Every investor and his aunty fret over the company not releasing yet the statistical data or “raw data” from PDD (Parkinson’s Disease Dementia) phase 2 Clinical trial. From what the Dr. Missling said even the cohort with 30mg dose responded and that all the targets have been achieved with statistical significance. The last two words require some explanation or rather some discussion. The threshold for statistical significance is the power of study threshold <= .05, that means there is only 5% or less chance that the results are a fluke.
I am not a statistician, but I was taught to think in terms of the smallest sample size for a large population. The number cited was 30 specimen. The PDD Phase 2 had cohorts of roughly 44 patients. I think the reason was twofold, on one hand the savings for the company on the cost of trial, usually about $10-30,000 per subject, and the other was that the there were expectation of strong signal so small cohort would not obscure the efficacy. The rather shocking truth is that $AVXL can not afford a drug with low efficacy and 1000’s of patients to justify approval. Biogen in point. $BIIB
The goal of a statistician is to detect signal, useful information in the sea of random change. That is why we have a placebo cohort to measure the noise. In medical practice outcomes are subject to variation and this creates a background noise against which one has to detect the signal in the cohorts receiving the drug. The larger the sampling population the more clearly the noise and the signal converge to their mean so there is less ambiguity in distinguishing one from another. Converging to the mean implies that with larger sample size approval is more likely even if the therapeutic signal is of lesser strength. On a very superficial level, you get a number with standard deviation, the mean and the spread around the mean. The ultimate goal is to separate the mean so much that standard deviations don’t overlap, and you have a clear signal. People cheering thousands of subjects in trial miss the point. It should rather be a sign that likely the drug is barely to make it through the trial, as it crafting the trial to the drug and not the drug to the trial.
The Phase 2a of Alzheimer’s did achieved statistical significance in few measurements, if my memory serves me well, PDD Phase2 even at this small sample size reached all targets, i.e., all measurements showed presence of a signal of the required strength for approval. Had the sample rate in AD (Alzheimer’s) Phase2a been ~44 patients per cohort the same could have been expected? Time again we, the longs, have another signal that efficacy is strong and undeniable. The periods of time between the old guard AD or PDD drugs and $AVXL drugs proves that SIGMAR1 receptor drug did open a new vistas in CNS therapeutics.
It seems that the projections in my blog were not entirely wishful thinking or extrapolations on unrealistic assumptions, but one of possible outcomes to which the reality of Blarcamesine clinical trials results seems to converge.
But as they say in the world of investing better to be stupid and lucky than wise and wrong.
I would like to thank you guys for all the views, this makes me feel great and useful. If you find this blog useful could you give me a small donation to cover the cost of at least the account with WordPress? Thank you!
Also if you can fill the poll at the end of page I would appreciate.
Make a one-time donation
Make a monthly donation
Make a yearly donation
Choose an amount
Or enter a custom amount
Your contribution is appreciated.
Your contribution is appreciated.
Your contribution is appreciated.DonateDonate monthlyDonate yearly