TDP-43 Research Tracker

Biomarkers: measuring TDP-43 problems

A biomarker is something doctors can measure — in blood, spinal fluid or a scan — that shows what's happening in the disease. For TDP-43 there is not yet a routine clinical test, but several are moving fast.

Why biomarkers matter for TDP-43 trials

TDP-43 trouble can only be seen directly under a microscope after death. Without a test in living people, trials can't easily tell who has TDP-43 disease, whether a drug reached its target, or whether it's working before muscle strength changes. Good biomarkers could make trials smaller, faster and more precise.

Deeper dive

Biomarkers serve different roles: diagnostic (does this person have TDP-43 pathology?), target-engagement / pharmacodynamic (did the drug change its target?), prognostic (how fast will disease progress?), and potentially surrogate endpoints for regulatory approval. TDP-43 loss-of-function markers such as cryptic neoepitopes were proposed explicitly to aid recruitment and measure target engagement[1]. A recent review summarizes the fluid biomarker landscape in ALS[2][3].

Cryptic exons and cryptic peptides

Where it stands: Research only

When TDP-43 stops working, genes like STMN2, UNC13A and HDGFL2 pick up cryptic exons[4]. Some of these get turned into abnormal protein pieces (cryptic peptides) that can leak into spinal fluid and even blood. Because healthy cells shouldn't make them, finding them is a direct sign of TDP-43 failure. In 2024, researchers detected an abnormal HDGFL2 protein in spinal fluid and blood of people with ALS/FTD — even in some gene carriers before symptoms[1].

Deeper dive

STMN2 and UNC13A: cryptic exon inclusion in STMN2 (premature polyadenylation, truncated transcript)[5][6] and UNC13A (nonsense-mediated decay)[7][8] was shown in postmortem tissue and neurons; these are measured mainly as RNA in tissue, and are also drug targets. Cryptic peptides: Seddighi et al. combined RNA sequencing and proteomics (mass spectrometry) in TDP-43-depleted human neurons, found 65 peptides mapping to 12 cryptic exons, and detected 18 de novo peptides from 13 genes in CSF of ALS/FTD patients[9]. HDGFL2: Irwin et al. built a monoclonal antibody against the HDGFL2 cryptic neoepitope; levels were higher in CSF in familial ALS-FTD and sporadic ALS than controls, rose earlier than NfL in familial disease, were detectable in blood and correlated with CSF[1]. A separate group (Citrano et al., Oct 2026) developed new anti-HDGFL2-CE immunoassays and linked brain HDGFL2-CE levels to TDP-43 pathology, earlier onset and shorter survival in FTLD-TDP[10].

⚠ Uncertain: we did not find a validated STMN2 or UNC13A test in blood or spinal fluid; these are measured in tissue/research settings. Cryptic HDGFL2 assays are research tests, not clinically available.

Neurofilament light (NfL)

Where it stands: Used in some clinics

NfL is a piece of the nerve fiber's skeleton that spills into spinal fluid and blood when nerves are damaged. It's high in ALS and tends to be higher in faster-progressing disease. It is not specific to TDP-43 — it rises in many nerve diseases — but it's the most established ALS fluid marker today.

Deeper dive

NfL can be measured in CSF and blood with sensitive immunoassays and is elevated across many neurological disorders[11]. In 2023 the FDA granted accelerated approval to tofersen for SOD1-ALS based on reduction in plasma NfL as a surrogate reasonably likely to predict benefit[12]. Many TDP-43 trials (e.g. QRL-201, VTx-002, NUZ-001) list NfL or pNfH as outcomes (see the pipeline). Recent reviews cover NfL alongside newer markers[3].

Phosphorylated TDP-43 and seeding assays (SAA / RT-QuIC)

Where it stands: Research only

Misfolded TDP-43 can act like a seed that makes normal TDP-43 misfold too. Seed amplification assays exploit this: a spinal fluid sample is mixed with normal TDP-43, and if seeds are present the clumping is amplified until it's detectable. Similar tests are already used for other brain proteins; TDP-43 versions are still being developed.

Deeper dive

Dellarole et al. (2025) found TDP-43 seeding activity in CSF of 67% of symptomatic TDP-43-linked genetic FTD/ALS patients (GRN, C9orf72) with 93% specificity, and in almost half of presymptomatic carriers, mostly GRN[13]. Borberg et al. (2026) described a digital SAA that counts single TDP-43 aggregates in CSF and found elevated seed concentrations in FTLD-TDP correlating with severity[14]. Phosphorylated TDP-43 (pS409/410) is the classic pathological form seen at autopsy[15].

⚠ Uncertain: published SAA cohorts are small and mostly genetic FTD; performance in sporadic ALS is not well established.

Extracellular vesicles

Where it stands: Research only

Brain cells release tiny bubbles called extracellular vesicles that can cross into the blood, carrying proteins like TDP-43. Fishing out vesicles that came from brain cells could give a blood test that reflects what's happening in the brain.

Deeper dive

Butt et al. (2026) isolated astrocyte-derived EVs from blood and plasma across three cohorts; the plasma EV pTDP-43/CD81 ratio distinguished ALS from healthy controls with mean AUC 0.89 (sensitivity 87%, specificity 89%). The authors note specificity versus ALS mimics and links to progression still need study[16].

⚠ Uncertain: single study, compared only with healthy controls.

PET tracers for TDP-43

Where it stands: Used in trials

A PET tracer would let doctors see TDP-43 clumps in a living brain on a scan, the way amyloid scans work in Alzheimer's. The first TDP-43 tracers are now in early human testing.

Deeper dive

AC Immune reported preliminary Phase 1 results for ACI-19626 showing increased uptake in brains of people with ALS (company statement, not yet peer-reviewed)[17]. Xia et al. (2026) described the development of another candidate tracer, [18F]JNJ-TDP43-1[18].

⚠ Uncertain: early, small, company-reported data.

Sources

  1. Irwin et al., Nat Med 2024 (PMID 38278991)
  2. Irwin et al., Mol Neurodegener 2024, review (PMID 38267984)
  3. Verde et al., Curr Opin Neurol 2025 (PMID 40832743)
  4. Ling et al., Science 2015 (PMID 26250685)
  5. Klim et al., Nat Neurosci 2019 (PMID 30643292)
  6. Melamed et al., Nat Neurosci 2019 (PMID 30643298)
  7. Ma et al., Nature 2022 (PMID 35197626)
  8. Brown et al., Nature 2022 (PMID 35197628)
  9. Seddighi et al., Sci Transl Med 2024 (PMID 38277467)
  10. Citrano et al., Cell Rep Med 2026 (PMID 42854693)
  11. Gaetani et al., J Neurol Neurosurg Psychiatry 2019 (PMID 30967444)
  12. US FDA: tofersen accelerated approval (2023)
  13. Dellarole et al., Alzheimers Dement 2025 (PMID 41399249)
  14. Borberg et al., Alzheimers Dement 2026 (PMID 42084118)
  15. Neumann et al., Science 2006 (PMID 17023659)
  16. Butt et al., Neurobiol Dis 2026 (PMID 42722112)
  17. AC Immune H1 2026 corporate update (press release)
  18. Xia et al., Alzheimers Dement 2026 (PMID 42471754)