A National Institutes of Health (NIH) National Institute on Aging-supported initiative will integrate genetics, exposome science, biomarkers, and aging clocks to transform dementia research, regional capacity building, and clinical trial readiness across Latin America.
The Multi-Partner Consortium to Expand Dementia Research in Latin America (ReDLat2) has entered a new phase, expanding one of the most ambitious international efforts to understand Alzheimer’s disease, frontotemporal lobar degeneration, and brain aging in Latin America and among Latino populations in the United States.
Recently highlighted by Nature Medicine, ReDLat2 will address a critical gap in global dementia research: we still know very little about how environmental factors (the exposome), genetic influences, and the biological pace of aging interact to shape dementia risk, progression, and resilience.
Most discoveries to date have emerged from populations of predominantly European ancestry living in relatively homogeneous environments. Meanwhile, Latin American populations remain significantly underrepresented despite facing a high burden of dementia, possessing complex genetic admixture, experiencing diverse environmental exposures, and confronting major disparities in diagnosis, biomarker use, and access to clinical trials.
Supported by a major investment from the U.S. National Institute on Aging (NIA) and complemented by multiple satellite projects—including two R01 grants and funding from the Alzheimer’s Association, CARD-NIA, The Bluefield Foundation, The TAU Consortium (Rainwater Charitable Foundation), Wellcome Trust, Wellcome Leap, and Horizon Europe—ReDLat2 will develop a genomically informed, environmentally contextualized, and complex-phenotype research platform across Latin America and the United States.
The consortium brings together 13 centers across 7 countries—Argentina, Brazil, Chile, Colombia, Mexico, Peru, and the United States—and involves approximately 90 investigators and multiple collaborative working groups. ReDLat2 builds upon the first phase of ReDLat, which established harmonized protocols, centralized data management, collaborative working groups, plasma biobanks, and a legal and scientific infrastructure that integrated more than 4,200 participants.
Figure 1 | Consortium overview. ReDLat2 is a multicenter platform designed to investigate how the exposome, genetic architecture, and accelerated aging shape dementia phenotypes across Latin America and within a comparative Latino cohort in the United States. The consortium brings together sites from Argentina, Brazil, Chile, Colombia, Mexico, Peru, and the United States, focusing on Alzheimer’s disease, frontotemporal lobar degeneration, and healthy control populations. Aim 1 will examine how the social exposome—including individual-level, socioeconomic, and aggregate social determinants of health—interacts with genetic risk to shape multimodal phenotypes. These analyses will assess group differences and test interactions between the social exposome and polygenic risk scores across clinical, cognitive, neuroimaging, and plasma-based measures. Aim 2 will characterize familial and sporadic genetic risk in Alzheimer’s disease and frontotemporal lobar degeneration through the study of large families, founder mutations, genetic admixture, and interactions between the social exposome and polygenic risk scores. This approach will link ancestry-informed autosomal dominant variation to clinical phenotype, age at onset, and disease progression. Aim 3 will develop multimodal models of accelerated aging and brain age gap using machine learning, deep learning, and generative biophysical modeling approaches. These models will enable the assessment of differences across regions, sex, and diagnostic groups, as well as interactions among brain age gap, the social exposome, and genetic risk.
Figure 1 | Consortium overview. ReDLat2 is a multicenter platform designed to investigate how the exposome, genetic architecture, and accelerated aging shape dementia phenotypes across Latin America and within a comparative Latino cohort in the United States. The consortium brings together sites from Argentina, Brazil, Chile, Colombia, Mexico, Peru, and the United States, focusing on Alzheimer’s disease, frontotemporal lobar degeneration, and healthy control populations. Aim 1 will examine how the social exposome—including individual-level, socioeconomic, and aggregate social determinants of health—interacts with genetic risk to shape multimodal phenotypes. These analyses will assess group differences and test interactions between the social exposome and polygenic risk scores across clinical, cognitive, neuroimaging, and plasma-based measures. Aim 2 will characterize familial and sporadic genetic risk in Alzheimer’s disease and frontotemporal lobar degeneration through the study of large families, founder mutations, genetic admixture, and interactions between the social exposome and polygenic risk scores. This approach will link ancestry-informed autosomal dominant variation to clinical phenotype, age at onset, and disease progression. Aim 3 will develop multimodal models of accelerated aging and brain age gap using machine learning, deep learning, and generative biophysical modeling approaches. These models will enable the assessment of differences across regions, sex, and diagnostic groups, as well as interactions among brain age gap, the social exposome, and genetic risk.
Together, the consortium’s three aims provide an integrated framework for modeling the multimodal diversity of dementia in underrepresented populations.
ReDLat2 will recruit 3,000 new participants, including individuals with Alzheimer’s disease, frontotemporal lobar degeneration, healthy controls, and members of large multiplex families. The project will integrate neuroimaging, cognition, spontaneous speech, circadian rhythms, plasma biomarkers, proteomics, epigenetics, whole-genome sequencing, ancestry, social determinants of health, environmental exposures, and advanced computational models.
At the heart of this initiative lies a simple yet transformative idea: dementia risk and resilience cannot be understood by studying genes or the environment in isolation. ReDLat2 will test how ancestry-related genetic risk, rare variants, founder effects, protective variants, and lifetime exposome adversity interact to shape disease onset, clinical symptoms, brain aging, and biological vulnerability.
One of ReDLat2’s major innovations is the development of multimodal aging clocks that incorporate brain, behavioral, speech, epigenetic, proteomic, and biomarker-based measures. These tools will help determine whether accelerated or delayed aging reflects shared biological pathways of vulnerability and resilience across diverse populations.
The initiative is led by its founders and program directors, Agustín Ibáñez (Latin American Brain Health Institute—BrainLat, Universidad Adolfo Ibáñez; Trinity College Dublin; Barcelonaβeta Brain Research Center; and Universidad de San Andrés) and Bruce L. Miller (University of California, San Francisco, UCSF), together with principal investigators Jennifer S. Yokoyama (UCSF) and Claudia Duran-Aniotz (BrainLat, Universidad Adolfo Ibáñez).
Investigators across ReDLat sites include Adolfo M. García (Universidad de San Andrés), Martín Bruno (Institute of Biomedical Sciences), Nahuel Magrath-Guimet (FLENI), Kenneth Kosik & Juliana Acosta Uribe (University of California, Santa Barbara), Nick Cochran (HudsonAlpha Institute for Biotechnology), Kate Possin & Victor Valcour (UCSF), Leonel Takada (University of São Paulo), Elisa Resende (Universidade Federal de Minas Gerais), Andrea Slachevsky (Universidad de Chile), María Isabel Behrens (Universidad de Chile), Diana Matallana (Hospital Universitario San Ignacio), Hernando Santamaría-García (Hospital Universitario San Ignacio/Pontificia Universidad Javeriana), David Aguillón (Grupo de Neurociencias de Antioquia), Lina Zapata (Universidad del Valle), José Alberto Avila Funes (Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán), and Nilton Custodio (Instituto Peruano de Neurociencias).
Beyond scientific discovery, ReDLat2 is designed as a regional platform for capacity building. The initiative will expand training opportunities for early-career researchers, strengthen harmonized research infrastructure, support secure data sharing, and create new opportunities for satellite studies led by investigators across Latin America. The project will also enhance regional readiness for future clinical trials by expanding access to biomarkers, genomics, and deeply characterized cohorts. “ReDLat2 is not only about understanding dementia in Latin America,” said Agustín Ibáñez. “It is about changing the way global dementia science is conducted. By integrating genetics, environment, biomarkers, and computational aging clocks, we can build precision-science models that better reflect real-world heterogeneity.”