PROTECT YOUR DNA WITH QUANTUM TECHNOLOGY
Orgo-Life the new way to the future Advertising by AdpathwayA comprehensive multi-omics investigation integrating large-scale human genomics with proteomic and transcriptomic datasets indicates that increased levels of the protein progranulin are causally associated with a reduced risk of developing Alzheimer’s disease and broader forms of dementia. Conducted by an international research team examining data from major population cohorts and disease-specific consortia, the study evaluated whether therapeutic elevation of progranulin could yield protective neurological effects. Progranulin, which is encoded by the GRN gene, has long been recognized for its critical involvement in neurodegenerative biology, as loss-of-function variants resulting in progranulin deficiency are established causal drivers of frontotemporal dementia and the lysosomal storage disorder neuronal ceroid lipofuscinosis. By systematically combining summary statistics from genome-wide association studies of Alzheimer’s disease with cis protein quantitative trait loci derived from blood plasma and cerebrospinal fluid, the researchers generated causal human evidence demonstrating that individuals with genetically sustained higher progranulin concentrations exhibit lower susceptibility to Alzheimer’s pathology.
The study design relied on integrating multiple genomic and proteomic platforms to interrogate common and rare variation across the GRN locus. Summary statistics for Alzheimer’s disease susceptibility were obtained from five published genome-wide association studies and the FinnGen release 9 biobank cohort encompassing approximately 500,000 Finnish participants. These investigations incorporated both clinically diagnosed true Alzheimer’s cases and proxy cases identified through parental family histories of dementia to maximize statistical discovery power while accounting for strong genetic correlations between diagnosed cases and proxy phenotypes. To assess gene expression and protein abundance in human tissues, the investigators evaluated expression quantitative trait loci from Genotype-Tissue Expression data and the MetaBrain project, alongside plasma protein quantitative trait loci from the UK Biobank Pharma Proteomics Project, a precompetitive consortium of thirteen pharmaceutical companies that profiled 2,923 proteins using the Olink Explore 3072 assay across 54,219 individuals. In parallel, cerebrospinal fluid proteomic datasets were compiled from 4,968 participants across eight neurodegenerative cohorts analyzed on SomaScan platforms, alongside enzyme-linked immunosorbent assay measures of progranulin from 818 participants in the Alzheimer’s Disease Neuroimaging Initiative.
To rigorously examine common variation, the researchers performed genetic colocalization analyses across the GRN genomic region to test whether genetic signals governing protein abundance, brain messenger RNA expression, and Alzheimer’s disease susceptibility share an identical causal biological mechanism. Regional association analyses highlighted the common variant rs5848 as the lead genetic signal across evaluated Alzheimer’s datasets, yielding strong statistical associations in both the Bellenguez et al. study at p = 1.8 × 10–12 and the FinnGen release 9 cohort at p = 1.6 × 10–13. Colocalization testing revealed a 100 percent posterior probability that the genetic signals responsible for progranulin protein abundance in plasma and cerebrospinal fluid share the exact causal variant underlying Alzheimer’s risk. Brain expression data confirmed that the disease-risk allele correlated with decreased progranulin transcript abundance across brain regions. Across both tissue compartments, the risk-increasing T allele of rs5848 consistently associated with decreased progranulin protein levels, demonstrating associations of p = 1.5 × 10–278 in plasma and p = 3.2 × 10–17 in cerebrospinal fluid, with each copy of the T allele conferring an average 10.8 percent reduction in cerebrospinal fluid and a 4.9 percent reduction in plasma.
To evaluate rare genetic variation beyond common single-nucleotide polymorphisms, the investigators analyzed whole-genome sequencing data from 500,000 UK Biobank participants generated by the UK Biobank Whole-Genome Sequencing Consortium. The researchers implemented an electronic health record phenotyping framework spanning five medical ontology systems and eight distinct clinical sources, establishing validated definitions for overall dementia cases, specific Alzheimer’s cases, and non-Alzheimer’s dementia cases while excluding frontotemporal dementia due to small case counts. Gene-based burden testing was performed with REGENIE software using logistic regression models among participants of non-Finnish European genetic ancestry, adjusting for baseline age, genetic sex, sequencing center, and twenty genetic principal components. Aggregating 54 rare predicted loss-of-function variants with minor allele frequencies below 0.001 annotated via LOFTEE and VEP revealed a profound association with elevated dementia and Alzheimer’s risk, showing p values below 6.58 × 10–7 and odds ratios ranging from 9.6 to 10.4. Furthermore, sixteen rare loss-of-function variants demonstrated a severe association with depleted plasma progranulin levels, producing an effect size beta of −3.14 at p = 1.08 × 10–58, establishing that rare progranulin-inactivating mutations mirror the deleterious risk conferred by common regulatory alleles.
The researchers subsequently deployed Mendelian randomization to model the prospective therapeutic impact of raising progranulin abundance, using the common GRN variant rs5848 as an instrumental variable in two-sample Wald ratio analyses. To establish clinically relevant estimates, protein measurements from the UK Biobank and the Alzheimer’s Disease Neuroimaging Initiative were analyzed on a log2 scale, where a single-unit alteration corresponds directly to a doubling or halving of progranulin abundance. The Mendelian randomization estimates were subsequently scaled to mimic the biological elevation of progranulin documented in recent Phase 1 human clinical trials of monoclonal antibodies targeting sortilin, such as latozinemab. Sortilin functions as a key clearance receptor that internalizes progranulin and directs it toward lysosomal degradation; blocking sortilin genetically or pharmacologically inhibits this clearance pathway. Phase 1 trials of anti-sortilin antibodies and small-molecule sortilin inhibitors such as VES001 confirmed substantial increases in systemic and central progranulin concentrations, providing real-world pharmacological benchmarks of two-fold increases in cerebrospinal fluid and three-fold increases in blood plasma.
Applying these empirically observed pharmacological scaling factors, the Mendelian randomization framework estimated that a genetically predicted two-fold increase in cerebrospinal fluid progranulin concentration is associated with approximately 35 percent lower odds of developing Alzheimer’s disease, corresponding to an odds ratio of 0.65 with a p value of 2.2 × 10–12. When evaluating systemic circulation, a genetically predicted three-fold increase in plasma progranulin abundance was associated with approximately 76 percent lower odds of developing Alzheimer’s disease, corresponding to an odds ratio of 0.24 at p = 2.0 × 10–12. Although circulating progranulin pools in blood plasma and cerebrospinal fluid display only moderate correlation and are subject to age and sex influences, both analytical models consistently pointed toward substantial disease protection. These statistical results indicate that common genetically determined variations that elevate progranulin protein production confer persistent protection against Alzheimer’s disease onset, reinforcing the rationale for developing therapeutic modalities designed to elevate progranulin levels in at-risk patient populations.
The study contextualized these findings across the broader spectrum of neurological disorders, highlighting how distinct genetic alterations in GRN generate an allelic series of clinical severity tied directly to progranulin dosage. Complete loss of functional progranulin caused by homozygous loss-of-function mutations leads to early-childhood neuronal ceroid lipofuscinosis, a severe lysosomal storage disorder characterized by retinal dystrophy, cerebellar ataxia and atrophy, and rapid cognitive deterioration. Heterozygous loss-of-function variants causing progranulin haploinsufficiency, which reduce circulating progranulin by approximately 50 percent, are well-documented causal drivers of adult-onset frontotemporal dementia and were shown here to drastically elevate general dementia risk. Common regulatory variants like rs5848 introduce modest changes of 5 to 11 percent in progranulin expression, yet cumulatively modulate lifetime Alzheimer’s vulnerability across the general population. The authors noted that this clear alignment between genetic variation, intermediate molecular protein biomarkers, and clinical disease risk places GRN within the highest tier of genetically validated Alzheimer’s targets, paralleling established mechanisms such as TREM2, PLCG2, and APOE.
In evaluating the translational potential of these findings, the authors emphasized several biological and methodological nuances inherent to genetic modeling. Mendelian randomization estimates reflect lifelong exposure to genetically elevated protein levels, whereas clinical pharmacological interventions typically occur acutely or later in life. Consequently, modeled genetic effects are likely to exceed the magnitude of clinical benefits observed in therapeutic practice, mirroring prior genetic studies where lifelong genetic suppression of PCSK9 or low-density lipoprotein cholesterol predicted larger cardiovascular risk reductions than those realized in short-term statin or antibody trials. Additionally, while therapeutic sortilin antibodies elicit robust biomarker responses in cerebrospinal fluid, fluid concentrations serve only as surrogates for brain penetration, and it remains uncertain whether engineered antibodies diffuse uniformly throughout the central nervous system parenchyma or penetrate all disease-relevant cell types equally. Furthermore, sortilin is not the sole receptor mediating progranulin uptake; alternative pathways involving low-density lipoprotein receptor-related protein 1 and mannose-6-phosphate receptors also facilitate cellular entry, and current datasets lack quantification of granulins, the biologically active cleavage fragments of progranulin that exert independent neuroprotective functions.
The researchers also addressed apparent biological complexities concerning progranulin dynamics across progressive stages of neurodegeneration. While genetic analyses confirm that lifelong progranulin deficiency increases the risk of Alzheimer’s onset, longitudinal biomarker observations from clinical cohorts reveal a biphasic pattern, wherein cerebrospinal fluid progranulin levels decrease during early asymptomatic phases but subsequently rise during advanced, late-stage Alzheimer’s disease. The authors suggested that initial progranulin deficits may drive disease vulnerability by precipitating microglial dysfunction and accelerating tau tangle formation, whereas secondary elevations in advanced disease might represent a reactive, compensatory microglial response to ongoing neuronal damage and amyloid-beta pathology. A similar counterintuitive elevation during disease progression has been established for soluble TREM2, indicating that the physiological impact of progranulin modulation may depend heavily on the specific disease stage and the brain’s evolving inflammatory milieu at the time of therapeutic administration.
The authors detailed several methodological limitations across the underlying data cohorts, noting that the incorporation of proxy Alzheimer’s cases from family history registries and multi-site meta-analyses introduces potential ascertainment and reporting heterogeneity, although sensitivity analyses restricted to clinically verified cases demonstrated consistent effect sizes. Moreover, sample sizes varied dramatically between proteomic platforms, as plasma Olink assays encompassed over 54,000 participants whereas cerebrospinal fluid profiling across SomaScan cohorts and ADNI ELISA assays involved smaller samples of 4,968 and 818 individuals, respectively. Discrepancies between measuring technologies and the current lack of adequately powered human brain parenchymal proteomic datasets also constrain direct comparisons between peripheral fluids and local tissue concentrations. Despite these constraints, the integration of common variant genome-wide associations, whole-genome sequencing burden tests, and multi-compartment proteomic analyses provides strong human genetic validation that progranulin deficiency contributes causally to Alzheimer’s disease, supporting targeted efforts to therapeutically elevate progranulin as a rational disease-modifying strategy.
Subject of Research: Genetic and proteomic evaluation of progranulin modulation in Alzheimer's disease risk
Article Title: Genetic evidence supports therapeutic modulation of progranulin levels in Alzheimer’s Disease
Article References: Robins, C., Eicher, J. D., Liu, J., Lo, Y., Richardson, T. G., Western, D., Davitte, J., Cruchaga, C., for the Alzheimer’s Disease Neuroimaging Initiative, Kumar, S., & Xu, C.-F. (2026). Genetic evidence supports therapeutic modulation of progranulin levels in Alzheimer’s Disease. Molecular Neurodegeneration Advances, 2(1), Article 12. https://doi.org/10.1186/s44477-025-00011-y
Image Credits: AI Generated
DOI: 10.1186/s44477-025-00011-y
Keywords: Alzheimer's disease, progranulin, GRN, Mendelian randomization, pQTL, genome-wide association studies, sortilin, latozinemab, UK Biobank, cerebrospinal fluid, colocalization, rare variants, neurodegeneration, ADNI, eQTL, dementia


3 hours ago
9




















English (US) ·
French (CA) ·