7/31/18

Biopharma/Investing
————————-
Well, Yescarta is doing quite well, actually. $68m last quarter as per Gilead’s most recent quarterly earnings. That’s not awful! This is one-time treatment, however. So unlike most drugs, these patients aren’t paying a subscription fee. For Gilead that kind of reminds me of hepatitis C, but of course, unlike HCV, cancer patients don’t come from a “pool”, they simply arrive every year. It’s far too early to tell but generally a performance like this would suggest a $1 billion medicine, at least. Makes KITE a decent deal. However, as in all things, you can never know for sure. Next quarter could be down! The drug may plateau at $500 million or go on to $2 billion+ performance. No idea.

Surprised how well ABBV is holding up in Hep C.

It is still fascinating to me that drug journalists not only are unfamiliar with statistics, but they also appear to be rooting for drug companies to do poorly. The hatred towards AbbVie for not having biosimilars in the US is particularly funny. Are these writers taking Humira for rheumatoid arthritis? Do they think that biosimilars to adalimumab, which represents ~2% of US drug spend will save what they perceive is a healthcare crisis? Have they looked at AbbVie’s patents and have pointed out the prior art that renders them meaningless despite the respect shown by the biosimilar companies? Emotion trumps inference for the cognitively challenged.

Papers I’ve Read
——————–
Semagacestat Is a Pseudo-Inhibitor of gamma-secretase. Tagami et al. Cell Reports 2017:21, 259-273.
Tagami and Shionogi workers go on an interesting journey to learn more about gamma secretase. They draw very speculative conclusions but have a foundation: semagacestat INCREASES beta-amyloid intracellularly. So, if you’ve been playing along from last week (and you should), we have to solve a mystery. One solution is a simple law of conservation. If you have a pre-peptide (endothelin, angiotensin are good examples) and it has to be “processed” (cleaved) to create its final form, and you inhibit the enzymes that do this, you still have the pre-peptide (sometimes called a pro-peptide, or even a pre-propeptide.) If that entity aggregates, you’re in trouble. Peptides are eventually proteolytically degraded but if they are aggregation-prone, you’re out of luck. The writers assume semagacestat is inhibiting some other functional part of the GS complex, the part they speculate trafficks AB out of the cell. More work is needed, but it neatly explains all the results seen so far. This is a great lesson in preclinical experimentation: make sure you know what you’re measuring. AB in the “soluble fraction” (extracellular) may drop but if its being retained, you haven’t solved the problem.

Primary Prevention of Cardiovascular Disease with a Mediterranean Diet. Estruch et al. NEJM 2013.
As we close the semagacestat puzzle, here is a new one for young, would-be drug hunters. What do you see in this publication that is strange? Are the conclusions reasonable? Can we reject the null hypothesis? Why not? Answers in the next blog. Guess away in the comments.

Phase 3 Trial of Ibrutinib plus Rituximab in Waldenstrom’s Macroglobulinemia. Dimopoulos et al. NEJM 2018.
The “iNNOVATE” trial demonstrated a death-defying HR=0.20. WM is a weird little disease, often an afterthought given its relatively indolent nature compared to its cousins. Still seems like there could be an opportunity here for another agent.

Dupilumab Efficacy and Safety in Moderate-to-Severe Uncontrolled Asthma. Castro et al. NEJM 2018.
Efficacy and Safety of Dupilumab in Glucocorticoid-Dependent Severe Asthma. Rabe et al. NEJM 2018.
New Biologics for Asthma. Drazen & Harrington. NEJM 2018.
Great data. Asthma is a bigger disease state than folks realize. Very much looking forward to the Dupixent Q2 number.

Personal
————–
The last post about intelligence requirements for STEM vs. the humanities was triggering. But, it was a trap! There has been enormous data generated recently that correlate intellect with educational achievement and even SAT performance (high R^2!). I must admit I’m probably going a bit too far when I say learning quantum physics is “harder” (requires more cognitive ability to perform adequately) than gender studies. One can be a master of each discipline. Was Einstein more of a genius than Shakespeare? No, definitely not. Was Shrodinger no less of a luminary than Duca? In all seriousness, I think a well-rounded education is important. I don’t write frequently anymore, but once-upon-a-time I thought I’d do it for a living. My problem with science business (pharmaceuticals, for instance) writers is they don’t know science or business. There are very few people who know both fields, and you can bet they’re not writing about it. Before you bicker about some prevailing inferiority complex, consider the above an advertisement: I’m doing it!

Glossary
———–
antibody – n – An immune system protein, “Y”-like in shape, composed of Fc and Fv regions. Binds to, and generally eliminates, a target molecule called an antigen, usually another large molecule/protein. Originally manufactured in mice (murine antibodies).
chimeric antibodies – n – Obsolete now, but antibodies created with a human Fc region and mouse Fv region.
complementarity-determining region (CDR) – region of an antibody that defines its binding to its target
consanguinity – adj – Related parents (inbreeding). Very common in rare genetic diseases. A restricted gene pool tends to result in autosomal recessive disease — the Amish, certain Jewish sects and other populations that simply don’t see much migration have a larger share of rare diseases.
epitope – n – exact binding site of antibody’s antigen.
eosinophils – n – a type of white blood cell that mediates an allergic/asthmatic response.
Fc region – n – “Framework constant” region of an antibody. The bottom part of the Y.
fully human antibody – n – generally created in transgenic mice where the mice are genetically modified to only create antibodies from human genes.
Fv region – n – “Framework variable” region of an antibody. The arms of the Y.
Humanized antibody – n – a chimeric antibody with a human CDR framework (much of which is constant).
intent-to-treat – n – for statistical purposes, the entire enrolled trial population, usually that was randomized and received at least one treatment dose. As opposed to per-protocol population, the pool of patients that completed the trial correctly. Test statistics are almost always conducted on the ITT population with missing data interpretation defined prospectively.
monoclonal antibody – n – See antibody. The monoclonal adjective distinguishes from polyclonal, in that monoclonals come from one source and have one epitope.
type 2 inflammatory response / immune response (Th2) – n – Elevated levels of eosinophils, IgE, IL-4, IL-5, IL-9, IL-13 and other immune markers that distinguish the response from Th1.

17 thoughts on “7/31/18”

  1. Will leave more detailed, specific comments later (when I have more time), but just want to say… THANK YOU! There is no one else providing this level of industry/academic commentary with the engaging, nonchalant and often humorous writing style that you have. I thoroughly look forward to every blog post!!

  2. “My problem with science business (pharmaceuticals, for instance) writers is they don’t know science or business.”

    Couldn’t agree more. My problem with your article is that you are writing about things you don’t know. Go back to school and learn something, if you have to.

  3. You should start a drug journalist/columnist website when you get out. Put that clown John Carroll out of business. That guy is freaking useless. My pubic hair has more intelligence. Stay strong buddy!

  4. I found the Mediterranean diet paper rather strange. I don’t think we can reject the null hypothesis (there is no relationship between major cardiovascular events and adherence to a Mediterranean diet) due to all of the participants living in Mediterranean countries and all participants have high cardiovascular risk.

    I’m brand new to this but I think you need to have a way more diverse set of participants. By taking every participant from the same geographic area, you have to consider genetic factors in how the diet interacts with the participants’ physiology. Would we have the same results if participants were taken from all over the world?

    The participants also all had numerous of health issues such as diabetes, hypertension, obesity, and cigarette addiction. It seems possible these health issues introduce a lot of variance into the study and maybe wasn’t sufficiently considered in the n=7400 estimation. Also, this doesn’t elucidate the risk reduction of adopting a Mediterranean diet as an already healthy individual.

    Looking forward to hearing the correct answer. Thanks for the blogs, Martin.

  5. I don’t know if someone sends you these comments but I would really love to see a longer take on the current situation with Alzheimer treatment.

  6. The primary and secondary endpoints don’t match for the same information. If the primary consists of MIs, Stroke, and death from cardiovascular cause, and the secondary consists of all of the above plus “death from any cause”, the sums don’t match when you subtract the “any cause” from the given info of the other three events.

    The first group had primary as 96, but when you take the details of the secondary, those 3 equal 112. Same issue with the other two groups.

    Also, they treated the control group differently from the experimental groups, fucking it up from the jump :p

  7. Degrees being awarded in the humanities are continuing to take a dive. Many factors at play, but intelligence requirements are the least of this trend’s worries. Infiltration of faux-subjects with the -studies suffix attached, ideological domination of leftist faculty and admin creating a no-go zone for right of center thinking, veering away from the classics, lack of scientific rigor when it comes to the social “sciences”.
    There is definitely a problem with the way a student is told how to think rather than how to hone critical thinking skills within the humanities of today. Tragic since the humanities are supposed to be the core of a post-sec education.

  8. Despite the science being completely over my head, which limits my patience for reading and digesting the technical passages, I enjoy this blog immensely. There is much to glean, not the least of which is your wit and humor. I hope you’ll continue writing and teaching for the long-term.

    As I write this at nearly 3am, I’m still eagerly awaiting the magic pharmaceutical that alleviates the need for sleep. Hope you have one of your teams working on that one. lol. Miss you, my friend.

    1. Elon created something. Martin simply bought drugs to profit off of. He’s a smart guy, but he’s fairly average in intelligence

  9. What do you see in this publication that is strange? The study was based in Spain for 10 years.
    Are the conclusions reasonable? Yes
    Can we reject the null hypothesis? Yes and no. Yes because the trial was based in Spain. A predominately Mediterranean eating country. If they eat Mediterranean regularly then switch up to a controlled diet then obviously it’ll change to a drastically different result compared to a regular diet.. No because it clearly showed and stated that consuming more Mediterranean food reduced the chances of Cardiovascular Disease.

    I’m not a scientist or anything so my opinion is most likely wrong and irrelevant

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.