OS02.5.A Alzheimer-type neuropathological changes in glioblastoma-adjacent cortex

Abstract

Abstract Background Glioblastoma (GBM) is an aggressive type of brain cancer that is prevalent and fatal in the elderly. Age is not only the most common risk factor for brain cancer but also for neurodegenerative diseases, and previous studies have indicated an excess risk of co-occurrence of both diseases. Here, we aim to map Alzheimer (AD)-related pathology in GBM-adjacent cortex. Material and Methods To this end, we have screened a cohort of 99 individuals with 200 tissue samples comprising tumor and adjacent cortex, including longitudinal samples in 13 patients. The samples were provided by the Division of Neuropathology and Neurochemistry, Medical University of Vienna from 2002 to 2021. Age and tumor location were abstracted from clinical data where available. All samples were stained for A-beta, tau-AT8 and NeuN using immunohistochemistry. Whole slide scans were segmented and protein deposits were quantified with QuPath. Further statistical analyses were conducted with R. Tau pathology was recorded as neurofibrillary tangles, neuropil threads, and astroglial pathology. Likewise, amyloid pathology was assessed as plaques and/or cerebral amyloid angiopathy (CAA). For both proteins, deposits were grouped into: absent, mild, moderate, and severe. Results In the total cohort, median age was 67.5 ys (range 20-92 ys), the female-to-male ratio 0.68. Overall 44.4 % (n=44/99) showed any type of A-beta and/or taupathology, which was strongly correlated with age (R=0.26, p= 0.001). Among them, 38.6 % (n=17/44) had combined pathology, while 36.4 % (n=16/44) displayed pure amyloid-beta, and 25 % (n=11/44) only tau pathology. A-beta pathology comprised plaques in 74.4 % (n=29/39) and CAA in 28.2 % (n=11/29). Consistent with the spatiotemporal evolution of AD, neurofibrillary tangle load was highest in the temporal lobe (42.9 % n=21/49), while plaque load was most prevalent in the occipital lobe (62.5 % n=5/8). This pattern was accentuated in patients above age 65, while it deviated in those below age 65. Over time, the AD-type pathology increased in 38.4 % (n=5/13) and remained stable in 53.8 % (n=7/13) of recurrent tumors. Total cell densities in tumor-infiltrated cortex ranged from 474 to 7,540 cells/mm2, being similar across all lobes. Higher cell density correlated with decreased neuronal counts (R= -0.46, p<0.0001) and decreased AD-load (R= -0.25, p= 0.002). Conclusion Collectively, our results establish frequent co-occurrence of Alzheimer disease neuropathological changes in the GBM-adjacent cortex. They prompt further investigation of shared pathogenic mechanisms and seek to raise awareness for synergistic effects on cognitive decline.