Simulated brain biopsy for diagnosing neurodegeneration using autopsy-confirmed cases.

Risks associated with brain biopsy limit availability of tissues and the role of brain biopsy in diagnosing neurodegeneration is unclear. We developed a simulated brain biopsy paradigm to comprehensively evaluate potential accuracy of detecting neurodegeneration in biopsies. Postmortem tissue from the frontal, temporal and parietal cortices and basal ganglia from 73 cases including Alzheimer's disease (AD), Lewy body disease (LBD), frontotemporal lobar degeneration-TDP43 (FTLD-TDP), multiple system atrophy (MSA), Pick's disease (PiD), corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP) were evaluated using H&E and immunostains. Brain biopsy was simulated in a blinded manner by masking each slide with opaque tape except for an area measuring 10 mm in diameter. Diagnoses obtained from frontal cortex only or all 4-brain regions were then compared with autopsy diagnoses. Diagnostic sensitivity in frontal cortex was highest in FTLD-TDP (88%), AD (80%) and LBD (79%); intermediate for MSA (71%), CBD (66%) and PiD (66%) and lowest for PSP (0%) (average 64%). Specificity was 43%. Sensitivities were enhanced with all 4-brain regions: FTLD-TDP (100%), AD (80%), LBD (100%), MSA (100%), CBD (83%), PiD (100%) and PSP (88%) (average 92%). Specificity was 71%. Simulated brain biopsy addressed limitations of standard brain biopsies such as tissue availability and lack of autopsy confirmation of diagnoses. These data could inform efforts to establish criteria for biopsy diagnosis of neurodegenerative disorders to guide care of individuals who undergo biopsy for enigmatic causes of cognitive impairment or when evidence of an underlying neurodegenerative disease may influence future therapy.