Thanks everyone for the interest in hearing an opposing point of view, and I hope those that aren’t interested don’t take this the wrong way. The DREAM-HF trial could very well be a resounding success, in which case I am happy to eat my words and more importantly be welcomed back as a shareholder long term for what would be an incredible ride. I would happily accept a 30 bagger while you all cash in 100 times. Obviously if I am wrong with any points I am happy to be corrected - If I am way off the mark I might still have time to buy back in before the readout.
Also this might be full of mistakes, I have just had a few beers and a good steak to celebrate a profit.
——
I’ve based my opinions on the likelihood of success with MSB trials based on:
- My interpretation of the data in the context of my experience interpreting medical research as well as my knowledge of physiology and pharmacology.
- How plausible I think the mechanism may be.
I thought I might structure this post based on these two points with the majority of time being spent on my interpretation of the data. I have also gone the ecoool2 way and highlighted headings and what I think are important points.
In addition to the below opinions I also wonder, and this is a key point for me, why are we injecting the cells into the myocardium instead of giving them IV? IV administration seems to work fine in GVHD.
My interpretation of the data:
General Stuff
What makes a good trial?
And what makes people use a drug once that trial is published and approved?
Put simply a well structured trial has a large sample size with adequate blinding as well as reasonable primary and secondary outcomes.
- The DREAM-HF trial seems to satisfy the first two criteria - 500+ Patients seems enough and the blinding seems ok. Obviously a criticism of the paper is that it is conducted by MSB and therefore isn’t truely free from bias.
- The outcome in this trial is a little blurred in that it us a composite endpoint (MACE). Any sort of composite outcome is a lower bar to set than hard outcomes such as 30 day mortality, number of hospital admissions, NYHA class or LVEF - We saw that in the LVAD trial.
- With positive results (especially if they are borderline) I would like to see the breakdown of what caused the MACE episodes. - I still don’t have a great idea of what was involved in the enrichment process
Put simply, a drug that works well, has minimal side effects and is cheap will be used widely.
- The really high price of MSCs is a limiting factor for me, which in my opinion means they had better work really well. That said, if the results are resoundingly positive then people will pay a premium.
The Phase 2 Trial:
Headline data that looks great, but let’s dig a little deeper. I guess it’s lucky that we didn’t cap the dose at 75M right?
Obviously phase 2 trials don’t have large population sizes or have adequate power for FDA approval, really they are pilot studies to find a dose and to support going on to an expensive phase 3 trial. As such, for me,
the phase 2 trial satisfies none of the above criteria for me to accept the results as significant regardless of P value.
In terms of looking at the validity of the data I guess when I look at a trial I am looking for results that fit with my expectations. Please interpret this in the context of traditional drug pharmacology and not specific for MSCs, although debatably the same principles apply:
Expectation 1 - The Dose-Response Curve
As a general rule the dose of a drug is proportional to its magnitude of effect. Although the relationship isn’t the same for all drugs and it happens in a non-linear fashion - the principle is almost universally true.
Take MSB’s phase 2 trial in RA, the results below almost always show a graded response to increased doses of MSC’s
Now how does this compare to the HF Phase 2?As you can see below, the data suggests that this relationship is not the case in the phase 2 HF trial. Why would injection of 25M and 75M cells result in worsening echo parameters compared with an improvement with 150M? It seems like 150M cells has just been cherry picked based on this. Also note that based on this data the phase 3 trail has been narrowed down to those with systolic heart failure (HFrEF).
The curve at the top of the post with combined MACE events clearly doesn’t display this principle either
Expectation 2 - Consistent Data If we accept that MSC’s do actually improve echo parameters, what I would care about (both as a doctor and a patient) is a functional improvement measured either by qualitative measure such as NYHA classification or by a quantitive test such as 6MWT.Interestingly, there actually seems to be very little in terms of
clinical outcomes that supports proceeding to a phase 3 trial. As you can see below, the patients didn’t show a significant improvement NYHA score or 6MWT. Note that the
150M MSC group had the lowest number of NYHA 3 patients, whereas the control had the highest. This results in better post-intervention results in the study groups but largely as a result of better pre-intervention function.
If true, why does an improvement in echo parameters not translate to an improvement in function?
Part 2 - How Plausible are the Results?
Without going into detail, let’s accept that MSCs largely exert their effects by immunomodulation and anti-inflammatory properties. If we accept that then we have Biological Plausibility as to why they reduce disease burden in GVHD and potentially work well in autoimmune diseases (RA) and in inflammatory diseases such as back pain.
But how does that translate to heart failure? These patients have structural heart disease. While I accept that there is emerging evidence that heart failure patients have inflammatory processes occurring which may be a result of (or worsen) their cardiac disease, in comparison to those with inflammatory diseases they don’t universally have raised serum markers such as CRP or ESR which validate the process.
One thing that always bothers me on this forum is the idea that MSC’s are going to treat any and every disease - It’s the same story as cannabis oil (amongst other things) which is effective in proven indications but not the magic bullet the crazies will have you believe.
In summary, the idea that injecting MSCs cells in the myocardium doesn’t (in my mind) plausibly result in improvement in cardiac function and, more importantly, is unsupported by the data we have available in the form of an MSB sponsored phase 2 trial. That isn’t to say that the DREAM- HF trail won’t produce decent results, but what makes anyone think that it will other than blind faith in the company? I think based on the above the likelihood of that is much less than 50% so I choose to protect my capital. I am sure others have differing opinions.
Best of luck everyone, I hope it all goes well. I'm sure I'll be a holder again one way or another - GVHD is a sure bet.
(20min delay)