Preliminary top-line results from XenoPort's XP21279 Phase 2 trial for advanced Parkinson’s disease

XenoPort, Inc. (Nasdaq:XNPT) announced today preliminary top-line results of a Phase 2, randomized, crossover clinical trial that compared optimized treatment with either Sinemet (immediate-release levodopa/carbidopa) or XP21279 co-formulated with carbidopa (279/CD) in advanced Parkinson's disease patients with motor fluctuations. 279/CD dosed three times per day reduced mean daily "off time" by 46% compared to baseline when the patients were taking their pre-trial Sinemet dosing regimen. However, in the primary analysis of the trial, the improvement with 279/CD was not statistically better than the improvement seen with optimized Sinemet dosed four or five times per day during the double-blind phase of the trial.

The trial enrolled subjects at 12 U.S. sites who were on a stable regimen of Sinemet dosed four or five times per day. Subjects were required to have "off time" in at least half of the inter-dose intervals between the first and last daily doses of Sinemet and an average daily "off time" greater than or equal to two hours during the three day baseline assessment period.

The trial consisted of an open-label, crossover optimization phase followed by a double-blind, crossover treatment phase. Thirty-five subjects entered the open-label phase of the trial, during which doses of Sinemet and 279/CD were each optimized for two weeks in a random order using the same protocol-specified guidelines. For Sinemet, doses were optimized while maintaining the same four or five times per day dosing frequency that the subject was taking during the baseline period. For 279/CD, doses were optimized using a fixed three times per day regimen. Qualified subjects then entered the double-blind phase, during which they received the optimized doses of Sinemet and 279/CD for two weeks each in random order.

The primary analysis was performed on the difference between Sinemet and 279/CD in the change from baseline in mean daily "off time" at the end of each period during the double-blind phase of the trial. The efficacy analysis included 28 subjects who completed the double-blind phase of the trial. The baseline mean daily "off time" for the analysis population was 6.4 hours. At the end of the open-label phase, mean daily "off time" was reduced from baseline by 2.0 hours for Sinemet compared to 3.4 hours for 279/CD. At the end of the double-blind phase, mean daily "off time" was reduced from baseline by 2.6 hours for Sinemet compared to 2.9 hours for 279/CD. The mean difference between Sinemet and 279/CD at the end of the double-blind phase of the trial was not statistically significant.

Robustness analysis included in the pre-specified statistical plan identified four subjects with outlier "off-time" values based on predefined criteria and evaluated the impact of these subjects on the trial results. Excluding these subjects, the baseline mean daily "off time" was 6.1 hours. At the end of the open-label phase, mean daily "off time" was reduced from baseline by 1.9 hours for Sinemet compared to 3.5 hours for 279/CD. At the end of the double-blind phase, mean daily "off time" was reduced from baseline by 2.4 hours for Sinemet compared to 3.3 hours for 279/CD. Excluding these outliers, the mean difference between Sinemet and 279/CD at the end of the double-blind phase of the trial was statistically significant (p<0.05).

All treatment-emergent adverse events were mild to moderate in severity. During the double-blind phase of the trial, dyskinesias were the most common adverse event. The incidence of new or worsening dyskinesias during the double-blind phase of the trial was 11% for Sinemet and 13% for 279/CD. There were no serious adverse events.

The trial employed a bi-layer tablet formulation containing both XP21279 and carbidopa. Pharmacokinetic analyses were performed on a subset of subjects during the open-label phase of the trial. The preliminary results indicate that the bi-layer tablet formulation provided a sustained levodopa exposure that was consistent with XenoPort's prior open-label trial that utilized separate extended-release XP21279 and commercially-available carbidopa tablets.

Ronald W. Barrett, Ph.D., chief executive officer of XenoPort, stated, "While we observed an impressive reduction from baseline in 'off time' for 279/CD dosed three times per day, we are disappointed that this trial did not provide a clear indication of the benefit, other than dosing frequency, of 279/CD for advanced Parkinson's disease patients compared to optimized doses of Sinemet. This may have been due to the large reduction from baseline in 'off time' observed once patients had their Sinemet dose optimized within the trial. Given that the goal of the trial was to demonstrate a clinical benefit over Sinemet, it was important that both treatments were equivalently optimized using specific guidelines. We intend to further analyze these data, including investigation of possible explanations for the outliers. We also plan to discuss these results with Parkinson's disease experts and possibly with regulatory authorities to determine next steps, if any. We will defer further investment in this program pending the outcome of our data analysis and these discussions."