Valid point, and I considered that. What's new/what has changed.
What has changed is the IL2Ralpha component as described in this patent here, Figures 4-7:
"FIG. 4 : Product manufactured with improved process has higher mean TNF-R1 levels and less variability in inhibition of IL-2Rα.
FIG. 5 : Manufacturing changes resulting in increased TNF-R1 levels have been accompanied by increased Day 100 survival.
FIG. 6 : Significantly improved survival in patients that received MSCs associated with IL-2Rα expression levels >60%.
FIG. 7 : IL2Ra inhibition in vitro correlates with in vivo reduction of Activated CD4+ T cells."
Danilkovich et al didn't mention the IL2Ra link
(
https://patents.google.com/patent/US11821004B2/en), which forms part of the pending patent of Itescu and Simmons.
In particular Table 2 [0228] is showing you how they have re-analysed ("mined") existing data from studies 265, 275, 280, 001/002 using the above link, which then showed an improvement between the old and new manufacturing process in survival which coincided with an increase of TNFR1 and IL2Ra from 58% to 75% survival on day 100.
Problem is, if you look for a connection, you will find one eventually in any data set, even if it's something silly and obviously unrelated, I believe
@JB1975 was using an example of ice cream sales correlation with shark attacks. But really, you should have a hypothesis first (in this case MoA of MSCs in SR-aGvHD), you then run the trials and the data obtained will either prove or disprove your hypothesis (nothing wrong however with fine-tuning - that's what the trials are there for).
BLA resubmission #2 included the following "new" data:
"long-term survival data from a phase 3 trial (NCT02336230) with 4 years of follow-up."
"new outcome data on the agent's activity in high-risk disease by propensity-matching patients in the phase 3 trial and controls from the Mount Sinai Acute GVHD International Consortium (MAGIC) database who were stratified based on biomarkers."
"highlighted data on the validated potency assay, primary mechanism of action, and in vivo bioactivity related to remestemcel-L."
"highlighted low variability and showed consistency and reproducibility in relation to manufacturing."
https://www.cancernetwork.com/view/fda-accepts-remestemcel-l-bla-resubmission-for-pediatric-acute-gvhdWhilst this at first glance sounds great and "new", it is still all based on old clinical data, and resulted in CRL #2.
In the Mesoblast presentations from November 2023 you can then see a different approach, and references to actual new data (new EAP data):
"In contrast, 100-day survival was 63% after remestemcel-L treatment was used under expanded access in 71 patients aged 12 and older with SR-aGVHD who failed to respond to at least one additional agent, such as ruxolitinib."
https://www.mesoblast.com/images/Mesoblast_-_Corporate_Deck_-_NOV_2023_FINAL.pdfFebruary 2024:
"In contrast, 100-day survival was 67% after RYONCIL treatment was used under expanded access in 51
adults and children with SR-aGVHD who failed to respond to at least one additional agent, such as
ruxolitinib."
https://investorsmedia.mesoblast.com/static-files/6e5bffe8-3485-4f66-a9a4-344388b68d5bAlso worth highlighting how it went from 71 patients aged 12 and older with 100 day survival of 63%, to then 51 patients (adults and children) with 67% survival on day 100. They went away from the 11 data points initially, the propensity-matched ones from GVHD-001 to use the prior trial data as their new/adjusted hypothesis and supported that with first 71, then 51 new data points in patients that fell into that "sweet spot" as outlined in I&S's pending patent.
Whilst the cynic in me looks at the decreasing patient and increasing survival number perhaps a bit differently than others, I believe that the above shows why MSB is confident this time around to avoid CRL #3 and instead gain approval.
I hope it all makes sense.
(20min delay)