InvestorsHub Logo
Post# of 251706
Next 10
Followers 3
Posts 138
Boards Moderated 0
Alias Born 07/03/2007

Re: jq1234 post# 183962

Sunday, 11/23/2014 10:23:45 AM

Sunday, November 23, 2014 10:23:45 AM

Post# of 251706
CLDX - With the latest ReACT data release, I was curious about the phase III ACT IV trial design. As I searched around, I found it strange that the company is coy on details of the powering assumptions, e.g. they are more forthcoming on the CDX-011 METRIC trial. Anyways for those interested, below is a summary of what I could find.

* Enrollment target = 374 for primary endpoint of OS in GTR patients, 700 enrolled overall
* Randomization 1:1
* Interim looks at 50% and 75% info fraction for futility and efficacy
* Stratification factors – MGMT status, RPA class, geographical region

The most specific info I could find on powering is a statement from 2Q14 earnings call "So, ultimately the hazard ratio of up to 0.79 would be able to detect a difference of three months at the median but it could also be positive if that median difference is even less [my comment-latter phrase in context of co's expectation of late KM curve tail separation]." Assuming a HR of .79 meets significance at the two sided .05 level, then the target is about 275 events, give or take. It also corresponds to a control arm MST of 11-12 months, which is less than the concurrent arm in the Stupp or control arms in the Bev 1st line trials even when accounting for randomization occurring after cCRT in ACT IV.

Some things I've been considering:
1) If EGFRvIII expression singularly has no prognostic significance, then ACT IV appears to underestimate control arm OS relative to contemporary trials.
2) The trial is running longer than co's earlier expectations, perhaps due in part to #1.
3) Other antigen-immune stimulant based cancer vaccines like Provenge, Stimuvax, GSK's MAGE-A3 have shown modest to weak effect on OS across diseases with HR's no better than .78. Is there anything about Rindo that clearly differentiates it from them?
4) The ReACT trial control arm in the Bev naive cohort may be be underperforming. If it were to have performed more like the Bev BRAIN trial, rough estimate of HR would be ~.8, in line with #3.

Lots of assumptions on my part and a long time until this plays out. But if the above has validity, then ACT IV could be underpowered.

http://seekingalpha.com/article/2392065-celldex-therapeutics-cldx-ceo-anthony-marucci-on-q2-2014-results-earnings-call-transcript






Join the InvestorsHub Community

Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.