Avanir’s STAR Trial: Offering Clarity to the FDA

| About: Avanir Pharmaceuticals, (AVNR)

In early 2006, Avanir (NASDAQ:AVNR) filed a drug application for Zenvia. It had filed a submission for Zenvia with the 30/30 dose combination. Two Phase III studies demonstrated safety and efficacy of Zenvia in PBA as well as 10 clinical trials. Priority review was received from the FDA as they acknowledged it’s a disorder with a high unmet medical need. In late 2006 AVNR received an approvable letter from the FDA.

The primary concern of the FDA was QT prolongation. The focus was not on the original therapeutic dose of Zenvia (that included 30mg quinidine) but the thorough QT study where the FDA concentrated on the supratherapeutic dose which was a 60/60 combination. The FDA saw a linear relationship in QT prolongation from the 60/60 combination. As the dose went from 30/30 BID to 60/60 BID there was almost a 2 times increase in QT prolongation. The FDA was concerned that with dosing errors, or drug to drug interactions, that many patients would have the QT risk associated with the 60/60 supratherapeutic dose.

In order to address this issue with the FDA Avanir agreed with the agency, via a Special Protocol Assessment, to reformulate the combination and reduce quinidine by 67% from 30mg in the original formulation down to 10mg in the new formulation. The FDA agreed upon the dose with Avanir for 30mg of dextromethorphan combined with 10mg of quinidine. That change reduced quinidine and thereby the cardiac risk that was associated with Zenvia. This also resulted in an improved the overall adverse event profile which was the 2nd area of concern by the FDA. Avanir spent four additional months with the FDA securing an SPA (Special Protocol Assessment) for the STAR trial.

This confirmatory Phase III trial called the STAR, agreed upon by the FDA, was a 12 week double-blind study followed by a 12 wk open label study. At the request of the agency they had 2 dosing arms. They had the 30/10 formulation which was the primary endpoint measured against placebo & the FDA also asked Avanir to explore a minimally effective dose. They added the 20/10 formulation as well. 20mg of dextromethorphan combined with 10mg of quinidine. Overall 326 pts enrolled in the study and 110 pts per arm. This was comprised mainly of ALS and MS (Multiple Sclerosis) patients. Approximately 197 ALS pts and 129 pts with MS. The primary endpoint was a reduction in episodes of laughing and crying relative to placebo.

The primary endpoint was very key from a regulatory perspective because the question the FDA wanted answered was:

1) What was the incremental benefit that Zenvia provided above and beyond placebo? With a majority of CNS disorders there is a high placebo effect so they wanted to demonstrate the incremental episode reduction that patients on Zenvia got above and beyond patients on placebo had.

Answer: Both doses performed extremely well w/ Zenvia 30/10 dose having almost a 47% reduction above and beyond placebo and the Zenvia 20/10 dose having almost a 50% incremental reduction above and beyond placebo with both achieving strong statistical significance. Starting at baseline Zenvia 30/10 placebo arms had about 33 episodes/week at baseline. Zenvia showed a sustained and quick response. By the 5th visit, Zenvia treatment arm had an 88% reduction in episodes vs. baseline. In the CNS arena there are very small differences often seen but with Zenvia you have a product with an 88% reduction vs baseline that shows a very significant demonstration of the benefit of Zenvia. What was prespecified in the protocol was to look at the remission defined as….”no episodes during the last 14 days of the study.” Half of the patients on Zenvia reported having zero episodes at the end of the study. This is also in lieu of the fact that the average patient was having 35 episodes/week at the start of the study but by near the conclusion of the study half reported no episodes.

The Secondary Endpoint was CNS lability scale. This is an instrument that measures the frequency and severity of PBA. Both the 30/10 and 20/10 doses had statistically significant differences vs. placebo in a double-blind study. At conclusion patients were given 14 days to decide whether or not to crossover into the open-label study. At that time 50% of the pts came off drug for an average of 3 days. The next step was to expose all pts to the 30/10 dose formulation. Another secondary endpoint was the SF36-Quality of life questionnaire. Zenvia had statistical significant benefit on the social functioning & on the mental health scale. This showed the benefit of episode reduction on the quality of life. The final secondary endpoint was moderate to severe pain secondary to MS. At the end of the study, there was a 1 pt difference on pain score which is strong because of the high unmet medical need for pain secondary to M.S. This data clearly showed that the efficacy was enhanced with the new 30/10 dose proposed by the FDA vs the original 30/30 dose in the STAR TRIAL.

Safety data for zenvia was clearly demonstrated in the STAR TRIAL and was extremely strong. There were basically no cardiovascular SAE’s. This is important because cardiovascular risks was a concern of the FDA. In the STAR Trial they took EKG’s at every office visit at peak plasma levels between 1.5 and 2.5 hrs after dosing. What was key is they did not see meaningful changes in the QT. There was no clinical or meaningful changes in QT, no pro-arrhythmic events, or no cardiovascular SAE’s observed with Zenvia in the STAR Trial. From an efficacy perspective from 30/30 to 30/10 there was no loss. By reducing the amount of quinidine from 30mg to 10mg the FDA believed they would see an improvement in safety.

Falls was also an area of concern for the FDA with respects to Zenvia. The FDA was concerned that patients with M.S. and ALS are already unstable as a result of their underlying diseases. They wanted to be confident that the addition of Zenvia did not lead to incremental falls for those patients. Across the study there was no difference in falls in these patient populations. There was also no difference in falls separately from underlying M.S. or ALS. There were also concerns of nausea, somnolence, and dizziness in the study but many of these symptoms went away as patients were in the study. Nausea went down dramatically & dizziness went down with the 30/10 formulation and 20/10 had a 10% incidence which is much lower than initially observed. Fatigue was also down dramatically. This resulted in a much lower discontinuation rate. What is very key to note is that on a 30/10 dose formulation that over 90% of the patients completed the study. That is a significant accomplishment given the large population in the study.

Conclusion: With the original application, Avanir had 800 pt subjects exposed. With the completion of the STAR trial, they now have over 1600 patients. So the amount of data that the FDA has to make a benefit/risk assessment is substantially higher. A majority of these patients were involved in the double-blind placebo controlled trials which allowed a much better assessment of a benefit/risk profile. Avanir believes that the risk/benefit profile was improved with Zenvia in the SPA designed STAR trial. The FDA was correct and the guidance that they gave Avanir with respect to Zenvia in putting together the new 30/10 formulation was solid. Come October 30th, Avanir now has a much safer product that is even more commercially viable and will be even more successful because of the reformulation. It seems the FDA did their job in providing the proper guidance with Zenvia and for Avanir Pharmaceuticals. Avanir should be very confident in their position with the FDA come October 30th.

Disclosure: Long AVNR