Replies to post #128000 on Biotech Values

[turtlepower]

That's a good catch. I hadn't noticed that. That does raise questions about the trial, at least to me. It looks like perhaps they may have excluded most of the 2nd line patients who may have enrolled, from the final data analysis?? Also CRC isn't that rare so being able to enroll only 38 patients in a span of 5 years is somewhat surprising. Especially for a treatment that appears to have had substantial benefit.

[mcbio]

KERX AEZS - Something kinda odd went on with the ph ii trial, at least with regards to what shows up on clinicaltrials.org

Do you mean clinicaltrials.gov here? I believe I've seen other people refer to a .org site before but that address doesn't come up for me and I have always assumed that clinicaltrials.gov is the official site that people reference when checking in on the status of clinical trials. Also, as people have noted before on here, there are often errors with how the trials are presented on this site (e.g. listed as active when they are actually completed, timelines for completion not being accurate, etc.)

1) The trial actually closed enrollment after enrolling multiple cancer types and allowing co-administered drugs of many types for about 2 years. (6 cancer types and thus 6 potential co-drugs)

2) AEZS Investor status gives status on PR/CR/ORR in Aug 2009 for 35 randomized patients per +/- cap.

3) In Sept 2009 it re-opens the trial to "enrollment by invitation" of only CRC patients

But the recently published data doesn't meaningfully expand the number of patients - what happened to the invited patients, or was it only 3 invitees (the latest paper, pointed to by genisi, had 38 patients)?

PS No discussion that I can find is ever made by the companies of the efficacy in the rest of the randomized ph ii (300 patients). So it can be presumed that they were all a complete bust. BUT, on the bright side, even if you adjust the p values for all of those failures the p values are still pretty good (especially in CRC PFS).

Is it possible that the Phase 2 trials in these other cancer patient populations are simply ongoing? AEZS' most recent investor presentation (http://www.aezsinc.com/pdfdyn/Presentation_%20September_2011.pdf ) lists perifosine in Phase 3 for both CRC and MM but also notes Phase 1 and 2 studies ongoing in several other tumor types (CLL, Renal Cancer, and others).

[jq1234]

1) The trial actually closed enrollment after enrolling multiple cancer types and allowing co-administered drugs of many types for about 2 years. (6 cancer types and thus 6 potential co-drugs)

It was exploratory trial, trying different drug combinations for different types of cancer.

2) AEZS Investor status gives status on PR/CR/ORR in Aug 2009 for 35 randomized patients per +/- cap.

3) In Sept 2009 it re-opens the trial to "enrollment by invitation" of only CRC patients

But the recently published data doesn't meaningfully expand the number of patients - what happened to the invited patients, or was it only 3 invitees (the latest paper, pointed to by genisi, had 38 patients)?

You can't depend on Clinicaltrial.gov for accurate timeline. More often than not, update time lag is signficant. Updated in Sept 2009 doesn't mean the evet actually happened in Sept 2009.

AEZS's 35 patients are the same as 38 patients in final analysis, the difference AEZS didn't include 3 patients couldn't be assessed for response.

I think the more important question is how number of 38 patients was decided, and what assumption, ie power calculation etc was made. These are important information companies should disclose.

Perifosine in phase I with different dosing regimens as single agent, found no response in CRC. Thus this could be the drug finally found the right combination for right disease setting, or the drug worked in a trial for patients that have certain hidden characteristics/inbalance that cannot be repeated in large trial. I don't think anyone knows at this point.

[genisi]

But the recently published data doesn't meaningfully expand the number of patients - what happened to the invited patients, or was it only 3 invitees (the latest paper, pointed to by genisi, had 38 patients)?

The 3 extra patients in the JCO paper (n=38) are 3 patients who dropped out from the trial and were excluded from the analysis presented at the ASCO abs. (n=35).

[NP1986]

KERX/AEZS Phase II trial

The phase II trial enrolled patients with other solid tumors as well. The JCO paper reports that an unplanned interim analysis was carried out because of limited resources. At the time, they had 25 CRC patients and I'm assuming this was the only group where they saw a benefit. So they then enrolled an additional 13 CRC patients to bring the final number to 38.

Three of the patients were not evaluable for response, so I think they are excluded from the TTP analyses. However, the OS analysis reported in JCO includes all 38 patients, IIRC.

There's no mention of controlling for multiple comparisons (since this was essentially like conducting a subgroup analysis in a larger trial), but as you've mentioned, the p-values are quite low.