cytosine deaminase which converts a non-toxic chemical into a toxic chemotherapeutic. We expect our research to identify the best hNSC + therapeutic gene combination to advance for clinical trial in patients with BTs, following our obtaining regulatory approval for using hNSC therapy at the end of this project. Because immunocompatibility of the hNSCs with recipient patients is not a concern in BT therapy, a limited number of hNSC lines can be used for treating all prospective patients. Furthermore, BT treatment does not require long-term NSC survival and can be combined with commonly used BT therapies. Finally, NSCs can be imaged in patients and therefore monitored after administration. Developing this approach for treatment of BT patients offers an ideal setting and opportunity for achieving dramatic results from stem cell therapy, and the results of this project will likely be applicable to the treatment of other cancers.
Brain tumors (BTs) are incurable, whether they start in the brain or spread there from other sites. Despite advances in surgical, radiation, drug, & gene therapies, survival with a BT is extremely short, because current therapies are limited by their inability to reach tumor cells that spread widely to normal brain structures. Interestingly, the therapeutic property that is needed to overcome this major treatment obstacle matches well with one of the better accepted attributes of neural stem cells (NSCs): an attraction for sites of disease in the adult brain, including primary & metastatic cancer. If engineered to be armed with a tumor-killing gene, NSCs (whether administered into the brain or into the bloodstream), that are attracted to cancers, could dramatically reduce patient tumor burden, and track after even single migrating tumor cells, in a manner that has never been achieved. The NSCs would perform this action without themselves causing tumors or increasing growth of the patient’s tumor, and this would be assured by engineering the NSCs to self-destruct. The tumor homing phenomenon of NSCs was first revealed by researchers on this proposed team and, in fact, the central concepts presented here have since been extended to many other kinds of disease. In this proposal, we will use a number of authentic mouse models of primary BTs to test therapeutic NSCs before testing them in humans. Human NSCs (hNSCs) will be derived from 3 distinct sources, with each having been proposed as therapeutic, but never having been compared head-to-head in treating cancer. Each of these stem cells will be modified using two different therapeutic genes: TRAIL, a protein that specifically kills tumor cells, but does not harm normal cells and tissues, and cytosine deaminase, which converts a non-toxic chemical into a chemotherapy drug that kills the tumor. We expect our research to identify the best hNSC + therapeutic gene combination to advance for evaluation in clinical trials in patients with intracranial BTs, after we have performed all necessary animal safety testing and submitted a complete plan for review by the US FDA and NIH. Members of this proposed team have experience in bringing cancer therapies to clinical trial, hold the IP surrounding the use of stem cells against cancer, have begun discussions with the FDA and NIH, and have enlisted a GMP facility. Because immune system compatibility between donor and recipient of the hNSCs with the recipient is not a concern in BT therapy, a small number of donors could be used to produce genetically modified hNSCs to treat all prospective patients. Developing this approach for treatment of BTs offers an ideal setting and opportunity for achieving dramatic results from stem cell therapy, and accomplishing substantial improvements in quantity and quality of life for BT patients would no doubt increase California's worldwide visibility in offering the best possible medical care for cancer patients.
During the first year of this project we have made substantial progress toward achieving the ultimate goal of developing a stem cell (SC) therapy for treating patients with recurrent glioblastoma (GBM). At the outset, we began investigating three SC candidates as the cellular vehicle to carry a therapeutic payload and disperse within the tumor of GBM patients: mesenchymal stem cells (MSCs); fetal neural stem cells (fNSCs); and adult neural stem cells (aNSCs). In addition, we were considering two therapeutic genes as the payload, cytosine deaminase (CD) and tumor necrosis factor related apoptosis-inducing ligand (TRAIL), and two routes of therapeutic SC administration for treating brain tumor patients, intravascular and direct intratumoral. Thus, at the start of the project, there were twelve possibilities (3 stem cell vehicles x 2 therapeutic genes x 2 routes of administration) to investigate and compare prior to determining the best combination to develop for a GBM clinical trial. From this starting point we have been able to rapidly eliminate the aNSCs from consideration due to their slow rate of proliferation that would limit their expansion to sufficient number for use in a clinical product for patients. Next we determined that SC access to intracranial tumor through intravascular injection was negligible, and that it is highly unlikely that SC administration by this route would result in a sufficient number of SCs reaching intracranial tumor for achieving therapeutic benefit in treating patients with recurrent GBM. Thus, our work to date has resulted in the narrowing options for SC + therapeutic gene + route of delivery to four: two cellular vehicle candidates (fNSCs and MSCs) and two therapeutic gene payloads (CD and TRAIL). During the first year of this project, each of the four combinations has been tested and have demonstrated anti-tumor activity. During early year 2 research we will determine the most effective combination based on preclinical testing results using multiple human GBM models. The decision regarding most effective therapeutic gene + stem cell vehicle will be achieved within six months, and from that point, in going forward, project emphasis will focus on the development of a specific product candidate, including manufacturing process and assay development, GLP/GMP product manufacturing, further preclinical animal studies to demonstrate efficacy and safety, and development of a clinical protocol. In association with the research accomplished to date we have developed and applied several approaches that will prove useful for SC research and clinical application in general. Foremost among these is the use of micron-sized particles of iron oxide (MPIOs) for labeling SCs prior to their injection into animal subjects, and then monitoring the movement of labeled SCs using magnetic resonance imaging (MRI). This is a powerful technique with implications for understanding the distribution and persistence of SCs in patients receiving SC therapies. For our project, this method is informing us about the distribution of labeled SCs within and around brain tumors, as well as with regard to how long the SCs remain in animal subjects. In addition to the MRI detection of iron particle labeled SCs, we have developed and refined a technique for determining the amount of human SC DNA in animal subject tissues, which has a sensitivity of detecting one human cell among more than a million host cells. Similar to the MRI detection of labeled SCs, the DNA detection method provides us a very sensitive approach for monitoring SC biodistribution and persistence in animal subjects, and it is broadly applicable to all SC research in which rodent models are used for pre-clinical investigation of SCs for treating disease. We are also developing novel approaches for the use of optical imaging to visualize stem cells labeled with fluorescent reporters, and for monitoring the anti-tumor activity of therapeutic stem cells administered to animal subjects. These novel approaches are contributing to the repertoire of techniques available to expedite the identification and application of therapeutic SCs in clinical settings. This project is a collaboration among outstanding scientists and clinicians at five of California’s leading medical research institutions: the Sanford-Burnham and Salk Institutes in La Jolla, and the San Francisco, Los Angeles, and San Diego campuses of the University of California (including Ludwig Institute at UCSD). By leveraging complementary expertise of these investigators, we have made rapid progress in the preclinical animal studies, design of the clinical trial protocol, and the product development studies that will lead to preparation of a gene-modified SC product for the clinical trial. These activities will culminate in an IND application to FDA that will allow us to test the safety and efficacy of our SC product in patients with this devastating illness.
This project was initiated in April of 2010, and was for comparing
• three types of stem cells
• two distinct therapeutic gene modifications of stem cells, and
• intravascular administration vs. direct tumor injection of stem cells
in order to identify the most efficacious stem cell + therapeutic gene + route of administration for treating patients with recurrent glioblastoma (GBM), a brain tumor that has a dismal prognosis, and that badly needs innovative approaches for improving treatment outcomes.
Major conclusions from this project, as concerns the objectives indicated above, are:
1. Stem cells administered by the vascular route do not reach brain tumors established in rodent subjects, to an extent which demonstrable therapeutic stem cell anti-tumor activity should be anticipated. In most instances, intravascular administration results in no detectable stem cells in intracranial tumor in rodent models. Therefore, therapeutic stem cells need to be administered direct into brain tumors in order to achieve a sufficient number and concentration of stem cells for observing anti-tumor effect.
2. Neural stem cells and mesenchymal stem cells delivered directly into intracranial tumor display similar extents of dispersion in tumor, indicating these stem cell types should perform comparably as concerns their ability to disseminate within, and deliver therapy to tumor.
3. However, unmodified (non-immortalized) neural stem cells, derived from single adult or fetal sources, have insufficient proliferative capacity for production as therapeutic stem cells to be used in clinical trials that enroll multiple patients. Because of the ready availability of mesenchymal stem cells (MSCs), from many donors, combined with the proliferative capacity of MSCs, MSCs were determined as the preferred candidate for developing therapeutic stem cells to treat patients with recurrent GBM.
4. Studies conducted with therapeutic stem cell + tumor cell mixtures indicated superior anti-tumor activity of cytosine deaminase modified stem cells + 5-fluorocytosine (FC), relative to secretable TRAIL modified stem cells, when anti-tumor activity is examined in liquid media (cell culture). The two types of therapeutic stem cells showed comparable anti-tumor activities when administered directly into brain tumor in animal (rodent) subjects.
5. In relation to other types of therapies (e.g., chemotherapeutics, antibodies, liposomal drugs) being tested by members of this disease team, manufactured therapeutic stem cells displayed low (modest) anti-tumor activity in animal subjects with brain tumor.
Technical advances, discovery, and products developed in association this project, and that can be shared/transferred in support of other CIRM funded research, include:
• Development of approaches for delivering stem cells through distinct routes of administration in rodent subjects.
• Development of a method, based on the use of polymerase chain reaction, for detecting human cells in rodent tissues, with a sensitivity of detection of one human cell per 100,000 mouse cells.
• Development of a cell labeling approach that enables tracking of stem cell migration in rodent subjects.
• Development of a histochemical method for detection of labeled human cells in rodent tissues.
• Development and characterization of multiple, tumorigenic human glioblastoma xenograft models for use in therapeutic testing.