...The other thread was getting a little long so I'll start another...
bdoglo - I modeled the 45 year old OS curve and had data points ranging from the week of Feb 17th to the week of Jun 9th with 80% of my data points falling between the week of Mar 24th to the week of May 5th – I picked this one because it was one of the longest curves that is still a reasonably feasible outcome that we could see (YES Disco – I know your thoughts on this but think of it as me looking at the upper limit on the range of possibility while you check out the lower limit – BTW, welcome back.) …And gaborrcz – back in Feb I did a “blended” curve of the UCLA 45 and 60 to get the same average age as the P2 trial and it is one of the curves that I keep in my file to reference. Call me overly cautious but I have seen too many trials where the control OS was way beyond what everyone thought that it would be and beyond historical norms - the good thing here is that we have some pretty clean and current data to use for curves and the company has provided us with a pretty good bit of trial details.
Just keep in mind that these numbers are high because they include dendritic vaccines like ICT-107, DC-Vax, and Celldex, as well as other experimental treatments which are not yet approved by the FDA and are not the SOC. I appreciate that you already acknowledged that you understand this. Their numbers are for marketing purposes, not for academic use. The control patients that drop out of our trial will not be eligible for dendritic vaccines, correct? At least I know that if you think you have the placebo, you certainly can't drop out and re-enroll in this trial and be re-randomized. I don't imagine that DC-Vax or Celldex could accept these patients either, or their numbers would be junk.
So one thing that we need to remember is that in most cancer trials, once there is progression the "trial" is off for that person and at that point the patient/doctor can try anything (not another trial but everything else is fair game.) I do not specifically see this listed for 107 but I don't see why this would be any different here - yes WE want info but the patient wants to live so this is why this is allowed - it is the same for the treatment arm and control arm so this is why it doesn't matter in the trial... if the drug worked there will still be a difference in OS.
Yes indeed. Here is the list of treatments at UCLA (from their webpage) - ICT-107 is not spelled out but other dendritic vaccines are on the list, so this can explain the relatively high survival numbers. In our trial the control group receives only SOC treatment so most likely the OS of this group is lower then the UCLA survival numbers.
"What chemotherapy agents does UCLA Neuro-Oncology use to treat patients with a GBM tumor?
Because each patient's treatment plan is unique, therapy normally is dictated by several factors including a person’s age, Karnofsky Score and any previous therapy they have received. Advances in our molecular diagnostics lab is enabling us to better predict what agents will benefit a particular patient group. In general, the following is an overview of agents we used to treat our GBM patients during the past 24 months. This list includes all patients treated at UCLA Neuro-Oncology between 5/28/2011 and 5/28/2013.
• 5FC • Accutane Hoffmann-La Roche • AEE788 Novartis
• AMG-102 • Anti Neoplaston • AQ4N (Banoxantrone)
• AVANDIA (Rosiglitazone Maleate) • Avastin • Avastin (Bevacizumab) Genetech
• BCNU • Bevacizumab • BiCNU Carmustine
• Carboplatin • CC-223 • CC223
• CCI-779 • CCNU • CCNU Lomustine
• Celecoxib (Systemic) • Celldex • Chloroquine
• Cilengitide (EMD 121974) • Cisplatin • CPT -11 (CAMPTOSAR, Irinotecan)
• CPT-11 • Cytoxan • Dasatinib (BMS-354825, Sprycel)
• DC Vax • Dendritic Cell Therapy • Etoposide (Eposin, Etopophos, Vepesid)
• GDC-0449 • Gleevec (imatinib mesylate) • GLIADEL Wafer
• Hydroxychloroquine • Hydroxyurea • IL-13
• IMC-3G3 • Immune Therapy • Iressa (ZD-1839)
• Lapatinib (GW572016) • LY317615 (Enzastaurin) • Methotrexate for Cancer (Systemic)
• Novocure TTF Therapy • ONARTUZUMAB VS PLACEBO • OSI-774
• PCV • Procarbazine • RAD001 Novartis (mTOR inhibitor)
• Rapamycin (Rapamune, Sirolimus) • RMP-7 • RTA 744
• Simvastatin • Sirolimus • Sorafenib
• SU-101 • SU5416 Sugen • Sulfasalazine (Azulfidi
So are you also simulating the trials? You seem to have a range of numbers like I get from the monte-carlo simulations. These numbers here are for no benefit from vaccine using the 45 year old average curve?
Yes bdoglo - this is a simulation model that I made years ago for bio trials, I just plug in the variables and run it several hundred times... and yes these numbers are the output for the 45 year old average curve with no benefit.
bdoglo, if I look at the UCLA curves I see two dataset, one for average age of 45 years old, total of 117 patients and OS survival (I assume this is median overall survival ) of 633 days (or exactly 21 months of median OS)
The other dataset is for average age of 60 years old, 362 patients and OS of 599 days (or 19.8 months).
Once I looked up our average age in this trial but I am confident it is some
where in the mid 50's so I would take the average of the above two dataset, which is about 20.5 month median OS, I think this is how far I would go in the model.
What OS advantage your model would give assuming 20.5 months OS for the control group and a mid May event date for 32 deaths ?
When would we reach 9 months OS assuming 20.5 months for the control group ?