From The Heart.org
May 31, 2010 | Steve Stiles
Berlin, Germany – In the first clinical test of a gene therapy for heart failure, administration of a gene that upregulates an enzyme involved in myocardial contraction and relaxation appeared to improve symptoms, functional status, and ventricular volumes in patients with severe systolic heart failure. The small dose-finding study, reported here at the Heart Failure Congress 2010, sponsored by the Heart Failure Association of the European Society of Cardiology, is something of milestone and, its researchers say, paves the way for studying the novel treatment in larger clinical-outcomes trials.
But an unconventional broad-based primary end point in the randomized, placebo-controlled study, called Calcium Upregulation by Percutaneous Administration of Gene Therapy in Cardiac Disease (CUPID), along with seemingly inconsistent effects at different dosage levels, left some observers scratching their heads about what the trial actually shows.
The treatment entails delivery of a gene associated with an enzyme that is central to controlling the flow of calcium ions between the sarcoplasmic reticulum and cytoplasm, and so plays a key role in regulating myocardial contractility. Deficiency of the enzyme, called sarcoplasmic reticulum CA2+ ATPase (SERCA2a), often occurs in advanced heart failure and is considered a mechanism of progressive systolic and diastolic dysfunction. The SERCA2a gene is administered using an adeno-associated viral vector in a single intracoronary dose.
Celladon Corporation (La Jolla, CA), the technology’s developer, announced in April [1] that the phase 2 CUPID trial had “met the prospectively defined safety and efficacy end points,” which it described as a composite of worsening heart failure, frequency of and time to cardiac transplantation or left ventricular assist device (LVAD) implantation, change in patient ability to exercise, echocardiographic measures, heart failure symptom status, and natriuretic peptide levels.
But the study’s actual primary end point was a bit more complicated than that. As Dr Barry H Greenberg (University of California, San Diego) described in his presentation of CUPID, the primary end point called for SERCA2a gene therapy to demonstrate an array of symptomatic, functional, and clinical improvements, compared with placebo, that individually could reach what’s generally considered only a weak level of significance—p<0.2 for many of the measures. But demonstration of a sufficient number of improvements at such a level of significance could in aggregate point to a meaningful effect.
“What we were looking for in this clinical trial was a concordance between different end points, and we were able to show that,” Greenberg told heartwire. As he reported, the patients treated with the gene-based enzyme-replacement strategy at the highest tested dosage level over six months showed enough improvement in six-minute walk distance, NYHA functional class, LVEF, LV end-systolic volume, natriuretic peptide levels, and other parameters to satisfy the prospectively defined criteria for treatment efficacy.
When interviewed, Greenberg explained that this kind of end point, if prospectively defined, “is actually a fairly rigorous way of looking at small databases.” He said a “permutation analysis” suggested only a one-in-1000 probability that the observed array of separate outcomes in the study would occur by chance.
CUPID randomized 39 patients at 16 US centers with NYHA class 3-4 heart failure and an LVEF <35% to receive intracoronary delivery of the gene vector at a low dosage (6×1011 particles, n=8), mid-range dosage (3×1012 particles, n=8), or high dosage (1×1013 particles, n=9), or placebo (n=14), along with standard heart-failure medications. Nearly all of the patients were men; about half had heart failure of ischemic etiology.
The six-month results, Greenberg said, “argue for taking this [treatment] to a phase 3 trial. I don’t think we’re going to learn much more about it by doing further phase 2 studies.”
Dr Andreas M Zeiher (Johann Wolfgang Goethe University, Frankfurt, Germany), the featured discussant for Greenberg’s presentation, agreed that the next step should be “a larger efficacy trial with clinical end points, including death and worsening of heart failure,” with a more conventionally defined statistical analysis.
But Zeiher said it was “surprising” that no dose-response was observed for many of the outcome measures, particularly LVEF and parameters that reflect symptom status, such as NYHA functional class, six-minute walk distance, and quality-of-life scores.
Indeed, although the highest SERCA2a gene-therapy dosage was associated with improvements in many of the measured outcomes, compared with placebo, sometimes the best responses were seen at the lowest or mid-range dosage levels, and sometimes the highest dosage gave the lowest response.
Still, in what Zeiher called “the most important message of this pilot trial,” there was evidence of a dose-response effect for two parameters that reflect ventricular remodeling—specifically, natriuretic peptide levels, which rose substantially less over six months in the SERCA2a groups than in the placebo group (they didn’t rise at all in the high-dose group), and LV end-systolic volume, which rose sharply in the placebo group and declined in the high-dose SERCA2a group.
Dr John J McMurray (University of Glasgow, Scotland), who wasn’t involved in CUPID, expressed reservations about CUPID’s completely novel primary end point, which to its credit was prospectively defined but has never been used before and hasn’t been well validated. And, he said, the frequent lack of a dose response in the trial was “concerning.”
For a treatment like this, one would expect to see greater responses with increasing dosages, or at least similar responses to all three dosages “if you’re already at the top of the dose-response curve,” he said to heartwire.
But, “this random scattering of changes that bore no relationship to dose, that I could see, is problematic for me.”
But Greenberg was optimistic about the prospects for SERCA2a gene therapy in heart failure. If the strategy proves successful in larger, more definitive trials, “I think it would have an enormous impact. It would give us another tool in our toolkit for treating the advanced heart-failure population, which is a considerable population for whom the only existing therapies include placement of an LVAD and cardiac transplantation.”