New Data on Subtyping Ovarian Cancer
Nashville-based molecular diagnostics company Insight Genetics presented new data on classifying distinct molecular subtypes of serous ovarian cancer at the 2015 ASCO Annual Meeting.
In a poster titled, “BL1 Gene Expression Subtype Predicts Outcome in Serous Ovarian Cancers,” the Insight Genetics team highlighted a novel approach in the diagnosis and treatment of ovarian cancer, applying the company’s proprietary Insight TNBCtypeTM algorithm to help classify which subtype of ovarian cancer a woman might have. The methodology builds upon subtyping technology for triple-negative breast cancer (TNBC) Insight Genetics exclusively licensed from Vanderbilt University Medical Center.
Based on prior observations of genetic abnormalities shared between TNBC and serous ovarian cancer, the diagnostic company applied Insight TNBCtype to existing serous ovarian cancer gene expression datasets and noted a similar distribution of molecular subtypes as previously found in TNBC. Moreover, based on treatment data studied to date, certain subtypes appear to have promise in predicting patient outcomes to various therapies in ovarian cancer.
“Similarities between triple-negative breast cancer and ovarian cancer have been observed in the past,” said Insight Genetics Chief Technology Officer Rob Seitz. “It is exciting to see that the Insight TNBCtype subtyping system is showing similar patterns in ovarian cancer. This potentially will allow personalized therapies in breast cancer to be more rapidly adopted in ovarian cancer and vice versa.”
According to the American Cancer Society, about 21,000 women will be diagnosed with ovarian cancer this year, and 14,000 will die of the disease.
PathGroup Launches SmartGenomics™ Cancer-Specific Testing
Last month, Brentwood-based PathGroup announced the launch of new cancer-specific SmartGenomics™ testing, expanding their genomic profiling menu for community-based oncologists. The addition of five diagnostic panels targeting cancers of lung, colon, brain, thyroid and melanoma/GIST represent testing options for newly diagnosed patients, the result of which will enable physicians to make more precise therapeutic recommendations and clinical management decisions.
SmartGenomics uses multiple technologies to provide a complete genomic picture of a patient’s individual cancer. Each panel offers cost-effective, tailored genomic testing for its respective disease, utilizing guideline recommendations from the National Comprehensive Cancer Network (NCCN). Standard of care diagnostic features include expanded RAS testing in colorectal cancer, biomarkers EGFR, ALK, ROS1, RET, MET in lung cancer, and BRAF, KIT mutation testing in melanoma.
“As evidenced by the plethora of new and upcoming genomically driven clinical trials, integrating genomic profiling into patient management and therapy selection is the future of clinical oncology practice and cancer pharmacotherapeutic development,” said Pranil K. Chandra, DO, medical director of Molecular Pathology Services at PathGroup.
Results from PathGroup’s cancer-specific panels provide treating clinicians with actionable information, linking the findings to available clinical trial options, therapies or both. Powered by the SmartGenomics platform, PathGroup offers comprehensive disease interrogation through all stages of cancer. The personalized medicine menu includes 62 and 85 gene panels for relapsed or refractory solid tumors and hematomalignancies, whole genome array analysis of greater than 22,000 genes, and disease-specific sequencing at initial diagnosis.
NIH Researchers Pilot Predictive Medicine by Studying Health People’s DNA
A new study by National Institutes of Health researchers has turned traditional genomics research on its head. Instead of trying to find a mutation in the genomic sequence of a person with a genetic disease, they sequenced the genomes of healthy participants, then analyzed the data to find “putative,” or presumed, mutations that would almost certainly lead to a genetic condition.
Out of almost 1,000 volunteers whose genomes were examined, about 100 had genomic variants predicting that they would have a rare disease. Almost half of them indeed had the disease when researchers went back and carefully evaluated them, said Leslie G. Biesecker, MD, chief of the Medical Genomics and Metabolic Genetics Branch (MGMGB) at the National Human Genome Research Institute (NHGRI) and corresponding author of the study published in the American Journal of Human Genetics on June 4.
“We were surprised that this many individuals had positive findings in a group of individuals that is basically healthy,” said Jennifer Johnston, PhD, lead author and staff scientist with the Clinical Genetics Section of the MGMGB. The research is part of ClinSeq, a large-scale, NIH research study that explores the fundamental medical, molecular and bioinformatic challenges facing individualized genome sequencing in a clinical research setting.
Once they identified participants with genomic mutations, researchers called them back to the clinic to give them a customized work-up. They called this method of looking at the person after looking at the genomic data “iterative phenotyping.”
Researchers sifted through more than 100,000 variants per participant – nearly all harmless – and studied only potentially harmful mutations that were found in about 100 of the 951 participants. Of those patients, 79 were followed up, and researchers confirmed that 34 had the specific condition linked to their genetic mutation. These findings indicate that 3 percent or more of the U.S. population might have a genetic condition compared to previous estimates of less than 0.02 percent.
“We achieved about a 50 percent accuracy of predicting disease in people not knowing anything about their health status beforehand,” Biesecker said. In other words, the researchers changed the odds of these patients having one of these diseases from something like 1 in 50,000 to 1 in 2 through iterative phenotyping.
Given this accuracy, Biesecker is upbeat about the future of genomic medicine. “These results show that you can dramatically improve your predictions based on genome sequence information.”
NHGRI Director Eric Green, MD, PhD, added, “Today, we tend to deliver medical care based on the expected response of the average patient, and yet we know that this is far from perfect. Eventually, we want to deliver medical care based on individual genomic differences that enable more precise ways to prevent and treat disease. These findings move us closer to that reality.”
The team also found having a mutation did not always lead to a condition that looked like a textbook case. Indeed, 20 of 79 of the participants with harmful mutations and an associated physical change didn’t know they might have a genetic condition. Sometimes these physical changes were so mild that the participant had neither sought a diagnosis nor reported them to the research team during enrollment in the study. While it might seem unnecessary to find these mild cases, a key aspect of genomics is that the same mutation can affect family members with different levels of severity.
“A couple of the participants with LDLR mutations thought they just had garden variety high cholesterol, when in fact they had familial hypercholesterolemia,” said Biesecker. This led to more aggressive cholesterol screening in other family members, including children as young as eight, because early treatment can delay heart attacks and prolong life.
