A small sample of a person’s genes may help doctors predict the risk of a common type of stroke in their patients, reports a preliminary study suggesting that this goal is within reach.
Through a technique that assesses multiple genetic variations at once, HMS researchers achieved 86 percent accuracy in predicting the people who suffered an artery-blocking stroke compared with those who did not. In this case, the computer generated the likely genetic signature that added up to the cardioembolic stroke suffered by 146 people treated in the Emergency Department or neurology clinic at Massachusetts General Hospital, compared with 423 other white adults who also were evaluated there.
“We did this study to personalize the prediction of stroke and to provide immediate clinical value to a genetic analysis,” said senior author Marco Ramoni, an HMS associate professor of pediatrics at Children’s Hospital Boston and member of the Partners Center for Personalized Genetic Medicine.
One of the big challenges in genomics research is developing clinically useful risk predictions for individuals. So far, the multiple genes rooted out by recent large gene association studies have been most valuable to scientists for revealing biological underpinnings of disease and potential new drug targets. Contrary to popular perception, the genes implicated in most common genetically complex diseases do not yet help doctors predict an individual’s risk of disease or inform consumers about the meaning of their personal genomes.
For their study, Ramoni and his co-authors used a statistical technique known as Bayesian networks to model the interactions of genes likely to underlie a phenotype. Starting with 1,313 genetic markers associated with stroke, the analysis identified 37 variations in 20 gene-coding DNA regions that, together with race, provided the most accurate risk estimate of stroke. The method can analyze larger numbers of interacting genes in smaller populations than gene association studies, Ramoni said.
The findings need to be validated in an independent population with more racial diversity, the authors said. If funding permits, the next phase may include assessing 1 million or more genetic markers plus environmental factors, such as diet and exercise, said first author Rachel Badovinac Ramoni, an HMS instructor in developmental biology at the Dental School.
Four years ago, Marco Ramoni and his colleagues used the same approach to predict stroke with 98 percent accuracy in people with sickle cell anemia, a genetically complex side effect in children and teenagers whose primary disease is caused by a change in one gene.
The researchers hope to develop a simple software tool for doctors to evaluate the genetic probability of stroke and other complex diseases that affect larger portions of the population. Those at high risk could be targeted for early intervention to prevent the disease or lower the risk, said Rachel Ramoni.
Students may contact Marco Ramoni at marco_ramoni@hms.harvard.edu for more information.
Conflict Disclosure: The authors declare no conflicts of interest.
Funding Sources: The National Human Genome Research Institute, National Institute of Dental and Craniofacial Research, National Institute of Neurological Diseases and Stroke, the Mallinckrodt General Clinical Research Center at Massachusetts General Hospital, National Center for Research Resources, and the Deane Institute of Integrative Research in Stroke and Atrial Fibrillation. DNA panels from the National Institute of Neurological Diseases and Stroke Human Genetics Resource Center DNA and Cell Line Repository (http://ccr.coriell.org/ninds) were used in this study.
