MOZZ OARSI 2023 - NOTES, page-15

While the snapshot argument is true in principle, PAR still need to be able to scientifically explain these results, which they should be able to do using the results from the canine study (and other pre-clinical studies), as well as the 6 month follow-up data.

The truth is PAR only need one of these biomarkers to act as a surrogate marker (for accelerated approval), and a select few to support DMOAD status. They don't need all of them. ARGS is proving to be the MOA and surrogate endpoint, whereas PAR may need to combine the overall effects of the remaining biomarkers to highlight DMOAD, as opposed to looking at each marker individually.

If I were PAR I would perform principle components analysis for twice weekly, once weekly and placebo, looking at proteins associated with cartilage degradation and inflammation (leaving out ARGS), and then look at the clusters to see if all three groups can be discerned from one another, which would indicate DMOAD status. They could also pull out individual markers that are clearly not showing any change which would make the clusters tighter, then follow this up with a statistical test such as a t test.