Evidence review

A literature search was conducted which identified 90 references (see search strategy for full details). These references were screened using their titles and abstracts and 4 references were obtained and assessed for relevance.

Two randomised controlled trials (RCTs) identified from the search (Singh D et al. 2016 [TRILOGY] and Vestbo J et al. 2017 [TRINITY]) were included in this evidence summary. A further study was published after the searches were undertaken and was also included (Papi A et al. 2018 [TRIBUTE]). A summary of the included studies is shown in table 2 (see evidence tables for full details).

Table 2 Summary of included studies

Study

Population

Intervention and comparison

Primary outcomes

Singh D et al. (2016) (TRILOGY)

RCT

159 sites in 14 countries

1,368 adults aged ≥40 years with COPD and

  • post-bronchodilator FEV1 <50%

  • FEV1/FVC <0.7

  • ≥1 moderate or severe exacerbationa in the last 12 months

  • symptomatic despite treatment with ICS/LABA, ICS/LAMA, LABA/LAMA or LAMA monotherapy.

Triple therapy with beclometasone/formoterol/ glycopyrronium bromide (100/6/12.5 micrograms) in a single metered dose inhaler (n=687)

versus

dual therapy with beclometasone/formoterol (100/6 micrograms) in a single matching metered dose inhaler (n=681b).

Both 2 puffs twice daily for 52 weeks.

Co-primary outcomes (assessed at week 26):

  • change from baseline in pre-dose (morning) FEV1

  • change from baseline in 2-hour post-dose FEV1 and

  • TDI focal score (change in dyspnoea severity from baseline.

Vestbo J et al. (2017) (TRINITY)

RCT

224 sites in 15 countries

2,691 adults aged ≥40 years with COPD and

  • post-bronchodilator FEV1 <50%

  • FEV1/FVC <0.7

  • ≥1 moderate or severe exacerbationa in the last 12 months

  • symptomatic despite treatment with ICS/LABA, ICS/LAMA, LABA/LAMA or LAMA monotherapy.

Fixed triple therapy with beclometasone/formoterol/ glycopyrronium bromide (100/6/12.5 micrograms) in a single metered dose inhaler plus a dummy dry powder inhaler (n=1,078b)

versus

tiotropium (18 micrograms) in a dry powder inhaler plus a dummy metered dose inhaler (n=1,075c)

and

open triple therapy with beclometasone/formoterol (100/6 micrograms) in a metered dose inhaler plus tiotropium (18 micrograms) in a dry powder inhaler (n=538b).

2 puffs twice daily for the metered dose inhalers and 1 puff daily for the dry powder inhalers for 52 weeks.

Moderate-to-severe COPD exacerbation frequency over 52 weeks of treatment for fixed triple therapy versus tiotropium.

Papi A et al. 2018 (TRIBUTE)

RCT

187 sites in 17 countries

1,532 adults aged ≥40 years with COPD and

  • post-bronchodilator FEV1 <50%

  • FEV1/FVC <0.7

  • ≥1 moderate or severe exacerbationa in the last 12 months

  • symptomatic despite treatment with ICS/LABA, ICS/LAMA, LABA/LAMA or LAMA monotherapy.

Triple therapy with beclometasone/formoterol/ glycopyrronium bromide (100/6/12.5 micrograms) in a single metered dose inhaler plus a dummy dry powder inhaler (n=764)

versus

dual therapy with indacaterol/glycopyrronium bromide (110/63 micrograms) in a single matching dry powder inhaler plus a dummy metered dose inhaler (n=768).

2 puffs twice daily for the metered dose inhalers and 1 puff daily for the dry powder inhalers for 52 weeks.

Moderate-to-severe COPD exacerbation frequency over 52 weeks of treatment.

a Defined as sustained worsening of respiratory symptoms that required treatment with systemic corticosteroids and/or antibiotics or need for hospitalisation.

b 1 person did not receive the allocated therapy.

c 1 person received the first dose of study medication, but withdrew consent before providing any post-baseline data, and is included in the safety population but not the efficacy population.

Abbreviations: COPD, chronic obstructive pulmonary disease; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 second; TDI, Transition Dyspnoea Index; RCT, randomised controlled trial.

The remaining 2 references were excluded. These are listed in excluded studies with reasons for their exclusion.

Clinical effectiveness

An overview of the results for clinical effectiveness can be found in the results tables.

Lung function

The TRILOGY study (Singh D et al. 2016) found that, at week 26, triple therapy with beclometasone/formoterol/glycopyrronium improved pre-dose forced expiratory volume in 1 second (FEV1; a co-primary outcome) by 0.081 litre more than dual therapy with beclometasone/formoterol (0.082 litre compared with 0.001 litre respectively; 95% confidence interval [CI] 0.052 litre to 0.109 litre; p<0.001). The European public assessment report for beclometasone/formoterol/glycopyrronium states that, although there is no consensus statement on what constitutes a minimally clinical important difference for pre-dose FEV1 for people with airflow obstruction, some published literature suggests a threshold of 0.100 litre for clinical trials may be appropriate. Also, a threshold for meaningful change is considered more meaningful for people with severe or very severe disease than for those with mild-to-moderate disease. In other words, small improvements in outcomes may be more noticeable for people with severe COPD than those with mild-to-moderate COPD. The Committee for Medicinal Products for Human Use (CHMP) concluded that, considering the severity of the condition of the target population, the improvement over baseline in pre-dose FEV1 of 0.081 litres for beclometasone/formoterol/glycopyrronium compared with beclometasone/formoterol is a meaningful benefit for this population who had severe COPD.

In TRILOGY, 2-hour post-dose FEV1 also improved more with triple therapy than with dual therapy at 26 weeks (a co-primary outcome), with a statistically significant difference between the groups (adjusted mean difference 0.117 litre, 95% CI 0.086 litre to 0.147 litre; p<0.001).

In responder analyses, pre-dose FEV1 improved by at least 0.100 litre in more participants using triple therapy compared with dual therapy at weeks 26 and 52 (for example, 38% compared with 23% respectively at 52 weeks; odds ratio [OR] 2.06, 95% CI 1.62 to 2.62; p<0.001).

In the TRINITY study (Vestbo J et al. 2017), a bigger improvement in pre-dose FEV1 (a secondary outcome) was seen with beclometasone/formoterol/glycopyrronium compared with tiotropium at week 52 (0.082 litre compared with 0.021 litre respectively; adjusted mean difference 0.061 litre, 95% CI 0.037 litre to 0.086 litre; p<0.0001) and the difference between the groups was statistically significant.

Fixed triple therapy with beclometasone/formoterol/glycopyrronium was found to be non-inferior to open triple therapy with beclometasone/formoterol and tiotropium (adjusted mean difference −0.003 litre, 95% CI −0.033 litre to 0.027 litre; p=0.85; the lower confidence interval is above the predefined non-inferiority margin of −0.050 litre).

Pre-dose FEV1 improved by at least 0.100 litre in more participants using triple therapy compared with tiotropium at weeks 26 and 52 (both p<0.0001; see the results tables for more details). No statistically significant differences were seen between fixed and open triple therapy at either time point (p=0.69 and p=0.63 respectively).

In the TRIBUTE study (Papi A et al. 2018), a statistically significant improvement in pre-dose FEV1 (a secondary outcome) was seen with beclometasone/formoterol/glycopyrronium compared with indacaterol/glycopyrronium at weeks 12 and 40 (both p<0.01), but not weeks 4, 26 and 52 (p values not reported). When averaged over 52 weeks, the mean improvement from baseline with the triple therapy compared with the dual therapy was 0.022 litre, which reached statistical significance (p<0.05).

Pre-dose FEV1 improved by at least 0.100 litre in around 20% of participants using beclometasone/formoterol/glycopyrronium and indacaterol/glycopyrronium at weeks 26 and 52, and there were no statistically significant differences between the groups (p=0.194 and p=0.198 respectively).

Dyspnoea

In the TRILOGY study, at week 26, there was no statistically significant difference between beclometasone/formoterol/glycopyrronium and beclometasone/formoterol in improvement in Transition Dyspnoea Index (TDI) focal scores (a co-primary outcome; p=0.160).

More than 50% of participants in each group reported clinically important improvements of at least 1 unit in TDI focal scores at week 26 and week 52. However, although there was a statistically significant difference between the groups at week 26 (57% compared with 52% respectively; OR 1.28, 95% CI 1.03 to 1.59; p=0.027), there was no significant difference at week 52 (p=0.430).

Dyspnoea was not investigated in the TRINITY or TRIBUTE studies.

Exacerbations of COPD

In the TRILOGY study, the rate of moderate-to-severe exacerbations was reduced by 23% over 52 weeks (annualised rate 0.1 exacerbation) with beclometasone/formoterol/glycopyrronium compared with beclometasone/formoterol (annualised rate 0.41 compared with 0.53 respectively; rate ratio [RR] 0.77, 95% CI 0.65 to 0.92; p=0.005).

Triple therapy prolonged the time to the first moderate-to-severe exacerbation compared with dual therapy (hazard ratio [HR] 0.80, 95% CI 0.67 to 0.97; p=0.020).

In the TRINITY study, beclometasone/formoterol/glycopyrronium reduced the rate of moderate-to-severe exacerbations by 20% over 52 weeks compared with tiotropium alone (the primary outcome; annualised rate 0.46 compared with 0.57 respectively; RR 0.80, 95% CI 0.69 to 0.92; p=0.0025). The European public assessment report for beclometasone/formoterol/glycopyrronium states that a 20% reduction in moderate-to-severe exacerbations is suggested to be a minimally clinically important difference in people with COPD. For this outcome, there was no statistically significant difference between fixed triple therapy with beclometasone/formoterol/glycopyrronium and open triple therapy with beclometasone/formoterol and tiotropium (p=0.89).

Fixed triple therapy prolonged the time to the first moderate-to-severe exacerbation compared with tiotropium (HR 0.84, 95% CI 0.72 to 0.97; p=0.0154). There was no statistically significant difference between fixed triple therapy and open triple therapy (p=0.57). Similar results were seen for the time to the first severe COPD exacerbation and time to the first moderate COPD exacerbation (see the results tables for more details).

In the TRIBUTE study, the rate of moderate-to-severe exacerbations (the primary outcome) was reduced by 15% over 52 weeks (annualised rate 0.1 exacerbation) with beclometasone/formoterol/glycopyrronium compared with indacaterol/glycopyrronium (annualised rate 0.50 compared with 0.59 respectively; RR 0.85, 95% CI 0.72 to 0.99; p=0.043). This reduction is not considered clinically important according to the criterion used in the European public assessment report (20% reduction).

There were no statistically significant differences between beclometasone/formoterol/glycopyrronium and indacaterol/glycopyrronium in the rate of moderate exacerbations or the rate of severe exacerbations (p=0.118 and p=0.189 respectively). However, the study was not statistically powered to compare the groups for these outcomes. There were also no statistically significant differences between the triple therapy and the dual therapy in the time to the first moderate-to-severe exacerbation or the time to the first severe exacerbation (p=0.219 and p=0.405 respectively).

Health-related quality of life

In the TRILOGY study, clinically relevant improvements of at least 4 units in St George's Respiratory Questionnaire (SGRQ) scores were reported with beclometasone/formoterol/glycopyrronium compared with beclometasone/formoterol, with statistically significant differences between the groups at weeks 26 and 52 (for example, 43% compared with 36% respectively at 52 weeks; OR 1.33, 95% CI 1.06 to 1.66; p=0.014).

In the TRINITY study, SGRQ scores improved by at least 4 units in more participants using triple therapy compared with tiotropium at weeks 26 and 52 (p=0.0024 and p=0.0019 respectively; see the results tables for more details). Fixed triple therapy was found to be better than open triple therapy at 26 weeks but not 52 weeks (p=0.0486 and p=0.37 respectively).

In the TRIBUTE study, SGRQ scores improved by at least 4 units in around 40% of participants using beclometasone/formoterol/glycopyrronium and indacaterol/glycopyrronium at weeks 26 and 52, and there were no statistically significant differences between the groups (p=0.255 and p=0.068 respectively).

Use of rescue medication

In the TRILOGY study, the use of rescue medication in puffs per day was significantly lower with beclometasone/formoterol/glycopyrronium compared with beclometasone/formoterol up to week 26 (for example, during weeks 13 to 26, −0.21 puffs compared with −0.02 puffs respectively; adjusted mean difference −0.19 puffs, 95% CI −0.35 puffs to −0.02 puffs; p=0.029). However, no significant difference was seen after this time point.

Participants using triple therapy had a higher proportion of days without rescue medication than those using dual therapy up to week 12 (for example, during weeks 5 to 12, 6.23% compared with 3.18% respectively; adjusted mean difference 3.05%, 95% CI 0.15% to 5.95%; p=0.039). No difference was seen after this time point.

In the TRINITY study, participants using triple therapy had fewer puffs of rescue medication per day compared with participants using tiotropium at all time points (all p<0.0001). No statistically significant differences were seen between fixed and open triple therapy at any time point (all p≥0.359).

Participants using triple therapy also had a higher proportion of days without rescue medication than those using tiotropium (all p<0.0001). No statistically significant differences were seen between fixed and open triple therapy at any time point (all p≥0.455).

In the TRIBUTE study, there was no statistically significant difference between beclometasone/formoterol/glycopyrronium and indacaterol/glycopyrronium in the average number of puffs of rescue medication (p=0.517) or the proportion of days without rescue medication (p=0.361).

Safety and tolerability

The European public assessment report states that, overall, the adverse effect profile of beclometasone/formoterol/glycopyrronium is well understood. None of the active substances are new and all have been used for many years, individually and in combination, for treating people with COPD of various grades of severity. The report concludes that the known adverse events and serious adverse events of the active substances were of a frequency and nature to be expected, and can all be considered reversible with change or modification of treatment.

In TRILOGY, a similar proportion of participants using beclometasone/formoterol/glycopyrronium and beclometasone/formoterol had treatment-emergent adverse events (54% compared with 56% respectively). Most of these were COPD-related (58% and 63% respectively), and most were mild or moderate in severity. About 3% of participants in each group had pneumonia, and about 2% of participants in each group had major adverse cardiovascular events. One treatment-related serious adverse event occurred (atrial fibrillation) in a person in the triple therapy group. This event resolved in 15 days and did not cause study medicine discontinuation.

Treatment-emergent adverse events resulted in death in a similar proportion of participants in the 2 groups (2%). None of the deaths were considered to be related to the study medicine. The number of treatment-emergent adverse events leading to discontinuation of treatment was also similar between the groups (5%).

In TRINITY, a similar proportion of participants using beclometasone/formoterol/glycopyrronium (fixed triple and open triple) and tiotropium had treatment-emergent adverse events (55%, 58% and 58% respectively). As in TRILOGY, most of these were COPD-related (59%, 54% and 62% respectively), and most were mild or moderate in severity. In each group, 2–3% of participants had pneumonia, and 1–2% of participants had major adverse cardiovascular events. One treatment-related serious adverse event was seen (angina pectoris) in the tiotropium group. Although the person recovered, the adverse event resulted in withdrawal from the study.

Treatment-emergent adverse events resulted in death in 1–3% of participants in the 3 groups. The most common treatment-emergent adverse events leading to death related to cardiac disorders or COPD exacerbations (European public assessment report). No deaths were considered related to study treatment. Fewer participants experienced adverse events leading to discontinuation of study treatment in the 2 triple therapy groups compared with the tiotropium group (both 3% compared with 6% respectively.

As in the other 2 studies, in TRIBUTE, a similar proportion of participants using beclometasone/formoterol/glycopyrronium and indacaterol/glycopyrronium had treatment-emergent adverse events (64% compared with 67% respectively). In both groups, 56% of these were COPD-related, and most were mild or moderate in severity. Pneumonia was seen in 4% of participants in each group, and adverse cardiovascular events were seen in 6% of participants using beclometasone/formoterol/glycopyrronium and 7% of participants using indacaterol/glycopyrronium. Two treatment-related serious adverse events were reported: dysuria in the triple therapy group and atrial fibrillation in the dual-therapy group.

In the beclometasone/formoterol/glycopyrronium group, 5% of participants had adverse events that led to discontinuation of treatment compared with 6% in the indacaterol/glycopyrronium group. The most common adverse events leading to treatment discontinuation were exacerbations of COPD. Adverse events resulted in 37 deaths across both groups, none of which were considered to be related to the study treatment.

According to the summary of product characteristics, the adverse events reported most frequently during the clinical development of beclometasone/formoterol/glycopyrronium were:

  • oral candidiasis (0.5%), which is normally associated with inhaled corticosteroids

  • muscle spasms (0.5%), which can be attributed to the long-acting beta-2agonist, and

  • dry mouth (0.5%), which is a typical anticholinergic effect.

An increase in the incidence of pneumonia, including pneumonia requiring hospitalisation, has been observed in people with COPD receiving ICS. Based on recommendations from the MHRA, the NICE guideline on COPD advises clinicians to be aware of the potential risk of developing adverse effects, including non-fatal pneumonia, in people with COPD treated with inhaled corticosteroids.

An overview of the results for safety and tolerability can be found in results tables.

Evidence strengths and limitations

TRILOGY and TRINITY were the key randomised controlled trials for licensing the beclometasone/formoterol/glycopyrronium triple-therapy inhaler. The European public assessment report states that their design and methodology was appropriate, and key outcomes were in line with guidance on undertaking studies in COPD. Randomisation and blinding were robust, statistical methods were acceptable, and the studies were of sufficient duration to measure exacerbation rate. The TRIBUTE study had a similar design and methodology.

All 3 studies included people with COPD with severe or very severe airflow limitation, symptoms despite treatment and a history of exacerbations who were current or ex-smokers. Just over half of the participants were aged less than 65 years, approximately three quarters were male, and most were of white ethnicity and had at least 1 concurrent disease. In both studies, the groups were comparable in terms of demography, lung function and symptoms. Most participants in the studies were using dual combination COPD therapy at study entry. In TRILOGY and TRINITY around 74% of participants were using ICS/LABA combinations compared with 61% in TRIBUTE. The European public assessment report notes that the participants enrolled in TRILOGY and TRINITY were representative of the target population for the therapeutic indication.

People with asthma, allergic rhinitis or atopy, clinically significant cardiovascular conditions or laboratory abnormalities, or unstable concurrent disease that might have affected the efficacy or safety of beclometasone/formoterol/glycopyrronium were not eligible to take part in TRILOGY or TRINITY, nor were people currently using triple therapy with an ICS, LABA and LAMA. Some of these populations were also excluded from TRIBUTE. It is unclear if the results of the studies apply to populations other than those recruited; for example, people with mild-to-moderate COPD for whom it is possible that any benefits of treatment might not outweigh the risks of adverse effects.

The TRILOGY study had 3 co-primary outcomes that were assessed according to a pre-specified hierarchy. The first 2 primary outcomes were disease-oriented and assessed changes in lung function. The third primary outcome assessed patient-oriented changes in dyspnoea using TDI score. Although TDI focal score improved more with beclometasone/formoterol/glycopyrronium than with beclometasone/formoterol, the difference between the groups was not statistically significant. According to the pre-specified success criterion, this means that TRILOGY failed to show that triple therapy was superior to dual therapy for this outcome.

The study authors stated that the need for participants to recall their previous symptoms is a potential problem when measuring TDI, and that this may have affected the result for this outcome, particularly as the study lasted 52 weeks. The authors of TRILOGY also observed a mean improvement in TDI focal score in the beclometasone/formoterol control group, which exceeded the minimum clinically important difference threshold, and an improvement in SGRQ total score which was close to the threshold, despite no change in treatment. They note that a trial effect on patient reported outcomes has been observed in previous COPD clinical trials.

Specialists involved in producing the evidence summary noted that triple therapy is generally considered only in people who have frequent exacerbations of COPD (more than 2 per year) despite dual inhaler therapy (for example a LABA/ICS or LABA/LAMA). At baseline, participants in the studies had, on average, 1 exacerbation per year, not 'frequent exacerbations'. The results of the studies may not apply to people who have 'frequent exacerbations'.

The exacerbation rate was reportedly lower than anticipated during all 3 studies during the 52-week follow-up, despite the requirement for participants to have a history of at least 1 exacerbation in the year before study entry. The authors of the studies consider that this might be because participants in interventional studies often receive improved care, with regular, detailed clinic visits. Also, treatment adherence in clinical trials is generally higher than that seen in clinical practice. Adherence was 94–95% in TRILOGY and TRINITY, and 98.5% in TRIBUTE.

According to the European public assessment report, evaluation of the safety of the beclometasone/formoterol/glycopyrronium triple inhaler was difficult because of the nature of the study populations. Participants were generally frail older people, with advanced lung disease and often other significant health problems.

TRILOGY compared triple therapy (beclometasone/formoterol/glycopyrronium) with dual therapy (beclometasone/formoterol) and TRINITY compared beclometasone/formoterol/glycopyrronium with tiotropium and open triple therapy with beclometasone/formoterol plus tiotropium. TRIBUTE compared beclometasone/formoterol/glycopyrronium with dual therapy (indacaterol/glycopyrronium). Therefore, it is not known how the efficacy and safety of fixed triple therapy compares with other treatments for COPD, such as other combinations of ICS/LABA (alone or with a separate LAMA) or other LABA/LAMAs.

It is also not known from the studies whether the fixed triple-therapy inhaler has any advantages over open triple therapy in terms of patient factors such as preference, adherence to treatment and ease of use of the device.

An overview of the quality assessment of each included study can be found in evidence tables.