Avid Bioservices Inc (CDMO)

[nuke661]

ex,

Thanks for the heads up on the math. However I don't think my mistake changes the facts; I just presented/described how to get them incorrectly.

When I stated the percentages for long term survival I said:

From the Abraxane drug insertAbraxane drug insert the number of long term survivors was 331 out of 431 total, ~23 %. The gemcitabine control group 359 out of 430, ~17 %.

The 23 and 17 percent numbers come directly from the Abraxane drug insert so these are Celgine's numbers not mine. I was just trying to simplify things for PPHMVERYLONG so he wouldn't have to put too much effort in understanding the information.

I should have stated that the ~23% number came from 1 minus 331 out of 431 total. It would have been simpler to just write 1-331/431=~23%. My bad.

Celgine specifically stated in the overall survival section of Table 13 that the number of deaths for the Abraxane group was 333. The MOS numbers I quoted were also from this same section of the table. Therefore, I can only interpret this to mean that the MOS numbers were encompassing of a death population of 333 (way more than needed to determine MOS I believe) but also that the final data for the trial, and therefore this is the only data that can be used for determination of long term survivors must have used overall deaths of 333 out of 431 for the duration of the trial (~4 years as I stated based on the date info on the clinicaltrials.gov site for which I also provided the link to). The survival curve provided in the insert implies the long term survival is much worse than what the NIC says to expect therefore, I again believe I'm not reading the survival graph correctly. Maybe you could help me with this inconsistency, it would be greatly appreciated.

Therefore, I can only assume these are the numbers to be used inclusive of any censoring events. I may be wrong on the inclusive of the censoring but the facts remain that of the 431 patients that went into the Abraxane trial group with operable stage 1 and 2 pancreatic cancer (operable was also an inclusion criteria in the Abraxane trial) they should have an expectation, based on NCI information, to have at least 18% of them live to 5 years. I would think that the NCI information is inclusive of an entire pancreatic cancer population, not just stage 1 and 2 patients accepted for a clinical trial. I believe what I'm hearing you say is that Celgine will most likely beat those 5 year numbers if the censoring is accounted for. However, I don't see that in the survival curve and if they did then why didn't they promote this information in trial data in the inserts or have it expressed it in the survival curve?

Please correct with details where I'm going wrong on this as I certainly want to work with the facts.

Also, it was not my intent to "diss" Abraxane but only to state that it wasn't all that great of an improvement (at least to me it wasn't), especially when considering the historical information provided by the NCI website. OF COURSE any improvement is better than none something that has been historically lacking in acknowledgement from a lot of posters in regards to the entire history of Bavituximab's cancer trials such as breast and lung cancer trials that I've been witness to since 2004.

It should also be remembered that my posting was in response to PPHMVERYLONG stating Celgine succeeded where Peregrine's Bavituximab failed. My major intent was to impress upon him and others how wrong it was to compare drugs that treated way different patient populations. Comparatively speaking, in the pancreatic cancer world, Abraxane did go after the lowest hanging fruit because they have the money and resources to conduct the large trial necessary to support the p value needed for such small percentage increases achieved, in absolute terms. As you stated 10 % vs 6% is better than no percent increase but not the increases one wants to see when considering that this is the best class of pancreatic patients that one is going to get for any trial.

I could also say that your characterization of 10% being almost double of 6% is overstated by at least 30%. I'm not being argumentative with you as I believe you provide good information most of the time. NOBODY gets it right all of the time including both of us. Thanks for your input ex.