COPD: triple therapy decreases mortality

This study augments the initial IMPACT study on triple therapy for COPD previously reported, by collecting and reporting additional data (see copd triple therapy dec mortality amjrespcritcare2020 in dropbox, https://www.atsjournals.org/doi/10.1164/rccm.201911-2207OC or https://www.atsjournals.org/doi/pdf/10.1164/rccm.201911-2207OC 

Details (elaborated further in blog of 2018 publication (http://gmodestmedblogs.blogspot.com/2018/05/copd-improvement-with-triple-inhaler.html):

--10,355 symptomatic COPD patients were randomized to fluticasone/umeclidinium/vilanterol (FF/UMEC/VI) vs UMEC/VI vs FF/VI in patients with COPD at risk of future exacerbations [fluticasone is an inhaled corticosteroid (ICS), umeclidinium is a long-acting muscarinic antagonist (LAMA), and vilanterol is a long-acting b₂-agonist (LABA)]. these combos were also referred to in the studies as ICS/LAMA/LABA vs LAMA/LABA vs ICS/LABA, respectively)

    --of note, they randomized only 20% to the LAMA/LABA group and 40% to each of the other groups, not an even divide: not sure why (though this was a drug-company sponsored study), but that does limit the strength of some of their conclusions below 

--trial in 37 countries, from 2014-17

--age 65, 34% female, BMI 27, 65% former smoker, 45% had one mod-to-severe COPD exacerbation in past yr/43% 2 exacerbations/11% >2, post-bronchodilator FEV1= 46% of predicted, CAT score of 20. 

--baseline meds: 81% were on at least 2 of the long-acting meds (40% on three). 18% had bronchodilator reversibility (FEV1>12%, >200ml)

    -- one important aspect of this study: these were all symptomatic patients with at least one moderate or severe COPD excebation in the past year. So, all were on some meds regularly at baseline. For some already on triple therapy, their meds were decreased through randomization in the study to one of the double med groups (and be considered “stepped down”). those who were on triple therapy prior to the study, not surprisingly, had lower lung function and more COPD hospitalizations in the prior 12 months

-- 574 (5.5%) patients were censored in the original analysis owing to incomplete vital status information [hence this new article to include them and see if their inclusion modified the prior results]

-- time to all-cause mortality was prespecified in the study, subsequent analyses were post-hoc (the primary outcome in the original study was annual rate of moderate or severe COPD exacerbations, at 52 weeks)

-- this report includes information by intention-to-treat analysis in 99.6% of the 10,355 people included

-- all-cause mortality was considered to be "on-treatment" if the patient's death was up to 7 days after the last day of treatment; otherwise "off-treatment"

Results:

-- all-cause mortality (27 additional deaths found in this post-hoc collection of vital signs info):

    -- FF/UMEC/VI: 98 deaths (2.36%)

    -- FF/VI: 109 deaths (2.64%)

    -- UMEC/VI: 66 deaths (3.19%)

-- comparisons for all-cause mortality:

    -- FF/UMEC/VI (ICS/LAMA/LABA) vs UMEC/VI (LAMA/LABA): HR 0.72 (0.53-0.99),p=0.042

    -- FF/VI (ICS/LABA) vs UMEC/VI (LAMA/LABA): HR 0.82 (0.60-1.11), p=0.190

-- on-treatment all-cause mortality:

    -- FF/UMEC/VI: 50 deaths (1.20%)

    -- FF/VI: 49 deaths (1.19%)

    -- UMEC/VI: 39 deaths (1.88%)

-- comparisons for all-cause mortality for on-treatment:

    -- FF/UMEC/VI (ICS/LAMA/LABA) vs UMEC/VI (LAMA/LABA): HR 0.58 (0.38-0.88),p=0.011 [ie, 42% less]

    -- FF/UMEC/VI vs UMEC/VI (LAMA/LABA): HR 0.61 (0.40-0.93), p=0.022 [ie, 39% less]

-- timing to all-cause mortality (within 30, 60, or 180 d after randomization):

    --statistically significant benefit for FF/UMEC/VI vs UMEC/VI within 60 and 180 days of randomization (the text referred to Figures E3 and E4 for details, but these figures did not exist in the article or supplemental material)

-- lower rates of cardiovascular and respiratory death, and death associated with COPD (the text referred to non-existent Table E1 for details)

-- those who had been on triple therapy and randomized to continue the triple therapy had a trend to lower risk of death than if they were stepped down from triple to either of the dual therapies, though those stepping down to LAMA/LABA seemed to do even less well, vs ICS/LABA

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

 

Commentary:

-- the only real interventions that have been shown to decrease mortality have been smoking cessation, oxygen therapy in severely hypoxemia patients, lung volume reduction surgery in appropriate patients, and pulmonary rehab

-- the original IMPACT article published in 2018 found that patients with moderate or severe COPD exacerbations on triple therapy with FF/UMEC/VI (ICS-LAMA-LABA) had fewer COPD exacerbations than those on dual therapy with either the ICS-LABA or the LAMA-LABA , with 15-25% fewer moderate-to-severe exacerbations and 13-34% fewer severe ones. For severe COPD, there was a very strong trend for triple therapy being better than for those on ICS-LABA, and clear benefit over LAMA-LABA; all-cause mortality was also lower on triple therapy (though the original results were disputed because of the missing data, now supplied). for more analysis of this initial study, see http://gmodestmedblogs.blogspot.com/2018/05/copd-improvement-with-triple-inhaler.html . But, it should be noted that the conclusion of this article is a bit different from the take-away message: there really was no difference in all-cause mortality outcomes between the triple therapy group and the dual-therapy group on ICS. a few issues here:

    -- some prior studies using ICS did find a strong trend to a mortality benefit, but not quite statistically significant (eg the TORCH, SIMMIT, and INSPIRE studies)

    -- most of the patients in this IMPACT study were on baseline ICS (77%), though it was evident that those not on ICS overall did less well after randomization (see commentary in http://gmodestmedblogs.blogspot.com/2018/05/copd-improvement-with-triple-inhaler.html ): mortality was lower if they stayed on an ICS-containing regimen than switching to a LAMA/LABA one (ie, continuing the steroid was helpful; though the fewer patients on LAMA/LABA at baseline had no signficant mortality difference, ??related to smaller numbers of patients in that group??)

    -- and, those on triple therapy were sicker at baseline, suggesting that they had progressed to triple therapy because of lack of adequate control on dual therapy

    -- but, there were 50% more cases of pneumonia in those on ICS, independent of whether they were on dual or triple therapy

    -- in this symptomatic COPD group in the above IMPACT trial, people tended to do less well if they had been on triple therapy and stepped down to dual; this does run counter to the suggestion in some guidelines to attempt to step down therapy, as suggested by the authors. However, I would add that in patients who are well-controlled on any of the therapies, stepping down therapy might still be reasonable (and potentially decrease the higher risk of pneumonia in those on ICS-containing regimens), with the option of increasing back to triple therapy (or ICS-containing dual therapy) if the COPD were not so well controlled. This is an untested hypothesis, and the above study does not include this option: hence, unclear if this would change the all-cause mortality outcome.

-- since the original study, I have put several patients on this triple therapy combination, with excellent results. One of the advantages of vilanterol is that it is a true long-acting LABA but has a very short onset of action (as opposed to salmeterol), advantageous when a person tends to have poorer COPD control in the evening, for example, so they could dose themselves with a single inhalation then.

-- in terms of the optimal combination (my guess):

    -- budesonide or beclomethasone would be the preferred ICS over fluticasone, since the former have higher tissue to blood levels (ie, less systemic absorption than fluticasone)

    -- formoterol or vilanterol would be preferred over salmeterol, since the former have much faster onset of action (which not only helps right away, but might prevent the patient erroneously using extra salmeterol in an attempt to avert a more stubborn COPD exacerbation, potentially leading to high cumulative b-agonist levels and dangerous cardiac arrhythmias)

    --I am unaware of any significant clinical differences between the LAMAs

    -- for prior blogs discussing this further:

        -- for a review of the original IMPACT study, see http://gmodestmedblogs.blogspot.com/2018/05/copd-improvement-with-triple-inhaler.html

        -- the new (and vastly improved) GINA asthma guidelines, which incorporate the game-changing recent SYGMA studies (the last US guidelines were in 2007!!), and includes the use of formoterol as the preferred LABA, as was used in the SYGMA studies because of its rapid onset of action, and even safely used with multiple inhalations: see  http://gmodestmedblogs.blogspot.com/2019/10/new-and-improved-asthma-guidelines.html [my review found that vilanterol is very similar to formoterol, and may have an even faster onset of actions]

        -- for a review of the safety of LABAs in the SUMMIT trial: see http://gmodestmedblogs.blogspot.com/2016/05/copd-safety-of-labas.html

limitations of study:

--only 52 weeks long. hard to know if some long-term serious adverse effects of any of the incorporated meds or their combination might affect the results

--there was a misleading and incorrect mortality noted in their abstract on results, not repeated in the body of the text, and their references to supplemental tables and figures were not available (neither were listed in the supplemental material nor with hyperlinks). this sloppiness does shadow the conclusions a bit, since not possible to see some seemingly useful info (eg: their commment on decreasing cardiovascular and respiratory mortality: would be useful to see the numbers...)

so, this study does add to the short list of interventions that improve survival in those with moderate to severe COPD. This adds increasing impetus to move to triple therapy in those with resistant symptoms, and provides an added talking point with patients about the importance of adherence to medications since that might also lead to increase life expectancy. And, there have been substantial advances in the development of combination therapies (3 drugs in one) as a single pretty easy-to-do inhalation daily. but this study did not find mortality benefit in those on triple therapy vs ICS/LABA. this all  suggests that the following strategy (not adequatly tested in this study), might still be a reasonable one, perhaps in terms of exacerbations, but more clearly in terms of mortality outcomes:

    -- starting with a LAMA-based inhaler (as argued in the blog: http://gmodestmedblogs.blogspot.com/2018/05/copd-improvement-with-triple-inhaler.html )

    -- add a LABA, preferably one of the faster-acting ones (formoterol or vilanterol), and sparing the use of ICS with its attendant increased association with pneumonia

    -- then progressing to triple therapy, especially with one of the easy-to-use, once-a-day formulations (though one could argue there may be advantage switching from LAMA/LABA to ICS/LABA as an intermediate step)

    -- and, it seems likely based on the constellation of data and a few assumptions from combining data from different studies, there should be both better symptom control and decreased all-cause mortality from this orderly progression, depending on symptom control

    -- and, if the patient is stable for awhile (?how long), to occasionally attempt to step-down therapy ....

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list