Jim Kling

May 29, 2013

ISTANBUL, Turkey — For patients on dialysis with hyperparathyroidism, cinacalcet therapy might reduce all-cause mortality, according to a new analysis of the EVOLVE trial.

That study failed to achieve its end point, but researchers on the new analysis argue that the finding was flawed and does not reflect the drug's effect in actual clinical practice (N Engl J Med. 2012;367:2482-2494).

The randomized controlled trial might have been undermined by restrictive entry criteria and the experimental setting, said Jürgen Floege, MD, professor of nephrology at the University of Aachen in Germany.

Dr. Floege presented the new analysis here at the European Renal Association-European Dialysis and Transplant Association 50th Congress.

The results were met with skepticism. "Are we just trying to put a positive spin on the negative EVOLVE results?" asked Muhammad Magdi Yaqoob, MD, professor of nephrology at London University, United Kingdom, during the discussion period at the meeting.

Dr. Floege defended the analysis and said the original trial missed statistical significance for 2 key reasons.

"Many subjects left their assigned treatment arms. If you adjust for that, you see that patients have a 15% mortality reduction. The other problem was a very unfortunate chance finding that patients in the cinacalcet arm were a median of 1 year older than those in the placebo arm. That apparently happened by chance, but 1 year in a dialysis patient adds 3.5 % to mortality. So if you simply correct for age, the study is positive and shows a 12% reduction in mortality," Dr. Floege told Medscape Medical News.

 
Are we just trying to put a positive spin on the negative EVOLVE results?
 

The researchers conducted a propensity-score-matched study of cinacalcet to determine its effect in clinical practice. They then compared their findings with those from the EVOLVE trial.

Hyperparathyroidism and other mineral metabolism disorders might contribute to vascular calcification. Cinacalcet is a calcimimetic agent that has the potential to reduce all-cause mortality and cardiovascular events in these patients.

The researchers examined 10,488 incident adult hemodialysis patients participating in the prospective Analyzing Data, Recognizing Excellence and Optimizing Outcomes (ARO) observational cohort.

They estimated the propensity for receiving cinacalcet, and then matched patients treated with and without cinacalcet, regardless of anticipated future exposure, which the researchers say imitates an intention-to-treat analysis in a randomized controlled trial.

The team used matched Cox proportional-hazards regression analysis to determine the association between cinacalcet exposure and all-cause mortality. They also used lag-censored analysis and inverse probability of censoring analysis to determine the effect of patient crossover.

The study matched 532 patients treated with cinacalcet to 1790 not treated with cinacalcet. Of these, 29% were subsequently treated with cinacalcet. Vitamin D use was similar in the 2 groups. Patients were followed for a median of 1.92 person-years.

On 6-month lag-censoring and inverse probability of censoring analyses, the hazard ratios for all-cause mortality decreased. The estimates of treatment effect were similar to those in the analysis of all-cause mortality in the EVOLVE trial.

Table. Hazard Ratios for All-Cause Mortality in the ARO and EVOLVE Cohorts

Analysis ARO (95% Confidence Interval) EVOLVE (95% Confidence Interval)
Cox proportional-hazards regression 1.03 (0.78–1.35) 0.94 (0.85–1.04)
6-month lag-censored 0.84 (0.60–1.18) 0.83 (0.73–0.96)
Inverse probability of censoring 0.79 (0.56–1.11) 0.81 (0.65–1.02)

 

"We mirrored the problems and the outcomes of the EVOLVE study in real life," said Dr. Floege. "It's very important to control a randomized controlled trial in real life because quite different things can happen. Investigators tend to use younger and healthier patients in randomized controlled trials, which doesn't necessarily mirror the dialysis population. This is why I'm stressing that the ARO trial was conducted in real life," he added.

During the discussion period, Dr. Yaqoob noted that statistical analyses can become so complex that "even journal editors may struggle to understand them."

Others said they agree, including session moderator Christoph Wanner, MD, from the University Clinic Würzburg in Germany.

He told Medscape Medical News that "sometimes we try to squeeze datasets to find a positive result to please us because the randomized controlled trial, at least EVOLVE, was negative with respect to the primary end point."

This study was funded by Amgen. Dr. Floege reports being an advisor to Amgen and being on the executive committee of the EVOLVE study. Dr. Yaqoob and Dr. Wanner have disclosed no relevant financial relationships.

European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) 50th Congress: Presented May 20, 2013.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....