Presentation of the first phase III study of TauRx Pharmaceuticals’ tau aggregation inhibitor LMTX at the Alzheimer’s Association International Conference yesterday divided opinion among seasoned observers and key opinion leaders alike. These viewpoints have ranged from it being another failed Alzheimer’s trial, where TauRx should be mocked for blatant attempts to spin positive data-dredge findings, to a promising development.
The truth is probably somewhere in between. And one byproduct is that the results have allowed TauRx to adjust the statistical analysis plan for its second phase III study, which could give the UK-Singaporean company a chance to prove its detractors wrong.
This first study, TRx-237-015, enrolled 891 subjects with mild or moderate Alzheimer’s disease (AD), who were treated for 15 months with one of two doses of LMTX or placebo. An interesting twist is that placebo in this case was a low dose of LMTX to stain bodily effluents blue and thus maintain blinding.
There is no dispute that this did not show an effect on its primary endpoints, but TauRx claims that as part of its secondary analysis it observed a profound effect on participants who were not on symptomatic treatment (acetylcholinesterase inhibitors or Namenda).
This was statistically significant in terms of cognitive assessment (ADAS-Cog), functional assessment (ADCS-ADL), and rate of brain atrophy, the last a purely quantitative measure and thus the most difficult to explain by chance. But TRx-237-015 only enrolled 136 patients who were not taking symptomatic AD drugs.
|Secondary analysis of TRx-237-015|
|Dose||ADAS-cog decline vs control||ADCS-ADL decline vs control||Brain atrophy by lateral ventricular volume vs control|
|LMTX monotherapy, 75mg bid||-6.3 +/-1.4 (p<0.0001)||6.5 +/-1.8 (p=0.0013)||38%+/-9% (p=0.0023)|
|LMTX monotherapy, 125mg bid||-5.8 +/-1.4 (p<0.0001)||6.9 +/-1.9 (p=0.0007)||33%+/-9% (p=0.0014)|
TauRx adds that an initial topline analysis of data from a second phase III trial, TRx-237-005, confirmed this finding of efficacy in patients who had received no symptomatic treatment. This study, also over 15 months, enrolled 800 mild AD patients; around 20%, or 160 patients, were not on symptomatic therapy.
TaurRx's chief executive, Claude Wischik, told EP Vantage that randomisation in the TRx-237-015 study had been balanced for geography, severity and use of symptomatic therapy, and thus each of these was tested as part of the prespecified secondary analysis.
There was no statistically significant difference in the first two but a clear one in the use of AD therapy, he said. Given this finding, the first thing that was done was to look at the decline in the control arm, but this was found to match the control arms of similar phase III studies with solanezumab and bapineuzumab.
Mr Wischik says the full data will soon be published in a peer-reviewed publication, which will allow a fuller critique.
Importantly, TauRx completed the analysis of TRx-237-015 before the database lock on the second study, and thus could legitimately change the latter’s primary endpoint to look at only those not on symptomatic therapy. Thus, Mr Wischik expects to report an unambiguously positive result for the second study at a future scientific meeting.
He says that since the first result TauRx has pulled out all the stops to find and validate a mechanistic hypothesis of why taking symptomatic treatments might interfere with the anti-Tau effect of LMTX. Animal studies have so far ruled out the most obvious possibilities of drug interactions, and the company is now looking at induction of transporters, which could eliminate one or other drug from the brain.
It is not the first time the company has had to do this. A phase II study run in 2008 with a precursor compound known as TRx-0014, or methylene blue, produced a bell-shape dose response that the company had to work hard to explain. Later it switched compounds anyway to LMTX, which had improved pharmacokinetics.
Mr Wischik says the company’s advisors suggest that it could file on the basis of the yet-to-be-reported TRx-237-005 with supportive data from TRx-237-015. But quite how regulators might treat results from what by AD standards would be two very small phase III studies is another matter; a request for a large confirmatory phase III would mean years more work.
However, further complicating matters, TauRx has conducted a third phase III study in the rarer behavioural-variant frontotemporal dementia (bvFTD), which affects younger patients than AD – usually those in their 50s – and causes an inability to modulate behaviour.
It had intended to present results of this at the AAIC, but withdrew at the last minute because of a pending patent application on a new discovery. It now plans to present the results at an FTD conference in late August. If this is positive then perhaps TauRx will file in this indication as well. Mr Wischik said the new patents covered a different observation found in bv-FTD to that seen in AD.
TauRx remains unconventional on many grounds, not least its funding, which has mostly come from Far East investors such as the Malaysian property/gaming conglomerate Genting Group. The company closed the final tranche of a previously announced $135m funding round last week, suggesting these investors continue to provide support in the light of the recent events. There might not be too long a wait to see if they are correct to do so.
|Mild to moderate AD||TRx-237-015||NCT01689246|
|Behavioural variant frontotemporal dementia||-||NCT01626378|