RE:RE:RE:RE:RE:RE:RE:RE:RE:RE:RE:RE:Patient #16, #17 & #18I'm bored so just writing things that I find interesting.....
I think the issue is all about stats and probability. The patient's treated so far are a "sample" from the population of all patients out there. Regulatory bodies need to get enough data so that the sample tells them something meaningful about the true response if approved and given to everyone. If a company gets lucky and a small patient sample looks good, the regulatory statistician would not hear that "CR rate" but rather, crunch the number to get the range of what is likely to be the TRUE CR (95% confidence). There is a wide margin of error if the sample is small.
As of now, TLT reports 5 patients (2PH1 plus 3PH2 Optimized drug and procedure) with a CR rate of 80%, then it means that there is a 95% chance that the true CR rate, when rolled out would be somewhere between 36.2% and the high 90% range. That's where we are now. Not a lot of room.
If the next newsletter report CR in 2 out of the most recent 3 that may be due for data reporting, then that would be 6 CR out of 8 in the future that would be 75% with the true value being somewhere between 40% and high 90% range.
If the next newsletter reports CR in only 1 on the most recent 3, that would mean 62.5% (5/8) and a true value somewhere between 28% and high 90% range.
Since FDA requires 30% CR to be a credible drug then getting 2 out of the next 3 would be really nice.
If we get 3/3 on the next 3 that takes us to 87.5% (7/8) with a true value of 52.5% on up.
While the other, non optimized drug or non optimized procedure, data is a jumble it does seem to show that when you get a CR you maintain it so that would be good too.
If the FDA looks at data as all 25, and doesn't accept the logic of separating out the optimized from the non-optimized (not saying that they will but who knows). Then it would take a 50% CR to mean that the true CR is 30% on up. I don't really understand the most recent newsletter data representation but there is a reference to 9 non responders (90day) and 8NR (180day) so that leaves the chance that the whole data set is flirting with the 50% line. I could see that leading to a request for more data when you get to bigger sample size. But who knows.
From an investor's standpoint, if you know that the optimized procedure and drug can reach 6/8 or better then the true value will come out, just may need to wait a while for the FDA to sign off on it.
Just my thought exercise...IMO