AMXT 1501 plus DFMO

[davidfinn]

https://www.ccia.org.au/blog/researc...ildhood-cancer

I am new to watching this kind of announcements, is this something to be excited about, or just hopeful until further testings is accomplished.

[musella]

I helped fund that study. It is exciting but way too early and I am not happy how they announce findings like this.. it builds up hopes. Many times in the past similiar things happen - something works great in the lab and doesn't work when tried in humans.. so it is something to watch, and we need to try to speed up the pace of the research

[davidfinn]

Glad to hear you helped in the research. Great job.

For someone just getting up to speed watching research, what is the next step after a finding like this?

[musella]

The next step is human trials. Ideally a phase 0 trial where they give the drugs before surgery then they take a sample of the tumor and see how much drug got in.. this lets you know if it can get to the tumor.. then once they know that, a phase 1 trial to figure out the best dosages and make sure it is safe, then a phase 2 trial to show how good it works. If it is really good they may be able to get fda approval. if good but not good enough then goes on to phase 3 trial then approval. (Or failure)

[davidfinn]

How hard is it to get Stage 0 approval? Does the FDA need to approve that or some other organization?

Sorry for the basic questions, I am new to all of this.

[musella]

The FDA has to approve it and the hospitals' Internet Review Board has to approve it. It is pretty hard to get approved. You have to prove to the FDA and the IRB that the drug is safe, and show the preclinical work, which could by itself coust a few million dollars and take a year or so.

[davidfinn]

Rather than start a new thread, I'll just ask if this is another one of these type of announcements. Great at this stage, but too early to tell if it will do well in patients?

https://www.ccia.org.au/blog/austral...ildhood-cancer

[musella]

I like that - and the Musella Foundation helped fund it through the DIPG collaborative. However it is way too early. Many things that looked great in the lab did not work on people. Especially DIPG. Obviously. I will be keeping an eye on it. Human trials will start soon.

[davidfinn]

Outside of going back and getting a MD. Is there anything I should be looking for in this stage research to tell me if they are really positive or just hopeful?

[musella]

I am probably getting old and jaded but nothing excites me that much at that stage of research any more. I have been burned too many times over the last 29 years I have been studyign this. Getting my hopes up and they fail in human trials. I know that any one of them could be that exception, but there really is no way to tell until they do human trials. Tocagen is a perfect example. They were able to cure all rats. They were able to cure brain tumors in most dogs (I forget the numbers but it was like 8 out of 10 dogs with brain tumors were cured). Early phase trials were exceedingly good. It failed phase 3 trials.
The only thing we can do is keep trying. Anything that looks good deserves to at least be tried. The more at bats we get the more chances of a home run. I realy think that the cure is going to be in combinations of treatments that we may already have. I am working on that angle with my "A Patient-Centric Platform Trial for Precision Oncology (XCELSIOR)" https://clinicaltrials.gov/ct2/show/NCT03793088 project.

[davidfinn]

Interesting, I know a person involved with Tocagen. Wasn't paying close attention because of Vivienne's disease, so I didn't realize it looked so good in Stage 1 and 2.

The idea of not knowing what is going to work, reminds me of some work in the investment fields. Starting with the idea that the future is unknowable, the correct choice is to distribute your investments across a wide variety. You further refine it by avoiding the obvious failures. It turns out it is much easier to pick the failures than the winners.

[musella]

Tocagen might not be completely dead just yet. By using what was learned in the trial that failed I think we could construct a trial that succeeds.
I think they just didn't have the money to continue adter they spent so much money on a failure.
This happens to a lot of trials. There is no flexibility. I saw bad things as they were happening but it was too late to make changes.

[davidfinn]

That is sad. Seems like their is no way to adapt trials once data starts coming in.

On another front, what do you think of this?

https://github.com/facebookresearch/CPA

[musella]

I never saw that before. Looks interesting. I will check it out. Do you know anything about it or them?

[davidfinn]

All I know is what I've read. It is the outgrowth of a collaboration between Facebook's AI group and Dr. Theis from Germany. They have created an open source AI model which combines the interpretability of linear models with the flexibility of deep-learning approaches for single-cell response modeling. CPA encodes and learns transcriptional drug response across different cell types, doses, and drug combinations.

Seems like if you have the single cell RNA of a tumor you can teach the AI to simulate millions of drug combinations and dosage levels to speculate at what might be most effective.


Here is Facebook's announcement and a link to the official paper: https://ai.facebook.com/blog/ai-pred...seases-faster/

Hopefully someone will pick this up and run with it for DIPG. If not I'll have to figure out how to program Python and work on it myself.