Cytokinetics: Clinical Investigators' Opinions On ATOMIC-AHF Trial Results

Introduction

Cytokinetics (NASDAQ:CYTK) issued a press release on the results of the Phase II trial of ATOMIC-AHF and then held a conference call on September 3. Before the results were discussed in the conference call, news wires reported that the trial had missed its primary endpoint and this caused an immediate sharp sell-off in the stock. It was well known among analysts who follow the stock that the trial was not likely to reach its primary endpoint and I wrote a report explaining why on August 26 called Cytokinetics: Previewing The Release Of Data From The ATOMIC-AHF Trial Of Omecamtiv Mecarbil.

In order to interpret the results, it is necessary to have details that were presented on the conference call. I was not surprised that the stock might initially trade down because of business wire reports, but the extent of the sell-off was more than I anticipated. My interpretation is that the details of the trial were better than my expectations. Pending the results of the ongoing COSMIC-HF trial in 1H, 2014, I believe that the ATOMIC-AHF results support moving omecamtiv mecarbil into Phase III. With success in the Phase III, omecamtiv mecarbil would likely be a several billion drug. I regard this as a buying opportunity.

I think that the best way of understanding the results of ATOMIC-AHF and their significance is to listen carefully to what clinical investigators in the trial said about the data. It doesn't really much matter what I think. It is the key opinion leaders and clinical investigators who are important. This note summarizes the views of the lead investigator on ATOMIC-AHF and another experienced investigator, both of whom spoke on the conference call.

As with any clinical trial there were positives and also issues that raise uncertainty. My read is that the investigators are very positive on omecamtiv mecarbil and believe that it should be studied in a Phase III trial, pending the results of the ongoing COSMIC-HF Phase II trial. Investors can make up their own minds after reading this report. An important event to watch for is the presentation of this same set of data at the Heart Failure Society of America on September 23. I think that the KOLs will have an opportunity to carefully review the data and opine on omecamtiv. I think that this could reverse the knee jerk negative reaction by investors on September 3.

Clinical Investigators for Omecamtiv

There were two clinical investigators who participated in the conference call that followed the release of ATOMIC-AHF data. Dr. John Teerlink was lead investigator on ATOMIC-AHF and Dr. John McMurray is also experienced with the drug. Both have been involved in the development and clinical research of omecamtiv mecarbil since the earliest clinical trials. Robert Blum, CEO of Cytokinetics, introduced Dr. Teerlink as knowing omecamtiv mecarbil as well or better than any other clinical investigator.

Reason for ATOMIC-AHF Trial

Dr. Teerlink said that prior clinical studies had clearly shown that in both healthy volunteers and in patients with stable chronic heart failure that omecamtiv mecarbil had shown increases in systolic ejection time. These had translated into improvements in heart performance as measured by increases in stroke volume, fractional shortening and improvements in ejection fraction.

The purpose of ATOMIC-HF was to see if similar results could be seen in a much sicker population of recently hospitalized heart failure patients. The main objective was to evaluate the safety, pharmacokinetics, pharmacodynamics, and efficacy of intravenous omecamtiv mecarbil in patients with acute heart failure.

Patients Enrolled in the Trial

The patients in the trial were all hospitalized for acute heart failure and were experiencing dyspnea (shortness of breath) despite the administration of an intravenous diuretic. Generally, diuretics are quite effective in these situations in relieving dyspnea so this indicates that these patients represented one of the sickest components of the heart failure population. There was a high prevalence of ischemic heart disease that made the patients at significant risk of heart attacks. Over half of the patients in the trial had troponin levels that were above the highest reference level for defining normal levels of troponin.

In this report, there is a great deal of discussion about troponin measurement so that it is important to understand what this means, Troponin is a complex of three regulatory proteins that are integral to cardiac muscle contraction. They are measured in the blood to differentiate between unstable angina and myocardial infarction (heart attack) in patients with chest pain or acute coronary syndrome. A patient who has suffered from a myocardial infarction has an area of damaged heart muscle and this results in elevated cardiac troponin levels in the blood.

Troponins are a marker of all heart muscle damage, not just myocardial infarction. Other conditions that directly or indirectly lead to heart muscle damage can also increase troponin levels such as severe tachycardia. Troponins are also increased in patients with heart failure, where they predict mortality and ventricular rhythm abnormalities.

ATOMIC-HF Results Missed Primary Endpoint or Did They?

The trial was conducted on three cohorts of patients using three different dosages targeted to achieve blood levels of 115, 230 and 320 ng/ml. The lowest dosage level was the minimum at which there were signs of physiological activity (although sub-therapeutic) and the highest dose was one which is close to that which will be used for Phase III. There were roughly 100 patients treated in each cohort with omecamtiv and 100 with placebo.

The statistical analysis was based on pooling all 300 or so omecamtiv and 300 or so placebo patients and did not achieve statistical significance on the endpoint. The primary endpoint which pertained to dyspnea was not reached.

However, in another prospectively defined statistical analysis, the 100 patients given omecamtiv in the third cohort were compared to the 100 placebo patients also dosed in that cohort. This is the probable Phase III dose and in this group the endpoint was reached. The response rate was 51% for omecamtiv and 41% for placebo. This was a 41% improvement in treatment effect and was statistically significant at a p value of 0.03. This indicates that when omecamtiv given at the dose that is likely to be used in Phase III and commercially, the dyspnea endpoint was achieved.

Dr. Teerlink also cited another analysis which looked at the relationship of increased blood level concentrations and dyspnea. This indicated that there was increasing dyspnea relief with increasing plasma levels of omecamtiv. This is consistent with the dyspnea relief seen in the cohort given the highest dose of omecamtiv.

Dr. Teerlink also emphasized the results in a secondary endpoint that he feels is perhaps more meaningful than dyspnea. The secondary endpoint was death or worsening heart failure within seven days of treatment. This showed a 46% decrease in the incidence of heart failure with a p value of 0.06, just missing statistical significance. This result was in the entire 600 patient trial, not just the high dose cohort.

Myocardial Infarctions in ATOMIC-AHF

As background, there have been numerous efforts to develop drugs that increase the blood volume output in heart failure patients. These drugs acted by increasing the contractility of the heart. However, in the badly damaged hearts of heart failure patients this can trigger arrhythmias or irregular heartbeats that can cause heart attacks and death. Despite this, two such drugs-digitalis and milrinone- are still widely used in heart failure because patients feel better on the drugs.

The factor that caught my attention from the first time I learned of omecamtiv was that it appeared to increase cardiac output without increasing the effort of the heart. It has a different and novel mechanism of action that causes the heart to contract longer and squeeze more blood into the arterial system without increasing the work load on the heart. This would be expected to reduce the incidence of arrhythmias and myocardial infarctions (MIs).

In this study, the incidence of supraventricular tachyarrhythmias was actually reduced by 50% in the omecamtiv mecarbil treated patients compared to placebo. Dr. Teerlink stated that MIs in the 600 patient trial were rare but that there were 3 in the roughly 300 placebo patients and seven in the roughly 300 omecamtiv patients; there were 2 in the low dose, 0 in the medium dose and 5 in the high dose. This numerical imbalance was not statistically significant, but the investigators stressed that this was something to be watched in future studies. There was also a modest increase in troponin levels in the omecamtiv patients.

Dr. Teerlink cited some reasons for not overly stressing the imbalance in MIs or the rise in troponin levels although he said that these are issues to be watched in future studies. In analyzing the modest increase in troponin levels, Dr. Teerlink and his co-investigators looked at the maximum change in troponin from when treatment was begun and compared this to omecamtiv plasma concentrations for the first 48 hours of administration. There was no correlation of troponin levels to omecamtiv plasma levels as measured by area under the curve. If omecamtiv were causing increased troponin, he would expect troponin to increase in line with area under the curve plasma concentrations of omecamtiv.

Dr. Teerlink noted that the pharmacology in ATOMIC-AH patients was similar to that seen in both healthy volunteers and stable heart failure patients. Importantly, he noted omecamtiv was able to demonstrate significant effects on heart rate and systolic blood pressure that are indicators of safety. The heart rate was decreased which means that the heart is working with less effort while the increase in blood pressure means that it is maintaining adequate blood supply. The reduced work load on the heart would be expected to reduce the potential for MIs. In Dr. Teerlink's judgment, these clinical outcomes are more predictive of the ability of omecamtiv to reduce mortality and hospital readmissions than dyspnea. Both Dr. Teerlink and Dr. McMurray stressed this point.

Dr. McMurray's Views on Omecamtiv

Dr. Teerlink was the principal investigator and presented the findings of the study. Dr. McMurray then followed with some general comments. He said that in looking at the results of the ATOMIC-AHF trial, there were several things that stood out. He stated that there was a trend toward dyspnea improvement, which he found interesting. However, he was more impressed by the results which suggested a trend toward improvement of the combined end point of death or worsening heart failure. This was defined as a secondary endpoint in the trial, but he noted that it as easily could have been defined as the primary endpoint with dyspnea as a secondary endpoint. He was also impressed by the reduction in arrhythmias and very significantly, the reduction in heart rate.

He believes that reduction in heart rate is critical. He has participated in numerous heart failure trials in his career. He has found that there is a uniform truth in both chronic and acute heart failure trials that an increase in heart rate is invariably associated with increased mortality. He said that almost all drugs that improve clinical outcomes in heart failure reduce the heart rate. He emphasized that the small decrease in heart rate was a "large and important" finding.

He said that there were safety issues to be looked at. There was a small rise in troponin levels and a trend toward increased troponin levels with higher doses. He questioned whether this was a finding that should be of concern or whether there might be another explanation such as chance. He felt that the analysis of area under the curve plasma concentrations of omecamtiv versus troponin levels suggested that it occurred by chance.

He concluded by saying that "as a Phase II trial this looks really as promising as I think that you can get at this stage in the development of drug for acute heart failure. I've looked at many other sets of Phase II data that haven't looked nearly as encouraging."

Question and Answer Session

After the presentations, there were several questions from the audience which added additional insight into the trial. These are listed below.

Q. Did the study meet its primary objective?

A. Dr. Teerlink answered that it certainly did. The objective of the trial was to determine how omecamtiv performed in patients with acute heart failure. There were encouraging signals of efficacy and safety that were needed to be answered before proceeding to a Phase III trial.

He said that the concern going into ATOMIC-AHF was that omecamtiv mecarbil would precipitate myocardial infarctions. This was why there was intensive troponin sampling that led to the largest data base of troponin values in acute heart failure patients in existence.

Q. What about the numerical imbalance of myocardial infarctions in the trial?

A. Dr. Teerlink said that there was a small and not statistically significant increase in myocardial infarctions and there is a small but potentially dose related changes in troponin levels. However, there was no correlation of troponin levels to omecamtiv plasma concentrations which is the most important and convincing data.

The myocardial infarctions must be carefully scrutinized but he believes that as investigators look carefully on a patient by patient basis they will become more comfortable that this is not a negative signal. He cited that some patients (he didn't state how many) that underwent percutaneous coronary interventions (PCIs) who had stable troponin levels before the PCI and increased troponin levels afterwards. These were classified as myocardial infarctions attributable to omecamtiv mecarbil, but might have been due to the PCI.

He also said that there is a data monitoring committee responsible for monitoring both the ATOMIC-AHF data and the COSMIC-HF data. They have reviewed all of the findings and read the case histories of each of the patients who were classified as having a myocardial infarction attributable to omecamtiv. They have allowed COSMIC-HF to continue which suggests that they are not alarmed by this issue.

Dr. McMurray said that there were not that many myocardial infarctions and that most occurred after infusion had been completed. He cited one case in which the MI occurred 22 days after the infusion. He noted that the half-life of omecamtiv is such that plasma concentration decreases 90% after four days. He questions whether this MI could in any way be linked to omecamtiv.

Q. What do you think of the data on dyspnea?

A. Dr. McMurray said that he is not inclined to put much emphasis on the dyspnea endpoint. He is more impressed by the way the drug works and there has been a consistent finding throughout the development program of omecamtiv that it can improve the amount of blood that is squeezed into the arterial tree, which is the fundamental problem to address in heart failure. It can do this without making the heart work harder as is indicated by the decrease in heart rate. He is also impressed by the overall tolerability.

Dr. Teerlink said that he agreed with Dr. McMurray and questions the ability of an improvement in dyspnea to predict long term outcomes in acute heart failure trials. He thinks that reduction in the incidence of worsening heart failure is a more important indicator. There was a 17% event rate for worsening heart failure for the 300 placebo patients versus 9% for the 100 patients in the high dose omecamtiv cohort. This was a 45% reduction in this event which he believes is the best predictor of reducing bad outcomes such as re-hospitalization and mortality. This signal was one of the more encouraging findings from the study.

Cytokinetics Take on ATOMIC-AHF

Cytokinetics' CEO Robert Blum said that Amgen and Cytokinetics have not yet made the decision to go forward into Phase III; however he said that Cytokinetics believes that the ATOMIC-AHF data supports continued progression and development. The ATOMIC-AHF results should be viewed together with the COSMIC-HF results in terms of how best to design a proper Phase III.

He stressed that this was clearly a Phase II study designed to ask and answer certain questions. In terms of this acutely ill heart failure patient population, which is a high risk and highly vulnerable group, the study was highly encouraging in terms of safety, tolerability, pharmacokinetics and pharmacodynamics that are important in planning a Phase III trial. He said that the next step will be to look at the COSMIC-HF data which they expect in 1H, 2014. This will provide the input to make the decision on whether to undertake a Phase III trial. Mr. Blum said that the results of ATOMIC-AHF, do support doing a Phase III trial

Amgen Statement on ATOMIC-AHF

An Amgen (NASDAQ:AMGN) representative was quoted in the Cytokinetics press release in which stated: "Although ATOMIC-AHF did not achieve its primary efficacy endpoint, we are encouraged by the data from this study," said Sean E. Harper, M.D., executive vice president of Research and Development at Amgen. "Omecamtiv mecarbil is a unique investigational therapy for patients with acute and chronic heart failure. We look forward to the data from the COSMIC-HF study, which together with the data from ATOMIC-AHF will inform our decision on whether to progress omecamtiv mecarbil into Phase 3 clinical trials."

Another way of gaining an insight into Amgen's view of the drug is a recent action that they took. On June 12, Amgen extended its collaboration with Cytokinetics on omecamtiv mecarbil to include Japan. This resulted in the upfront payment of $15 million and the purchase by Amgen of $10 million of CYTK stock. At the time, they had seen the data set on ATOMIC-AHF. It is hard to fathom Amgen taking this action if they thought that the data on ATOMIC-HF was negative.

Disclosure: I am long CYTK. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.