Mepolizumab for Eosinophilic COPD--Biologics Still Have Miles to Go

Andrew Shorr, MD, MPH


December 27, 2017

This is Andy Shorr from Washington, DC, with the pulmonary and critical care update. There has been a lot of interest in biologics, which are focused on the cytokines that modify eosinophil biology in asthma. We have seen a number of studies in the past 2 years and the approval of molecules that target the eosinophilic phenotype in asthma.[1] These studies show that these biologics are able to offer a reduction in exacerbation rates; they have steroid-sparing effects and generally improve the quality of life for our patients. The different molecules have different attributes in terms of pharmacokinetics, mechanisms, and dosing, but it's good to have a number of drugs to choose from.

At the same time, we've seen an evolution in our thoughts about chronic obstructive pulmonary disease (COPD) and a focus on an eosinophilic phenotype within COPD. This eosinophilic phenotype is thought to perhaps respond better to inhaled corticosteroids and is in many ways different from the noneosinophilic patients in the COPD population.

Two studies by Pavord and colleagues[2] were reported in the October 26, 2017, issue of the New England Journal of Medicine, focusing on mepolizumab, a cytokine monoclonal antibody focused on eosinophil biology. These two trials, METREX and METRIO, explored whether there is a role for these antieosinophil monoclonals to improve outcomes for patients with the eosinophilic phenotype within COPD.

The report of the two trials is very complicated to pull out and piece together. In METREX, they randomized patients with advanced COPD who were on triple therapy (GOLD group D), and who all had recent exacerbations, to either 100 mg of mepolizumab or placebo. Within this trial they stratified patients on the basis of an eosinophilic or noneosinophilic phenotype. They defined the eosinophil phenotype as an eosinophil count > 150/mm3 at the time of enrollment, or > 300 mm3 in the previous year. This was an all-comer population—both eosinophil and noneosinophilic phenotypes—with a prespecified stratification and randomization to 100 mg mepolizumab or placebo. The primary endpoint was the annual rate of exacerbations.

In the second trial (METREO), they took only patients with the eosinophilic phenotype and randomized them to one of three arms (placebo, 100 mg mepolizumab, or 300 mg mepolizumab.) For this trial, they enrolled only GOLD stage D patients and looked at the rate of exacerbations.

In METREX, they enrolled approximately 840 patients. Half of these patients had an eosinophilic COPD phenotype and half did not. When looking at the overall population, there was no difference in the annual rate of exacerbations. However, the primary endpoint was the population with the eosinophilic phenotype, and in this group, there was a significant reduction in exacerbations in patients who received the 100-mg dose of mepolizumab.

The investigators didn't see any other differences. There were no differences in the rate of emergency room visits, hospitalizations, lung function, or quality-of-life measures. By using mepolizumab, they were preventing corticosteroid and antibiotic exposure.

The METREO trial was a directed trial, such that all patients had the eosinophilic phenotype and were then randomized to placebo or to one of two doses of mepolizumab. They did not see a significant difference during the evaluation period in the rate of annual exacerbations. So we have one trial that might be positive and a second trial that was clearly negative.

As part of the analysis plan, the authors stated that they were going to do a prespecified meta-analysis at the patient level focusing on patients with eosinophilic phenotype. When they pooled the data for all eosinophilic phenotype patients, there was a signal indicating that in patients with the eosinophilic phenotype, antieosinophil therapy offers a benefit.

I have several concerns here. First, it's not clear that you should ever plan to take data from a negative trial and pool it with data from a positive trial, except purely for purposes of generating a hypothesis. We can't draw anything clinically meaningful from that kind of analysis. That analysis was very interesting in terms of hypothesis generation; it showed that the eosinophil count increased and the protective effect increased with drug exposure. There is clearly some connection between eosinophil measurement as a biomarker and exacerbation rates. Lowering the eosinophil count seems to have some protective effect.

We basically have a well-done double-blind, randomized controlled trial, stratified by phenotype, looking at the 100- and 300-mg dose or placebo but showing no signal. It's hard to dismiss—through a pooling mechanism—the negative findings of a well-done, 600-plus patient study. More important, the treatment effect was exacerbations, which are of unclear clinical significance. If you're not preventing severe exacerbations (although hard to show in a clinical trial because you need a much larger sample size) and are showing no other benefit in patient-reported outcomes, lung function, or quality of life, even if you find a difference, you have to wonder whether it's a distinction without a difference, and what is the benefit to the patient?

Although the authors do some important work here in suggesting a role for the eosinophils in the biology of COPD and COPD exacerbations, it's unclear that they have a very clean signal that even by inhibiting the eosinophil, they're actually improving outcomes. That is the true cause-and-effect nexus they want to demonstrate.

Realize that earlier trials of other monoclonals in this disease state, and other antieosinophil-directed monoclonals, have shown slightly different outcomes. Those were earlier and much smaller studies. From a clinician perspective, the use of these agents in COPD, certainly in the eosinophilic phenotype, is not yet proven to be of benefit. I don't think we should be progressing that way, given the expense of these agents, at least in the short term.

In terms of a question about the role for eosinophil biology, it's very thought-provoking, but in the words of Robert Frost, we have miles to go before we sleep. It comes back to the debate that's been raging in the literature for 40 years about the Dutch hypothesis and the real cause of COPD. Unfortunately, although these trials were very well done, and kudos to the investigators for doing this work and thinking about these agents as a way to get to this question, the data don't crystallize the issue one way or the other.


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: