The margin of victory in the IMPRESS-2 study on the PDQ bother score was even smaller, with a p value of 0.0496. If a patient sneezed at the wrong time, this portion of the study would have failed. Obviously, the marginal Peyronie's "bothersomness" improvement attributable to Xiaflex is a worrisome issue with respect to FDA approval. Auxilium needs to demonstrate a clinical benefit for Peyronie's patients beyond just some reduction in crooked penis angles. The company chose "bother" as a co-primary endpoint but results are far from a home run. I'm sure someone at the FDA is also asking Auxilium to explain why the other two components of the PDQ score were excluded from the primary efficacy analysis of the phase III studies. The two other components of the PDQ scale were included in the studies as secondary endpoints. Xiaflex failed to improve psychological and physical symptoms of Peyronie's compared to sham injections when the two studies were assessed individually. When Auxilium combined data from the two studies, this measure reached statistically significance. Xiaflex did not reduce penile pain compared to sham injection when the two studies were analyzed individually or combined. Before I finish up on the Xiaflex efficacy data, take another look at the two slides above. In the bottom left-hand corner of each, Auxilium notes the analyses were conducted on an "MITT" or modified intent-to-treat patient population. Uh oh. Take a look at this slide, which charts the patient disposition from the IMPRESS studies.
I'll do the math for you. Across both studies, 832 patients were enrolled and comprise the intent-to-treat (ITT) patient population. However, 220 of these patients, or 26 percent, were excluded from the primary endpoint analysis. Auxilium says a majority of these patients were excluded intentionally (and within the statistical rules of the studies) because they were sexually inactive within three months of baseline PDQ measurements. Since sexuality activity was a pre-requisite for PDQ scores, they couldn't be included in the primary analysis. Auxilium announced top-line results from the IMPRESS 1 and 2 studies in June 2012. At that time, the company's rationalization for excluding 26 percent of the enrolled patients from the primary efficacy analysis did not go over well with investors. One benefit of Xiaflex touted by Auxilium is the ability of men to resume sexual activity. If that's true, then why would the company exclude from its efficacy analysis men who were unable to have sex at study baseline but may have started to have sex after a year of Xiaflex therapy?