The current study demonstrated that QD dosing with UMEC provided statistically and clinically significant improvements in lung function over 24 hours. These results extend the findings from two previous dose-ranging studies [6, 8] by demonstrating a dose ordering over a range of UMEC 15.6 to 125 μg QD and further substantiating that UMEC is efficacious when administered QD. In addition, trough FEV1 results for the overlapping doses of UMEC 125 μg QD and UMEC 62.5 μg QD from this study were similar to that observed in the previous dose-ranging studies (UMEC 125 μg QD: 0.147 L [95% CI: 0.077, 0.216]  and 0.159 L [95% CI: 0.088, 0.229]  and UMEC 62.5 μg QD: 0.128 L [95% CI: 0.060, 0.196] ).
Consistent with published dose–response models for bronchodilators [19, 20], a population physiological Emax model characterised the relationship between UMEC dose and trough FEV1 on Day 8. A clear monotonic UMEC dose response was observed with dose ordering from the lowest (15.6 μg) to highest dose (125 μg). The maximum dose (125 μg UMEC QD) equated to approximately 3.4-fold ED50 of 37 μg estimated in this study. The trough FEV1 response for UMEC 125 μg QD (0.183 L) was near the maximum predicted response from the model (0.185 L) and was similar to the trough FEV1 response observed in the previous dose-ranging studies in patients with COPD, which ranged from 0.147 to 0.159 L [6, 8].
A dose ordering was observed across UMEC QD dosing regimens with the UMEC 125 μg QD dose demonstrating the greatest improvement in lung function measurements compared with the lower QD doses. This pattern was further supported by Bayesian analysis of trough FEV1. Although all UMEC QD doses exhibited greater improvements in serial FEV1 measures compared with placebo at Day 7, UMEC 125 μg QD maintained greater increases from baseline in FEV1 across serial time points over 24 hours compared with the other QD doses.
Interestingly, greater improvements in trough FEV1 were observed on Day 8 compared to improvements observed on Day 7. This may be explained by the time of drug administration. Measurements on Days 7 and 8 were taken 23 and 24 hours after dosing; however, only dosing before the Day 8 measures was observed by study staff. Furthermore, patients were required to stay overnight between Days 7 and 8. Both of these factors may explain the differences observed in trough FEV1 across the two study days. Modelling of the pooled Day 7 and 8 data provided similar results to those on Day 8 only, and the ANCOVA results on Day 7 supported those on Day 8. Although patients with COPD are known to change their responsiveness to treatment over time, individual heterogeneity in response to treatment was not examined during this study, which may be viewed as a limitation. Individual response to treatment may have affected the interpretation of the study only because the study was small; however, the analysis was adjusted for the effect of patient, treatment and period.
The effects of UMEC QD versus BID dosing were also examined and BID regimens demonstrated small increases in the difference from placebo in some measures over QD regimens of the same total daily dose. Our study was not powered for formal comparisons between QD and BID doses. Given the small magnitude of improvement observed with BID dosing, it is considered unlikely to translate into a clinically significant benefit in efficacy. Improvements in serial FEV1 from placebo over the first 12 hours were maintained at time points over the second 12 hours for all UMEC QD doses, suggesting that QD dosing is supported for UMEC. Additionally, the ratios of evening (12–24-hour) to morning (0–12-hour) weighted mean FEV1 values were approximately 1.0 for all QD doses indicating that the 24-hour duration of effect is an intrinsic property of UMEC rather than being of a dose-related nature. The model-based analysis did not find dose regimen (QD versus BID) to be an important factor influencing dose response, which is also consistent with the previous study results . This finding is not unexpected given that preclinical UMEC data demonstrated slow off-rate receptor kinetics in in-vitro human M3 receptor and bronchial tissue, and support a QD regimen . It is noteworthy that the BID regimen showed small numerical increases in FEV1 values compared with those following QD and this finding is reflected in the probability values for 15.6 μg BID versus 31.25 μg OD. These numerical differences in FEV1 response between regimens are generally accounted for by the model-based analysis that integrated the variability estimates between- and within-patient across all doses . Plausible reasons for the marked lower probability estimates at 31.25 μg BID compared with the 62.5 μg OD regimen may include study design factors, namely imbalance in patient dose allocation due to the incomplete block study design, or possible clinical under-performance of the 31.25 μg BID formulation.
The LAMA tiotropium was used as an active comparator in this study as it is a well-established medication for the treatment of COPD . Although the observed improvement in trough FEV1 over placebo with tiotropium was within the range of those observed with UMEC doses, tiotropium was at the lower end of the range. Nearly all UMEC QD and BID doses demonstrated improvements over placebo in trough FEV1 on Day 8 (range: 0.101–0.183 L) that were numerically similar to or greater than tiotropium (0.101 L). However, trough FEV1 does not convey the full picture of bronchodilation offered to patients, and thus the 24-hour weighted mean and 24-hour serial FEV1 profiles were also evaluated. The 24-hour weighted mean and 24-hour serial FEV1 profiles showed UMEC doses of 62.5 and 125 μg were the most similar in efficacy to tiotropium, with the bronchodilator effect of the 62.5 μg dose generally less and the effect of the 125 μg dose generally greater than that of tiotropium over time. These data suggest that UMEC 125 μg and UMEC 62.5 μg would be appropriate doses to use in future clinical studies to further assess efficacy and safety of UMEC in COPD patients.
No assessments of COPD-related symptoms were included in this study. However, rescue salbutamol on an as-needed basis for breakthrough symptoms provided a secondary marker of symptom control. Treatment with UMEC reduced rescue salbutamol use compared with placebo at all doses and a dose ordering of reduction in salbutamol use was observed for UMEC QD doses, with UMEC 125 μg QD doses showing the greatest improvement. Rescue salbutamol use data are consistent with the observed improvements in lung function (i.e. the greatest reduction in use was observed with UMEC 125 μg QD, which had the greatest improvement in lung function parameters). Interestingly, the greatest reduction in salbutamol occurred with tiotropium. This may be due to the design of the study in that tiotropium was administered in an open-label fashion.
This was the first study to examine the PK of low-dose UMEC (15.6 μg and 31.25 μg QD and the BID dosing regimen) in patients with COPD. In spite of the large percentage of non-quantifiable data, increased UMEC systemic exposure (in terms of AUC and Cmax) was seen with lower doses. Median Cmax occurred early for all UMEC doses. Systemic exposure was higher following QD dosing than BID dosing for the same total daily dose. Cross-study comparison of the steady-state data with previous studies of UMEC 62.5 μg to 1000 μg QD suggest that systemic exposure of UMEC was within the expected range [6, 8].
The safety profile was comparable between treatment groups. Overall, AEs were higher with UMEC 125 μg QD dosing (18%) than placebo or the other active treatments; however, the individual AEs observed with the UMEC 125 μg QD consisted of reports of sinusitis, dysgeusia, nasopharyngitis, and headache. Headache was the most common AE across all treatments and did not appear related to active treatment. Although the AEs of nasopharyngitis, dysgeusia, and sinusitis could potentially be topical effects of LAMA treatment, these events occurred only with UMEC 15.6 μg and 125 μg QD dosing and so did not appear to be dose-related. Overall, UMEC was well tolerated across all dose regimens.