Conference Coverage
Identifying these hazard ratios is important to understanding the general risk of Alzheimer’s disease progression, based on these measures. Despite evidence of global trends that were useful, however, no single measure demonstrated enough accuracy to predict MCI in specific individuals. “What we really want to know is who, on an individual basis, is likely to be at risk,” Dr. Albert said, because “those are the people that we want to treat and track for the impact of medication.”The researchers therefore searched for combinations of domains that could be used on a large scale in the general population and that would be highly sensitive and specific for the development of MCI. The investigators wanted to identify the best measures from each of the domains using time-dependent receiver operating characteristic. They conducted this analysis from several perspectives, evaluating issues such as invasiveness and cost. “The best combination is the one that uses the least number of measures and seems to be the most predictive,” she explained, adding that the optimal biomarker would need to have both a sensitivity and specificity of at least 0.80.
Although no individual measure was able to achieve this threshold, the combination of multiple domains significantly enhanced both the sensitivity and specificity. The combination of the best measures from each domain—a genetic variable (APOE-4), cognitive variables (DSS and PA Immediate Recall), a CSF variable (p-tau), and MRI variables (right EC thickness and right hippocampal volume)—had a sensitivity of 0.80 and a specificity of 0.75 and an area under the curve of 0.85 in relation to the baseline, with respect to the progression of symptoms five years after enrollment in the study, Dr. Albert reported. “This is obviously approaching [the optimum], and we are quite pleased with this result,” she said.
These results would need to be replicated to validate this or other combinations for use in individual cases, Dr. Albert said. Researchers are developing a consortium consisting of five centers around the world that will be combining their data together to make diagnosis and treatment of Alzheimer’s disease in its earliest phases possible. Future goals are to identify less-invasive testing tools and to improve the accuracy of prediction.
In 2014, the BIOCARD study received new funding to continue follow-up and collections of CSF, MRI, and amyloid imaging from patients in the database. “I think it’s clear that biomarkers are critical for early identification,” Dr. Albert said. “We are hopeful [that] this approach will be useful to selecting patients for future clinical trials.”
—Linda Peckel