Ok little mouse. I'll talk a bit about the measured biomarkers and endpoints.
Your concern is that MSB data dredged and moved the target to suit the data and used evidence that this >2mg/L hsCRP was pulled out of nowhere.Here are the other Outcome Measures they measured in the trial:
1) Pharmacodynamics Measures (NT-proBNP and hsCRP) [ Time Frame: Screening, months 3, 6, 12 and every 12 months thereafter until study conclusion ]
2) Pharmacogenomics (PGx) Analysis [ Time Frame: Screening (only from those subjects who provide consent to participate in PGx sample collection) ]
3) Immunogenicity Measures (panel reactive antibodies, donor-specific antibody, if PRA test is positive, antibodies against bovine and murine products) [ Time Frame: Screening, day 10, months 1, 3, 6, and 12 ]
I'll break down 2) and 3) first.
3) Why did we not see any reported immunogenicity measures? Because the product didn't induce a foreign body response.
Yay!2) PGx is trying to identify groups within the studied population to identify genetic responders/non-responders. Nothing reported, but that's not to say something won't have been signalled, wait for the Journal Article to find out. Regardless, with nothing reported, there doesn't seem to be a genetic requirement in the patients as a precursor for efficacy.
Yay!1) Now the measure of NTproBNP and hsCRP
N-terminal pro b-type natriuretic peptides (NTproBNP) are peptides (small proteins) that are either hormones or part of the peptide that contained the hormone at one time. They are continually produced in small quantities in the heart and released in larger quantities when the heart senses that it needs to work harder.
High-sensitivity C-reactive protein (hsCRP) is an important risk factor for systemic atherosclerosis and is related to its clinical complications. Small studies have shown that plasma CRP is elevated in patients with HF. In several community studies, plasma CRP predicted the development of HF and other adverse events.
First of all, the hsCRP level is just a predictor of efficacy and helps MSB develop a target population and develop potency assays if required.
Taking the hsCRP analysis away we still had "60% reduction in cardiac death in NYHA class II patients (p=0.037) and prevention of progression to NYHA class III rate of cardiac death (p=0.004)", so your concerns over hsCRP are already moot as MSB substantially decreased death in a group that was clearly pre-defined as shown by its exclusion criteria.
As
@col69 already pointed out,
Accelerated Approvals are designed for approvals for severe diseases with significant unmet medical needs based on a surrogate endpoint. The defined Primary Endpoint was a surrogate endpoint for... death. MSB skipped the surrogate and went straight to improvement in death. The FDA takes into consideration the duration of these trials - don't get me wrong we're definitely not going to get a full approval off this, but a phase 4 trial with accelerated approval is easily on the cards to stop 5-7 years worth of CHF deaths. We already should have feedback from the FDA on our CHF pathway as of the 13th of November, so I'm sure that conversation will be ongoing considering
the diseaseappears on 13.4% of death certificates in the USA.
Second of all, the disclosures on clinicaltrials.gov are to provide transparency to researchers and participants on the trial design. When has a trial EVER disclosed how they will perform their statistics or how they are planning on breaking down their analysis on a clinicaltrials.gov registration. They don't. Those sorts of things are included in the application to the FDA to run the trial, ie.
they are pre-specified with the FDA as sub-groups for analysis, but are never included on clinicaltrials.gov because that's not its purpose. It seems pretty clear that the pre-specified value for NTproBNP was not indicative of success, however, hsCRP levels at the trial initiation were. Change hsCRP levels throughout the trial will be mentioned in the publication, but are hardly important enough for a 10-minute slot.
Third of all,hsCRP is an inflammatory marker. While it is not well researched or understood, I am betting that hsCRP will be present in patients with CHF at some point, probably when their disease is early and more acute (as we have seen with MSC therapies before) - hence the success seen in NYHA class II but this drops off into NYHA class III. However, when you look at hsCRP levels as a predictor of efficacy, the 'n' of the sample goes up to 301 as it includes some of the patients in NYHA class III that are still inflammatory.
This is super encouraging as this will likely provide an 'intervention period' denoted by elevated hsCRP.
Fourth of all, you're hung up on where they got >2mg/L from. It is well characterised that hsCRP is predictive of death for CHF patients. If you look at the graph below,
patients with LESS THAN hsCRP 2mg/L still die at a high rate.
So this 2mg/L line in the sand is the prespecified measure MSB agreed with the FDA, which is why the results are so stark with an 80% reduction in death, but it does not mean that it is not beneficial in patients with 1-2mg/L hsCRP - so efficacy isn't even likely to be limited to patients with levels over 2mg/L.
https://www.karger.com/Article/Fulltext/351115Altogether, this whole argument that they data dredged to try and get meaningful data and the product is dead in the water without another trial has no basis.Accelerated approval is still on the cards, let us see if the FDA see enough need for the product.
Meanwhile, I hope whatever short positions people have are at a level where they're comfortable exiting, a risky game with what's around the corner.
Gang Gang
)
(20min delay)