Researchers Find Better Way to Predict Stroke Risk in Sickle Cell Anemia Patients
FOR IMMEDIATE RELEASE:
March 20, 2005
For Further Information:
BUSPH Contact: Jodi Petrie, 617-638-5432, firstname.lastname@example.org
Children's Hospital Boston Contact: Bess Andrews, 617-355-5332, Elizabeth.Andrews@childrens.harvard.edu
BUSM/BMC Contact: Gina DiGravio, 617-638-8491, email@example.com
Stroke is a major vascular complication of sickle cell anemia, especially in patients under the age of 20. Currently, the chances of stroke for a child with sickle cell anemia are estimated through Transcranial Doppler (TCD) flow studies, a method of estimating blood flow to the brain. However, only 10 percent of individuals with abnormal TCD will have a stroke in the first year of observation and about 20 percent of individuals with normal TCD will have a stroke.
Researchers from Children's Hospital Boston, Boston University School of Public Health (BUSPH), Boston University School of Medicine (BUSM), Boston Medical Center (BMC) and Harvard Medical School have developed a novel approach to predict the risk of stroke in these patients using their genetic variations. This approach, reported in the March 20, 2005 issue of Nature Genetics, is based on a method called Bayesian networks, and is able to examine the complex structure of associations between a disease complication and genetic variations to produce accurate predictive models.
Bayesian networks, a data mining method born at the confluence of Artificial Intelligence and statistics, are used today in a variety of applications, ranging from spam email filtering to robotics. This study showed that these networks allow the integration of clinical information with the genetic profile of a patient to quantify his or her risk of developing a disease.
Using this method, the researchers analyzed the genetic differences in genes that might protect from or increase the chance of stroke in sickle cell anemia. The analysis showed that genetic differences in 12 genes interact with fetal hemoglobin level, another prognostic indicator in sickle cell anemia, to modulate the risk of stroke.
The model includes one gene already associated with stroke in the general population and the researchers theorized that the presence in the model of genes already associated with stroke suggests that some genetic factors predisposing patients to stroke are shared by both sickle cell anemia patients and stroke victims in the general population.
To test their conclusions, the researchers validated their results in an independent population by predicting the occurrence of stroke in a set of 114 subjects not included in the original study: seven stroke patients and 107 control patients, a proportion consistent with the original study group. Their model predicted the correct outcome for all seven stroke subjects, and for 105 of 107 non-stroke patients, with an overall predictive accuracy of 98.2 percent.
''This is an exciting step forward for researchers. Our results show the promise of this novel approach for the discovery, representation, and prognostic use of the genetic basis of complex traits,'' said author Paola Sebastiani, PhD, an associate professor at BUSPH and lead investigator of the study.
According to Martin Steinberg, MD, director Center of Excellence in Sickle Cell Disease at BMC and professor of medicine at BUSM, the ultimate outcome of sickle cell anemia is likely to be determined by the actions of many genes that all modify the abnormalities triggered by the sickle cell mutation. ''Our method should have many applications as we are better able to examine the genetic diversity among patients with sickle cell disease,'' he added
Marco Ramoni, PhD, a researcher in the Children's Hospital Informatics Program and Harvard Medical School, who co-led the study with Sebastiani, says ''This is an important contribution of the 'Bayesian revolution in genetics' to one of the most critical problems of genetic analysis, and one more example of what can be accomplished by a Bayesian approach - and by Bayesian networks in particular - and that cannot be achieved otherwise.''
Children's Hospital Boston is home to the world's largest research enterprise based at a pediatric medical center, where its discoveries have benefited both children and adults for 136 years. More than 500 scientists, including eight members of the National Academy of Sciences, nine members of the Institute of Medicine and ten members of the Howard Hughes Medical Institute comprise Children's research community. Founded in 1869 as a 20-bed hospital for children, Children's Hospital Boston today is a 325-bed comprehensive center for pediatric and adolescent health care grounded in the values of excellence in patient care and sensitivity to the complex needs and diversity of children and families. Children's also is the primary pediatric teaching affiliate of Harvard Medical School. For more information about the hospital visit: http://www.childrenshospital.org.
Established in 1873, Boston University School of Medicine is a leading academic and research institution, with an enrollment of nearly 630 students and more than 1,100 full and part-time faculty members. It is known for its programs in arthritis, cardiovascular disease, cancer, human genetics, pulmonary disease and dermatology, among others, and is one of the major biomedical research institutions in the United States. The School is affiliated with Boston Medical Center, its principal teaching hospital, and Boston Veterans Administration Medical Center. Along with Boston Medical Center and 15 community health centers, the School of Medicine is a partner in Boston HealthNet.
Boston Medical Center is a private, not-for-profit, academic medical center affiliated with Boston University School of Medicine. Committed to providing high-quality health care to all, the hospital offers a full spectrum of pediatric and adult care services including primary and family medicine and advanced specialty care with an emphasis on community-based care. Boston Medical Center offers specialized care for complex health problems and is a leading research institution. Boston Medical Center and Boston University School of Medicine are partners in the Boston HealthNet--15 community health centers focused on providing exceptional health care to residents of Boston. For more information, please go to www.bmc.org.