CAP Today has two interesting deep dive articles. One is on Alzheimer's disease, as FDA-cleared tau and amyloid tests gather real-world experience. Fujirebio and Roche are featured. The other is a deep dive article on the challenges of digital pathology. Paige AI, Fuji, are featured venders and experienced pathologists at several institutions are profiled.
CAP: ALZHEIMER TESTS
https://www.captodayonline.com/inside-labs-experience-with-assays-for-alzheimers/
CAP: DIGITAL PATHOLOGY UPDATE
https://www.captodayonline.com/digital-paths-star-rises-from-the-mists/
CONCISE SUMMARIES
ALZHEIMER
Mayo Clinic's experience with Alzheimer's disease assays in 2023 highlights significant advancements in diagnosis and biomarker utilization. Cerebrospinal fluid testing orders have surged since 2020, especially following the approvals of aducanumab and lecanemab. Dr. Joshua Bornhorst emphasized that FDA-approved Alzheimer's biomarkers demonstrate high precision, though challenges like preanalytical protocols and unresolved payment issues persist.
Novel aspects include the recent incorporation of blood-based biomarkers in proposed diagnostic criteria by the National Institute on Aging and the Alzheimer's Association, expanding diagnostic possibilities. Biomarkers such as Aβ42, p-tau181, and NfL are all playing crucial roles in improving Alzheimer's diagnosis, with different biomarker profiles aiding in identifying disease stages. Mayo Clinic offers a range of CSF biomarker tests, including Roche's and Fujirebio's FDA-cleared assays, with ongoing refinements to enhance accuracy and diagnostic capabilities.
###
- SIDEBAR. Other current ALZ news include Quanterix introducing a lab test based on J&J pTau217 assets (here, here), and Quest licensing a Diadem Alzheimer blood test based, more unusually, on p53 (here, here). Diadem's AlzoSure has FDA breakthrough status from January 2022, but it sounds like it would be revalidated as a Quest LDT under the license.
- SIDEBAR. I've heard since August that CMS would release some sort of guidance to MACs about Alzheimer tests (related to the newly released Alzheimer drugs), but I haven't heard of anything concrete.
DIGITAL PATHOLOGY
The article discusses the current state and future prospects of digital pathology, highlighting the slow adoption of this technology in clinical practice in the United States. Despite the initial predictions of rapid adoption, traditional glass slides remain the primary method for diagnosing cancer in the U.S. Only a small percentage of hospitals have fully embraced digital pathology.
The articles also note some positive developments in the field, such as the FDA approval of the first product combining digital pathology with artificial intelligence (AI) for cancer diagnosis. Companies like Paige and Fujifilm are investing in digital pathology and AI technologies, signaling a growing interest in the clinical market.
- TIP: FDA just updated the public on the rapid pace of its AI approvals of software - here.
Partnerships among digital pathology companies are increasing, facilitating interoperability and streamlined workflows between scanners, image management software, and AI tools. DeciBio Consulting estimates the global clinical digital pathology market to be worth around $270 million in 2023, with significant growth expected.
However, challenges remain, including the lack of clear reimbursement models for digital pathology and concerns about job displacement among pathologists. Some pathologists argue that certain aspects of diagnosis, especially those requiring nuanced interpretation, are irreplaceable by machines.
- SIDEBAR: Digital Radiology received a final push when CMS began to cut payments for non-digital radiology (noted in CAP Today).
Despite these challenges, experts believe that digital pathology, when used as a tool alongside AI, can enhance the efficiency and accuracy of pathology practices. Digital pathology is seen as the next frontier in digitization within the medical field, offering the potential to improve patient care while allowing pathologists to focus on more complex cases.
AI corner.
Summaries, AI assisted (GPT3.5).
Halloween Special, Chat GPT rewrites an Edgar Allen Poe story here.
