COPD improvement with triple inhaler therapy

​Triple long-acting therapy is best for moderate-to-severe COPD (See copd daily triple therapy helps nejm2018 in dropbox, or DOI: 10.1056/NEJMoa1713901).

Details:

--10,355 patients with COPD: all symptomatic, with COPD Assessment score (CAT) of >10 (range 0-40 with higher being worse and the minimal clinically important difference being 2 units), and either FEV1 <50% of predicted and at least 1 moderate or severe COPD exacerbation in prior year, or FEV1 50-80% and at least 2 exacerbations.  88% completed the study

--trial in 37 countries, from 2014-17

--age 65, 34% female, BMI 27, 65% former smoker, 45% one mod-to-severe COPD exacerb in past yr/43% 2 exacerb, 11% >2, post-bronchodilator FEV1= 46% of predicted, CAT score of 20. 75% were on baseline at least 2 of the long-acting meds below. 18% had bronchodilator reversibility (FEV1>12%, >200ml)

--all patients had baseline CXR (used particularly to document any subsequent pneumonia)

--3 therapies compared, for 52 weeks, of once-daily:

    --triple therapy of fluticosone furoate 100mcg (inhaled corticosteroid, ICS) plus umeclinidium 62.5 mcg (long-acting muscarinic antagonist, LAMA) plus vilanterol 25mcg (long-acting b₂-agonist, LABA)

    --dual therapy of ICS plus LABA (above meds)

    --dual therapy of LAMA plus LABA (above meds)

--COPD exacerbation: mild (needs to increase prn albuterol), moderate (needs treatment with antibiotics or systemic steroids), or severe (leading to hospitalization or death)

--primary outcome: annual rate of moderate or severe COPD exacerbations, comparing triple therapy to each ICS-LABA and LAMA-LABA

--secondary outcomes:

    --trough FEV1 and St George’s Respiratory Questionnaire (SGRQ) for COPD (0-100, with lower scores reflecting better health quality, and 4 point difference being minimally clinically important)

    --annual rate of and time to first moderate or severe COPD exacerbation in patients with blood eosinophil counts >150

    --annual rate of severe exacerbations

--drug company-sponsored study, and lead author is employee of the company

Results:

--treatment was discontinued prematurely in 18% on triple therapy, 25% on ICS-LABA, and 27% on LAMA-LABA

--rate of moderate or severe COPD exacerbations:

    --triple therapy: 0.91 per year

    --dual ICS-LABA therapy: 1.07/yr; 15% less with triple therapy, RR 0.85 (0.80-0.90), p<0.001

    --dual LAMA-LABA therapy: 1.21/yr; 25% less with triple therapy, RR 0.75 (0.70-0.81), p<0.001

--rate of severe COPD exacerbations:

    --triple therapy: 0.13 per year

    --dual ICS-LABA therapy: 0.15/yr; almost significant 13% less with triple therapy, RR 0.87 (0.76-1.01), p=0.06 [note: confidence intervals are skewed so probability of benefit greatly exceeds that of harm]

    --dual LAMA-LABA therapy: 0.19/yr; 34% less with triple therapy, RR 0.66 (0.56-0.78), p<0.001

--trough FEV1, as change from baseline: triple vs ICS-LABA 97 ml (85-109, p<0.001) and 54 ml for triple vs LAMA-LABA (39-69, p<0.001)

--change in SGRQ score clinically significant for triple therapy (decreased 5.5 units), not so for the 2 dual-therapy groups (-3.7 for each)

--no difference in outcomes as stratified by eosinophil count between groups, though overall those with higher eos counts had fewer exacerbations

--rate of pneumonia: higher in both ICS groups (96/1000 patient-yrs for each, vs 61/1000 pt-yrs for LAMA-LABA)

--all-cause mortality was significantly lower in the ICS groups than the LAMA-LABA group (1% in the former but 2% in the latter), including both respiratory and cardiovascular deaths

Commentary:

--one advantage of the above therapies is that meds can now be given as a one-daily single-puff combination. A pharmacy call today found that this combo (Trelegy Ellipta) cost (AWP) is $637/month, significantly cheaper than the equivalent combo of Breo Ellipta ($395/month) plus either tiotropium ($467/mo) or umeclinidium​ ($393/mo)

--there had been (to me) pretty impressive mechanistic and clinical data suggesting that LAMAs were the preferred initial therapy for patients with more than minimally symptomatic COPD (ie, requiring only prn bronchodilators), given that it is topical and specific (inhaled, and the quaternary muscarinic blockers have minimal CNS effect anyway), directly improves bronchomotor tone (as a parasympathetic nervous system inhibitor, focusing on the vagus nerve), has been documented to be better than b-agonists, and has been shown to improve lung function and quality of life.The next step was using the combo of LAMA and LABA over adding an inhaled steroid inhaler, since there are documented benefits of LAMA/LABA and much lower risk of pneumonia found than with using ICS (as was also found in above study). and there have been some studies (eg the FLAME study: see copd LAMA LABA better than LABA steroid nejm2016 in dropbox, or DOI: 10.1056/NEJMoa1516385​) showing that LABA/LAMA was better than LABA/ICS (for review, see http://gmodestmedblogs.blogspot.com/2016/05/copd-superiority-of-laba-and-lama-combo.html  .)  The 2017 GOLD COPD recommendations (see copd gold guidelines 2017 in dropbox, or https://www.ncbi.nlm.nih.gov/pubmed/28128970 ) do seem to put ICS as the third choice (behind LAMA and LABA), though those with higher eosinophil counts seemed to derive more benefit from ICS than those with lower counts. the GOLD guidelines also do suggest the triple therapy approach in patients on 2 drugs with exacerbations, though they also support switching from LABA/ICS if on LABA/LAMA (this current study seems to support the more aggressive approach of putting all of these patients on triple therapy).​

--so, the poor performance of the LAMA-LABA group in the above study was a bit surprising (at least to me), with both the higher overall morality (also found in some but not all other studies, though i'm not sure how to explain this mechanistically for a topical anticholinergic) and less overall efficacy than ICS added to LABA (this was statistically significant). the authors suggest that the contrary efficacy results of the FLAME trial above were distorted because there was an important selection bias in that study.

--in terms of adverse events: the rate of pneumonia in the ICS groups was about 35 more per 1000 patient-years. However, the rate of moderate or severe COPD exacerbations was much less in those on triple therapy (923/1000 patient-years) and those on ICS-LABA (1052/1000 patient-years) vs those on LAMA-LABA (1148/1000 patient-years).

--one concern about even this large a study is the necessary combination of different baseline patients without the granular data to really tease out how these different individuals might benefit. For example, those already on triple therapy were unlikely to benefit if randomized to triple therapy and more likely to do worse stepping down to dual therapy; those on only one or 2 meds previously would more likely benefit. In addition, what is the effect of being on a med, then stopping it to be randomized to a different regimen? Any issues with rebound exacerbations, or subsequent decreased effectiveness if randomized to the same drug later?

So, this study seemed to be a large, well-designed one which strongly reinforced the clinical effectiveness of triple therapy in patients with moderate to severe COPD, in preventing both moderate and severe exacerbations as well as improving their COPD-related quality of life, with the expected downside of more clinical pneumonia. It suggests that we might have a more accelerated step-wise approach to putting patients on triple therapy . That being said, i always do have some angst about drug-company led and supported studies, both in terms of their reliability (this one does seem reasonable on the surface), the much increased hassle of getting prior approvals for these new drugs, and the stability of the drug costs (ie, this new combo is currently cheaper.  will that last?? especially if this drug becomes more popular based on this study??)  There may well be an added benefit for some patients to take a single inhaler once-a-day to achieve this benefit, though for many patients with symptomatic COPD, they seem to do just fine with the current formulations.