David Sobek

David Sobek
Contributor since: 2012
Company: Sobek Analytics
Are you really critiquing the company because they have not cured cancer? If curing cancer were the hurdle then nothing in oncology would be worth an investment. Vintafolide is a significant improvement over currently available chemotherapies for PROC.
If you are looking for cancer cures before investing in an oncology company then you will be disappointed.
Yes. Tedizolid is active against linezolid resistance strains. I wrote on that while back. You can check it out here.
Thanks. I think you could make a base case assumption of $400-$500M in peak sales and that would leave room for upside as well. Even with those sales, TSRX is undervalued.
FDA generally requires two trials for approval. Antimicrobial stewardship is good in that they need new antibiotics. The antibiotics market itself is very large.
Right, which is why I look at a range of different studies to see which are most similar to what PICASSO III looks like. Obviously no apples to apples but if the vast bulk are saying less than (or around) 5 months, then I'll go with that as my base case. Plus, there is a margin of error to where even a 5.5 month median PFS for dox would be fine, although that would be getting too close for comfort.
I'm sorry but I know the bear case about the stabilization of pali not being good but it seems like a tempest in a teapot.
Even excluding the PICASSO II the vast experience with dox in STS is a median PFS of around 5 months. There are certainly exceptions but it is what it is. You might want to not believe it but a dox median PFS significantly over 5, let alone six is rare. Not impossible but rare. I'd rather invest on most likely scenarios not least likely.
IFOS is an old drug but your reference is really meaningless. IPM had been prepared before but not stabilized. You seem to know enough about there area to also know that.
Keep in mind that 2 of the 3 studies I look at are double blind, so you first point is just not accurate.
Yes, if you simply cherry pick the Maurel study you can get to over 6 months median PFS but it is not an apple to apple comparison and it also ignores all the other studies that have median PFS much lower than that, so I would prefer not to cherry pick a study to support a view but look at the entirety of the data.
I think the switch to TRIS was because it was more stable but I'd have to go back and double check that.
FRR is a rule of thumb and not a law.
Pali should be better than IFOS. It is not only the side effects but IFOS is metabolized in a more variable manner than pali. So you get more consistency in the dose with pali than you would see with IFOS.
My impression of the CLSN analyses was that they did not really present a good model of the control group (as there was really not enough data to make that kind of analysis). Plus, the bigger problem with CLSN was that its mechanism of action was not targeting the cause of progression. That is not the case with ZIOP.
Glad to hear it worked out well. The problem is not when Zyvox is used within the guidelines (under 28 days), which is what you seemed to experience. The problem is long term use (months instead of weeks).
You bring up and interesting point and one I thought about noting in the article. Vancomycin has the largest share of the prescription market but by sales Zyvox is king. What we don't know is how TZD would do in a market with a generic Zyvox. I think it will be fine and still see a slow and steady rise of revnue but I've not gathered enough data yet to provide any meat onto that statement. Hopefully soon.
Thanks for the comment. I'm actually working now on the questions you are asking, i.e. why are these drugs having different launches and what can we glean from that in terms of TZD.
Thanks for the comment and what you say is accurate but misses the point of my analysis. There are a lot of good analysis that uses the method you describe to get at an estimate. While those are good, I think there is something to learn about antibiotic launches in general (even those less similar to TZD). Despite the drugs that I look like being quite different, the model still explains 97%+ of the variation. That is pretty good. In the model fit was not nearly as good, then it would point to these drugs being too different.
So this was not meant to be THE way to look at TZD but a different way. When looking at a drug or company I find it useful to use a number of different methods to value it. If they converge around the same value, then you get more confidence. Relying on one way I find risks tunnel vision.
So that is a long winded way of saying you are right to a certain extent but looking at a broader set of drugs is adding useful information. Best.
Nice buy. I think you buy ECYT on dips, sell covered calls on rips, and hold shares long term.
Thanks. I think the collaboration speaks to the broad activity of Cabo, which simply reinforces my view that it will work in mCRPC. We know it works in MTC. We shall see more evidence of activity in mCRPC at ASCO. With the NCI-CTEP we should see activity in additional cancers as well.
I think over the long-term you are right that Cabo and other indications will add value. I've not run the numbers but I think it would certainly support a larger valuation. That being said, I think sentiment towards Cabo in mCRPC is going to drive the share price.
Right, so the US filing would not be until after data in 2014 and not after the EU filing. Also, that application would be for accelerated approval and then final approval would be based on the OS data in 2015.
Yes, the May calls will expire after the abstracts are released. Often the abstracts will have some of the pertinent information, although clearly you get more details with the presentation at the conference.
You are right that pricing matters but the 26% market share estimate that I use gives it $500M+ in peak sales. So rather than think in terms of market share, the dream case is based on $500M+ in peak sales. The $4+ fair value is based off of $250M+ in peak sales.
It was tough to figure out the exact split, so I might be wrong on the 50/50 divide. There was a line in the 10-Q that seemed to indicate that profits and losses would be "equally divided." So I was basing the split on that line.
Good question. When I changed the model to $36 million in 2016 (which is about 2% market share), I also then had it grow to 10% market share in 2020. Under those assumptions discounted cash flow falls to $1.41.
FYI, the FR++ was not retrospective. It was prospectively designed to look at FR-, FR+, and FR++ groups. The company always noted that FR++ group was where they expected the benefit. They included the other groups in the study because FDA requested it. The FDA wanted to see if the benefit of the drug increased as the patients went from FR- to FR+ to FR++.
Good summary. I would also note that anyone interested in CELG to check out the retail shareholder effort. While we all believe in the CELG story, it would be that much greater if the management were willing to pursue more shareholer friendly actions.
See our full arguments at:
I think you need to do some more diligence in terms of the earnings of CELG. The trailing twelve months is just under $2/share and your chart has them at well under $1/share. The current estimate for this year is $2.82/share again well above the number you use.
So of course CELG looks overvalued when you are using earnings numbers that are 1/4 of the actual numbers.