Exhibit 99.1

Poster No. D3

A repeat dose, double-blind, placebo-controlled ‘Thorough QT/QTc study’ to assess the cardiac safety
of fluticasone furoate (FF) and vilanterol (VI) administered in combination

Rodger Kempsford,(1) Ann Allen,(2) Kathryn Kelly,(3) Parminder Saggu,(4) Courtney Crim(5)

GlaxoSmithKline R&D: (1)Respiratory Medicines Development Centre and (2)Clinical Pharmacology Modelling & Simulation, Stevenage, UK; (3)Clinical Pharmacology Science & Study Operations, Uxbridge, UK; (4)Quantitative Sciences, Uxbridge, UK; (5)Respiratory Medicines Development Center, Research Triangle Park, NC, USA

INTRODUCTION

·                  A combination of the novel inhaled corticosteroid FF and long-acting beta2 adrenoceptor agonist VI (FF/VI) is currently under development as a once-daily inhaled treatment for asthma and COPD.

·                  Beta2 adrenoceptor agonists can prolong the electrocardiogram (ECG) QTc interval, particularly at supra-therapeutic doses.

·                  This study was conducted to comply with United States and European regulatory requirements that all new non-antiarrhythmic drugs must be assessed for potential effects on cardiac conduction in a ‘Thorough QT/QTc study’.(1)

OBJECTIVES

Primary

·                  To demonstrate the lack of effect of FF/VI (200/25mcg) on the QTc interval (Fridericia’s correction; QTcF) compared with placebo after 7 days’ dosing. This represented the highest FF/VI dose administered in phase III studies in asthma and COPD.

Secondary

·                  To estimate the effect of FF/VI (800/100mcg) on the QTcF interval compared with placebo after 7 days’ dosing. This FF/VI dose represented four-times the highest dose administered in phase III studies in asthma and COPD.

·                  To estimate the effect of FF/VI (200/25 and 800/100mcg) on the QTc interval (individual correction; QTci) compared with placebo after 7 days’ dosing.

·                  To estimate the effect of single dose oral moxifloxacin (400mg; positive control for assay sensitivity) on the QTcF interval compared with placebo after 7 days’ dosing.

METHODS

·                  This was a single center, randomized, double-blind (single-blind moxifloxacin), four-way crossover study in healthy male and female non-smoking subjects.

·                  Subjects each received:

· inhaled FF/VI (200/25mcg) for 7 days + placebo tablet (Day 7 only)

· inhaled FF/VI (800/100mcg) for 7 days + placebo tablet (Day 7 only)

· inhaled placebo for 7 days + placebo tablet (Day 7 only)

· inhaled placebo for 7 days + moxifloxacin 400mg tablet (Day 7 only).

·                  Continuous ECG data (Mortara H12+ [12-lead; 1000Hz]) was collected on Day –1 (0–12h; period 1) and on Day 7 (0–24h; periods 1–4).

·                  Subjects remained on bed rest in a semi-supine position for 30min pre-dose to 4h post-dose and were then placed on bed rest in the supine or semi-supine position for at least 30min prior to each remaining time point.

·                  QTc values were calculated as the mean of 15 ECG complexes (five ECG complexes from each of three 10-second recordings approximately 1min apart). Each subject’s ECGs were analyzed by the same blinded independent cardiologist.

·                  QTcF (msec) was calculated as QT/(R-R)1/3.

·                  Each individual’s QTci correction factor was determined from 0–12h Holter ECGs collected pre-treatment (Day –1 treatment period 1), pre-treatment (Day 1 periods 1–4) and on placebo (total = 72 ECG readings [24 time points in triplicate]).

·                  QTc data were analyzed as the mean time-matched difference from placebo in change from baseline. A clinically significant effect was defined as where the upper bound of the upper 90% confidence interval (CI) for this difference exceeded 10msec at any time point.(1)

·                  Adverse events (AEs) and laboratory safety screens were recorded throughout the study.

RESULTS

Table 1. Demographics (N=85).

SD = standard deviation

Pharmacodynamics

·                  At an FF/VI dose of 200/25mcg there was a lack of effect on QTcF or QTci: all time-matched mean differences in change from baseline from placebo were <5msec with all upper 90% CIs <10msec (Figure 1).

·                  At an FF/VI dose of 800/100mcg (representing four times the proposed therapeutic FF/VI dose of 200/25mcg), there was no clinically significant effect on QTcF (all upper 90% CIs were <10msec) at any time point except 30min post-dose where the upper 90% CI exceeded 10msec but with a mean value <10msec (mean [90% CI]: 9.6msec [7.2, 12.0]).

·                  At an FF/VI dose of 800/100mcg there was no clinically significant effect on QTci: all time-matched mean differences in change from baseline from placebo were <5msec with all upper 90% CIs <10msec (Figure 1). There was no marked difference in 0–24h QTci for FF/VI 200/25mcg and 800/100mcg (Figure 1) suggesting that QTci more effectively corrected for heart rate than QTcF.

·                  Categorical analysis showed that there were no QTcF values >450msec after either FF/VI 200/25mcg or 800/100mcg (Table 2). There were no QTcF increases from baseline of >30msec after FF/VI 200/25mcg and only three subjects (4%) had an increase from baseline of between 30–60msec (Table 3).

·                  Assay sensitivity was confirmed: moxifloxacin (400mg) prolonged both QTcF and QTci time-matched mean differences in change from baseline from placebo >10msec at 1–8h post-dose (Figure 1). The magnitude of the effect was comparable for QTcF and QTci and was also similar to that reported previously.(2) Moxifloxacin was also associated with increases in QTcF as indicated by categorical analysis (Tables 2 and 3).

·                  Heart rate changes were seen after both doses of FF/VI. Maximum heart rate (0–4h mean difference from placebo [90% CI]) was 3.9bpm [2.7, 5.1] and 12.4bpm [11.2, 13.6] following administration of FF/VI 200/25mcg and 800/100mcg, respectively. Weighted mean heart rate (0–4h mean difference from placebo [90% CI]) was 2.6bpm [1.6, 3.5] and 7.5bpm [6.6, 8.5] following administration of FF/VI 200/25mcg and 800/100mcg, respectively. Peak effects on heart rate were seen at 10min after dosing with FF/VI, which correlated with the VI Cmax at 5–10min. Maximum and weighted mean heart rate (0–4h mean difference from placebo [90% CI]) following moxifloxacin 400mg were 0.3 bpm [–0.9, 1.6] and 1.1bpm [0.2, 2.1], respectively.

·                  All treatments were well tolerated; there were no AEs that led to withdrawal. The frequency (%) of AEs was higher for the FF/VI 800/100mcg (61%) and moxifloxacin (58%) treatment regimens compared with FF/VI 200/25mcg (47%) and placebo (42%) treatment regimens (Table 4). The most frequent AE was headache, which had a similar frequency across all treatment regimens (18–23%). FF/VI 800/100 was associated with a higher incidence of palpitations (15%), dizziness (10%) and tremor (8%) compared with placebo or FF/VI 200/25mcg. Single-dose moxifloxacin administration was associated with nausea (11%).

Figure 1. QTcF and QTci on Day 7 (0–24h) following administration of FF/VI and moxifloxacin. Adjusted mean change from baseline (difference from placebo [90% CI]).

Table 2. QTcF categorical analysis: absolute values (0–24h post-dose; Day 7).

Per protocol population

Table 3. Categorical QTcF analysis: change from baseline (0–24h post-dose; Day 7).

Per protocol population

Table 4. Most frequent AEs (>3% in any treatment group).

URTI = upper respiratory tract infection

CONCLUSIONS

·                  At the proposed highest therapeutic dose of FF/VI (200/25mcg) there was a lack of effect on the QTc interval as measured by either QTcF or QTci. All time-matched mean differences from placebo (0–24h) were <5msec with no upper 90% CI values greater than 10msec.

·                  At a four-fold multiple of the FF/VI therapeutic dose (800/100mcg) there was a small, predictable and transient effect on QTcF during the 1st hour after dosing. The upper 90% CI exceeded 10msec at one time-point (30min) after dosing. All mean differences from placebo were <10msec.

·                  At a four-fold multiple of the FF/VI therapeutic dose (800/100mcg) there was a lack of effect on QTci. All time-matched mean differences from placebo (0–24h) were <5msec with no upper 90% CI values greater than 10msec. There were no marked differences in QTci between FF/VI 200/25mcg and 800/100mcg.

REFERENCES

(1)          ICH Harmonised Tripartite Guideline, May 2005. Available at: http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/ Guidelines/Efficacy/E14/Step4/E14_Guideline.pdf.

(2)          Stylianou A, Roger K, Stephens K. A statistical assessment of QT data following placebo and moxifloxacin dosing in thorough QT studies. J Biopharmaceutical Statistics 2008;18:502–16.

ACKNOWLEDGMENTS

·                  This study was sponsored by GlaxoSmithKline (ClinicalTrials.gov: NCT01209026; HZA102936).

·                  R. Kempsford is an employee of GlaxoSmithKline.

·                  The clinical phase of this study was conducted by Hammersmith Medicines Research, London NW10 7EW, UK.

·                  Editorial support (in the form of copy-editing) was provided by Geoff Weller, PhD, at Gardiner-Caldwell Communications and was funded by GlaxoSmithKline.

Presented at the Annual Conference of the American Thoracic Society (ATS), San Francisco, CA, USA, May 18–23 2012