Evidence review

This evidence review is based on 2 phase III studies with similar designs: GLOW1 and GLOW2 (see tables 2 and 3).

GLOW1( D'Urzo et al. 2011 ) and GLOW2 ( Kerwin et al. 2012 )

  • Design: randomised, double-blind, placebo-controlled trials.

  • Population: adults aged 40 years and over with stable moderate to severe COPD (post-bronchodilator FEV1 30–79% predicted and FEV1/FVC less than 0.7) and a smoking history of at least 10 pack-years. Patients with a history of asthma were excluded, but no limits on airway reversibility were used. GLOW1 and GLOW2 included 822 and 1066 patients respectively. Long-acting bronchodilators were stopped but ICS and short-acting beta2 agonists could be continued throughout the study. Approximately half of the patients in each group were using ICS at baseline, 59.6% were ex-smokers, and the overall mean smoking history was 47.6 pack-years. The mean post-bronchodilator % predicted FEV1 was 55.5%[5].

  • Intervention and comparison: following a screening and 2 week run-in period, patients were randomised to glycopyrronium bromide 50 micrograms once daily or placebo using a single-dose dry powder inhaler (Breezhaler). GLOW2 also included an open-label tiotropium arm (18 micrograms daily delivered using the Handihaler). The duration of treatment was 26 weeks in GLOW1 and 52 weeks in GLOW2.

  • Outcome: the primary outcome measure was trough FEV1 at week 12. Key secondary outcomes were breathlessness measured using the Transition Dyspnoea Index (TDI; week 26 in both studies) and health status measured using the St George's Respiratory Questionnaire (SGRQ; week 26 in GLOW1 and week 52 in GLOW2).

Table 2 Summary of the GLOW1 study: D'Urzo et al. 2011 [6]

Glycopyrronium bromide

50 micrograms

Placebo

Analysis

Randomised

n=552

n=270

Efficacy (full analysis population)a

n=534

n=260

Primary outcome: trough FEV1 at week 12 (LS mean [±SE])

1408±105 L

1301±137 L

Difference 108±14.8 ml (p<0.001)

Selected secondary outcomes:

TDI focal score at week 26 (LS mean [±SE])

1.84±0.257

0.80±0.294

Difference 1.04±0.235 (p<0.001)

SGRQ score at week 26 (LS mean [±SE])

39.50±0.813

42.31±0.992

Difference
–2.81±0.961 (p=0.004)

Safety b

n=550

n=267

Patients reporting serious adverse events

8.4% (46/550)

9.0% (24/267)

Significance of difference vs. placebo not given

Discontinuations due to adverse events

5.8% (32/550)

7.1% (19/267)

Significance of difference vs. placebo not given

Abbreviations: FEV1, forced expired volume in 1 second; LS, least square; SE, standard error; SGRQ, St. George's Respiratory Questionnaire; TDI, Transition Dyspnoea Index.

a All randomised patients who took at least 1 dose of study treatment.

b Patients who took 1 dose of study drug, regardless of whether they were randomised.

Table 3 Summary of the GLOW2 study: Kerwin et al. 2012 [7]

Glycopyrronium bromide

50 micrograms

Placebo

Tiotropium 18 micrograms

Analysis

Randomised

n=529

n=269

n=268

Efficacy (full analysis population)a

n=525

n=268

n=267

Primary outcome: trough FEV1 at week 12 (LS mean [±SE])

1469±141 mlb

1372±173 mlb

1455±170 mlb

Glycopyrronium bromide vs. placebo

Difference 97±16.7 ml (p<0.001)

Tiotropium vs. placebo

Difference 83±19.3 ml (p<0.001)

Glycopyrronium bromide vs. tiotropium

No significant difference

Selected secondary outcomes:

TDI focal score at week 26 (LS mean [±SE])

2.13±0.240b

1.32±0.289b

2.26±0.281b

Glycopyrronium bromide vs. placebo

Difference 0.81±0.260 (p=0.002)b

Tiotropium vs. placebo

Difference 0.94±0.297 (p=0.002)b

Glycopyrronium bromide vs. tiotropium

No significant difference

SGRQ score at week 52 (LS mean [±SE])

40.85±0.854b

44.16±1.040b

41.32±1.024b

Glycopyrronium bromide vs. placebo

Difference -3.32±1.004 (p<0.001)b

Tiotropium vs. placebo

Difference
–2.84±1.155 (p=0.014)b

Glycopyrronium bromide vs. tiotropium

No significant difference

Safety c

n=525

n=268

n=267

Patients reporting serious adverse events

12.6% (66/525)

16.0% (43/268)

15.4% (41/267)

Significance of difference vs. placebo/tiotropium not given

Discontinuations due to adverse events

8.0% (42/525)

11.6% (31/268)

7.5% (20/267)

Significance of difference vs. placebo/tiotropium not given

Abbreviations: FEV1, forced expired volume in 1 second; LS, least square; SE, standard error; SGRQ, St. George's Respiratory Questionnaire; TDI, Transition Dyspnoea Index.

a All randomised patients who took at least 1 dose of study treatment.

b European Medicines Agency (2012) European public assessment report: Seebri Breezhaler

c Patients who took 1 dose of study drug, regardless of whether they were randomised.

Clinical effectiveness

The results of both studies are consistent and show statistically significant improvements in most outcome measures with glycopyrronium bromide, compared with placebo. The absolute difference in least square mean 12-week trough FEV1 was 108 ml in GLOW1 and 97 ml in GLOW2 with glycopyrronium bromide compared with placebo (p<0.001 in both studies). These differences are around the value which is considered to be the minimum clinically important improvement in trough FEV1 (100 ml)[8]. In GLOW2, at week 12 the increase in trough FEV1 compared with placebo was 83 ml for tiotropium.

Statistically significant improvements in breathlessness and health status were seen with glycopyrronium bromide compared with placebo in both studies. However, the differences were not all clinically significant. At week 26, the difference in absolute TDI focal score value was 1.04 in GLOW1 and 0.81 in GLOW2 for glycopyrronium bromide compared with placebo. For tiotropium the difference in absolute TDI focal score after 26 weeks was 0.94 compared with placebo in GLOW2. The minimum clinically important improvement in TDI score (on a scale that ranges from –9 to 9) is considered to be 1[5].

The difference between glycopyrronium bromide and placebo in SGRQ total score was –2.81 at week 26 in GLOW1 and –3.32 at week 52 in GLOW2. In GLOW2, the difference between tiotropium and placebo was -2.84 at week 52. The minimum clinically important improvement in SGRQ total score (on a 100-point scale) is –4[5].

In GLOW2, there was no statistically significant difference between tiotropium and glycopyrronium bromide for any of these outcomes.

Both studies showed a statistically significant decrease in the risk of moderate or severe COPD exacerbations in terms of time to first exacerbation in patients treated with glycopyrronium bromide compared with placebo (GLOW1: hazard ratio [HR] 0.69, 95% confidence interval [CI] 0.50 to 0.95, p=0.023; GLOW2: HR 0.66, 95% CI 0.52 to 0.85, p=0.001). Improvements were also seen with glycopyrronium bromide in terms of use of rescue medication in both studies (difference compared with placebo GLOW1: 0.46 puffs per day, p=0.005; GLOW2: 0.37 puffs per day, p=0.039).

Safety

The percentage of patients reporting adverse events was similar between groups in both studies. Anticholinergic adverse events were most commonly seen. However, for most of these the incidences in excess of placebo were low. According to the summary of product characteristics, the most common anticholinergic adverse reaction reported with glycopyrronium bromide in studies was dry mouth (2.4%). Gastrointestinal effects including gastroenteritis and dyspepsia were also observed, as were adverse reactions related to local tolerability such as throat irritation, nasopharyngitis, rhinitis and sinusitis[9].

In their assessment of glycopyrronium bromide, the European Medicines Agency (EMA) found that atrial fibrillation was reported more frequently with glycopyrronium bromide compared with placebo. However, when they considered adjudicated electrocardiogram (ECG) recordings, the number of patients with a new or worsening ECG finding of atrial fibrillation was similar for glycopyrronium bromide and placebo. Nevertheless, information on atrial fibrillation has been included in the summary of product characteristics and the EMA has advised that cardiovascular outcomes should be monitored closely post marketing[5].

Evidence strengths and limitations

Although these 2 phase III studies were well conducted, they had some limitations. In their assessment of glycopyrronium bromide, the EMA noted that the most robust clinical criteria were not used to exclude patients suspected of having asthma. Nevertheless, the inclusion and exclusion criteria were considered sufficient to ensure that study participants reflect a general population of patients with moderate to severe COPD[5].

There was also a 2-week run-in phase, after which 38% of patients were excluded in GLOW1 and 47% of patients were excluded in GLOW2, primarily because of failure to meet the diagnostic or severity criteria. In addition, 19% of patients in GLOW1 and 24% of patients in GLOW2 did not complete the studies. The most frequent reasons for this were withdrawal of consent and adverse events[5]. For both studies allocation concealment was unclear, which may have introduced bias.

The majority of patients in these studies had moderate COPD (about 61% in GLOW1 and 64% in GLOW2); no published data are available on patients with mild or very severe disease. Caution is needed when translating phase III trial results to clinical practice in a diverse population of people with COPD.

It is not yet known whether the effects of glycopyrronium bromide are maintained beyond 1 year. Also, more robust evidence comparing glycopyrronium bromide with other active treatments for COPD, such as a LAMA or a LABA (with or without an ICS), would enable its place in therapy to be more clearly established. In GLOW2, glycopyrronium bromide and open-label tiotropium were found to be of similar efficacy. However, the inability to blind tiotropium treatment means these results could be biased. It is difficult to blind tiotropium in clinical trials because the commercial capsules (marked with a logo) contain a hygroscopic powder that cannot be removed and repackaged into unmarked capsules.

The long-term safety of glycopyrronium bromide is still unclear. In view of the potential risk of adverse cardiovascular outcomes related to the mechanism of action of anticholinergic quaternary ammonium compounds and the concerns on cardiovascular safety of same class products[10], the Committee for Medicinal Products for Human Use (CHMP) recommended that adverse cardiovascular outcomes should be followed closely in the post-marketing setting. They also requested a post-authorisation safety study to monitor cardiovascular and cerebrovascular outcomes post-marketing[5].



[5] European Medicines Agency (2012) European public assessment report: Seebri Breezhaler

[6] D'Urzo A, Ferguson GT, van Noord JA et al. (2011) Efficacy and safety of once-daily NVA237 in patients with moderate-to-severe COPD: the GLOW1 trial. Respiratory Research 12: 156

[7] Kerwin E, Hébert J, Gallagher N et al. (2012) Efficacy and safety of NVA237 versus placebo and tiotropium in patients with COPD: the GLOW2 study. European Respiratory Journal 40: 1106–14

[9] Novartis Pharmaceuticals UK Ltd (2012) Seebri Breezhaler summary of product characteristics

[10] MHRA (2010) Drug Safety Update