Author + information
- Christopher M. O’Connor, MD, FACC, Editor-in-Chief, JACC: Heart Failure∗ ()
- ↵∗Address for correspondence:
Dr. Christopher M. O’Connor, Editor-in-Chief, JACC: Heart Failure, Heart House, 2400 N Street NW, Washington, DC 20037.
I was recently asked to present the following topic at a cardiovascular symposium: “Do We Need Phase II Clinical Trials in the Development of Important Therapeutics in Cardiology?” The presenters reviewed the program of LCZ696, a novel angiotensin receptor neprilysin inhibitor (ARNI) with outstanding phase III clinical trial results, and provided other examples of how drug development can be streamlined to shorten development times and costs, but more importantly, to bring life-saving therapies to our patients in a timely fashion. Traditional drug development has been advocated to proceed in a stepwise fashion including phase I: early exposure safety trials; phase II: early trials to assess treatment efficacy, exploratory examination of error rates, and selection of doses for further development; and phase III: large-scale controlled trials with clinical outcomes and confirmatory in nature, with error rates controlled to give definitive conclusions.
The PARADIGM trial (1) examined the novel therapeutic ARNI combination, which showed improved cardiovascular outcomes in patients with heart failure with reduced ejection fraction (HFrEF). This trial did not have a classical phase II study done in the HFrEF population. They used inferential data from a sister compound known as omapatrilat, in which a large clinical outcome study showed no significant efficacy on the primary outcome, but a secondary post-hoc analysis showed a promising signal. An adverse safety signal emerged, angioedema, that occurred in black patients, which was presumed to be due to the angiotensin-converting enzyme inhibitor component of the combination therapy; however, the overall adverse events favored the novel omapatrilat therapy. This encouraged the development of LCZ696, a novel direct ARNI, with the intent to eliminate the angioedema safety side effect but continue to advance the efficacy. The drug was explored in 1,320 patients with hypertension in a phase II study with multiple doses, and the study was also conducted in a preserved ejection fraction heart failure group looking at surrogate markers for efficacy.
Based on this inferential data, a large-scale clinical trial was conducted in over 8,000 patients, which resulted in a clear signal of efficacy without a detrimental safety signal. It is believed that this strategy of skipping over phase II, particularly in the HFrEF population, resulted in a reduction of approximately 3 years in drug development time. This, however, is not the first example of accelerating drug development. The development of the intravenous prostacyclin, epoprostenol, in advanced heart failure utilized a small study of 33 patients with severe heart failure to look at the feasibility of conducting a clinical trial, and then moved quickly to the FIRST (Flolan International Randomized Survival Trial), which was an outcomes study (2). This trial was stopped early after 450 patients were randomized because of an excess risk of harm and evidence of futility. Although this did not result in a new therapy for heart failure, it did stop the drug development program with <500 patients enrolled and saved millions of dollars to the program. How did this drug get to this level without a true phase II study? Again, they used inferential data from the pulmonary hypertension program and careful data safety monitoring in the early stages of the phase III study.
Another example of how phase II can be merged with phase III was the PURSUIT study of eptifibatide versus a placebo in acute coronary syndromes (3). This study was designed as a large outcomes study with 2 doses of therapy: 1.3 μg/kg/min versus 2.0 μg/kg/min versus placebo. This study was stopped in the lower-dose group after an independent data safety monitoring committee reviewed the safety data. This review provided that the higher dose had an acceptable safety profile, and therefore, after 3,200 patients had been randomly assigned to treatment groups, the committee recommended stopping the lower dose. This is an excellent example of dropping the “loser” in the early phase trials and doing phase II work in the phase III study. The overall results of the trial showed a positive efficacy signal on the end-point of death or nonfatal myocardial infarction at 30 days with the treatment.
Thus, one may ask “how do we move forward in this complicated world of drug development?” There are 3 types of strategies that go beyond the traditional approach:
1. The seamless phase II–III design, which refers to a design that combines the traditional phase IIB study and traditional phase III study into 1 trial.
2. An adaptive design, which refers to a design that uses accumulating data to modify aspects of the study to continue without undermining the validity and integrity of the trial.
3. Finally, the most progressive is the adaptive seamless phase II–III design, which refers to a design that uses data from before and after the adaption in the final analysis. The primary objective in the adaptive seamless design is to combine dose selection and efficacy into 1 trial.
There are multiple potential benefits to these designs, which include: more efficacy and dose information prior to triggering phase III by responsive adaptive treatment allocation and dropping treatment arms; reducing development timelines and cost by cutting out the wide space between phase II and III using the same operational infrastructure; more safety information and long-term patient exposure by allowing longer-term follow up of patients enrolled in the earlier stages; and a higher chance of patients within the trial to be treated with efficacious and safe doses by making timely use of available information in anticipation of benefit.
I think we can say that the hop, skip, and jump approach is, indeed, a strategy we should use so that we can move away from traditional models and accept adaptive, seamless designs in more of our clinical development programs.
- American College of Cardiology Foundation