One of the challenges of Alzheimer's Disease is the presence of co-pathologies; this means patients with AD tend to have other neurodegenerative pathologies present in their brains in addition to tau. To address this, researchers at Penn, in collaboration with researchers at University of Castilla-La Mancha, Spain are working to improve imaging-based biomarkers for AD by better understanding the spread of tau and its specific effects on neurodegeneration using postmortem, or ex vivo, MRI and histological imaging.
Using these unique imaging techniques, Sadhana Ravikumar, PhD, a postdoc working in the Penn Image and Computing Science Lab, and her team are able to look directly at the underlying pathology in patients to understand what co-morbid pathologies they have, and link this information to structural changes they are seeing with MRI. Much of their work is developing tools and methods for analyzing and linking 3D postmortem MRI and histology images, and doing further analyses to identify which regions of the brain are most tied to tau pathology in Alzheimer’s disease versus other neurodegenerative pathologies.
While the lab also does whole hemisphere high resolution postmortem imaging, Ravikumar’s work focuses mostly on MRI and histology imaging of the medial temporal lobe (MTL) – the earliest cortical region affected by Alzheimer’s disease. Through this method she is able to visualize the MTL anatomy and pathology in great detail and describe for the first time, differences in the 3-D topography of tau pathology between early and late Braak stages.
In this particular research project, Ravikumar and her team are studying 25 subjects postmortem, which is a relatively large number in this type of research. To handle this data, Ravikumar developed a computational atlas of the MTL to combine the data from the 25 subjects and perform group comparisons.
What she is finding is that a lot of the work points to the entorhinal cortex (EC) as being affected very early on within the MTL, as well as high levels of pathology very early on in certain subfields in the hippocampus such as the CA1 subfield. “We also found that we see a lot more pathology in the anterior MTL than the posterior MTL in the early stages, so it is kind of pinpointing the trajectory of tau pathology accumulation across different Braak stages,” she explains.
“Since this is all done post mortem, the hope is to translate our findings into an in vivo space where we’re actually applying it to clinical data.”
Ravikumar’s hope is that identifying these regions that are affected early on in the disease will inform biomarker development and help pinpoint what regions can be used as biomarkers within the clinical trial framework. For example, when looking at clinical trials in patients where you want to track if a drug is working or not, you can evaluate the reduction in neurodegeneration in these specific regions that you know are affected by tau pathology in AD.
“Some studies have already started using some of the regions that we’ve pinpointed through these ex vivo analyses to inform other in vivo analyses,” she said. “Since this is all done post mortem, the hope is to translate our findings into an in vivo space where we’re actually applying it to clinical data.”
This research project, Ex vivo MRI atlas of the human medial temporal lobe: characterizing neurodegeneration due to tau pathology, was published in Acta Neuropathologica Communications (2021) and presented as a Featured Research Session by Ravikumar at this year’s Alzheimer’s Association International Conference (AAIC) 2023.