Cel Sci Corporation (CVM)

[sushifishman]

FROM FOSCO ON YAHOO

Some of you know me. In the company I founded, we are dealing daily with millions of data from our customers. We are not dealing with medical data from clinical trials though, only with consumer data. So, I am lacking a bit this knowledge.

Still, my first reaction when thinking about data lock, is that a statistician could in a very few days / weeks once he gets the dataset, determine whether the primary endpoint is successful or not. I do think this is still true, eg I believe the guys who have the data and are due to a complete confidentiality should be in the known by now. This is an important factor to take into consideration as well : we have two populations who have seen the data and know whether it is good or not : the IDMC and the independent statistician. While I believe in confidentiality agreements and work ethics, I do believe as well that the perspective and profound joy of dealing with a successful trial in the making cannot be let completely unnoticed.

However, I understand as well that this kind of analysis defined in the initial SAP and required by the FDA has to be much more thorough than a simple assessment of success, and should take more time : that’s frustrating.
John A. from the board, who is leading an investment firm and is quite knowledgeable, said that he believes April could make it for the data.

As an example of study, we have this excellent document Sushi posted that I would like to comment.

https://www.dropbox.com/s/cnku7it96z42mg5/Baas%20et%20al.%20Meso.%20CM%20743.%20IASLC%20WCLC%202020.%20Oral.pdf?dl=0

I do like very much this document : this is perfect synthesis of a successful trial, well presented, concise and clear.

First, it shows that the sponsor did wait until they have a complete analysis to disclose it to the public. Have a look at it : there are many pages and many analysis on different axis.

The part of interest in addition to showing what is a complete analysis, is that it indicates that the database lock was achieved as of April 2020 and the publication of the positive (p<0.05) interim analysis plus all the detailed study has been done by August 2020, so 4 full months after.

It took them 4 months from Datalock to publish the Analysis. It seems a lot but then the FDA approved it in October 2020 !
(see this link https://www.fda.gov/drugs/drug-approvals-and-databases/fda-approves-nivolumab-and-ipilimumab-unresectable-malignant-pleural-mesothelioma )

Analysis took long, (4 months) after data lock but then the FDA approved right away. Worth noting as well that the sponsor wanted it to be approved as of Standard of Care, hence they have allowed themselves enough time and no shortcuts to produce a perfect analysis. It saves them time with the FDA to produce a complete analysis that was so overwhelmingly positive and well done that the FDA approved a few months later.

So, getting back to Geert statements : He said that he wanted a complete analysis because he wants to prove that MK is suitable for Standard of Care.

The shown example study is consistent with what CVM is saying, complete analysis is essential and takes time and more importantly saves time when well done. Datalock was announced by 7th of December, let’s wait a few months more.

Fosco