Expert Financial Analysis and Reporting

Investment Thesis

In listening carefully to Cytokinetics’ (CYTK) comments over the last three months, I am convinced that the Company will consult with the FDA and EMEA about the design of a new phase 3 trial of tirasemtiv in ALS. They will ask for approval to use the respiratory measure of sustained vital capacity (SVC) as the primary endpoint.  They will almost certainly seek a Special Protocol Assessment from the FDA which acknowledges that SVC is an acceptable primary endpoint for a new phase 3 trial.

If the FDA agrees, CYTK can probably conduct a phase 3 trial with perhaps 700 or so patients and use the recently completed phase 2b BENEFIT-ALS trial as a supporting study. Based on the results of BENEFIT-ALS, the probability of success in a phase 3 trial using SVC as an endpoint is quite high in my opinion. The great uncertainty is whether regulators will agree with CYTK that SVC is an acceptable primary endpoint. This is after all a surrogate for respiratory function and not a clear clinical endpoint. This would also represent a sharp demarcation for the FDA which has accepted the ALSFRS-r scale as the primary endpoint for all recent ALS trials; this would be a difficult decision for the agency.

I am unsure as to what the FDA will decide. If they accept SVC as the primary endpoint, I think there is a high probability that tirasemtiv will succeed in the phase 3 trial. On the other hand, if the FDA insists on using ALSFRS-r as the primary endpoint, there is almost no chance that CYTK would launch another phase 3 trial.

I believe that there is very little expectation on the part of investors that tirasemtiv will proceed to a phase 3 study and be successful. Most of the $200 million valuation of CYTK is now based on the potential for omecamtiv mecarbil, CYTK’s drug for congestive heart failure, which could move into phase 3 trials in 2015. Hence, there seems to be not much risk if tirasemtiv is abandoned as weighed against very significant upside if it is ultimately successful in a phase 3 trial. This seems to be an asymmetric opportunity for investors.

As for potential upside if tirasemtiv is successful, I would site the examples of Acadia (ACAD) and Puma Biotechnology (PBYI) which I discuss in more detail shortly. Both companies are somewhat comparable in that they succeeded in phase 3 trials for drugs that had been regarded by investors as having no potential. In the case of Acadia this led to a $2 billion market capitalization and in the case of Puma to a $7 billion valuation. These are only indicative of what might happen to CYTK if it is successful in a phase 3 trial with tirasemtiv. With 49 million fully diluted shares outstanding each $1 billion of market value translates into $20 per share. The current market capitalization is $200 million and seems based solely on omecamtiv mecarbil’s prospects.

The first key inflection point for the stock is the decision of the FDA and EMEA on whether to accept SVC as an endpoint. If they agree, I could see the stock trading to the $5.00 to $6.00 range. We should know this by the end of 2014. If the agencies give thumbs down, I think that tirasemtiv will be abandoned and this could cause a 5% to 10% sell off in the stock as a guess.

The second and major inflection point for the stock would be success for tirasemtiv in a phase 3 trial in ALS using SVC as an endpoint. The trial could probably report out topline data in 2H, 2016 or 2017 and with success, which I think is probable, we could see a $1 billion or $2 billion market capitalization or $20 to $40 per share due to tirasemtiv alone at that time.

Investment Overview

Let me start out by saying that I have been around the biotechnology block and I recognize that one of the worst mistakes for companies can be in trying to resurrect a drug that failed in its initial clinical studies. A failure means that the hypotheses that management used to conduct the trial were wrong or that they failed to take key factors properly into account. History has shown that small biotechnology companies are reluctant to give up on a drug for both emotional reasons and because it might put management’s jobs at risk. As a result, they will often perpetuate development of a drug that is ultimately doomed to fail.

It is always a hard judgment for investors to make as to whether management has a reasonable basis for continuing development of a drug following disappointing trial results. Nevertheless, failure in a clinical trial does not always mean that the drug is ineffective. Trial design can sometimes be the cause of a trial failure that can be corrected and lead to success in subsequent trials. The recent experience with Acadia Pharmaceutical’s (ACAD) pimavanserin is an example of this. Two phase 3 attempts failed which caused the stock after the second failure to drop to $0.67 in November 2011. Then a well-designed phase 3 trial based on knowledge gained from the two previous trials resulted in a successful outcome and the stock is currently trading at about $21.

In a somewhat different, but instructive situation, Puma Biotechnology (PBYI) licensed a drug called neratinib from Pfizer; it is an oral small molecule HER2 inhibitor being studied in patients with selected types of breast and lung cancer. Pfizer conducted phase 2 trials involving over 3000 patients that while demonstrating safety and efficacy couldn’t identify a clear path forward for clinical development. There were also concerns about a 30% incidence of grade 3 diarrhea. A Pfizer internal program aimed at narrowing the product development focus led to a decision to out-license neratinib to Puma.

Puma narrowed the clinical focus of development and in the recently reported phase 3 ExteNET trial in breast cancer, it was demonstrated that patients who were given a maintenance monotherapy dose of neratinib following 12 months of Herceptin experienced a very meaningful improvement in disease free survival. Wall Street believes this is a blockbuster commercial opportunity. Moreover, using Imodium prophylaxis, Puma was able to reduce the incidence of grade 3 diarrhea to less than 10%. Following the announcement of this data on July 23, 2014 the stock jumped from $59 to $233.

The Acadia and Puma examples are obviously the rare exceptions. However, it does show that it is possible to use information gained in a failed trial to design a new trial that goes on to be successful. There may be a similar potential with Cytokinetics and tirasemtiv which failed to achieve its primary endpoint of reduction in rate of decline in the ALSFRS-r scale; this was in the phase 2b trial in ALS called BENEFIT-ALS. In fact, patients on tirasemtiv showed a slightly greater decline in ALSFRS-r than control patients although this did not reach statistical significance. Over a three month period the average decline in ALSFRS-r for tirasemtiv patients was 2.98 versus 2.40 for placebo. This scale measures the functional decline in ALS patients so that a lesser rate of decline is a favorable outcome.

Tirasemtiv also failed to show statistically significant improvement in all secondary endpoints except for one. It showed no effect on two measures of pulmonary function; sniff nasal inspiratory pressure (SNIP), and maximum voluntary ventilation (MVV). It also showed no effect on handgrip fatigue and muscle strength. The only statistically significant effect was on the secondary endpoint of sustained vital capacity (SVC). Not surprisingly, these disappointing results reported in April 2014 caused a sharp one day crash in the stock price from $13.00 to $4.60. Most investors wrote off tirasemtiv and attributed most of the investment value of the Company to omecamtiv, the congestive heart failure drug that could begin phase 3 trials in 2H, 2015.

As early on as April when the results of BENEFIT-ALS were announced, CYTK management strongly indicated that it did not believe that the trial results signaled the death of tirasemtiv. They believe that SVC trumps all other measures. Respiratory failure is the primary cause of death in ALS patients and slow vital capacity (SVC), which measures the amount of air that can be slowly exhaled after a deep breath, is the most widely used and relied upon measure of breathing function. ALS patients know their SVC scores like most people know their social security numbers. Survival of ALS patients is closely correlated with SVCand it is used for decision making on whether to put a patient on ventilatory assistance, to surgically intervene or to decide to put a patient in hospice.

Should Cytokinetics Advance Tirasemtiv to Phase 3 Based on SVC Results

In the phase 2b Benefit-ALS study, the rate of decline in SVC was measured for tirasemtiv in comparison to control. The results were strikingly positive as the rate of decline for tirasemtiv was only one-third that of control patients and the p value was a very strong p=0.0006.  Cytokinetics believes that tirasemtiv is the only drug that has ever demonstrated the ability to slow the decline in respiratory function in a large (over 200 patients) clinical trial.

The side effect profile of tirasemtiv may have played an important, negative role in the outcome of the trial as the drug was not well tolerated. Management was surprised at the extent and severity of side effects. They believe that through greater attention to control of side effects and by just encouraging patients to tough it out other efficacy measures than SVC can be improved. The belief is that most dropouts occur in the first month or so and if patients can get through that period, they are more likely to get through the trial and experience benefit.

There were 303 patients randomized to tirasemtiv in the trial and 73 dropped out due to side effects as compared to 302 control patients of whom only 12 dropped out because of side effects. In the tirasemtiv group 96.7% of patients reported adverse events versus 87.5% in the control group. The side effects that occurred most frequently for tirasemtiv were dizziness (50.8% of patients), fatigue (33.2%) and nausea (21.9%). Those side effects which occurred more frequently with tirasemtiv were dizziness (31.2% difference in incidence from placebo), fatigue (19.0%), nausea (14.1%), muscle spasm (9.5%), confusional state (9.9%), decreased appetite (6.9%), headache (6.8%), and insomnia (6.2%). An indicator of the impact of side effects on the trial is that it was difficult getting patients to the highest dose. In the clinical trials only 50% of patients on tirasemtiv could tolerate the highest dose of 250 mg BID.

The ALSFRS-r scale is essentially made up of 12 different factors that determine quality of life for an ALS patient. Each of these is evaluated on a scale of 1 to 4 by the patient or the physician to determine the final score. It is highly probable that the side effect profile of tirasemtiv resulted in patients feeling worse and negatively affecting the resultant ALSFRS-r score. There was also a negative effect of tirasemtiv in causing weight loss and in prior clinical trials of other drugs this has correlated with a poorer outcome. The study showed that weight loss at each week of the three month trial was twice as great on tirasemtiv as control. In addition to focusing on management of side effects, any potential future trial must also try to maintain the weight of patients. Prior studies have shown that weight loss has an incremental negative effect on ALS patients

Additional Data from BENEFIT-ALS Further Strengthens the Case for SVC

Cytokinetics first reported the results from BENEFIT-ALS at the American Academy of Neurology meeting in Philadelphia in April. Additional analysis was presented in June at the Joint Congress of European Neurology in Istanbul. Investigators presented data obtained at one and four weeks after the last dose of double-blind study medication. This showed that statistically significant smaller reduction from baseline in SVC in patients treated with tirasemtiv versus those on placebo persisted for at least four weeks following the last dose of double-blind study medication. This suggests a durable effect of tirasemtiv on SVC.

In July, at the 13th International Congress on Neuromuscular Diseases in Nice, France, analysis of the primary and secondary endpoints of specified sub groups of the patients enrolled in BENEFIT-ALS were reported. The authors concluded that tirasemtiv reduced the decline in SVC versus placebo by a similar magnitude regardless of age, sex, geographic region, and riluzole use, anatomic site of ALS onset, baseline pulmonary function, base line weight, and baseline body mass index. This further strengthens the case for tirasemtiv.

Tirasemtiv is the first drug candidate to consistently reduce the rate of decline in SVC in ALS patients regardless of age, sex, geographic region, riluzole use, site of disease onset, baseline pulmonary function and baseline weight and body mass index. The reduced decline in SVC versus placebo was statistically significant within each subgroup examined except patients enrolled in Europe, those with bulbar onset, and those with a percent predicted SVC less than the median at baseline, although, directionally, those groups still showed a similar magnitude of improvement on tirasemtiv versus placebo.

On a disappointing note, further analysis of BENEFIT-ALS did not show an improvement in three measures of breathing function that are part of the twelve components of the ALSFRS-r scale. These are orthopnea, dyspnea and use of mechanical ventilation. This makes SVC even more isolated as the one measure that showed a strong effect.

Why Was Efficacy Seen in SVC and Not Other Measures

It is puzzling and concerning to see such striking effects with SVC and no effect on other parameters, particularly those measuring pulmonary function. This is not something that I can assess independently so that I can only relate management’s thinking on this subject. Their consultants tell them that SVC is so regularly measured that there is great reproducibility from test to test which greatly increases its reliability. This is much less the case with SNIP and MVC so that they are more subject to variability. As CYTK discussed this with physicians they also found that physicians don’t focus much on SNIP and MVV. They regard SVC as the most important measure and place much less weight on SNIP and MVV.

The three respiratory measures included in ALSFRS-r are orthopnea, dyspnea and use of mechanical ventilation;  they also showed no improvement. Management believes that the brief 12 week duration of the trial was too short to glean much data. There just wasn’t much of a change observed either in placebo or tirasemtiv-treated patients from baseline over these 12 weeks. Management is convinced and the community of experts with whom they have consulted are convinced that SVC is the most important of any of the measures that were made.

What Key Opinion Leaders Think

Management believes that tirasemtiv had a clear and meaningful clinical effect on ALS. They emphasize that this is the first drug that has shown a positive effect in slowing the rate of decline in respiratory function. They also believe that the side effect profile and weight loss can be handled more effectively and keep more patients on the drug which should improve the clinical results.

Cytokinetics has consulted extensively with experts in neurology and pulmonology on data from BENEFIT-ALS. Their feedback provides support for management’s position that the effects of tirasemtiv on pulmonary function observed in BENEFIT-ALS are robust and potentially clinically meaningful. CYTK has conducted an extensive series of conversations not only with neuromuscular specialists but also with their counterpart pulmonary consultants. CEO Robert Blum stated in the 2Q, 2014 conference call that “there is unanimous support for these data (SVC results), meaning everyone with whom I’ve engaged in conversation around these data is indicating that they believe that these effects are real robust and could be predictive of clinical benefit and therefore should be the subject of further study in Phase III.”

Design of a New Phase 3 Trial

As I have previously stressed, CYTK will only conduct a phase 3 trial if the FDA and EMEA will accept SVC as the primary endpoint. If this agreement is reached in the second half of 2014, I could see a phase 3 trial beginning in early 2015. Management has not discussed in any detail what the design of the trial might be and the timelines for conducting the trial. It will probably be quite similar to BENEFIT-ALS so we can look to that trial to guess at how long the trial might take to conduct.

The BENEFIT-ALS trial was a randomized trial in ALS patients over a period of 12 week course of therapy. It began enrollment in November of 2012 and was scheduled to enroll up to 500 patients and complete enrollment in mid-2013. Topline data originally was expected to be presented in December 2013. However, an error in enrollment by the CRO conducting the trial caused a delay and resulted in increasing enrollment by 200 patients. This delayed the release of topline data until April 2014.

The phase 3 trial should be similar in size to BENEFIT-ALS. If the enrollment period was the same as 12 weeks used in BENEFIT-ALS and if the trial began in 1Q, 2015, we might see topline data in 1Q or 2Q, 2016. However, I suspect that the duration of treatment may be extended beyond 12 weeks so that topline data might not be available until early 2017.

Because the slowing in SVC was consistent across all sub-groups, I don’t see much of a change in inclusion and exclusion criteria from BENEFIT-ALS. Hence, the population studied in a new phase 3 would be much the same and this increases the probability of replicating the SVC results. Muscle strength may be dropped as a measured endpoint because the trial would be focusing on respiratory function. In BENEFIT-ALS, they had a number of patients excluded because they had excellent function in their hands and upper limbs but not in lower limbs and vice versa.

There will also be some thought given to titration of the dose. In BENEFIT-ALS, they gave all enrollees tirasemtiv in an open label design to start out. Those patients who could not tolerate the drug were then excluded and the remaining patients were randomized to tirasemtiv and control. Nevertheless, there were a sizable number of dropouts due to dizziness that occurred in the first four weeks after randomization. By extending the initial open label phase of the trial, they might be able to identify and exclude patients who cannot tolerate tirasemtiv at doses that are clinically effective.

Finance

Cytokinetics ended the second quarter with approximately $92 million in cash, cash equivalents and investments. Management guided that this represents approximately 20 to 22 months of going-forward cash burn based on our current financial guidance. This would take them into 1Q, 2016. However, this guidance doesn’t take into account a phase 3 trial for tirasemtiv. In order to complete a phase 3 trial of tirasemtiv, I think that the company will have to either partner the drug or raise significant amounts of capital. While management has not been explicit, I get the impression that they may look to partner tirasemtiv as a first option.