Unfortunately I tend to agree, people see the results and apply what they know about statistics, which tends to be little and mostly wrong.
In my opinion the results are good and promising but not great, that would have made Leo to put some p values in the press release.
However, for comparison:
Compared to Dusquetide (now in P3) phase 2 trial
Incidence of Severe Oral Mucositis
Dusquetide:
Placebo (n = 38): 74 %
Dusquetide (n = 36) 1.5 mg/kg: 69 %
Risk Ratio: 0.93
2 tailed p value (Fisher): 0.798
Brilacidin:
Placebo (n = 25): 60 %
Brilacidin 45 mg/15 ml (n = 21): 42.9 %
Risk Ratio: 0.715
2 tailed p value (Fisher): 0.375
What 95 % of 'professional' analysts will do is to pluck the proportions into Fisher's formula and judge based on that. In this case that judgement is not so good.
Big pharma looks things a bit differently (law of large numbers is known there and I am not referring just to finances :). At The Shameless Swiss Co we used to look at the risk ratio, which in this case is OK for Brilacidin. With cancer anything below 0.80 was good to begin with. Then we would try to figure out 2 things:
1. How likely it is that the risk ratio will hold when the trial is scaled up. One can get an idea of the direction of possible change by comparing patient population and care in the present trial to historical trials.
2. What should be the minimum size of the trial for the probable risk ratio to translate into significant p value.
I can't say a thing about the stability of risk ratio regarding Brilacidin without access to baseline patient characteristics. However, assuming that RR is stable a trial of 140 or more subjects and RR around 0.7 would yield satisfactory p-values. Actually rather small number but maybe, still, beyond IPIX alone. And that's where Leo's problem sets in:
Even if big pharma thinks (and don't think they don't) that the results are promising and do warrant going into P3 they will use the uncertainty of the IPIX's financial situation to bring Leo's demands down, a lot. They can wait for a really long time while Leo can't.
It is like IPIX fell in the same trap as Polymedix, too cheap/small validation trials for their own good.
But let's hope the best - maybe there is one white knight pharma in dire need of potential blockbuster.
