The following is my opinion of the briefing documents' content. No-one should use the following to make investment decisions without conducting their own research.
Glossary of Terms: BLA Biologics License Approval CQA Critical Quality Attributes (the things MSB claim make Rem-L have potency and that they intend to control in manufacture DP Drug Product MAGIC. Mount Sinai Acute GVHD International Consortium MSB-GVHD001MSB single-arm pediatric SR-aGVHD clinical trial for Remestemcel-L ODAC Oncological Drug Advisory Committee ORR. Overall Response Rate RCT Randomised Clinical Trial Rem-L Remestemcel-L SAP. Statistical Analysis Plan SR-aGVHD Steroid Refractory acute Graft Versus Host Disease My comments inItalics.
Following on from the previous post, I'll now examineFDA-ODAC-PMthe briefing document for ODAC afternoon session, this continues on from Section 2.2
Section 2.2:This section deals with the history of Rem-L through a series of trials going back to Osiris in 2008 to the present day, with the information behind 14 trials provided by MSB. Protocols 265 and 280 by Osiris didn't meet their primary goal of CR last more than 28 days. Although the FDA lists these trials as failures, MSB used them to inform future trials (post trial analysis showed some statistically significant improvement in pediatric response and some responses for particular GVHD symptoms (liver etc). What follows are key statements that the FDA makes about MSC trials:
You can do a single-arm trial and satisfy FDA needs and you might be able to define a "control" based on modelling of historical patient controls or studies. But such a study needs to be designed and reviewed prior to its conduct.
Proto(col) 275 is not a well-controlled trial and doesn't supply confirmatory evidence to support BLA
Proto 280 is a negative trial
Proto 275 and 280 can be used to inform future trials - MSB should consider an RCT to provide confirmatory evidence of efficacy
FDA recommended an adult RCT versus SoC for SR-aGVHD - could provide confirmation of pediatric benefit
MSB-GVHD001 allows us of other prophylactic and other agents that may effect efficacy outcomes - confounding interpretation of treatment effect
Null Hypothesis for MSB-GVHD001 not based on historical controls - FDA not able to agree that it is acceptable.
Given absence of concurrent controls FDA doesn't think MSB-GVHD001 is an adequate well-controlled trial so probably not sufficient to support BLA.
Any claim of efficacy based on MSB-GVHD001 needs to take into account all studies of Rem-L for aGVHD including the failed trials.
MSB flied BLA with MSB-GVHD001 as sole basis of efficacy.
OK. This seems to be the real "meat" in the FDA's position which didn't seem to shift despite being considered by the ODAC, which overwhelmingly disagreed with their position (it should be noted that although the ODAC vote was 9-1, some members expressed surprise at the result (they thought it would be closer) and one said she voted yes because her internal balance was 51% for to 49% against - which means it was not quite as strong as the stark figures imply).
So let's start at the bottom of that list, "efficacy meeds to take into account all other studies". I find this rather strange. The FDA has stated earlier that Rem-L in MSB-GVHD001 was modified by MSB from that used in prior studies, so their results are largely irrelevant. It seems to make no sense to dismiss every new CT based on a failed trial with a modified earlier version of the drug. Perhaps MSB have erred in not designating this as an entirely new drug - i.e. toss out Rem-L as a name (Ryoncil is a marketing name) for the drug and refer to any royalty stream owed to Osiris as patent licensing arrangements. The Osiris trials seem to weigh heavily in the minds of the FDA. In particular, because they were placebo controlled RCT, they seem to think they establish the efficacy (or lack thereof) above all other trials. It clearly indicates that a positive well-controlled RCT would appear to be a necessary (and perhaps sufficient) step before the FDA would consider the results of any single-arm study. I really believe this opens the door for something like the Covid-ARDS trial providing a clear efficacy signal for Rem-L in immunosuppression which once accepted would the FDA to reconsider the findings of the single-arm studies positively, just as they were able to use the single-arm study on Jakafi, based on a prior positive RCT for Jakafi (albeit in what appears to be a rather different MoA).
I personally don't see why the FDA is so insistent on this, since a reading of the post hoc analysis of Proto 265 & 280 showed clear statistical efficacy for the target population of this BLA and they were used to inform the later trials, but that seems to be the reality of the FDA position.
The first point, seems to concede that a single-arm trial would be enough for an approval so the reason for rejection must lie in the latter points.
The second point, rejecting Proto 275 out of hand appears very weak and should be the subject of MSB scrutiny. the FDA state that it isn't well-controlled or adequate, but produces no evidence or direct criticisms in the document to support their statements. Proto 275 had multi-centre support (which should reduce any severity diagnosis bias) and was an EAP - there was no control - i.e. everyone received the treatment, however all of those participating hospitals have a wealth of prior history of outcomes with SoC with aGVHD. In addition the study clearly states that the 69% ORR was obtained despite the fact that the MAJORITY of the 241 subject had failed with other agents - i.e. it was a rescue therapy. Given a number of studies that correlate ORR @ 28 days with eventual outcome, and given it was being used as a rescue therapy, moving the ORR dial up to 69% of the population was actually highly significant. As for lack of controls, a number of studies, including REACH2 for Jakafi point to a SoC result for ORR of less than 40%, once again showing that 69% is a clearly improved result. For the FDA to reject this would mean that they would envisage either a significantly better ORR for SoC (for which they have zero evidence) or they believe that the gradings of severity across multiple institutions with strong experience in this disease are incredibly biased towards overscoring - once again, they provide no evidence to justify this position.
Proto 280 is negative, while true is also largely irrelevant. Different drug, different target population from the BLA.
Point 5 is asking for an adult RCT - but even there it equivocates as to whether this would inform us about pediatric benefit by using the word "potentially". However this seems to be the course they are asking for in the CRL.
Point 6 claims that some of the other concurrent treatments may be confounding the analysis of Rem-L efficacy. Once again does the FDA possess and data pointing to a 69% ORR in pediatric aGVHD through the use of SoC? If they do, they should send it out to all these participating hospitals because they aren't seeing that level of response. While the point is "true" it is also insignificant unless SoC is seeing these sorts of ORR rates. It's the equivalent of pointing out a minor niggle that the efficacy might have been 68% rather than 69% because of other drugs (it could have also been 70%). They have zero evidence to support any major influence this would have had because all historical studies support much lower ORR.
Point 7 Is a criticism of the null hypothesis - once again the FDA approved Jakafi which used a 40% ORR as the base no-Jakafi ORR.They didn't seem to have a problem with that, but they seem to have a problem with 45%. They offer no null hypothesis of their own. Multiple lines of historical data were used to judge the appropriateness of 45% as the null hypothesis - the FDA provide zero data in support of their "Nah!"
Point 8 is just a summary of their negativity on the trial based on they zero-data rejection of the null hypothesis and their inkling that some of the complementary drugs might have influenced the result. It's all very subjective with no actual objective data introduced by the FDA.
So that's the end of Section 2. It clearly shows the FDA's position and the hurdles that MSB needs to jump to satisfy them. I don't agree with the majority of the FDA's conclusions based on the fact that they are either trivial (i.e. confounding drugs would have not moved the ORR needle anywhere near 69% based on their inability to produce these results outside of this trial) or totally subjective - i.e. rejection of null hypothesis without providing a justification of any higher figure. The continued reference to Osiris trials is also an issue that doesn't seem like it will go away without a placebo controlled RCT that produces a positive result in some immunological disease.
I'll tackle section 3 in my next post if time permits.
MSB Price at posting:
$3.36 Sentiment: Buy Disclosure: Held
(20min delay)
