A New Approach to Brain Cancers: Murine Models

Courtesy of Charles River Laboratories Compared to the treatment for other malignancies, brain cancer therapies exist in the dark ages, with prognoses and mortality rates similar to those from 30 years ago. A new generation of murine models, however, could change that situation in relatively rapid order, experts predict. "To put it into perspective how ridiculously bad the [current] situation is, there have been over 100 clinical trials over the past two decades that have translated into approximately one month additional survival for glioblastoma patients, which is unacceptable," says Ronald DePinho, Department of Medical Oncology, Dana-Farber Cancer Center, Boston. DePinho and others want to decipher the causes of malignancies and the unique biological characteristics that impede successful treatment. To do this, they are devising murine models, which recapitulate the human disease, by inserting genetic mutations resembling those detected in human tumors. Scientists now can watch predictable brain tumors develop in mice; they can see how the tumors start and from what cell types they develop. Working with mice that lack the genes patched and p53, Tom Curran, Department of Developmental Neurobiology, St. Jude Children's Research Hospital, Memphis, Tenn., generated a murine strain in which more than 95% develop a tumor that resembles human medulloblastoma and has many of its pathological hallmarks.1 "We don't know that the double mutation, that is, the loss of one copy of patched and two copies of p53, is seen in human tumors," says Curran, "but we do know that there are defects in both the patched pathway and p53 pathway in human tumors." OTHER GENETIC MODELS Not everyone is sold on these mice. Francis Ali-Osman, Department of Neurosurgery Research, MD Anderson Cancer Center, Houston, is not sure that scientists know enough about the critical pathways in developing human central nervous system tumors to accurately model them. "We have to be sure that the mutations are the ones that drive the tumor" before using them for developing new therapies, says Ali-Osman. While the older xenograph models, in which human tumor cells are injected into murine brains, lack the infiltrating tendrils of an endogenous brain tumor, the xenographs do provide scientists the opportunity to work with human tumors, he says. At least five other genetic models of similar quality exist, according to researchers: A pRb-PTEN mutant model for astrocytoma2 Several Ink4a-Arf models that generate oligodendromas, astrocytomas or glioblastomas,3,4 depending on where the mutation is expressed and what other genetic lesions accompany it An NF1-p53 model for grade III astrocytoma5 These models provide a reliable testing ground for new therapies, supporters say; until now, testing potential treatments has primarily occurred in a clinical setting. Luis Parada, Center for Developmental Biology, University of Texas Southwestern Medical Center, who developed the grade III astrocytoma model, says that the reproducible timing and high percentage of animals that develop disease will allow them to do a high-throughput screen of potential therapeutic agents. "We can throw the virtual kitchen sink at them and ask what works," said Parada. In contrast, Curran expects the mice to be most valuable in testing targeted therapies that directly compensate for the mutation, or block the pathways, which are disrupted in the model. His team is already testing a number of small molecules, such as cyclopamine, to see if they can inhibit tumor formation. DEVELOPING ACCURATE SURROGATE MARKERS DePinho expects that the new models will break open the bottlenecks in drug development, illuminating the genes involved in tumor initiation and maintenance, which will enable scientists to make educated decisions about the lead compounds to test in animals and clinical patients. Ultimately, the new models will aid researchers in developing accurate surrogate markers for tumor activity and in determining whether a therapy is working. "In clinical trials, we want to know within one or two doses what a drug is doing," says DePinho. "We want markers of efficacy and specificity as opposed to [markers of] tumor shrinkage and patient survival, which are the Stone Age markers currently in use." With the murine models, researchers can identify proteomic profiles from patients' serum or from noninvasive imaging findings that will let them know if a drug is hitting its target. Though the critics appear to be few, no one thinks that they have generated the definitive model. DePinho agrees with Ali-Osman that not all pathways are known: "We need to know the full spectrum of targets; we have an incomplete view of what genes are involved. But a lot of work going on now will change this in the next year." Rabiya S. Tuma (rabiya@nasw.org) is a freelance writer in Brooklyn, New York. References 1. C. Wetmore et al., "Loss of p53 but not ARF accelerates medulloblastoma in mice heterozygous for patched," Cancer Res, 61:513-6, 2001. 2. A. Xiao et al., "Astrocyte inactivation of the pRb pathway predisposes mice to malignant astrocytoma development that is accelerated by PTEN mutation," Cancer Cell, 1:157-68, 2002. 3. R.M. Bachoo et al., "Epidermal growth factor receptor and Ink4a/Arf: convergent mechanisms governing terminal differentiation and transformation along the neural stem cell to astrocyte axis," Cancer Cell, 1:269-77, 2002. 4. L. Uhrbom et al., "Ink4a-Arf loss cooperates with KRas activation in astrocytes and neural progenitors to generate glioblastomas of various morphologies depending on activated Akt," Cancer Res, 62:5551-8, 2002. 5. H. Li et al., "Epidermal growth factor receptor signaling pathways are associated with tumorigenesis in the Nf1:p53 mouse tumor model," Cancer Res, 62:4507-13, 2002.

A New Approach to Brain Cancers: Murine Models

Interested in reading more?

Become a Member of