APOE ε4 carriership status enhances age-related changes in Alzheimer’s disease plasma biomarkers concentrations

APOE ε4 carriership status enhances age-related changes in Alzheimer’s disease plasma biomarkers concentrations | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article APOE ε4 carriership status enhances age-related changes in Alzheimer’s disease plasma biomarkers concentrations Rebecca Rousset, Anouk den Braber, Bram Bongers, Lynn Boonkamp, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8426026/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Alzheimer’s disease (AD) plasma biomarkers are cost-effective tools to detect early stage amyloid pathology. We investigated how interactions between AD genetic risk (apolipoprotein E ( APOE ) ε4, AD polygenic risk score (excluding APOE; PRS AD )), age, and sex affected AD plasma biomarkers concentrations in a population-based cohort. We included 5913 participants age 30 – 80 (64.1% females, 29.8% ε4 carriers) and measured plasma phosphorylated-tau-217 (p-tau217), amyloid-β-42/40 (Aβ-42/40), neurofilament light (NfL), and glial fibrillary acidic protein (GFAP). PRS AD did not associate with biomarker concentrations over age. In ε4 carriers, decreases in Aβ-42/40 became steeper after age 50, and increases in p-tau217 and GFAP became steeper after age 55 and 60 compared to non-carriers, respectively. Analyses in a separate longitudinal cohort confirmed that p-tau217 concentration changes in ε4 carriers reflect AD pathology. These results highlight the potential of plasma biomarkers, especially p-tau217, for early detection of preclinical AD pathology. Biological sciences/Neuroscience Biological sciences/Genetics Health sciences/Biomarkers/Predictive markers Figures Figure 1 Figure 2 Figure 3 Figure 4 INTRODUCTION Alzheimer’s disease (AD) is a progressive neurodegenerative disorder, the primary cause of dementia and a leading cause of death worldwide (1, 2). The preclinical stage of AD, during which the pathology develops without measurable cognitive changes, can last up to 20 years, and can start as early as age 40 (3, 4). Many risk factors for AD have been identified, the main ones being age, with AD incidence increasing exponentially in those over 65 years (2, 5); sex, with about two thirds of AD cases being female (2, 6); and the Apolipoprotein E ( APOE ) ε4 allele (5-7). Additionally, these different AD risk factors have been shown to interact in their effect on overall AD risk and onset (8). The APOE gene codes the apoE protein, which is involved in lipid metabolism and cholesterol transport (9). There are three main alleles of the APOE gene, namely ε2, ε3, and ε4. The APOE ε4 allele has a population frequency of 13.7% (10, 11), and exerts a dose-dependent risk effect on lifetime risk of developing AD (12, 13). Compared to the ε3ε3 genotype, heterozygous and homozygous carriership of the ε4 allele confer a 2- to 3-fold and up to 13-fold increased risk of developing AD, respectively (7). APOE ε4 also partially moderates the effect of age on AD risk and onset, with onset of clinical AD being 8 years younger in heterozygous ε4 carriers and 16 years younger in homozygous ε4 carriers compared to non-carriers (12). Sex differences in the effect of APOE ε4 have also been reported, with the increased AD risk conferred by APOE ε4 being greater in females under 70 years old (6). Plasma biomarkers for diagnosis of AD include phosphorylated tau-217 (p-tau217) (14) and the amyloid-β 42/40 (Aβ-42/40) ratio (15-17). Plasma p-tau217 concentrations begin to increase in the very early disease stages, before measurable changes in cognition (18). At the clinical stage, plasma p-tau217 can accurately differentiate between AD-dementia and other forms of dementia, such as frontotemporal dementia, and is able to detect amyloid co-pathology in Lewy body disorders (14, 19, 20). For Aβ-42/40, decreases in plasma concentration ratio are associated with AD pathology in both early and more advanced disease stages (16, 17). Other biomarkers that are elevated in AD include neurofilament light (NfL), a broad marker of neurodegeneration, and glial fibrillary acidic protein (GFAP), a marker of astrocyte reactivity (15, 21). NfL is significantly elevated in clinical AD and other dementia syndromes (22), while plasma GFAP concentrations are elevated in early AD stages (16, 23) and prognostic of atrophy or cognitive decline (24, 25). Previous studies have reported an association between carriership of the APOE ε4 allele and elevated concentrations of plasma p-tau217 (26-29), NfL (30-32), and GFAP (33, 34), and lower plasma Aβ-42/40 (34). However, findings for plasma NfL, GFAP, and Aβ-42/40 have been inconsistent across studies (31, 33-37), possibly due to insufficient power related to small sample sizes, or from not accounting for sex or age effects and the complex interaction between age, sex, and APOE ε4 carriership as AD risk factors (8). In addition to APOE ε4, the association between other genetic risk factors for AD and plasma biomarker concentrations may also be informative. Polygenic risk scores for AD (PRS AD ) excluding the APOE gene have previously been reported to weakly associate with plasma p-tau181, Aβ-42/40, and GFAP concentrations (38, 39), potentially reflecting of AD genetic risk beyond APOE . A better understanding of how demographic and genetic risk factors for AD relate to AD plasma biomarker concentrations would facilitate the understanding of these plasma biomarkers in clinical practice and research. The aim of this study was to provide insight on the complex interaction between the AD risk factors APOE ε4 carriership, age, and sex on concentrations of plasma p-tau217, Aβ-42/40, NfL, and GFAP. This investigation was conducted in the Netherlands Twin Registry (NTR) cohort, consisting of 5913 cognitively normal participants spanning a large age range of 30 to 80 years old, enabling us to accurately map the genetic effects on plasma biomarkers across life, and to accurately estimate the age at which biomarker concentrations begin to differ between ε4 carriers and non-carriers in both males and females. Using data from a longitudinal study of older individuals (age 50 to 95), EMIF-AD, where longitudinal data on plasma biomarkers and Aβ positron emission tomography (PET) were collected, we explored whether changes in plasma biomarker concentrations in ε4 carriers and non-carriers aligned with the presence of AD pathology. Additionally, we tested whether a PRS AD that excluded the APOE gene region associated with plasma biomarker change over age. RESULTS 1. Cohort description 1.1. NTR demographics A total of 5913 participants with no reported cognitive decline was included from the NTR cohort (see Methods). The majority of participants had the ε3ε3 genotype (n = 3377, 57.1%), followed by the ε3ε4 genotype (n = 1459, 24.7%) and the ε2ε3 genotype (n = 729, 12.3%). The remaining genotypes were much less frequent, with ε2ε4 present in 170 individuals (2.88%), ε4ε4 in 136 individuals (2.30%), and ε2ε2 in 42 individuals (0.71%). This amounted to 1765 individuals with at least one ε4 allele (29.8%). These proportions match those previously reported in white populations (40). PRS AD values were normally distributed, and no outliers were identified (Figure S1 ). NTR cohort demographic information is reported in Table 1 . Overall, ε4 carriers were significantly younger than non-carriers (mean age difference: 1 year, p = 0.004). There were no differences in proportions of females ( p = 0.87), body mass index (BMI) ( p = 0.83) or estimated glomerular filtration rate (eGFR) ( p = 0.10) between the groups. E4 carriers had significantly higher p-tau217 (mean difference: 0.002 pg/mL; p = 0.003) and significantly lower Aβ-42/40 (mean difference: 0.002; p = 0.006). There were no differences in NfL ( p = 0.51) and GFAP ( p = 0.56) between ε4 carriers and non-carriers. Cohort demographic per PRS AD score group (excluding APOE ), are presented in Table S1 . There were no differences in age, proportion of females, proportion of ε4 carriers, or any of the biomarker concentrations between the low, medium, and high PRS AD groups (p > 0.02 for all). 1.2. EMIF-AD demographics For the EMIF-AD cohort, 202 participants were included, of which 77 also had an earlier plasma biomarker measurement as part of the NTR cohort. Samples were collected at baseline biennially over 6 years/3 follow-up points. Of these 202 participants, 195 (96.5%) had at least one available Aβ PET measurement, 66 (32.8%) were ε4 carriers (6 ε2ε4, 55 ε3ε4, 5 ε4ε4) and 135 (67.2%) were non-carriers (2 ε2ε2, 21 ε2ε3, 112 ε3ε3). Participants were further categorized as Aβ PET negative (Aβ PET negative at all Aβ PET assessments) and Aβ PET positive (Aβ PET positive at any of the Aβ PET assessments). Stratifying by Aβ PET status resulted in 109 Aβ PET negative non-carriers (56%), 40 Aβ PET negative ε4 carriers (21%), 20 Aβ PET positive non-carriers (10%), and 26 Aβ PET positive ε4 carriers (13%). EMIF-AD cohort demographics are reported in Table 2 . At baseline, ε4 carriers were significantly younger (mean difference: 4 years, p < 0.001) than non-carriers and this group contained a higher proportion of Aβ PET positive participants (39.4% vs 14.8%, p < 0.001). There were no other significant differences (Table S2). Table 1 Demographic information for the NTR cohort. n Total ε4 non-carriers ε4 carriers Group difference 5913 4148 1765 - Age [year] Mean [range] 49.8 [30–80] 50.1 [30–80] 49.1 [30–80] 0.004 ± Sex [-] Female (%) 3793 (64.1%) 2664 (64.2%) 1129 (64.0%) 0.87* BMI [kg/m 2 ] Mean (SD) 25.7 (4.2) 25.8 (4.2) 25.7 (4.3) 0.83 ± eGFR [mL/min/1.73m 2 ] Mean (SD) 81.2 (14.6) 81.0 (14.4) 81.7 (15.1) 0.10 ± p-tau217 [pg/mL] Mean (SD) 0.030 (0.023) 0.030 (0.023) 0.032 (0.024) 0.003 ± Aβ-42/40 [-] Mean (SD) 0.067 (0.018) 0.068 (0.018) 0.066 (0.018) 0.006 ± NfL [pg/mL] Mean (SD) 9.91 (9.69) 9.96 (9.84) 9.79 (9.31) 0.51 ± GFAP [pg/mL] Mean (SD) 68.3 (56.0) 68.0 (51.1) 69.0 (66.0) 0.56 ± ± t-test ; * Chi-square Table 2 Demographic information for the longitudinal EMIF-AD cohort, per measurement wave. ε4 non-carriers ε4 carriers NTR (n = 46) BL (n = 135) FU1 (n = 107) FU2 (n = 86) FU3 (n = 73) NTR (n = 31) BL (n = 66) FU1 (n = 66) FU2 (n = 57) FU3 (n = 49) Age [year] Mean [range] 59.5 [50–75] 72.2 [61–94] 73.0 [63–95] 73.8 [65–94] 75.0 [67–86] 57.3 [51–72] 67.5 [60–83] 69.4 [63–86] 71.1 [65–88] 73.0 [67–90] Sex [-] Female (%) 27 (58.7%) 86 (63.7%) 64 (59.8%) 51 (59.3%) 46 (63.0%) 14 (45.2%) 30 (45.5%) 30 (45.5%) 24 (42.1%) 19 (38.8%) MMSE score Median [range] NA 29 [24–30] 29 [25–30] 29 [24–30] 29 [26–30] NA 29 [25–30] 29 [25–30] 29 [26–30] 29 [24–30] Aβ PET status Aβ+, % 3 (6.52%) 20 (14.8%) 18 (16.8%) 13 (15.1%) 9 (12.3%) 14 (45.2%) 26 (39.4%) 26 (39.4%) 26 (38.6%) 19 (38.8%) p-tau217 [pg/mL] Mean (SD) 0.031 (0.021) 0.041 (0.024) 0.040 (0.021) 0.042 (0.028) 0.040 (0.020) 0.041 (0.025) 0.047 (0.031) 0.051 (0.045) 0.059 (0.053) 0.059 (0.041) Aβ-42/40 [-] Mean (SD) 0.071 (0.014) 0.061 (0.011) 0.060 (0.012) 0.060 (0.012) 0.059 (0.012) 0.069 (0.016) 0.060 (0.012) 0.058 (0.011) 0.059 (0.011) 0.059 (0.009) NfL [pg/mL] Mean (SD) 13.0 (8.03) 18.0 (9.52) 18.2 (9.45) 18.0 (9.51) 19.4 (8.26) 8.75 (3.38) 15.4 (10.7) 16.8 (18.7) 16.8 (7.89) 20.3 (11.6) GFAP [pg/mL] Mean (SD) 72.9 (36.0) 115 (55.6) 115 (51.9) 121 (67.0) 127 (51.1) 74.1 (47.7) 112 (71.0) 116 (80.7) 132 (98.8) 132 (74.7) Participants were followed over a maximum of five timepoints: Netherlands Twin Register (NTR), which took place prior to the EMIF-AD study and in which only a fraction of EMIF-AD participants also participated; Baseline (BL), the baseline assessment for EMIF-AD; Follow-up 1 (FU1, two years after BL) of the EMIF-AD; Follow-up 2 (FU2, four years after BL) of the EMIF-AD; Follow-up 3 (FU3, six years after BL) of the EMIF-AD. 2. Age-dependent effect of APOE ε4 on plasma biomarker concentrations 2.1. Main cohort (NTR) Plasma biomarker concentrations over age are presented in Fig. 1 . We first applied GALMSS. A main effect of APOE ε4 carriership was observed for p-tau217 (LRT = 17.5, p < 0.001) and Aβ-42/40 (LRT = 20.0, p < 0.001), with p-tau217 concentrations being higher and Aβ-42/40 lower in ε4 carriers compared to non-carriers. An interaction effect between age and APOE ε4 carriership was observed for p-tau217 (LRT = 50.3, p < 0.001), Aβ-42/40 (LRT = 33.2, p < 0.001), and GFAP (LRT = 18.5, p = 0.007). There was no APOE ε4 carriership effect on NfL ( p ≥ 0.76 for all). Using GAM, which allows the computation of confidence intervals, we were further able to determine the age at which biomarker concentrations began to significantly diverge between ε4 carriers and non-carriers. This divergence age between ε4 carriers and non-carriers was determined to be 53.2 for Aβ-42/40, followed by 58.8 for p-tau217, and 62.8 for GFAP (Figure S2), with the magnitude of difference being most pronounced for p-tau217. At age 50, mean p-tau217 concentrations were virtually the same between ε4 carriers and non-carriers (mean difference = 0.0001 pg/mL), while at age 75 mean concentrations were 0.02 pg/mL higher in ε4 carriers. The difference in ratio of Aβ-42/40 increased from 0.001 at age 50 to 0.006 at age 75. Differences in GFAP concentrations between ε4 carriers and non-carriers increased from 2.1 pg/mL at age 50 to 14.5 pg/mL at 75. Model specifications are reported in Table S3. When considering heterozygous and homozygous ε4 carriers separately, significant age-dependent effects of APOE ε4 carriership were again found for p-tau217, Aβ-42/40, and GFAP ( p ≤ 0.005 for all) (Table S4). For all three biomarkers, concentrations appeared to start increasing between 5 and 10 years earlier for homozygous carriers compared to heterozygous carriers (Figure S3). We repeated the analyses in a subset of participants who were genetically unrelated. This subset consisted of 3624 individuals (ε4 carriers: 1075 (29.7%)). Results from this sensitivity analysis mirrored the results in the full NTR cohort, aside from the age by APOE interaction effect initially found for GFAP no longer being significant ( p = 0.02) (Table S5). FIGURE 1 2.2. Stratification by sex Concentrations over age by sex are presented in Fig. 2 . Model specifications are reported in Table S6. There was a main effect of APOE ε4 carriership on p-tau217 in both males (LRT = 6.89, p = 0.009) and females (LRT = 8.51, p = 0.004), as well as an interaction effect between age and APOE ε4 carriership in males (LRT = 35.9, p < 0.001) but not females (LRT = 9.63, p = 0.04), although a trend was observed. Using GAM analysis, the age at which concentrations began to significantly differ between ε4 carriers and non-carriers in males was determined to be 61.3 years (Figures S4–S5). In males, p-tau217 concentrations began to increase with age for both ε4 carriers and non-carriers, but do so markedly more steeply in ε4 carriers. At age 50, the difference in p-tau217 concentrations between male ε4 carriers and non-carriers was less than 0.0001 pg/mL, while at age 75 p-tau217 concentrations were 0.03 pg/mL higher among ε4 carriers. Looking at heterozygous and homozygous ε4 carriers separately, a main APOE ε4 carriership effect was found for both sexes, but the age by APOE ε4 carriership interaction was only significant in males, where p-tau217 concentrations started increasing at an earlier age in homozygous carriers compared to heterozygous carriers (Table S7, Figures S6–S7). APOE ε4 carriership also exerted a main effect on Aβ-42/40 in both males (LRT = 10.5, p = 0.001) and females (LRT = 10.9, p = 0.001), and also interacted with age (male: LRT = 15.3, p = 0.004; females: LRT = 16.1, p = 0.004). Using GAM, the age at which Aβ-42/40 concentrations begin to differ between male ε4 carriers and non-carriers was calculated to be 52.7 years old (Figures S4–S5). The GAM approach did not detect a demarcation age between female ε4 carriers and non-carriers. Difference in Aβ-42/40 increased from 0.001 at age 50 to 0.006 at age 75 in males and to 0.007 in females. When treating heterozygous and homozygous carriers separately (Table S7), concentrations began to decrease 5 to 10 years earlier for homozygous carriers (Figures S6–S7). In line with the results in the full sample, APOE ε4 carriership did not influence NfL concentrations in either males or females ( p ≥ 0.17 for both), nor did it have an interaction with age in either sex ( p ≥ 0.44 for both). Interestingly, when evaluating heterozygous and homozygous ε4 carriers separately, a main effect of APOE ε4 carriership was found in males only (LRT = 15.9, p < 0.001), with NfL concentrations being overall higher in homozygous carriers compared to heterozygous carriers and non-carriers. For GFAP, APOE ε4 carriership did not exert a main effect in either males or females ( p ≥ 0.43 for both), and there no longer was an age-dependent effect of APOE ε4 in either sexes (males: LRT = 14.8, p = 0.02; females: LRT = 5.62, p = 0.45). An age-dependent effect of APOE ε4 carriership effect was observed for males only when handling heterozygous and homozygous carriers separately, showing a more linear increase in concentrations over age compared to the exponential increase previously observed (Figures S6–S7). FIGURE 2 3. Age-dependent effect of PRS on plasma biomarker concentrations All results relating to the PRS AD (with and without APOE ε4 correction) are presented in the eMaterials (Tables S9–S12/Figures S8–S9). The PRS AD was not significantly associated with any of the biomarker concentrations in the NTR. In males only, a higher PRS AD was associated with higher p-tau217 concentrations at all ages ( p < 0.001). No age-dependent effect was found in either the full cohort or in the male and female subsamples. Adjusting for APOE ε4 did not lead to different results. When repeating the analysis in the subset of genetically unrelated individuals, results were similar to those from the NTR, though an age-independent effect on p-tau217 concentrations was additionally detected ( p = 0.008), with a higher PRS AD being associated with overall higher p-tau217 concentrations (Table S12). 4. Additional analysis in the longitudinal EMIF-AD cohort with available Amyloid PET results Concentrations in the EMIF-AD cohort aligned well with concentrations from the NTR. At age 60, p-tau217 concentrations were significantly higher in ε4 carriers compared to non-carriers ( p < 0.001). These results indicate an enduring ε4 carriership effect on p-tau217 concentrations in participants 60 and older. Other biomarker concentrations did not significantly differ between ε4 carriers and non-carriers at cross-sectional age 60. However, NfL concentrations were observed to increase more steeply over age in ε4 carriers than non-carriers ( p = 0.001). Unlike the results from the NTR cohort, rate of change in p-tau217, Aβ-42/40, and GFAP concentrations over age was not affected by APOE ε4 carriership ( p ≥ 0.14 for all) (Fig. 3 /Figure S10–S11), which is expected as EMIF-AD is an older cohort - monitoring started at ages above the ages where biomarkers started to differ in the NTR. Results are reported in Table S13. Participants were next stratified on Aβ PET status. At age 60, cross-sectional differences were found for Aβ PET positive/ε4 carriers, who had higher p-tau217 concentrations than both Aβ PET positive/non-carriers and Aβ PET negative/non-carriers ( p < 0.001 for both). Aβ PET negative/ε4 carriers also had higher p-tau217 concentrations than Aβ PET positive/non-carriers ( p = 0.003). Longitudinally, p-tau217 concentrations were further observed to increase more steeply over age for Aβ PET positive/ε4 carriers than for both Aβ PET negative/ε4 carriers and Aβ PET negative/non-carriers ( p < 0.01 for both). Aβ PET positive/ε4 non-carriers were also observed to have steeper p-tau217 slopes over age compared to both Aβ PET negative/ε4 carriers and Aβ PET negative/non-carriers ( p < 0.001 for both). Together, these results show that steeper p-tau217 slopes over age are reflective of Aβ PET status, and therefore AD pathology, rather than APOE ε4 carriership, while APOE ε4 carriership appears to affect the timing of increase in p-tau217 concentrations, with concentrations increasing 20 to 25 years earlier in ε4 carriers compared to non-carriers. Though no cross-sectional differences in NfL concentrations were found at age 60, a steeper increase in NfL concentrations was found for Aβ PET negative/ε4 carriers compared to Aβ PET negative/non-carries ( p = 0.002). There were no differences in Aβ-42/40 concentrations between any of the Aβ PET and ε4-carriership groups ( p ≥ 0.11 for all) (Fig. 4 /Figure S12). Results are reported in Table S14. FIGURE 3 FIGURE 4 DISCUSSION In this large cross-sectional cohort of cognitively normal individuals, covering a broad part of the age range where amyloid pathology changes begin to occur, we showed an effect of the APOE ε4 allele on plasma concentrations of p-tau217, Aβ-42/40, and GFAP, but not NfL. By taking into account age and sex, which are two major risk factors for AD, we were able to show a three-way age, sex, and APOE ε4 carriership interaction on p-tau217 concentrations and the Aβ-42/40 ratio, as well as an interaction between age and APOE ε4 carriership on GFAP concentrations, with the effect of APOE ε4 carriership being most pronounced for p-tau217. We observed that in ε4 carriers, the Aβ-42/40 ratio starts to decrease more steeply after age 50, p-tau217 concentrations increased more steeply after 55, and GFAP concentrations after 60, compared to non-carriers. Our supplementary analyses in a cohort containing longitudinal biomarker and Aβ PET information allowed us to show that increases in p-tau217 concentrations over age are due to early AD pathology, and exacerbated by ε4 carriership, which affects the timing of offset. Additionally, by stratifying by sex in our cross-sectional cohort, we were able to show a male-specific effect in the change in p-tau217 concentration over age, with carriership of the APOE ε4 allele being associated with steeper increase in p-tau217 from its inflection point, while in females the difference between ε4 carriers and non-carriers was age independent, although a trend was observed. For Aβ-42/40, we observed an age-dependent effect of APOE ε4 carriership in both sexes. No effect was observed on GFAP nor NfL in the sex-specific subsets. When handling heterozygous and homozygous ε4 carriers separately, we additionally observed male-specific dose-response effects of APOE ε4 carriership on both NfL (age-independent effect) and GFAP (age-dependent effect). The PRS AD did not show a strong association with any of the biomarkers. Our results align with previous literature, where differences in p-tau217 concentrations between ε4 carriers and non-carriers were reported (26–29), though without accounting for age- or sex-dependent effects. Adding to this, we observe an age-dependent effect on Aβ-42/40 and GFAP concentrations in the NTR. Finally, while NfL concentrations have previously been reported to be more elevated in ε4 carriers compared to non-carriers (30, 31), we here report that homozygous ε4 carriers have consistently elevated NfL concentrations across their lifespan compared to heterozygous ε4 carriers and non-carriers. Discrepancies between our results and previous results on the association between APOE ε4 carriership and plasma Aβ-42/40, NfL and GFAP (31, 33–37) may be due to small sample sizes or not adjusting for age and/or sex in previous studies. Our observation of a lack of effect of the PRS AD on the age-related slopes for Aβ-42/40, NfL, or GFAP is in line with previous research, which report that PRS AD associates more strongly with changes further along the disease continuum (39, 41), such as cognitive decline in individuals with mild cognitive impairment due to AD (41). We did however detect a male-specific age-independent effect of the PRS AD on p-tau217 concentrations in the NTR, and a similar effect in both sexes when performing the analysis in a genetically unrelated subset of the population. Such weak associations between the PRS AD and both p-tau217 and p-tau181 had previously been reported (38, 42). Here, we were able to demonstrate that the association between PRS AD and p-tau217 concentrations is independent of age, which is in contrast to the effect of APOE where the effects were exacerbated by age. Plasma concentrations of AD biomarkers have been reported to change before the clinical onset of AD. A previous study on plasma p-tau217 concentrations over age in participants with and without mild cognitive impairment reported that after age 55, p-tau217 concentrations in participants with AD pathology (as defined by Aβ PET) increase 12 times faster than in participants without AD pathology (28). Using APOE ε4 carriership as a proxy for early AD pathology, we here were able to show a similar pattern of steeper increase starting from around age 55 in ε4 carriers compared to non-carriers even in a cohort of cognitively normal participants only. This was further confirmed by comparing the plasma data in relation to Amyloid PET positivity in the longitudinal EMIF-AD cohort. Here, we observed that increases in p-tau217 concentrations over age are principally due to AD pathology, and not solely ε4 carriership. It is likely that differences in plasma p-tau217 concentrations between ε4 carriers and non-carriers in the NTR cohort were likewise due to developing AD pathology. Our results are in line with a recently published study, which reported that p-tau217 concentrations increase with increased AD pathological burden (29). Stratifying by sex further enabled us to show that the pattern and timing of p-tau217 increase differs between males and females. In males, increase in p-tau217 occurred after age 55 and seemed to be almost entirely due to APOE ε4 carriership (and, by proxy, early AD pathology), and the increase in p-tau217 concentrations was much steeper in male ε4 carriers compared to female ε4 carriers. In females, an increase in p-tau217 concentrations independent of APOE ε4 carriership could be observed between age 50 and 55, before concentrations began to differ between the carriership groups, though this divergence between APOE carriership groups was only a trend. The association of plasma Aβ-42/40, NfL, and GFAP with Aβ positivity over age in pre-clinical AD populations has so far rarely been investigated, since such longitudinal cohorts covering a broad age range are not available. However, predictive models studying cognitive decline have been able to estimate that the plasma Aβ-42/40 ratio start decreasing between 14 and 41 years before clinical AD onset (43, 44). For NfL, concentration changes are detected up to 12 years prior to clinical onset (44), but are clearer and increase more steeply during and after onset of cognitive impairment (43). The same has been reported for GFAP concentrations (25, 43, 44). Here, in a cohort of cognitively normal individuals, we similarly observed that Aβ-42/40 concentrations are the first to begin to differ between APOE ε4 carriers and non-carriers, followed by GFAP 10 to 15 years later, though we did not detect difference in NfL concentrations. However, we did not determine if the decrease in Aβ-42/40 and increase in GFAP concentrations were due to age, AD pathology, or an APOE ε4 effect independent of AD pathology. Additionally, among Aβ PET negative individuals only, we showed a steeper increase in NfL concentrations over age in ε4 carriers compared to non-carriers. This may be a reflection of a dose-dependent effect of APOE ε4, since we observed higher NfL concentrations in homozygous carriers compared to both heterozygous carriers and non-carriers in the NTR cohort. This ε4 carriership effect independent of AD pathology, which we also did not observe in the NTR, may be a reflection of the higher vascular risk associated with APOE ε4 carriership (45). One limitation is that our cohort consists of families, which may increase the risk of false positive results (46) – however, the results of our sensitivity analysis in a subset of non-genetically related individuals were highly comparable to the results of our primary analysis. Another limitation is that we did not formally test cognitive function, however we were able to exclude all participants who reported memory complaints. Thus, future studies should focus on longitudinal assessment of amyloid status and cognition, to further expand our understanding of p-tau217 as a predictive marker for future amyloid pathology. The strengths of our study were that our cohort spanned an age range of 50 years, allowing us to narrow down the age at which the rate of concentration change begins to differ between ε4 carriers and non-carriers. Our use of GAMLSS also allowed us to more accurately capture the non-linear relationship between age and the AD biomarker concentrations without data transformation. The GAM analysis further allowed us to determine the likely age of divergence in concentrations between ε4 carriers and non-carriers. Using a cohort with available longitudinal biomarker information and Aβ PET, we were also able to relate our APOE ε4 carriership results to actual AD pathology. Taken together, we demonstrated that APOE ε4 carriership influences plasma concentrations of p-tau217, Aβ-42/40, and GFAP, but not NfL, in an age-dependent manner in cognitively normal individuals, as well as a sex-dependent manner for p-tau217 and Aβ-42/40. We observed that p-tau217, Aβ-42/40, and GFAP concentrations begin to differ between ε4 carriers and non-carriers 15 to 20 years before estimated mean age of clinical AD onset. In a longitudinal cohort, we were able to show that increases in p-tau217 concentrations over age are principally due to AD pathology, while carriership of the APOE ε4 allele causes in a younger age of increase onset. We were also able to demonstrate that PRS AD does not affect AD plasma biomarker concentrations, except for a male-specific effect on p-tau217. These results highlights the potential of AD plasma biomarkers, particularly p-tau217, for detection of early AD pathological changes, which in turn will have implications for inclusion in trials targeting AD. Online Methods 1. Cohorts 1.1. The Netherlands Twin Register cohort Participants were included from the Netherland Twin Register (NTR) biobank, for which blood collection took place between 2004 and 2011 (47). Details on blood collection and storage procedures have been published elsewhere (47, 48). The NTR collects genotypic and longitudinal phenotypic data of Dutch twins and other multiples and their family members (47, 48). Although cognition is not formally assessed as part of the NTR, multiple surveys contain questions on memory problems (time of survey ranges from 22 days prior to biobank measurement to 9.3 years after biobank measurement, with a mean (SD) of 3.4 years after measurement (2 years)). At time of blood collection, participants also completed a general health survey, which provided information on any medical condition the participants may have or any medication they may be taking. All participants in this cohort were at least 30 years of age. Participants were eligible for inclusion if information on age, sex, and APOE genotype was available. Participants were excluded if they had reported having problematic or very problematic memory complaints on any NTR survey (n = 77), or if they reported having a diagnosis of dementia due to Alzheimer’s disease (n = 1), Parkinson’s disease (n = 6), extrapyramidal and movement disorders (n = 5), multiple sclerosis (n = 14), or speech disturbance (n = 1) (total n = 27) during the biobank-specific survey. Based on the age distribution, and to ensure reliable estimates, participants above 80 years of age were excluded (n = 43). 1.2. The European Medical Information Framework for Alzheimer's disease cohort To further strengthen our analyses, participants from the Dutch subset of the European Medical Information Framework for Alzheimer's disease (EMIF-AD) cohort were included (49, 50). The EMIF-AD cohort consists of 206 monozygotic twins followed longitudinally, with follow-up data being collected every 2 years. Cognition was assessed using the Mini Mental State Exam (MMSE), which resulted in 4 participants being excluded for having an MMSE score < 24 at any timepoint. For each participant, blood is collected at baseline and every follow-up point, while an Aβ [18F]flutemetamol-PET (Aβ PET) scan is performed at baseline and every other follow-up point. As part of the EMIF-AD cohort, participants were followed from 0 to 7.5 years (mean time: 4.8 years (SD = 2.5)). A subset of the EMIF-AD participants (n = 80) were also part of the NTR. These participants were followed from 8.5 to 17.5 years (mean time: 14.5 years (SD = 2.5)). The mean time between NTR participation and baseline EMIF-AD measurement was 9 years (SD = 2 years). 2. Alzheimer’s disease biomarker measurements All biomarker measurements were performed on the Simoa HD-X platform. Biomarker concentrations were measured using the same kits and on the same Simoa HD-X platforms for both the NTR and EMIF-AD cohorts. Concentrations of p-tau217 were measured using the ACC pTau217 Assay (reference #104740), and concentrations of NfL, GFAP, Aβ-42 and Aβ-40 were assessed using the Neurology-4-plex E Advantage kit (reference #103670). For p-tau217 in the NTR, the first 337 samples were measured in duplicate. Quality of the duplicate measurements was assessed (Figure S13), after which it was decided the remaining 5857 samples would be measured in single. In-house quality controls (QCs) were included in each run, at three levels (low, medium, and high concentrations). Additionally, kit QCs (also low, medium, and high concentrations) were included every five runs. Inter-assay coefficients of variation (CVs) were 9.1%. Seventy seven samples had concentrations below the lower limit of detection. For EMIF-AD, all samples were measured in duplicate over the course of 6 runs, with inter-assay CVs of 5.6% and intra-assay CVs of 7.1%. For measurements of NfL, GFAP, Aβ-42, and Aβ-40, two QCs (low, high) per biomarker were included in each run. In the NTR, inter-assay CVs were 4.5% for NfL, 4.6% for GFAP, 5.2% for Aβ-42, and 4.1% for Aβ-40. In the EMIF-AD cohort, inter-assay CVs were 5.5% for NfL, 5.0% for GFAP, 7.3% for Aβ42, and 5.0% for Aβ40. 3. APOE genotyping and PRS AD calculation The same methods and platforms were used for APOE genotyping in both the NTR and EMIF-AD cohorts. DNA was extracted from peripheral blood samples. Participants were genotyped on the Affymetrix Axiom and Affymetrix 6 arrays (51). Participants’ APOE genotype was determined using imputed dosages of single nucleotide polymorphisms rs429358 (APOE ε4) and rs7412 (APOE ε2) (52). The PRS AD were calculated for NTR participants only based on the genome-wide association study by Bellenguez et al. (53) for AD, excluding APOE. For calculation of the PRS AD , we retained variants for which the effect allele frequency (EAF) was 0.01 ≤ EAF ≤ 0.99. Variant EAF and effect sizes were aligned with the NTR reference for the 1000 genomes variants. We excluded the APOE region by removing variants between base pair positions 45384477 and 45432606 on chromosome 19 (build 37/hg 19) from the above GWAS summary statistics, in keeping with Ware et al. (54). Discovery variants that were not part of this reference were discarded. The LDpred 0.9 software (55) was used to estimate the target linkage disequilibrium structure, using a selection of unrelated NTR individuals and selecting a set of well-imputed variants in the NTR sample. The PRS AD generated using the infinitesimal model were preserved for analysis. 4. Amyloid-β status assessment In EMIF-AD participants only, an Aβ PET scan was performed at baseline and follow-up 2 (mean time to follow-up 2 = 4.33 years (SD = 0.42)). The Aβ PET imaging protocol has been described in detail elsewhere (50). Based on the Aβ PET scan, participants were scored as either Aβ- (no significant AD pathology) or Aβ+ (significant AD pathology) by three trained readers following General Electric Healthcare guidelines. In the case of a disagreement, the consensus rating of two readers was used. 5. Statistical analyses 5.1. Main investigation in the NTR cohort All statistical analyses were performed in R version 4.3.2. Participants were defined “ε4 carriers” (ε2ε4, ε3ε4, ε4ε4) and “ε4 non-carriers” (ε2ε2, ε2ε3, ε3ε3). All analyses included body mass index (BMI), estimated Glomerular Filtration Rate (eGFR, a measure of kidney function), and genetic principal components 1 to 20 as covariates. Our decision to include BMI and eGFR as covariates is based on previous literature showing that these two factors may influence plasma biomarker concentrations (56–58). For the statistical analyses, the PRS AD was handled as continuous. For visual representation, participants were classified as low, medium, or high PRS AD using tertile cutoffs. Generalized additive models for location, scale and shape (GAMLSS) were fit for each AD biomarker using the gamlss package (version 5.4–22). Using the fitDist function, it was determined that a Box-Cox t distribution (59) would be the most appropriate for each biomarker. The analysis was then conducted in two parts: first, test whether APOE ε4 carriership influenced biomarker concentrations over age, and second, test whether low, medium, or high PRS AD influenced plasma biomarker concentrations over age. For all analyses, nested models were compared using a likelihood ratio test, whereas a p < 0.01 was taken to indicate the more complex model was a better fit of the data. To test the effect of APOE ε4 carriership on plasma biomarkers concentrations over age, three models were fit: 1, with age as the only predictor; 2, with age and the main effect of APOE ε4 carriership; and 3, with age, APOE ε4 carriership, and the age* APOE ε4 carriership interaction. Model 1 and 2 were compared to determine whether there was a main effect of APOE ε4 carriership on plasma biomarker concentrations, and Model 2 and 3 were compared to determine whether, in addition to a main effect, APOE ε4 carriership also affected the concentration change over age. Considering the reported dose-dependent effect of APOE ε4 carriership on the risk of developing AD, we further investigated whether a dose-dependent effect was observed on plasma biomarker concentrations by repeating the analyses modelling APOE ε4 carriership as ordinal (non-carrier, heterozygous ε4 carrier, homozygous ε4 carrier). The same approach was taken to test the effect of PRS AD on plasma biomarker concentrations over age, replacing APOE carriership with PRS AD . The analyses were then repeated, this time accounting for APOE ε4 carriership. The three models were then: 1, age and main effect of APOE ; 2, age, APOE , and main effect of PRS AD ; and 3, age, APOE , main effect of PRS AD , and the age*PRS AD interaction. All analysis were performed first in the main cohort (NTR) and then in sex-specific subsets to test for sex differences (Figure S14). Analyses were repeated using a generalized additive models (GAM), which allow us to generate confidence intervals and statistically estimate the age at divergence (if any) between the different AD genetic risk groups, but are more sensitive to overfitting. 5.1.1. Sensitivity analysis for genetic relatedness Because our cohort consists of genetically related individuals, we performed a sensitivity analysis randomly including only one person per family (with the exceptions of families consisting only of spouses, in which case both spouses were included). Within this cohort of individuals that are not genetically related, we repeated our analyses to test the effect of APOE ε4 carriership and PRS AD (excluding APOE ) on AD biomarker concentrations over age. 5.2. Additional analysis in the EMIF-AD longitudinal cohort Similarly to the main analyses, participants in the EMIF-AD cohort were defined as “ε4 carriers” (ε2ε4, ε3ε4, ε4ε4) and “ε4 non-carriers” (ε2ε2, ε2ε3, ε3ε3). Participants were further classified as Aβ PET negative if they were scored as Aβ- at all of their assessments (baseline and/or follow-up 2). Participants that were scored as Aβ + at either of their assessment were classified as Aβ PET positive. Therefore, the Aβ PET positive group includes both participants that were Aβ + at baseline, but also participants who were Aβ- at baseline and Aβ + at follow-up. Including participants who converted from Aβ- to Aβ + over a four year range into the Aβ PET positive group was done 1) to determine if plasma biomarker concentrations begin to change in the earliest AD stages and 2) because the specific timing of conversion from Aβ- to Aβ + could not be determined. To model individual longitudinal biomarker changes, linear mixed-effects models (LMMs) were fit, allowing for both random intercepts and random slopes, as well as a random effect for family to account for the familial clustering within monozygotic twin pairs. The LMMs included the measurements from the NTR biobank when available. In all models, age and sex were added as covariates, and biomarker concentrations were log2 transformed to fulfil linear regression assumptions. In the first set of analyses, the LMMs were fit with an age* APOE ε4 carriership interaction without accounting for Aβ PET status. The emmeans function was used to estimate mean biomarker concentrations over age per APOE carriership group (ε4 carrier or non-carrier). These estimated biomarker concentration trajectories were visually compared to the estimated concentration changes in the NTR cohort. Then, a second set of LMMs were fit, with a three-way age* APOE ε4 carriership* Aβ PET status interaction. The emmeans function was used to estimate mean biomarker concentrations over age per APOE carriership group and per Aβ PET status, resulting in four groups: Aβ PET negative non-carriers, Aβ PET negative ε4 carriers, Aβ PET positive non-carriers, and Aβ PET positive ε4 carriers. The emmeans and emtrends functions were used to compare biomarker trajectories in the four aforementioned groups, with a p < 0.01 taken as the significance threshold. Declarations For the Netherlands Twin Register: "Written informed consent was obtained from all participants. The study was approved by the Central Ethics Committee on Research Involving Human Subjects of the VU University Medical Centre, Amsterdam, an Institutional Review Board certified by the U.S. Office of Human Research Protections (IRB Number IRB00002991 under Federal-wide Assurance-FWA00017598; IRB/ institute codes, NTR 03-180)." For the EMIF cohort: "The Medical Ethics Review Committee of the Amsterdum University Medical Center performed approved the study. Research was performed according to the principles of the Declaration of Helsinki and in accordance with the Medical Research Involving Human Subjects Act and codes on ‘good use’ of clinical data and biological samples as developed by the Dutch Federation of Medical Scientific Societies. All participants gave written informed consent (IRB registration number: 2014.210." Acknowledgments Funding for blood sample collection was obtained from the Netherlands Organization for Scientific Research (NWO) and The Netherlands Organisation for Health Research and Development (ZonMW) grants 904-61-090, 985-10-002, 904-61-193,480-04-004, 400-05-717, Addiction-31160008, 016-115-035, 400-07-080, Middelgroot-911-09-032, NWO-Groot 480-15-001/674; the European Community's Fifth and Seventh Framework Program (FP5- LIFE QUALITY-CT-2002-2006, FP7- HEALTH-F4-2007-2013, grant 01254: GenomEUtwin, grant 01413: ENGAGE). RR , CT and AdB receive funding from the CANTATE project. The CANTATE project is funded by the Alzheimer Drug Discovery Foundation. Research of CT is supported by the European Commission (Marie Curie International Training Network, grant agreement No 860197 (MIRIADE), Innovative Medicines Initiatives 3TR (Horizon 2020, grant no 831434) EPND (IMI 2 Joint Undertaking (JU), grant No. 101034344) and JPND (bPRIDE), National MS Society (Progressive MS alliance), Alzheimer Association, Health Holland, the Dutch Research Council (ZonMW), Alzheimer Drug Discovery Foundation, The Selfridges Group Foundation, and Alzheimer Netherlands. CT is recipient of ABOARD, which is a public-private partnership receiving funding from ZonMW (#73305095007) and Health~Holland, Topsector Life Sciences & Health (PPP-allowance; #LSHM20106). CT is recipient of TAP-dementia, a ZonMw funded project (#10510032120003) in the context of the Dutch National Dementia Strategy. CT has research contracts with Acumen, ADx Neurosciences, AC-Immune, Alamar, Aribio, Axon Neurosciences, Beckman-Coulter, BioConnect, Bioorchestra, Brainstorm Therapeutics, C2N diagnostics, Celgene, Cognition Therapeutics, EIP Pharma, Eisai, Eli Lilly, Fujirebio, Instant Nano Biosensors, Merck, Muna, Nitrase Therapeutics, Novo Nordisk, Olink, PeopleBio, Quanterix, Roche, Sysmex, Toyama, Vaccinex, Vivoryon. She is editor in chief of Alzheimer Research and Therapy, and serves on editorial boards of Molecular Neurodegeneration, Alzheimer’s & Dementia, Neurology: Neuroimmunology & Neuroinflammation, Medidact Neurologie/Springer, and is committee member to define guidelines for Cognitive disturbances, and one for acute Neurology in the Netherlands. She has consultancy/speaker contracts for Aribio, Biogen, Beckman-Coulter, Cognition Therapeutics, Danaher, Eisai, Eli Lilly, Janssen, Merck, Neurogen Biomarking, Nordic Biosciences, Novo Nordisk, Novartis, Olink, Quanterix, Roche, Sanofi and Veravas. BB, LB, DW, LL, RP, PJV , SMvdL and EdG has nothing to disclose. References Organization WH. Global Health Estimates 2020: Deaths by Cause, Age, Sex, by Country and by Region, 2000-2019 2020 [Available from: https://www.who.int/data/gho/data/themes/mortality-and-global-health-estimates/ghe-leading-causes-of-death. Association As. 2025 Alzheimer's disease facts and figures. Alzheimer's & Dementia. 2025;'21(4):e70235. Rafii MS, Aisen PS. Detection and treatment of Alzheimer's disease in its preclinical stage. nature aging. 2023;3:520–31. Jia J, Ning Y, Chen M, Wang S, Yang H, Li F, et al. Biomarker Changes during 20 Years Preceding Alzheimer's Disease. The New England Journal of Medicine. 2024;390(8):712–22. Therriault J, Pascoal TA, Benedet AL, Tissot C, Savard M, Chamoun M, et al. Frequency of Biologically Defined Alzheimer Disease in Relation to Age, Sex, APOE ε4, and Cognitive Impairment. Neurology. 2021;96(7):e975–85. Neu SC, Pa J, Kukull W, Beekly D, Kuzma A, Gangadharan P, et al. Apolipoprotein E Genotype and Sex Risk Factors for Alzheimer Disease. JAMA Neurology. 2017;74(10):1178–89. Belloy ME, Andrews SJ, Guen YL, Cuccaro M, Farrer LA, Napolioni V, et al. APOE Genotype and Alzheimer Disease Risk Across Age, Sex, and Population Ancestry. JAMA Neurology. 2023;80(12):1284–94. Riedel BC, Thompson PM, Brinton RD. Age, APOE and Sex: Triad of Risk of Alzheimer's Disease. The Journal of Steroid Biochemistry and Molecular Biology. 2016;160:134–47. Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988;240(4852):622–30. Abondio P, Sazzini M, Garagnani P, Boattini A, Monti D, Franceschi C, et al. The Genetic Variability of APOE in Different Human Populations and Its Implications for Longevity. Genes (Basel). 2019;10(3):222. Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of Age, Sex, and Ethnicity on the Association Between Apolipoprotein E Genotype and Alzheimer Disease. JAMA. 1997;278(1349-56). Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer's Disease in Late Onset Families. Science. 1993;261:921–23. Blacker D, Haines JL, Rodes L, Terwedow H, Go RC, Harrell LE, et al. ApoE-4 and age at onset of Alzheimer's disease: the NIMH genetics initiative. Neurology. 1997;48(1):139–47. Ashton NJ, Brum WS, Molfetta GD, Benedet AL, Arslan B, Jonaitis E, et al. Diagnostic Accuracy of a Plasma Phosphorylated Tau 217 Immunoassay for Alzheimer Disease Pathology. JAMA Neurology. 2024:e235319. Altomare D, Stampacchia S, Ribaldi F, Tomczyk S, Chevalier C, Poulain G, et al. Plasma biomarkers for Alzheimer’s disease: a field-test in a memory clinic. Journal of Neurology, Neurosurgery, and Psychiatry. 2023;94:420–7. Chatterjee P, Pedrini S, Doecke JD, Thota R, Villemagne VL, Doré V, et al. Plasma Aβ42/40 ratio, p-tau181, GFAP, and NfL across the Alzheimer's disease continuum: A cross-sectional and longitudinal study in the AIBL cohort. Alzheimer's & Dementia. 2022;19(4):1117–34. Wojdała AL, Bellomo G, Toja A, Gaetani L, Parnetti L, Chiasserini D. CSF and plasma Aβ42/40 across Alzheimer’s disease continuum: comparison of two ultrasensitive Simoa® assays targeting distinct amyloid regions. Clinical Chemistry and Laboratory Medicine. 2023;62(2). Ossenkoppele R, Salvadó G, Janelidze S, Binette AP, Bali D, Karlsson L, et al. Plasma p-tau217 and tau-PET predict future cognitive decline among cognitively unimpaired individuals: implications for clinical trials. nature aging. 2025;5:883–96. Eratne D, Kang M, Malpas CB, Dang C, Lewis C, Bhalala OG, et al. Plasma p‐tau217, NfL, GFAP diagnostic performance and biomarker profiles in Alzheimer's disease, frontotemporal dementia, and psychiatric disorders, in a prospective unselected neuropsychiatry memory clinic. Alzheimer's & Dementia. 2025;21(10):e70717. Woo KA, Yoon EJ, Kim R, Kim H, Kim S, Jin B, et al. Plasma p-tau217 and Aβ42/40 as markers of Aβ pathology in the Lewy body continuum. Alzheimer's & Dementia. 2025;21(10):e70788. Teunissen CE, Verberk IMW, Thijssen EH, Vermunt L, Hansson O, Zetterberg H, et al. Blood-based biomarkers for Alzheimer's disease towards clinical implementation. Lancet Neurology. 2022;21(1):66–77. Forgrave LM, Ma M, Best JR, DeMarco ML. The diagnostic performance of neurofilament light chain in CSF and blood for Alzheimer's disease, frontotemporal dementia, and amyotrophic lateral sclerosis: A systematic review and meta-analysis. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2019;11:730–43. Shir D, Graff‐Radford J, Hofrenning EI, Lesnick TG, Przybelski SA, Lowe VJ, et al. Association of plasma glial fibrillary acidic protein (GFAP) with neuroimaging of Alzheimer's disease and vascular pathology. Alzheimer's & Dementia. 2022;14(1):e12291. Lin J, Ou R, Li C, Hou Y, Zhang L, Wei Q, et al. Plasma glial fibrillary acidic protein as a biomarker of disease progression in Parkinson’s disease: a prospective cohort study. BMC Medicine. 2023;21:420. Chatterjee P, Vermunt L, Gordon BA, Pedrini S, Boonkamp L, Armstrong NJ, et al. Plasma glial fibrillary acidic protein in autosomal dominant Alzheimer's disease: Associations with Aβ-PET, neurodegeneration, and cognition. Alzheimer's & Dementia. 2023;19(7):2790–804. Panday N, Yang Z, Cieza B, Reyes-Dumeyer D, Kang MS, Montesinos R, et al. Plasma phospho-tau217 as a predictive biomarker for Alzheimer’s disease in a large south American cohort. Alzheimer's Research & Therapy. 2025;17:1. Lehmann S, Schraen-Maschke S, Vidal J-S, Delaby C, Buee L, Blanc F, et al. Clinical value of plasma ALZpath pTau217 immunoassay for assessing mild cognitive impairment. Journal of Neurology, Neurosurgery & Psychiatry. 2024;95:1046–53. Mielke MM, Dage JL, Frank RD, Algeciras-Schimnich A, Knopman DS, Lowe VJ, et al. Performance of plasma phosphorylated tau 181 and 217 in the community. Nature Medicine. 2022;28(7):1398–405. Aarsland D, Sunde AL, Tovar-Rios DA, Leuzy A, Fladby T, Zetterberg H, et al. Prevalence of Alzheimer's disease pathology in the community. Nature. 2025. Malek-Ahmadi M, Su Y, Ghisays V, Luo J, Devadas V, Chen Y, et al. Plasma NfL is associated with the APOE ε4 allele, brain imaging measurements of neurodegeneration, and lower recall memory scores in cognitively unimpaired late-middle-aged and older adults. Alzheimer's Research & Therapy. 2023;15(74). Ingannato A, Bagnoli S, Mazzeo S, Giacomucci G, Bessi V, Ferrari C, et al. Plasma GFAP, NfL and pTau 181 detect preclinical stages of dementia. Frontiers in Endocrinology. 2024;15. Langella S, Bonta K, Chen Y, Su Y, Vasquez D, Aguillon D, et al. Impact of APOE ε4 and ε2 on plasma neurofilament light chain and cognition in autosomal dominant Alzheimer’s disease. Alzheimer's Research & Therapy. 2024;16(208). Varma VR, An Y, Kac PR, Bilgel M, Moghekar A, Loeffler T, et al. Longitudinal progression of blood biomarkers reveals a key role of astrocyte reactivity in preclinical Alzheimer’s disease. medRxiv. 2024. Yakoub Y, Ashton NJ, Strikwerda-Brown C, Montoliu-Gaya L, Karikari TK, Kac PR, et al. Longitudinal blood biomarker trajectories in preclinical Alzheimer's disease. Alzheimer's & Dementia. 2023;19(12):5620–31. Snellman A, Ekblad LL, Ashton NJ, Karikari TK, Lantero-Rodriguez J, Pietilä E, et al. Head-to-head comparison of plasma p-tau181, p-tau231 and glial fibrillary acidic protein in clinically unimpaired elderly with three levels of APOE4-related risk for Alzheimer's disease. Neurobiology of Disease. 2023;183(106175). Snellman A, Ekblad LL, Tuisku J, Koivumäki M, Ashton NJ, Lantero-Rodriguez J, et al. APOE ε4 gene dose effect on imaging and blood biomarkers of neuroinflammation and beta-amyloid in cognitively unimpaired elderly. Alzheimer's Research & Therapy. 2023;15(71). Cooper JG, Ghodsi M, Stukas S, Leach S, Brooks-Wilson A, Wellington CL. APOE ε4 carrier status modifies plasma p‐tau181 concentrations in cognitively healthy super‐seniors. Alzheimer's & Dementia. 2024;20(6):4373–80. Xu Y, Qiao M, Gunasekaran TI, Gu Y, Reyes-Dumeyer D, Piriz A, et al. Evaluating polygenic risk score prediction performance for Alzheimer's disease in a population-based Hispanic cohort using single- and multi-ancestry models. The Lancet Regional Health - Americas. 2025;49:101198. Stocker H, Trares K, Beyer L, Perna L, Rujescu D, Holleczek B, et al. Alzheimer’s polygenic risk scores, APOE, Alzheimer’s disease risk, and dementia-related blood biomarker levels in a population-based cohort study followed over 17 years. Alzheimer's Research & Therapy. 2023;15:129. Lumsden AL, Mulugeta A, Zhou A, Hyppönen E. Apolipoprotein E (APOE) genotype-associated disease risks: a phenome-wide, registry-based, case-control study utilising the UK Biobank. eBioMedicine. 2020;59:102954. Daunt P, Ballard CG, Creese B, Davidson G, Hardy J, Oshota O, et al. Polygenic Risk Scoring is an Effective Approach to Predict Those Individuals Most Likely to Decline Cognitively Due to Alzheimer's Disease. The Journal of Prevention of Alzheimer's Disease. 2021;8(1):78–83. Kasuga K, Kikuchi M, Kikkawa-Saito E, Tsukie T, Ishiguro T, Miyashita A, et al. Evaluation of plasma p-tau217 biomarkers in detecting amyloid pathology and predicting cognitive outcomes: Observations from Japanese Alzheimer's disease neuroimaging initiative cohort. The Journal of Prevention of Alzheimer's Disease. 2026;13(4):100502. Bilgel M, An Y, Walker KA, Moghekar AR, Ashton NJ, Kac PR, et al. Longitudinal changes in Alzheimer's-related plasma biomarkers and brain amyloid. Alzheimer's & Dementia. 2023;19(10):4335–45. Li Y, Yen D, Hendrix RD, Gordon BA, Dlamini S, Barthélemy NR, et al. Timing of Biomarker Changes in Sporadic Alzheimer's Disease in Estimated Years from Symptom Onset. Annals of Neurology. 2024;95(5):951–65. Shaaban CE, Jia Y, Chang C-CH, Ganguli M. Independent and joint effects of vascular and cardiovascular risk factor pairs for risk of all-cause dementia: a prospective population-based study. International Psychogeriatrics. 2019;31(10):1421–32. Sillanpää MJ. Overview of techniques to account for confounding due to population stratification and cryptic relatedness in genomic data association analyses. Heredity. 2010;106(4):511–19. Willemsen G, Geus EJCd, Bartels M, Beijsterveldt CEMTv, Brooks AI, Burk GFE-v, et al. The Netherlands Twin Register biobank: a resource for genetic epidemiological studies. Twin Research and Human Genetics. 2010;13(3):231–45. Ligthart L, Beijsterveldt CEMv, Kevenaar ST, Zeeuw Ed, Bergen Ev, Bruins S, et al. The Netherlands Twin Register: Longitudinal Research Based on Twin and Twin-Family Designs. Twin Research and Human Genetics. 2019;22(6):623–36. Verberk IMW, Misdorp EO, Koelewijn J, Ball AJ, Blennow K, Dage JL, et al. Characterization of pre-analytical sample handling effects on a panel of Alzheimer's disease–related blood-based biomarkers: Results from the Standardization of Alzheimer's Blood Biomarkers (SABB) working group. Alzheimer's & Dementia. 2022;18(8):1484–97. Konijnenberg E, Carter SF, Kate MT, Braber Ad, Tomassen J, Amadi C, et al. The EMIF-AD PreclinAD study: study design and baseline cohort overview. Alzheimer's Research & Therapy. 2018;10(1):75. Ehli EA, Abdellaoui A, Fedko IO, Grieser C, Nohzadeh-Malakshah S, Willemsen G, et al. A method to customize population-specific arrays for genome-wide association testing. European Journal of Human Genetics. 2017;25(2):267–70. Lee SJvd, Wolters FJ, Ikram MK, Hofman A, Ikram MA, Amin N, et al. The effect of APOE and other common genetic variants on the onset of Alzheimer's disease and dementia: a community-based cohort study. Lancet Neurology. 2018;17(5):434–44. Bellenguez C, Küçükali F, Jansen IE, Kleineidam L, Moreno-Grau S, Amin N, et al. New insights into the genetic etiology of Alzheimer’s disease and related dementias. Nature Genetics. 2022;54(4):412–36. Ware EB, Faul JD, Mitchell CM, Bakulski KM. Considering the APOE locus in Alzheimer's disease polygenic scores in the Health and Retirement Study: a longitudinal panel study. BMC Medical Genomics. 2020;13:164. Vilhjálmsson BJ, Yang J, Finucane HK, Gusev A, Lindström S, Ripke S, et al. Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk Scores. American Journal of Human Genetics. 2015;97(4):576–92. Jacobs T, O'Brien C, Figueredo L, Gogola A, Gaggi NL, Hurwitz B, et al. Body mass index and blood volume influence plasma biomarkers and positron emission tomography classification in preclinical Alzheimer's disease. Alzheimer's & Dementia. 2025;21(10):e70765. Hermesdorf M, Esselmann H, Morgado B, Jahn-Brodmann A, Herrera-Rivero M, Wiltfang J, et al. The association of body mass index and body composition with plasma amyloid beta levels. Brain Communications. 2023;5(5):fcad263. Gasparini F, Valletta M, Vetrano DL, Beridze G, Rizzuto D, Calderón-Larrañaga A, et al. Kidney Function, Alzheimer Disease Blood Biomarkers, and Dementia Risk in Community-Dwelling Older Adults. Neurology. 2026;106(1):e214446. Rigby RA, Stasinopoulos DM. Using the Box-Cox t distribution in GAMLSS to model skewness and kurtosis. Statistical Modelling. 2006;6(3):209–29. Additional Declarations There is NO Competing Interest. Supplementary Files natmedRRoussetPlasmabiomarkersAPOESupplements.docx eMaterials Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8426026","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":636733867,"identity":"b2547b7a-fdd7-4f04-bed9-293bb1698f7b","order_by":0,"name":"Rebecca Rousset","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABB0lEQVRIiWNgGAWjYJADxgdQhgGUZm4goIOZGaRUAkkLI0EtbBJEaTFvYH/AXFFx2F6+vf9Yxc+cujp+6cMbP91s2ybPIN2IVYvMAYYExjNnDiduOHOY7WbvtsMSkn1pxdK5bbcNG2QOYtUCdMcBxsa2wwkGEslsN3i3HZAwOMNjANKSwCCRiEMLYwNIi738/MdshX+31UnYn+Ex/o1fCzMDSAtjww1mNmbebcwSBjw8ZvhtYWZjONhwJh3ol2RjadlthyVnnGErs845d9uwDZcW9vaHDxsqrIEhdvDhx7fb6vj5e5g3384puy3PL5F8AJsWoLsYsEswsGEXHgWjYBSMglFABAAA6jlbezO8r1sAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-9355-7286","institution":"Amsterdam UMC, Location VUMC","correspondingAuthor":true,"prefix":"","firstName":"Rebecca","middleName":"","lastName":"Rousset","suffix":""},{"id":636733868,"identity":"ec6fba8c-7fb6-4232-a599-3a5ee740e097","order_by":1,"name":"Anouk den Braber","email":"","orcid":"","institution":"Amsterdam UMC, location VUmc","correspondingAuthor":false,"prefix":"","firstName":"Anouk","middleName":"den","lastName":"Braber","suffix":""},{"id":636733870,"identity":"a716b363-554b-408d-8478-309874e271ea","order_by":2,"name":"Bram Bongers","email":"","orcid":"https://orcid.org/0009-0006-0122-4243","institution":"Department of Laboratory Medicine, Neurochemistry Laboratory, Amsterdam, UMC location Vrije Universiteit Amsterdam, Boelelaan 1117, Amsterdam 1081 HV, The Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Bram","middleName":"","lastName":"Bongers","suffix":""},{"id":636733872,"identity":"a77e6f6d-8532-483e-b7cf-647e74c9adf5","order_by":3,"name":"Lynn Boonkamp","email":"","orcid":"","institution":"Amsterdam University Medical Centers","correspondingAuthor":false,"prefix":"","firstName":"Lynn","middleName":"","lastName":"Boonkamp","suffix":""},{"id":636733874,"identity":"14c42897-506c-41a5-a598-4367d168e9d8","order_by":4,"name":"David Wilson","email":"","orcid":"","institution":"Quanterix Corporation","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Wilson","suffix":""},{"id":636733877,"identity":"2aecb08a-265d-4f40-a86b-9b9fcc75cb5b","order_by":5,"name":"Lannie Ligthart","email":"","orcid":"https://orcid.org/0000-0002-6570-3319","institution":"Vrije Universiteit Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Lannie","middleName":"","lastName":"Ligthart","suffix":""},{"id":636733878,"identity":"b6f9a835-a97c-49b7-942a-69ad0933ec2d","order_by":6,"name":"René Pool","email":"","orcid":"https://orcid.org/0000-0001-5579-0933","institution":"Department of Biological Psychology, Vrije Universiteit Amsterdam, 1081 BT Amsterdam, The Netherlands","correspondingAuthor":false,"prefix":"","firstName":"René","middleName":"","lastName":"Pool","suffix":""},{"id":636733881,"identity":"abfc95c6-78ca-44c0-afeb-e4eeb64c1a69","order_by":7,"name":"Pieter Jelle Visser","email":"","orcid":"","institution":"Maastricht University","correspondingAuthor":false,"prefix":"","firstName":"Pieter","middleName":"Jelle","lastName":"Visser","suffix":""},{"id":636733883,"identity":"701407c2-2f54-43c3-8684-adf114283524","order_by":8,"name":"Sophie van der Landen","email":"","orcid":"","institution":"Alzheimer Center Amsterdam, Neurology, Vrije Universiteit Amsterdam, Amsterdam UMC location VUmc, Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Sophie","middleName":"van der","lastName":"Landen","suffix":""},{"id":636733884,"identity":"5e1aa56d-4709-4df0-897a-17115b1dc67d","order_by":9,"name":"Eco de Geus","email":"","orcid":"https://orcid.org/0000-0001-6022-2666","institution":"Vrije Universiteit Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Eco","middleName":"","lastName":"de Geus","suffix":""},{"id":636733885,"identity":"89c43aad-aeea-46a4-8fc9-3c87f9d0d36d","order_by":10,"name":"Charlotte Teunissen","email":"","orcid":"https://orcid.org/0000-0002-4061-0837","institution":"Neurochemistry Laboratory and Biobank","correspondingAuthor":false,"prefix":"","firstName":"Charlotte","middleName":"","lastName":"Teunissen","suffix":""}],"badges":[],"createdAt":"2025-12-22 14:35:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8426026/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8426026/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109102178,"identity":"45f246a6-cf5b-4755-ab7d-28e29e15eca1","added_by":"auto","created_at":"2026-05-12 14:31:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":766455,"visible":true,"origin":"","legend":"\u003cp\u003ePlasma phosphorylated tau-217 (A), amyloid-β-42/40 (B), neurofilament light (C), and glial fibrillary acidic protein (D) over age, stratified by \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership. \u003cem\u003eAPOE\u003c/em\u003e = Apolipoprotein gene. The regression lines represent the association between the biomarker concentrations and age. Points represent individual biomarker concentrations and are coloured in blue for participants who are not \u003cem\u003eAPOE\u003c/em\u003e ε4-carriers and in orange for participants who are \u003cem\u003eAPOE\u003c/em\u003e ε4-carriers.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8426026/v1/716bfc58ac3eeedcc3bc66d1.png"},{"id":109102206,"identity":"2c6ff118-cb15-40e9-8d60-d4a7e93ccce8","added_by":"auto","created_at":"2026-05-12 14:31:35","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":550491,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of APOE ε4 carriership on plasma phosphorylated tau-217, amyloid-β-42/40, neurofilament light, and glial fibrillary acidic protein, stratified by sex. A: p-tau217 in males; B: Aβ-42/40 in males; C: NfL in males; D: GFAP in males; E: p-tau217 in females; F: Aβ-42/40 in females; G: NfL in females; H: GFAP in females. APOE: Apolipoprotein gene. The regression lines represent the association between the biomarker concentrations and age. Points represent individual biomarker concentrations and are coloured in light green (males) or orange (females) for participants who are not \u003cem\u003eAPOE\u003c/em\u003e ε4-carriers and in dark blue (males) or dark red (females) for participants who are \u003cem\u003eAPOE\u003c/em\u003e ε4-carriers.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8426026/v1/e0ef2c1028143f8cef95cba9.png"},{"id":109102205,"identity":"e99b837b-7e15-4528-9396-70f1bc278ea8","added_by":"auto","created_at":"2026-05-12 14:31:35","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":304965,"visible":true,"origin":"","legend":"\u003cp\u003eBiomarker concentrations over age in the NTR cohort and EMIF-AD longitudinal cohort, stratified by APOE ε4 carriership. A: p-tau217. B: Aβ-42/40. C: NfL. D: GFAP. Solid lines are derived from cross-sectional data from the NTR, while dashed lines are derived from longitudinal data from the EMIF-AD longitudinal cohort.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8426026/v1/b3faf5bfea4c5bd9f0ca7351.png"},{"id":109102124,"identity":"b34a8fef-4ff0-4e3f-a660-00ceac1fffba","added_by":"auto","created_at":"2026-05-12 14:31:16","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":372003,"visible":true,"origin":"","legend":"\u003cp\u003eBiomarker concentrations over age in the NTR cohort and EMIF-AD longitudinal cohort, stratified by APOE ε4 carriership and Aβ status (EMIF-AD only). A: p-tau217. B: Aβ-42/40. C: NfL. D: GFAP. Solid lines are derived from cross-sectional data from the NTR, while dashed lines are derived from longitudinal data from the EMIF-AD longitudinal cohort.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8426026/v1/ac147865269421f82b2f4fc0.png"},{"id":109204619,"identity":"b23c62e2-af65-4807-92d4-e4af18c93004","added_by":"auto","created_at":"2026-05-13 15:01:29","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2009989,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8426026/v1/be375cf6-ced0-4f46-b6e1-5723c1990a72.pdf"},{"id":109102153,"identity":"6a4088cf-0998-4855-87d9-eff65b7ab776","added_by":"auto","created_at":"2026-05-12 14:31:23","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1820699,"visible":true,"origin":"","legend":"eMaterials","description":"","filename":"natmedRRoussetPlasmabiomarkersAPOESupplements.docx","url":"https://assets-eu.researchsquare.com/files/rs-8426026/v1/a74f60eabb06815dcc1d28b4.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"APOE ε4 carriership status enhances age-related changes in Alzheimer’s disease plasma biomarkers concentrations","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eAlzheimer\u0026rsquo;s disease (AD) is a progressive neurodegenerative disorder, the primary cause of dementia and a leading cause of death worldwide (1, 2). The preclinical stage of AD, during which the pathology develops without measurable cognitive changes, can last up to 20 years, and can start as early as age 40 (3, 4). Many risk factors for AD have been identified, the main ones being age, with AD incidence increasing exponentially in those over 65 years (2, 5); sex, with about two thirds of AD cases being female (2, 6); and the Apolipoprotein E (\u003cem\u003eAPOE\u003c/em\u003e) \u0026epsilon;4 allele (5-7). Additionally, these different AD risk factors have been shown to interact in their effect on overall AD risk and onset (8).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003eAPOE\u003c/em\u003e gene codes the apoE protein, which is involved in lipid metabolism and cholesterol transport (9). There are three main alleles of the \u003cem\u003eAPOE\u003c/em\u003e gene, namely \u0026epsilon;2, \u0026epsilon;3, and \u0026epsilon;4. The \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 allele has a population frequency of 13.7% (10, 11), and exerts a dose-dependent risk effect on lifetime risk of developing AD (12, 13). Compared to the \u0026epsilon;3\u0026epsilon;3 genotype, heterozygous and homozygous carriership of the \u0026epsilon;4 allele confer a 2- to 3-fold and up to 13-fold increased risk of developing AD, respectively (7). \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 also partially moderates the effect of age on AD risk and onset, with onset of clinical AD being 8 years younger in heterozygous \u0026epsilon;4 carriers and 16 years younger in homozygous \u0026epsilon;4 carriers compared to non-carriers (12). Sex differences in the effect of \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003e\u0026epsilon;4 have also been reported, with the increased AD risk conferred by \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 being greater in females under 70 years old (6).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePlasma biomarkers for diagnosis of AD include phosphorylated tau-217 (p-tau217) (14)\u0026nbsp;and the amyloid-\u0026beta; 42/40 (A\u0026beta;-42/40) ratio\u0026nbsp;(15-17). Plasma p-tau217 concentrations begin to increase in the very early disease stages, before measurable changes in cognition\u0026nbsp;(18). At the clinical stage, plasma p-tau217 can accurately differentiate between AD-dementia and other forms of dementia, such as frontotemporal dementia, and is able to detect amyloid co-pathology in Lewy body disorders\u0026nbsp;(14, 19, 20). For A\u0026beta;-42/40, decreases in plasma concentration ratio are associated with AD pathology in both early and more advanced disease stages\u0026nbsp;(16, 17). Other biomarkers that are elevated in AD include neurofilament light (NfL), a broad marker of neurodegeneration, and glial fibrillary acidic protein (GFAP), a marker of astrocyte reactivity\u0026nbsp;(15, 21). NfL is significantly elevated in clinical AD and other dementia syndromes\u0026nbsp;(22), while plasma GFAP concentrations are elevated in early AD stages\u0026nbsp;(16, 23)\u0026nbsp;and prognostic of atrophy or cognitive decline\u0026nbsp;(24, 25).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePrevious studies have reported an association between carriership of the \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 allele and elevated concentrations of plasma p-tau217 (26-29), NfL (30-32), and GFAP (33, 34), and lower plasma A\u0026beta;-42/40 (34). However, findings for plasma NfL, GFAP, and A\u0026beta;-42/40 have been inconsistent across studies (31, 33-37), possibly due to insufficient power related to small sample sizes, or from not accounting for sex or age effects and the complex interaction between age, sex, and \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 carriership as AD risk factors (8).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn addition to \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4, the association between other genetic risk factors for AD and plasma biomarker concentrations may also be informative. Polygenic risk scores for AD (PRS\u003csub\u003eAD\u003c/sub\u003e) excluding the \u003cem\u003eAPOE\u003c/em\u003e gene have previously been reported to weakly associate with plasma p-tau181, A\u0026beta;-42/40, and GFAP concentrations (38, 39), potentially reflecting of AD genetic risk beyond \u003cem\u003eAPOE\u003c/em\u003e. A better understanding of how demographic and genetic risk factors for AD relate to AD plasma biomarker concentrations would facilitate the understanding of these plasma biomarkers in clinical practice and research.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe aim of this study was to provide insight on the complex interaction between the AD risk factors \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003e\u0026epsilon;4 carriership, age, and sex on concentrations of plasma p-tau217, A\u0026beta;-42/40, NfL, and GFAP. This investigation was conducted in the Netherlands Twin Registry (NTR) cohort, consisting of 5913 cognitively normal participants spanning a large age range of 30 to 80 years old, enabling us to accurately map the genetic effects on plasma biomarkers across life, and to accurately estimate the age at which biomarker concentrations begin to differ between \u0026epsilon;4 carriers and non-carriers in both males and females. Using data from a longitudinal study of older individuals (age 50 to 95), EMIF-AD, where longitudinal data on plasma biomarkers and A\u0026beta; positron emission tomography (PET) were collected, we explored whether changes in plasma biomarker concentrations in \u0026epsilon;4 carriers and non-carriers aligned with the presence of AD pathology. Additionally, we tested whether a PRS\u003csub\u003eAD\u003c/sub\u003e that excluded the\u0026nbsp;\u003cem\u003eAPOE\u003c/em\u003e gene region associated with plasma biomarker change over age.\u0026nbsp;\u003cbr clear=\"all\"\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"RESULTS","content":"\n\u003ch3\u003e1. Cohort description\u003c/h3\u003e\n\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e1.1. NTR demographics\u003c/h2\u003e \u003cp\u003eA total of 5913 participants with no reported cognitive decline was included from the NTR cohort (see Methods). The majority of participants had the ε3ε3 genotype (n\u0026thinsp;=\u0026thinsp;3377, 57.1%), followed by the ε3ε4 genotype (n\u0026thinsp;=\u0026thinsp;1459, 24.7%) and the ε2ε3 genotype (n\u0026thinsp;=\u0026thinsp;729, 12.3%). The remaining genotypes were much less frequent, with ε2ε4 present in 170 individuals (2.88%), ε4ε4 in 136 individuals (2.30%), and ε2ε2 in 42 individuals (0.71%). This amounted to 1765 individuals with at least one ε4 allele (29.8%). These proportions match those previously reported in white populations (40). PRS\u003csub\u003eAD\u003c/sub\u003e values were normally distributed, and no outliers were identified (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNTR cohort demographic information is reported in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Overall, ε4 carriers were significantly younger than non-carriers (mean age difference: 1 year, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.004). There were no differences in proportions of females (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.87), body mass index (BMI) (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.83) or estimated glomerular filtration rate (eGFR) (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.10) between the groups. E4 carriers had significantly higher p-tau217 (mean difference: 0.002 pg/mL; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003) and significantly lower Aβ-42/40 (mean difference: 0.002; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.006). There were no differences in NfL (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.51) and GFAP (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.56) between ε4 carriers and non-carriers. Cohort demographic per PRS\u003csub\u003eAD\u003c/sub\u003e score group (excluding \u003cem\u003eAPOE\u003c/em\u003e), are presented in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. There were no differences in age, proportion of females, proportion of ε4 carriers, or any of the biomarker concentrations between the low, medium, and high PRS\u003csub\u003eAD\u003c/sub\u003e groups (p\u0026thinsp;\u0026gt;\u0026thinsp;0.02 for all).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e1.2. EMIF-AD demographics\u003c/h2\u003e \u003cp\u003eFor the EMIF-AD cohort, 202 participants were included, of which 77 also had an earlier plasma biomarker measurement as part of the NTR cohort. Samples were collected at baseline biennially over 6 years/3 follow-up points. Of these 202 participants, 195 (96.5%) had at least one available Aβ PET measurement, 66 (32.8%) were ε4 carriers (6 ε2ε4, 55 ε3ε4, 5 ε4ε4) and 135 (67.2%) were non-carriers (2 ε2ε2, 21 ε2ε3, 112 ε3ε3). Participants were further categorized as Aβ PET negative (Aβ PET negative at all Aβ PET assessments) and Aβ PET positive (Aβ PET positive at any of the Aβ PET assessments). Stratifying by Aβ PET status resulted in 109 Aβ PET negative non-carriers (56%), 40 Aβ PET negative ε4 carriers (21%), 20 Aβ PET positive non-carriers (10%), and 26 Aβ PET positive ε4 carriers (13%). EMIF-AD cohort demographics are reported in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. At baseline, ε4 carriers were significantly younger (mean difference: 4 years, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) than non-carriers and this group contained a higher proportion of Aβ PET positive participants (39.4% vs 14.8%, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). There were no other significant differences (Table S2).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDemographic information for the NTR cohort.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003en\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eTotal\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eε4 non-carriers\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eε4 carriers\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eGroup difference\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5913\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4148\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1765\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge [year]\u003c/p\u003e \u003cp\u003eMean [range]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e49.8 [30\u0026ndash;80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e50.1 [30\u0026ndash;80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e49.1 [30\u0026ndash;80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.004\u003c/b\u003e\u003csup\u003e\u003cb\u003e\u0026plusmn;\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex [-]\u003c/p\u003e \u003cp\u003eFemale (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3793 (64.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2664 (64.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1129 (64.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.87*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI [kg/m\u003csup\u003e2\u003c/sup\u003e]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25.7 (4.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e25.8 (4.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25.7 (4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.83\u003csup\u003e\u003cb\u003e\u0026plusmn;\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eeGFR [mL/min/1.73m\u003csup\u003e2\u003c/sup\u003e]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e81.2 (14.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e81.0 (14.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e81.7 (15.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.10\u003csup\u003e\u003cb\u003e\u0026plusmn;\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ep-tau217 [pg/mL]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.030 (0.023)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.030 (0.023)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.032 (0.024)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003csup\u003e\u003cb\u003e\u0026plusmn;\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAβ-42/40 [-]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.067 (0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.068 (0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.066 (0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.006\u003c/b\u003e\u003csup\u003e\u003cb\u003e\u0026plusmn;\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNfL [pg/mL]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e9.91 (9.69)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9.96 (9.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.79 (9.31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.51\u003csup\u003e\u0026plusmn;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGFAP [pg/mL]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e68.3 (56.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e68.0 (51.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e69.0 (66.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.56\u003csup\u003e\u0026plusmn;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003csup\u003e\u0026plusmn;\u003c/sup\u003e t-test ; * Chi-square\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDemographic information for the longitudinal EMIF-AD cohort, per measurement wave.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"11\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e \u003cp\u003e\u003cem\u003eε4 non-carriers\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"5\" nameend=\"c11\" namest=\"c7\"\u003e \u003cp\u003e\u003cem\u003eε4 carriers\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNTR\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;46)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBL\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;135)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFU1\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;107)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFU2\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;86)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFU3\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;73)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNTR\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;31)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBL\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;66)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eFU1\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;66)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eFU2\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;57)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eFU3\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;49)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge [year]\u003c/p\u003e \u003cp\u003eMean [range]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e59.5\u003c/p\u003e \u003cp\u003e[50\u0026ndash;75]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e72.2\u003c/p\u003e \u003cp\u003e[61\u0026ndash;94]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e73.0\u003c/p\u003e \u003cp\u003e[63\u0026ndash;95]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e73.8\u003c/p\u003e \u003cp\u003e[65\u0026ndash;94]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.0\u003c/p\u003e \u003cp\u003e[67\u0026ndash;86]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e57.3\u003c/p\u003e \u003cp\u003e[51\u0026ndash;72]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003cp\u003e[60\u0026ndash;83]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e69.4\u003c/p\u003e \u003cp\u003e[63\u0026ndash;86]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e71.1\u003c/p\u003e \u003cp\u003e[65\u0026ndash;88]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e73.0\u003c/p\u003e \u003cp\u003e[67\u0026ndash;90]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex [-]\u003c/p\u003e \u003cp\u003eFemale (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27\u003c/p\u003e \u003cp\u003e(58.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e86\u003c/p\u003e \u003cp\u003e(63.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64\u003c/p\u003e \u003cp\u003e(59.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e51\u003c/p\u003e \u003cp\u003e(59.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e46\u003c/p\u003e \u003cp\u003e(63.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e14 (45.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e30\u003c/p\u003e \u003cp\u003e(45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e30\u003c/p\u003e \u003cp\u003e(45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e24\u003c/p\u003e \u003cp\u003e(42.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e19\u003c/p\u003e \u003cp\u003e(38.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMMSE score\u003c/p\u003e \u003cp\u003eMedian\u003c/p\u003e \u003cp\u003e[range]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e29\u003c/p\u003e \u003cp\u003e[24\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29\u003c/p\u003e \u003cp\u003e[25\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e29\u003c/p\u003e \u003cp\u003e[24\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e29\u003c/p\u003e \u003cp\u003e[26\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e29 [25\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e29 [25\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e29\u003c/p\u003e \u003cp\u003e[26\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e29\u003c/p\u003e \u003cp\u003e[24\u0026ndash;30]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAβ PET status\u003c/p\u003e \u003cp\u003eAβ+, %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003cp\u003e(6.52%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003cp\u003e(14.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003cp\u003e(16.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e13\u003c/p\u003e \u003cp\u003e(15.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9\u003c/p\u003e \u003cp\u003e(12.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e14\u003c/p\u003e \u003cp\u003e(45.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e26\u003c/p\u003e \u003cp\u003e(39.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e26\u003c/p\u003e \u003cp\u003e(39.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e26\u003c/p\u003e \u003cp\u003e(38.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e19\u003c/p\u003e \u003cp\u003e(38.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ep-tau217 [pg/mL]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.031 (0.021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.041 (0.024)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.040 (0.021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.042 (0.028)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.040 (0.020)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.041 (0.025)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.047 (0.031)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003cp\u003e(0.045)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003cp\u003e(0.053)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003cp\u003e(0.041)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAβ-42/40 [-]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.071 (0.014)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.061 (0.011)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.060 (0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.060 (0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.059 (0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.069 (0.016)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.060 (0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.058 (0.011)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.059 (0.011)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.059 (0.009)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNfL [pg/mL]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.0 (8.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18.0 (9.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18.2 (9.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18.0 (9.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e19.4 (8.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8.75 (3.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e15.4 (10.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e16.8 (18.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e16.8 (7.89)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e20.3 (11.6)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGFAP [pg/mL]\u003c/p\u003e \u003cp\u003eMean (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e72.9 (36.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e115 (55.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e115\u003c/p\u003e \u003cp\u003e(51.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e121 (67.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e127\u003c/p\u003e \u003cp\u003e(51.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e74.1 (47.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e112\u003c/p\u003e \u003cp\u003e(71.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e116\u003c/p\u003e \u003cp\u003e(80.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e132\u003c/p\u003e \u003cp\u003e(98.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e132\u003c/p\u003e \u003cp\u003e(74.7)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eParticipants were followed over a maximum of five timepoints: Netherlands Twin Register (NTR), which took place prior to the EMIF-AD study and in which only a fraction of EMIF-AD participants also participated; Baseline (BL), the baseline assessment for EMIF-AD; Follow-up 1 (FU1, two years after BL) of the EMIF-AD; Follow-up 2 (FU2, four years after BL) of the EMIF-AD; Follow-up 3 (FU3, six years after BL) of the EMIF-AD.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2. Age-dependent effect of APOE ε4 on plasma biomarker concentrations\u003c/h3\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Main cohort (NTR)\u003c/h2\u003e \u003cp\u003ePlasma biomarker concentrations over age are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. We first applied GALMSS. A main effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership was observed for p-tau217 (LRT\u0026thinsp;=\u0026thinsp;17.5, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and Aβ-42/40 (LRT\u0026thinsp;=\u0026thinsp;20.0, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with p-tau217 concentrations being higher and Aβ-42/40 lower in ε4 carriers compared to non-carriers. An interaction effect between age and APOE ε4 carriership was observed for p-tau217 (LRT\u0026thinsp;=\u0026thinsp;50.3, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), Aβ-42/40 (LRT\u0026thinsp;=\u0026thinsp;33.2, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and GFAP (LRT\u0026thinsp;=\u0026thinsp;18.5, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.007). There was no APOE ε4 carriership effect on NfL (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026ge;\u0026thinsp;0.76 for all). Using GAM, which allows the computation of confidence intervals, we were further able to determine the age at which biomarker concentrations began to significantly diverge between ε4 carriers and non-carriers. This divergence age between ε4 carriers and non-carriers was determined to be 53.2 for Aβ-42/40, followed by 58.8 for p-tau217, and 62.8 for GFAP (Figure S2), with the magnitude of difference being most pronounced for p-tau217. At age 50, mean p-tau217 concentrations were virtually the same between ε4 carriers and non-carriers (mean difference\u0026thinsp;=\u0026thinsp;0.0001 pg/mL), while at age 75 mean concentrations were 0.02 pg/mL higher in ε4 carriers. The difference in ratio of Aβ-42/40 increased from 0.001 at age 50 to 0.006 at age 75. Differences in GFAP concentrations between ε4 carriers and non-carriers increased from 2.1 pg/mL at age 50 to 14.5 pg/mL at 75. Model specifications are reported in Table S3.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhen considering heterozygous and homozygous ε4 carriers separately, significant age-dependent effects of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership were again found for p-tau217, Aβ-42/40, and GFAP (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.005 for all) (Table S4). For all three biomarkers, concentrations appeared to start increasing between 5 and 10 years earlier for homozygous carriers compared to heterozygous carriers (Figure S3).\u003c/p\u003e \u003cp\u003eWe repeated the analyses in a subset of participants who were genetically unrelated. This subset consisted of 3624 individuals (ε4 carriers: 1075 (29.7%)). Results from this sensitivity analysis mirrored the results in the full NTR cohort, aside from the age by \u003cem\u003eAPOE\u003c/em\u003e interaction effect initially found for GFAP no longer being significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02) (Table S5).\u003c/p\u003e \u003cp\u003eFIGURE \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Stratification by sex\u003c/h2\u003e \u003cp\u003eConcentrations over age by sex are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Model specifications are reported in Table S6.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThere was a main effect of APOE ε4 carriership on p-tau217 in both males (LRT\u0026thinsp;=\u0026thinsp;6.89, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.009) and females (LRT\u0026thinsp;=\u0026thinsp;8.51, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.004), as well as an interaction effect between age and \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership in males (LRT\u0026thinsp;=\u0026thinsp;35.9, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) but not females (LRT\u0026thinsp;=\u0026thinsp;9.63, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.04), although a trend was observed. Using GAM analysis, the age at which concentrations began to significantly differ between ε4 carriers and non-carriers in males was determined to be 61.3 years (Figures S4\u0026ndash;S5). In males, p-tau217 concentrations began to increase with age for both ε4 carriers and non-carriers, but do so markedly more steeply in ε4 carriers. At age 50, the difference in p-tau217 concentrations between male ε4 carriers and non-carriers was less than 0.0001 pg/mL, while at age 75 p-tau217 concentrations were 0.03 pg/mL higher among ε4 carriers. Looking at heterozygous and homozygous ε4 carriers separately, a main \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership effect was found for both sexes, but the age by \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership interaction was only significant in males, where p-tau217 concentrations started increasing at an earlier age in homozygous carriers compared to heterozygous carriers (Table S7, Figures S6\u0026ndash;S7).\u003c/p\u003e \u003cp\u003e \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership also exerted a main effect on Aβ-42/40 in both males (LRT\u0026thinsp;=\u0026thinsp;10.5, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001) and females (LRT\u0026thinsp;=\u0026thinsp;10.9, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001), and also interacted with age (male: LRT\u0026thinsp;=\u0026thinsp;15.3, \u003cem\u003ep\u0026thinsp;=\u003c/em\u003e\u0026thinsp;0.004; females: LRT\u0026thinsp;=\u0026thinsp;16.1, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.004). Using GAM, the age at which Aβ-42/40 concentrations begin to differ between male ε4 carriers and non-carriers was calculated to be 52.7 years old (Figures S4\u0026ndash;S5). The GAM approach did not detect a demarcation age between female ε4 carriers and non-carriers. Difference in Aβ-42/40 increased from 0.001 at age 50 to 0.006 at age 75 in males and to 0.007 in females. When treating heterozygous and homozygous carriers separately (Table S7), concentrations began to decrease 5 to 10 years earlier for homozygous carriers (Figures S6\u0026ndash;S7).\u003c/p\u003e \u003cp\u003eIn line with the results in the full sample, \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership did not influence NfL concentrations in either males or females (\u003cem\u003ep\u0026thinsp;\u0026ge;\u003c/em\u003e\u0026thinsp;0.17 for both), nor did it have an interaction with age in either sex (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026ge;\u0026thinsp;0.44 for both). Interestingly, when evaluating heterozygous and homozygous ε4 carriers separately, a main effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership was found in males only (LRT\u0026thinsp;=\u0026thinsp;15.9, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with NfL concentrations being overall higher in homozygous carriers compared to heterozygous carriers and non-carriers.\u003c/p\u003e \u003cp\u003eFor GFAP, \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership did not exert a main effect in either males or females (\u003cem\u003ep\u0026thinsp;\u0026ge;\u003c/em\u003e\u0026thinsp;0.43 for both), and there no longer was an age-dependent effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 in either sexes (males: LRT\u0026thinsp;=\u0026thinsp;14.8, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02; females: LRT\u0026thinsp;=\u0026thinsp;5.62, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.45). An age-dependent effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership effect was observed for males only when handling heterozygous and homozygous carriers separately, showing a more linear increase in concentrations over age compared to the exponential increase previously observed (Figures S6\u0026ndash;S7).\u003c/p\u003e \u003cp\u003eFIGURE \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e3. Age-dependent effect of PRS on plasma biomarker concentrations\u003c/h3\u003e\n\u003cp\u003eAll results relating to the PRS\u003csub\u003eAD\u003c/sub\u003e (with and without \u003cem\u003eAPOE\u003c/em\u003e ε4 correction) are presented in the eMaterials (Tables S9\u0026ndash;S12/Figures S8\u0026ndash;S9). The PRS\u003csub\u003eAD\u003c/sub\u003e was not significantly associated with any of the biomarker concentrations in the NTR. In males only, a higher PRS\u003csub\u003eAD\u003c/sub\u003e was associated with higher p-tau217 concentrations at all ages (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). No age-dependent effect was found in either the full cohort or in the male and female subsamples. Adjusting for \u003cem\u003eAPOE\u003c/em\u003e ε4 did not lead to different results. When repeating the analysis in the subset of genetically unrelated individuals, results were similar to those from the NTR, though an age-independent effect on p-tau217 concentrations was additionally detected (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.008), with a higher PRS\u003csub\u003eAD\u003c/sub\u003e being associated with overall higher p-tau217 concentrations (Table S12).\u003c/p\u003e\n\u003ch3\u003e4. Additional analysis in the longitudinal EMIF-AD cohort with available Amyloid PET results\u003c/h3\u003e\n\u003cp\u003eConcentrations in the EMIF-AD cohort aligned well with concentrations from the NTR. At age 60, p-tau217 concentrations were significantly higher in ε4 carriers compared to non-carriers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). These results indicate an enduring ε4 carriership effect on p-tau217 concentrations in participants 60 and older. Other biomarker concentrations did not significantly differ between ε4 carriers and non-carriers at cross-sectional age 60. However, NfL concentrations were observed to increase more steeply over age in ε4 carriers than non-carriers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001). Unlike the results from the NTR cohort, rate of change in p-tau217, Aβ-42/40, and GFAP concentrations over age was not affected by \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026ge;\u0026thinsp;0.14 for all) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e/Figure S10\u0026ndash;S11), which is expected as EMIF-AD is an older cohort - monitoring started at ages above the ages where biomarkers started to differ in the NTR. Results are reported in Table S13.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eParticipants were next stratified on Aβ PET status. At age 60, cross-sectional differences were found for Aβ PET positive/ε4 carriers, who had higher p-tau217 concentrations than both Aβ PET positive/non-carriers and Aβ PET negative/non-carriers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for both). Aβ PET negative/ε4 carriers also had higher p-tau217 concentrations than Aβ PET positive/non-carriers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003). Longitudinally, p-tau217 concentrations were further observed to increase more steeply over age for Aβ PET positive/ε4 carriers than for both Aβ PET negative/ε4 carriers and Aβ PET negative/non-carriers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01 for both). Aβ PET positive/ε4 non-carriers were also observed to have steeper p-tau217 slopes over age compared to both Aβ PET negative/ε4 carriers and Aβ PET negative/non-carriers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for both). Together, these results show that steeper p-tau217 slopes over age are reflective of Aβ PET status, and therefore AD pathology, rather than \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership, while \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership appears to affect the timing of increase in p-tau217 concentrations, with concentrations increasing 20 to 25 years earlier in ε4 carriers compared to non-carriers. Though no cross-sectional differences in NfL concentrations were found at age 60, a steeper increase in NfL concentrations was found for Aβ PET negative/ε4 carriers compared to Aβ PET negative/non-carries (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002). There were no differences in Aβ-42/40 concentrations between any of the Aβ PET and ε4-carriership groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026ge;\u0026thinsp;0.11 for all) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e/Figure S12). Results are reported in Table S14.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFIGURE \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003c/p\u003e \u003cp\u003eFIGURE \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn this large cross-sectional cohort of cognitively normal individuals, covering a broad part of the age range where amyloid pathology changes begin to occur, we showed an effect of the \u003cem\u003eAPOE\u003c/em\u003e ε4 allele on plasma concentrations of p-tau217, Aβ-42/40, and GFAP, but not NfL. By taking into account age and sex, which are two major risk factors for AD, we were able to show a three-way age, sex, and \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership interaction on p-tau217 concentrations and the Aβ-42/40 ratio, as well as an interaction between age and \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership on GFAP concentrations, with the effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership being most pronounced for p-tau217. We observed that in ε4 carriers, the Aβ-42/40 ratio starts to decrease more steeply after age 50, p-tau217 concentrations increased more steeply after 55, and GFAP concentrations after 60, compared to non-carriers. Our supplementary analyses in a cohort containing longitudinal biomarker and Aβ PET information allowed us to show that increases in p-tau217 concentrations over age are due to early AD pathology, and exacerbated by ε4 carriership, which affects the timing of offset. Additionally, by stratifying by sex in our cross-sectional cohort, we were able to show a male-specific effect in the change in p-tau217 concentration over age, with carriership of the \u003cem\u003eAPOE\u003c/em\u003e ε4 allele being associated with steeper increase in p-tau217 from its inflection point, while in females the difference between ε4 carriers and non-carriers was age independent, although a trend was observed. For Aβ-42/40, we observed an age-dependent effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership in both sexes. No effect was observed on GFAP nor NfL in the sex-specific subsets. When handling heterozygous and homozygous ε4 carriers separately, we additionally observed male-specific dose-response effects of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership on both NfL (age-independent effect) and GFAP (age-dependent effect). The PRS\u003csub\u003eAD\u003c/sub\u003e did not show a strong association with any of the biomarkers.\u003c/p\u003e \u003cp\u003eOur results align with previous literature, where differences in p-tau217 concentrations between ε4 carriers and non-carriers were reported (26–29), though without accounting for age- or sex-dependent effects. Adding to this, we observe an age-dependent effect on Aβ-42/40 and GFAP concentrations in the NTR. Finally, while NfL concentrations have previously been reported to be more elevated in ε4 carriers compared to non-carriers (30, 31), we here report that homozygous ε4 carriers have consistently elevated NfL concentrations across their lifespan compared to heterozygous ε4 carriers and non-carriers. Discrepancies between our results and previous results on the association between \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership and plasma Aβ-42/40, NfL and GFAP (31, 33–37) may be due to small sample sizes or not adjusting for age and/or sex in previous studies. Our observation of a lack of effect of the PRS\u003csub\u003eAD\u003c/sub\u003e on the age-related slopes for Aβ-42/40, NfL, or GFAP is in line with previous research, which report that PRS\u003csub\u003eAD\u003c/sub\u003e associates more strongly with changes further along the disease continuum (39, 41), such as cognitive decline in individuals with mild cognitive impairment due to AD (41). We did however detect a male-specific age-independent effect of the PRS\u003csub\u003eAD\u003c/sub\u003e on p-tau217 concentrations in the NTR, and a similar effect in both sexes when performing the analysis in a genetically unrelated subset of the population. Such weak associations between the PRS\u003csub\u003eAD\u003c/sub\u003e and both p-tau217 and p-tau181 had previously been reported (38, 42). Here, we were able to demonstrate that the association between PRS\u003csub\u003eAD\u003c/sub\u003e and p-tau217 concentrations is independent of age, which is in contrast to the effect of APOE where the effects were exacerbated by age.\u003c/p\u003e \u003cp\u003ePlasma concentrations of AD biomarkers have been reported to change before the clinical onset of AD. A previous study on plasma p-tau217 concentrations over age in participants with and without mild cognitive impairment reported that after age 55, p-tau217 concentrations in participants with AD pathology (as defined by Aβ PET) increase 12 times faster than in participants without AD pathology (28). Using \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership as a proxy for early AD pathology, we here were able to show a similar pattern of steeper increase starting from around age 55 in ε4 carriers compared to non-carriers even in a cohort of cognitively normal participants only. This was further confirmed by comparing the plasma data in relation to Amyloid PET positivity in the longitudinal EMIF-AD cohort. Here, we observed that increases in p-tau217 concentrations over age are principally due to AD pathology, and not solely ε4 carriership. It is likely that differences in plasma p-tau217 concentrations between ε4 carriers and non-carriers in the NTR cohort were likewise due to developing AD pathology. Our results are in line with a recently published study, which reported that p-tau217 concentrations increase with increased AD pathological burden (29).\u003c/p\u003e \u003cp\u003eStratifying by sex further enabled us to show that the pattern and timing of p-tau217 increase differs between males and females. In males, increase in p-tau217 occurred after age 55 and seemed to be almost entirely due to \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership (and, by proxy, early AD pathology), and the increase in p-tau217 concentrations was much steeper in male ε4 carriers compared to female ε4 carriers. In females, an increase in p-tau217 concentrations independent of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership could be observed between age 50 and 55, before concentrations began to differ between the carriership groups, though this divergence between \u003cem\u003eAPOE\u003c/em\u003e carriership groups was only a trend.\u003c/p\u003e \u003cp\u003eThe association of plasma Aβ-42/40, NfL, and GFAP with Aβ positivity over age in pre-clinical AD populations has so far rarely been investigated, since such longitudinal cohorts covering a broad age range are not available. However, predictive models studying cognitive decline have been able to estimate that the plasma Aβ-42/40 ratio start decreasing between 14 and 41 years before clinical AD onset (43, 44). For NfL, concentration changes are detected up to 12 years prior to clinical onset (44), but are clearer and increase more steeply during and after onset of cognitive impairment (43). The same has been reported for GFAP concentrations (25, 43, 44). Here, in a cohort of cognitively normal individuals, we similarly observed that Aβ-42/40 concentrations are the first to begin to differ between \u003cem\u003eAPOE\u003c/em\u003e ε4 carriers and non-carriers, followed by GFAP 10 to 15 years later, though we did not detect difference in NfL concentrations. However, we did not determine if the decrease in Aβ-42/40 and increase in GFAP concentrations were due to age, AD pathology, or an \u003cem\u003eAPOE\u003c/em\u003e ε4 effect independent of AD pathology. Additionally, among Aβ PET negative individuals only, we showed a steeper increase in NfL concentrations over age in ε4 carriers compared to non-carriers. This may be a reflection of a dose-dependent effect of \u003cem\u003eAPOE\u003c/em\u003e ε4, since we observed higher NfL concentrations in homozygous carriers compared to both heterozygous carriers and non-carriers in the NTR cohort. This ε4 carriership effect independent of AD pathology, which we also did not observe in the NTR, may be a reflection of the higher vascular risk associated with \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership (45).\u003c/p\u003e \u003cp\u003eOne limitation is that our cohort consists of families, which may increase the risk of false positive results (46) – however, the results of our sensitivity analysis in a subset of non-genetically related individuals were highly comparable to the results of our primary analysis. Another limitation is that we did not formally test cognitive function, however we were able to exclude all participants who reported memory complaints. Thus, future studies should focus on longitudinal assessment of amyloid status and cognition, to further expand our understanding of p-tau217 as a predictive marker for future amyloid pathology. The strengths of our study were that our cohort spanned an age range of 50 years, allowing us to narrow down the age at which the rate of concentration change begins to differ between ε4 carriers and non-carriers. Our use of GAMLSS also allowed us to more accurately capture the non-linear relationship between age and the AD biomarker concentrations without data transformation. The GAM analysis further allowed us to determine the likely age of divergence in concentrations between ε4 carriers and non-carriers. Using a cohort with available longitudinal biomarker information and Aβ PET, we were also able to relate our \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership results to actual AD pathology.\u003c/p\u003e \u003cp\u003eTaken together, we demonstrated that \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership influences plasma concentrations of p-tau217, Aβ-42/40, and GFAP, but not NfL, in an age-dependent manner in cognitively normal individuals, as well as a sex-dependent manner for p-tau217 and Aβ-42/40. We observed that p-tau217, Aβ-42/40, and GFAP concentrations begin to differ between ε4 carriers and non-carriers 15 to 20 years before estimated mean age of clinical AD onset. In a longitudinal cohort, we were able to show that increases in p-tau217 concentrations over age are principally due to AD pathology, while carriership of the \u003cem\u003eAPOE\u003c/em\u003e ε4 allele causes in a younger age of increase onset. We were also able to demonstrate that PRS\u003csub\u003eAD\u003c/sub\u003e does not affect AD plasma biomarker concentrations, except for a male-specific effect on p-tau217. These results highlights the potential of AD plasma biomarkers, particularly p-tau217, for detection of early AD pathological changes, which in turn will have implications for inclusion in trials targeting AD.\u003c/p\u003e \n\n \n\n\n\u003ch3\u003e \u003c/h3\u003e\n\u003cdiv class=\"Heading\"\u003e\u003csub\u003e\u003c/sub\u003e \u003c/div\u003e \n\n\n\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e "},{"header":"Online Methods","content":"\u003ch3\u003e1. Cohorts\u003c/h3\u003e\u003ch2\u003e1.1. The Netherlands Twin Register cohort\u003c/h2\u003e\u003cp\u003eParticipants were included from the Netherland Twin Register (NTR) biobank, for which blood collection took place between 2004 and 2011 (47). Details on blood collection and storage procedures have been published elsewhere (47, 48). The NTR collects genotypic and longitudinal phenotypic data of Dutch twins and other multiples and their family members (47, 48). Although cognition is not formally assessed as part of the NTR, multiple surveys contain questions on memory problems (time of survey ranges from 22 days prior to biobank measurement to 9.3 years after biobank measurement, with a mean (SD) of 3.4 years after measurement (2 years)). At time of blood collection, participants also completed a general health survey, which provided information on any medical condition the participants may have or any medication they may be taking. All participants in this cohort were at least 30 years of age. Participants were eligible for inclusion if information on age, sex, and \u003cem\u003eAPOE\u003c/em\u003e genotype was available. Participants were excluded if they had reported having problematic or very problematic memory complaints on any NTR survey (n = 77), or if they reported having a diagnosis of dementia due to Alzheimer’s disease (n = 1), Parkinson’s disease (n = 6), extrapyramidal and movement disorders (n = 5), multiple sclerosis (n = 14), or speech disturbance (n = 1) (total n = 27) during the biobank-specific survey. Based on the age distribution, and to ensure reliable estimates, participants above 80 years of age were excluded (n = 43).\u003c/p\u003e\u003ch2\u003e1.2. The European Medical Information Framework for Alzheimer's disease cohort\u003c/h2\u003e\u003cp\u003eTo further strengthen our analyses, participants from the Dutch subset of the European Medical Information Framework for Alzheimer's disease (EMIF-AD) cohort were included (49, 50). The EMIF-AD cohort consists of 206 monozygotic twins followed longitudinally, with follow-up data being collected every 2 years. Cognition was assessed using the Mini Mental State Exam (MMSE), which resulted in 4 participants being excluded for having an MMSE score \u0026lt; 24 at any timepoint. For each participant, blood is collected at baseline and every follow-up point, while an Aβ [18F]flutemetamol-PET (Aβ PET) scan is performed at baseline and every other follow-up point. As part of the EMIF-AD cohort, participants were followed from 0 to 7.5 years (mean time: 4.8 years (SD = 2.5)). A subset of the EMIF-AD participants (n = 80) were also part of the NTR. These participants were followed from 8.5 to 17.5 years (mean time: 14.5 years (SD = 2.5)). The mean time between NTR participation and baseline EMIF-AD measurement was 9 years (SD = 2 years).\u003c/p\u003e\u003ch3\u003e2. Alzheimer’s disease biomarker measurements\u003c/h3\u003e\u003cp\u003eAll biomarker measurements were performed on the Simoa HD-X platform. Biomarker concentrations were measured using the same kits and on the same Simoa HD-X platforms for both the NTR and EMIF-AD cohorts. Concentrations of p-tau217 were measured using the ACC pTau217 Assay (reference #104740), and concentrations of NfL, GFAP, Aβ-42 and Aβ-40 were assessed using the Neurology-4-plex E Advantage kit (reference #103670). For p-tau217 in the NTR, the first 337 samples were measured in duplicate. Quality of the duplicate measurements was assessed (Figure S13), after which it was decided the remaining 5857 samples would be measured in single. In-house quality controls (QCs) were included in each run, at three levels (low, medium, and high concentrations). Additionally, kit QCs (also low, medium, and high concentrations) were included every five runs. Inter-assay coefficients of variation (CVs) were 9.1%. Seventy seven samples had concentrations below the lower limit of detection. For EMIF-AD, all samples were measured in duplicate over the course of 6 runs, with inter-assay CVs of 5.6% and intra-assay CVs of 7.1%. For measurements of NfL, GFAP, Aβ-42, and Aβ-40, two QCs (low, high) per biomarker were included in each run. In the NTR, inter-assay CVs were 4.5% for NfL, 4.6% for GFAP, 5.2% for Aβ-42, and 4.1% for Aβ-40. In the EMIF-AD cohort, inter-assay CVs were 5.5% for NfL, 5.0% for GFAP, 7.3% for Aβ42, and 5.0% for Aβ40.\u003c/p\u003e\u003cem\u003e3. APOE genotyping and PRS\u003c/em\u003e\u003cem\u003eAD\u003c/em\u003e\u003cem\u003ecalculation\u003c/em\u003e\u003cp\u003eThe same methods and platforms were used for \u003cem\u003eAPOE\u003c/em\u003e genotyping in both the NTR and EMIF-AD cohorts. DNA was extracted from peripheral blood samples. Participants were genotyped on the Affymetrix Axiom and Affymetrix 6 arrays (51). Participants’ APOE genotype was determined using imputed dosages of single nucleotide polymorphisms rs429358 (APOE ε4) and rs7412 (APOE ε2) (52). The PRS\u003csub\u003eAD\u003c/sub\u003e were calculated for NTR participants only based on the genome-wide association study by Bellenguez et al. (53) for AD, excluding APOE. For calculation of the PRS\u003csub\u003eAD\u003c/sub\u003e, we retained variants for which the effect allele frequency (EAF) was 0.01 ≤ EAF ≤ 0.99. Variant EAF and effect sizes were aligned with the NTR reference for the 1000 genomes variants. We excluded the APOE region by removing variants between base pair positions 45384477 and 45432606 on chromosome 19 (build 37/hg 19) from the above GWAS summary statistics, in keeping with Ware et al. (54). Discovery variants that were not part of this reference were discarded. The LDpred 0.9 software (55) was used to estimate the target linkage disequilibrium structure, using a selection of unrelated NTR individuals and selecting a set of well-imputed variants in the NTR sample. The PRS\u003csub\u003eAD\u003c/sub\u003e generated using the infinitesimal model were preserved for analysis.\u003c/p\u003e\u003ch3\u003e4. Amyloid-β status assessment\u003c/h3\u003e\u003cp\u003eIn EMIF-AD participants only, an Aβ PET scan was performed at baseline and follow-up 2 (mean time to follow-up 2 = 4.33 years (SD = 0.42)). The Aβ PET imaging protocol has been described in detail elsewhere (50). Based on the Aβ PET scan, participants were scored as either Aβ- (no significant AD pathology) or Aβ+ (significant AD pathology) by three trained readers following General Electric Healthcare guidelines. In the case of a disagreement, the consensus rating of two readers was used.\u003c/p\u003e\u003ch3\u003e5. Statistical analyses\u003c/h3\u003e\u003ch2\u003e5.1. Main investigation in the NTR cohort\u003c/h2\u003e\u003cp\u003eAll statistical analyses were performed in R version 4.3.2. Participants were defined “ε4 carriers” (ε2ε4, ε3ε4, ε4ε4) and “ε4 non-carriers” (ε2ε2, ε2ε3, ε3ε3). All analyses included body mass index (BMI), estimated Glomerular Filtration Rate (eGFR, a measure of kidney function), and genetic principal components 1 to 20 as covariates. Our decision to include BMI and eGFR as covariates is based on previous literature showing that these two factors may influence plasma biomarker concentrations (56–58). For the statistical analyses, the PRS\u003csub\u003eAD\u003c/sub\u003e was handled as continuous. For visual representation, participants were classified as low, medium, or high PRS\u003csub\u003eAD\u003c/sub\u003e using tertile cutoffs.\u003c/p\u003e\u003cp\u003eGeneralized additive models for location, scale and shape (GAMLSS) were fit for each AD biomarker using the \u003cem\u003egamlss\u003c/em\u003e package (version 5.4–22). Using the \u003cem\u003efitDist\u003c/em\u003e function, it was determined that a Box-Cox t distribution (59) would be the most appropriate for each biomarker. The analysis was then conducted in two parts: first, test whether \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership influenced biomarker concentrations over age, and second, test whether low, medium, or high PRS\u003csub\u003eAD\u003c/sub\u003e influenced plasma biomarker concentrations over age. For all analyses, nested models were compared using a likelihood ratio test, whereas a \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01 was taken to indicate the more complex model was a better fit of the data.\u003c/p\u003e\u003cp\u003eTo test the effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership on plasma biomarkers concentrations over age, three models were fit: 1, with age as the only predictor; 2, with age and the main effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership; and 3, with age, \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership, and the age*\u003cem\u003eAPOE\u003c/em\u003e ε4 carriership interaction. Model 1 and 2 were compared to determine whether there was a main effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership on plasma biomarker concentrations, and Model 2 and 3 were compared to determine whether, in addition to a main effect, \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership also affected the concentration change over age. Considering the reported dose-dependent effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership on the risk of developing AD, we further investigated whether a dose-dependent effect was observed on plasma biomarker concentrations by repeating the analyses modelling \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership as ordinal (non-carrier, heterozygous ε4 carrier, homozygous ε4 carrier). The same approach was taken to test the effect of PRS\u003csub\u003eAD\u003c/sub\u003e on plasma biomarker concentrations over age, replacing \u003cem\u003eAPOE\u003c/em\u003e carriership with PRS\u003csub\u003eAD\u003c/sub\u003e. The analyses were then repeated, this time accounting for \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership. The three models were then: 1, age and main effect of \u003cem\u003eAPOE\u003c/em\u003e; 2, age, \u003cem\u003eAPOE\u003c/em\u003e, and main effect of PRS\u003csub\u003eAD\u003c/sub\u003e; and 3, age, \u003cem\u003eAPOE\u003c/em\u003e, main effect of PRS\u003csub\u003eAD\u003c/sub\u003e, and the age*PRS\u003csub\u003eAD\u003c/sub\u003e interaction. All analysis were performed first in the main cohort (NTR) and then in sex-specific subsets to test for sex differences (Figure S14). Analyses were repeated using a generalized additive models (GAM), which allow us to generate confidence intervals and statistically estimate the age at divergence (if any) between the different AD genetic risk groups, but are more sensitive to overfitting.\u003c/p\u003e\u003ch2\u003e5.1.1. Sensitivity analysis for genetic relatedness\u003c/h2\u003e\u003cp\u003eBecause our cohort consists of genetically related individuals, we performed a sensitivity analysis randomly including only one person per family (with the exceptions of families consisting only of spouses, in which case both spouses were included). Within this cohort of individuals that are not genetically related, we repeated our analyses to test the effect of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriership and PRS\u003csub\u003eAD\u003c/sub\u003e (excluding \u003cem\u003eAPOE\u003c/em\u003e) on AD biomarker concentrations over age.\u003c/p\u003e\u003ch2\u003e5.2. Additional analysis in the EMIF-AD longitudinal cohort\u003c/h2\u003e\u003cp\u003eSimilarly to the main analyses, participants in the EMIF-AD cohort were defined as “ε4 carriers” (ε2ε4, ε3ε4, ε4ε4) and “ε4 non-carriers” (ε2ε2, ε2ε3, ε3ε3). Participants were further classified as Aβ PET negative if they were scored as Aβ- at all of their assessments (baseline and/or follow-up 2). Participants that were scored as Aβ + at either of their assessment were classified as Aβ PET positive. Therefore, the Aβ PET positive group includes both participants that were Aβ + at baseline, but also participants who were Aβ- at baseline and Aβ + at follow-up. Including participants who converted from Aβ- to Aβ + over a four year range into the Aβ PET positive group was done 1) to determine if plasma biomarker concentrations begin to change in the earliest AD stages and 2) because the specific timing of conversion from Aβ- to Aβ + could not be determined.\u003c/p\u003e\u003cp\u003eTo model individual longitudinal biomarker changes, linear mixed-effects models (LMMs) were fit, allowing for both random intercepts and random slopes, as well as a random effect for family to account for the familial clustering within monozygotic twin pairs. The LMMs included the measurements from the NTR biobank when available. In all models, age and sex were added as covariates, and biomarker concentrations were \u003cem\u003elog2\u003c/em\u003e transformed to fulfil linear regression assumptions. In the first set of analyses, the LMMs were fit with an age*\u003cem\u003eAPOE\u003c/em\u003e ε4 carriership interaction without accounting for Aβ PET status. The \u003cem\u003eemmeans\u003c/em\u003e function was used to estimate mean biomarker concentrations over age per \u003cem\u003eAPOE\u003c/em\u003e carriership group (ε4 carrier or non-carrier). These estimated biomarker concentration trajectories were visually compared to the estimated concentration changes in the NTR cohort. Then, a second set of LMMs were fit, with a three-way age*\u003cem\u003eAPOE\u003c/em\u003e ε4 carriership* Aβ PET status interaction. The \u003cem\u003eemmeans\u003c/em\u003e function was used to estimate mean biomarker concentrations over age per \u003cem\u003eAPOE\u003c/em\u003e carriership group and per Aβ PET status, resulting in four groups: Aβ PET negative non-carriers, Aβ PET negative ε4 carriers, Aβ PET positive non-carriers, and Aβ PET positive ε4 carriers. The \u003cem\u003eemmeans\u003c/em\u003e and \u003cem\u003eemtrends\u003c/em\u003e functions were used to compare biomarker trajectories in the four aforementioned groups, with a \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01 taken as the significance threshold.\u003c/p\u003e"},{"header":"Declarations","content":" \u003cp\u003eFor the Netherlands Twin Register: \u0026quot;Written informed consent was obtained from all participants. The study was approved by the Central Ethics Committee on Research Involving Human Subjects of the VU University Medical Centre, Amsterdam, an Institutional Review Board certified by the U.S. Office of Human Research Protections (IRB Number IRB00002991 under Federal-wide Assurance-FWA00017598; IRB/ institute codes, NTR 03-180).\u0026quot;\u003c/p\u003e\n \u003cp\u003eFor the EMIF cohort: \u0026quot;The Medical Ethics Review Committee of the Amsterdum University Medical Center performed approved the study. Research was performed according to the principles of the Declaration of Helsinki and in accordance with the Medical Research Involving Human Subjects Act and codes on \u0026lsquo;good use\u0026rsquo; of clinical data and biological samples as developed by the Dutch Federation of Medical Scientific Societies. All participants gave written informed consent (IRB registration number: 2014.210.\u0026quot;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFunding for blood sample collection was obtained from the Netherlands Organization for Scientific Research (NWO) and The Netherlands Organisation for Health Research and Development (ZonMW) grants 904-61-090, 985-10-002, 904-61-193,480-04-004, 400-05-717, \u0026nbsp;Addiction-31160008, 016-115-035, 400-07-080, Middelgroot-911-09-032, NWO-Groot 480-15-001/674; the European Community\u0026apos;s Fifth and Seventh Framework Program (FP5- LIFE QUALITY-CT-2002-2006, FP7- HEALTH-F4-2007-2013, grant 01254: GenomEUtwin, grant 01413: ENGAGE).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRR\u003c/strong\u003e, \u003cstrong\u003eCT\u003c/strong\u003e and \u003cstrong\u003eAdB\u003c/strong\u003e receive funding from the CANTATE project. The CANTATE project is funded by the Alzheimer Drug Discovery Foundation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eResearch of \u003cstrong\u003eCT\u003c/strong\u003e is supported by the European Commission (Marie Curie International Training Network, grant agreement No 860197 (MIRIADE), Innovative Medicines Initiatives 3TR (Horizon 2020, grant no 831434) EPND (IMI 2 Joint Undertaking (JU), grant No. 101034344) and JPND (bPRIDE), National MS Society (Progressive MS alliance), Alzheimer Association, Health Holland, the Dutch Research Council (ZonMW), Alzheimer Drug Discovery Foundation, The Selfridges Group Foundation, and Alzheimer Netherlands.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCT\u003c/strong\u003e is recipient of ABOARD, which is a public-private partnership receiving funding from ZonMW (#73305095007) and Health~Holland, Topsector Life Sciences \u0026amp; Health (PPP-allowance; #LSHM20106).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCT\u003c/strong\u003e is recipient of TAP-dementia, a ZonMw funded project (#10510032120003) in the context of the Dutch National Dementia Strategy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCT\u003c/strong\u003e has research contracts with Acumen, ADx Neurosciences, AC-Immune, Alamar, Aribio, Axon Neurosciences, Beckman-Coulter, BioConnect, Bioorchestra, Brainstorm Therapeutics, C2N diagnostics, Celgene, Cognition Therapeutics, EIP Pharma, Eisai, Eli Lilly, Fujirebio, Instant Nano Biosensors, Merck, Muna, Nitrase Therapeutics, Novo Nordisk, Olink, PeopleBio, Quanterix, Roche, Sysmex, Toyama, Vaccinex, Vivoryon. She is editor in chief of Alzheimer Research and Therapy, and serves on editorial boards of Molecular Neurodegeneration, Alzheimer\u0026rsquo;s \u0026amp; Dementia, Neurology: Neuroimmunology \u0026amp; Neuroinflammation, Medidact Neurologie/Springer, and is committee member to define guidelines for Cognitive disturbances, and one for acute Neurology in the Netherlands. \u0026nbsp;She has consultancy/speaker contracts for Aribio, Biogen, Beckman-Coulter, Cognition Therapeutics, Danaher, Eisai, Eli Lilly, Janssen, Merck, Neurogen Biomarking, Nordic Biosciences, Novo Nordisk, Novartis, Olink, Quanterix, Roche, Sanofi and Veravas.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBB, LB, DW, LL, RP,\u003c/strong\u003e \u003cstrong\u003ePJV\u003c/strong\u003e, \u003cstrong\u003eSMvdL\u003c/strong\u003e and \u003cstrong\u003eEdG\u003c/strong\u003e has nothing to disclose. \u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eOrganization WH. Global Health Estimates 2020: Deaths by Cause, Age, Sex, by Country and by Region, 2000-2019 2020 [Available from: https://www.who.int/data/gho/data/themes/mortality-and-global-health-estimates/ghe-leading-causes-of-death.\u003c/li\u003e\n\u003cli\u003eAssociation As. 2025 Alzheimer\u0026apos;s disease facts and figures. Alzheimer\u0026apos;s \u0026amp; Dementia. 2025;\u0026apos;21(4):e70235.\u003c/li\u003e\n\u003cli\u003eRafii MS, Aisen PS. Detection and treatment of Alzheimer\u0026apos;s disease in its preclinical stage. nature aging. 2023;3:520\u0026ndash;31.\u003c/li\u003e\n\u003cli\u003eJia J, Ning Y, Chen M, Wang S, Yang H, Li F, et al. Biomarker Changes during 20 Years Preceding Alzheimer\u0026apos;s Disease. The New England Journal of Medicine. 2024;390(8):712\u0026ndash;22.\u003c/li\u003e\n\u003cli\u003eTherriault J, Pascoal TA, Benedet AL, Tissot C, Savard M, Chamoun M, et al. Frequency of Biologically Defined Alzheimer Disease in Relation to Age, Sex, APOE \u0026epsilon;4, and Cognitive Impairment. Neurology. 2021;96(7):e975\u0026ndash;85.\u003c/li\u003e\n\u003cli\u003eNeu SC, Pa J, Kukull W, Beekly D, Kuzma A, Gangadharan P, et al. Apolipoprotein E Genotype and Sex Risk Factors for Alzheimer Disease. JAMA Neurology. 2017;74(10):1178\u0026ndash;89.\u003c/li\u003e\n\u003cli\u003eBelloy ME, Andrews SJ, Guen YL, Cuccaro M, Farrer LA, Napolioni V, et al. APOE Genotype and Alzheimer Disease Risk Across Age, Sex, and Population Ancestry. JAMA Neurology. 2023;80(12):1284\u0026ndash;94.\u003c/li\u003e\n\u003cli\u003eRiedel BC, Thompson PM, Brinton RD. Age, APOE and Sex: Triad of Risk of Alzheimer\u0026apos;s Disease. The Journal of Steroid Biochemistry and Molecular Biology. 2016;160:134\u0026ndash;47.\u003c/li\u003e\n\u003cli\u003eMahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988;240(4852):622\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eAbondio P, Sazzini M, Garagnani P, Boattini A, Monti D, Franceschi C, et al. The Genetic Variability of APOE in Different Human Populations and Its Implications for Longevity. Genes (Basel). 2019;10(3):222.\u003c/li\u003e\n\u003cli\u003eFarrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of Age, Sex, and Ethnicity on the Association Between Apolipoprotein E Genotype and Alzheimer Disease. JAMA. 1997;278(1349-56).\u003c/li\u003e\n\u003cli\u003eCorder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer\u0026apos;s Disease in Late Onset Families. Science. 1993;261:921\u0026ndash;23.\u003c/li\u003e\n\u003cli\u003eBlacker D, Haines JL, Rodes L, Terwedow H, Go RC, Harrell LE, et al. ApoE-4 and age at onset of Alzheimer\u0026apos;s disease: the NIMH genetics initiative. Neurology. 1997;48(1):139\u0026ndash;47.\u003c/li\u003e\n\u003cli\u003eAshton NJ, Brum WS, Molfetta GD, Benedet AL, Arslan B, Jonaitis E, et al. Diagnostic Accuracy of a Plasma Phosphorylated Tau 217 Immunoassay for Alzheimer Disease Pathology. JAMA Neurology. 2024:e235319.\u003c/li\u003e\n\u003cli\u003eAltomare D, Stampacchia S, Ribaldi F, Tomczyk S, Chevalier C, Poulain G, et al. Plasma biomarkers for Alzheimer\u0026rsquo;s disease: a field-test in a memory clinic. Journal of Neurology, Neurosurgery, and Psychiatry. 2023;94:420\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eChatterjee P, Pedrini S, Doecke JD, Thota R, Villemagne VL, Dor\u0026eacute; V, et al. Plasma A\u0026beta;42/40 ratio, p-tau181, GFAP, and NfL across the Alzheimer\u0026apos;s disease continuum: A cross-sectional and longitudinal study in the AIBL cohort. Alzheimer\u0026apos;s \u0026amp; Dementia. 2022;19(4):1117\u0026ndash;34.\u003c/li\u003e\n\u003cli\u003eWojdała AL, Bellomo G, Toja A, Gaetani L, Parnetti L, Chiasserini D. CSF and plasma A\u0026beta;42/40 across Alzheimer\u0026rsquo;s disease continuum: comparison of two ultrasensitive Simoa\u0026reg; assays targeting distinct amyloid regions. Clinical Chemistry and Laboratory Medicine. 2023;62(2).\u003c/li\u003e\n\u003cli\u003eOssenkoppele R, Salvad\u0026oacute; G, Janelidze S, Binette AP, Bali D, Karlsson L, et al. Plasma p-tau217 and tau-PET predict future cognitive decline among cognitively unimpaired individuals: implications for clinical trials. nature aging. 2025;5:883\u0026ndash;96.\u003c/li\u003e\n\u003cli\u003eEratne D, Kang M, Malpas CB, Dang C, Lewis C, Bhalala OG, et al. Plasma p‐tau217, NfL, GFAP diagnostic performance and biomarker profiles in Alzheimer\u0026apos;s disease, frontotemporal dementia, and psychiatric disorders, in a prospective unselected neuropsychiatry memory clinic. Alzheimer\u0026apos;s \u0026amp; Dementia. 2025;21(10):e70717.\u003c/li\u003e\n\u003cli\u003eWoo KA, Yoon EJ, Kim R, Kim H, Kim S, Jin B, et al. Plasma p-tau217 and A\u0026beta;42/40 as markers of A\u0026beta; pathology in the Lewy body continuum. Alzheimer\u0026apos;s \u0026amp; Dementia. 2025;21(10):e70788.\u003c/li\u003e\n\u003cli\u003eTeunissen CE, Verberk IMW, Thijssen EH, Vermunt L, Hansson O, Zetterberg H, et al. Blood-based biomarkers for Alzheimer\u0026apos;s disease towards clinical implementation. Lancet Neurology. 2022;21(1):66\u0026ndash;77.\u003c/li\u003e\n\u003cli\u003eForgrave LM, Ma M, Best JR, DeMarco ML. The diagnostic performance of neurofilament light chain in CSF and blood for Alzheimer\u0026apos;s disease, frontotemporal dementia, and amyotrophic lateral sclerosis: A systematic review and meta-analysis. Alzheimer\u0026apos;s \u0026amp; Dementia: Diagnosis, Assessment \u0026amp; Disease Monitoring. 2019;11:730\u0026ndash;43.\u003c/li\u003e\n\u003cli\u003eShir D, Graff‐Radford J, Hofrenning EI, Lesnick TG, Przybelski SA, Lowe VJ, et al. Association of plasma glial fibrillary acidic protein (GFAP) with neuroimaging of Alzheimer\u0026apos;s disease and vascular pathology. Alzheimer\u0026apos;s \u0026amp; Dementia. 2022;14(1):e12291.\u003c/li\u003e\n\u003cli\u003eLin J, Ou R, Li C, Hou Y, Zhang L, Wei Q, et al. Plasma glial fibrillary acidic protein as a biomarker of disease progression in Parkinson\u0026rsquo;s disease: a prospective cohort study. BMC Medicine. 2023;21:420.\u003c/li\u003e\n\u003cli\u003eChatterjee P, Vermunt L, Gordon BA, Pedrini S, Boonkamp L, Armstrong NJ, et al. Plasma glial fibrillary acidic protein in autosomal dominant Alzheimer\u0026apos;s disease: Associations with A\u0026beta;-PET, neurodegeneration, and cognition. Alzheimer\u0026apos;s \u0026amp; Dementia. 2023;19(7):2790\u0026ndash;804.\u003c/li\u003e\n\u003cli\u003ePanday N, Yang Z, Cieza B, Reyes-Dumeyer D, Kang MS, Montesinos R, et al. Plasma phospho-tau217 as a predictive biomarker for Alzheimer\u0026rsquo;s disease in a large south American cohort. Alzheimer\u0026apos;s Research \u0026amp; Therapy. 2025;17:1.\u003c/li\u003e\n\u003cli\u003eLehmann S, Schraen-Maschke S, Vidal J-S, Delaby C, Buee L, Blanc F, et al. Clinical value of plasma ALZpath pTau217 immunoassay for assessing mild cognitive impairment. Journal of Neurology, Neurosurgery \u0026amp; Psychiatry. 2024;95:1046\u0026ndash;53.\u003c/li\u003e\n\u003cli\u003eMielke MM, Dage JL, Frank RD, Algeciras-Schimnich A, Knopman DS, Lowe VJ, et al. Performance of plasma phosphorylated tau 181 and 217 in the community. Nature Medicine. 2022;28(7):1398\u0026ndash;405.\u003c/li\u003e\n\u003cli\u003eAarsland D, Sunde AL, Tovar-Rios DA, Leuzy A, Fladby T, Zetterberg H, et al. Prevalence of Alzheimer\u0026apos;s disease pathology in the community. Nature. 2025.\u003c/li\u003e\n\u003cli\u003eMalek-Ahmadi M, Su Y, Ghisays V, Luo J, Devadas V, Chen Y, et al. Plasma NfL is associated with the APOE \u0026epsilon;4 allele, brain imaging measurements of neurodegeneration, and lower recall memory scores in cognitively unimpaired late-middle-aged and older adults. Alzheimer\u0026apos;s Research \u0026amp; Therapy. 2023;15(74).\u003c/li\u003e\n\u003cli\u003eIngannato A, Bagnoli S, Mazzeo S, Giacomucci G, Bessi V, Ferrari C, et al. Plasma GFAP, NfL and pTau 181 detect preclinical stages of dementia. Frontiers in Endocrinology. 2024;15.\u003c/li\u003e\n\u003cli\u003eLangella S, Bonta K, Chen Y, Su Y, Vasquez D, Aguillon D, et al. Impact of APOE \u0026epsilon;4 and \u0026epsilon;2 on plasma neurofilament light chain and cognition in autosomal dominant Alzheimer\u0026rsquo;s disease. Alzheimer\u0026apos;s Research \u0026amp; Therapy. 2024;16(208).\u003c/li\u003e\n\u003cli\u003eVarma VR, An Y, Kac PR, Bilgel M, Moghekar A, Loeffler T, et al. Longitudinal progression of blood biomarkers reveals a key role of astrocyte reactivity in preclinical Alzheimer\u0026rsquo;s disease. medRxiv. 2024.\u003c/li\u003e\n\u003cli\u003eYakoub Y, Ashton NJ, Strikwerda-Brown C, Montoliu-Gaya L, Karikari TK, Kac PR, et al. Longitudinal blood biomarker trajectories in preclinical Alzheimer\u0026apos;s disease. Alzheimer\u0026apos;s \u0026amp; Dementia. 2023;19(12):5620\u0026ndash;31.\u003c/li\u003e\n\u003cli\u003eSnellman A, Ekblad LL, Ashton NJ, Karikari TK, Lantero-Rodriguez J, Pietil\u0026auml; E, et al. Head-to-head comparison of plasma p-tau181, p-tau231 and glial fibrillary acidic protein in clinically unimpaired elderly with three levels of APOE4-related risk for Alzheimer\u0026apos;s disease. Neurobiology of Disease. 2023;183(106175).\u003c/li\u003e\n\u003cli\u003eSnellman A, Ekblad LL, Tuisku J, Koivum\u0026auml;ki M, Ashton NJ, Lantero-Rodriguez J, et al. APOE \u0026epsilon;4 gene dose effect on imaging and blood biomarkers of neuroinflammation and beta-amyloid in cognitively unimpaired elderly. Alzheimer\u0026apos;s Research \u0026amp; Therapy. 2023;15(71).\u003c/li\u003e\n\u003cli\u003eCooper JG, Ghodsi M, Stukas S, Leach S, Brooks-Wilson A, Wellington CL. APOE \u0026epsilon;4 carrier status modifies plasma p‐tau181 concentrations in cognitively healthy super‐seniors. Alzheimer\u0026apos;s \u0026amp; Dementia. 2024;20(6):4373\u0026ndash;80.\u003c/li\u003e\n\u003cli\u003eXu Y, Qiao M, Gunasekaran TI, Gu Y, Reyes-Dumeyer D, Piriz A, et al. Evaluating polygenic risk score prediction performance for Alzheimer\u0026apos;s disease in a population-based Hispanic cohort using single- and multi-ancestry models. The Lancet Regional Health - Americas. 2025;49:101198.\u003c/li\u003e\n\u003cli\u003eStocker H, Trares K, Beyer L, Perna L, Rujescu D, Holleczek B, et al. Alzheimer\u0026rsquo;s polygenic risk scores, APOE, Alzheimer\u0026rsquo;s disease risk, and dementia-related blood biomarker levels in a population-based cohort study followed over 17 years. Alzheimer\u0026apos;s Research \u0026amp; Therapy. 2023;15:129.\u003c/li\u003e\n\u003cli\u003eLumsden AL, Mulugeta A, Zhou A, Hypp\u0026ouml;nen E. Apolipoprotein E (APOE) genotype-associated disease risks: a phenome-wide, registry-based, case-control study utilising the UK Biobank. eBioMedicine. 2020;59:102954.\u003c/li\u003e\n\u003cli\u003eDaunt P, Ballard CG, Creese B, Davidson G, Hardy J, Oshota O, et al. Polygenic Risk Scoring is an Effective Approach to Predict Those Individuals Most Likely to Decline Cognitively Due to Alzheimer\u0026apos;s Disease. The Journal of Prevention of Alzheimer\u0026apos;s Disease. 2021;8(1):78\u0026ndash;83.\u003c/li\u003e\n\u003cli\u003eKasuga K, Kikuchi M, Kikkawa-Saito E, Tsukie T, Ishiguro T, Miyashita A, et al. Evaluation of plasma p-tau217 biomarkers in detecting amyloid pathology and predicting cognitive outcomes: Observations from Japanese Alzheimer\u0026apos;s disease neuroimaging initiative cohort. The Journal of Prevention of Alzheimer\u0026apos;s Disease. 2026;13(4):100502.\u003c/li\u003e\n\u003cli\u003eBilgel M, An Y, Walker KA, Moghekar AR, Ashton NJ, Kac PR, et al. Longitudinal changes in Alzheimer\u0026apos;s-related plasma biomarkers and brain amyloid. Alzheimer\u0026apos;s \u0026amp; Dementia. 2023;19(10):4335\u0026ndash;45.\u003c/li\u003e\n\u003cli\u003eLi Y, Yen D, Hendrix RD, Gordon BA, Dlamini S, Barth\u0026eacute;lemy NR, et al. Timing of Biomarker Changes in Sporadic Alzheimer\u0026apos;s Disease in Estimated Years from Symptom Onset. Annals of Neurology. 2024;95(5):951\u0026ndash;65.\u003c/li\u003e\n\u003cli\u003eShaaban CE, Jia Y, Chang C-CH, Ganguli M. Independent and joint effects of vascular and cardiovascular risk factor pairs for risk of all-cause dementia: a prospective population-based study. International Psychogeriatrics. 2019;31(10):1421\u0026ndash;32.\u003c/li\u003e\n\u003cli\u003eSillanp\u0026auml;\u0026auml; MJ. Overview of techniques to account for confounding due to population stratification and cryptic relatedness in genomic data association analyses. Heredity. 2010;106(4):511\u0026ndash;19.\u003c/li\u003e\n\u003cli\u003eWillemsen G, Geus EJCd, Bartels M, Beijsterveldt CEMTv, Brooks AI, Burk GFE-v, et al. The Netherlands Twin Register biobank: a resource for genetic epidemiological studies. Twin Research and Human Genetics. 2010;13(3):231\u0026ndash;45.\u003c/li\u003e\n\u003cli\u003eLigthart L, Beijsterveldt CEMv, Kevenaar ST, Zeeuw Ed, Bergen Ev, Bruins S, et al. The Netherlands Twin Register: Longitudinal Research Based on Twin and Twin-Family Designs. Twin Research and Human Genetics. 2019;22(6):623\u0026ndash;36.\u003c/li\u003e\n\u003cli\u003eVerberk IMW, Misdorp EO, Koelewijn J, Ball AJ, Blennow K, Dage JL, et al. Characterization of pre-analytical sample handling effects on a panel of Alzheimer\u0026apos;s disease\u0026ndash;related blood-based biomarkers: Results from the Standardization of Alzheimer\u0026apos;s Blood Biomarkers (SABB) working group. Alzheimer\u0026apos;s \u0026amp; Dementia. 2022;18(8):1484\u0026ndash;97.\u003c/li\u003e\n\u003cli\u003eKonijnenberg E, Carter SF, Kate MT, Braber Ad, Tomassen J, Amadi C, et al. The EMIF-AD PreclinAD study: study design and baseline cohort overview. Alzheimer\u0026apos;s Research \u0026amp; Therapy. 2018;10(1):75.\u003c/li\u003e\n\u003cli\u003eEhli EA, Abdellaoui A, Fedko IO, Grieser C, Nohzadeh-Malakshah S, Willemsen G, et al. A method to customize population-specific arrays for genome-wide association testing. European Journal of Human Genetics. 2017;25(2):267\u0026ndash;70.\u003c/li\u003e\n\u003cli\u003eLee SJvd, Wolters FJ, Ikram MK, Hofman A, Ikram MA, Amin N, et al. The effect of APOE and other common genetic variants on the onset of Alzheimer\u0026apos;s disease and dementia: a community-based cohort study. Lancet Neurology. 2018;17(5):434\u0026ndash;44.\u003c/li\u003e\n\u003cli\u003eBellenguez C, K\u0026uuml;\u0026ccedil;\u0026uuml;kali F, Jansen IE, Kleineidam L, Moreno-Grau S, Amin N, et al. New insights into the genetic etiology of Alzheimer\u0026rsquo;s disease and related dementias. Nature Genetics. 2022;54(4):412\u0026ndash;36.\u003c/li\u003e\n\u003cli\u003eWare EB, Faul JD, Mitchell CM, Bakulski KM. Considering the \u003cem\u003eAPOE\u003c/em\u003e locus in Alzheimer\u0026apos;s disease polygenic scores in the Health and Retirement Study: a longitudinal panel study. BMC Medical Genomics. 2020;13:164.\u003c/li\u003e\n\u003cli\u003eVilhj\u0026aacute;lmsson BJ, Yang J, Finucane HK, Gusev A, Lindstr\u0026ouml;m S, Ripke S, et al. Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk Scores. American Journal of Human Genetics. 2015;97(4):576\u0026ndash;92.\u003c/li\u003e\n\u003cli\u003eJacobs T, O\u0026apos;Brien C, Figueredo L, Gogola A, Gaggi NL, Hurwitz B, et al. Body mass index and blood volume influence plasma biomarkers and positron emission tomography classification in preclinical Alzheimer\u0026apos;s disease. Alzheimer\u0026apos;s \u0026amp; Dementia. 2025;21(10):e70765.\u003c/li\u003e\n\u003cli\u003eHermesdorf M, Esselmann H, Morgado B, Jahn-Brodmann A, Herrera-Rivero M, Wiltfang J, et al. The association of body mass index and body composition with plasma amyloid beta levels. Brain Communications. 2023;5(5):fcad263.\u003c/li\u003e\n\u003cli\u003eGasparini F, Valletta M, Vetrano DL, Beridze G, Rizzuto D, Calder\u0026oacute;n-Larra\u0026ntilde;aga A, et al. Kidney Function, Alzheimer Disease Blood Biomarkers, and Dementia Risk in Community-Dwelling Older Adults. Neurology. 2026;106(1):e214446.\u003c/li\u003e\n\u003cli\u003eRigby RA, Stasinopoulos DM. Using the Box-Cox t distribution in GAMLSS to model skewness and kurtosis. Statistical Modelling. 2006;6(3):209\u0026ndash;29.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8426026/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8426026/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAlzheimer’s disease (AD) plasma biomarkers are cost-effective tools to detect early stage amyloid pathology. We investigated how interactions between AD genetic risk (apolipoprotein E (\u003cem\u003eAPOE\u003c/em\u003e) ε4, AD polygenic risk score (excluding \u003cem\u003eAPOE; \u003c/em\u003ePRS\u003csub\u003eAD\u003c/sub\u003e)), age, and sex affected AD plasma biomarkers concentrations in a population-based cohort. We included 5913 participants age 30 – 80 (64.1% females, 29.8% ε4 carriers) and measured plasma phosphorylated-tau-217 (p-tau217), amyloid-β-42/40 (Aβ-42/40), neurofilament light (NfL), and glial fibrillary acidic protein (GFAP). PRS\u003csub\u003eAD\u003c/sub\u003e did not associate with biomarker concentrations over age. In ε4 carriers, decreases in Aβ-42/40 became steeper after age 50, and increases in p-tau217 and GFAP became steeper after age 55 and 60 compared to non-carriers, respectively. Analyses in a separate longitudinal cohort confirmed that p-tau217 concentration changes in ε4 carriers reflect AD pathology. These results highlight the potential of plasma biomarkers, especially p-tau217, for early detection of preclinical AD pathology.\u003c/p\u003e","manuscriptTitle":"APOE ε4 carriership status enhances age-related changes in Alzheimer’s disease plasma biomarkers concentrations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-12 14:30:33","doi":"10.21203/rs.3.rs-8426026/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7b926a63-8214-46fe-aaa0-8fd51b86c08e","owner":[],"postedDate":"May 12th, 2026","published":true,"recentEditorialEvents":[{"type":"editorAssigned","content":"","date":"2026-05-08T08:56:33+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-05-07T13:40:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Nature Communications","date":"2026-05-07T13:07:18+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":67770650,"name":"Biological sciences/Neuroscience"},{"id":67770651,"name":"Biological sciences/Genetics"},{"id":67770652,"name":"Health sciences/Biomarkers/Predictive markers"}],"tags":[],"updatedAt":"2026-05-12T14:30:33+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-12 14:30:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8426026","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8426026","identity":"rs-8426026","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}