Will Tau Drug Show Its True Colors in Phase 3 Trials?

image See Q&A With Claude Wischik .

2 October 2012. Remember Rember®—the blue dye that made headlines in 2008 when Phase 2 clinical trial data suggested it slowed decline in people with Alzheimer’s disease (see ARF related news story)? A revamped version of this compound, which presumably inhibits tau aggregation, is now heading toward Phase 3 testing, though in a different tauopathy. TauRx Therapeutics—a Singapore-based biotech company with research operations in Scotland—announced last month that it received U.S. and European regulatory approval to launch a 12-month, global Phase 3 trial of its experimental drug in behavioral-variant frontotemporal dementia (bvFTD). “The FTD field is excited to finally have an agent worthy of a clinical trial as a potential disease-modifying agent. This is a big deal,” said Bradley Boeve of Mayo Clinic, Rochester, Minnesota.

The company issued a news release on the bvFTD trial in September, just after researchers and clinicians gathered in Manchester, U.K., for the 8th International Conference on Frontotemporal Dementias (see ARF related news story). The company did not announce the study, or report any preclinical data backing it, at the conference. Instead, Claude Wischik of the University of Aberdeen, U.K., the company’s CEO and founder, briefed some 30 scientists—potential site investigators in the Phase 3 bvFTD trial—in a closed meeting held before the main conference.

TauRx also plans to test its compound in two Phase 3 AD trials: a 12-month study of 833 people with mild to moderate AD, and an 18-month study of 500 people with mild AD. These will begin enrolling in late 2012, Wischik said.

In the previous Phase 2 AD trial, 12-month data appeared to suggest that 60 mg Rember taken orally three times a day slowed cognitive decline as measured by the ADAS-Cog. The 100 mg dose showed no benefit. The active ingredient in Rember is methylthionine (MT), dosed in capsules in the form of its chloride salt (MTC), which is commonly known as methylene blue. Wischik and colleagues told Alzforum that a formulation defect delayed release of MT from high-dose capsules and prevented absorption (see Wischik Q&A). Even after correcting the formulation defect, the researchers found that high doses were not absorbed properly in the presence of food. “People need to take [the compound] with food, or else they get nauseous,” Wischik said.

The company has since developed a modified version of the drug, LMTX™, which it will use in the upcoming Phase 3 trials. The new compound is a stabilized, reduced form of methylthionine, and is more readily absorbed and tolerated at a 10-fold higher dose than Rember®, Wischik said. So far, TauRx has secured LMTX patents in Europe, the U.S., Canada, South Korea, Hong Kong, and Singapore.

Why launch a Phase 3 trial in bvFTD before AD? It is logical to test a compound targeting tau aggregation in multiple tauopathies, and a bvFTD trial seems justified by the current lack of treatment options for this disorder, suggested Dirk Beher of Merck Serono S.A., Geneva, Switzerland, in an e-mail to Alzforum. In 2010, TauRx received orphan designation for LMTX, which enables faster approval of experimental therapies for serious, rare diseases such as FTD. Beyond company incentives, “these indications may provide an earlier and more cost-effective proof of concept before moving into more expensive Phase 3 AD trials,” Beher noted.

The bvFTD trial will take place at 25 to 30 sites worldwide, enrolling 180 adults with clinical and brain scan evidence of probable bvFTD. The trial will test a single dose, 100 mg LMTX twice a day for 12 months. Half of the participants will be on this regimen, while the other half will take placebo tablets that, curiously, will also include 4 mg LMTX as a colorant to maintain blinding. After the 2008 AD trial, researchers suspected that patients knew they were on Rember because it tints the urine, and sometimes eyes, blue-green, breaking the study blinding. Though colorless in solution, LMT/LMTX converts to its oxidized, colored form (MT+) when exposed to air. (For more on the practice of adding study compounds to placebo, see author Q&A below.)

Primary outcome measures will be global and cognitive changes from baseline in the Modified Alzheimer’s Disease Cooperative Study—Clinical Global Impression of Change (ADCS-CGIC) rating and Addenbrooke’s Cognitive Examination—Revised (ACE-R); other scales constitute secondary outcome measures. According to the study’s ClinicalTrials.gov entry, this Phase 3 bvFTD trial will not include biomarkers, such as CSF tau measurement, to verify target engagement. It is unclear whether changes in CSF tau would reflect cognitive improvement, Wischik said. Other researchers agreed that target engagement is hard to demonstrate for tau. One scientist who attended the closed meeting in Manchester and spoke on condition of anonymity said spinal taps were considered a threat to recruitment. (For more on CSF, see author Q&A below.)

Prior to the Phase 3 trial, Wischik said his team did experiments in minipigs and monkeys, as well as in healthy elderly volunteers (see, e.g., Phase 1 trial), to determine bioavailability and other pharmacokinetic properties of the compound. That work is unpublished. In addition, the scientists showed that methylene blue mitigates pathology and cognitive defects in several lines of tau transgenic mice. TauRx scientists presented the mouse data at the 2008 International Conference on Alzheimer’s Disease in Chicago. Since then, preclinical findings have been confirmed in other tauopathy mouse strains by several labs in the privately funded Consortium to Study Tauopathies, established in 2010 at the University of California, San Francisco. These data, as yet unpublished, “make me feel that [testing LMTX in FTD patients] is a worthwhile endeavor,” said another scientist who is involved in the Phase 3 bvFTD trial and requested anonymity while final ethics review and approvals are pending.

The company expects U.S. sites to begin enrolling for the Phase 3 bvFTD study in November or December of this year. Enrollment should start shortly thereafter in the U.K., Germany, Singapore, Australia, the Netherlands, and Canada. The sites are not listed on ClinicalTrials.gov; according to sources familiar with the trial, site IRBs have not granted permissions yet.

In 2008, researchers expressed concern about the design and analysis of the earlier Phase 2 AD trial (see ARF related news story). Even so, it spurred other research groups to explore the mechanism of action of methylene blue in model systems. Some studies suggest the compound blocks aggregation of not only tau, but also several other proteins involved in neurodegeneration—TDP-43, huntingtin, α-synuclein, even Aβ to some extent (see ARF related news story; Sontag et al., 2012), whereas other studies find the compound to be ineffective (van Bebber et al., 2010). Other work suggests methylene blue may relieve tauopathy by stimulating autophagy (Congdon et al., 2012), boosting proteasome activity (Medina et al., 2011), or generally showing a combination of pleiotropic effects that may make it beneficial (O’Leary et al., 2010).

A clinical piece of support for the Phase 3 bvFTD trial comes from a small, unpublished study Wischik presented at the closed investigators’ meeting in Manchester. Based on the Phase 2 data in AD, the principal investigator of that trial—Peter Bentham of Birmingham and Solihull Mental Health NHS Foundation Trust, UK—treated seven of his FTD patients with 60 mg Rember three times daily for 15 months under provisions for compassionate use in the U.K. These patients had magnetic resonance imaging (MRI) scans suggesting frontal and temporal lobe atrophy that met bvFTD diagnosis criteria. Over the 15-month treatment, their ACE-R scores hardly dropped, while historical data suggest they should have declined by 15 points (Kipps et al., 2008). “Even with just seven cases, the result was statistically significant at the nine-month timepoint,” Wischik said. Compared to the decline expected from historical data, he noted, “the implied effect size was statistically significant at all three-month timepoints measured.” Since it was a small, uncontrolled study, those data have not been published, he said. Other scientists cautioned against interpreting open-label treatment of seven patients, noting that in the AD field, for example, there was a single-center analysis of open-label anti-amyloid treatment with AN1792 that was not reflected in the overall clinical trial (Hock and Nitsch, 2005).

PubMed contains no publications on Rember’s preclinical or clinical development, nor on LMTX. Wischik’s team has not yet submitted the Phase 2 AD trial results to a peer-reviewed journal. “Clinicians in general don’t believe Phase 2 data in AD anymore, and publishing won’t make any difference. All they are going to believe are Phase 3 results,” Wischik told Alzforum. Besides, “there is nothing to hide,” he said. “The data have been presented in excruciating detail to regulatory authorities in the U.S. and Europe. They have given the go-ahead for our Phase 3 program based on the Phase 2 data, and we are not required to repeat a Phase 2 study with LMTX.”

Funding for the bvFTD and AD Phase 3 programs comes largely from individual and private investors in Singapore and East Asia, as well as North America. Wischik said that, as of December 2011, TauRx had raised $170 million, and has commitments for an additional $120 million this year. He cofounded the company in 2002 with Kwang Meng Seng, a physician and venture capitalist.—Esther Landhuis.

Q&A With Claude Wischik. Questions by Esther Landhuis.

Q: The ClinicalTrials.gov entry says placebo tablets will contain 4 mg LMTX as "colorant to maintain blinding." Does this truly maintain blinding? It seems that 100 mg drug would appear darker blue, while the 4 mg in the placebo tablet would come out as a lighter blue color in urine. Has this been tested? Are the actual drug and placebo pills, in fact, indistinguishable in color when they come out in urine/feces?

A: We have undertaken dose-ranging studies and measured the degree of urinary and fecal discoloration. Urinary discoloration is not dose-dependent (DiSanto and Wagner, 1972). We have been able to demonstrate sufficient overlap in terms of discoloration of excreta to permit the blind to be maintained. This point was discussed with and approved by the FDA.

Q: It is unusual for clinical trials to mix drug into the placebo pill. Did the company try food coloring (or some other form of colorant besides the drug itself) to satisfy blinding?

A: We did test a range of potential food colorants. The parent proved to be most reliable and appropriate.

Q: Will the trial use any biomarkers, such as CSF tau measurement, to verify target engagement?

A: There are still many unanswered questions about what the cerebrospinal fluid (CSF) tau measurement biomarker is showing, and how precisely this correlates to verification of target engagement. The assay claims to measure phosphorylated tau (p-tau), and doesn’t appear to measure the tau aggregation, which LMTX is targeting. P-tau is very proteolytically susceptible, and it is unclear how such an antigen could get to and survive in CSF. The only tau released from extracellular tangles is a short fragment from the core of the tau molecule (Mena et al., 1995). Furthermore, the CSF tau measurements only look at antibody immunoreactivity, and we do not know as yet exactly what protein species is being measured. Until it has been sequenced, we can’t even say for sure it is tau. Until these questions are answered, it could potentially be misleading to use such biomarkers to evaluate the efficacy of LMTX in clinical studies.

Increased levels of total tau and p-tau and decreased levels of Aβ immunoreactivity in CSF are indicative of AD, and in combination they can provide good sensitivity and specificity for diagnosis. Nonetheless, evidence that changes in the levels of these CSF biomarkers follow the clinical progression of the disease is lacking. The corollary to this is that decreases in CSF tau may not reflect cognitive improvement. Recent data from the EFNS Stockholm meeting supporting the engagement of bapineuzumab with its target based upon changes on CSF Aβ in the absence of clinical efficacy have been discussed on Alzforum (ARF related news story) and have been treated with mixed reactions. It is not clear why a drug targeting amyloid should result in a lowering of CSF tau immunoreactivity.

It is assumed that a drug that prevents tau aggregation and dissolves tau aggregates will lead to the release of tau into CSF. This is untested, and several other possibilities exist that would refute such a hypothesis. The source of tau may be extracellular, and its removal may only affect transneuronal spread within the brain, whereas a major objective for TAI would be to prevent intracellular aggregation of tau before there is loss of neuronal function on its way to becoming an extracellular tangle.

Antibodies to different phosphorylated and non-phosphorylated tau epitopes were used to identify three histologically distinct types of neurofibrillary tangles in Alzheimer's disease by Bondareff et al. (Bondareff et al., 1994). Intracellular tangles (Type 1) were identified by antibodies recognizing epitopes throughout the tau molecule, including the amino-terminus. Compact extracellular tangles (Type 2) were characterized by the loss of amino-terminal and, hence, phosphorylation-dependent immunoreactivity, but retention of other tau epitopes. Dispersed extracellular tangles (Type 3) were characterized by the presence of epitopes associated only with the microtubule binding region and the C-terminus. The three stages of neurofibrillary degeneration can be understood as a sequential stripping of the proteolytically sensitive fuzzy outer coat of the paired helical filament, leaving behind the proteolytically stable core which comprises solely the repeat-domain core fragment that constitutes the core structure of the PHF (Wischik et al., 1988). Phosphorylated tau epitopes are located exclusively in the fuzzy outer coat of the PHF. Thus, when they become extracellular after neuronal death, the neurofibrillary tangles are non-phosphorylated (Bondareff et al., 1994), and so clearance of p-tau into CSF from this source is not possible. It is, therefore, difficult to understand in terms of an underlying mechanism how presence of p-tau in CSF is a marker of neurodegeneration.

The regulators have not accepted any validated biomarker for AD, but are obviously keen that such a biomarker or combination of biomarkers will eventually be validated. There are, however, many unanswered problems in relation to CSF biomarkers. Sample timing, collection, storage, and transport introduce variability. The nature of the analyte to be measured is still being investigated, with antibodies to different epitopes being used in assays and novel assays still under development.

Our PET and SPECT findings from the Phase 2 trial suggest that such molecular imaging may be a more direct biomarker of disease progression, and we are including imaging as an endpoint in one of our Phase 3 AD trials. Indeed, TauRx is developing novel imaging markers to permit direct detection of aggregated tau in the brain by PET imaging.

Q: Did the company do dose-ranging to determine which doses reach the brain and engage the target, and generally satisfy CNS PK/PD requirements? This has become standard for drug trials in the field.

A: We have undertaken extensive animal studies and Phase 1 studies in humans. We have determined plasma and brain levels in minipig, which is the most representative species for PK purposes and for measuring MT absorption and disposition. We have also undertaken radiolabeled MT distribution studies, including brain in minipig. Finally, we have determined efficacy on tau pathology in two different tau transgenic mouse models that produce tau pathology. We have also measured the brain concentration at which effect on tau pathology is observed, and have related this back to plasma levels in humans, and plasma and brain levels in minipigs.

Q: The 12-month Phase 3 in mild to moderate AD and the 18-month Phase 3 in mild AD are slated to begin enrolling in October, according to the ClinicalTrials.gov postings. Correct?

A: We are on track for 2012, and will make an announcement regarding the AD trial program soon.

Q: The Phase 2 study in people already taking medications for mild to moderate AD began enrollment in August, according to the ClinicalTrials.gov postings. Correct?

A: Yes. This is a small safety study to determine the safety of co-medicating with LMTX and pre-existing stable acetylcholinesterase inhibitor and/or memantine therapy in mild to moderate Alzheimer’s. The study has regulatory approval and is due to start dosing shortly in the U.K. and Germany.

Q: You mentioned that the $170 million TauRx has raised thus far, plus the $120 million committed, will cover Phase 3. Do you mean all three trials in bvFTD and AD, or only the FTD one?

A: We have enough raised/committed to cover all three trials in their basic form as currently planned. We plan further fundraising.

Q: Most of the funding comes from private investors, not the more typical drug development investors, i.e., pharma or the biotech venture capital industry. Correct?

A: Most of the funding for our Phase 3 FTD and AD trials comes from private and institutional investors. We have a small discovery partnership in the field of development of novel tau imaging markers, which was with Bayer Healthcare and, following their divestment of this division, is now with Piramal Imaging, with whom we are currently in discussions. We have no immediate plans or needs to partner with a major pharma company. At the moment, we are able to carry the bulk of the risk of our own programs. We always remain open to discussion.

Q: Except for a 2009 Psychopharmacology paper, a mouse study on MTC/rivastigmine, and some conference abstracts, I found no publications on preclinical or clinical development of Rember or LMTX. Correct?

A: We will consider publishing some of the extensive body of work on preclinical and clinical development for the general scientific audience in due course. We published seven abstracts relating to preclinical and clinical development on Rember for ICAD in 2008. Reviews originally published as book chapters are available on the TauRx website. Our objective has been completion and presentation of full reports of the preclinical and clinical development program for the purpose of regulatory approval of the Phase 3 program.

Comments

  1. This is a well-reported story. I can offer some general regulatory remarks regarding the movement of a drug from Phase 2 to 3. There is no such thing as regulatory approval for Phase 3. In Europe, a drug sponsor can do all its development without getting any blessing from the authorities. The only thing sponsors need is to get their clinical trials approved. This is mostly a matter of having a safety profile that is reassuring and a rationale that suggests that efficacy is likely. So in large part, the sponsors' burden is mostly on showing safety.

    In the U.S., it is slightly different. Sponsors meet with the FDA at the end of Phase 2 to discuss the criteria and conditions that would make their Phase 3 be considered successful. At that time point, they make an agreement with the FDA, which more or less means that if sponsors achieve certain efficacy landmarks, they can go ahead. The FDA might share with sponsors its impression based on prior data whether Phase 3 is likely to succeed or not, but this is confidential. If the safety does not raise any concerns, sponsors can always go on. The agreement between the FDA and the sponsor may be harder to make if the data are shaky at that meeting. But the bottom line is, the sponsor is free to go to Phase 3 provided there are no safety concerns that cannot be managed.

    A word on the science. FTD is a good model for tauopathy, so it makes sense to try a drug that potentially affects tau in FTD. It is less immediate that one can translate a positive result to AD, because AD is a mixed condition. I have recommended for some time that in AD, a combination of a tau and an amyloid drug is most likely to yield a measurable result.

    View all comments by Cristina Sampaio

  2. It is notable that a tau-directed therapeutic agent is progressing into Phase 3 clinical testing, particularly given the recent failures of drugs designed to affect amyloid-β levels. However, it is somewhat disappointing that CSF tau/p-tau measurements are not included in all of these trials. Apparently, TauRx believes there are too many uncertainties in how to interpret CSF tau levels. This may be correct for bvFTD, since unlike AD, there doesn’t appear to be an elevation of tau/p-tau in these patients for unknown reasons. However, determining whether drug treatment causes any upward or downward movement in CSF tau/p-tau, particularly if CSF changes correlate with behavioral changes, would be informative for future trials. Moreover, ADNI and other data would seem to justify looking at CSF tau in the AD trials, as was done in the recent bapineuzumab Phase 3 study.

    There appears to be some concern that including lumbar punctures may affect recruiting. This may be true to some extent, although many clinicians suggest that this can be overcome by providing patients with information on the procedure. Arguably, the likely modest effect of CSF draws on recruitment is outweighed by the potential value that CSF tau/p-tau measurements might provide regarding target engagement/efficacy in these trials. It is noted that the smaller of the two planned AD trials lists CSF markers of AD under "Other Outcome Measures," so perhaps we can look forward to some data on CSF tau/p-tau in TRx0237-treated patients.

    View all comments by Kurt Brunden

  3. Informative and thorough discussion, very valuable since Phase 3 trials of LMTX have started in the U.S.

    View all comments by Rebecca Price

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Chicago: Out of the Blue—A Tau-based Treatment for AD?
  2. C9ORF72 Steals the Show at Frontotemporal Dementia Meeting
  3. Vienna (and Burkina Faso): What's New With Methylene Blue?
  4. Bapineuzumab Phase 3: Target Engagement, But No Benefit

Paper Citations

  1. Sontag EM, Lotz GP, Agrawal N, Tran A, Aron R, Yang G, Necula M, Lau A, Finkbeiner S, Glabe C, Marsh JL, Muchowski PJ, Thompson LM. Methylene blue modulates huntingtin aggregation intermediates and is protective in Huntington's disease models. J Neurosci. 2012 Aug 8;32(32):11109-19. PubMed.
  2. van Bebber F, Paquet D, Hruscha A, Schmid B, Haass C. Methylene blue fails to inhibit Tau and polyglutamine protein dependent toxicity in zebrafish. Neurobiol Dis. 2010 Sep;39(3):265-71. PubMed.
  3. Congdon EE, Wu JW, Myeku N, Figueroa YH, Herman M, Marinec PS, Gestwicki JE, Dickey CA, Yu WH, Duff KE. Methylthioninium chloride (methylene blue) induces autophagy and attenuates tauopathy in vitro and in vivo. Autophagy. 2012 Apr;8(4):609-22. PubMed.
  4. Medina DX, Caccamo A, Oddo S. Methylene blue reduces aβ levels and rescues early cognitive deficit by increasing proteasome activity. Brain Pathol. 2011 Mar;21(2):140-9. PubMed.
  5. O'Leary JC, Li Q, Marinec P, Blair LJ, Congdon EE, Johnson AG, Jinwal UK, Koren J, Jones JR, Kraft C, Peters M, Abisambra JF, Duff KE, Weeber EJ, Gestwicki JE, Dickey CA. Phenothiazine-mediated rescue of cognition in tau transgenic mice requires neuroprotection and reduced soluble tau burden. Mol Neurodegener. 2010;5:45. PubMed.
  6. Kipps CM, Nestor PJ, Dawson CE, Mitchell J, Hodges JR. Measuring progression in frontotemporal dementia: implications for therapeutic interventions. Neurology. 2008 May 27;70(22):2046-52. PubMed.
  7. Hock C, Nitsch RM. Clinical observations with AN-1792 using TAPIR analyses. Neurodegener Dis. 2005;2(5):273-6. PubMed.
  8. DiSanto AR, Wagner JG. Pharmacokinetics of highly ionized drugs. I. Methylene blue--whole blood, urine, and tissue assays. J Pharm Sci. 1972 Apr;61(4):598-602. PubMed.
  9. Mena R, Edwards P, Pérez-Olvera O, Wischik CM. Monitoring pathological assembly of tau and beta-amyloid proteins in Alzheimer's disease. Acta Neuropathol. 1995;89(1):50-6. PubMed.
  10. Bondareff W, Harrington C, Wischik CM, Hauser DL, Roth M. Immunohistochemical staging of neurofibrillary degeneration in Alzheimer's disease. J Neuropathol Exp Neurol. 1994 Mar;53(2):158-64. PubMed.
  11. Wischik CM, Novak M, Edwards PC, Klug A, Tichelaar W, Crowther RA. Structural characterization of the core of the paired helical filament of Alzheimer disease. Proc Natl Acad Sci U S A. 1988 Jul;85(13):4884-8. PubMed.

Other Citations

  1. See Q&A With Claude Wischik

External Citations

  1. Phase 3 trial
  2. news release
  3. 12-month study
  4. 18-month study
  5. Phase 1 trial
  6. Consortium to Study Tauopathies

Further Reading

Annotate

To make an annotation you must Login or Register.