Innovation Pharmaceuticals Inc (IPIX)

[PlentyParanoid]

I am late into the fray as usual. LR, all the additional info needed is in the presentation given in
MEDICAL DERMATOLOGY THERAPEUTIC R&D AND TECHNICAL INNOVATION, Boston, MA, 19 September 2016
source:
https://static1.squarespace.com/static/5715352e20c647639137f992/t/583f7f14bebafbe20767391b/1480556310644/Prurisol-Presentation-for-Med-Derm-RD-Workshop_19Sep2016-u.pdf

Slide 10: Per Protocol population: 20, Subjects achieving at least IGA shift 2: 7.
Slide 12 Right hand chart:
IGA 3 at start: 65 % --> 0.65 * 20 = 13
IGA 2 at start: 35 % --> 0.35 * 20 = 7

Because there seems to be widespread misunderstanding about Intent to Treat and Per Protocol groups, at least in terms of validity, I am going say a bit about them here.

Intent to Treat: all subjects that were randomized for a trial.
What one expects a "real world" experience with a treatment to be. Intent to Treat is preferred because using it eliminates one source of bias. But it is also generally considered that ITT analysis has a tendency to underestimate treatment effect.

Per Protocol: all subjects that completed a trial without (major) protocol violation.
What one expects the effect of a drug to be with perfectly executed treatment regimen and perfectly behaving patients. Eliminating subjects introduces the possibility of an adding bias and underestimating adverse effects. Usually regarded as too optimistic.

The "truth" is out there somewhere in between ITT and PP.

There is occasionally some "fudging" with ITT, trial runners may eliminate subjects that never received any treatment and like. There is a much bigger problem with PP - what constitutes Per Protocol group is not well defined and varies from trial to trial. I, personally, don't use PP data in my evaluations UNLESS it is crystal clear to me what got eliminated from Per Protocol group.

In this case I stick with ITT analysis until IPIX tells me why 8 subjects (original size 28 not 27 as in the table on slide 10) from prurisol 200 mg group and 12 subject from placebo group got eliminated for PP. Reasons could be valid and benign, I just don't know.

BTW: It is not true that FDA abhors PP analysis. FDA official position is that they will consider results from reasonable PP analysis as long as it is accompanied with ITT analysis. If my rotten memory serves me correctly FDA has approved at least one rheumatoid arthritis drug based mostly on results from PP analysis.

some sources
ICH
https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E9/Step4/E9_Guideline.pdf
FDA
https://www.fda.gov/downloads/Drugs/.../Guidances/UCM071665.pdf
SPIRIT 2013 Statement for Trial Protocols:
http://annals.org/aim/fullarticle/1556168/spirit-2013-statement-defining-standard-protocol-items-clinical-trials

interesting commentary at the end of a short article:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4936074/