Article Text

S35 Efficacy and safety of the direct switch from various previous treatments to glycopyrronium or indacaterol/glycopyrronium in patients with moderate copd: the crystal study
  1. C Vogelmeier1,
  2. M Aalamian-Mattheis2,
  3. T Greulich1,
  4. J M Marin3,
  5. W Castellani4,
  6. T Similowski5,
  7. V Ninane6,
  8. M Gaga7,
  9. S Lane8,
  10. X Nunez9,
  11. F Patalano2,
  12. A Clemens2,
  13. K Kostikas2
  1. 1University Medical Centre Giessen and Marburg, Marburg, Germany
  2. 2Novartis Pharma AG, Basel, Switzerland
  3. 3Respiratory Medicine, Hospital Universitario Miguel Servet, Zaragoza, Spain
  4. 4Department of Respiratory Physiopathology, Palagi Hospital, Florence, Italy
  5. 5Respiratory Medicine and Intensive Care, Pitié – Salpêtrière Hospital, Paris, France
  6. 6CHU Saint-Pierre – Service de Pneumologie, Brussells, Belgium
  7. 7Athens Chest Hospital “Sotiria”, Athens, Greece
  8. 8Adelaide and Meath Hospital, Dublin, Ireland
  9. 9TFS Develop, Barcelona, Spain


Introduction and objectives In contrast to clinical trials, changes to new therapies in clinical practice occur without any washout period. The CRYSTAL study was designed to mimic clinical practice. Patients with symptomatic, non-frequently exacerbating, moderate COPD treated with various drugs were directly switched to glycopyrronium 50 μg (GLY) or indacaterol/glycopyrronium 110/50 μg (IND/GLY). Lung function and symptoms were evaluated.

Methods CRYSTAL was a prospective, multicentre, 12-week, randomised, pragmatic, open-label trial. Patients were recruited into 4 Groups according to previous medication and symptoms (mMRC) and randomised to a direct switch to GLY or IND/GLY vs. continuation of baseline therapy (3:1). Co-primary objectives were superiority of GLY vs. previous SABA and/or SAMA, non-inferiority of GLY vs. previous LABA or LAMA, and superiority of IND/GLY vs. LABA, LAMA and LABA+ICS regarding trough FEV1 and transition dyspnoea index (TDI) at Week 12. Due to slow recruitment, Groups A and B were prematurely discontinued at the time of completion of Groups C and D.

Results Of the 4,389 patients randomised, 2,159 patients received IND/GLY (C2: n = 811; D2: n = 811) or continued their previous treatment (LABA + ICS C1: n = 269; LABA or LAMA D1: n = 268). IND/GLY provided superior improvement in trough FEV1 at Week 12 vs. LABA + ICS (treatment difference (Δ) = 71 mL, p < 0.0001) and LABA or LAMA (Δ = 101 mL, p < 0.0001). IND/GLY also improved TDI vs. LABA + ICS (Δ = 1.10 units, p < 0.0001) and vs. LABA or LAMA (Δ = 1.26 units, p < 0.0001). Significantly more patients on IND/GLY reached the minimally clinically important difference (MCID) of 100 mL for trough FEV1 and 1 point for TDI vs. comparators (Table 1). In the Groups A and B that were underpowered due to sample size, GLY was superior to previous SABA and/or SAMA and was non-inferior to previous LABA or LAMA on trough FEV1 and TDI (Table 1). GLY and IND/GLY were well tolerated.

Conclusions In the pragmatic CRYSTAL trial, IND/GLY demonstrated superior improvement in lung function (trough FEV1) and dyspnoea (TDI) after 12 weeks, in symptomatic patients with moderate COPD and a history of up to 1 exacerbation in the previous year, after direct switch from previous treatment with either LABA+ICS or with a LABA or LAMA.

Abstract S35 Table 1

CRYSTAL Study Results

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.