Does the Anti‐Tau Strategy in Progressive Supranuclear Palsy Need to Be Reconsidered? Yes

The prevailing framework for neurodegeneration posits that a wide range of clinical manifestations converging on a given pathology constitutes a disease. Further, the pathology of such disease must be pathogenic. Two logical consequences arise from this framework: (1) a measurement of the disease‐defining pathology represents a biomarker of that disease, and (2) the targeted elimination of that pathology would mean the end of that disease.

Theory does not meet practice. Validation of diseases as toxic proteinopathies has been challenged from a variety of sources. These include the lack of correlation between brain burden of these pathologies and disease severity or neuronal loss;¹, ², ³ the generation of behavioral and degenerative phenotypes in the absence of protein aggregation in knockout and knockdown animal models for tau,⁴ APP (whose cleavage produces amyloid‐β),⁵ and α‐synuclein;⁶ and the rarity for a single pathology to dominate the disease it defines.⁷, ⁸ In progressive supranuclear palsy (PSP), for example, tau‐only pathology occurs in only 8% of cases whereas co‐pathology with several other proteins is the rule.⁹ The most important argument against equating pathology with pathogenesis arises from trials of putative disease‐modifying interventions: success in brain protein reduction has yielded either no behavioral changes or worsening.¹⁰, ¹¹ While most of that evidence over the past two decades has come from anti‐amyloid approaches, anti‐tau and anti‐synuclein interventions appear set to dominate in this decade (Fig. 1).

FIG. 1.

Anti‐amyloid, anti‐tau, and anti‐synuclein therapies: molecular characteristics, cohorts, and outcomes. Alpha‐syn, alpha‐synuclein; Aβ, Amyloid beta; BACE, Beta‐secretase 1; NA, not available; PD, Parkinson's disease; FTD, frontotemporal dementia; AD, Alzheimer's disease; MSA, Multiple system atrophy; PSP, Progressive supranuclear palsy; eAD, early Alzheimer's disease; ePD, early PD; Mod, moderate; pro, prodromal; pre, preclinical; agg, aggregation; inhib, inhibitor; phosph, phosphorylation; ab, antibody.

Tau accumulation is a terminal event in neurodegenerative disorders. Because the increase of tau in the cerebrospinal fluid (CSF) of patients with Alzheimer's disease (AD) occurs closer to the onset of neurodegeneration than the decrease of amyloid‐β 42 (Aβ42), tau has been shown to correlate to a greater extent with dementia than Aβ42,¹² with insoluble tau predicting dementia better than insoluble amyloid‐β.¹³ However, a reduction in CSF (soluble) Aβ42 levels is the strongest predictor of conversion to AD (hazard ratio [HR], 16), an effect size four times greater than the increase in t‐tau (HR, 2.8) or p‐tau 2.6 (HR, 2.6),¹⁴ and of greater magnitude than the corresponding increase in brain amyloid burden.¹⁵ In the ‘primary tauopathy’ of PSP, it is not tau but a different marker of axonal cytostructural integrity, neurofilament light (NF‐L), which exhibits the closest association with neuronal loss and therefore the best prediction of PSP progression.¹⁶ But like tau, NF‐L is an undifferentiated marker of neuronal loss, a common end to many disorders, not the beginning of any.

Despite a combined negative output of 40 Phase II or III anti‐amyloid trials in AD,¹⁷ the strong allure of the gain‐of‐function toxic protein model of neurodegeneration has fostered a robust anti‐protein clinical trial pipeline, including one against tau. Of 12 anti‐tau therapies in development for AD, the first phase II trial of semorinemab, an IgG4 humanized anti‐tau antibody against the N‐terminal epitope, showed no differences in the rates of cognitive decline in any subgroup analyses of prodromal or mild AD patients compared with placebo.¹⁸ Two monoclonal antibodies binding to the N‐terminus of human tau have already shown futility in phase II trials in PSP: gosuranemab (BIIB092, PASSPORT trial, NCT03068468) and tilavonemab (C2N‐8E12, ARISE trial, NCT02985879).

The results of the gosuranemab trial remain unpublished but we suspect may be similar to those from tilavonemab.¹⁹ As with gosuranemab, tilavonemab was terminated earlier than anticipated at an interim evaluation by virtue of futility in changing the primary efficacy endpoint, the Progressive Supranuclear Palsy Rating Scale (PSPRS). This double‐blind, placebo‐controlled trial randomly assigned 378 PSP patients to placebo, tilavonemab 2000 mg, or tilavonemab 4000 mg, infused on days 1, 15, and 29, and every 28 days throughout the 52‐week study. There are three key takeaways from this study. First, the drug acted as designed: target engagement was demonstrated by a significant reduction in CSF tau. Second, the study would not have had a different outcome with a larger sample size. In fact, the PSPRS scores trended toward worsening in the treated arm compared to the placebo arm, a decline that seems significant by week 36 (by visual inspection of the slopes of decline). Third, brain volume changes also trended in the opposite direction to that hypothesized, with greater atrophy in the treated arms: compared with placebo, the least squares mean difference was −7.2 mm³ for tilavonemab 2000 mg and −6.3 mm³ for tilavonemab 4000 mg.

In sum, tilavonemab successfully lowered the CSF concentration of free tau, as intended, but without parallel improvement in clinical outcomes. The investigators offered the following explanations for these results: (1) extracellular tau may not be the “key to progression;” (2) “tilavonemab might not target the specific tau species that drives spreading of pathology;” and (3) “not enough tilavonemab enters the CNS to elicit a therapeutic effect.”¹⁹ Not included is the possibility that the hypothesis of tau accumulation as a primary pathogenic event in PSP may be false. Clinical trials should serve to confirm or reject their inspiring hypotheses. The tilavonemab trial was well designed and executed, attaining tau lowering but not the clinical benefits such tau change was hypothesized to generate. We disagree with the authors' proposal that “the findings provide potentially useful information for future investigations of passive immunization using tau antibodies for PSP.” The authors encouraged further anti‐protein trials with higher dosages, more sensitive endpoints, alternative epitopes, and earlier populations. While clinical trials cannot define underlying pathogenic mechanisms, a trial yielding clinical futility can and should be taken as evidence against the hypothesis if the desired biomarker change was attained by the intervention.

It behooves our field to take a step back to examine the big picture on our conceptual framework of neurodegeneration rather than continue searching for solutions in better trial designs, better measures, or better targets within protein epitopes or “strains.” The evidence accumulated suggests that protein aggregation represents a universal outcome of a broad range of biological, infectious, inflammatory, or toxic stressors; that amyloid‐β, α‐synuclein, tau, and other proteins, most often co‐aggregated together, signal the presence of an active source of neuronal stress upstream but are not themselves the source of stress; that they are associated with diseases but are not their cause.²⁰ In the case of tau, consideration must be given to shifting research efforts from a gain‐of‐function model where tau is targeted by modifying it (phosphorylation, cleavage, etc.), preventing its aggregation, or enhancing its clearance, to a loss‐of‐function model where normal tau levels and function are restored. The importance of tau in its native configuration is highlighted by the fact that tau knockout models exhibit a neurological phenotype, including dopaminergic dysfunction.²¹

Protein aggregation is a physical problem pertaining to loss of solubility, with sequestration into insoluble cross‐β fibers, also known as amyloids. Amyloid aggregation thus represents the end product of the soluble‐to‐insoluble phase transition of proteins, a functioning‐to‐nonfunctioning biological change. Many of the proteins that become amyloids are evolutionarily conserved and highly expressed in the nervous system, which not only points out to their important neurological functions, but also render them prone to aggregation at high or supersaturated concentrations.²² Thus, when exposed to unwanted nucleating agents (virus, pollutants, or other nanoparticles), they can easily precipitate into amyloid fibrils.²³ The high thermodynamic stability of amyloids renders them relatively inactive in terms of their ability to interact with other substrates; hence their inability to induce toxic effects.²⁴ This largely inert nature of amyloids can explain the lack of therapeutic effect from removing amyloids. However, as proteins transition into insoluble amyloids, there is a corresponding decrease in the pool of soluble precursors, which is associated with a loss of the normal functions they provide.

It is time to question our long‐held belief in the pathogenic role of aggregated proteins, falsify the toxic proteinopathy hypothesis of neurodegeneration, and cease further trials of anti‐tau, anti‐amyloid, and anti‐synuclein treatments. Rather than continuing to expose more lives to the iatrogenic harm of anti‐protein interventions, we can begin the process of examining the alternative hypothesis: the loss of soluble peptides as a more relevant source of neuronal toxicity than their accrual into amyloids. This will require steering efforts toward assessing the safety and efficacy of protein replacement using less‐ or non‐aggregating soluble peptide analogs. This type of “rescue medicine” can be evaluated in preclinical models, with confirmation or falsification in future proof‐of‐concept clinical trials. Protein replacement could precede or complement the launch of “precision medicine,” the matching of therapies to affected individuals whose etiologic factors such therapies can correct in order to limit the protein aggregation they trigger.

Author Roles

(1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the first draft, B. Review and Critique.

A.J.E.: 3A

K.E.: 3B

A.S.: 3B

Disclosures

Ethical Compliance Statement

This work did not require informed patient consent or the approval of an institutional review board. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest

The authors declare that there are no funding sources or conflicts of interest relevant to this work.

Financial Disclosures for the Previous 12 Months

Dr. Espay reports grants from NIH, grants from the Michael J. Fox Foundation, personal fees from Abbvie, personal fees from Neuroderm, personal fees from Neurocrine, personal fees from Acadia, personal fees from Acorda, personal fees from Sunovion, personal fees from Lundbeck, personal fees from USWorldMeds, personal fees from UCB, personal fees from Lippincott Williams & Wilkins, personal fees from Cambridge University Press, personal fees from Springer, personal fees from Kyowa Kirin, personal fees from Amneal. In addition, Dr. Espay is cofounder of REGAIN Therapeutics, owner of a provisional patent on compositions and methods for treatment and/or prophylaxis of proteinopathies. Dr. Ezzat is cofounder of REGAIN Therapeutics, owner of a provisional patent on compositions and methods for treatment and/or prophylaxis of proteinopathies. Dr. Sturchio is cofounder of REGAIN Therapeutics, owner of a provisional patent on compositions and methods for treatment and/or prophylaxis of proteinopathies.