MSB 2022 - The road to commercialisation, page-1307

In addition to the above, you also said:

In my opinion it is possible that MSB will succeed in getting a BLA approved without another trial. I can't say that it is certain but I can say that I think its possible with ST2, Reg3alpha and MAP surrogate biomarkers related to survival being used to shore up confidence in the single armed pediatric clinicial trial that has already taken place.

If MSB adds new potency assays besides TNFR1 and IL2Ralpha to the release assays though, which it could, unless those assays are somehow linked to the trial that has already taken place, (ST2, Reg3alpha and MAP could be linked because they are markers from the patients not the products) then I can't see how the new potency assays would help make the already completed trial be sufficient as an adequate and well controlled study.

I think the FDA needs to have an adequate and well controlled study (at least one) to grant a BLA so I think MSB has to rescue the last one to have at least one without having to do another one. So potency assays that are new and on the product side rather than one the patient side would seem to be too late to do that now. You can't use new potency assays on product that has already been administered in a trial already completed to further bulk up the credibility of that already completed trial. You can use new patient date though (ST2, Reg3alpha and MAP). In my opinion.
https://hotcopper.com.au/threads/msb-all-time-low-party-thread.6816848/page-334?post_id=62368237

And I think this is exactly what is going to happen: they may have established a way to link an assay to ST2, Reg3alpha and MAP (on the patient side), data collected as part of the elective biomarker substudy, compared with a "closely matched" MAGIC cohort to provide a control, showing an improved survival outcome. The combination may be something that the FDA considers reasonable as it suggests to have matched the existing P3 data. Sure, it is not ideal and the term "data mining" comes to mind, but it may be the best shot to save it.

That is how I understand the announcement here (underline & [ ] = me):
"The objective of the study was to evaluate whether outcomes differed according to treatment with remestemcel-L vs other therapies in children at highest risk of death, namely those with baseline MAGIC Algorithm Probability (MAP) biomarker levels ≥0.291, a level predictive of very high mortality and poor responses to therapy in SR-aGVHD. MAP combines the serum concentrations of two biomarkers, Reg3α and ST2, into a single value that predicts long-term outcomes and significant GI tract damage.

MAP levels ≥0.291 were present in 48% of remestemcel-L treated children (12/25) and 37% of the MAGIC cohort (10/27). Treatment with remestemcel-L resulted in 67% Day 28 Overall Response and 64% Day 180 overall survival compared with 10% Day 28 Overall Response and 10% Day 180 survival in the MAGIC cohort (both p=0.01) when treated with various biologics, including ruxolitinib [a control of some sort and Jakafi mentioned for a reason given that it has been approved in cGvHD a few months prior]. These results extend previous observations showing that children who achieved clinically meaningful responses and survival after treatment with remestemcel-L had significant reductions in the ST2 biomarker of inflammation, consistent with healing of the GI tract.4 [Note: even the P3 Osiris trial back in 2009 stated, "Prochymal significantly improved response rates in patients with steroid-refractory gastrointestinal GvHD (88% vs. 64%, p=0.018, n=71)" and also, "In patients with steroid-refractory liver GvHD, treatment with Prochymal significantly improved response (76% vs. 47%, p=0.026, n=61) and durable complete response (29% vs. 5%, p=0.046)". Both liver and GI apparently show and increased level of Reg3α in aGvHD].

These data provide further support for the proposed anti-inflammatory mechanism of action of remestemcel-L and its immunomodulatory activity [the poster called it "mechanistic evidence of immunomodulation'] in patients with SR-aGVHD, resulting in improved survival outcomes. At its upcoming scheduled meeting with FDA’s OTAT, Mesoblast will address the appropriateness of potency assays related to remestemcel-L’s proposed anti-inflammatory mechanism of action as well as the outstanding CMC items which could support a resubmission of the current BLA for remestemcel-L in the treatment of SR-aGVHD in children with a six month review."
https://www.globenewswire.com/en/news-release/2021/11/23/2340233/0/en/Operational-Highlights-and-Financial-Results-for-the-Period-Ended-September-30-2021.html

Dr Rose didn't further elaborate on what the problems are, but said that they have reverted to an Assay "that speaks to the mechanism of action of the cells" (the poster called it and he also mentioned that they have now "also been able to demonstrate a clinical correlation between the potency assay and outcomes". Hence I agree with your initial assumption that "new" patient data could be key, obtained by what appears to be an extensive data mining approach. We will have to wait and see if it holds up.