Exhibit 99.2

Poster Board No. 415

Cardiovascular safety of umeclidinium/vilanterol in COPD: results from eight randomized clinical trials

Gerald Naccarelli(1), John Finkle(2), Bikramjit Chopra(3), Jean Brooks(3), Stephanie Harris(4), Alison Church(4)

(1)Penn State Heart and Vascular Institute, Milton S. Hershey Medical Center, Penn State University, Hershey, PA, USA; (2)GlaxoSmithKline, Upper Providence, Collegeville, PA, USA;

(3)GlaxoSmithKline, Stockley Park, Uxbridge, Middlesex, UK; (4)GlaxoSmithKline, Respiratory and Immuno-Inflammation, Research Triangle Park, NC, USA

INTRODUCTION

·                  Concerns have been raised around the cardiovascular (CV) safety of long-acting muscarinic antagonists (LAMAs) and long-acting b-agonists (LABAs) in patients with chronic obstructive pulmonary disease (COPD).(1), (2)

·                  Umeclidinium (UMEC)/vilanterol (VI) (ANORO™ ELLIPTA™) is an approved maintenance treatment for COPD in the US. It is not indicated for treatment of asthma.

OBJECTIVES

·                  To assess the CV safety of once-daily UMEC/VI in patients with COPD.

METHODS

·                  Major adverse CV events (MACE) and CV adverse events of special interest (AESI) were assessed in a pooled analysis of patients with COPD enrolled in Phase III efficacy and safety studies of UMEC/VI 125/25 mcg (delivering 113/22 mcg), UMEC/VI 62.5/25 mcg (delivering 55/22 mcg), UMEC 62.5 mcg (delivering 55 mcg), UMEC 125 mcg (delivering 113 mcg), VI 25 mcg (delivering 22 mcg), or active-comparator (tiotropium [TIO] via Handihaler®) (Table 1).

MACE analysis (Studies 1–8)

·                  MACE evaluations were divided into ‘broad’ (planned) and ‘narrow’ (post hoc) analyses and included:

·                  Broad MACE analysis included all the Medical Dictionary for Regulatory Activities (MedDRA) preferred terms (PTs) in the ‘Myocardial infarction (MI)’ and ‘Other ischemic heart disease’ standardized MedDRA queries (SMQs).

·                  Narrow MACE analysis specified the preferred terms of ‘acute MI’ and ‘MI only’

·                  Both analyses included adjudicated CV death and non-fatal stroke in addition to the terms listed above.

AESI analysis (Studies 1–4, and 7)

·                  CV AESI are presented from the four 24-week primary efficacy studies and a 52-week long-term safety study.

·                  CV AESI categories were: acquired long QT interval, cardiac arrhythmias, cardiac failure, cardiac ischemia, hypertension, sudden death, and stroke.

·                  Appropriate SMQs or MedDRA Higher Level Terms (HLTs) were used to define AE terms.

·                  When MedDRA SMQs or HLTs were not available, an appropriate selection of MedDRA PTs were used.

RESULTS

Patients

·                  Overall, 6156 patients (2615 subject years [SY]) were included in the MACE analyses and 5295 patients (2315 SY) in the CV AESI analysis.

MACE analyses

·                  A similar or lower number of patients experienced an event with UMEC/VI or their monotherapy components compared with placebo in both the broad and narrow MACE analyses (broad: UMEC/VI or their monotherapy components 9–22 [1–2%] patients vs 20 [2%] patients for placebo; narrow: UMEC/VI or their monotherapy components 2–8 [<1%] patients vs 7 patients [<1%] for placebo; Table 2).

·                  Exposure-adjusted incidence rates (i.e. number of subjects with an event per 1000 SY of exposure) were lower in both the broad and narrow analyses for UMEC/VI or their monotherapy components than for placebo (broad: UMEC/VI treatment 31.2–44.5 vs 54.3 for placebo; narrow: UMEC/VI or their monotherapy components 9.9–18.1 vs 19.0 for placebo; Table 2).

·                  Incidence of CV death, non-fatal stroke, non-fatal cardiac ischemia, and non-fatal MI were low and similar across all treatment groups including placebo(<1%).

·                  There was a small numerical imbalance in exposure-adjusted incidence of non-fatal
MI (UMEC/VI 125/25 mcg, 5.2; UMEC/VI 62.5/25 mcg, 7.4; placebo, 2.7 patients with events/1000-patient-years) based on the low actual number of MIs (UMEC/VI 125/25 mcg, 3; UMEC/VI 62.5/25 mcg, 3; placebo, 1).

·                  This was not observed for non-fatal cardiac ischemia.

TABLE 1. SUMMARY OF CLINICAL STUDIES

TABLE 2. SUMMARY OF MACE (STUDIES 1–8)(a)

(a)DB2113361; DB2113373; DB2113360; DB2113374; DB2114417; DB2114418; DB2113359; AC4115408, subjects in crossover studies were counted once under each treatment received; (b)Independently adjudicated; (c)The following MedDRA SMQ contributed to the non-fatal stroke AESI category: central nervous system haemorrhages and cerebrovascular conditions SMQ; (d)The following MedDRA SMQs contributed to non-fatal cardiac ischemia: Myocardial Infarction SMQ; Other Ischaemic Heart Disease SMQ. (e)The following MedDRA PTs contributed to non-fatal MI: MI and acute MI.

CV AESI

·                  In the 6-month studies, small imbalances in the incidence of any CV AESI was noted in some of the active treatment groups compared with placebo. However, incidence in the UMEC/VI 125/25 mcg group was similar and had a lower exposure adjusted rate than the placebo group (Table 3).

·                  Incidence of cardiac arrhythmia, cardiac failure, cardiac ischemia, and hypertension was low; small differences in exposure-adjusted incidence rates were observed.

·                  Stroke was reported in <1% of patients across all treatment groups including placebo.

TABLE 3. CV AESI (STUDIES 1–4, 6-MONTH STUDIES)(a)

(a)DB2113361; DB2113373; DB2113360; DB2113374

·                  In the long-term safety study, the incidence of any CV AESI was lower for UMEC 125 mcg and UMEC/VI 125/25 mcg (22% and 15%) compared with placebo (23%) (Table 4).

·                  Incidence of cardiac arrhythmias was lower in the UMEC/VI 125/25 mcg group compared with UMEC 125 mcg and placebo (12% vs 17% and 16%).

·                  Incidence of cardiac ischemia was lower in the UMEC/VI 125/25 mcg and UMEC 125 mcg group compared with placebo (2% vs 4%).

·                  Incidence of hypertension was lower in UMEC/VI 125/25 mcg and UMEC 125 mcg compared with placebo (4% and 3% vs 6%).

·                  No additive effects were observed with combination treatment compared with the individual components and no dose response was evident between either UMEC/VI or UMEC doses.

TABLE 4. CV AESI (STUDY 7, LONG-TERM SAFETY)(a)

(a)DB2113359

CONCLUSIONS

·                  No increased risk of MACE was observed with active treatments versus placebo

·                  Overall, the number of cardiac ischemia events were low with inconsistent small imbalances appearing in some studies but not others.

·                  No evidence of dose response for either UMEC/VI or UMEC and no additive effect with combination treatment over individual components was observed.

·                  No clinically-relevant increase in CV events was apparent with UMEC/VI, UMEC or VI compared with placebo.

REFERENCES

(1) Singh S et al. JAMA. 2008;300:1439–50.

(2) Gershon A, et al. JAMA Intern Med. 2013;173:1175–85.

(3) Celli B, et al. CHEST.. 2014 [Epub ahead of print].

(4) Donohue J, et al. Respir Med. 2013;107:1538–46.

(5) DeCramer M, et al. Lancet Resp Med. In press.

(6) Maltais F, et al. Submitted.

(7) Donohue J, et al. Submitted.

(8) Trivedi R, et al. Eur Resp J 2014;43:72–81.

ACKNOWLEDGMENTS

·                  The presenting author, Gerald Naccarelli, declares the following real or perceived conflicts of interest during the last 3 years in relation to this presentation: consultancies for Xention, Daiichi-Sankyo, Biosense Webster, Janssen, Otsuka Pharmaceutical, Bristol-Myers Squibb, GlaxoSmithKline, Boehringer-Ingelheim, Pfizer, sanofi aventis, and Forest. JF, BC, JB, SH, and AC are employees of GlaxoSmithKline and hold stocks/shares in GlaxoSmithKline.

·                  The studies described in this presentation were funded by GlaxoSmithKline.

·                  Editorial support in the form of writing assistance, assembling tables and figures, collating author comments, grammatical editing, and referencing was provided by Joanne Parker of Fishawack Indicia Ltd and was funded by GlaxoSmithKline.

Presented at the Annual Congress of the American Thoracic Society (ATS), San Diego, CA, USA, May 16–21, 2014