Treatable traits and the relevance of eosinophils in COPD

Tuesday, 18 Jun 2019

Dr. Konstantinos Kostikas

The limbic spoke with Associate Professor Konstantinos Kostikas, Head of the Respiratory Medicine Department of the University Hospital of Ioannina, Greece, to discuss the latest evidence on managing COPD, as well as some of the challenges that clinicians face. Associate Professor Kostikas was in Australia for the TSANZSRS 2019 Annual Scientific Meeting on the Gold Coast.

1. What are the biggest challenges in treating COPD at the moment?

One of the biggest challenges is we still have a lot of undiagnosed patients today despite the fact we have been implementing spirometry throughout the world for many, many years. A diagnosis of COPD requires exposure, mostly cigarette smoking, symptoms and spirometry. If you don’t have those elements, you don’t have an appropriate diagnosis. If you don’t have access to spirometry, technically you don’t have a diagnosis of COPD. So we still need to increase implementation of spirometry.

A second challenge is how to provide personalised management to patients because the easiest thing with COPD is to give them what we would call everything – long-acting bronchodilators and inhaled corticosteroids (ICS), and then treat exacerbations – but this is a generic management. We probably now have all the data we need to move to a more personalised management of COPD patients and this involves mainly the appropriate use of ICS to the patients who really need that and treat the rest of the patients effectively with long-acting bronchodilator in order to alleviate their symptoms and reduce exacerbations.

2. There’s a lot of talk about treatable traits in COPD. Can we and should we phenotype patients?

I think there is no way out of phenotyping patients now. We have almost all the data we need based on the recent studies of triple therapy and the LABA/LAMA combinations. We have a lot of data to understand which patients will benefit from what treatment based on the two main treatable traits that have been identified recently, including in the GOLD document. One is symptoms, predominately dyspnoea, and the other is exacerbations.

For patients with dyspnea without a history of frequent exacerbations, the appropriate treatment nowadays is the selection of the most appropriate long acting bronchodilator. Whether this is going to be one bronchodilator or two – so LABA or a LAMA or both of them – this depends on the level of dyspnea of the patients. We have data we’re presenting here [at TSANZ 2019] – a pooled analysis of the ARISE[1], SHINE[2] and SPARK[3] trials – which shows that in treatment-naive patients with COPD, dual bronchodilation is better than monotherapy with a LAMA in terms of improving lung function, reducing rescue medication and also improving symptoms. Which patients are the most appropriate to receive two bronchodilators? Probably the ones who are more dyspnoeic at baseline and we have a lot of patients with predominately emphysema and hyperinflation that will need dual bronchodilation from the beginning. Not only for the improvement in FEV1 but also for what we would call lung deflation – a term that is not very popular but I think is appropriate because we don’t want only to open the bronchi but we also want to reduce the air trapping and lung hyperinflation.

In the patients who have a history of exacerbations, the GOLD[4] recommended pathways are quite complex but there is a simple way out. If you have a patient who has high blood eosinophils (>300 cells/uL), this patient is highly likely to respond to an ICS. In patients with really low eosinophils (<150 cells/uL) those patients are highly unlikely to respond to ICS. And we have a group in the middle where clinical judgment is required.

3. What should the treatment plan be for people with moderate to severe COPD?

Nowadays in COPD we have decided not to use spirometry to make additional treatment decisions during the course of the disease. We classify the patients then we use treatable traits. I have to defend the GOLD statement here.

Moderate to severe COPD refers to airflow limitation so it goes to the old classification of GOLD stages 1, 2, 3 and 4. Moderate to severe COPD is an FEV1 of 30-80%. So for those patients, if they do exacerbate, you can consider an ICS if they have high blood eosinophils and if they don’t exacerbate and they don’t have high blood eosinophils, we shouldn’t consider ICS. And again, we have some patients in the middle where clinical judgement plays a role.

 4. Where do LABA/LAMA FDCs fit in the treatment of COPD patients?

This treatment is for the majority of COPD patients who have dyspnoea and is also the treatment choice for patients who exacerbate but do not have increased blood eosinophils. In these patients the next step might be to go for treatment options that don’t include ICS. Such options could be roflumilast, a PDE4 inhibitor, or low dose azithromycin as an immunomodulating drug, or even mucolytics in those with chronic sputum production.

5. Is there now a consensus on the use of ICS in COPD?

 We are close to a consensus for the extremes of COPD. Very low blood eosinophils? No ICS. Very high eosinophils and exacerbations? Yes to ICS. We still haven’t covered the middle grounds where clinical judgement will continue playing a role.

I would say that the additional treatable trait is a history of asthma. If you have asthma and blood eosinophils, the patients probably needs ICS. If you don’t have them, the patient probably doesn’t need ICS. But there is not a universal consensus on the cut points for blood eosinophils. We are getting closer but we are not there yet.

We probably need more research targeting patients in each specific blood eosinophil group because the majority of the data that we have are post-hoc analyses or secondary analyses of large clinical trials. We don’t have any data or any study that has used blood eosinophils as the treatment decision maker.

6. What has the SUNSET trial and subsequent analysis taught us about the use of ICS in managing COPD?

The SUNSET study[5] was designed to answer one specific question – what should we do with patients who are not frequently exacerbating and are on triple therapy? Patients on long-term triple therapy for at least six months, and had 0 or 1 exacerbations in the previous year and a FEV1 of 40-80% predicted – stable patients on triple therapy not frequently exacerbating – what do we do with them?

So we withdrew the ICS and technically we moved to a newer generation LABA/LAMA in that case. We started with a combination of salmeterol/fluticasone propionate plus tiotropium – we standardised for that and then we moved to either that or indacaterol/glycopyrronium. What we saw in that study was a difference of 26 ml between the two treatments in favour of triple therapy. The study did not meet the non-inferiority primary end point on FEV1.

However, in a pre-specified secondary analysis we saw that the patients who drove that effect were the patients with high blood eosinophils. These patients had a larger lung function loss. An important secondary end point of that study was exacerbations. There was no difference between the two groups however the patients with more than 300 blood eosinophils/uL, they had an 80% higher chance of exacerbating. The patients who exacerbated in the end were the ones who had consistently, in two consecutive measurements, more than 300 blood eosinophils.

This is an important message because that study tells us in which patients on long-term triple therapy, we can withdraw ICS. These patients are the ones who are not frequently exacerbating, and they had less than 300 blood eosinophils/uL at the time they were receiving triple therapy. In those patients, we can safely withdraw the ICS without any increase risk of exacerbation or loss of lung function. For about 70% of the population in the moderate to severe COPD patients, not frequently exacerbating and on long-term triple therapy, SUNSET trial we can withdraw the ICS safely.

7. Can you tell us about your research on ICS and the risk of diabetes?

We are presenting some real world evidence data from two large UK databases and in those analyses we were able to show that long-term high dose ICS use is related to an increased risk of new onset diabetes, and also with a dose-response manner.

A high dose of ICS ≥500 ug of fluticasone propionate daily is related to an increased risk of diabetes in that UK database. It increased the risk of incidence diabetes approximately 30%. A dose ≥1,000 ug/day lead to a 50% increased risk of diabetes.

We are quite confident that, correcting for confounding factors, this is a real effect because we did an appropriate case-control analysis with appropriate matching with a long follow-up of patients. I believe that these are concrete results – so there is an increased risk of new onset diabetes in patients who received high dose ICS for a long time.

I am concerned because these are elderly COPD patients with a lot of comorbidities and with a lot of diabetes predisposition. If we were talking about young patients with asthma, it would be a different thing. Nevertheless the long-term use of the ICS probably has some systemic effect and we can see that in the new onset of diabetes. It’s yet another reason to minimise unnecessary use of ICS.



  1. Asai K et al. Efficacy and safety of indacaterol/glycopyrronium in Japanese patients with COPD: Pooled analysis of SHINE and ARISE. Respir Investig. 2016; 54, 428-35.
  2. Bateman ED et al. Dual bronchodilation with QVA149 versus single bronchodilator therapy: the SHINE study. Eur Respir J. 2013; 42, 1484-94.
  3. Wedzicha JA et al. Analysis of chronic obstructive pulmonary disease exacerbations with the dual bronchodilator QVA149 compared with glycopyrronium and tiotropium (SPARK): a randomised, double-blind, parallel-group study. Lancet Resp Med. 2013; 1, 199-209.
  4. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Prevention, Diagnosis and management of COPD. 2019.
  5. Chapman KR et al. Long-Term Triple Therapy De-escalation to Indacaterol/Glycopyrronium in Patients with Chronic Obstructive Pulmonary Disease (SUNSET): A Randomized, Double-Blind, Triple-Dummy Clinical Trial. Am J Respir Crit Care Med. 2018: 198, 329–39.

Already a member?

Login to keep reading.

Email me a login link

© 2022 the limbic