Altering Cancer Diagnosis

Researchers at Harvard Medical School have made a landmark discovery in cancer genomics that may permanently shift the way cancer is currently diagnosed.

The study, published in the Feb. issue of Nature, identified 158 regions of the cancer genome that were most frequently altered in cancer.

“What was unique in our approach was that we were interested in alterations across different cancer types,” said HMS student Craig H. Mermel. “We want to stop looking at brain cancer or lung cancer but instead diagnose cancer by mutations common to the disease.”

In a sample of 3,131 cancer specimens of 26 cancer types, the group identified 158 regions of focal somatic copy number alterations (SCNAs) that were frequently altered across several cancer types. Somatic copy number alterations occur when there are abnormal numbers of chromosomes within a tumor’s genome, Mermel said.

These SCNAs include amplifications, in which more than one copy of a gene is present, and deletions, in which the copy is not present at all.

Although the exact role of SCNAs is still unclear, the researchers found that a large number of SCNAs identified in individual cancer types are present over several cancers of different origin. The common SCNAs may act as important markers in the diagnosis and treatment of cancer, according to Rameen Beroukhim, a clinical fellow in oncology at Harvard-affiliate Dana-Farber Cancer Institute.

“Each cancer type obviously has a very complicated copy number profile,” Beroukhim said. “Two cancer types individually don’t look that similar, but we found that over such a large number of samples, copy number profiles looked more similar than different.”

Matthew L. Meyerson ’85, professor of pathology at HMS, emphasized the shift to viewing the diagnosis of cancer on a genetic basis.

“I think that it is part of the change in cancer diagnosis as no longer looking at cancer as lung cancer or breast cancer, but as caused by a particular type of genetic alteration,” Meyerson said.

Although this study is unique in that it indicates a promising universal view of the cancer landscape, Beroukhim said it would not have been possible without the context of current cancer research.

“This study happened now because there are more tumor samples available and in higher resolutions,” Beroukhim said. “We can really determine what gives rise to different cancer types.”

The promise of their findings has already prompted funding from the National Institutes of Health, the Doris Duke Charitable Foundation, the Sarah Thomas Monopoli Lung Cancer Research Fund, the Lucas Foundation, and several drug companies—most notably Novartis—in the hope of developing drugs that may specifically target common genetic alterations across cancer types.

But the researchers said that more work needs to be done before patients can benefit from the findings.

“This is only one particular milestone on a hypothesis that we thought was particularly powerful,” Mermel said.