Clinical trajectories are highly variable among patients with Alzheimer’s disease at all stages, even when those patients have been screened for amyloid positivity. This has presented a challenge in clinical trials, diluting statistical power and preventing balanced comparisons of placebo and treatment. Rates of worsening are even more difficult to predict, and fewer subjects decline rapidly, with earlier stages of disease.
ADMdx has developed pattern-recognition based classifiers, using FDG PET, structural MRI, and other imaging modalities, that objectively quantify a patient’s neurodegenerative status. The numeric score assigned to the patient’s scan by the classifier reflects that subject’s expression of disease-related changes, and is predictive of the subsequent rate of clinical worsening. The classifiers can be used for:
- Accurate staging of patients
- Prediction of clinical trajectory for enrichment
- Detection of asymptomatic subjects who will convert to MCI
- Pre-defined endpoints that have increased statistical power over other imaging measures
Pattern associated with progression toward Alzheimer’s dementia. Blue = hypometabolism, Red = preservation relative to whole brain. The greater the numeric score assigned to the scan, the greater the pattern expression. Plot shows mean scores (bars = SEM) of over 570 independent test subjects blindly scored by the classifier. Note the cascade like progression from Normal amyloid negative (NL Am-) status to Early MCI and progressive stages of late MCI, to Alzheimer’s (AD) dementia.