Recommended if funds allow
Statement of Benefit to California:
SYNOPSIS: Several highly aggressive human cancers, including melanoma, neuroblastoma and peripheral primitive neuroectodermal tumors—also known as Ewing’s sarcoma family tumors (ESFT)— are believed to be of neural crest origin. All ESFT express characteristic fusion oncogenes, most commonly EWS-FLI1. This proposal will test the hypothesis that expression of this oncogene in neural crest stem cells initiates Ewing’s Sarcoma by disrupting the normal pathway of neural crest differentiation. In Aim 1 the PI will use published protocols to induce human ES cells to undergo neural crest differentiation, and will develop tools to purify those cells. In Aim 2 hES and NC cells will be transduced with the EWS-FLI1 oncogene. The PI proposes to assess the effect of the oncogene on differentiation in vitro, and on tumorigenicity in vivo. Gene expression profiling will be used to determine the extent to which the resulting tumors resemble Ewing Sarcoma’s from pediatric patients. INNOVATION AND SIGNIFICANCE: Studies on neural crest biology have been limited almost entirely to model organisms. These studies represent a pathway for generating tools to study human neural crest development and tumorigenesis. The goal of deriving neural crest stem cells (NCSCs) from hES cells, and then examining their response to an oncogene that causes Ewings sarcoma is novel and important. The PI makes a good case that NCSCs with physiological properties are unlikely to become available, except through derivation from hES cells. Therefore the general approach taken in this grant is right on the mark. Furthermore, one of the great values of hES cells lies in their ability to model disease. The goal of using hES-derived neural crest cells to study the origin and development of Ewing's sarcoma is significant and worthwhile and would also be of benefit for the development of new therapeutic agents. STRENGTHS: The experiments described here offer a new opportunity for understanding the oncogenic mechanism of the EWS-FLI1 fusion oncogene, which remains poorly understood. The PI presents a strong argument that ESFT arise from the expression of this oncogene in progenitor cells, specifically those in the neural crest lineage. Whereas expression of EWS-FLI1 is lethal in most normal cells, progenitor cells that also express Bmi-1 are tolerant of EWS-FLI1 and indeed the two genes cooperate to cause oncogenic transformation. The approach of deriving NCSC from hESC is a good approach and the assays proposed to isolate and identify NCSCs are the right ones, short of more extensive in vivo analyses The PI and colleagues are highly qualified to perform the proposed experiments, the PI having assembled a team of scientists with complementary expertise in neural crest biology, in the biology and clinical oncology of neural crest-derived tumors, and in human stem cell biology. In Aim 2 the PI proposes to use gene expression profiling to determine the extent to which tumors generated after EWS-FLI1 expression in human stem cells resemble ESFT from pediatric patients. This will be greatly facilitated by ongoing NIH funded work in the PI’s laboratory to generate gene expression datasets from clinical ESFT samples. WEAKNESSES: The PI is not yet established as a leader in this area, and has few publications since 1998. Nonetheless, the proposal suggests she is on the right track. A minor consideration is that the PI of this grant is already performing virtually identical experiments in murine stem cells, and is fully funded for that work. The applicant states that the purpose of this new proposal to CIRM is to translate that work into humans, but this seems premature. It could be argued that it would make more sense to wait to see if the murine work is successful rather than trying to run the redundant programs concurrently. Two issues that the PI should think about (if hasn't already) are the following. 1. It is possible that the ability to tolerate EWS-FL1 expression is caused by other mutations, such as deletion of Ink4a, Arf, or p53, rather than Bmi-1 expression or stem cell identity. Are any of these genes normally deleted in Ewing's sarcoma? These experiments might work a lot better if done in Ink4a/Arf or p53 deficient cells. The reviewer realizes that such hES cells are not yet available and offers this as a future option to think about, not as a criticism of this proposal. 2. It might work better to infect NCSCs with the EWS-FL1 lentivirus after derivation from hES cells, rather than infecting the hES cell first. DISSCUSSION: The question was raised as to what are the in vivo assays for generating neural crest cells. Answer was transplantation into chick embryos; this was then followed by question as to whether this can be done with human cells. The reviewers did not know whether this had been done or would work with human cells but noted that thus far, those cells which differentiate and function in the in vitro assay have worked in the in vivo assay. PROGRAMMATIC REVIEW: The Working Group voted to recommend this application for special consideration for funding if additional funds become available based on: 1) the applicant is a young investigator with a lot of potential and 2) the proposal is for a stem cell model of childhood cancer that was scientifically well regarded.