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.
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.
PLRX - is getting crushed today because they are clearly do not have data analysis integrity. Every presentation uses different endpoints, the data changes without clear explanations, …
This Twitter thread is a good summary:
https://twitter.com/whatprobability/status/1652874063260336130?s=61&t=-DGxXBpsKJ3DTYuLilvNuw
Even for early stage assets it’s important to always have a standard primary endpoint for which you show ITT. And, perhaps, a *few* other exploratory endpoints that are clearly labeled as from a short, prespecified, list of secondaries. Everything after that is tertiary. Sure, the company can track it. But even internally they should take it with a big grain of salt.
Hypothesis: companies that emphasize fancy new, non standardized, scan/measurement techniques (like QLF for PLRX) seem especially prone to poor data hygiene?
FGEN
Dosing
There are multiple interesting things about the dosing of the various Botulinum drugs. One, which has been discussed multiple times, is that the amount of actual toxin in the Glabellar Lines therapeutic dose is substantially different in the different drugs. This has been discussed before, but I haven't seen discussion of the fact that the scaling into different indications is quite different between the two primary indications (Glabellar Lines and Cervical Dystonia):
For Daxi CD it is 3x to 6x the GL dose (want to see label when it comes out)
For Botox it is 12x (avg)
For Dysport it is 10x
For Xeomin it is 6x (recommended starting dose) to 12x
There are a variety of potential implications from this:
a) determining dose off-label (eg Frontalis) will require some experimentation
b) the revenue for Daxi in CD will not be the same multiple of GL as seen in other botulinum drugs
Random comments:
1) This is largely a function of diffusion? Which may be different in different tissues for different drugs?
1a) Doctors that use their own dilutions may assume that the same dilutions that worked on Botox etc will work here - but that seems highly suspect?
2) Weird scaling may also be a function of the fact that Revance had ability to run CD Phase 3 using all the latest targetting techniques, so they could use smaller dose? Similarly... current treatment using older drugs also saves drug quantity? (Would love to have current data on CD drug use vs label)
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8648634/pdf/40265_2021_Article_1631.pdf
FWIW I believe he’s (indirectly) asserting that dosing required isn’t what Revance claims, although I don’t know how you’d know that this quickly after usage starting?
He’s also asserting that practices will want ppl to come in more frequently, so from their perspective it’s a negative that if dosing moves out to every 5 or 6 months. And many patients already come in more frequently, so for them it’s not very important whether dosing is every 3 or 6 months?
FWIW - the kidneys are difficult. Lots of things, like potassium absorbers, ‘should’ work, but don’t. Latest is for nephrolithiasis treatment:
https://www.nejm.org/doi/full/10.1056/nejmoa2209275
While interesting that brain volume was lower in patients treated with amyloid drugs, brain volume is complicated. Eg inflammation etc will affect it. So as long as cognition improves, and continues improving (as seems to be true), not sure it matters?
Forehead was the least duration of any of the major areas of face in the below daxi paper (used 32 U for frontalis) - about 4 to 8 weeks less depending upon exactly what metrics are used.
https://journals.lww.com/dermatologicsurgery/Fulltext/2023/01000/Treatment_of_Upper_Facial_Lines_With.12.aspx?context=LatestArticles
But also true that it’s a harder treatment area? Eg see this (old) paper for Dysport
https://pubmed.ncbi.nlm.nih.gov/16549707/
On Atlantic article:
Authors of the preprint that was source of Atlantic article are now locked out of the database from which the data item came. Almost certainly because they violated TOS for front running a paper in review using the data loaded onto the database to enable review.
And note the authors are a Who’s Who of zoonoti who’ve been ultra nasty to anyone with a different view and, as FOIA emails make clear, view their primary mission as controlling the message by any means necessary. (The Gao paper, which is the source of the data they front ran, won’t say what they want it to. Hence violating TOS, publishing in Atlantic before preprint out, bizarre message in preprint, …)
Nope.
A) they are clearly trying to front run a George Gao (China CDC head) paper using the data he released for a review of the paper. They’ve done this before. In this case their urgency made them write an article before even having a preprint out. (Gao’s paper is pretty unlikely to reach the conclusion they want)
B) WHO Director explicitly said the data they reference does not support the conclusions of the Atlantic article.
Background: suggest need to read the FOIA emails etc. Emily Kopp has kept a very good timeline. Pretty clear that many of the important zoonoti virologists actively believed in early 2020 that the FCS (wrong codons etc) indicated it was likely engineered. Then switched, over a very few number of days and for no reasons stated in any FOIA’d material, to managing a message that Lab Leak was a conspiracy theory.
Twitter thread of biotechs that have made statements about SVB
other $SIVB-related filings this afternoon $XBI
— . (@OxAnalyst) March 10, 2023
non-exhaustive pic.twitter.com/6me4uP3mzL
Yeah - this below was just stupid for a bank (which live or die on customer confidence):
Long COVID:
I had hopes that the higher prevalence and higher severity of long term sequelae from COVID wld spur a lot of useful research in some things that are similar to CFS. But at this point that seems highly unlikely. We aren’t even doing good research in much more measurable long term sequelae, like long term clotting disorders.
The problem is that while some long term sequelae are likely real, even some of the squishy ones, … it’s very heterogeneous. The only way to make progress is focus, ideally first and foremost on subsets with the most measurable issues. Hopefully that then leads into insight in squishier systems. But instead we’ve focused on:
A) trying to include everyone
B) focusing on loudest complaints first, even if extremely squishy (eg brain fog). And note that I don’t doubt brain fog exists for some ppl, but it’s almost impossible to measure (see CABG pump head for similar issues), and is further muddied by what is almost certainly a high percentage of attention seekers. Sometimes you have to look under the light first.
>>There is no error;<<
The actual text of the letter (??) is conspicuously missing any mention of duration. Just ‘after injection’. Undoubtedly they will claim they meant after 24 weeks, but as a hit piece that won’t get traction with the FDA anyway, better to just leave it out.
GTHX
GTHX has leaned way out over the front of their skis. Eg on this terminated trial they had not done a ph2, or indeed had any human data in this protocol or adjacent. They just relied on a speculative MOA w very sketchy data.
They are similarly ahead of themselves on the upcoming ph3 interim result. Huge change in protocol/population from ph2 and massively underpowered (it’s effectively two entirely separate trials in a single trial).
It’s a pity because it’s an interesting drug. It’s probably the single best poster child for the biotech aphorism that ‘even with a very good drug you need better than average management (similarly … vice versa)’.