Expert Financial Analysis and Reporting

Overview

Cytokinetics held a conference call on the results of the BENEFIT-ALS trial on the morning of April 30. The Company announced on April 25, that the trial had failed to meet its primary endpoint of change in ALSFRS-r and that results on secondary endpoints were mixed. Based on this, investors concluded that tirasemtiv was “dead” and that there was no path forward for the drug. This was also my initial impression.

At this price, I think that CYTK is a buy based solely on the prospects for omecamtiv mecarbil in congestive heart failure. However, I think that there is the potential for investors to regain excitement about the potential for tirasemtiv in ALS. This meeting put a much more positive light on the results of the trial and suggested a possible path forward, although the Company needs to do more analysis of BENEFIT-ALS results before talking about the potential for a new phase 3 trial(s).

Investors have learned from long experience that managements are very reluctant to give up on drugs even if they fail in clinical trials. Hence, there will be lots of skepticism on any decision on the part of CYTK to move this drug forward into phase 3. While this skepticism is generally warranted, there are some examples in which better understanding of the reasons for the failure of a trial enabled the company to design future trials that were successful. The recent example of the success of Acadia Pharmaceutical (ACAD) in conducting a successful phase 3 trial with pimavanserin after two successive phase 3 failures is a clear case in point.

My best judgment is that there is a high probability that tirasemtiv will progress to a phase 3 trial. This judgment is most influenced by what I heard from three key opinion leaders on the call who feel that this is an effective drug that can benefit ALS patients and should be moved into phase 3 development.  I would not have reached this conclusion if my only source of input were management. This is not to say that I distrust management (indeed I have great confidence in them), but I wanted to have input from physicians who have no financial interest in the success of tirasemtiv and are motivated by what they think will benefit ALS patients. However, I must caution that physicians can also develop positive biases for drugs that they are deeply involved with.

Views of Physicians Participating in BENEFIT-ALS

Three key opinion leaders spoke on the call. Dr. Jeremy M. Shefner, Professor and Chair, Department of Neurology at the Upstate Medical University, State University of New York was the Principal Investigator of BENEFIT-ALS. Dr. Robert G. Miller, Clinical Professor of Neurology and Neurological Sciences at Stanford University, and Director of the Forbes Norris ALS Research Center at the California Pacific Medical Center; and Dr. Merit E. Cudkowicz, Professor of Neurology, Harvard Medical School and Chief of Neurology Service at Massachusetts General Hospital (MGH) and Director of the ALS Clinic at MGH were physician participants in the trial

They emphasized that the trial did successfully meet statistical significance for the pre-specified secondary endpoint of Slow Vital Capacity (SVC) in which the p value was a very strong p=0.0006. SVC is a measure of the strength of the skeletal muscles responsible for breathing that has been shown to be an important predictor of disease progression and survival in prior trials of patients with ALS. This pre-specified secondary efficacy endpoint also declined less on tirasemtiv than on placebo at each assessment time point of 4, 5 and 12 weeks.

The analyses of other pre-specified secondary efficacy endpoints in BENEFIT-ALS produced mixed results. The Muscle Strength Mega-Score, a measure of strength combining the data from several muscle groups in each patient, declined more slowly on tirasemtiv versus placebo (p = 0.016 for the difference in slope of decline); however, there were no differences at any time point that reached statistical significance.

The rate of decline for Sniff Nasal Inspiratory Pressure (SNIP) was not statistically significantly different (p = 0.21); however, SNIP decreased more on tirasemtiv compared with placebo in a statistically significant manner at 4 and 12 weeks (p values at 4, 8, and 12 weeks were 0.012, 0.066, 0.050, respectively). No differences in Maximum Voluntary Ventilation and Hand Grip Fatigue were observed on tirasemtiv versus placebo.

The physicians feel that SCV is the best measure of respiratory function and hence the most important variable in determining whether patients can survive longer on tirasemtiv than placebo. Similarly they felt that Muscle Strength Mega-score was the best measure of improvement in muscle strength and these two variables were the most important factors to consider in judging whether this drug has clear biological activity.

They felt that gastrointestinal side effects associated with tirasemtiv probably affected the emotional and physical state of ALS patients so that it had a negative effect on the ALSFRS-r scale which measures a wide variety of activities that are important to quality of life. They also felt that two other secondary endpoints measured in the trial, SNIP and MVV, which did not reach statistical significance, were less important factors for them. In their opinions (it was unanimous), the improvement in SVC in particular and Muscle Strength Mega-score trumped all other measures in the trial.

They unanimously felt that this drug has clear biological activity and the ability to help their patients. They want to see the Company do another trial(s). The primary endpoint of a future phase 3 trial are likely to be SVC with a potential co-primary endpoint of Muscle Stregth Mega-score, not ALSFRS-r. I have included comments made by the three physicians who participated in the trial that give an insight into their thought processes.

Dr. Miller

His ALS patients live on average two to five years after diagnosis. The ALS community has seen nothing but trial failures over the 20 years since Riluzole was approved. Riluzole improved overall survival by 2 to 5 months in its phase 3 trials, but had no effect on a scale that was a forerunner to ALSFRS-r, had no effect on breathing and had no effect on muscle strength.

Physicians aren’t sure as to why Riluzole produces an effect; they don’t know the mechanism of action. With tirasemtiv, they see clear effect on improving respiration and strengthening limb muscles. This is as expected given that its mechanism of action is to improve muscle contractility.

In BENEFIT-ALS, tolerability of GI side effects which contributed to weight loss was a key issue. This obscured the ALSFRS-r results as it is well understood that patients who suffer weight loss perform more poorly on this scale as this affects quality of life.

No other drug ever tested has shown an improvement on the scale of what was seen with tirasemtiv in BENEFIT-ALS. It improves muscle strength and breathing. It is a very positive drug despite the weight loss issues that have dragged down results.

MVV as a secondary endpoint is new to the ALS community. It is a hard test to administer and interpret. The relationship of SNIP to disease is less clear than SVC. On the other hand, SVC is well established as a measure of breathing function and predictor of survival.

In the earlier phase 2a studies, he had two patients who showed dramatic improvement in muscle strength during the three week treatment period. One patient could not raise her arms from her side, but after tirasemtiv was able raise her hand. Another wheelchair bound patient was able to improve from a weak to a very strong handshake. He took this as an indication that tirasemtiv has a rapid and clear biological effect in some patients.

His patients who participated in BENEFIT-ALS would clearly participate in a new phase 3 trial. His patients were disappointed that they could not continue in the trial.

He believes that further studies should be done. He feels that SVC and Muscle Strength Mega-strength are very important endpoints with SVC being a predictor of survival. SVC could be the primary endpoint for a new phase 3 trial or potentially there could be a co-primary endpoint of SVC and muscle strength Mega-score.

Dr. Cudkowicz

This drug improves muscle strength. No other drug has ever done this.

Patients want access to drugs that may help them and will tolerate side effects if they think tirasemtiv will improve breathing and muscle strength.

The side effects with tirasemtiv are manageable. Patients will accept the side effect risks of tirasemtiv. Also, the side effects can be managed.

Just getting a few months more of life with some quality is meaningful to ALS patients.

She is excited about what has been learned with tirasemtiv and wants to move the drug forward to regulatory approval.

She is not too concerned about the failure to reach significance in the ALSFRS-r scale. This is a broad and crude scale.

Her patients would be eager to participate in a new phase 3 trial.

Dr. Schefner

This was a positive study and there should be a new study undertaken. Physicians can make the drug more tolerable through proper management of side effects and through better dose titration.

The endpoint of a potential phase 3 trial would have to be discussed with the FDA. However, he feels that the FDA is not set on ALSFRS-r as an endpoint and might accept SVC and Muscle Strength Mega-score as primary endpoints in a new phase 3.

The GI side effects were a surprise. They knew that dizziness was an issue, but in the three week phase 2a trials, they had not seen the problem with GI side effects that materialized during BENEFIT-ALS. The GI side effects are slower to materialize and are difficult to sort out from the many other ailments of ALS patients.