The following is a transcript from the interview recorded in the Section of Neuroradiology at the Medical College of Wisconsin at Milwaukee.
Participants included:
- Dr. John Ulmer, Professor of Radiology and Section Chief of Neuroradiology
- Dr. Gosia Franczak, Professor of Neurology and Director of Memory Disorders Clinic
- Laura Umfleet, Assistant Professor of Neuropsychology and an expert in memory disorders
- Dr. Mohit Agarwal, Assistant Professor in Neuroradiology
Agarwal — Hello and welcome to this edition of the AJNR podcast. This is your host Mohit Agarwal, Asst Prof of Neuroradiology at the Medical College of Wisconsin at Milwaukee and I am joined today by our section chief Prof John Ulmer, by our neurologist Prof Gosia Franczak, who is Director of the Memory Disorders clinic at MCW and our Neuropsychologist Laura Umfleet, Asst Prof in Neuropsychology and an expert in memory disorders. Together, we have developed a team here at MCW where we are taking a multi-prong approach to the diagnosis, follow up and management of patients coming in cognitive decline, memory problems and other issues.
The world of dementia and cognitive disorders has changed a lot in the past few years and is still rapidly changing with the development and incorporation of biomarkers in the diagnosis. A lot of centers across the country are using PET for the diagnosis of these patients. We, at MCW however have been using MRI, especially volumetric MRI as one of the main tools in the management of these patients and that will be our primary focus today. I want to start this discussion today with our neurologist Dr. Franczak and I want to ask her about her perspective of how this field has changed in the past years and how her approach has changed in recent times.
Franczak — Alzheimer’s Disease (AD) is a neurodegenerative disease and the most common cause of dementia. AD is the sixth-leading cause of death in the US. In the US, an estimated 5.7 million people pf all ages have Alzheimer’s Disease. By 2025, the number of Americans age 65 and older with AD is estimated to reach 7.1 million. In Alzheimer’s Disease two main proteins deposit in the brain beta amyloid and tau protein. Following numerous drug trial failures, researchers concluded that anti-amyloid therapies have proven disappointing because patients receive them too late. Had patients obtained these therapies earlier, in the years preceding irreversible brain damage, perhaps the disease could have been prevented.
We do know that there is are long preclinical and prodromal (Mild Cognitive Impairment) phases of AD and other dementias. This period of 10-20 years provides a critical opportunity for potential intervention with disease-modifying therapy. Determining the pathology underlying the presentation of MCI can be challenging for clinicians. Patients with MCI may progress to AD, but also to other types of dementia such as vascular dementia or Lewy Body Dementia etc. Establishing the cause of MCI is also very critical in initiating proper therapeutic strategies.
Current imaging biomarkers including structural (MRI, CT) or functional (fMRI, FDG- PET and Amyloid PET) are either not specific enough for detecting early changes (CT or MR) or very expensive and not widely available. Thus, there is an urgent need for biomarkers which can identify patients in the preclinical and prodromal phases of AD and other dementias.
Here at MCW we are excited to use multimodal MRI techniques including volumetrics as the potential biomarkers for the early detection and monitoring of people suspected to have MCI and dementia.
Agarwal — That is very interesting. So, you mentioned MRI being highly useful to you, I want to turn our discussion to Dr. Ulmer, who recently delivered a talk at RSNA on this topic, to tell us more about the MRI biomarkers and how we can use these to help our neurocognitive disorders patients.
Ulmer — Volumetric MRI: We acquire 3D SPGR images on each patient and use FDA cleared automated segmentation software to segment 43 brain substructures providing Z-scores and percentiles compared to age matched normative data. Structural biomarkers mildly precede the onset of cognitive deficits. Patterns of volume loss reflects affected networks, and so it has the potential to serve as a structural correlate to network dysfunction. In effect, concurrence between the volumetric profiles and the cognitive profiles can support specific neurodegenerative diagnoses. We may even differentiate disease from normal aging in patients with subjective cognitive impairment, where no cognitive deficit can be detected by Neuropsych testing.
Agarwal — So, Dr. Ulmer, besides volumetric MRI, how do you think the other biomarkers like ASL and DTI metrics can contribute to the diagnosis?
Ulmer — We collect mean diffusivity data to assess white matter microstructure. This currently has 2 practical applications: First, we are finding that MD maps reveal a greater extent of chronic cerebral microangiopathic disease. This is important given that microvascular disease is a common confounding and possibly predisposing variable in the diagnosis of AD and DLB and may cause cognitive impairment in its own right. Secondly, increased MD emanating from the cortical-white matter interface reflects Wallerian degeneration associated with cortical degeneration and can serve to strengthen the diagnosis of neurodegeneration of a particular region.
We also collect ASL data, though the reproducibility across individuals is hampered by individual variation in cardiovascular hemodynamics in the elderly patients. We are in the process of implement time-resolved ASL, which should increase the power of this sensitive biomarker. Currently though, ASL helps us in about 20% of patients by the diagnosis of neurodegeneration.
Agarwal — Exciting times for imaging in dementia diagnosis. It looks like the role of Neuroradiologists have entirely changed in the present times. Traditionally, the role of the neuroradiologist was just to exclude a treatable cause of dementia but it looks like now they have a lot to contribute
Ulmer — Role of the radiologist: At this point, our role is to support and strengthen a clinical diagnosis. In time, when as a field have gained more experience and understanding, we may take a more substantial role in making early diagnoses. Currently, most cases provide correlative imaging biomarker patterns to the clinical presentation and thereby increase certainty of a diagnosis at an earlier stage than might otherwise be the case. I anticipate that through greater understanding of network patterns, perhaps with the aid of AI, we may actually begin to help define diagnoses.
Agarwal — we also have with us today, our neuropsychologist Laura Umfleet, who is part of our team and as most of us know, neuropsychological evaluation plays a crucial role in dementia diagnosis. So, Laura, what is your take on this?
Umfleet — I think advances in MRI and other biomarkers are essential to move us forward in early detection of neurodegenerative disorders. As a neuropsychologist, I specialize in understanding the relationship between the brain and behavior which relies on standardized cognitive testing combined with integration of patient history and available biomarkers—typically MRI. As Dr. Ulmer discussed, we are seeing much potential in the use of cutting-edge MRI techniques, and it is hoped that these techniques will increase the predictive power of MRI in early detection of both AD and non-AD processes.
Agarwal — so, in this discussion, I have heard a lot about neurocognitive conditions other than AD. Since, you see a wide spectrum of cognitive profiles in your practice, I want to ask you how the approach to MCI patients is different at MCW with regards to differentiating AD from other conditions.
Umfleet — The diagnosis of MCI requires cognitive changes that are serious enough to be noticed but the changes are not severe enough to interfere with daily life or independent function. Impairment on testing must also be present. When determining objective cognitive impairment to support a diagnosis of MCI, neuropsychologists often use widely accepted criteria that is based on research such as Petersen’s criteria which requires scores to be at least 1.5 SD below the population mean to indicate abnormality.
There are 4 MCI subtypes (amnestic, single domain; amnestic multidoman; non-amnestic single domain, non-amnestic multidomain). aMCI requires impaired performance in delayed recall. If memory and one or more cognitive domains fall in the abnormal range compared to a standardization sample, multidomain aMCI is diagnosed. In naMCI, abnormalities in non-memory domains are required.
At MCW, we take a team approach to diagnosing and treating our patients. This involves multiple disciplines including neurology, neuroradiology, and neuropsychology. I want to provide an example to illustrate how we work together. MCW neurologists and neuropsychologists work side-by-side in our multidisciplinary memory disorders assessment program where each patient is examined by both disciplines. Patients with memory concerns with abnormal memory scores on formal testing are referred for neuroimaging to help clarify etiology of aMCI. Should neuroimaging show hippocampal atrophy, AD is the likely diagnosis and the neurologist may decide to start treatment. Both aMCI and naMCI can be due to a variety of etiologies, though the strongest association is between aMCI and AD. naMCI is a more heterogeneous group and can be due to behavioral variant FTD, language variant FTD, vascular disease, parkinsonism, etc. There is much work that needs to be done to detect non-AD conditions early on in the disease process. Our team is dedicated to better understanding biomarkers for early detection of AD and non-AD conditions.
Agarwal — Finally, as we close, I want to take a brief two-line summary or a take-home message from all three of you on the topic, lets start with Dr. Franczak.
Franczak — Integration of imaging biomarkers with clinical and neuropsychological data can be critical in the accurate diagnosis in patients with preclinical and prodromal phases of AD and other types of dementias. Improved understanding of preclinical and prodromal phase of AD and other dementias as well as the role of these biomarkers in the detection and tracking these stages may also have important implications for the development of effective treatments for AD and other types of dementia.
Umfleet — A multidisciplinary team approach to clinical research and patient care is essential for moving the field forward in early detection and prevention of neurodegenerative diseases.
Ulmer — I’d like to echo Gosia’s and Laura’s sentiments: The development, refinement, and translation of cognitive neuroimaging biomarkers is a rapidly evolving endeavor. We don’t know more than we do know about what these techniques can offer patients. It is important for neuroradiologists to get together with neurology and neuropsychology colleagues, so that each subspecialist can more fully understand the perspective and approach of the others. This allows the group to hit the target in clinical care. There is still craft and art in drawing inferences from imprecise and variable clinical and imaging data. Concurrence of data from all sources and consensus opinions are integral in securing a clinical diagnosis.
If you are interested in gearing up a cognitive neuroimaging service, my advice is to seek out champions in neurology and neuropsychology and begin discussions. From the neuroradiologist’s standpoint, this may require some additional learning about neurodegenerative diseases. Also, work closely with the PET folks in your department to foster integration of MRI and PET biomarkers.
Agarwal — Very exciting discussion! This brings us to the end of this podcast and I want to thank our listeners and in want to thank Wende Gibbs and AJNR for giving us space on this platform. Thank you!
Dr. Ulmer’s presentation on dementia from the 2018 RSNA Annual Meeting is part of the Virtual Meeting, which is available through April 30. The course number is RC105 and a link is here: https://meeting.rsna.org/program/index.cfm
