Unraveling the association between retinal thickness and Alzheimer's disease, and circulating total-tau levels: Insights from genetic evidence

Unraveling the association between retinal thickness and Alzheimer's disease, and circulating total-tau levels: Insights from genetic evidence | 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 Research Article Unraveling the association between retinal thickness and Alzheimer's disease, and circulating total-tau levels: Insights from genetic evidence Dandan Sheng, Song Wang, Zheng Xiao, Weiping Liu, Bo Xiao, Luo Zhou This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5869988/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 Background Observational studies have reported associations between retinal thickness and Alzheimer’s disease (AD); however, the causal relationship remains uncertain. Methods We investigated the potential causal relationship between retinal thickness and AD using Mendelian randomization (MR) and genetic colocalization analyses. Multiple genome-wide association studies (GWAS) of European ancestry were used for MR causal inference. Summary-level data on retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GC-IPL) thickness were sourced from the UK Biobank (N = 31,434). AD data were obtained from a GWAS meta-analysis conducted by the International Genomics of Alzheimer’s Project for discovery (N = 63,926) and an independent European GWAS cohort for replication (N = 85,934). Circulating total-tau biomarker data were derived from another GWAS in individuals of European ancestry (N = 14,721). The inverse-variance weighted (IVW) method served as the primary analytical approach, supplemented by MR-Egger, robust adjusted profile score, maximum likelihood, and weighted median methods. Sensitivity analyses included Cochran’s Q test, MR-Egger intercept test, leave-one-out analysis, and MR pleiotropy residual sum and outlier analysis to ensure robustness. Genetic colocalization analysis was performed to identify potential shared causal variants between retinal thickness and AD. Results The IVW estimates from the discovery MR analysis indicated no statistically significant causal effect of genetically predicted RNFL or GC-IPL thickness on AD or circulating total-tau levels, and reverse MR analysis found no causal link either (P IVW > 0.05). Replication bidirectional MR analysis produced consistent negative results (P IVW > 0.05). Sensitivity analyses demonstrated robustness across all MR methods, with no evidence of heterogeneity, horizontal pleiotropy, or instrumental variable outliers. Genetic colocalization analysis identified no shared causal variants between RNFL or GC-IPL thickness and AD or circulating total-tau (posterior probability H4 < 0.75). Conclusion Our study does not support a genetic causal link between retinal thickness, AD, and circulating total-tau levels, despite previous observational studies suggesting an association between retinal thinning and higher AD risk. Further research is needed to clarify the relationship between RNFL and GC-IPL thickness and AD, as well as the underlying biological mechanisms. Alzheimer's disease tau retinal nerve fiber layer ganglion cell-inner plexiform layer Mendelian randomization Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction With the increase in life expectancy, there is an anticipated upsurge in the prevalence of neurodegenerative diseases, among which Alzheimer's disease (AD) stands out as the most prominent form of neurodegenerative dementia observed in older individuals. The pathological hallmark of this condition involves the progressive accumulation of misfolded amyloid-beta (Aβ) and tau proteins in the brain during an extended, subtle, and insidious pre-symptomatic phase, ultimately leading to a gradual decline in cognitive performance( 1 ). Early identification and prevention can help limit progression from asymptomatic to clinical stages. Given that the retina is an extension of the brain and shares a common embryonic origin with the central nervous system, there is significant interest in utilizing the expanding retinal imaging technologies to understand, diagnose, and monitor AD( 2 ). Optical coherence tomography (OCT) is widely used due to its high spatial resolution and relatively low cost. The alterations of retinal structures, particularly the thinning of the retinal nerve fiber layer (RNFL) and the ganglion cell-inner plexiform layer (GC-IPL), have attracted the attentions to be associated with a decline in cognitive performance based on multiple previous studies, which have identified measurable changes in retinal thickness during an early stage of AD( 3 ). However, in contrast to the findings mentioned above, several well-conducted studies have reported minimal changes in retinal thickness (RNFL, GCL-IPL) associated with early dementia or mild cognitive impairment. Additionally, no significant associations were found between retinal thickness and CSF biomarkers (Aβ-42, tau-181, phosphorylated tau, and the tau-181/Aβ1–42 ratio)( 4 – 6 ). Maria et al. recently investigated the association between cognitive decline spanning both AD and non-AD cases, and retinal thickness in a large Spanish cohort. Utilizing genome-wide association study (GWAS) and polygenic risk score analysis, their findings did not reveal a significant genetic association between dementia and retinal thickness, either of AD or non-AD dementia types( 7 ). It is important to note that the outputs of OCT are influenced by a complex interplay of technical parameters that vary across vendors and operators, potentially leading to inconsistent estimates of retinal layer thickness. Additionally, due to limitations in study design and imaging equipment, the field of view is frequently restricted to the macula or optic disc, thereby excluding substantial portions of the peripheral retina from analysis. Furthermore, a significant challenge in longitudinal studies is the co-registration of images across different subjects and time points, which is a critical factor in the analysis of retinal imaging data. Finally, although OCT devices provide micron-level resolution, the reproducibility of measurements is heavily contingent upon the experience and training of the operators( 8 ). Thus, current researches are insufficient to fully elucidate the causal relationship between AD, pathological protein deposits, and the thinning of RNFL and GC-IPL. Mendelian Randomization (MR) is a roubust research methodology that relies on single nucleotide polymorphisms (SNPs) from GWAS to create instrumental variables (IVs) for establishment of causation. The random assignment of alleles to offspring effectively mitigates confounding bias. MR employs IVs to roubustly identify risk factors( 9 ). On the other hand, genetic colocalization analysis is commonly employed to ascertain whether two phenotypes are driven by the same causal variant within a specific gene region, thereby bolstering the evidence of a genetic causal association between the exposure and outcome. In this study, we aimed to comprehensively assess the causality between RNFL, GC-IPL thickness and AD, circulating total-tau through MR and genetic colocalization analyses. 2. Materials and Methods 2.1 Study design We first conducted a bidirectional two-sample MR study to ascertain the causality between the traits of retinal thickness and AD, and its biomarker circulating total-tau. In the forward MR analysis, we assessed the causal impact of RNFL and GC-IPL thickness on AD and circulating total-tau level, while in the reverse MR analysis, we examined the causal impact of AD, circulating total-tau level on RNFL and GC-IPL thickness. Ensuring the reliability of MR results relies on fulfilling three fundamental assumptions: (a) the correlation hypothesis, which posits that IVs possess significant genetic connections to the exposure; (b) the independence hypothesis, which stipulates that IVs should not be directly associated with confounding factors related to exposure or outcome; (c) the exclusivity hypothesis, which implies that IVs should not exhibit direct associations with the outcome( 9 ). To investigate the presence of a co-driven causal variant, we subsequently conducted additional genetic colocalization analysis. The datasets utilized in the present study were obtained from publicly accessible repositories, and ethical approval was previously acquired from the local ethics committee during the original GWAS. No additional ethical approval or informed consent was necessary. The flowchart of the study is presented in Fig. 1 . 2.2 Data sources The summary-level datasets of RNFL and GC-IPL thickness were derived from the recent largest GWAS investigating inner retinal morphology using phenotypes obtained from OCT images of 31,434 UK Biobank participants authored by Kunkle et al., encompassing 9,121,075 genetic variants of SNPs( 10 ). Summary statistics regarding circulating levels of total-tau were derived from a large-scale meta-analysis of GWAS involving 14,721 individuals of European ancestry with 8,360,926 SNPs( 11 ). For the discovery stage, the summary-level dataset of the AD was derived from a GWAS meta-analysis conducted by International Genomics of Alzheimer’s Project (IGAP), which included 63,926 participants (21,982 cases and 41,944 controls) with 10,528,610 SNPs( 12 ). For further validation, an additional dataset on AD with a sample size of 85,934 participants (39,106 cases and 46,828 controls) with 20,921,626 SNPs was utilized from a GWAS conducted by Bellenguez et al. for replication MR validation( 13 ). The participants included in our datasets are exclusively of European ancestry. The GWAS-GWAS genetic colocalization analysis was subsequently conducted within these datasets. The detailed information of GWAS datasets utilized in the MR and genetic colocalization is summarized in Supplementary Table S1 . 2.3 Mendelian randomization analysis We employed a threshold of P < 5E-08 and pairwise r 2 < 0.001 within a 10,000 kb window for the selection of IVs in both forward and reverse MR analyses, using the European population as the reference group. For the two traits related to retinal thickness, we applied a more lenient threshold of P < 5E-06 to ensure an adequate number of SNPs were available as IVs. SNPs with a minor allele frequency less than 0.01 and palindromic allele frequencies were excluded. All SNPs designated as IVs for scrutiny in this investigation exhibited a minimum F-statistic of 10 to counteract the influence of weak genetic instruments. PhenoScanner was utilized to check all IVs to remove potential confounding factors. We employed the inverse-variance weighted (IVW) method as the primary MR estimate. Several complementary methods, including MR-Egger, Robust adjusted profile score (RAPS), Maximum likelihood, and weighted median were also utilized for validation. The IVs are described in detail in Supplementary Table S2 . 2.4 Sensitivity analysis We employed Cochran's Q and the MR-Egger intercept tests to assess heterogeneity and pleiotropy within IVs. In order to identify potential SNP outliers among the IVs, we further utilized the assessment of MR pleiotropy residual sum and outlier (MR-PRESSO) as well as leave-one-out analysis. If an SNP outlier was detected through either MR-PRESSO or leave-one-out, it was excluded, after which we may perform a re-evaluation of the MR and sensitivity analysis to obtain robust and definitive findings. After applying the Bonferroni correction, a significance level of P < 0.0125 (0.05/4) would be considered statistically significant. All MR statistical analysis in this study were performed using the TwoSampleMR package (Version 0.5.7) in the RStudio (Version 2023.06.0). 2.5 Genetic colocalization analysis To determine if the SNPs associated with the exposure also impact the outcome, we additionally conducted a genetic colocalization analysis using the aforementioned GWAS datasets. This approach allows us to assess genetic correlations by assigning prior probabilities to random SNPs within the genomic region of interest, considering their associations with either the exposure, outcome, or both( 14 ). We extracted SNPs within a ± 50kb gene window surrounding the identified IVs for each potential causal variant from both exposure and outcome GWAS datasets. The posterior probability was considered under five hypotheses: H0 indicating no association with either trait; H1 indicating an association with trait 1; H2 indicating an association with trait 2; H3 indicating an association with both traits, involving two independent SNPs; H4 indicating an association with both traits, linked to one shared SNP. The colocalization was considered positive when the posterior probability H4 exceeded 0.75, indicating strong evidence for the presence of a shared causal variant( 15 ). The R package Coloc (Version 5.2.3) was employed for conducting genetic colocalization analysis. 3. Results 3.1 Causal relationship between retinal thickness and AD 3.1.1 Results of Mendelian randomization in the discovery stage In the discovery stage, the IVW estimates derived from IGAP dataset found no evidence supporting a causal effect of genetically predicted RNFL and GC-IPL thickness on the risk of AD (RNFL-AD: odds ratio [OR] = 1.015, 95% confidence interval [CI] = 0.987–1.043, P = 0.305; GC-IPL-AD: OR = 1.014, 95% CI = 0.991–1.037, P = 0.237). Likewise, there was no indication of any causal effect of genetically predicted AD on RNFL and GC-IPL thickness (AD-RNFL: OR = 0.930, 95% CI = 0.824–1.050, P = 0.239; AD-GC-IPL: OR = 0.982, 95% CI = 0.849–1.136, P = 0.807). The estimates calculated through IVW were further supported by alternative methods, including MR-Egger, RAPS, Maximum likelihood, and weighted median. In the MR-Egger method, we observed a nominal causal effect of genetically predicted AD on RNFL thickness (P = 0.041). However, after applying the Bonferroni correction, this estimate did not reach statistical significance. The forest plots illustrating the summarized causal estimates are presented in Fig. 2 . These findings remained robust across all sensitivity analysis, and detailed results are summarized in Supplementary Table S3. Scatter and leave-one-out plots in the the discovery MR analysis are presented in Supplementary Figure S1 and S2. 3.1.2 Results of Mendelian randomization in the replication stage The replication MR analysis consistently revealed no causal association between RNFL, GC-IPL thickness and AD in either direction, as all MR causal estimates yielded a P -value greater than 0.0125 in IVW (RNFL-AD: OR = 1.001, 95% CI = 0.986–1.017, P = 0.883; GC-IPL-AD: OR = 1.008, 95% CI = 0.998–1.018, P = 0.129; AD-RNFL: OR = 0.943, 95% CI = 0.826–1.077, P = 0.389; AD-GC-IPL: OR = 0.990, 95% CI = 0.816-1.200, P = 0.917). The causal eatimates calculated through MR-Egger, Robust adjusted profile score (RAPS), Maximum likelihood, and weighted median further supported the negative finding. The summarized causal estimates in the replication MR analysis are presented in Table 1 . Furthermore, the results from the replication MR analysis remained robust across all sensitivity analysis, as summarized in Supplementary Table S3. Supplementary Figure S3 and S4 depict scatter and leave-one-out plots illustrating the resluts of the replication MR analysis. Table 1 The causal estimates between RNFL, GC-IPL thickness and AD in replication MR analysis Exposure Outcome nSNPs Methods OR (95%CI) P RNFL thickness AD 20 IVW 1.001(0.986–1.017) 0.883 20 Weighted median 1.010(0.989–1.031) 0.365 20 Maximum likelihood 1.001(0.987–1.016) 0.870 20 RAPS 1.003(0.987–1.020) 0.715 20 MR Egger 1.024(0.979–1.071) 0.322 GC-IPL thickness AD 21 IVW 1.008(0.998–1.018) 0.129 21 Weighted median 1.009(0.995–1.023) 0.202 21 Maximum likelihood 1.008(0.998–1.018) 0.121 21 RAPS 1.006(0.996–1.017) 0.234 21 MR Egger 1.022(0.994–1.052) 0.140 AD RNFL thickness 54 IVW 0.943(0.826–1.077) 0.389 54 Weighted median 0.979(0.807–1.188) 0.830 54 Maximum likelihood 0.943(0.834–1.067) 0.350 54 RAPS 0.948(0.830–1.084) 0.435 54 MR Egger 0.842(0.667–1.062) 0.153 AD GC-IPL thickness 57 Inverse variance weighted 0.990(0.816-1.200) 0.917 57 Weighted median 1.082(0.831–1.410) 0.559 57 Maximum likelihood 0.990(0.842–1.164) 0.902 57 RAPS 0.966(0.799–1.169) 0.722 57 MR Egger 0.905(0.644–1.271) 0.566 nSNPs, number of single-nucleotide polymorphisms; OR, odds ratio; CI, confidence interval; AD, Alzheimer's disease; RNFL, retinal nerve fiber layer; GC-IPL, ganglion cell-inner plexiform layer; IVW, Inverse variance weighted; RAPS, Robust adjusted profile score; MR-Egger, Mendelian randomization Egger. 3.3 Causal Relationship between retinal thickness and circulating total-tau The IVW analysis revealed that RNFL thickness (OR, 0.993 [0.977–1.008]; P = 0.355) and GC-IPL thickness (OR, 1.001 [95% CI, 0.992–1.009]; P = 0.908) had no significant causal effect on circulating levels of total-tau. Similarly, there was no evidence of a causal effect of genetically predicted circulating total-tau levels on RNFL thickness (OR, 0.989 [0.886–1.105]; P = 0.849) or GC-IPL thickness (OR, 1.129 [95% CI, 0.977–1.305]; P = 0.099). These estimates derived from the IVW method were corroborated by alternative approaches, including MR-Egger, RAPS, Maximum Likelihood, and weighted median methods. The forest plots displaying the summarized causal estimates are shown in Fig. 3 . These findings remained robust across all sensitivity analyses, with detailed results summarized in Supplementary Table S4. Scatter and leave-one-out plots from the MR analysis are presented in Supplementary Figures S5 and S6. 3.4 Results of genetic colocalization analysis The posterior probability H4 obtained from GWAS-GWAS genetic colocalization analysis within the datasets of retianl thickness and AD ranged from 0.006 to 0.047, which fell below the predefined threshold of 0.75, suggesting a limited possibility for the presence of a shared causal variant underlying both the exposure and outcome traits. The regional association plots for colocalization analysis are further elaborated in Figs. 4 . The posterior probabilities H0-H4 are presented in Supplementary Table S5. 4. Discussion Our study represents a pioneering effort to elucidate the causal relationship between AD, the biomarker of circulating levels of total-tau, and retinal thickness from a genetic perspective, effectively accounting for confounding factors. Through bidirectional MR analysis, we found no evidence of a genetically predicted causal link between RNFL, GC-IPL thickness, in relation to circulating levels of total-tau and AD. Numerous early studies have identified a potential link between AD and retinal thickness; however, the results often lack consistency. Some studies have generally indicated that individuals with AD have exhibited thinner RNFL and GC-IPL in comparison to non-AD control groups. A population-based prospective cohort study conducted by Mutlu et al. has demonstrated a significant association between decreased RNFL thickness measured using OCT and an increased risk of AD dementia. Additionally, thinner GC-IPL thickness has been found to be correlated with prevalent dementia including AD( 16 ). Ko et al. have similarly discovered that thinner RNFL may be linked to poorer cognitive performance in individuals without neurodegenerative diseases and an increased likelihood of future cognitive decline within a substantial community-based cohort of healthy individuals( 17 ). A recent histopathologic, morphometric analysis of human postmortem retinas included eight patients with AD and eleven age-matched controls revealed a remarkable thinning of both the RNFL and GC-IPL correlated with AD( 18 ). Furthermore, Aβ protein deposition has been observed in the retinas of both AD patients and AD mouse models( 19 ). These observational findings seemed to hint that the retinal thickness may serve as a potential biomarker for early diagnosis and monitoring of AD. Relevant studies, however, have also presented divergent conclusions. Two recent population-based surveys conducted by Ito and Van Koolwijk et al. failed to yield evidence supporting the association between RNFL thickness and cognitive decline or dementia( 6 , 20 ). A Dutch cohort study was conducted involving 165 cognitively healthy monozygotic twins who underwent [18F] flutemetamol positron emission tomography, which was visually assessed to determine the presence of cortical Aβ. The findings revealed no discernible differences in retinal layer thickness in the macula or peripapillary RNFL between Aβ + and Aβ- participants, indicating limited application of RNFL thickness as a biomarker for early detecting of AD( 21 ). Similarly, in a study involving 57 participants with CSF amyloid-positive and PET amyloid-positive late-onset Alzheimer's disease (LOAD), no correlation was found between Mini-mental State Examination (MMSE) scores and any retinal thickness parameters. In patients with posterior cortical atrophy, an AD variant primarily affecting the visual cortex, no association was reported between retinal thickness measurements and LOAD status, including MMSE scores( 4 ). These inconsistent findings may be attributed to biases stemming from differences in study design, imaging equipment, and variability among researchers. Elevated levels of total-tau in CSF are a well-recognized biomarker of AD, with research indicating that a decline in total-tau levels may occur up to 10 years before the onset of AD( 22 ). However, studies examining the relationship between circulating levels of total-tau and AD have produced inconsistent results. Eunjoo et al. reported that circulating levels of total-tau in the mild AD group were significantly higher than those in age-matched controls and found a significant correlation between serum total-tau and both MMSE scores( 23 ). In a study by Niklas et al., 1,284 participants were included, and associations between plasma-tau, diagnosis, CSF biomarkers, MRI measures, 18fluorodeoxyglucose-PET, and cognition were tested. The findings indicated that plasma-tau partially reflected AD pathology, but there was considerable overlap between normal aging and AD, particularly in individuals without dementia. Although group-level differences were observed, these results do not support plasma-tau as a reliable biomarker for individual AD diagnosis( 24 ). In AD patients, elevated levels of pathological tau subtypes have been detected in the retina, including increased microglial uptake of retinal phosphorylated tau. These tau subtypes were primarily localized to the synapse-rich inner plexiform layer (IPL) and outer plexiform layer (OPL) in AD patients, but were also distinctly observed in the inner retinal layers, including the RNFL, INL, and GCL( 25 – 27 ). However, a few studies have failed to confirm the presence of phosphorylated tau in the retinas of human AD patients( 28 , 29 ). In summary, the relationship between abnormal retinal tau forms, AD brain pathology, and cognitive function remains insufficiently studied. Moreover, no current research has examined the correlation between circulating levels of total-tau levels and retinal thickness. The MR approach emphasizes the necessity of a robust correlation between genetic variants and exposure, while ensuring their lack of any causal impact on the outcome. In contrast, genetic colocalization analysis aims to identify a singular genetic variant of SNP that may jointly influence both exposure and outcome. To further explore the genetic causal association between RNFL, GC-IPL thickness and AD, circulating levels of total-tau from an alternative conceptual framework as a supplement to MR, we conducted the GWAS-GWAS genetic colocalization analysis for RNFL, GC-IPL thickness and AD, circulating levels of total-tau. The findings from genetic colocalization revealed no shared causal variants for RNFL, GC-IPL thickness and AD, circulating levels of total-tau either, indicating a low likelihood of a causal linkage between them. It's worth noting that our research did not refute the co-occurrence observed in previous studies regarding RNFL, GC-IPL thickness and AD. Nevertheless, such thinning of inner retinal structures may not be attributed to the same genetic etiology as AD. Mathew et al. reported a positive correlation between RNFL thickness in both eyes and brain volumes, including global volume, temporal lobe, and hippocampal volume, among individuals with subjective and objective cognitive decline. However, no significant association was observed between RNFL thickness and the progression of cognitive decline or dementia with AD( 30 ). Another recent prospective cohort study conducted by Grace et al. investigated the association between RNFL thickness and the incidence of all-cause dementia in a European population comprising 6,239 participants. The glaucoma detection with variable corneal compensation (GDx-VCC) and Heidelberg Retinal Tomograph II (HRT II) methods were employed for assessment. However, none of the four calculation models used demonstrated any significant correlation between RNFL thickness derived from either GDx-VCC or HRT II and the risk of incident dementia. These findings suggested that RNFL thickness could not serve as a powerful predictor for AD( 31 ). Morover, thinning of the RNFL and GC-IPL should not be considered as a specific manifestation of AD, as its presence has also been observed in conjunction with other conditions, such as glaucoma, Parkinson's disease, atypical parkinsonism, and Lewy body dementia( 32 – 35 ). The accumulation of Aβ protein in the retina has also been observed in animal models of glaucoma and in postmortem studies of glaucoma patients( 36 ). Previous studies have reported an increased risk of glaucoma among AD patients( 37 ), although contradictory results also existed( 38 ). The literature suggested that retinal thinning and deposition of Aβ protein may be present in various age-related conditions with highly intricate underlying mechanisms. It is intriguing that a recent MR analysis conducted by Currant et al. has suggested that the genetic causal pathway for glaucoma itself may not be directly responsible for the reduction of inner retinal thickness. However, this analysis has highlighted the potential role of elevated intraocular pressure as a causative factor for retinal thinning. These findings implied that genetic influences on retinal thinning in glaucoma were primarily mediated through increased intraocular pressure, which may often occur in the older individuals with cognitive impairment( 10 ). As yet, there’s a paucity of literature elucidating the relationship between retinal thinning in AD patients and their relationship with intraocular pressure. In summary, the precise biological functions as well as the underlying mechanisms of RNFL, GC-IPL thinning with AD, necessitate further investigation in future studies. Our study has several notable strengths. Firstly, it represented the inaugural bidirectional MR investigation into the causality between RNFL, GC-IPL thickness and AD, circulating levels of total-tau. Secondly, the MR along with genetic colocalization analyses provided more robust outcomes by addressing potential biases, including reverse causation and residual confounding, which were often encountered in conventional observational studies. Thirdly, this study leveraged GWAS datasets with a substantial sample size and meticulous precision for IVs identification. Nevertheless, it is essential to acknowledge several limitations. Firstly, the absence of initial sociodemographic data, such as comorbidities and staging of course of disease, may hinder the feasibility of conducting further subgroup analysis. Additionally, the GWAS datasets pertaining to retinal changes and AD, circulating levels of total-tau were originated from individuals of European ancestry, thus limiting the generalizability of our findings to other ethnic groups. 5. Conclusion Our MR and genetic colocalization results do not support a genetic causal link between retinal thickness, AD, and circulating total-tau levels, despite previous observational studies suggesting an association between retinal thinning and higher AD risk. Further research is needed to clarify the relationship between RNFL and GC-IPL thickness and AD, as well as the underlying biological mechanisms. Declarations Conflict of interest The authors declare that they have no competing interests. Ethics approval and consent to participate The used data were publicly available and approved by their corresponding institutions. An ethics approval is not required for the present study. No animal subjects were used in this work. Consent for publication Not applicable. Funding This work was supported by the following funding: National Natural Science Foundation of China (Grant No. 81601139) and Natural Science Foundation of Hunan Province (Grant No. 2024JJ5576). Author Contribution Dandan Sheng: Writing–review & editing, Writing–original draft, Visualization, Methodology, Investigation, Formal analysis. Song Wang: Writing–review & editing, Writing–original draft, Data curation, Methodology, Formal analysis. Zheng Xiao: Writing–review & editing, Methodology. Weiping Liu and Bo Xiao: Writing–review & editing, Conceptualization. Luo Zhou: Writing–review & editing, Writing–original draft, Visualization, Funding acquisition, Project administration, Conceptualization. Acknowledgements We express our gratitude to the authors and participants who contributed to the original GWAS and release the public summary datasets. Data Availability The data supporting this study's findings are available from the corresponding author upon reasonable request. References (2022) : 2022 Alzheimer's disease facts and figures. Alzheimers Dement. 18:700–789 Frise´n CBJ J, Lothian C, Lendahl U (1998) Central nervous system stem cells in the embryo and adult. CMLS Cell Mol Life Sci 54:9 Gupta VB, Chitranshi N, den Haan J, Mirzaei M, You Y, Lim JK et al (2021) Retinal changes in Alzheimer's disease- integrated prospects of imaging, functional and molecular advances. Prog Retin Eye Res 82:100899 den Haan J, van de Kreeke JA, Konijnenberg E, ten Kate M, den Braber A, Barkhof F et al (2019) Retinal thickness as a potential biomarker in patients with amyloid-proven early- and late-onset Alzheimer's disease. Alzheimers Dement (Amst) 11:463–471 Sánchez D, Castilla-Marti M, Marquié M, Valero S, Moreno-Grau S, Rodríguez-Gómez O et al (2020) Evaluation of macular thickness and volume tested by optical coherence tomography as biomarkers for Alzheimer's disease in a memory clinic. Sci Rep 10:1580 Ito Y, Sasaki M, Takahashi H, Nozaki S, Matsuguma S, Motomura K et al (2020) Quantitative Assessment of the Retina Using OCT and Associations with Cognitive Function. Ophthalmology 127:107–118 Sáez ME, García-Sánchez A, de Rojas I, Alarcón-Martín E, Martínez J, Cano A et al (2024) Genome-wide association study and polygenic risk scores of retinal thickness across the cognitive continuum: data from the NORFACE cohort. Alzheimers Res Ther 16:38 Kashani AH, Asanad S, Chan JW, Singer MB, Zhang J, Sharifi M et al (2021) Past, present and future role of retinal imaging in neurodegenerative disease. Prog Retin Eye Res 83:100938 Emdin CA, Khera AV, Kathiresan S (2017) Mendelian Randomization. JAMA 318:19 Currant H, Hysi P, Fitzgerald TW, Gharahkhani P, Bonnemaijer PWM, Senabouth A et al (2021) Genetic variation affects morphological retinal phenotypes extracted from UK Biobank optical coherence tomography images. PLoS Genet 17:e1009497 Sarnowski C, Ghanbari M, Bis JC, Logue M, Fornage M, Mishra A et al (2022) Meta-analysis of genome-wide association studies identifies ancestry-specific associations underlying circulating total tau levels. Commun Biol 5:336 Kunkle BW, Grenier-Boley B, Sims R, Bis JC, Damotte V, Naj AC et al (2019) Genetic meta-analysis of diagnosed Alzheimer's disease identifies new risk loci and implicates Abeta, tau, immunity and lipid processing. Nat Genet 51:414–430 Bellenguez C, Kucukali F, Jansen IE, Kleineidam L, Moreno-Grau S, Amin N et al (2022) New insights into the genetic etiology of Alzheimer's disease and related dementias. Nat Genet 54:412–436 Zuber V, Grinberg NF, Gill D, Manipur I, Slob EAW, Patel A et al (2022) Combining evidence from Mendelian randomization and colocalization: Review and comparison of approaches. Am J Hum Genet 109:767–782 Hu J, Lu J, Lu Q, Weng W, Guan Z, Wang Z (2023) Mendelian randomization and colocalization analyses reveal an association between short sleep duration or morning chronotype and altered leukocyte telomere length. Commun Biol 6:1014 Mutlu U, Colijn JM, Ikram MA, Bonnemaijer PWM, Licher S, Wolters FJ et al (2018) Association of Retinal Neurodegeneration on Optical Coherence Tomography With Dementia: A Population-Based Study. JAMA Neurol 75:1256–1263 Ko F, Muthy ZA, Gallacher J, Sudlow C, Rees G, Yang Q et al (2018) Association of Retinal Nerve Fiber Layer Thinning With Current and Future Cognitive Decline: A Study Using Optical Coherence Tomography. JAMA Neurol 75:1198–1205 Asanad S, Ross-Cisneros FN, Nassisi M, Barron E, Karanjia R, Sadun AA (2019) : The Retina in Alzheimer's Disease: Histomorphometric Analysis of an Ophthalmologic Biomarker. Invest Opthalmology Visual Sci. 60 Koronyo-Hamaoui M, Koronyo Y, Ljubimov AV, Miller CA, Ko MK, Black KL et al (2011) Identification of amyloid plaques in retinas from Alzheimer's patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. NeuroImage 54(Suppl 1):S204–217 van Koolwijk LME, Despriet DDG, Van Duijn CM, Oostra BA, van Swieten JC, de Koning I et al (2009) Association of cognitive functioning with retinal nerve fiber layer thickness. Investig Ophthalmol Vis Sci 50:4576–4580 van de Kreeke JA, Nguyen HT, den Haan J, Konijnenberg E, Tomassen J, den Braber A et al (2019) Retinal layer thickness in preclinical Alzheimer's disease. Acta Ophthalmol 97:798–804 Jia J, Ning Y, Chen M, Wang S, Yang H, Li F et al (2024) Biomarker Changes during 20 Years Preceding Alzheimer's Disease. N Engl J Med 390:712–722 Nam E, Lee Y-B, Moon C, Chang K-A (2020) : Serum Tau Proteins as Potential Biomarkers for the Assessment of Alzheimer's Disease Progression. Int J Mol Sci. 21 Mattsson N, Zetterberg H, Janelidze S, Insel PS, Andreasson U, Stomrud E et al (2016) Plasma tau in Alzheimer disease. Neurology 87:1827–1835 den Haan J, Morrema THJ, Verbraak FD, de Boer JF, Scheltens P, Rozemuller AJ et al (2018) Amyloid-beta and phosphorylated tau in post-mortem Alzheimer's disease retinas. Acta Neuropathol Commun 6:147 Hart de Ruyter FJ, Morrema THJ, den Haan J, Twisk JWR, de Boer JF, Scheltens P et al (2023) Phosphorylated tau in the retina correlates with tau pathology in the brain in Alzheimer's disease and primary tauopathies. Acta Neuropathol 145:197–218 Walkiewicz G, Ronisz A, Van Ginderdeuren R, Lemmens S, Bouwman FH, Hoozemans JJM et al (2024) Primary retinal tauopathy: A tauopathy with a distinct molecular pattern. Alzheimer's Dement J Alzheimer's Assoc 20:330–340 Ho C-Y, Troncoso JC, Knox D, Stark W, Eberhart CG (2014) Beta-amyloid, phospho-tau and alpha-synuclein deposits similar to those in the brain are not identified in the eyes of Alzheimer's and Parkinson's disease patients. Brain Pathol 24:25–32 Williams EA, McGuone D, Frosch MP, Hyman BT, Laver N, Stemmer-Rachamimov A (2017) Absence of Alzheimer Disease Neuropathologic Changes in Eyes of Subjects With Alzheimer Disease. J Neuropathol Exp Neurol 76:376–383 Mathew S, WuDunn D, Mackay DD, Vosmeier A, Tallman EF, Deardorff R et al (2023) Association of Brain Volume and Retinal Thickness in the Early Stages of Alzheimer's Disease. J Alzheimers Dis 91:743–752 Yin GS, van der Heide F, Littlejohns TJ, Kuźma E, Hayat S, Brayne C et al (2023) Association Between Retinal Nerve Fiber Layer Thickness and Incident Dementia in the European Prospective Investigation into Cancer in Norfolk Cohort. J Alzheimers Dis 95:691–702 Tatham AJ, Medeiros FA (2017) Detecting Structural Progression in Glaucoma with Optical Coherence Tomography. Ophthalmology 124:S57–S65 Chrysou A, Jansonius NM, van Laar T (2019) Retinal layers in Parkinson's disease: A meta-analysis of spectral-domain optical coherence tomography studies. Parkinsonism Relat Disord 64:40–49 Ma X, Wang Y, Wang N, Zhang R (2021) Retina thickness in atypical parkinsonism: a systematic review and meta-analysis. J Neurol 269:1272–1281 Moreno-Ramos T, Benito-León J, Villarejo A, Bermejo-Pareja F (2013) Retinal Nerve Fiber Layer Thinning in Dementia Associated with Parkinson's Disease, Dementia with Lewy Bodies, and Alzheimer's Disease. J Alzheimers Dis 34:659–664 Guo LST, Luong V, Wood N, Cheung W, Maass A, Ferrari G, Russo-Marie F, Sillito AM, Cheetham ME, Moss SE, Fitzke FW, Cordeiro MF (2007) Targeting amyloid-beta in glaucoma treatment. Proc Natl Acad Sci U S A 104:33 Helmer C, Malet F, Rougier M-B, Schweitzer C, Colin J, Delyfer M-N et al (2013) Is there a link between open-angle glaucoma and dementia? Annals of Neurology.n/. a-n/a Janssen S, Jansonius NM, Bouwman F, Verbraak FD, Bergen AA (2015): Systematic review of the association between Alzheimer’s disease and chronic glaucoma. Clinical Ophthalmology Additional Declarations No competing interests reported. Supplementary Files SupplementaryTableS1S4.docx SupplementaryFigureS1S6.pdf 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-5869988","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":405584053,"identity":"83dc5bf1-3def-442b-af23-d796e180d018","order_by":0,"name":"Dandan Sheng","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Dandan","middleName":"","lastName":"Sheng","suffix":""},{"id":405584054,"identity":"55ee554e-b90f-4c87-a38a-76caee30f20d","order_by":1,"name":"Song Wang","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Song","middleName":"","lastName":"Wang","suffix":""},{"id":405584055,"identity":"409b643d-9b3f-414f-b54c-a5f6b87bec09","order_by":2,"name":"Zheng Xiao","email":"","orcid":"","institution":"First Hospital of Changsha","correspondingAuthor":false,"prefix":"","firstName":"Zheng","middleName":"","lastName":"Xiao","suffix":""},{"id":405584056,"identity":"bebdca8d-ba03-4ed4-980f-b4a7fdfc9797","order_by":3,"name":"Weiping Liu","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Weiping","middleName":"","lastName":"Liu","suffix":""},{"id":405584057,"identity":"26042a20-4018-44f1-baf8-3c4e84e92d6a","order_by":4,"name":"Bo Xiao","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Xiao","suffix":""},{"id":405584058,"identity":"ad0cfa3b-8737-43d8-8089-b2612c909868","order_by":5,"name":"Luo Zhou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxElEQVRIiWNgGAWjYDACCSB+wMAgB+GxEaslgYHBmHQtiQ1Ea5Gf3XxMIrHNJr2//4wBw4eywwz8sxvwa2GccywNqCUtd8aNHAPGGecOM0jcOYBfC7NEjhlQy+HcDRI8Bsy8bYcZDCQS8Gthg2pJN+A/Y8D8lxgtPFAtCQYMOQbMjMRokZBIS7ZIOJdmOONGWsHBnnPpPBI3CGiRn5F88MaHMht5/v7DGx/8KLOW459BQAsKOAByKQnqR8EoGAWjYBTgAgA16DyNQso5bAAAAABJRU5ErkJggg==","orcid":"","institution":"Central South University","correspondingAuthor":true,"prefix":"","firstName":"Luo","middleName":"","lastName":"Zhou","suffix":""}],"badges":[],"createdAt":"2025-01-21 04:23:03","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5869988/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5869988/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":74693944,"identity":"39de07e7-bba1-48d5-9790-ac4a231a8c7d","added_by":"auto","created_at":"2025-01-24 19:34:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":276199,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMendelian randomization and genetic colocalization analyses exploring the causality between the thickness of RNFL, GC-IPL and AD, circulating levels of total-tau.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSNPs: single-nucleotide polymorphisms; IVs: instrumental variables; RNFL thickness: Retinal nerve fibre layer thickness; GC-IPL thickness: Ganglion cell-inner plexiform layer thickness; IVW: inverse-variance weighted; MR-Egger: Mendelian randomization Egger; RAPS: Robust adjusted profile score; MR-PRESSO: Mendelian randomization pleiotropy residual sum and outlier.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/b65249de845df8e2b64968fc.png"},{"id":74694361,"identity":"dabecb36-20b3-4f60-b327-793c44aee70f","added_by":"auto","created_at":"2025-01-24 19:42:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":479006,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForest plots of Mendelian randomization causal estimates between the thickness of RNFL, GC-IPL and AD in discovery stage.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSNPs, single-nucleotide polymorphisms; IVW, inverse-variance weighted; AD, Alzheimer's disease; RNFL, retinal nerve fiber layer; GC-IPL, ganglion cell-inner plexiform layer.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/d70dc631ab1553683137b116.png"},{"id":74693946,"identity":"c6bfac15-9e62-4e6f-8852-c9866ded9fce","added_by":"auto","created_at":"2025-01-24 19:34:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":487284,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForest plots of Mendelian randomization causal estimates between the thickness of RNFL, GC-IPL and circulating levels of total-tau\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSNPs, single-nucleotide polymorphisms; IVW, inverse-variance weighted; RNFL, retinal nerve fiber layer; GC-IPL, ganglion cell-inner plexiform layer.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/597c2cad6b6a15ebade5d72c.png"},{"id":74693949,"identity":"a4008880-f2e5-46fc-8285-54e8eb6c5ccc","added_by":"auto","created_at":"2025-01-24 19:34:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1172027,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRegional association plots for genetic colocalization analysis of RNFL, GC-IPL thickness and AD, circulating levels of total-tau. The lead SNP is shown as a purple diamond. SNPs within ±50kb of the corresponding instrumental SNP were included.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA: RNFL thickness and AD (Discovery), B: GC-IPL thickness and AD (Discovery), C: RNFL thickness and AD (Replication), D: GC-IPL thickness and AD (Replication), E：RNFL thickness and circulating levels of total-tau, F: GC-IPL thickness and circulating levels of total-tau.\u003c/p\u003e\n\u003cp\u003eAD, Alzheimer's disease; RNFL, retinal nerve fiber layer; GC-IPL, ganglion cell-inner plexiform layer.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/110872132f248f82cee58260.png"},{"id":74702267,"identity":"5dd3e903-a4ff-45c6-a489-aff6631d6075","added_by":"auto","created_at":"2025-01-25 01:01:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3462350,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/4b3e888a-e03d-461b-a683-14e3b81372fb.pdf"},{"id":74693947,"identity":"abbf8e51-b03d-4a1f-b29e-b20e4f675fbe","added_by":"auto","created_at":"2025-01-24 19:34:24","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":120284,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTableS1S4.docx","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/7e7b4d97eeb3dd12c0caf084.docx"},{"id":74693950,"identity":"ccbc5187-e12a-489b-ae8a-26007fa821f7","added_by":"auto","created_at":"2025-01-24 19:34:24","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":689450,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigureS1S6.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5869988/v1/29d4f4df219afeb3b8e9bccd.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Unraveling the association between retinal thickness and Alzheimer's disease, and circulating total-tau levels: Insights from genetic evidence","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eWith the increase in life expectancy, there is an anticipated upsurge in the prevalence of neurodegenerative diseases, among which Alzheimer's disease (AD) stands out as the most prominent form of neurodegenerative dementia observed in older individuals. The pathological hallmark of this condition involves the progressive accumulation of misfolded amyloid-beta (Aβ) and tau proteins in the brain during an extended, subtle, and insidious pre-symptomatic phase, ultimately leading to a gradual decline in cognitive performance(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Early identification and prevention can help limit progression from asymptomatic to clinical stages.\u003c/p\u003e \u003cp\u003eGiven that the retina is an extension of the brain and shares a common embryonic origin with the central nervous system, there is significant interest in utilizing the expanding retinal imaging technologies to understand, diagnose, and monitor AD(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Optical coherence tomography (OCT) is widely used due to its high spatial resolution and relatively low cost. The alterations of retinal structures, particularly the thinning of the retinal nerve fiber layer (RNFL) and the ganglion cell-inner plexiform layer (GC-IPL), have attracted the attentions to be associated with a decline in cognitive performance based on multiple previous studies, which have identified measurable changes in retinal thickness during an early stage of AD(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). However, in contrast to the findings mentioned above, several well-conducted studies have reported minimal changes in retinal thickness (RNFL, GCL-IPL) associated with early dementia or mild cognitive impairment. Additionally, no significant associations were found between retinal thickness and CSF biomarkers (Aβ-42, tau-181, phosphorylated tau, and the tau-181/Aβ1\u0026ndash;42 ratio)(\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Maria et al. recently investigated the association between cognitive decline spanning both AD and non-AD cases, and retinal thickness in a large Spanish cohort. Utilizing genome-wide association study (GWAS) and polygenic risk score analysis, their findings did not reveal a significant genetic association between dementia and retinal thickness, either of AD or non-AD dementia types(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). It is important to note that the outputs of OCT are influenced by a complex interplay of technical parameters that vary across vendors and operators, potentially leading to inconsistent estimates of retinal layer thickness. Additionally, due to limitations in study design and imaging equipment, the field of view is frequently restricted to the macula or optic disc, thereby excluding substantial portions of the peripheral retina from analysis. Furthermore, a significant challenge in longitudinal studies is the co-registration of images across different subjects and time points, which is a critical factor in the analysis of retinal imaging data. Finally, although OCT devices provide micron-level resolution, the reproducibility of measurements is heavily contingent upon the experience and training of the operators(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Thus, current researches are insufficient to fully elucidate the causal relationship between AD, pathological protein deposits, and the thinning of RNFL and GC-IPL.\u003c/p\u003e \u003cp\u003eMendelian Randomization (MR) is a roubust research methodology that relies on single nucleotide polymorphisms (SNPs) from GWAS to create instrumental variables (IVs) for establishment of causation. The random assignment of alleles to offspring effectively mitigates confounding bias. MR employs IVs to roubustly identify risk factors(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). On the other hand, genetic colocalization analysis is commonly employed to ascertain whether two phenotypes are driven by the same causal variant within a specific gene region, thereby bolstering the evidence of a genetic causal association between the exposure and outcome. In this study, we aimed to comprehensively assess the causality between RNFL, GC-IPL thickness and AD, circulating total-tau through MR and genetic colocalization analyses.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study design\u003c/h2\u003e \u003cp\u003eWe first conducted a bidirectional two-sample MR study to ascertain the causality between the traits of retinal thickness and AD, and its biomarker circulating total-tau. In the forward MR analysis, we assessed the causal impact of RNFL and GC-IPL thickness on AD and circulating total-tau level, while in the reverse MR analysis, we examined the causal impact of AD, circulating total-tau level on RNFL and GC-IPL thickness. Ensuring the reliability of MR results relies on fulfilling three fundamental assumptions: (a) the correlation hypothesis, which posits that IVs possess significant genetic connections to the exposure; (b) the independence hypothesis, which stipulates that IVs should not be directly associated with confounding factors related to exposure or outcome; (c) the exclusivity hypothesis, which implies that IVs should not exhibit direct associations with the outcome(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). To investigate the presence of a co-driven causal variant, we subsequently conducted additional genetic colocalization analysis. The datasets utilized in the present study were obtained from publicly accessible repositories, and ethical approval was previously acquired from the local ethics committee during the original GWAS. No additional ethical approval or informed consent was necessary. The flowchart of the study is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data sources\u003c/h2\u003e \u003cp\u003eThe summary-level datasets of RNFL and GC-IPL thickness were derived from the recent largest GWAS investigating inner retinal morphology using phenotypes obtained from OCT images of 31,434 UK Biobank participants authored by Kunkle et al., encompassing 9,121,075 genetic variants of SNPs(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Summary statistics regarding circulating levels of total-tau were derived from a large-scale meta-analysis of GWAS involving 14,721 individuals of European ancestry with 8,360,926 SNPs(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). For the discovery stage, the summary-level dataset of the AD was derived from a GWAS meta-analysis conducted by International Genomics of Alzheimer\u0026rsquo;s Project (IGAP), which included 63,926 participants (21,982 cases and 41,944 controls) with 10,528,610 SNPs(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). For further validation, an additional dataset on AD with a sample size of 85,934 participants (39,106 cases and 46,828 controls) with 20,921,626 SNPs was utilized from a GWAS conducted by Bellenguez et al. for replication MR validation(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). The participants included in our datasets are exclusively of European ancestry. The GWAS-GWAS genetic colocalization analysis was subsequently conducted within these datasets. The detailed information of GWAS datasets utilized in the MR and genetic colocalization is summarized in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Mendelian randomization analysis\u003c/h2\u003e \u003cp\u003eWe employed a threshold of \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;5E-08 and pairwise r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 within a 10,000 kb window for the selection of IVs in both forward and reverse MR analyses, using the European population as the reference group. For the two traits related to retinal thickness, we applied a more lenient threshold of \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;5E-06 to ensure an adequate number of SNPs were available as IVs. SNPs with a minor allele frequency less than 0.01 and palindromic allele frequencies were excluded. All SNPs designated as IVs for scrutiny in this investigation exhibited a minimum F-statistic of 10 to counteract the influence of weak genetic instruments. PhenoScanner was utilized to check all IVs to remove potential confounding factors. We employed the inverse-variance weighted (IVW) method as the primary MR estimate. Several complementary methods, including MR-Egger, Robust adjusted profile score (RAPS), Maximum likelihood, and weighted median were also utilized for validation. The IVs are described in detail in Supplementary Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Sensitivity analysis\u003c/h2\u003e \u003cp\u003eWe employed Cochran's Q and the MR-Egger intercept tests to assess heterogeneity and pleiotropy within IVs. In order to identify potential SNP outliers among the IVs, we further utilized the assessment of MR pleiotropy residual sum and outlier (MR-PRESSO) as well as leave-one-out analysis. If an SNP outlier was detected through either MR-PRESSO or leave-one-out, it was excluded, after which we may perform a re-evaluation of the MR and sensitivity analysis to obtain robust and definitive findings. After applying the Bonferroni correction, a significance level of \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0125 (0.05/4) would be considered statistically significant. All MR statistical analysis in this study were performed using the TwoSampleMR package (Version 0.5.7) in the RStudio (Version 2023.06.0).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Genetic colocalization analysis\u003c/h2\u003e \u003cp\u003eTo determine if the SNPs associated with the exposure also impact the outcome, we additionally conducted a genetic colocalization analysis using the aforementioned GWAS datasets. This approach allows us to assess genetic correlations by assigning prior probabilities to random SNPs within the genomic region of interest, considering their associations with either the exposure, outcome, or both(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). We extracted SNPs within a\u0026thinsp;\u0026plusmn;\u0026thinsp;50kb gene window surrounding the identified IVs for each potential causal variant from both exposure and outcome GWAS datasets. The posterior probability was considered under five hypotheses: H0 indicating no association with either trait; H1 indicating an association with trait 1; H2 indicating an association with trait 2; H3 indicating an association with both traits, involving two independent SNPs; H4 indicating an association with both traits, linked to one shared SNP. The colocalization was considered positive when the posterior probability H4 exceeded 0.75, indicating strong evidence for the presence of a shared causal variant(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). The R package Coloc (Version 5.2.3) was employed for conducting genetic colocalization analysis.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Causal relationship between retinal thickness and AD\u003c/h2\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e3.1.1 Results of Mendelian randomization in the discovery stage\u003c/h2\u003e \u003cp\u003eIn the discovery stage, the IVW estimates derived from IGAP dataset found no evidence supporting a causal effect of genetically predicted RNFL and GC-IPL thickness on the risk of AD (RNFL-AD: odds ratio [OR]\u0026thinsp;=\u0026thinsp;1.015, 95% confidence interval [CI]\u0026thinsp;=\u0026thinsp;0.987\u0026ndash;1.043, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.305; GC-IPL-AD: OR\u0026thinsp;=\u0026thinsp;1.014, 95% CI\u0026thinsp;=\u0026thinsp;0.991\u0026ndash;1.037, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.237). Likewise, there was no indication of any causal effect of genetically predicted AD on RNFL and GC-IPL thickness (AD-RNFL: OR\u0026thinsp;=\u0026thinsp;0.930, 95% CI\u0026thinsp;=\u0026thinsp;0.824\u0026ndash;1.050, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.239; AD-GC-IPL: OR\u0026thinsp;=\u0026thinsp;0.982, 95% CI\u0026thinsp;=\u0026thinsp;0.849\u0026ndash;1.136, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.807). The estimates calculated through IVW were further supported by alternative methods, including MR-Egger, RAPS, Maximum likelihood, and weighted median. In the MR-Egger method, we observed a nominal causal effect of genetically predicted AD on RNFL thickness (P\u0026thinsp;=\u0026thinsp;0.041). However, after applying the Bonferroni correction, this estimate did not reach statistical significance. The forest plots illustrating the summarized causal estimates are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. These findings remained robust across all sensitivity analysis, and detailed results are summarized in Supplementary Table S3. Scatter and leave-one-out plots in the the discovery MR analysis are presented in Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e and S2.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2 Results of Mendelian randomization in the replication stage\u003c/h2\u003e \u003cp\u003eThe replication MR analysis consistently revealed no causal association between RNFL, GC-IPL thickness and AD in either direction, as all MR causal estimates yielded a \u003cem\u003eP\u003c/em\u003e-value greater than 0.0125 in IVW (RNFL-AD: OR\u0026thinsp;=\u0026thinsp;1.001, 95% CI\u0026thinsp;=\u0026thinsp;0.986\u0026ndash;1.017, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.883; GC-IPL-AD: OR\u0026thinsp;=\u0026thinsp;1.008, 95% CI\u0026thinsp;=\u0026thinsp;0.998\u0026ndash;1.018, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.129; AD-RNFL: OR\u0026thinsp;=\u0026thinsp;0.943, 95% CI\u0026thinsp;=\u0026thinsp;0.826\u0026ndash;1.077, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.389; AD-GC-IPL: OR\u0026thinsp;=\u0026thinsp;0.990, 95% CI\u0026thinsp;=\u0026thinsp;0.816-1.200, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.917). The causal eatimates calculated through MR-Egger, Robust adjusted profile score (RAPS), Maximum likelihood, and weighted median further supported the negative finding. The summarized causal estimates in the replication MR analysis are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Furthermore, the results from the replication MR analysis remained robust across all sensitivity analysis, as summarized in Supplementary Table S3. Supplementary Figure S3 and S4 depict scatter and leave-one-out plots illustrating the resluts of the replication MR analysis.\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\u003eThe causal estimates between RNFL, GC-IPL thickness and AD in replication MR analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\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=\"char\" char=\".\" 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=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExposure\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOutcome\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003enSNPs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOR (95%CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRNFL thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIVW\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.001(0.986\u0026ndash;1.017)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.883\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWeighted median\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.010(0.989\u0026ndash;1.031)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.365\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMaximum likelihood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.001(0.987\u0026ndash;1.016)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.870\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRAPS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.003(0.987\u0026ndash;1.020)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.715\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMR Egger\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.024(0.979\u0026ndash;1.071)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.322\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGC-IPL thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIVW\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.008(0.998\u0026ndash;1.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.129\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWeighted median\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.009(0.995\u0026ndash;1.023)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.202\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMaximum likelihood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.008(0.998\u0026ndash;1.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.121\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRAPS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.006(0.996\u0026ndash;1.017)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.234\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMR Egger\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.022(0.994\u0026ndash;1.052)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.140\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRNFL thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIVW\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.943(0.826\u0026ndash;1.077)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.389\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWeighted median\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.979(0.807\u0026ndash;1.188)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.830\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMaximum likelihood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.943(0.834\u0026ndash;1.067)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.350\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRAPS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.948(0.830\u0026ndash;1.084)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.435\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMR Egger\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.842(0.667\u0026ndash;1.062)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.153\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGC-IPL thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eInverse variance weighted\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.990(0.816-1.200)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.917\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWeighted median\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.082(0.831\u0026ndash;1.410)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.559\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMaximum likelihood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.990(0.842\u0026ndash;1.164)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.902\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRAPS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.966(0.799\u0026ndash;1.169)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.722\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMR Egger\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.905(0.644\u0026ndash;1.271)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.566\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003enSNPs, number of single-nucleotide polymorphisms; OR, odds ratio; CI, confidence interval; AD, Alzheimer's disease; RNFL, retinal nerve fiber layer; GC-IPL, ganglion cell-inner plexiform layer; IVW, Inverse variance weighted; RAPS, Robust adjusted profile score; MR-Egger, Mendelian randomization Egger.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Causal Relationship between retinal thickness and circulating total-tau\u003c/h2\u003e \u003cp\u003eThe IVW analysis revealed that RNFL thickness (OR, 0.993 [0.977\u0026ndash;1.008]; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.355) and GC-IPL thickness (OR, 1.001 [95% CI, 0.992\u0026ndash;1.009]; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.908) had no significant causal effect on circulating levels of total-tau. Similarly, there was no evidence of a causal effect of genetically predicted circulating total-tau levels on RNFL thickness (OR, 0.989 [0.886\u0026ndash;1.105]; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.849) or GC-IPL thickness (OR, 1.129 [95% CI, 0.977\u0026ndash;1.305]; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.099). These estimates derived from the IVW method were corroborated by alternative approaches, including MR-Egger, RAPS, Maximum Likelihood, and weighted median methods. The forest plots displaying the summarized causal estimates are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. These findings remained robust across all sensitivity analyses, with detailed results summarized in Supplementary Table S4. Scatter and leave-one-out plots from the MR analysis are presented in Supplementary Figures S5 and S6.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Results of genetic colocalization analysis\u003c/h2\u003e \u003cp\u003eThe posterior probability H4 obtained from GWAS-GWAS genetic colocalization analysis within the datasets of retianl thickness and AD ranged from 0.006 to 0.047, which fell below the predefined threshold of 0.75, suggesting a limited possibility for the presence of a shared causal variant underlying both the exposure and outcome traits. The regional association plots for colocalization analysis are further elaborated in Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. The posterior probabilities H0-H4 are presented in Supplementary Table S5.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eOur study represents a pioneering effort to elucidate the causal relationship between AD, the biomarker of circulating levels of total-tau, and retinal thickness from a genetic perspective, effectively accounting for confounding factors. Through bidirectional MR analysis, we found no evidence of a genetically predicted causal link between RNFL, GC-IPL thickness, in relation to circulating levels of total-tau and AD.\u003c/p\u003e \u003cp\u003eNumerous early studies have identified a potential link between AD and retinal thickness; however, the results often lack consistency. Some studies have generally indicated that individuals with AD have exhibited thinner RNFL and GC-IPL in comparison to non-AD control groups. A population-based prospective cohort study conducted by Mutlu et al. has demonstrated a significant association between decreased RNFL thickness measured using OCT and an increased risk of AD dementia. Additionally, thinner GC-IPL thickness has been found to be correlated with prevalent dementia including AD(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Ko et al. have similarly discovered that thinner RNFL may be linked to poorer cognitive performance in individuals without neurodegenerative diseases and an increased likelihood of future cognitive decline within a substantial community-based cohort of healthy individuals(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). A recent histopathologic, morphometric analysis of human postmortem retinas included eight patients with AD and eleven age-matched controls revealed a remarkable thinning of both the RNFL and GC-IPL correlated with AD(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Furthermore, Aβ protein deposition has been observed in the retinas of both AD patients and AD mouse models(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). These observational findings seemed to hint that the retinal thickness may serve as a potential biomarker for early diagnosis and monitoring of AD. Relevant studies, however, have also presented divergent conclusions. Two recent population-based surveys conducted by Ito and Van Koolwijk et al. failed to yield evidence supporting the association between RNFL thickness and cognitive decline or dementia(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). A Dutch cohort study was conducted involving 165 cognitively healthy monozygotic twins who underwent [18F] flutemetamol positron emission tomography, which was visually assessed to determine the presence of cortical Aβ. The findings revealed no discernible differences in retinal layer thickness in the macula or peripapillary RNFL between Aβ\u0026thinsp;+\u0026thinsp;and Aβ- participants, indicating limited application of RNFL thickness as a biomarker for early detecting of AD(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Similarly, in a study involving 57 participants with CSF amyloid-positive and PET amyloid-positive late-onset Alzheimer's disease (LOAD), no correlation was found between Mini-mental State Examination (MMSE) scores and any retinal thickness parameters. In patients with posterior cortical atrophy, an AD variant primarily affecting the visual cortex, no association was reported between retinal thickness measurements and LOAD status, including MMSE scores(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). These inconsistent findings may be attributed to biases stemming from differences in study design, imaging equipment, and variability among researchers.\u003c/p\u003e \u003cp\u003eElevated levels of total-tau in CSF are a well-recognized biomarker of AD, with research indicating that a decline in total-tau levels may occur up to 10 years before the onset of AD(\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). However, studies examining the relationship between circulating levels of total-tau and AD have produced inconsistent results. Eunjoo et al. reported that circulating levels of total-tau in the mild AD group were significantly higher than those in age-matched controls and found a significant correlation between serum total-tau and both MMSE scores(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). In a study by Niklas et al., 1,284 participants were included, and associations between plasma-tau, diagnosis, CSF biomarkers, MRI measures, 18fluorodeoxyglucose-PET, and cognition were tested. The findings indicated that plasma-tau partially reflected AD pathology, but there was considerable overlap between normal aging and AD, particularly in individuals without dementia. Although group-level differences were observed, these results do not support plasma-tau as a reliable biomarker for individual AD diagnosis(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn AD patients, elevated levels of pathological tau subtypes have been detected in the retina, including increased microglial uptake of retinal phosphorylated tau. These tau subtypes were primarily localized to the synapse-rich inner plexiform layer (IPL) and outer plexiform layer (OPL) in AD patients, but were also distinctly observed in the inner retinal layers, including the RNFL, INL, and GCL(\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). However, a few studies have failed to confirm the presence of phosphorylated tau in the retinas of human AD patients(\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). In summary, the relationship between abnormal retinal tau forms, AD brain pathology, and cognitive function remains insufficiently studied. Moreover, no current research has examined the correlation between circulating levels of total-tau levels and retinal thickness.\u003c/p\u003e \u003cp\u003eThe MR approach emphasizes the necessity of a robust correlation between genetic variants and exposure, while ensuring their lack of any causal impact on the outcome. In contrast, genetic colocalization analysis aims to identify a singular genetic variant of SNP that may jointly influence both exposure and outcome. To further explore the genetic causal association between RNFL, GC-IPL thickness and AD, circulating levels of total-tau from an alternative conceptual framework as a supplement to MR, we conducted the GWAS-GWAS genetic colocalization analysis for RNFL, GC-IPL thickness and AD, circulating levels of total-tau. The findings from genetic colocalization revealed no shared causal variants for RNFL, GC-IPL thickness and AD, circulating levels of total-tau either, indicating a low likelihood of a causal linkage between them. It's worth noting that our research did not refute the co-occurrence observed in previous studies regarding RNFL, GC-IPL thickness and AD. Nevertheless, such thinning of inner retinal structures may not be attributed to the same genetic etiology as AD. Mathew et al. reported a positive correlation between RNFL thickness in both eyes and brain volumes, including global volume, temporal lobe, and hippocampal volume, among individuals with subjective and objective cognitive decline. However, no significant association was observed between RNFL thickness and the progression of cognitive decline or dementia with AD(\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). Another recent prospective cohort study conducted by Grace et al. investigated the association between RNFL thickness and the incidence of all-cause dementia in a European population comprising 6,239 participants. The glaucoma detection with variable corneal compensation (GDx-VCC) and Heidelberg Retinal Tomograph II (HRT II) methods were employed for assessment. However, none of the four calculation models used demonstrated any significant correlation between RNFL thickness derived from either GDx-VCC or HRT II and the risk of incident dementia. These findings suggested that RNFL thickness could not serve as a powerful predictor for AD(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Morover, thinning of the RNFL and GC-IPL should not be considered as a specific manifestation of AD, as its presence has also been observed in conjunction with other conditions, such as glaucoma, Parkinson's disease, atypical parkinsonism, and Lewy body dementia(\u003cspan additionalcitationids=\"CR33 CR34\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). The accumulation of Aβ protein in the retina has also been observed in animal models of glaucoma and in postmortem studies of glaucoma patients(\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Previous studies have reported an increased risk of glaucoma among AD patients(\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e), although contradictory results also existed(\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). The literature suggested that retinal thinning and deposition of Aβ protein may be present in various age-related conditions with highly intricate underlying mechanisms. It is intriguing that a recent MR analysis conducted by Currant et al. has suggested that the genetic causal pathway for glaucoma itself may not be directly responsible for the reduction of inner retinal thickness. However, this analysis has highlighted the potential role of elevated intraocular pressure as a causative factor for retinal thinning. These findings implied that genetic influences on retinal thinning in glaucoma were primarily mediated through increased intraocular pressure, which may often occur in the older individuals with cognitive impairment(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). As yet, there\u0026rsquo;s a paucity of literature elucidating the relationship between retinal thinning in AD patients and their relationship with intraocular pressure. In summary, the precise biological functions as well as the underlying mechanisms of RNFL, GC-IPL thinning with AD, necessitate further investigation in future studies.\u003c/p\u003e \u003cp\u003eOur study has several notable strengths. Firstly, it represented the inaugural bidirectional MR investigation into the causality between RNFL, GC-IPL thickness and AD, circulating levels of total-tau. Secondly, the MR along with genetic colocalization analyses provided more robust outcomes by addressing potential biases, including reverse causation and residual confounding, which were often encountered in conventional observational studies. Thirdly, this study leveraged GWAS datasets with a substantial sample size and meticulous precision for IVs identification. Nevertheless, it is essential to acknowledge several limitations. Firstly, the absence of initial sociodemographic data, such as comorbidities and staging of course of disease, may hinder the feasibility of conducting further subgroup analysis. Additionally, the GWAS datasets pertaining to retinal changes and AD, circulating levels of total-tau were originated from individuals of European ancestry, thus limiting the generalizability of our findings to other ethnic groups.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eOur MR and genetic colocalization results do not support a genetic causal link between retinal thickness, AD, and circulating total-tau levels, despite previous observational studies suggesting an association between retinal thinning and higher AD risk. Further research is needed to clarify the relationship between RNFL and GC-IPL thickness and AD, as well as the underlying biological mechanisms.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflict of interest\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e \u003cp\u003eThe used data were publicly available and approved by their corresponding institutions. An ethics approval is not required for the present study. No animal subjects were used in this work.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis work was supported by the following funding: National Natural Science Foundation of China (Grant No. 81601139) and Natural Science Foundation of Hunan Province (Grant No. 2024JJ5576).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eDandan Sheng: Writing\u0026ndash;review \u0026amp; editing, Writing\u0026ndash;original draft, Visualization, Methodology, Investigation, Formal analysis. Song Wang: Writing\u0026ndash;review \u0026amp; editing, Writing\u0026ndash;original draft, Data curation, Methodology, Formal analysis. Zheng Xiao: Writing\u0026ndash;review \u0026amp; editing, Methodology. Weiping Liu and Bo Xiao: Writing\u0026ndash;review \u0026amp; editing, Conceptualization. Luo Zhou: Writing\u0026ndash;review \u0026amp; editing, Writing\u0026ndash;original draft, Visualization, Funding acquisition, Project administration, Conceptualization.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eWe express our gratitude to the authors and participants who contributed to the original GWAS and release the public summary datasets.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data supporting this study's findings are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e(2022) : 2022 Alzheimer's disease facts and figures. Alzheimers Dement. 18:700\u0026ndash;789\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrise\u0026acute;n CBJ J, Lothian C, Lendahl U (1998) Central nervous system stem cells in the embryo and adult. CMLS Cell Mol Life Sci 54:9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGupta VB, Chitranshi N, den Haan J, Mirzaei M, You Y, Lim JK et al (2021) Retinal changes in Alzheimer's disease- integrated prospects of imaging, functional and molecular advances. Prog Retin Eye Res 82:100899\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eden Haan J, van de Kreeke JA, Konijnenberg E, ten Kate M, den Braber A, Barkhof F et al (2019) Retinal thickness as a potential biomarker in patients with amyloid-proven early- and late-onset Alzheimer's disease. Alzheimers Dement (Amst) 11:463\u0026ndash;471\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS\u0026aacute;nchez D, Castilla-Marti M, Marqui\u0026eacute; M, Valero S, Moreno-Grau S, Rodr\u0026iacute;guez-G\u0026oacute;mez O et al (2020) Evaluation of macular thickness and volume tested by optical coherence tomography as biomarkers for Alzheimer's disease in a memory clinic. Sci Rep 10:1580\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIto Y, Sasaki M, Takahashi H, Nozaki S, Matsuguma S, Motomura K et al (2020) Quantitative Assessment of the Retina Using OCT and Associations with Cognitive Function. Ophthalmology 127:107\u0026ndash;118\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS\u0026aacute;ez ME, Garc\u0026iacute;a-S\u0026aacute;nchez A, de Rojas I, Alarc\u0026oacute;n-Mart\u0026iacute;n E, Mart\u0026iacute;nez J, Cano A et al (2024) Genome-wide association study and polygenic risk scores of retinal thickness across the cognitive continuum: data from the NORFACE cohort. Alzheimers Res Ther 16:38\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKashani AH, Asanad S, Chan JW, Singer MB, Zhang J, Sharifi M et al (2021) Past, present and future role of retinal imaging in neurodegenerative disease. Prog Retin Eye Res 83:100938\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEmdin CA, Khera AV, Kathiresan S (2017) Mendelian Randomization. JAMA 318:19\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCurrant H, Hysi P, Fitzgerald TW, Gharahkhani P, Bonnemaijer PWM, Senabouth A et al (2021) Genetic variation affects morphological retinal phenotypes extracted from UK Biobank optical coherence tomography images. PLoS Genet 17:e1009497\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSarnowski C, Ghanbari M, Bis JC, Logue M, Fornage M, Mishra A et al (2022) Meta-analysis of genome-wide association studies identifies ancestry-specific associations underlying circulating total tau levels. Commun Biol 5:336\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKunkle BW, Grenier-Boley B, Sims R, Bis JC, Damotte V, Naj AC et al (2019) Genetic meta-analysis of diagnosed Alzheimer's disease identifies new risk loci and implicates Abeta, tau, immunity and lipid processing. Nat Genet 51:414\u0026ndash;430\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBellenguez C, Kucukali F, Jansen IE, Kleineidam L, Moreno-Grau S, Amin N et al (2022) New insights into the genetic etiology of Alzheimer's disease and related dementias. Nat Genet 54:412\u0026ndash;436\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZuber V, Grinberg NF, Gill D, Manipur I, Slob EAW, Patel A et al (2022) Combining evidence from Mendelian randomization and colocalization: Review and comparison of approaches. Am J Hum Genet 109:767\u0026ndash;782\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu J, Lu J, Lu Q, Weng W, Guan Z, Wang Z (2023) Mendelian randomization and colocalization analyses reveal an association between short sleep duration or morning chronotype and altered leukocyte telomere length. Commun Biol 6:1014\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMutlu U, Colijn JM, Ikram MA, Bonnemaijer PWM, Licher S, Wolters FJ et al (2018) Association of Retinal Neurodegeneration on Optical Coherence Tomography With Dementia: A Population-Based Study. JAMA Neurol 75:1256\u0026ndash;1263\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKo F, Muthy ZA, Gallacher J, Sudlow C, Rees G, Yang Q et al (2018) Association of Retinal Nerve Fiber Layer Thinning With Current and Future Cognitive Decline: A Study Using Optical Coherence Tomography. JAMA Neurol 75:1198\u0026ndash;1205\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsanad S, Ross-Cisneros FN, Nassisi M, Barron E, Karanjia R, Sadun AA (2019) : The Retina in Alzheimer's Disease: Histomorphometric Analysis of an Ophthalmologic Biomarker. Invest Opthalmology Visual Sci. 60\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoronyo-Hamaoui M, Koronyo Y, Ljubimov AV, Miller CA, Ko MK, Black KL et al (2011) Identification of amyloid plaques in retinas from Alzheimer's patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. NeuroImage 54(Suppl 1):S204\u0026ndash;217\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Koolwijk LME, Despriet DDG, Van Duijn CM, Oostra BA, van Swieten JC, de Koning I et al (2009) Association of cognitive functioning with retinal nerve fiber layer thickness. Investig Ophthalmol Vis Sci 50:4576\u0026ndash;4580\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan de Kreeke JA, Nguyen HT, den Haan J, Konijnenberg E, Tomassen J, den Braber A et al (2019) Retinal layer thickness in preclinical Alzheimer's disease. Acta Ophthalmol 97:798\u0026ndash;804\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJia J, Ning Y, Chen M, Wang S, Yang H, Li F et al (2024) Biomarker Changes during 20 Years Preceding Alzheimer's Disease. N Engl J Med 390:712\u0026ndash;722\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNam E, Lee Y-B, Moon C, Chang K-A (2020) : Serum Tau Proteins as Potential Biomarkers for the Assessment of Alzheimer's Disease Progression. Int J Mol Sci. 21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMattsson N, Zetterberg H, Janelidze S, Insel PS, Andreasson U, Stomrud E et al (2016) Plasma tau in Alzheimer disease. Neurology 87:1827\u0026ndash;1835\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eden Haan J, Morrema THJ, Verbraak FD, de Boer JF, Scheltens P, Rozemuller AJ et al (2018) Amyloid-beta and phosphorylated tau in post-mortem Alzheimer's disease retinas. Acta Neuropathol Commun 6:147\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHart de Ruyter FJ, Morrema THJ, den Haan J, Twisk JWR, de Boer JF, Scheltens P et al (2023) Phosphorylated tau in the retina correlates with tau pathology in the brain in Alzheimer's disease and primary tauopathies. Acta Neuropathol 145:197\u0026ndash;218\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalkiewicz G, Ronisz A, Van Ginderdeuren R, Lemmens S, Bouwman FH, Hoozemans JJM et al (2024) Primary retinal tauopathy: A tauopathy with a distinct molecular pattern. Alzheimer's Dement J Alzheimer's Assoc 20:330\u0026ndash;340\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHo C-Y, Troncoso JC, Knox D, Stark W, Eberhart CG (2014) Beta-amyloid, phospho-tau and alpha-synuclein deposits similar to those in the brain are not identified in the eyes of Alzheimer's and Parkinson's disease patients. Brain Pathol 24:25\u0026ndash;32\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams EA, McGuone D, Frosch MP, Hyman BT, Laver N, Stemmer-Rachamimov A (2017) Absence of Alzheimer Disease Neuropathologic Changes in Eyes of Subjects With Alzheimer Disease. J Neuropathol Exp Neurol 76:376\u0026ndash;383\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMathew S, WuDunn D, Mackay DD, Vosmeier A, Tallman EF, Deardorff R et al (2023) Association of Brain Volume and Retinal Thickness in the Early Stages of Alzheimer's Disease. J Alzheimers Dis 91:743\u0026ndash;752\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin GS, van der Heide F, Littlejohns TJ, Kuźma E, Hayat S, Brayne C et al (2023) Association Between Retinal Nerve Fiber Layer Thickness and Incident Dementia in the European Prospective Investigation into Cancer in Norfolk Cohort. J Alzheimers Dis 95:691\u0026ndash;702\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTatham AJ, Medeiros FA (2017) Detecting Structural Progression in Glaucoma with Optical Coherence Tomography. Ophthalmology 124:S57\u0026ndash;S65\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChrysou A, Jansonius NM, van Laar T (2019) Retinal layers in Parkinson's disease: A meta-analysis of spectral-domain optical coherence tomography studies. Parkinsonism Relat Disord 64:40\u0026ndash;49\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa X, Wang Y, Wang N, Zhang R (2021) Retina thickness in atypical parkinsonism: a systematic review and meta-analysis. J Neurol 269:1272\u0026ndash;1281\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoreno-Ramos T, Benito-Le\u0026oacute;n J, Villarejo A, Bermejo-Pareja F (2013) Retinal Nerve Fiber Layer Thinning in Dementia Associated with Parkinson's Disease, Dementia with Lewy Bodies, and Alzheimer's Disease. J Alzheimers Dis 34:659\u0026ndash;664\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo LST, Luong V, Wood N, Cheung W, Maass A, Ferrari G, Russo-Marie F, Sillito AM, Cheetham ME, Moss SE, Fitzke FW, Cordeiro MF (2007) Targeting amyloid-beta in glaucoma treatment. Proc Natl Acad Sci U S A 104:33\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHelmer C, Malet F, Rougier M-B, Schweitzer C, Colin J, Delyfer M-N et al (2013) Is there a link between open-angle glaucoma and dementia? Annals of Neurology.n/. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ea-n/a\u003c/span\u003e\u003cspan address=\"http://a-n/a\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJanssen S, Jansonius NM, Bouwman F, Verbraak FD, Bergen AA (2015): Systematic review of the association between Alzheimer\u0026rsquo;s disease and chronic glaucoma. \u003cem\u003eClinical Ophthalmology\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"Alzheimer's disease, tau, retinal nerve fiber layer, ganglion cell-inner plexiform layer, Mendelian randomization","lastPublishedDoi":"10.21203/rs.3.rs-5869988/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5869988/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eObservational studies have reported associations between retinal thickness and Alzheimer\u0026rsquo;s disease (AD); however, the causal relationship remains uncertain.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe investigated the potential causal relationship between retinal thickness and AD using Mendelian randomization (MR) and genetic colocalization analyses. Multiple genome-wide association studies (GWAS) of European ancestry were used for MR causal inference. Summary-level data on retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GC-IPL) thickness were sourced from the UK Biobank (N\u0026thinsp;=\u0026thinsp;31,434). AD data were obtained from a GWAS meta-analysis conducted by the International Genomics of Alzheimer\u0026rsquo;s Project for discovery (N\u0026thinsp;=\u0026thinsp;63,926) and an independent European GWAS cohort for replication (N\u0026thinsp;=\u0026thinsp;85,934). Circulating total-tau biomarker data were derived from another GWAS in individuals of European ancestry (N\u0026thinsp;=\u0026thinsp;14,721). The inverse-variance weighted (IVW) method served as the primary analytical approach, supplemented by MR-Egger, robust adjusted profile score, maximum likelihood, and weighted median methods. Sensitivity analyses included Cochran\u0026rsquo;s Q test, MR-Egger intercept test, leave-one-out analysis, and MR pleiotropy residual sum and outlier analysis to ensure robustness. Genetic colocalization analysis was performed to identify potential shared causal variants between retinal thickness and AD.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe IVW estimates from the discovery MR analysis indicated no statistically significant causal effect of genetically predicted RNFL or GC-IPL thickness on AD or circulating total-tau levels, and reverse MR analysis found no causal link either (P\u003csub\u003eIVW\u003c/sub\u003e \u0026gt; 0.05). Replication bidirectional MR analysis produced consistent negative results (P\u003csub\u003eIVW\u003c/sub\u003e \u0026gt; 0.05). Sensitivity analyses demonstrated robustness across all MR methods, with no evidence of heterogeneity, horizontal pleiotropy, or instrumental variable outliers. Genetic colocalization analysis identified no shared causal variants between RNFL or GC-IPL thickness and AD or circulating total-tau (posterior probability H4\u0026thinsp;\u0026lt;\u0026thinsp;0.75).\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eOur study does not support a genetic causal link between retinal thickness, AD, and circulating total-tau levels, despite previous observational studies suggesting an association between retinal thinning and higher AD risk. Further research is needed to clarify the relationship between RNFL and GC-IPL thickness and AD, as well as the underlying biological mechanisms.\u003c/p\u003e","manuscriptTitle":"Unraveling the association between retinal thickness and Alzheimer's disease, and circulating total-tau levels: Insights from genetic evidence","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-24 19:34:19","doi":"10.21203/rs.3.rs-5869988/v1","editorialEvents":[{"type":"communityComments","content":0}],"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":"15b67f4c-def7-42a6-abdf-f3ee3a46a93a","owner":[],"postedDate":"January 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-01-25T00:53:09+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-24 19:34:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5869988","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5869988","identity":"rs-5869988","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}