Are Radiology Enterprises Prepared to Meet the Growing Demand for Alzheimer Disease Imaging?
Editor’s Note: Moderated by Lawrence Tanenbaum, MD, FACR, Applied Radiology hosted a series of conversations with experts on Alzheimer disease imaging and management at the 2024 Radiological Society of North America Annual Meeting and Scientific Exhibition.
In this discussion, Dr Tanenbaum, Tammie L.S. Benzinger, MD, PhD, a professor of radiology who serves as chief of MRI service at Mallinckrodt Institute of Radiology at Washington University School of Medicine in St. Louis, Missouri, and Saurabh Sharma, MBA, a senior director for MR Strategy at Siemens Healthineers, explore whether imaging enterprises are prepared for the expected growth in the demand for brain MR imaging in patients with Alzheimer disease.
The surge in development of anti-amyloid immunotherapies for Alzheimer disease (AD) in recent years has increased the volume of MR imaging performed at the Mallinckrodt Institute of Radiology in St. Louis, Missouri, says Tammie Benzinger, MD, a professor of radiology and chief of MRI service at the facility.
“For the first 180 patients that started the therapy, we generated a thousand additional brain MRI visits,” says Dr Benzinger. Mallinckrodt draws patients from a 300-mile radius; she anticipates volume will reach some 15,000 patients, generating an immense need for optimized imaging operations. Mallinckrodt is not alone. Thanks to the growing demand for Alzheimer disease immunotherapy, radiology departments and clinics across the US are experiencing concurrent demand for brain MR imaging.
There's 6-7 million Americans that suffer from Alzheimer disease today and the number is expected to double to 13 million by 2050. That underscores the need for early detection and treatment,” says Saurabh Sharma MBA, senior director for MR Strategy at Siemens Healthineers. Given the growing population of AD patients eligible for immunotherapy, Dr Benzinger says, there is no question that healthcare enterprises urgently need to scale up their imaging capacity. Moreover, the challenge lies not just in accommodating the sheer volume of imaging, but also in doing so in a way that puts the patient’s quality of experience first.
Indeed, Dr Benzinger emphasizes the importance of delivering a quality exam tailored to the unique requirements of Alzheimer disease patients, as well as a timely report focused on the key issues of eligibility and surveillance with respect to immunotherapy.
Using Artificial Intelligence to Enhance Reproducibility and Efficiency
Manufacturers like Siemens Healthineers are stepping up to the challenge with artificial intelligence (AI), says Mr Sharma. Deep learning, for example, can be used to reduce scan time, a critical need for AD patients, who must undergo multiple MRI studies during their treatment journey. He also emphasizes the importance of making scans more consistent and reproducible to facilitate longitudinal analysis.
“It's important that scans are consistent and fast, and the way we are doing it is leveraging a lot of artificial intelligence. So we're using tools such as automated patient positioning, automated brain slice positioning to make sure there's reproducibility. And then of course on the fast scan side, we're using a lot of our deep learning reconstruction technology to make them faster,” he says.
Standardized Protocols for Consistency
Advances like these are important to monitor for amyloid-related imaging abnormalities (ARIA), a potential side effect of anti-amyloid therapy. ARIA manifests as vasogenic edema (ARIA-E) or micro-hemorrhages (ARIA-H), signs that require precise and repeatable imaging to track their response to treatment. Even minor variability can adversely affect the detection of ARIA, says Dr Tanenbaum.
Experience [has demonstrated] that left to our own devices, we really won't get it quite right,” he says, noting that that Siemens and other companies are collaborating with the American Society of Neuroradiology (ASNR) to develop and supply downloadable, standardized protocols. Standardization ensures that even subtle findings such as micro-hemorrhages can be detected regardless of scanner field strength or platform.
“We had different working groups at different sites running test protocols across real patients to identify and make sure that the same minute micro hemorrhage findings could be seen between the SWI and the GRE between the 3 T and the 1.5 T, and that if you hopped between Siemens and GE, you'd get something as close as possible to being the same,” Dr Benzinger says. “The protocols that came out are about 10 minutes long, so easy to run, so easy to install, so easy to read because you don't have to look at a thousand other images as well.”
Addressing Workflow and Reporting Challenges
It’s worth noting that some challenges persist in ARIA detection; for example, in reporting findings. Dr Benzinger notes that while radiologists can identify ARIA, their reports are not always structured in a way that aligns with clinical decision-making criteria at the treatment provider level.
“In these clinics, the nurse practitioners, the physician assistants, they're the ones taking all the patient phone calls, ordering the scans, calling the infusion center to go, no go. We need to give them a structured report that spells out very clearly what the findings are on the exam so radiologists aren't missing the findings,” she says.
Failing to categorize ARIA severity or omitting comparisons with prior exams, in particular, can delay treatment. Dr Tanenbaum echoes this concern and believes AI has the potential to improve reporting capabilities.
“There’s clearly a need for education and possibly AI assistance in reading these exams,” he says, observing that AI tools that highlight and quantify ARIA findings can enhance accuracy and consistency, especially when radiologists are managing high caseloads.
Genetic Screening and Imaging Integration
Screening for APOE4, a genetic variant and known risk factor for ARIA, is becoming more common in clinical practice. Dr. Benzinger reported that insurance coverage for APOE genotyping has improved, facilitating more informed risk discussions. “The APOE screening guides the risk discussion that neurologists have with patients,” she explained, noting that some institutions restrict therapy for APOE4 homozygous patients.
The integration of PET with MRI (PET-MR) was another topic of interest. Given the demands on patients and their families for frequent visits, PET-MR offers a more patient-centered approach by combining both modalities in a single session. “We’re asking a lot from patients,” Dr. Benzinger noted. “PET-MR allows us to get comprehensive information in one visit, making it easier for patients to access therapy.”
Innovations in Accessibility and Low-Field MRI
Accessibility remains a challenge, particularly for patients in remote or underserved areas. Mid-field MRI systems, such as Siemens’ MAGNETOM Free.Max offer a potential solution. These systems are easier to install and more comfortable for patients. While lower field strength may reduce sensitivity to micro-hemorrhages, Dr. Benzinger suggested that ARIA-E, which is more clinically relevant, can still be effectively monitored with these systems.
“In a situation where you've got a really good relationship with the patient and you're confident that they're asymptomatic, you could feel comfortable using ultra-low field to look for moderate ARIA-E,” she explained. Dr Benzinger shared her positive experience with the Siemens MAGNETOM Free.Max, noting that in a year and a half of experience, she did not find any cases of missed micro-hemorrhages with the protocols she was using.
Ongoing Dialogue, Shared Learning Needed
Alzheimer disease imaging presents opportunities and challenges. To meet the rising demand, enterprises must adopt AI-driven workflows, standardized protocols, and patient-centered innovations like PET-MRI and lower-field MRI systems. Collaboration amongst radiologists, manufacturers, and providers will be key to delivering efficient and accurate care.
“The formula is straightforward but navigating the gray areas requires ongoing dialogue and shared learning,” Dr Tanenbaum says. “This collaborative approach will ensure that imaging enterprises are prepared for the evolving landscape of Alzheimer disease management.”