The investigators, from The Children's Hospital of Philadelphia and Thomas Jefferson University, say their technique may be readily adapted for other types of cancer. The proof-of-principle study appears in the August issue of Genome Research.
One advantage of their technique is its flexibility, said co-author John M. Maris, M.D., a pediatric oncologist at The Children's Hospital of Philadelphia. "As future research identifies other genes active in neuroblastoma, we can modify the microarray to include such regions," he added.
"We have customized this tool for neuroblastoma, but the approach might also be adapted to other types of cancer in which DNA changes are important," said co-author Paolo Fortina, M.D., Ph.D., professor of medicine at Jefferson Medical College of Thomas Jefferson University in Philadelphia and section chief, Genomics and Diagnostics, in the Jefferson Department of Medicine's Center for Translational Medicine.
The most common cancer found in infants, neuroblastoma strikes the peripheral nervous system, often appearing as a solid tumor in a child's chest or abdomen. Some types of neuroblastoma are low risk, resolving after surgeons remove the tumor, while others are much more aggressive. Identifying the correct risk level allows doctors to treat aggressive cancers appropriately, while not subjecting children with low-risk cancer to overtreatment.
Cancer researchers have pinpointed specific genetic abnormalities that influence the aggressiveness of neuroblastoma. An important abnormality is loss of heterozygosity (LOH), the deletion of o ne copy of a pair of genes. When the gene involved is a tumor suppressor gene, LOH removes a brake on uncontrolled cell growth, the growth that is the hallmark of cancer.
Researchers in Dr. Maris' laboratory previously established that LOH in a region of chromosome 11 allows aggressive neuroblastoma to take hold. The new microarray can detect such gene defects on chromosome 11 and other genetic regions implicated in neuroblastoma.
Microarrays are silicon chips that contain tightly ordered selections of genetic material upon which sample material can be tested. When DNA bases from a sample bind to complementary sequences on the microarray, they cause fluorescent tags to shine under laser light. This is a signal that a particular gene variation is present in the sample.
"We can test DNA from peripheral blood and from the tumor, and we should see a loss of signal in the cancer," said Dr. Fortina. He noted that the researchers can simultaneously evaluate seven chromosomal regions known to be involved in neuroblastoma.
Unlike gene expression microarrays, which detect varying levels of RNA to measure the activity levels of different genes as DNA transfers information to RNA, the current microarray directly identifies changes in DNA. "These DNA changes, involving gain or loss of genetic material, are important for neuroblastoma prognosis," said Dr. Maris.
In pinpointing specific regions of chromosomes with loss in DNA, the technology may help confirm a clinical diagnosis, said Saul Surrey, Ph.D., professor of medicine and Associate Director of Research at the Cardeza Foundation for Hematologic Research and the Division of Hematology at Jefferson Medical College. If a clinical diagnosis isn't known, the method might provide some clues.
The microarrray described in the paper has only been used in their laboratory study, but the researchers hope that with further study it may become more widely available as a diagnostic tool for oncologists treating patients with neuroblastoma, and possibly for other cancers.
In addition to Drs. Maris, Fortina, and Surrey, other co-authors are George Hii, Peter S. White, Ph.D., and Eric Rappaport, Ph.D., of The Children's Hospital of Philadelphia; and Craig A. Gelfand, Ph.D., and Shobha Varde, M.S., of Orchid Biosciences, Princeton, N.J. Grants from the National Institutes of Health and the Children's Oncology Group supported the work.