The current tau PET tracers bind best to the fibrils found in Alzheimer’s disease, and now, finally, comes progress on non-AD tauopathies. In the June 14 Nature Communications, researchers led by Neil Vasdev at the University of Toronto, in collaboration with Chester Mathis at the University of Pittsburgh and Samuel Svensson at the biotech Oxiant Discovery in Södertälje, Sweden, debuted a new tau tracer. In postmortem brain, OXD-2314 bound the four-repeat tau deposits characteristic of progressive supranuclear palsy and corticobasal degeneration, as well as 3R tau in Pick’s disease. In healthy rodents and nonhuman primates, radiolabeled OXD-2314 entered the brain and cleared quickly. The tracer is now in a Phase 1 study in healthy people.

  • OXD-2314 binds 3R, 4R, and 3R/4R tau with high affinity in vitro.
  • Phase 1 trial has begun.
  • APN-1607 is starting a Phase 3 registration trial for 4R tauopathy PSP.

Rather than being derived from existing tracers, OXD-2314 was designed from scratch, with scientists generating a new chemical structure computationally predicted to bind 4R tau. “This is a new class of compound that can be explored. It’s a bench-to-bedside example of how we develop a new PET tracer,” Vasdev told Alzforum.

Further along in development, Aprinoia Therapeutics received Fast Track designation from the U.S. Food and Drug Administration to test APN-1607 in the most common 4R tauopathy, PSP. This FDA designation speeds up regulatory review, enabling products to come to market sooner. APN-1607, a pan-tau tracer, meaning it binds to all forms of tau fibrils, already has been tested in nearly 500 PSP patients. A worldwide Phase 3 study in PSP will begin enrolling later this year. If it succeeds, APN-1607 could become the first 4R tau tracer approved for clinical use. “We’re trying to bring it to patients as soon as possible,” Mark Shearman at Aprinoia told Alzforum, noting the urgent need for better ways to diagnose PSP. This disease is often mistaken for other disorders based on clinical symptoms.

In it Goes. In healthy rhesus macaques, OXD-2314 easily entered the brain, and washed out quickly. [Courtesy of Lindberg et al., 2024.]

Tau tracers already in research use, such as flortaucipir, were optimized to detect the mixed three- and four-repeat-containing fibrils found in AD brain. Some of them, for example Life Molecular Imaging’s PI-2620, also bind 4R tau, but the signal is typically weaker (May 2024 news). For this reason, Vasdev and colleagues set out to design a new compound that specifically bound 4R tau. Built around a pyridinyl-indole scaffold, their first candidate, OXD-2115, bound 4R fibrils in vitro but bounced off the blood-brain barrier (Lindberg et al., 2021). 

Trying again, the authors synthesized and tested more than 150 analogues. First author Anton Lindberg identified OXD-2314 as the most promising one. It bound to homogenates from PSP brain with a dissociation constant of 2.4 nM, about twice as strongly as did OXD-2115, APN-1607, or PI-2620 in this assay. OXD-2314 bound equally well to homogenates from CBD brain, which contain 4R tau fibrils with a related, though distinct, fold from that in PSP.

Unexpectedly, OXD-2314 also bound 3R tau from Pick’s disease brain, and 3R/4R fibrils from AD homogenates, with a similar affinity. Computational modeling suggested it would bind the 3R/4R fibrils from chronic traumatic encephalopathy brain as well. AD and CTE folds resemble each other, and are more akin to the Pick’s fold than to the CBD and PSP folds (see Oct 2021 news for the tau fold family tree).

The Right Scaffold? The OXD-2314 pyridinyl-indole backbone (bottom right), distinguishes it from the carbazole of flortaucipir and PI-2620 (top left), the benzothiazole of PBB3 and APN-1607, aka florzolotau (bottom left), and the scaffolds of other PET tracers (top right). [Courtesy of Lindberg et al., 2024.]

The results suggest that, like APN-1607, OXD-2314 is a pan-tau tracer. The plan is to develop it for all the non-AD tauopathies, Vasdev told Alzforum. He does not know where OXD-2314 binds on tau fibrils, but competition assays showed the site is separate from where APN-1607, PI-2620, flortaucipir, and MK-6240 latch on.

Makoto Higuchi at the National Institutes for Quantum and Radiological Science and Technology in Chiba, Japan, said these data imply that OXD-2314 might bind two separate sites, with the lower-affinity binding accounting for perhaps a quarter of the total. Future work should identify this second site, which could potentially reflect binding to something other than tau, he suggested (comment below).

In mice, OXD-2314 injected into the blood readily entered the central nervous system, with a total brain-to-plasma ratio of 1.7. Values above 1 are desirable, Vasdev noted. The authors conjugated OXD-2314 to radiolabeled fluorine and tested the tracer in two rats, two rhesus macaque monkeys, and two baboons. Pharmacokinetics were favorable for a PET ligand, with the tracer peaking in the brain at two minutes, and half of it gone by around 20 minutes, as expected for healthy brain without tau aggregates (see image above).

In OXD-2314’s P1, the first two participants were scanned in June. Vasdev said the tracer entered the brain, produced a stable signal, and spawned no troublesome radiometabolites. “I’m delighted to say the tracer looks excellent in healthy controls,” he told Alzforum. The study is expected to conclude this summer, and by fall, the authors will test OXD-2314 in PSP patients. They hope to expand to CBD, Pick’s disease, and CTE.

Each is Different. Tau tracer APN-1607 lights up the non-AD tauopathies PSP (left), CBD (middle), and Pick's disease (right) with distinct regional binding patterns. [Courtesy of Aprinoia.]

For its part, APN-1607 has been in research use for PSP and other disorders for seven years. More than 3,000 people have been scanned with it. Developed by Higuchi from the precursor PBB3, and originally known as PM-PBB3, the tracer now goes by the moniker florzolotau.

In PSP, its signal matches known patterns of fibril distribution, and correlates with clinical severity (Mar 2020 conference news). In addition, the PET signal matched subsequent autopsy findings from one PSP, one CBD, and one Pick’s patient (Tagai et al., 2020).

Based on these data, and the unmet need for 4R tracers, the FDA granted APN-1607 orphan drug designation for PSP in 2017. Fast Track was added May 8 2024 (see press release). The agency decided that a single Phase 3 trial, without formal autopsy confirmation, would be sufficient to demonstrate that this tracer would help diagnose PSP, Aprinoia’s Brad Navia told Alzforum. The pivotal trial will take place at several sites in Canada, Germany, Japan, Taiwan, the U.K., and the U.S., and will enroll 130 people with suspected PSP.

Beyond PSP, APN-1607 has also been used to scan about 70 people with CBD, and 140 people suspected of having tauopathy-driven frontotemporal lobar degeneration. Aprinoia will conduct Phase 3 trials in these conditions as well, Navia said. The company is also developing APN-1607 for AD, with a Phase 3 AD trial underway in China, and a Phase 2 study starting in the U.S.—Madolyn Bowman Rogers


  1. Mathis, Vasdev, and colleagues provide a model protocol for designing and evaluating PET radioligands based on chemical, pharmacokinetic, and pharmacodynamic properties. The validity of this approach will soon be examined by a clinical PET study for AD and non-AD tauopathy cases. The new tracer, 18F-OXD-2314, has a relatively long backbone structure (indol-pyridine-piperidin BBQ) similar to 18F-florzolotau, and presumably these compounds bind to hydrophobic grooves spanning stacked β-strands of tau molecules (Shi et al., 2021). As these groove-shaped binding pockets are present across the cross-beta motif of diverse tau fibril folds, such “long” chemicals are potentially reactive with a wide range of tau pathologies.

    The notable difference between OXD-2314 and florzolotau is that the former contains a pyridine ring instead of a butadiene linker to circumvent photoisomerization issues. I sense that the florzolotau backbone, including the slender linker portion in the middle, may show a lock-and-key fit to the wavy surface of the groove, but this view needs to be tested by in-depth molecular docking.

    Among various assessments in the current study, homogenate binding assays to determine the affinity of the compounds have illustrated the most intriguing features, as the dissociation constants (KDs) of OXD-2314 in PSP, CBD, and PiD were considerably lower than those of florzolotau. A slightly complicated issue is that the homologous blockade curves for OXD-2314 were well described by a two-site binding model, and KDs of this compound for the high-affinity binding site were compared to KDs of florzolotau.

    There is no information on whether a one- or two-binding model was employed for the fit to the florzolotau binding curves to determine KDs. It is also unclear whether the low-affinity, high-capacity binding of OXD-2314, which accounts for 20 percent to 30 percent of the total binding, reflects tau or non-tau pathologies. Another interesting point is the lack of displacement of 3H-OXD-2314 binding with florzolotau, contrary to the notion that both ligands could be docked with the grooves in the cross-beta spine of tau fibrils. The Bmax/KD values of OXD-2314 were also very high in a three-digit range and should be compared with those of florzolotau.

    Moreover, the Bmax/KD values of OXD-2314 were 550 in the AD cortex, 580 in the PSP cortex, and 1,000 in the PSP subcortex, indicating greater binding of this compound in PSP than in AD. This might not be consistent with autoradiographic findings since intense labeling was noted in AD rather than PSP slices. The Bmax/KD values of OXD-2314 were 371 in the CBD frontal cortex, while very strong displaceable binding was homogeneously observed in the autoradiography of a CBD frontal cortical sample.

    The rodent and monkey PET data revealed ideal pharmacokinetic and metabolic profiles of 18F-OXD-2314. This ligand was also shown to react with tau aggregates in transgenic mice, and it would be nice to demonstrate the in vivo labeling of tau deposits in rTg4510 mice with 18F-OXD-2314, which could be compared to the florzolotau data in the literature (Kimura et al., 2022). For the advancement of diagnostic technologies, it will be crucial to carry out comparative tests of different tau PET probes with proper methodologies in non-clinical and clinical settings.


    . Cryo-EM structures of tau filaments from Alzheimer's disease with PET ligand APN-1607. Acta Neuropathol. 2021 May;141(5):697-708. Epub 2021 Mar 16 PubMed. Correction.

    . A quantitative in vivo imaging platform for tracking pathological tau depositions and resultant neuronal death in a mouse model. Eur J Nucl Med Mol Imaging. 2022 Nov;49(13):4298-4311. Epub 2022 Jul 8 PubMed.

  2. The development of [18F]OXD-2314 heralds a promising advance in the imaging of tauopathies, particularly the non-Alzheimer's disease (non-AD) type. This new tracer, designed through a ligand-based approach, shows high affinity and selectivity for 4R-tau aggregates, characteristic of diseases such as progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD). While [18F]OXD-2314 has shown promising preclinical results with higher binding affinity in PSP and CBD tissues and favorable pharmacokinetic properties, it has not yet been tested in human subjects.

    Therefore, it will be important to compare it clinically with [18F]PI-2620 and [18F]PM-PBB3, aka APN-1607, which have been established as reliable tracers for imaging tau pathology in both AD and non-AD tauopathies. [18F]PI-2620 and [18F]PM-PBB3 have been evaluated in patients and have shown the ability to image tau aggregates effectively in conditions such as PSP and corticobasal syndrome (CBS) (Brendel et al., 2020Palleis et al., 2021; Tagai et al., 2021). In addition, a small number of patients who have been imaged with [18F]PI-2620 and undergone autopsy upon death, corroborated the binding of the tracer to regions of the brain with 4R tau accumulation (Selmann et al., 2024). 

    The next steps for [18F]OXD-2314 involve crucial, human PET studies to validate its preclinical potential. If [18F]OXD-2314 can match or exceed the performance of [18F]PI-2620 or [18F]PM-PBB3 in clinical trials, it may become a valuable addition to the tool kit for imaging tau pathology. That [18F]OXD-2314 appears to have a unique binding site different than existing tau radiotracers is quite interesting. This may provide a different imaging profile, with respect to both on target and off target binding.

    The success of [18F]OXD-2314 in upcoming human studies will be eagerly anticipated, potentially enhancing the landscape of tau imaging in neurodegenerative diseases.


    . Assessment of 18F-PI-2620 as a Biomarker in Progressive Supranuclear Palsy. JAMA Neurol. 2020 Nov 1;77(11):1408-1419. PubMed.

    . In Vivo Assessment of Neuroinflammation in 4-Repeat Tauopathies. Mov Disord. 2021 Apr;36(4):883-894. Epub 2020 Nov 27 PubMed.

    . In Vivo Assessment of Neuroinflammation in 4-Repeat Tauopathies. Mov Disord. 2021 Apr;36(4):883-894. Epub 2020 Nov 27 PubMed.

    . Neuronal and oligodendroglial but not astroglial tau translates to in vivo tau-PET signals in primary tauopathies. 2024 May 07 10.1101/2024.05.04.592508 (version 1) bioRxiv.

  3. Lindberg et al. provide compelling evidence for OXD-2314 as a new lead candidate for imaging non-AD tauopathies, and AD. Indeed, 3HOXD-2314 had selective and stronger binding affinity for tau aggregates of PSP and CBD in postmortem tissue compared with other tritium-labeled PET ligands, PI-2620, florzolotau, OXD-2115. The authors also provided evidence that OXD-2314 penetrates brain. To date, tau PET radiotracers have been optimized for the detection of the 3R/4R tau aggregates related to AD.

    Research has been studying tracers that work for AD to determine their usefulness in non-AD tauopathies, but I don’t think that this is the best approach. For example, currently available radiotracers (e.g. TAUVID) show low-intensity, variable binding with small effects in humans with CTE tau aggregates. While CTE is a 3R/4R tauopathy and was not examined here, existing data suggest that tau might be predominantly 4R in the early stages and therefore [18F]OXD-2314 could be a potential game-changer for the in-vivo detection of CTE tau. The same could be true for PSP and CBD.

    There needs to be discovery efforts, like those of Lindberg et al., that develop and optimize tau radiotracers for non-AD tauopathies instead of relying on what works for AD. Overall, this is an important development for the AD/ADRD field, and in vivo data from human trials will be much anticipated. 

  4. Since the development of tau PET tracers capable of detecting the 3R/4R tau pathology in Alzheimer’s disease, there have been increased efforts to find new tracers that specifically detect 4R tau in PSP and CBD. 

    This work by Lindberg et al. from the Vasdev lab represents a significant contribution to the field with the addition a novel chemical group of tau PET tracer compounds. The preclinical evaluation and characterization of these compounds are thorough. The lead compound, OXD-2314, demonstrates good selectivity for tau over Aβ and α-synuclein, with no obvious off-target binding in vitro.

    However, OXD-2314 faces similar challenges regarding specificity for 4R tau versus mixed 3R/4R tau and 3R tau, as seen with previous 4R-tau PET compounds (PI-2620 and florzolotau). Nonetheless, OXD-2314 appears to have a slightly higher affinity for 4R tau, particularly in the case of CBD-related tau. A key difference between these tracers is that OXD-2314 seems to bind to a distinct site on tau aggregates compared to PI-2620, florzolotau, tauvid, and MK-6240, as its binding cannot be blocked by these compounds in competition experiments. Potentially this could explain the increase in affinity.

    Although the ideal 4R tau tracer should be specific and preferably not bind to mixed 3R/4R or 3R tau, an increase in affinity that allows for the detection of cortical and subcortical 4R tau would still be a significant advancement. Differences in cortical and subcortical binding patterns could still help differentiate the various tau-related neurodegenerative disorders from one another. In vivo, OXD-2314 shows rapid brain uptake and washout in rodent and non-human primate models, along with low non-specific binding in these models, which is promising.

    Overall, this novel PET tracer class is a valuable addition to the array of tau PET tracers and shows promise in preclinical testing. The critical question now is whether the tracer will perform as well in vivo in humans. I look forward to seeing the results from these scans.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Autopsies Confirm That PI-2620 Binds 4R Tau Deposits
  2. Flock of New Folds Fills in Tauopathy Family Tree
  3. Primary Tauopathies Get New PET Ligands

Paper Citations

  1. . Radiosynthesis, In Vitro and In Vivo Evaluation of [18F]CBD-2115 as a First-in-Class Radiotracer for Imaging 4R-Tauopathies. ACS Chem Neurosci. 2021 Feb 17;12(4):596-602. Epub 2021 Jan 26 PubMed.
  2. . High-Contrast In Vivo Imaging of Tau Pathologies in Alzheimer's and Non-Alzheimer's Disease Tauopathies. Neuron. 2021 Jan 6;109(1):42-58.e8. Epub 2020 Oct 29 PubMed.

External Citations

  1. press release

Further Reading

Primary Papers

  1. . Ligand-based design of [18F]OXD-2314 for PET imaging in non-Alzheimer's disease tauopathies. Nat Commun. 2024 Jun 14;15(1):5109. PubMed.