Identification of PIK3CA as an oncogene associated with brain cancers opens the door to new screening processes

Hotspots in two areas of a gene that encodes a specific signaling enzyme, or kinase, are vulnerable to a variety of mutations found in five types of brain cancers, according to a report published in the August 1 issue of the journal Cancer Research.

Mutations in the gene PIK3CA occur spontaneously as part of the brain tumor development rather than being passed genetically between generations, said Hai Yan, M.D., Ph.D., the senior scientist of the studies conducted by a collaborative research team from Duke University, Johns Hopkins University, and the University of Utah.

"PIK3CA mutations are known to occur in as much as 30 percent of colorectal and gastric cancers and glioblastomas and they are also present, to a lesser extent, in breast and lung cancer," Yan noted. "Our studies defined the association of mutant PIK3CA gene in a wider spectrum of adult and pediatric brain tumors as well."

PIK3CA is part of a family of genes that encode lipid kinases, enzymes that modify fatty molecules and direct cells to grow, change shape and move. Kinases have been the focus of recent drug development strategies, with some tumor-inhibiting compounds such as Gleevec, which is a protein kinase inhibitor already in use clinically to thwart tumor growth.

Yan and colleagues pinpointed a cluster of 13 mutations on two particular areas of the PIK3CA gene, exons 9 and 20. The mutations were identified in 14 percent of anaplastic oligodendrogliomas, 5 percent of medulloblastomas, 5 percent of glioblastomas and 3 percent of anaplastic astrocytomas. No PIK3CA mutation variants were found in samples of ependymomas or low-grade astrocytomas.

Nine of the eleven PIK3CA mutations were consistent with alterations observed in the colorectal cancers. Two additional, new mutations were also observed.

Identification of PIK3CA as an oncogene associated with brain cancers opens the door to screening processes that can identify patients for treatment strategies, as well as development of targeted molecular therapeutics aimed controlling brain cancer development through regulation of the errant gene, Yan said.

Yan is an assistant professor of pathology, Duke University Medical Center. His colleagues who contributed to this work include Daniel Broderick, Chunhui Di, Timothy Parrett, Roger McLendon, and Darell Bigner, Duke University; and Yardena Samuels, Jordan Cummins and Victor Velculescu, The Johns Hopkins University Medical Institutions; and Daniel Fults, the University of Utah School of Medicine. This work is supported by the National Institute of Health and Pediatric Brain Tumor Foundation.