Intention to Treat
Cinacalcet, a calcimimetic agent, was approved for clinical use after randomized, controlled trials demonstrated its safety and efficacy in reducing parathyroid hormone (PTH) levels. Serum calcium and phosphorus levels also declined with the agent.
“We've been treating patients for elevated levels of parathyroid hormone for years and years and years without any data demonstrating that a treatment actually affects any hard clinical endpoints, so it's definitely the right time to conduct a study like this,” said Sharon M. Moe, MD, a member of the EVOLVE Executive Committee, in a phone interview. Dr. Moe is the Stuart A. Kleit Professor of Medicine and Division Director of Nephrology at Indiana University School of Medicine.
“I've been involved in both the KDOQI [National Kidney Foundation Kidney Disease Outcomes Quality Initiative] and the KDIGO [Kidney Disease: Improving Global Outcomes] guidelines, and, without hard endpoint studies, it's very hard to make firm recommendations in clinical practice guidelines.”
In the EVOLVE trial, which was funded by Amgen, 3,883 adults who had moderate-to-severe secondary hyperparathyroidism and were undergoing dialysis were randomized to receive cinacalcet or placebo between Aug. 22, 2006, and Jan. 31, 2008. Only 2.1% of patients were lost to follow-up in the trial.
“To my knowledge, this was the largest in number and longest in duration clinical trial ever conducted in the hemodialysis population,” Dr. Chertow said.
The trial included participants from the United States, Europe, Latin America, Russia, Australia, and Canada. The primary composite endpoint was time to death or first nonfatal cardiovascular event, which included myocardial infarction, hospitalization for unstable angina, heart failure, or peripheral vascular event. The use of phosphate binders, vitamin D sterols, calcium supplements, and other medications was left to the treating physician's discretion.
The median age of patients randomized to receive cinacalcet was 55.0, compared with 54.0 in the placebo group.
“The difference in age distribution, while small in an absolute amount, just missed statistical significance, and age in a dialysis population is very different than age in an ambulatory population,” Dr. Wetmore said. “Differences in ages really, really matter.
“The overwhelming likelihood is that had the age been similar between the two arms and had the randomization, which we count on to equalize the two arms, worked in terms of the age, I am somewhat confident that the primary endpoint would have barely met statistical significance. In short, I think it was really horrible luck that the ages differed between the two arms.”
A higher-than-expected rate of study-drug discontinuation posed another challenge for the trial. In the cinacalcet and placebo groups, 66.7% and 70.5% of patients, respectively, discontinued the study drug. One in five patients in the placebo group began receiving commercially available cinacalcet.
“In the intention-to-treat analysis, those 20% of patients in the placebo arm who also got cinacalcet were included in the analysis,” said W. Charles O'Neill, MD, Professor of Medicine at Emory University School of Medicine, in a phone interview. Dr. O'Neill, whose areas of expertise include vascular calcification in kidney disease, was not involved in the trial.
“That's my problem with the intention to treat: patients get included in the analysis whether or not they take the medicine. I think that's a flawed way to perform some of these studies. I would characterize the study as a positive study masquerading as a negative study.”
Reduction in Parathyroidectomy
Cinacalcet was shown in the trial to cut the rates of parathyroidectomy and severe, unremitting hyperparathyroidism in half. The latter endpoint was defined by the investigators after a review of blinded baseline data and the recognition that parathyroidectomy decisions were likely to differ across countries, clinical sites, and treating physicians.
“For me, the EVOLVE study strengthens the data supporting cinacalcet's use in the management of secondary hyperparathyroidism in ESRD [end-stage renal disease],” Dr. Quarles said. “Use of cinacalcet gives us an opportunity to reduce PTH to low levels and prevent parathyroidectomy.”
There was no statistically significant decrease in fracture, with fracture events adjudicated in 12% of patients in the cinacalcet group and 13% of patients in the placebo group.
Hypocalcemia was seven times more common in the cinacalcet group than in the placebo group, while nausea and vomiting were twice as common with cinacalcet. Adverse effects resulted in study-drug discontinuation for 18.1% of patients in the cinacalcet group and 13.0% of patients in the placebo group.
“Secondary analyses suggest some concerns about tolerability, with high risks of adverse events, such as nausea and vomiting and hypocalcemia, and also a small imbalance in the number of cancers between the two groups,” said Vlado Perkovic, MBBS, PhD, coauthor of the editorial that accompanied the study in the New England Journal of Medicine. Dr. Perkovic is Executive Director of the George Institute, Australia, and George Clinical, as well as Professor of Medicine at the University of Sydney.
He also noted the proportion of patients stopping or switching study treatment and the baseline imbalances across the groups.
“Importantly, both of those factors will also tend to underestimate any harmful effects of the drug, so that's important to keep in mind,” he said in a phone interview.
There is some degree of apprehension about how the absence of a statistically significant reduction in the primary composite endpoint of EVOLVE will affect cinacalcet use.
“I suspect what's going to happen is people will use this as a justification for using less cinacalcet,” said Geoffrey A. Block, MD, a member of the EVOLVE Executive Committee and Director of Clinical Research at Denver Nephrology, in a phone interview. “It gets wrapped up in a complex manner with 2014 and the inclusion of this drug in the bundle, and the fact that it's an expensive medicine.” Cinacalcet and other oral-only drugs are currently separately billable items in ESRD, but they are to become part of the bundled payment for outpatient dialysis services in January 2014.
“I can tell you it's had the opposite effect on my practice,” Dr. Block said. “My interpretation of the data, having spent a lot of time with it, is I am prescribing cinacalcet to all of my patients who meet the criteria to get cinacalcet, and we can reasonably find a way to pay for it.”
In addition to any potential influence on cinacalcet use, the EVOLVE trial holds lessons for the future conduct of clinical trials in the dialysis patient population.
“I think all of us on the steering committee were a little humbled by what happened with regard to the age,” Dr. Block said. “We tend to assume that randomization will take care of problems like this.
“In retrospect, we looked at some of the large randomized trials that have been conducted in dialysis, and you end up with an imbalance in age about eight percent of the time. Consequently, it seems clear that any primary analysis in a dialysis population, and probably a CKD population, needs to have age adjusted.
“I also happen to believe that while the global nature of EVOLVE is a strength overall, it did highlight for me personally some of the potential pitfalls of conducting trials in countries with very different practice patterns than the countries in which you powered your trial.
“Eastern Europe was quite a bit different than Western Europe or the US in this trial, and Latin America was quite a bit different in its characteristics than US and Western Europe, so I think those are important trial design issues we need to take into account.”
As cinacalcet was already an approved drug, its availability outside of the trial posed another complication.
“We learned that it's very difficult for patients and their providers to stand on the sidelines allowing patients to remain in long-term trials when drugs that are approved for use and commercially available can correct some of the laboratory or other intermediate outcomes,” Dr. Chertow said.
EVOLVE also had to be extended from 48 to 64 months because of a lower-than-anticipated event rate.
“I think one of the messages for dialysis trials is that we need big numbers and probably shorter follow-up times,” said David C. Wheeler, MD, a member of the EVOLVE Executive Committee and Reader in Nephrology at University College London, in a phone interview.
“You put dialysis patients into trials, and they basically become more and more ill as time goes by because of the very nature of the disease, and they accumulate comorbidities along the way. These patients are on so many other therapies, and sometimes the trial drug gets forgotten.
“I think this trial encourages us to do studies in dialysis patients. It is possible; we just need to get the design right.”
One way to improve clinical trial design in dialysis lies in the study population selected.
“In the context of all of these studies in patients on dialysis that use prevalent patients—a very heterogeneous group of patients with multiple comorbidities—really no studies have shown a positive result, mainly because, I think, their fatality rates are determined by the comorbidities,” said Kevin J. Martin, MB, BCh, Professor of Internal Medicine and Director of the Division of Nephrology at Saint Louis University School of Medicine, who was not a trial investigator but has published research on the agent.
“I think studies in dialysis patients might need a rethink as to how they should be approached and how to select the patients a little better so we can get some meaningful answers,” Dr. Martin said in a phone interview.
The results also support a move from single-intervention trials to those evaluating combination therapy.
“This and prior studies may suggest that a single silver bullet for dialysis patients is difficult to find,” said Kamyar Kalantar-Zadeh, MD, MPH, PhD, Professor and Chief of the Division of Nephrology at the University of California, Irvine, who was not involved in EVOLVE but has published research on cinacalcet, in a phone interview.
“Cinacalcet combined with other things should be considered as an intervention combination that could offer survival benefits and outcome benefits.”© 2012 Lippincott Williams & Wilkins, Inc.