g'luck but there is nothing ahead in the way of substantive news that will make it go lower and quite a few events that may make it go higher. You are in at the bottom here.
Absolutely. That is the point. At some point survivors can no longer affect the median OS but if the survivors live a much longer time than control, that also will be important to the FDA as well. But the more important point is that this is a P2. Its sole purpose was to elicit a strong clinical signal that gets them to a p3. Stat sig is way overblown in this context. IMUC could have waited which might have been better for the stock price, but not the company or patients. The goal is to get to a p3 as quickly and cheaply as possible. Waiting for data to mature while burning 3mm a month is bad business. They got the signal they needed to proceed to a p3. Everything from here on in is gravy for the end of p2 meeting with the FDA. My wild #$%$ guess is that the p3 will need about 350 patients at 100k per patient. Expect one more round of dilution late 2014 or early 2015.
Pretty close -- stat sig is always a function of BOTH variables so you can't say (at least on this data set) whether the sample or the effect was "too small" Together they did not reach stat sig (yet). It may be that just more data points (assuming similar results) will be enough to push the results into stat sig (which I think is the case although it may not be in the p2 trial) I will say this much, a 2 or 3 month improvement in OS in this patient group in larger trial will prove stat sig AND clinical significance. You need both in the context of FDA.
It's not that simple which is what most of the loud mouthed donkeys don't understand. Stat sig is a function 2 pieces of data --the sample size and the magnitude of the effect being measured. Magnitude is further a function of the control or baseline data compared to the treatment arm data. So trying to pick a month is like trying to clap with one hand--Doesn't work. It may, in fact, be the case that the mere addition of additional data points (larger sample) will be sufficient to translate the results into stat sig. Remember the OS data is based on only 67 events whereas the PFS is based on 103. PFS was stat sig with 2 or 3 month improvement. Wonder what the OS data will look like when there are 103 data points on it???
Last but not least if there is ANY objective difference between two parameters, in any setting, you can always prove stat sig with a large enough sample. But proving stat sig in the clinical trial context does NOT mean it has clinical significance. So obsessing about stat sig without knowing more is for more-ons Take this to the bank if 107 shows a 3 month OS benefit in a large trial, with their benign safety profile, approval is a slam dunk
I'm rarely wrong and certainly not wrong in this instance donkey. You just make an #$%$ of yourself when you post duuumchit like this. OS is 2-3 months greater in the treatment arm than control. Clearly stated in the PR. You are also clueless as the p values, the difference between clinical signifcance and statistical significance, and the difference between p2 and p3 trials. You should keep yourcockhole shut to avoid further humiliation.
huh? Spike some more Krokodil cupcake. you have no idea what you are talking about
Not angry - just derisive. Still lurking looking for an opportunity to buy some puts on the outside chance that the mets comes in with misleading results. as far as your os and pfs story, I see you are buying into the company censoring the data of 30 patients because they did too well. I don't remember seeing anything like this in the last decade (or ever) Don't you wonder just a little what the data would look like if they actually told you the truth.
PFS unlikely to move much (if at all) since it is based on 103 data points and we are over a year on study with remaining pts, the data is pretty mature---OS is another story. Only based on 67 events and is currently only an estimate. In addition you still have the "long tail" to consider. Median is very important but if 107 exhibits a long tail that would be very significant as well.
No. 1/3 were given placebo and 2/3 ICT-107- enrollment closed sept 2012 so the 57 still alive have all been on study for more than a year, 21 of whom have not even progressed yet. OS number need not necessarily move in IMUC's favor but SHOULD move right on KM and many folks don't realize that, assuming data comes in like the reported data the P value will improve with the mere addition of more data points (larger sample). Hear this! and remember where you read it. IMUC may reach stat sig as OS data matures. YOU have been given a rare opportunity to get on board at fire sale prices. But IMUC will still need a p3- this is fact.
Can you read donkey? Short answer- people DID live longer on 107 2 months in the ITT chart and 3 months in the per protocol group. Explaining anything else is a waste of my time (and yours since you wouldn't understand it anyway)