search "brain chancery" for Chad Peacocks blog. He is on trial and at nearly 17 mos PFS. That is about the same as the median PFS in phase I. Very good news for him!
Also from the few people I have found on trial the median OS (for both arms since it is blinded) is now at 18 months. Interestingly, according to my models and simulations, at about an 18 mo median OS, the 64 events would occur this week. I kind of view 18 as a reasonable estimate of what the control arm might look like. Of course if the median was at 18 we would have seen the interim analysis much earlier also.
Doesn't mean anything. He might as well be in the placebo group. Since he's young, it's normal for him to see better results than most people with brain cancer. Remember when Dendreon scammed FDA by using a bunch of young people in the experiment group?
All of my code is publicly available. Search github for Imuc-montecarlo.
I use interpolated survival curves from the literature and then do a time stretch to modify the curves for different median survivals. I used the enrollment numbers taken from the charts from the IMUC presentations and randomize the enrollments uniformly throughout the months. I then randomize the enrolled patients for HLA1/2+ and then to the trial arms. For each enrolled person I randomize their survival according to the survival distribution of the arm. Basically I simulate the drug trial for the given assumptions and then see when 64 events occur. I repeat this trial simulation thousands of times. This provides me with a distribution of when the 64 events would occur given the assumptions I have made.
The actual survival curves are unknown but the time-stretching seems to fit different data to a reasonable degree.
There are some others on this board like phishes and disco that have some models also, and they get similar but slightly different numbers. I don't think they are doing monte-carlo simulations but rather they are using the mean values for all the parameters (I think).
"at about an 18 mo median OS, the 64 events would occur this week."
This is in line with the company projection. They did say 64 the end of the 3rd quarter. Their estimate of 4 months to read out is way out of line. I read enough bio boards to know 4-6 weeks to issue report, another week [4 days] for the company to release results.
Thnks for your contributions to the board. Appreciated.
I read the news a short while ago - : ) !
With no results posted for nearly 5 days, I was bummed. Kind of depressed actually. Thought maybe things went the other way.
Excellent news. He deserves every bit of it.