Anavex has seen a huge sell-off since last Friday.

The sell-off has come even as data released from the Phase 2a trial was satisfactory.

I interviewed Dr. Stephen Macfarlane, the trial's investigator, about the data.

Anavex Life Sciences (NASDAQ:AVXL) shares are down more than 30% today. The stock has seen an unprecedented drop since Friday. The reason for the sell-off is not clear. Based on Saturday's data, I am still positive about the science behind Anavex 2-73. I also had a chance to interview Dr. Stephen Macfarlane by email, who conducted the Phase 2a study. Dr. Macfarlane, who is the director of Aged Psychiatry at Alfred Health, has commented on the recent data. I will be providing further data to subscribers tomorrow.

AVXL is still one of the most profitable picks for me this year. However, I would recommend readers to consider their personal situation before considering to add to their positions during the current sell-off to bring down their average cost of acquisition. I opened a position in AVXL at $0.53 (pre-split) so despite the current sell-off, I am in green with the shares I am holding now. This is why I am continuing to hold despite the recent sell-off.

Below are my questions to Dr. Macfarlane and his responses.

1. I would love to know what you think of the data in terms of efficacy.

Whilst there is clearly a lot of interest in the cognitive data the study has generated thus far, I think people are losing sight of the fact that this was designed as a study to test safety, tolerability and bioavailability. Part A of the trial, if you examine the study flow chart, was extremely onerous for our participants, who had to attend daily for 12 days to receive study drug and pharmacokinetic sampling procedures that lasted several hours. They then had 12 days drug-free before the process was repeated. When designing the protocol the question was asked "what's in this for patients? They go through this onerous process, but have no chance to continue to receive the drug if they perceive benefit from it during Part A." Part B was thus added to the protocol partly to provide some incentive for volunteers to participate in the trial.
I think it's fair to say that we were not expecting such strong cognitive signals to emerge from a small n study that was not designed to capture statistical significance within the resultant cognitive dataset. To me it is very significant that all those who completed Part A elected to enrol in the 26-week extension, and as the first few patients began to return to the study sites for Part B visits we received the message from patients and carers that they wanted to continue on the study drug for longer, which prompted me to approach the sponsor to extend the study for a further 26 weeks.
To me, it was extremely encouraging to see any sign of a cognitive signal in a five-week study, and the strength of these signals was remarkable. I've been running Alzheimer's trials for over 16 years, and have never seen such a strong cognitive signal in any of the trials I've been involved in (including solanezumab). We've collected a number of verbatim quotes from study participants and their carers, which I'd be happy to provide to you in a deidentified form upon request.

2. Are you satisfied that data presented at 5 weeks is any indication of results after a full-fledged 52-week study?

As above, to see a cognitive signal of such strength after only 5 weeks is truly remarkable. Only time will tell if these signals remain over the full 52 weeks, but participants will be undergoing cognitive evaluations at the 12, 24, 36 and 52-week time points, and we'd expect to report further on this data as it becomes available.

3. Can you explain the various scores in the data, specifically where Anavex 2-73 showed promise, and where it didn't?

Five out of the 6 cogstate domains showed improvement, with 3 of these showing large effect sizes. The 'one-back' test within cogitate showed an effect size of 1.1 (meaning the standard distribution curve of the results was improved by 1.1 standard deviations, a huge effect size. A published meta-analysis of donepezil's benefit on the same test showed an effect size of 0.28. The significance of this particular change is underlined by the fact that 75% of our sample had already been taking a stable dose of donepezil for at least three months, so the improvements on the Cogitate battery (and on all the other cognitive outcome measures, were achieved over and above the gains our participants might already have achieved through being on donepezil. The same is true for the improvements noted on MMSE and ADCS-ADL, and the EEG markers (of which one ERP was statistically significant). I gather much is being made in certain circles that the remainder of the EEG markers, as well as the MMSE and ADCS-ADL scores failed to achieve statistical significance, but in a 5-week study of 32 patients it would almost have been beyond belief if this were to have been shown to be the case. The study was not powered to demonstrate significance, but the response on certain outcome measures was so marked that significance was achieved anyway (at a level of p=0.001 in the case of the Cogitate one-back task.

4. Of all the battery of tests which do you think most powerful for AD, and how did Anavex 2-73 perform in that test?

The One-back test (which measures working memory, a key domain of impairment in AD) within the Cogstate battery is the most difficult test, yet showed the greatest improvement. This outcome is unlikely to have been an artefact of any placebo effect...when placebo effects occur they are typically larger for the easier tests. In addition, whilst placebo effects are common with CNS sedative drugs, they are less likely to be present in tests of drugs used to enhance cognition...individuals cannot anticipate, or 'imagine' what a better score would consist of, and are unable to produce improved results through mere 'wishful thinking.'

5. Can you compare these results with results from other AD trials in similar clinical stage?

A similar outcome in a trial such as this is unprecedented in my experience. I'd cite again the fact size of 1.1 for ANAVEX 2-73 on the Cogstate One-back test, and compare this to the published metaanalysis of donepezil's effect on the same test (0.28). Bear in mind, once again, that 75% of our subjects were already on a stable dose of donepezil for at least 3 months prior to baseline.

6. How would you compare this with Toyama's study? Are there critical differences between the two drugs?

As a clinician, I'm unfamiliar with the Toyama drug to any great extent, though I understand it's a pure sigma-1 receptor drug. ANAVEX 2-73 is different in that it targets both sigma-1 and cholinergic muscarinic receptors. The muscarinic activity is thought to underly the cognitive benefits of ANAVEX 2-73, with the sigma-1 component believed to reduce protein misfolding (including Abeta and Tau) and inflammation.

7. Do you think you will have even better results by adding donepezil to 2-73? Why? Are there plans to do that?

The population pharmacokinetic and pharmacodynamic analysis will reveal that, since we included both donepezil-treated (24/32) and donepezil-naive (8/32) patients in our trial. In other words, 75% of our sample have had ANAVEX 2-73 added to donepezil as part of the study protocol.

Author's note and conclusion - I believe this is a decisive interview and conclusively answers bear on this stock. I have been in this stock since it was a pre-$100mn market cap company. They have a drug that - as I have kept on saying - may fail or may succeed. I have no clear idea whether it will. However, this disease is Alzheimer's, hence there is too much expectation and too much volatility. A misinterpreted trial can wreck havoc on the stock. And shorts probably sense that. Hence the current raging debate and the huge drop in price. However, nobody bashing the stock seems to be covering the science, telling us, scientifically, why the stock should go down because the results were bad. Nobody is able to say that the results were bad - because they were not. Now we have here the actual expert behind the trials telling us these results were astoundingly good. To paraphrase, in his 16 years of trial experience (Dr Macfarlane conducts many trials, including one sponsored by Eli Lilly), he hasn't seen AD results like this.