146 Chapter 6 change was calculated by subtracting the baseline value from the follow-up value and subsequently dividing it by the time between PET scans. All models included a random intercept, and a random slope was added if it significantly improved the model based on the Akaike Information Criterium (AIC). To assess whether findings for tau PET BPND were independent of R1 and vice versa, we additionally ran LMMs in which both (tau PET BPND (per ROI) and R1 (global cortical)) determinants were included simultaneously. As sensitivity analyses, we repeated all analyses with only neuropsychological data collected from the date of the baseline PET scan onward. All analyses were run in CI and CU individuals separately. All results were corrected for multiple comparisons using the Bejamini Hochbergs false discovery rate (FDR) and only FDR-corrected results are described in the main text of the manuscript. All analyses were performed in R version 2022.07. RESULTS Sample Sample characteristics are shown in Table 1. Average follow-up time for cognitive assessments was 4.4 ± 2.1 years ( 6.2 ± 2.4 years in CU individuals and 3.1 ± 1.5 years in CI individuals). At baseline, CU individuals showed very minimal tau PET BPND (-0.03 ± 0.17 Braak I, 0.05 ± 0.09 Braak III/IV, and 0.05 ± 0.07 Braak V/VI), while CI individuals showed substantial tau PET BPND (0.34 ± 0.22 Braak I, 0.33 ± 0.20 Braak III/IV, and 0.38 ± 0.30 Braak V/VI). Tau PET BPND increased over time, with largest increases in Braak I (Table 1). Relative CBF was lower in CI individuals compared to CU individuals (0.87 ± 0.05 versus 0.91 ± 0.04, respectively (p<0.001) at baseline, and decreased over time in CI individuals, but not in CU (Table 1). Baseline PET as determinant for cognitive decline Tau pathology (BPND) Linear mixed models with baseline tau PET BPND (3 ROIs in separate models) as determinant and longitudinal cognitive test scores as outcome were run in CI and CU individuals separately. In CI individuals, higher baseline tau PET BPND in all three ROIs was associated with lower concurrent scores on MMSE, RAVLT immediate recall and TMT-B, except for tau PET BPND in Braak I and MMSE and TMT-B (Figure 1). In addition, higher baseline tau PET BPND in all three ROIs was associated with a steeper rate of decline on MMSE, RAVLT immediate recall and TMT-B, except for tau PET BPND in Braak I and MMSE. Beta’s ranged from -0.15 (SE 0.07; Braak V/VI~RAVLT) to -0.83 (SE 0.15; Braak V/VI~TMT-B) (Figure 1). In CU individuals, higher baseline tau PET
RkJQdWJsaXNoZXIy MjY0ODMw