A new paradigm for neurodegenerative diseases classification: A clinical perspective

A vast progress has been made in the understanding of neurodegenerative diseases during the past few years. However, clinical diagnostic accuracy continues to be very low, despite the introduction of various diagnostic tools and repeated revisions of diagnostic criteria. For instance, patients with Alzheimer's disease (AD) may present with symptoms that overlap with other neurodegenerative conditions like dementia with Lewy bodies (DLB), making accurate diagnosis challenging. This diagnostic uncertainty can lead to delayed or incorrect treatment, significantly impacting patients' quality of life and prognosis. Thus, the definite diagnosis still relies on post-mortem pathological findings, placing a significant burden on both clinicians and researchers. As a growing body of evidence indicates, co-pathology seems to be the rule among neurodegenerative diseases. Additionally, a single pathological diagnosis, such as AD, can manifest in various clinical presentations, ranging from predominantly cognitive impairment to significant motor symptoms. Each of these presentations currently requires its own set of complicated diagnostic criteria. Perhaps, the time has come for a much-needed radical revision of existing clinical diagnostic criteria. Inclusion of patients do not neatly fit into existing diagnostic categories for neurodegenerative diseases, in future large-scale, longitudinal studies and/or clinical trials, and systematic assessment of their clinical features and disease progression using machine learning could generate valuable data on patients with mixed pathologies and improve our understanding of how to effectively treat these complex cases.