Exhibit 99.4

Poster No. D68

Safety and tolerability of the GSK573719/vilanterol combination in patients with COPD

Feldman G(1), Walker R(2), Brooks J(3), Mehta R(4), Crater G(5)

(1)S. Carolina Pharmaceutical Research, Spartanburg, USA; (2)Greenville Pharmaceutical Research, Greenville, USA; (3)GlaxoSmithKline, Uxbridge, UK; (4)GlaxoSmithKline, Research Triangle Park, USA; (5)GlaxoSmithKline, Mississauga, Canada

INTRODUCTION

·                  Bronchodilators are central to the pharmacological management of chronic obstructive pulmonary disease (COPD) symptoms. Treatment with long-acting bronchodilators, a beta2 agonist (LABA) or a muscarinic antagonist (LAMA), has been shown to be more effective than a short-acting bronchodilator.(1)

·                  Long-acting bronchodilators, however, may increase the risk of cardiovascular side effects such as tachycardia and prolongation of the QTc interval.(2)–(4)

·                  Combination therapy with different classes of pharmacological agents has the potential to increase efficacy and decrease the risk of side effects, compared with increasing the dose of a single bronchodilator when symptoms are not adequately controlled.(1)

·                  GSK573719 (umeclidinium, UMEC) is an investigational inhaled LAMA currently in development as a once-daily combination product with vilanterol (GW642444, VI), an investigational, potent and selective LABA, for treatment of COPD.

·                  Previous studies of UMEC(5)–(6) and VI(7) monotherapies in healthy volunteers and in patients with COPD(8)–(12) have demonstrated that both treatments are well tolerated and have favorable safety profiles.

·                  This was the first clinical study conducted to evaluate the safety and tolerability of the UMEC/VI combination in patients with COPD.

OBJECTIVES

·                  Evaluate the safety, tolerability, pharmacodynamics (PD), and pharmacokinetics (PK) of repeat inhaled doses of UMEC/VI (500/25mcg) administered once daily for 28 days in patients with COPD.

METHODS

Study population

·                  Males and females >40 years of age with an established clinical history of COPD, a history of smoking >10 packs of cigarettes a year, a post-albuterol/salbutamol forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) ratio of <0.70 and a post-albuterol/salbutamol FEV1 of <0.80 of predicted normal values.

·                  Patients were excluded if they had a current diagnosis of asthma, alpha1-antitrypsin deficiency, or an abnormal 12-lead electrocardiogram (ECG) that resulted in an active medical problem or had clinically significant abnormalities from 24-h Holter ECG monitoring at screening, or had used oral corticosteroids, antibiotics, or had been hospitalized due to exacerbation of COPD or a lower respiratory tract infection within 3 months prior to screening.

Study design and treatment

·                  Multicenter, randomized, placebo-controlled, double-blind, parallel-group study at four centers in the USA from January 14 2010 to April 20 2010 (ClinicalTrials.gov: NCT01039675; protocol number: DB2113120).

·                  Patients who met the eligibility criteria completed a run-in period of 5–8 days. Following the run-in period, patients who met the randomization criteria received 4 weeks of treatment with a 7-day follow-up.

·                  Patients were randomized 4:1 to UMEC/VI (500/25mcg) or placebo administered via a novel dry powder inhaler

·  this ratio provided 90% power to demonstrate the non-inferiority of UMEC/VI to placebo in terms of weighted mean pulse rate assuming at least 36 patients in the UMEC/VI group and 9 patients in the placebo group completed the study.

Endpoints

·                  Primary: weighted mean pulse rate during 0–6h post dose on Day 28.

·                  Secondary: weighted mean pulse rate during 0–6h post dose on Days 1 and 14; and maximum and minimum pulse rate during 0–6h post dose on Days 1, 14 and 28.

·                  Blood pressure, ECG, 24-h Holter monitoring, adverse events (AEs), incidence of COPD exacerbations, spirometry, hematology, biochemistry, urinalysis, and PK were assessed.

RESULTS

Demographics

·                  77 patients were screened, 51 were randomized (intent-to-treat population [ITT]), 44 completed the study

·  seven patients prematurely withdrew from the study: 2 AEs (swollen tongue and upper abdominal pain); 2 lack of efficacy(exacerbation); 1 lost to follow-up; 1 investigator discretion; 1 withdrew consent.

· Baseline patient characteristics and screening lung function are shown in Table 1.

Safety endpoints

Pulse rate

·                  The adjusted mean changes from baseline in weighted mean pulse rate during 0–6h post dose on Day 28 (primary endpoint) were similar between UMEC/VI and placebo (–0.5 beats per min [bpm]; 95% confidence interval [CI]: –5.5, 4.5; Figure 1)

·  UMEC/VI was non-inferior to placebo as the upper confidence limit was below the prespecified non-inferior limit of 10bpm.

·                  The adjusted mean changes from baseline in weighted mean pulse rate during 0–6h post dose on Day 1 were similar for UMEC/VI (–0.6bpm) and placebo (–1.2bpm). On Day 14, the values were 3.1bpm (UMEC/VI) and –1.7bpm (placebo).

·                  The adjusted mean treatment differences between UMEC/VI and placebo in maximum pulse rate during 0–6h post dose were small

·  2.0bpm (95% CI: –3.7, 7.6) on Day 1; 4.8bpm (95% CI: –1.8, 11.3) on Day 14, and –1.3bpm (95% CI: –6.9, 4.3) on Day 28 (Figure 2a).

·                  The adjusted mean treatment differences between UMEC/VI and placebo in minimum pulse rate during 0–6h post dose were also small

·  0.3bpm (95% CI: –4.3, 5.0) on Day 1; 4.0bpm (95% CI: –2.3, 10.2) on Day 14, and 1.7bpm (95% CI: –3.6, 7.1) on Day 28 (Figure 2b).

TABLE 1: BASELINE DEMOGRAPHICS AND SCREENING LUNG FUNCTION (ITT POPULATION)

SD = standard deviation

FIGURE 1: CHANGE FROM BASELINE IN 0–6H WEIGHTED MEAN PULSE RATE (ITT POPULATION)

Analyses were performed using a repeated measures model with covariates of baseline pulse rate, sex, age, smoking status, treatment day, and day by treatment and day by baseline interactions

LS = least squares

Blood pressure, ECG, and clinical laboratory assessments

·                  No clinically significant differences were observed for blood pressure and minimum and maximum heart rate.

·                  At screening, no patient had an abnormal clinically significant finding in 24-h Holter ECG assessments. At Day 28, 86% of patients on UMEC/VI and 89% of patients on placebo had no change or insignificant changes, while 11% of patients in each treatment group had clinically significant unfavorable changes.

·                  The proportion of patients receiving UMEC/VI with clinically significant unfavorable changes from baseline in the 12-lead ECG finding at any time post baseline was low and comparable with placebo (29% vs 22%).

·                  No clinically significant difference was observed between UMEC/VI and placebo in the change from baseline in 0–6h maximum QTcF.

·                  No clinically relevant treatment effects were noted in any laboratory assessments.

AEs

·                  A greater proportion of patients receiving UMEC/VI reported AEs than patients receiving placebo (26% vs 11%; Table 2).

·                  No single AE was reported in more than one patient; no serious AEs were reported.

FIGURE 2. CHANGE FROM BASELINE IN 0–6H MAXIMUM AND MINIMUM PULSE RATES (ITT POPULATION)

Analyses were performed using a repeated measures model with covariates of baseline pulse rate, sex, age, smoking status, treatment day, and day by treatment and day by baseline interactions

TABLE 2: REPORTED ON-TREATMENT ADVERSE EVENTS (ITT POPULATION)

(a)AE was reported as right upper quadrant pain

(b)One patient reported treatment-related AEs of abnormal dreams, swollen tongue and dry mouth

(c)Not considered related to study treatment

COPD exacerbation

· Three patients receiving UMEC/VI had a COPD exacerbation

· lack of efficacy was reported by the investigator as the cause of two exacerbations; these patients were withdrawn from the study

· an upper respiratory infection other than the common cold was the cause of the other exacerbation

· all three patients were treated with antibiotics and corticosteroids.

Spirometry

·                  Greater improvements were observed for UMEC/VI compared with placebo for change from baseline for both trough FEV1 and serial FEV1 over 0–6h post dose

·  raw mean change from baseline in trough FEV1 on Day 29 with UMEC/VI was 163mL and with placebo was 9mL

·  increases in serial FEV1 from baseline occurred on Day 1 at 1h, 3h, and 6h (Day 1: 194–265mL), and on Days 14 and 28 at pre-dose, 1h, 3h and 6h (Day 14: 155–271mL; Day 28: 160–273mL) with UMEC/VI.

PK/PD analysis

·                  Both UMEC and VI were rapidly absorbed (median time to maximum plasma concentration [tmax] ~6min for both drugs) following single and repeat doses of UMEC/VI. No accumulation for area under the curve (AUC) or maximum plasma concentration (Cmax) on Day 28 vs Day 1 was observed.

·                  No relationship between steady-state Cmax and change from baseline in pulse rate on Day 28 was observed for UMEC or VI (Figure 3). Furthermore, the level of change on Day 28 with active treatment was similar to that observed with placebo.

FIGURE 3: INDIVIDUAL CHANGE FROM BASELINE IN PULSE RATE VS MAXIMUM PLASMA CONCENTRATION (CMAX) ON DAY 28(a) (PK POPULATION)

(a)Linear regression: UMEC Cmax vs change in baseline pulse rate, Y=0.95X + 3.70

r2=0.0034; VI Cmax vs change in baseline pulse rate, Y=–0.696X + 7.79 r2=0.00115

CONCLUSIONS

·                  The UMEC/VI (500/25mcg) combination administered once daily for 28 days in patients with moderate-to-very severe COPD was not associated with any clinically significant effects on pulse rate, blood pressure, heart rate or incidence of AEs relative to placebo.

·                  Both UMEC and VI showed rapid absorption followed by a quick decline in plasma concentration indicating rapid distribution and elimination. No association between steady-state Cmax and change from baseline in pulse rate occurred for either UMEC or VI.

·                  The safety profile of UMEC/VI demonstrated in this study further supports the development of this once-daily LAMA/LABA combination as a COPD therapy.

REFERENCES

(1)          GOLD. Global Strategy for Diagnosis, Management, and Prevention of COPD – updated December 2011. Available at:www.goldcopd.org. Last accessed 24 April 2012.

(2)          Spiriva (tiotropium) prescribing information. http://bidocs.boehringer-ingelheim.com/BIWebAccess/ViewServlet.ser?docBase=renetnt&folder Path=/Prescribing+Information/Pls/Spiriva/Spiriva.pdf. Last accessed 24 April 2012.

(3)          Serevent (salmeterol) prescribing information. http://us.gsk.com/products/assets/us_serevent_diskus.pdf. Last accessed 24 April 2012.

(4)          Arcapta (indacaterol) prescribing information. http://www.pharma.us.novartis.com/product/pi/pdf/arcapta.pdf. Last accessed 24 April 2012.

(5)          Cahn A, et al. Eur Resp J 2011;38(Suppl. 55):723s.

(6)          Mehta R, et al. Eur Resp J 2011;38(Suppl. 5):723s.

(7)          Kempsford RD, et al. Am J Respir Crit Care Med 2010;181:A446.

(8)          Decramer M, et al. Eur Resp J 2011;38(Suppl. 55):150s.

(9)          Kelleher D, et al. Eur Resp J 2011;38(Suppl. 55):140s.

(10)    Mehta R, et al. Eur Resp J 2011;38(Suppl. 55):138s.

(11)    Kempsford RD, et al. Am J Respir Crit Care Med 2010;181:A4447.

(12)    Hanania N, et al. Chest 2012. Epub ahead of print.

ACKNOWLEDGMENTS

·                  This study was sponsored by GlaxoSmithKline.

·                  GF and RW received funds as Principal Investigators in this clinical trial sponsored by GlaxoSmithKline.

·                  Editorial support (in the form of writing assistance, assembling tables and figures, collating author comments, grammatical editing and referencing) was provided by Tara N Miller, PhD at Gardiner-Caldwell Communications (Lyndhurst, NJ) and was funded by GlaxoSmithKline.

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