Michael Donohue, PhD
CTRI Supports UCSD Biostatistician Donohue in Alzheimer's Disease Research
June 17, 2014 – Numbers. Patterns. Models. With support from UC San Diego Clinical and Translational Research Institute (CTRI), Michael Donohue, PhD, is using mathematics to help untangle the mysteries of Alzheimer’s disease.
Donohue, an assistant professor at UC San Diego’s Division of Biostatistics and Bioinformatics, received CTRI’s KL2 Post-doctoral Scholar Award to develop models to improve Alzheimer’s disease clinical trials. He investigated efficient analysis methods for longitudinal clinical trials, particularly in the face of missing data typical in Alzheimer’s disease trials. Additional Alzheimer’s disease research under the three-year grant focused on the amyloid cascade hypothesis and transforming short-term snapshots of disease progression into longer term pictures, as well as justifying a new outcome measure for very early Alzheimer’s.
Findings from his research are published in two papers this spring. The first, “Estimating long-term multivariate progression from short-term data,” was published in March 2014 in the journal, Alzheimer’s & Dementia. The second paper, “The Preclinical Alzheimer Cognitive Composite: Measuring Amyloid-Related Decline,” was published in the Journal of the American Medical Association (JAMA) Neurology in early June. A related editorial in JAMA Neurology, “Secondary Prevention Trials in Alzheimer Disease: The Challenge of Identifying a Meaningful Endpoint,” by Richard J. Kryscio, PhD, congratulated Donohue and the team for “conducting a careful and detailed retrospective data analysis.”
Estimating long-term progression from short-term data
Alzheimer’s disease, a neurodegenerative disorder that affects 4.5 million people in the U.S., is characterized by the accumulation of amyloid plaques and neurofibrillary tangles in the brain. These plaques and tangles, formed from irregular proteins, damage neurons and block signaling between the nerve cells.
Through Donohue’s initial research under the KL2, he performed a quantitative review applying competing models to data from five clinical trials conducted by the Alzheimer’s Disease Cooperative Study (ADCS) at UC San Diego. This work was published in The Journal of Nutrition, Health & Aging in November 2012.
“The KL2 application had an emphasis on missingness in clinical trials,” Donohue said. Missingness – or missing data – plagues Alzheimer’s disease studies and could be associated with the decline of the human subjects enrolled in the trials. “As the disease progresses, it becomes more difficult for the subject to come in for assessments. Some people are lost to follow up,” he said.
The concern is that missing data can invalidate the results of a clinical trial. Donohue combed the ADCS studies that began during the early 2000s, analyzing data to assess the impact of missingness. “We actually found that the typical models we’re using appear to be fairly robust to the missingness we see in the trials,” Donohue said.
Under the KL2 award, he next studied the amyloid cascade hypothesis, which suggests that Alzheimer’s disease begins with amyloid accumulation long before neuron death occurs in the brain and symptoms appear. “It’s been depicted as these successive sigmoid curves – a succession of abnormalities as the disease progresses from normal cognition to dementia,” Donohue said. Researchers applied data to try to estimate these curves, but methodological challenges exist because the disease is believed to evolve over 20 or 30 years and subjects have been followed with novel disease markers for at most eight years, according to the biostatistician. “We haven’t been able to follow volunteers for that long because these markers are relatively new,” he explained.
Through simulations, Donohue and his team demonstrated that the method used could actually recover long-term disease trends in Alzheimer’s from short-term observations. “We were able to discern the long-term picture from short-term snapshots,” he said.
Additional research under the KL2 focused on measuring the cognitive performance of participants in the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s study, a clinical study for older people who have normal thinking and memory function but may be at risk for developing Alzheimer’s disease. “The question is whether we can measure the effectiveness of a treatment, especially in a population that doesn’t yet exhibit symptoms,” Donohue said. “We found, based on natural history studies, that we could see some decline in people who had amyloid relative to people who didn’t. That difference in decline suggests that we might be able to detect a treatment effect with the proposed outcome measure.”
Donohue received a bachelor’s degree in mathematics from Georgetown University in 1999 and a PhD in mathematics from UC San Diego in 2005 before joining UC San Diego. While in graduate school he realized pure math would not be enough to keep him motivated. “I needed something tangible, so that’s how I ended up in biostatistics,” said Donohue, whose experience includes serving on CTRI’s Biostatistics team. “I had a desire to work on relevant problems.”
His present focus is on design, monitoring, and analysis of Alzheimer’s disease clinical trials and observational studies, working closely with ADCS Director Paul S. Aisen, MD.
This month, Donohue concludes his work under the KL2 grant, which was awarded in 2011. “It (the KL2) has been immensely valuable in giving me the opportunity to focus on problems that I believe are important,” Donohue said.
Written by Patti Wieser
About UC San Diego Altman Clinical and Translational Research Institute:
UC San Diego Altman Clinical and Translational Research Institute (ACTRI) is part of a national Clinical and Translational Science Award consortium, led by the National Institutes of Health National Center for Advancing Translational Science. Established in 2010, ACTRI provides infrastructure and support for basic, translational and clinical research throughout the San Diego region to bring discoveries from the laboratory to the bedside, and facilitates training and education of the next generation of researchers. ACTRI carries out its activities in collaboration with institutional and corporate partners and currently has more than 1,400 members.