Clinical Burden of Chronic Obstructive Pulmonary Disease in Patients with Suboptimal Peak Inspiratory Flow

Introduction Many patients with chronic obstructive pulmonary disease (COPD) may derive inadequate benefit from dry powder inhalers (DPIs) because of suboptimal peak inspiratory flow (sPIF). Objectives To assess the clinical burden of COPD by characterizing the clinical characteristics of participants with sPIF against medium-low resistance DPIs versus those with optimal PIF (oPIF) from two phase 3 clinical trials. Methods Baseline data were collected from two randomized, controlled, phase 3 trials (NCT03095456; NCT02518139) in participants with moderate-to-severe COPD. oPIF (60 L/min) against the medium-low resistance DPIs was used as the threshold for defining the PIF subgroups (<60 L/min (sPIF) vs ≥60 L/min (oPIF)). Results Most participants included in this analysis were White (92%) and male (63%); the mean (range) age was 65 (43–87) years. Participants with sPIF had significantly greater dyspnea than those with oPIF as measured using the modified Medical Research Council scoring (mean (95% CI): 2.1 (2.0–2.2) vs 1.6 (1.4–1.7); P  < 0.001) and baseline dyspnea index (mean (95% CI): 5.1 (4.9–5.4) vs 6.1 (5.8–6.3); P  < 0.001). Based on COPD Assessment Test scores, participants with sPIF had a higher COPD symptom burden than those with oPIF (mean (95% CI): 21.5 (19.7–23.3) vs 19.5 (18.6–20.4); P = 0.05). Conclusion In these trials, participants with COPD who had sPIF against the medium-low resistance DPIs had more dyspnea and worse health status than those with oPIF. These results demonstrate that sPIF is associated with a higher clinical burden as measured by patient-reported outcomes.


Introduction
Treatment with inhaled bronchodilators is the foundation of pharmacologic management of symptoms in patients with chronic obstructive pulmonary disease (COPD) [1].Dry powder inhalers (DPIs), pressurized metered-dose inhalers (pMDIs), soft mist inhalers (SMIs), and nebulizers are the most commonly prescribed inhalation devices for the delivery of bronchodilators [2].Each device requires a unique inhalation technique for optimal delivery of medication to the lower respiratory tract [3].For optimal use of DPIs, patients must be able to generate sufcient peak inspiratory fow (PIF) against the internal resistance of the device to disaggregate powdered drugs into fne particles for lung deposition [4].However, many patients with COPD have a suboptimal PIF (sPIF) and may not derive optimal beneft from DPIs.In observational studies, sPIF was observed in 19%-78% of stable outpatients with COPD and 32%-52% of inpatients before hospital discharge after treatment for COPD exacerbation [5][6][7][8][9][10].
sPIF in patients with COPD is associated with female sex, older age, shorter height, and lung function impairment [5][6][7][9][10][11]. Low forced vital capacity (FVC) percent predicted and inspiratory capacity (IC) percent predicted are independent predictors of sPIF [9].Despite several studies demonstrating the efects of age, sex, lung function parameters, and device resistance on PIF in patients with COPD, little is known about the association between inspiratory fow and severity of dyspnea and respiratory health status.In this analysis, we assessed the demographics and baseline clinical characteristics of participants with moderate-to-severe COPD from two randomized, controlled phase 3 trials of revefenacin [12][13][14] according to their PIF status (sPIF vs optimal PIF (oPIF)) to compare the population diferences between the PIF subgroups.We used the optimal PIF of medium-low resistance DPIs such as Diskus ® , Diskhaler ® , and Ellipta ® (60 L/min) [15, 16] as a cut- of value for these analyses because medium-low resistance has been used most frequently for reporting the prevalence of sPIF [7-10, 17, 18] and a PIF of ≥60 L/min is generally considered optimal for most DPI devices [7,9,17,19].

Trial Design.
Demographics and baseline clinical characteristics of participants with COPD were pooled from two randomized, controlled phase 3 trials, 0128 and 0149; both have been described previously [12][13][14].In brief, trial 0128 (NCT02518139) was a 52-week, tiotropium-controlled, parallel-group phase 3 safety trial evaluating the safety and tolerability of revefenacin for nebulization in participants with moderate-to-very severe COPD [12,13].Trial 0149 (NCT03095456) was a 28-day, double-blind, double-dummy, parallel-group phase 3b trial comparing the efect of once-daily revefenacin for nebulization administered via the PARI LC ® Sprint jet nebulizer with tiotropium administered via Handi-Haler ® on lung function in participants with moderate-to-very severe COPD and a PIF of <60 L/min against the medium-low resistance DPIs [14].Te trials were conducted in accordance with the principles of the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use guidelines for good clinical practice and the code of ethics of the World Medical Association's Declaration of Helsinki, and all patients provided written informed consent.

Participants.
Both trials enrolled participants diagnosed with moderate-to-severe COPD.Eligible participants had a smoking history of ≥10 packs per year, a postipratropium forced expiratory volume in 1 second (FEV 1 ) to a FVC ratio of <0.7 at screening, and a postipratropium FEV 1 of <80% of predicted normal and >700 mL at screening in trial 0128 and >400 mL in trial 0149.In addition, participants in trial 0149 had a PIF of <60 L/min.

PIF and Pulmonary Function Measurements.
Baseline PIF was measured using the In-Check ™ DIAL device (Al- liance Tech Medical, Inc.) set to medium-low resistance DPI (R-2) and high resistance DPI (R-5) in trial 0149 and to R-5 alone in trial 0128.To measure PIF, participants were instructed to exhale completely, place the mouthpiece of the device into their mouths, and inhale as forcefully and deeply as possible.Participants repeated the PIF maneuver three times after adequate rest and recovery from each efort, with their PIF measurement refecting the highest recorded value.
Measurements were conducted at zero resistance followed by R-5 resistance in trial 0128 and at R-2 resistance followed by R-5 resistance in trial 0149.Data from trial 0149 were used to develop an algorithm to predictively correlate resistance in the R-2 device to resistance in the R-5 device.Tese values were then used to defne PIF against the R-2 device in trial 0128.Te methodology used to defne the correlation between PIF against the R-2 device and PIF against the R-5 device has been described previously [20].On the basis of this correlation analysis, a PIF value of 40 L/min against the R-5 device is approximately equivalent to a PIF of 60 L/min against the R-2 device [20].oPIF against the resistance of the R-2 device was defned as >60 L/min, and sPIF was defned as ≤60 L/min.

Patient-Reported
Outcomes.Dyspnea at baseline was assessed using the modifed Medical Research Council (mMRC) dyspnea scale and the baseline dyspnea index (BDI) using standard methods [21][22][23][24].A higher score on the mMRC scale and a lower score on the BDI represented greater dyspnea.An mMRC score of ≥2 was used as a threshold for distinguishing participants with more dyspnea from those with less dyspnea [1].
Participants' quality of life was assessed using the COPD Assessment Test (CAT) and St. George's Respiratory Questionnaire (SGRQ) [25,26].Participants with a CAT score of ≥10 were categorized as symptomatic [27] and those with a CAT score of ≥20 [1] as having more severe COPD symptoms.Participants with an SGRQ score of ≥40 were considered to have severe COPD symptoms [28].CAT and SGRQ scores are reported only for trial 0128, as they were not assessed during trial 0149.

Statistical Analysis.
Baseline characteristics, such as age, time since COPD diagnosis, smoking duration, height, weight, body mass index (BMI), PIF, percent predicted FEV 1 and FVC, SGRQ score, and CAT score, are reported as mean values with 95% confdence intervals.Diferences between the oPIF and sPIF subgroups were compared using a twosample t-test.

Baseline Demographics and Clinical Characteristics.
Of the total number of participants enrolled, PIF data were available for 525 participants (actual measurements from 206 participants enrolled in trial 0149 and derived values for participants enrolled in trial 0128).Of these participants, 273 (52.0%) had sPIF (mean (95% CI): 44.6 L/min ( Participants with sPIF had a signifcantly higher COPD symptom burden than participants with oPIF on the basis of CAT's total score (P � 0.05; Figure 2; Table 2).Tere was no signifcant diference in SGRQ scores between participants with sPIF and those with oPIF (P � 0.22; Table 2).

Discussion
In this analysis of data from two phase 3 trials of participants with moderate-to-very severe COPD, we have demonstrated that sPIF in patients with COPD is associated with high levels of dyspnea.Chronic dyspnea is one of the most common symptoms of COPD [1] and may be caused by a variety of mechanisms, including increased ventilatory demand, dynamic airway compression, lung hyperinfation, and respiratory muscle weakness [29].PIF is determined by the patient's inspiratory efort and the strength of the inspiratory muscles [30].Janssens and colleagues have demonstrated a signifcant correlation between PIF and both the inspiratory and expiratory mouth pressures, measures of respiratory muscle strength [7].Respiratory muscle function is often compromised in COPD because of lung hyperinfation, hypoxemia, and muscle wasting [8].Lung hyperinfation can afect PIF by shortening the vertical muscle fbers of the diaphragm, which in turn reduces the inspiratory muscle strength, and by adding an elastic load that must be overcome during inspiration [30,31].In addition, weight loss caused by poor nutrition and muscle wasting in patients with COPD can also lead to lower respiratory and peripheral muscle strength [32][33][34][35], resulting in dyspnea and sPIF.Tus, reduced inspiratory muscle strength in patients with COPD may be the common mechanism responsible for sPIF and dyspnea.
In addition to experiencing greater dyspnea, participants with sPIF against the medium-low resistance DPIs also had a higher perceived symptom burden than those with oPIF, on the basis of participants' CAT scores.More participants with sPIF than with oPIF reported SGRQ and CAT scores above the threshold for the disease with uncontrolled symptoms.SGRQ and CAT provide a comprehensive assessment of the COPD-specifc health status of patients [1]; therefore, a signifcant diference between CAT scores in participants with sPIF and those with oPIF suggests that suboptimal PIF may be generally associated with poor health status.A signifcant association between sPIF and high levels of dyspnea and poor COPD-related health status was also recently demonstrated in outpatients with stable moderateto-very severe COPD and sPIF [36].
In this analysis, participants with sPIF had a signifcantly lower FEV 1 percent predicted and FVC percent predicted than did participants with oPIF.Other studies have not shown a consistent diference in spirometric measurements (FEV 1 and FEV 1 percent predicted) between participants with sPIF and oPIF [7-9, 11, 37], although Price and colleagues reported a weak correlation between FEV 1 and PIF among participants who were discharged after hospitalization for a COPD exacerbation in a small retrospective observational study [38].Results of another observational study that included 213 participants with advanced COPD also demonstrated signifcantly lower values for FVC percent predicted and IC percent predicted, but not for FEV 1 percent predicted, in participants with sPIF (defned as <60 L/min against the simulated resistance of Diskus, a medium-low DPI) when compared with those with oPIF (≥60 L/min against the simulated resistance of Diskus) [9].Te diferences in FVC percent predicted and IC percent predicted could be due to the greater air trapping and hyperinfation.Alternatively, lower lung volumes may be a result of lower inspiratory efort [9].
Tis analysis demonstrated that participants with sPIF had a signifcantly lower height, weight, and BMI, longer smoking history and COPD duration, and signifcantly more severe airfow obstruction than participants with oPIF.Previous studies have also shown that characteristics such as Canadian Respiratory Journal age, sex (female), and markers of hyperinfation are consistently associated with the presence of sPIF [5-7, 9, 11, 39].Height and measures of lung function such as FVC percent predicted and IC percent predicted may also be associated with the presence of sPIF [5,9].Tus, our results generally support the published data.
Tis analysis has some limitations.In trial 0149, only participants with sPIF were enrolled; therefore, an estimate of prevalence cannot be provided from this analysis.Data for CAT and SGRQ tests were collected only in trial 0128, leading to considerably fewer participants in the sPIF subgroup than in the oPIF subgroup.Peak inspiratory fow against the resistance of medium-low resistance DPIs was measured only in trial 0149; in trial 0128, these values were estimated using the predictive model correlating PIF with medium-low resistance DPIs and high resistance DPIs from trial 0149 [20].An additional limitation is that the trials on which this analysis was based had strict inclusion and exclusion criteria (e.g., exclusion of patients with signifcant comorbid pulmonary conditions or, for trial 0128, elevated cardiovascular risk) [12][13][14], and the trial population may therefore not be representative of the real-world population with COPD.
According to the 2024 GOLD strategy report, an individualized assessment of each patient's symptoms and future risk of exacerbations should be made before prescribing treatment for COPD [1].Discovering measures and approaches that predict readmission following COPD exacerbation is needed to improve patient health.One study found that reduced PIF rate at discharge, a higher CAT score at discharge, frailty, and previous exacerbations were associated with hospital readmissions in patients with COPD [40].Although some studies have found no relationship between the likelihood of readmission and the presence of sPIF among participants hospitalized for an acute COPD exacerbation [10,41], others have shown that participants with sPIF are at increased risk of readmission [8].Specifcally, one retrospective analysis of patients who were     Canadian Respiratory Journal  Canadian Respiratory Journal hospitalized for acute exacerbation of COPD found that sPIF was common in these patients and sPIF predicted all-cause and COPD readmissions [8].Participants with sPIF may not receive an adequate dose of bronchodilators through DPI devices to ameliorate their symptoms; therefore, in addition to measuring airfow limitation by spirometry, a healthcare provider should consider measuring PIF against the simulated resistance of the DPI before prescribing treatment.It has been suggested that PIF measured against the simulated resistance of a specifc DPI may be used as a biomarker to identify patients who are likely or not likely to beneft from the DPI [42,43].In addition, characteristics such as sex, age, height, weight, BMI, and the markers of hyperinfation may also have utility in identifying participants at a higher risk for sPIF.Patients with sPIF are less likely to have a favorable response to DPIs and may be candidates for bronchodilator therapies administered via delivery systems that require low inspiratory eforts, such as pMDI, SMIs, and nebulizers.
In conclusion, in this analysis of pooled data from two phase 3 clinical trials, participants with COPD who had sPIF had signifcantly more dyspnea and worse health status than patients with oPIF, suggesting that sPIF is associated with a higher clinical burden than oPIF.