7 August 2008. Of all clinical trials reported at the International Conference on Alzheimer’s Disease (ICAD) held 26-31 July in Chicago, one targeting the protein tau was the therapy du jour, both in terms of buzz among scientists and media coverage. Never mind that the scientific buzz ranged from puzzled to intensely skeptical, whereas the media coverage was mostly breathless (see Chicago Sun Times headline Breakthrough Drug Fights Alzheimer’s; Alzheimer’s ”wonder drug”). This first tau-based Phase 2 trial reported at an AD conference fell on fertile ground in part because the research community is so urgently awaiting the development of tau-based therapies. None as yet directly address tau pathology, the second hallmark of AD, and consensus is growing that truly effective, disease-modifying therapy will have to attack both amyloid and tau. Some tau-related targets, such as its phosphorylating kinases or the GSK3β enzyme, have been extensively studied in academia and industry, and pharmaceutical anti-tau programs in general are said to be revitalized; however, no drugs based on these programs have yet posted any success in Phase 2. Moreover, the definitive failure of Flurizan in Phase 3 and the complex, mixed results of Bapineuzumab in Phase 2, made it easy for this newcomer to steal the show.
And a newcomer it is. PubMed lists no publications on the new drug, or on preclinical research to prepare its human testing. From the perspective of AD researchers, the sponsors of the new drug burst on the scene out of left field, with eight abstracts at ICAD and a press briefing that garnered multiple television interviews (see, e.g., CNN clip).
The name behind the potential drug is Claude Wischik. Younger readers may not remember that Wischik made his mark in the AD field 20 years ago, when he published two groundbreaking papers on tau with Nobel Laureate Aaron Klug, Sir Martin Roth, and tau experts Michel Goedert and R. Anthony Crowther (Wischik et al., 1988; Goedert et al., 1988). After that, Wischik continued publishing in the tau field, though he has not reported new AD research findings in the public literature since 2001. Besides holding an appointment at the University of Aberdeen, Scotland, Wischik now is co-founder and chairman of TauRx Therapeutics, a privately held Scottish-Singaporean biotech company that officially launched its website 30 July this year during the ICAD conference.
Here is a summary of the drug and the data. Its trade name is RemberTM, its active compound is methyl thioninium chloride (MTC), a reducing agent better known as methylene blue. This is a deep blue dye used in analytical chemistry, as a tissue stain in biology, and in various industrial products such as ink, for example. Wischik told reporters and scientists that MTC interferes with tau aggregation by acting on self-aggregating truncated tau fragments.
The company conducted a Phase 2 study randomizing 321 people with mild or moderate AD to treatment with either placebo or one of three oral doses of MTC: 30 mg, 60 mg, or 100 mg three times a day. People taking AD drugs, i.e., acetylcholinesterase inhibitors or memantine, were excluded. The trial’s primary objective was to compare the effect of MTC to placebo on cognitive abilities measured by the ADAS-Cog battery at 24 weeks. In this phase of the study, patients were assessed every six weeks. Wischik did not show results of the entire cohort at 24 weeks; he showed results stratified by mild and moderate AD. For patients with mild AD, there was no difference between the groups at six months. For patients with moderate AD, Wischik reported a roughly 5.5-point decline on ADAS-Cog for the placebo patients versus a 1.5-point decline in the treated groups, resulting in an approximately four-point treatment effect. (If this holds up, it would be larger than currently approved drugs typically achieve.)
Secondary objectives included assessments at 50 and 84 weeks, brain imaging at 25 weeks, and safety and tolerability. For the 50-week time point, Wischik presented pooled results of mild to moderate patients. By 50 weeks, people on placebo had declined seven points on ADAS-Cog and those on 60 mg MTC by one point, yielding a treatment effect size of about six points. This translated to an 81 percent reduction in the rate of decline on the higher dose, Wischik said. (The low dose slope showed a 3.5-point decline.)
At 84 weeks/21 months, 30 mg data were not shown, but people on 60 mg appeared to have stabilized. If this finding stands the test of time, it would be a major advance in AD therapy. However, leading scientists cautioned that this was a large claim to stake on a single Phase 2 study, not to mention the first human study publicly reported on the compound. There were no placebo data by 84 weeks anymore, as the placebo period ended at 50 weeks; presumptive comparisons would have to be made against historical controls.
This clinical trial featured an imaging marker component, whose data looked supportive at first glance. Some patients (138 as per ICAD abstract, 125 as per presented poster) received a SPECT scan at baseline and again between 18 and 28 weeks into the study. Some patients (18 as per online abstract, 19 as per presented poster) received an FDG-PET scan at baseline and at around 26 weeks. The images were taken at participating sites in England and Singapore, and analyzed at the University of Aberdeen. With SPECT, patients on placebo showed reduced regional blood flow in AD-relevant areas, whereas patients on MTC did not. With FDG-PET, patients on MTC showed increased glucose use, whereas patients on placebo did not. (In both imaging modalities, the data on the posters indicated that the placebo groups were two to five years older than the MTC groups.)
On safety and tolerability, Wischik showed no data to the press but said in the subsequent scientific session that the major side effects were diarrhea, urinary urgency, and painful urination; there were also some dizziness and falls. Wischik said the side effect profile overall is similar to the three acetylcholinesterase inhibitors that are in wide use, but that diarrhea was more common.
On that point, it is worth noting that the talks excluded certain data that have become standard for pharmaceutical company presentations. For example, it is unclear how many people completed each arm of the trial, and what were the reasons for discontinuation. Other scientists in the field later wondered about how intent-to-treat analysis was handled, and how dropouts may have affected the power of the final data.
The 100 mg dose was ineffective. Wischik said that was due to interactions between the study drug and gelatine in the capsule wall. According to Wischik, this delayed absorption of the drug from the stomach to the intestine. However, rather than analyzing this group for what it was, warts and all, or leaving this group of patients out of the analysis, the investigators instead decided to combine the 100 mg group with the placebo group and compared this pooled set to the 30 mg and 60 groups. This is unusual. It means that all side effects people sustained from 300 mg MTC per day were tallied on the placebo side, making the 30 and 60 mg doses look better by comparison. (For sake of argument, it’s fair to assume that if Elan/Wyeth had pooled the placebo group with the ApoE4 carrier group, in whom the immunotherapy did not work so well, and then compared safety in this combined group with the ApoE4 non-carriers, that practice would have drawn protest.)
On MTC’s efficacy, as well, it is possible that including the highest dose with placebo affected the progression of the true placebo group, inflating the effect size. “If the investigators could have gotten statistical significance or a statistic with a p value <0.05 with the placebo group alone, then they would have done that. The 100 mg dose was associated with adverse events. It looks like the investigators decided that because it was ineffective, none of it was absorbed, and then they came up with the explanation about drug bound to the matrix of the capsule,” said Lon Schneider of University of Southern California in Los Angeles.
Wischik noted that the study was double-blind for the full 84 weeks. However, methylene blue colors the urine green, raising the question of how a study with this substance can stay blinded at all. MTC also has been reported to sometimes tint eyes blue. (Sorry folks, not the iris. It goes into the white part.)
During the press briefing and again after his talk, Wischik was asked whether the company had included CSF tau/phospho-tau measurements in the trial. (Biomarker data are increasingly built into clinical trials in AD to establish evidence that the drug at hand hits its intended target. Research shows that CSF tau and phospho-tau go up before AD diagnosis and during its course.) Wischik replied that TauRx did not, because prior research made it impossible to predict whether CSF tau should go up or down in response to treatment with MTC, and that he would have had to pre-specify this issue prior to the trial. In fact, biomarker pre-specification is not necessary for all Phase 2 trials. For example, a Phase 2 trial of Prana’s PBT2 compound, conducted in Europe, measured CSF Aβ42 precisely to find out which way it would change in response to the drug, as well as to obtain some measure of how well the drug crosses the blood-brain barrier in humans (see subsequent ICAD story). It’s unclear at this point how well methylene blue enters the human brain.
Scientists from TauRx did, however, present posters on tau-transgenic mouse lines developed by scientists at University of Aberdeen who are also employees at WisTa Laboratories, a company also headed by Wischik. According to the ICAD abstracts, MTC given intravenously to these mice at 5 mg/kg for 17 days ameliorated tau pathology and cognitive deficits in the Morris water maze. Interested AD researchers can attempt to reproduce and expand these findings in a variety of tau-transgenic mouse lines known to the AD research community. For comparative listings, see tau-transgenic mice, double-crosses, others; strains available through The Jackson Laboratory.
In terms of the compound’s potency, Wischik’s presentation stated that MTC dissolves tangle filaments isolated from brain with an effective concentration (EC50) of 0.15 μMolar and stops tau aggregation in cells with an EC50 of 0.56 μMolar in cells. For background, a compound’s half-maximal inhibitory concentration (IC50) is used for in-vitro assays where there is no absorption, distribution, metabolism, excretion (ADME) component. That is, the assay simply contains a target and a drug, and the investigator can assume that 100 percent of the drug finds the target. The EC50 is typically used in cell culture and is lower than the IC50 because the compound has to cross the cell membrane and distribute through the cells. The key parameter for humans is the effective dose (ED50). That is because the ADME component is large as the drug spreads in the body, and only a small fraction finds the desired target, especially if it is in the brain and intracellular. Between the EC50 and the ED50, there can be two to three orders of magnitude difference. Because of these issues, companies typically prefer compounds that start out with an IC50 in the low nanomolar range.
Other researchers at ICAD noted that TauRx plans to exploit MTC interactions with α-synuclein, as well, to develop RemberTM for Parkinson disease. The company’s website states this as well. These scientists questioned how specific the underlying molecular mechanism of a single pan-AD/tauopathy/synucleinopathy medication will prove to be.
MTC is available in the U.S. in tablet form. Called Urolene Blue, it is a grandfathered drug that predates the existence of the FDA, Wischik said. Urolene Blue is being used to treat the blood disorder methemoglobinemia. According to Wischik, it is also a mild antibiotic and in its earlier days used to be prescribed to treat urinary tract infections. According to Wikipedia, methylene blue was used against malaria a century ago.
TauRx has a use patent on a particular formulation of methylene blue, which Wischik said is purer, and hence possibly safer, than Urolene Blue. Both RemberTM and TauRx have new Wikipedia pages, though the former was on the site’s watch list for deletion on August 3.
In summary, Wischik said: “These Phase 2 data show for the first time that it is possible to halt the progression of AD with an anti-tangle treatment.” Most commentators felt that if there is a real treatment effect in this data, even just a kernel, this would be exciting as a general signal that targeting tau aggregation can in fact make people better. Researchers tend to express healthy skepticism when data are not available in peer-reviewed publications. Here, too, they reserved judgment of whether, or how much of, this signal is real. Many will attempt to reproduce the reported effects in tau models with available forms of methylene blue. Scientists voiced enthusiasm about the general approach of inhibiting tau aggregation, but tempered it with some doubt about whether this particular dye formulation will prove to be enough like a drug to be approvable for chronic consumption. The ICAD presentation has raised awareness of this approach; both regulators and investors will scrutinize the quality of the data in detail before deciding on next steps.—Gabrielle Strobel.