I know it's not new but some of it's new to me. I was not present for the presentation. It was just uploaded to the journal site. I had not seen some of the break downs and others here may not have seen it. So, act as cool as you want. There it is for those not at the conference last Nov.
Actually it looks to be in the range of 15-19 months (immature but likely closer to 19 than 15).
"METEOR: Results from the Randomized Phase 3 pdf "
it should be top result.
There is a poster presentation PDF apparently uploaded 3 days ago at kidney-cancer-journal tagged as meteor_poster and it shows the kaplan-meler projection of OS of over 18 months and everolimus around 14.5 months. I'll try to post better search term in reply to this.
leerink partners is in the business of investment banking and asset management. It's very highly unlikely they opened up any new data. Think of it as a presentation to their investors and potential future investors.
Great for BMY. Hats off to a new MOA. It means science is progressing.
It's a mono-clonal antibody. Do you think development analong the antibody lines will get cheaper? I don't think so.
Anything that is comparable is going to be favored if it's cheaper and small molecules are cheaper. I think Cabo is on the range of $4k/month cheaper or $48K/ year cheaper than Nivo; that does not include initial treatment costs.
Antibodies will survive only if they can be heads and shoulder above cheaper treatments if not cures. Then they go to last in line.
Yes, they can be cheap to make but they're only going to get more complex and specialized.
Good for BMY for making a better drug and a new class of drug. Good for EXEL for improving known drugs based on new data.
Unfortunately, the only way to get rid of cancer is cut out the offending piece before it spreads and that's not always possible which means that combon treatements will the the key going forward since one drug can never to everything. There is room for a lot of winners.
This is the trial to watch in phase 1:
Safety and Pharmacology of SNX-5422 Plus Everolimus in Subjects with Neuroendocrine Tumors
If this looks good, cabo should follow since it whipped everoliums
I agree that one has to be careful but data is starting to be generated from human trials so literature should start giving us some data to make more educated bets on investments in the future.
I haven't looked yet but do you know of data for AUY922, preferably from the P2 trial.
Yep. There is a paper a few years old now that looked at VEGF and MET inhibition in combination with HSP90 inhibition whih showed promise in models (up to xenographs if I recall properly). This was an good indication that the MET pathway was upregulated downstream from the TKI MET inhibition, at least over time.
Shutting it down more completely may help and on the other hand, it could be worse as when a drug blocking VEGFRs 1-3 with high potentency was too much.
However, that was multiple modes of inhibition which combines toxcitiy. Since Cabo targets VEGF and MET all in one, that lowers the bar.
Development of Heat Shock Protein (Hsp90) Inhibitors to Combat
Resistance to Tyrosine Kinase Inhibitors through Hsp90-Kinase Interactions
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01106 • Publication Date (Web): 04 Feb 2016
I think it would be a good read for those here that have been following the science. As an investor, as well, it tells us where the next investment might be.
It doesn't mention Cabozantinib but shows that HSP90 inhibitors block MET as well which works to overcome resistance to TKIs
"A study of the molecular mechanism revealed that the growth inhibition of  was associated with the degradation of EGFR, MET and their downstream signaling intermediates."
I know others will agree but I take it as the FDA wanting the data. We'll get the data at the end of the secondary endpoint. (planed 36 months, and primary was 17 months). The FDA probably wanted to make sure that they weren't gong to approve something that kills people faster despite having good PFS as seen in the past. It's called an interim analysis for a reason.
I hope it's 8 months. No wait, I hope it outright cured some of them...
The point it that they are words and nothing more. When investing in biotech, it's important to understand what they can represent. However, I'm a paranoid realist by hoping for the best but looking and fearing the worst.
Statistically significant only means the error bars of the test arm don't overlap the error bars of the control arm.
For "clinically meaningful", see my response to ernie; but it likely is a response to "MCID" which refers to issue the patient finds important such as reduce pain that is unmeasurable but qualitative and can refer to quality of life
There are literally papers written about the wishy washy statement "clinically meaningful" and "clinically meaningful" is often used to denote quality of life improvements (HR-QOL).
The use of "clinically meanngful" most likely refers to "MCID" which are issues the patient finds important and is unlikely to refer to OS.
My fear is that it's only a little OS improvement. Let's say 2 weeks. The implication would be another survival pathway is upregulated giving a poor increase in OS.
I don't think that it would prevent approval as along as OS was increase because it's improving the quality of life without more harm.
It's entirely possible for the cMet pathway to be upregulated downtream from where the TKI is blocking it and usurps the affect of the drug. Then combination treatment with HSP90 might be prove useful
If it's a combo VEGF, MET, mTor or HSP treament that's needed, Exelixis may be in a good spot; however, it would have been better to be chasing the data via research this whole time and developing the next gen drug.
They said "meaningful" and not "significant" which I'm sure that they cannot say for legal reasons.
So, one might assume without more information that stat. sig. has not been reached. However, they did say "increase" which means it's at least as good as the control arm.
"Why is the bar set so high?"
Simple answer: Error bars.
Extended answer: The FDA needs statistical significance. If the error bars of the test result and the control overlap, the result is not statistically significant.
So, to make it statistically significant, the confidence levels need to be reduced in both sets until there is no overlap of the error bars.
So, if they give a value and have not reached statistical significance, it can tell you something within the above paradigm.
Hope that helps.