Having just participated in the Prana CC, there are a couple of informational pieces worth mentioning:
1) The statement that all trends seen on the range of measures administered were in the “positive direction” gives some reassurance that the two specific measures where benefit was statistically significant or approached significance (Trails B and the TFC test of functional capability) are not just straws randomly emerging from a haystack of molecular irrelevance. It supports the hypothesis–and that is what it is– that there is a trend towards improvement on a range of functions that may be more salient, and substantiated, in a trial that tests more people, for a longer duration. Six months may have not been long enough for the control group to decline in some areas to the degree that can be detected by the tests assessing those areas.
2) When one sees a neuropsychological testing result, it always begs the question of its relevance to real-world functionality. The good news here is that improvement was seen on both the Trails B neuropsych measure and the TFC functional composite score. The open question is whether the magnitude of benefit on the TFC (a decrease of 0.3 for the high dose group compared with a decrease of 0.6 points for the control group) would translate into tangible real-world benefit. One of the researchers on the call said that the TFC, which taps six different areas on a scale that ends up covering a range of 0-13, has subcomponents that would not be expected to decline over six months, and some that would (managing finances and occupation function)–the implication being that the benefit may have been in specific behavioral subsets, and that what looks like a small treatment effect may actually be more salient within that context.
3) The presentation reiterated a point made on this blog this morning: The fact that the greatest improvement was seen on the same measure where it was observed in an entirely different patient population (Alzheimer’s) makes it less likely that it was a fluke–because it was replicated.
As was noted earlier, this was a small n, short-duration trial, and should be thought of as a pilot study. But the report that the overall gestalt of trial results was of positive effects, even if not approaching statistical significance, makes it more likely that this is a real signal, not the product of ‘data-fracking.’ The results provide hint-of-concept, not proof-of concept, and now requires replication in a much larger study, hopefully one that would utilize at least a one-year treatment duration (probably with an open-label extension).
While we had been wondering whether a 40 pt Alzheimer’s trial (due to report next month) can shed much light on anything due to its small size, it does have an imaging component that might be very interesting.