Ann: CEO Presentation to Bioshares Conference, page-24

Yesterday's FTT presentation at Bioshares contained a couple of updates that I think further reinforce the chance that the trial will hit it's primary.

The first is that the healing curve of the blinded cohort has remained more or less stable with the addition of 22 patients (91 to 113), indicating that data variability is probably not high.

The second is slightly more subtle, but still important. Specifically, we may have gotten some insight into the consistency of matched control subgroups between studies. This is a critical consideration in assessing the risk of a shift in the control arm compared to the historical subgroup analysis that FTT ran alongside the average healing curve of the blind cohort:-

Previous presentations from FTT have included comparative 'probability of non-healing' curves for subgroups between VF001 and matched controls. Ros's presentation at Bioshares also included a comparative non-healing analysis but it used what looks like a different data-source (which I assume to be VenUS-III, but can't be sure because the chart labeling was poor). The control curve's presented by Ros yesterday (slide 6) had a different n to previous analysis and a slightly different shape.

The reason this is interesting is that we now have access to what looks like two independent subgroup control curves (one from Ros's presentation n=23 and one from previous presentations n=68). This in turn means that it is possible to get an idea of the expected variability of control subgroups by comparing the probability non-healing curves from each of the samples....and the result is very encouraging.

As you can see below, the two control curves are fairly similar. This is good as it gives a hint that the risk of a significant departure of the control arm from the sample analysis presented by FTT is lower than I thought it might have been in the absence of this cross sample analysis (remember we need a 17% improvement in the control arm over the VenUS-III sub-sample to disrupt the inference of success from the currently available blinded healing curves).