A study led by Fred Hutchinson researchers, Drs. Andrew Hallahan and James Olson showed that retinoids killed cancer cells from medulloblastoma tumors that had been surgically removed from patients as well as tumors that had been grafted onto mice. Through genome analysis, the scientists also identified a protein in medulloblastoma cells that is triggered by retinoids to initiate cell death, a finding that is likely to lead to the development of additional therapies for the disease. The study appears in the Aug. 3 issue of Nature Medicine.
Because retinoids already have received approval from the Food and Drug Administration for treatment of another childhood cancer, neuroblastoma, researchers expect that the drugs will enter clinical trials in pediatric medulloblastoma patients with minimal delay.
Medulloblastomas arise from primitive cells in the back of the brain, or cerebellum, a region important for motor control and spatial orientation. The disease primarily strikes children under the age of 7. Standard therapy, which includes surgical removal of the tumor followed by radiation and a year of chemotherapy, offers about a 70 percent chance of survival for children over age 3 who do not have recurrent cancer.
High-risk children, who include those under age 3 or who have recurrent disease, have a much lower chance of survival with standard therapy. For that reason-and because of the toxic side effects of radiation and chemotherapy in young children-scientists are eager to find new treatments.
According to Hallahan, retinoids could prove to be even more effective at treating medulloblastoma than neuroblastoma.
"These compounds work against neuroblastoma and other cancers because they trigger cells to differentiate (form specialized cells) and stop dividing," he said. "But we observed that when the compounds are applied to medulloblastoma tumors, a large percentage of cancer cells actually die."
Retinoids are molecules naturally produced by the human body, where they play a critical role in normal development by triggering primitive cells to become specialized cells characteristic of a particular tissue, such as nerve cells in the brain. Scientists also have created synthetic retinoids, which have proved to be effective against some tumors because they drive cancer cells from their relatively primitive, undifferentiated state into specialized cells that cease to divide.
Hallahan and colleagues examined the effect of three retinoids on medulloblastoma tumor specimens obtained from surgeries.
"Thanks to our colleagues at Children's Hospital and Regional Medical Center in Seattle we were able to start our experiments right in the operating room by getting surgical sections that we could immediately put into culture medium," James Olson said. "This was a critical step because it was uncertain whether the available medulloblastoma cell lines (cells previously extracted from tumors and grown indefinitely in the laboratory) would accurately reflect the disease."
The experiment led to a surprising result, Hallahan said. "We began by looking for signs of differentiation, which is what we expected based on how the compounds affect neuroblastomas," he said. "About 5 to 10 percent of the medulloblastoma cells did differentiate. But what was immediately obvious to us was that there was a huge wave of cell death."
The researchers also found that mice transplanted with medulloblastoma tumors that were treated with retinoids developed tumors that were about a third of the size as those that grew on untreated mice.
To identify genes responsible for the cell death, researchers looked at patterns of gene expression in medulloblastoma cells that were exposed to retinoids compared to untreated cells. Using DNA microarrays, which permit the analysis of thousands of genes simultaneously, they identified common sets of genes that were turned on or off in response to treatment with each of the three compounds.
Among the genes found to be switched on by retinoids was BMP-2 (bone morphogenetic protein-2), which codes for a protein thought to play a role in cell death and in the development of nervous system tissue that gives rise to medulloblastomas.
The addition of purified BMP-2 protein to medulloblastoma cells caused significant cell death, even to cells that were resistant to retinoids. Further, when retinoid-senstive cells were grown in close proximity to resistant cells, both types of cells were killed after addition of retinoids. This result indicates that BMP-2 protein secreted by the sensitive cells triggers killing of the neighboring drug-resistant cells.
"This has significant clinical implications," Hallahan said. "Tumors contain mixture of cell types, some of which may be resistant to certain drugs. Our results suggest that retinoid treatment could still manage to kill resistant cells that are in proximity to drug-responsive cells in the tumor."
Olson said that it also suggests that BMP-2 or other proteins "downstream" in the retinoid-induced cell-death cascade could be potential targets for new anticancer drugs.
A proposed clinical trial for high-risk pediatric medulloblastoma patients to compare standard therapy plus retinoids to standard therapy alone is now under review by the Children's Oncology Group, a National Cancer Institute-supported clinical trials cooperative group whose member institutions are devoted exclusively to childhood and adolescent cancer research. Olson, who will serve as principal investigator of the study, hopes to begin the trial at 235 hospitals sometime within the next year.
"It's tremendously satisfying to see a new therapy move so quickly through the pipeline," he said. "One major reason this has progressed so rapidly is that retinoids already have been approved for use in childhood cancers. But equally important has been the center's unique research environment-with strengths in laboratory science and pediatric oncology-combined with the opportunity for collaboration with physician's at Children's, which allowed us to take this study all the way from basic genetics to propose a clinical trial," Olson said.
Coauthors include Olson lab members Joel Pritchard, Ryan Overland and Dr. Andrew Strand; Dr. Stephen Tapscott of the Human Biology and Clinical Research divisions; and Drs. Richard Ellenbogen and Russell Geyer of Children's Hospital and Regional Medical Center. Funding for the study included grants from the Emily Dorfman Foundation, the Burroughs Wellcome Fund and the Damon Runyon Cancer Research Foundation.
The Fred Hutchinson Cancer Research Center, home of two Nobel Prize laureates, is an independent, nonprofit research institution dedicated to the development and advancement of biomedical research to eliminate cancer and other potentially fatal diseases. Fred Hutchinson receives more funding from the National Institutes of Health than any other independent U.S. research center. Recognized internationally for its pioneering work in bone-marrow transplantation, the center's four scientific divisions collaborate to form a unique environment for conducting basic and applied science. Fred Hutchinson, in collaboration with its clinical and research partners, the University of Washington and Children's Hospital and Regional Medical Center, is the only National Cancer Institute-designated comprehensive cancer center in the Pacific Northwest and is one of 39 nationwide. For more information, visit the center's Web site at http://www.fhcrc.org.
Journal
Nature Medicine