Genomics And Digital neuropathological phenotyping of Iberian bRains - Project GADIR
GADIR is a collaborative project between neurological tissue banks (BTN) in Spain and Portugal, coordinated by the CIEN (Madrid) and ACE (Barcelona) Foundations, and developed in the period 2025 - 2027, with funding from the Pasqual Maragall Researcher's Programme. PI: Alberto Rábano (CIEN); co-IP: Victoria Fernández (FACE).
Background, problem statement and main objective
Dementia is one of the main health problems worldwide, and its most prevalent neuropathological substrate is Alzheimer's disease (AD), vascular dementia (VaD) and dementia with Lewy bodies (DLB). Except in cases of genetic origin (< 3%), definitive diagnosis of the disease still depends on neuropathological (NP) study, usually in the context of post-mortem brain donation. In recent years, the work of the BTNs has shown that the coexistence of different pathologies is the norm in dementias associated with advanced age. In addition to the three pathologies mentioned (AD, VaD and MCI), limbic-predominant age-related TDP-43 encephalopathy (LATE) has been added as a pathological condition with evidence of clinical impact, while in the case of other co-pathologies of varying prevalence (mainly tauopathies) the clinical correlation is still uncertain.
In the current context of dementia research inaugurated by the new anti-amyloid therapies for AD with the capacity to modify the cognitive data of patients, precision medicine is rapidly advancing in the definition of biomarkers in cerebrospinal fluid and blood, and by means of various neuroimaging techniques (mainly MRI and PET), for the detection, also early, of the main pathologies and copathologies. These markers are complemented by genetic risk factors, which in cases of sporadic dementia depend on a growing number of genes (polygenic risk scores, PRS). While there are extensive recent studies of GWAS in dementia, mainly based on clinical samples, studies with neuropathological confirmation are scarce, and it is the latter that have the greatest potential to establish direct associations with pathology-specific NP traits and related pathogenic processes.
The main objective of the GADIR project is the massive genetic sequencing of donated brains in the Iberian Peninsula, and the association analysis of genomic data with diagnostic NP data and brain classification (endophenotypes), in order to advance in the discovery of new genetic risk factors and associated metabolic and cellular pathways.
Methodology and work plan
The project will be developed according to the following specific objectives:
Objective 1. Creation of the largest Iberian database of brains with neuropathological characterisation.
The CIEN Tissue Bank (BT-CIEN) will coordinate the selection of cases for inclusion in the project, the recording of associated data, the necessary sampling, as well as all the activity related to the updating of the NP diagnoses of the cases. Brains up to 30 years old will be included, and, taking into account that NP diagnostic criteria have changed radically in the last 10-15 years, a high percentage of cases will require new immunostaining techniques and diagnostic evaluations. The new NP information resulting from the project will be made available to the participating BTNs. The BT-CIEN will also coordinate the shipment of frozen tissue samples (frontal cortex) from each of the selected brains to FACE for DNA extraction. All case management and classification data will be recorded in a database in REDCap, which will be open to all BTNs participating in the project. Inclusion criteria will be 1) the availability of fresh frozen brain tissue, and 2) the availability of a minimum set of associated data (pre-analytical, socio-demographic, clinical and neuropathological). It is estimated that at least 3,500 brains will be included in the study.
Objective 2. Development of polygenic risk scores to discover pathogenic pathways.
Genomic DNA extraction will be performed on frozen brain tissue samples submitted to FACE. The genomic sequencing study (GWAS) will be performed at the Centro Nacional de Genotyping (CEGEN), in Santiago de Compostela, using the Axiom 815K Spanish biobank array (Thermo Fisher) which includes rare variants specific to the Spanish population. The genotyping data will be processed in FACE, where, on the basis of SNP analysis, different PRS will be calculated for the main neurodegenerative pathologies (AD, FTLD, MCI, ALS, PD) and their most frequent combinations. Likewise, agnostic analysis of PRS associated with biological processes (pPRS) in all pathologies will be carried out using the Molecular Signatures Database (https://www.gsea-msigdb.org/gsea/msigdb/), as a reference, and the relationship between clinical diagnosis, PRS and neuropathological diagnosis will be analysed. Specifically, SNPs contributing to copathologies will be analysed.
Objective 3. To assess neuropathological phenotypic heterogeneity in highly prevalent pathologies by digital pathology.
250 cases of late-onset AD (LOAD) (>65 years) and 100 cases of early-onset AD (EOAD) will be selected. The LOAD cases will be distributed between brains with low burden (50%) and brains with moderate/high burden of vascular pathology (50%), the most prevalent copathology. In all of them, immunostained preparations for β-amyloid or tau will be selected from 8 representative cortical and subcortical regions. All of them will be digitised using a 3D-Histech midi digital pathology scanner and subjected to analysis using an AI/machine learning programme developed in the laboratory of Diego Sepúlveda-Falla, UKE Hamburg. For plaque pathology, CAA and neurofibrillary pathology, the total pathological burden will be analysed, the main lesions will be quantified and regular lesion patterns (phenotypic heterogeneity of each type of pathology) will be searched for. The different morphological variables obtained will be analysed in correlation with the genomic data of the cases.
Objective 4. To determine the genetic factors contributing to the main neuropathological features.
Several analysis strategies (single variant analysis, gene-based analysis, principal component analysis, network and pathway analysis) will be used to contrast the genomic information with the different types of neuropathological variables available: i) qualitative variables due to the presence/absence of specific pathologies, or combinations of these pathologies, ii) ordinal variables derived from the staging/classification systems of the most prevalent pathologies, and iii) quantitative variables derived from the morphometric analyses described in Objective 3.
An attempt will be made to replicate the results obtained in existing databases (NACC, ROSMAP, ACT). Furthermore, in-silico functional assays will be performed to analyse the risk variants detected.
Availability and public access to the results
All information generated during the update of the neuropathological classification of the included cases will be made available to the participating BTNs. Genomic sequencing data will also be available for the BTNs of origin of the samples. In addition, the final neuropathological classification data of the included cases, digitised images and sequencing data will be made available to external researchers in a GADIR data repository.
The GADIR project is supported by the ISCIII Platform of Biomodels and Biobanks, the Neurological Tissue Banks Working Group, the Spanish Consortium of Genetics of Dementias (DEGESCO), the Spanish Neuropathology Club (belonging to the Spanish Society of Neurology and the Spanish Society of Pathological Anatomy), and the Portuguese Society of Neuropathology.
For further information, please contact the following email address: gadir@fundacioncien.es
