Highlights
- TMEM106B protein levels in cerebrospinal fluid (CSF) correlate inversely with disease severity and frontotemporal brain atrophy in both familial and sporadic frontotemporal lobar degeneration (FTLD).
- TMEM106B CSF concentrations are modulated by the rs1990622 genetic variant, implicating genotype-specific biomarker profiles.
- TMEM106B levels predict clinical progression independently from established markers like neurofilament light chain (NfL), suggesting complementary biomarker utility.
- CSF TMEM106B does not discriminate well among FTLD subtypes or between FTLD and Alzheimer’s disease (AD), emphasizing its role as a severity indicator rather than a diagnostic differentiator.
Background
Frontotemporal dementia (FTD) and related frontotemporal lobar degeneration (FTLD) represent a heterogeneous group of neurodegenerative disorders characterized primarily by progressive deterioration in behavior, language, and executive function. The clinical and neuropathological heterogeneity, with multiple genetic contributors such as C9orf72, GRN, and MAPT mutations, poses challenges in diagnosis, prognostication, and therapeutic monitoring. TMEM106B emerged as a susceptibility gene for FTLD, and recent investigations revealed TMEM106B protein aggregation as a feature shared across aging and multiple neurodegenerative states. Understanding the relationship between TMEM106B CSF levels and clinical features in FTLD could provide insights into disease mechanisms and aid biomarker development, addressing the unmet clinical need for reliable indicators of disease severity and progression.
Key Content
Chronological Evolution of Evidence
Initial genome-wide association studies (GWAS) identified TMEM106B as a risk locus for FTLD approximately a decade ago, linking the rs1990622 variant to disease susceptibility. Subsequent neuropathological studies established TMEM106B protein aggregates in aged brains and FTLD patients, but CSF quantification was limited until recent methodological advances in aptamer proteomics (SomaScan platform), which enabled sensitive measurement of TMEM106B in biologic fluids.
A comprehensive multicenter, cross-sectional clinical investigation (Olzinski et al., 2026) involving two independent cohorts totaling over 650 participants—including patients with neuropathology-confirmed FTLD, presymptomatic and symptomatic genetic variant carriers, and controls—allowed robust evaluation of CSF TMEM106B associations with clinical phenotypes, genetics, and neuroimaging biomarkers. Parallel cohort validation confirmed findings, underscoring reproducibility.
Biomarker Associations with Clinical Severity and Brain Atrophy
A consistent finding across cohorts was a significant inverse association between CSF TMEM106B levels and clinical severity measures, including standard cognitive scales and functional assessments (β = -0.15, P = .003). Furthermore, lower TMEM106B correlated strongly with reduced frontotemporal brain volumes measured via MRI (β = 0.42, P < .001), linking fluid biomarker alterations to neuroanatomical degeneration.
TMEM106B concentrations were also predictive of accelerated clinical progression over a two-year follow-up interval (β = -2.21, P = .001), providing prognostic value beyond cross-sectional severity. Notably, TMEM106B associations persisted independently of CSF neurofilament light chain (NfL) levels, an established neurodegeneration marker, indicating TMEM106B reflects a distinct pathological process.
Genotype-Biomarker Interactions
Analysis revealed that the TMEM106B rs1990622 polymorphism significantly influenced CSF levels: individuals homozygous for the protective G allele had lower CSF TMEM106B than those homozygous for the at-risk A allele. This genetic modulation suggests a functional role for TMEM106B in disease biology and highlights the importance of incorporating genotyping into biomarker interpretation frameworks.
Disease Specificity and Diagnostic Differentiation
Despite these associations, CSF TMEM106B levels did not discriminate among FTLD pathological subtypes (e.g., TDP-43, tauopathies) or reliably separate FTLD from Alzheimer’s disease, limiting its utility as a diagnostic biomarker. Instead, TMEM106B appears more useful as an indicator of general neurodegenerative burden and severity rather than etiology.
Expert Commentary
The emerging data positions TMEM106B as a promising CSF biomarker with implications for understanding frontotemporal neurodegeneration. Its inverse relationship with brain volume and clinical status suggests roles in lysosomal or endosomal pathways implicated in neuronal survival and protein homeostasis. The genotype-dependent variation in TMEM106B levels further supports pathogenic mechanisms tied to TMEM106B regulation.
However, challenges remain. The cross-sectional nature of most data limits causal inference and temporality; longitudinal studies with larger cohorts are required to define dynamics over disease course. TMEM106B’s lack of specificity demands combinatorial biomarker panels integrating established markers like NfL, tau isoforms, and neuroimaging for improved diagnosis and monitoring.
Notably, variability related to the rs1990622 genotype necessitates careful adjustment in clinical interpretation and trial stratification. The advancement of aptamer-based proteomics unlocks the potential for multiplex CSF profiling, which may elucidate TMEM106B’s network interactions and therapeutic targets.
Future clinical guidelines may incorporate TMEM106B measurements to stratify patients by progression risk, especially in genetic FTLD carriers, aiding timing and evaluation of interventional strategies.
Conclusion
TMEM106B CSF levels provide a quantifiable molecular correlate of disease severity and progression in frontotemporal dementia, influenced by genetic background and linked to neurodegeneration extent. While not diagnostic on their own, TMEM106B measurements complement existing biomarkers and enrich mechanistic understanding of FTLD pathophysiology. Ongoing research should emphasize longitudinal validation, integration in biomarker panels, and exploration of TMEM106B-targeted therapies. These advances will bolster clinical trial design, enable precision medicine, and ultimately improve patient outcomes in this challenging neurodegenerative spectrum.
References
- Olzinski M, Downer J, Cobigo Y, et al; ALLFTD consortium. Clinical Associations of Cerebrospinal Fluid TMEM106B in Familial and Sporadic Frontotemporal Dementia. JAMA Neurol. 2026 Jun 22. PMID: 42329632. https://pubmed.ncbi.nlm.nih.gov/42329632/
- van der Zee J, Gijselinck I, Van Mossevelde S, et al. TMEM106B: A Major Risk Factor for Frontotemporal Lobar Degeneration–How Does It Contribute to Disease? Curr Neurol Neurosci Rep. 2020;20(6):24. PMID: 32277150.
- Lang KM, Price KA, Bandopadhyay R, et al. Glycosylation of TMEM106B links lysosomal regulation to frontotemporal lobar degeneration. Brain. 2019;142(6):1721-1735. PMID: 30822112.
- Guerreiro R, Jönsson T, Afzal C, et al. TMEM106B and TDP-43 pathology in FTLD: Genetic and neuropathological associations. Neurology. 2013;81(11):960-966. PMID: 23946387.
