Sage Journals: Discover world-class research

Abstract

Background:

Roflumilast, a phosphodiesterase 4 inhibitor, has been shown to improve lung function and reduce exacerbation rates, but is associated with adverse events (AEs). The purpose of this study was to systematically review the clinical effectiveness and safety of roflumilast.

Methods:

A systematic search was made of MEDLINE, Cochrane trials database, DARE and CINAHL. Randomized, controlled trials of more than 12 weeks’ duration comparing roflumilast with placebo were reviewed. Studies were pooled to yield relative risk (RR), incident rate difference or weighted mean differences with 95% confidence intervals (CIs).

Results:

Eight trials (8698 patients) met the inclusion criteria. Roflumilast significantly reduced moderate to severe exacerbations (RR 0.85; 95% CI 0.80−0.91) compared with placebo, but not severe exacerbations (RR 0.83; 95% CI 0.68–1.01) or mortality (RR 0.90; 95% CI 0.63–1.28). Roflumilast significantly improved lung function relative to placebo, but not quality of life measures. AEs (RR 1.11; 95% CI 1.03–1.19) and discontinuations of treatment due to AEs (RR 1.63; 95% CI 1.45–1.84) were significantly more frequent with roflumilast than placebo. In the chronic obstructive pulmonary disease (COPD) Safety Pool (12,054 patients), the overall incidence of serious AEs did not differ between groups. However, atrial fibrillation (0.4% versus 0.2%; p = 0.02) and suicidality (0.08% versus 0%) were more frequent with roflumilast than placebo.

Conclusions:

The efficacy of roflumilast appears modest compared with other available therapies for COPD. Further studies are needed to investigate the risk–benefit ratio and long-term safety of roflumilast before its wider use.

Keywords

chronic obstructive meta-analysis outcome assessment (health care)phosphodiesterase 4 inhibitors pulmonary disease

Introduction

Chronic obstructive pulmonary disease (COPD) continues to be a major cause of disability, mortality and healthcare costs worldwide. In 2005, 210 million people were diagnosed with COPD worldwide and 3 million died of the disease. In the USA, it is estimated that more than 12 million people are affected by the disease and COPD has become the third leading cause of death after cardiovascular disease and malignancy [Kochanek et al. 2011]. It is predicted that COPD will also become the third leading cause of death worldwide by 2030 [Halbert et al. 2006].

The clinical efficacy of currently available therapies is still suboptimal despite the advances in medical technology. Continuous oxygen therapy and smoking cessation are the only interventions shown to reduce mortality. There is a pressing need for additional intervention for patients with COPD, as no existing treatment has been shown to reduce disease progression [Barnes, 2008].

Roflumilast is a selective phosphodiesterase 4 (PDE4) inhibitor that leads to decreased recruitment and activation of proinflammatory cells and decreased production and release of proinflammatory cytokines in lung cells. Although the exact mechanism by which roflumilast exerts its therapeutic action is not completely understood, it is thought to be related to increased intracellular cyclic adenosine monophosphate production in the lung tissues and inflammatory cells. Clinical studies have shown that roflumilast improves lung function and reduces exacerbations [Kelly Freeman, 2012].

The US Food and Drug Administration (FDA) approved roflumilast in March 2011 for clinical use to reduce the risk of COPD exacerbations, although an advisory panel of outside experts had opposed it by a 5 to 10 vote. The advisory committee felt that a small improvement in lung function did not outweigh its significant side effects, which primarily consisted of gastrointestinal and psychiatric problems, including suicidal behavior, which led to increased withdrawal from clinical studies [FDA, 2010]. Roflumilast was finally approved by the FDA after its indication was made more restrictive and a warning about the psychiatric side effects was added to the drug label. The place and role of roflumilast in COPD management are currently unclear. The purpose of this study was to systematically review the clinical effectiveness and safety of roflumilast, the first-in-class orally active PDE4 inhibitor in the USA and a number of European countries.

Methods

Identification of trials

We identified all relevant clinical trials that studied clinical efficacies and safety of roflumilast. We independently searched the National Library of Medicine’s Medline database for studies in any language published from 1946 to 2 March 2012 using the Medical Subject Headings and keywords: Phosphodiesterase Inhibitors or roflumilast AND Pulmonary Disease, Chronic Obstructive or COPD. In addition, we searched the Cochrane Central Register of Controlled Trials, the Database of Abstracts of Reviews of Effects and the Cumulative Index to Nursing and Allied Health Literature (CINAHL). We also reviewed our personal files and searched the internet for relevant information. Bibliographies of all selected articles and review articles that included information on roflumilast in COPD were reviewed for other relevant articles.

Study selection and data extraction

Both authors independently abstracted data from all studies using a standardized form. Data were abstracted on study design, study size, population, severity of illness and the impact of roflumilast on the endpoints of interest. Disagreements regarding values or analyses were resolved by discussion.

To be included in the analysis, a study had to be a randomized, controlled trial comparing the clinical effectiveness of roflumilast with that of placebo and had to have at least one of the following as a primary outcome variable: forced expiratory volume in 1 s (FEV₁), COPD exacerbation, quality of life score or mortality rate. The quality of included studies was evaluated using the Consolidated Standards of Reporting Trials (CONSORT) criteria [Moher et al. 2010]. Each study was given a score out of 25, reflecting how many of the 25 CONSORT items were complied with (each item being given equal weighting). Our meta-analysis was conducted in accordance with the Quality of Reporting of Meta-analyses of Randomized Control Trials statement [Moher et al. 1999].

Data analysis

Mortality, adverse event (AE) and exacerbation rates were dichotomous variables. FEV₁ and quality of life scores were continuous variables. The data analysis was performed using meta-analysis software (StatsDirect 2.7.8, StatsDirect Ltd, Cheshire, UK). The results were expressed as relative risk (RR) or incidence rate difference (IRD) for dichotomous outcomes and weighted mean difference (WMD) for continuous outcomes, along with their 95% confidence intervals (CIs). A Z-test was performed to examine the overall effect. We tested heterogeneity between trials with the I² statistic. I² indicates the proportion of between-study heterogeneity that is beyond chance and ranges between 0% and 100%; higher I² values indicate higher amounts of heterogeneity. We used the DerSimonian and Laird random-effects model to pool rate differences across studies.

Sensitivity analyses

Sensitivity analyses were performed to assure the robustness of the results by excluding the clinical trials one by one from the pooled analyses and using odds ratio and risk difference. A multivariate regression meta-analysis was performed to identify a subgroup in which roflumilast could be more efficacious to reduce moderate to severe exacerbations. The variables included were mean age, baseline FEV₁, duration of the clinical trials, concomitant use of tiotropium or salmeterol, and history of prior COPD exacerbations. Meta-regression analyses were performed using Stata 10 (StataCorp LP, College Station, TX, USA).

Results

Study selection

The electronic database searches identified 104 citations. Ninety-five studies were excluded on more detailed review. The remaining nine studies were reviewed for duplicate publications. Two studies were excluded for duplication [Bateman et al. 2011; Rutten-van Molken et al. 2007]. One study was excluded due to crossover design [Grootendorst et al. 2007]. We found that two studies included two separate trials. We included six publications reporting the total of eight clinical trials (Figure 1) [Calverley et al. 2007, 2009; Fabbri et al. 2009; Lee et al. 2011; Rabe et al. 2005; Rennard et al. 2011]. Detailed information on AEs which were not included in the original studies was found on the internet (COPD Safety Pool) [Forest Research Institute, 2010]. These safety data included the total of 12,054 patients (6972 receiving roflumilast) enrolled in 14 controlled parallel group studies including all eight studies used in our meta-analysis except for the study by Lee and colleagues [Lee et al. 2011]. This database was used in additional analyses for AEs and mortality.

Figure 1.

Flow of Study Selection.

Study characteristics and patients

Characteristics of included studies are summarized in Table 1. The duration of studies ranged from 12 to 52 weeks. A total of 8698 patients with COPD were included in this analysis; 4459 were treated with roflumilast (500 μg a day) and 4239 received placebo. The mean age of included patients was 64.2 and 74% were men. The mean prebronchodilator FEV₁ was 1.14 liters and 41% of predicted. Short-acting β agonists were allowed for rescue treatment in all studies. Concomitant use of inhaled corticosteroid (ICS; ≤2000 μg equivalent to beclomethasone) was allowed in two trials. The CONSORT score ranged from 20 to 24 and the mean score was 22. All studies were sponsored by the same pharmaceutical company and the sponsor was involved in the data interpretation and preparation of manuscripts.

Table 1.

Clinical trials.

Study [year] (code)	n	Duration (weeks)	Mean age	Men	Mean baseline FEV₁, liters (% pred)	Treatment arms	Concomitant ICS allowed	Relevant exclusion criteria	CONSORT score
Rabe [2005] (M2-107)	835	24	64	74%	1.42 (51%)	Roflumilast versus placebo	No	LTOT	20
Calverley [2007] (M2-112)	1514	52	65	76%	1.04 (41%)	Roflumilast versus placebo	Yes	LTOT, CPA	21
Calverley [2009] (M2-124)	1523	52	64	71%	1.07 (35%)	Roflumilast versus placebo	No	CPA, ECG	24
Calverley [2009] (M2-125)	1568	52	64	80%	0.97 (32%)	Roflumilast versus placebo	No	CPA, ECG	24
Fabbri [2009] (M2-127)	933	24	65	66%	1.42 (52%)	Roflumilast + salmeterol versus placebo + salmeterol	No	CPA	24
Fabbri [2009] (M2-128)	743	24	64	71%	1.48 (53%)	Roflumilast + tiotropium versus placebo + tiotropium	No	CPA	24
Rennard [2011] (M2-111)	1173	52	64	67%	0.94 (36%)	Roflumilast versus placebo	Yes	LTOT*	20
Lee [2011]	410	12	67	93%	1.28 (50%)	Roflumilast versus placebo	No	LTOT*, CPA	20

Requiring over 16 h a day.

CONSORT, Consolidated Standards of Reporting Trials; CPA, clinically significant cardiopulmonary abnormalities; ECG, clinically relevant electrocardiogram findings not related to chronic obstructive pulmonary disease; FEV₁, forced expiratory volume in 1 s; ICS, inhaled corticosteroid; LTOT, long-term oxygen therapy.

Exacerbations

The definitions of exacerbations varied among studies and are summarized in Table 2. We examined the event rates per patient year as well as the number of patients experiencing at least one exacerbation. The results are shown in Table 3. The rate of overall exacerbations (mild, moderate or severe) was significantly reduced with roflumilast compared with placebo (IRD −0.41 events/patient year; 95% CI −0.72 to −0.11). The rate of moderate or severe exacerbations was also lower with roflumilast than placebo (IRD −0.14 events/patient year; 95% CI −0.22 to −0.06) (Figure 2). However, this was primarily driven by moderate exacerbations as evidenced by no significant difference in the rate of severe exacerbations between groups (IRD −0.01; 95% CI −0.04 to 0.02; p = 0.49) (Table 3).

Table 2.

Definitions of exacerbations.

Study	Definitions
Rabe (M2-107)	Mild=increase in bronchodilator use on ≥2 consecutive days (>4 puffs SABA/day above mean use during week before randomization) without additional healthcare contact.
	Moderate=home management with administration of oral corticosteroid or unscheduled healthcare contact, or both.
	Severe=required hospital admission or ER treatment.
Calverley (M2-124, M2-125)	Moderate=requiring systemic corticosteroidsSevere=associated with admission or death
Fabbri (M2-127, M2-128)	Mild=increase in SABA of ≥3 puffs/day on at least 2 consecutive days
	Moderate=requiring oral corticosteroids (not antibiotics)
	Severe=associated with hospitalization or death
Rennard (M2-111, M2-112)	Moderate=symptomatic deterioration treated with systemic steroids and/or antibiotics
	Severe=requiring hospitalization
Lee	Moderate=requiring systemic corticosteroids and/or antibiotics
	Severe=requiring hospitalization

SABA, short-acting β agonist.

Table 3.

Exacerbation rates.

Number of patients experiencing at least one exacerbation	RR Rof: plbo	95% CI	p value	Study
All	0.81	0.72 to 0.91	0.0004	M2-127/128, M2-107, Lee
Moderate or severe	0.86	0.79 to 0.93	0.0002	M2-127/128, M2-124/125, M2-107, Lee
Severe	0.83	0.68 to 1.00	0.05	M2-124/125, M2-107, Lee

Number of events per patient-year	IRD Rof–plbo	95% CI	p value

All	−0.41	−0.72 to −0.11	0.007	M2-127/128, M2-107, Lee
Moderate or severe	−0.14	−0.22 to −0.06	0.0005	M2-124/125, M2-111/112, M2-107
Severe	−0.01	−0.04 to 0.02	0.49	M2-124/125, M2-107

IRD, incidence rate difference; plbo, placebo; Rof, roflumilast; RR, relative risk.

Figure 2.

Summary effects of roflumilast on relative risk (a) and incident rate difference (b) of moderate and severe exacerbations.

Health status

Three trials evaluated health-related quality of life using St George’s Respiratory Questionnaire (SGRQ) score (range 0–100, with lower scores indicating a better quality of life; a minimum clinically important difference in the total score being 4 points) [Calverley et al. 2007; Rabe et al. 2005; Rennard et al. 2011]. Improvement in SGRQ scores was greater with roflumilast compared with placebo. However, the difference was not statistically or clinically significant (pooled WMD −1.32; 95% CI −3.20 to 0.55; p = 0.07). M2-124 and M2-125 trials [Calverley et al. 2009] reported Euroquol 5-dimension scores and found no difference between roflumilast and placebo groups (pooled WMD 0.002; 95% CI −0.01 to 0.017; p = 0.76).

Lung function

Small improvement in mean prebronchodilator FEV₁ was observed with roflumilast (range 9 to 65 ml) in all studies whereas it was decreased or relatively unchanged with placebo (range −42 to 8 ml). Compared with placebo, improvement in prebronchodilator FEV₁ was significantly greater with roflumilast (pooled effect size WMD 55 ml; 95% CI 42–68; p < 0.0001) (Figure 3). A similar result was noted for other lung function parameters (Table 4).

Figure 3.

Summary effects of roflumilast on prebronchodilator forced expiratory volume in 1 s.

Table 4.

Improvement in lung function with roflumilast compared with placebo.

Lung function	WMD (ml)	95% CI	p value	Study
FEV₁ prebronchodilator	55	42–68	<0.0001	M2-107, M2-111/112, M2-124/125, M2-127/128, Lee
FEV₁ postbronchodilator	59	47–71	<0.0001	M2-107, M2-111/112, M2-124/125, M2-127/128, Lee
FVC prebronchodilator	89	60–118	<0.0001	M2-124/125, M2-127/128, Lee
FVC postbronchodilator	84	65–103	<0.0001	M2-107, M2-112, M2-124/125, M2-127/128, Lee

FEV₁, forced expiratory volume in 1 s; FVC, forced vital capacity; WMD, weighted mean difference.

Safety

All included studies reported AEs occurring in at least 2% of patients [Calverley et al. 2007, 2009; Fabbri et al. 2009; Lee et al. 2011; Rabe et al. 2005; Rennard et al. 2011]. A total of 8696 patients were included in the safety analysis. Among 4459 patients receiving roflumilast 500 μg, 1392 and 2864 were exposed for 6 months and 1 year, respectively. In the pooled analysis, 47% and 28% of patients in the roflumilast and placebo groups reported at least one AE (RR 1.11; 95% CI 1.03–1.19). Discontinuation of treatment due to AEs was significantly more frequent with roflumilast compared with placebo (15% versus 9.2 %; RR 1.62; 95% CI 1.44–1.83; p < 0.0001) (Table 5). Gastrointestinal (diarrhea, nausea and decreased appetite), nervous system (headache, dizziness and tremor), psychiatric (insomnia) and constitutional AEs (weight loss) were significantly more common with roflumilast compared with placebo (Table 5). Infectious AEs were more common with placebo compared with roflumilast but the difference was not statistically significant.

Table 5.

Adverse events that occurred in 2% or more of patients.

	Roflumilast (%) (n = 4459)	Placebo (%) (n = 4239)	RR (95% CI)	Risk difference (95% CI)	NNH (95% CI)
All AEs	2096 (47)	1172 (28)	1.11 (1.03 to 1.19)	0.19 (0.17 to 0.21)	5 (5 to 6)
Gastrointestinal	715 (16)	181 (4.7)	4.50 (3.13 to 6.47)	0.12 (0.11 to 0.13)	8 (8 to 9)
Musculoskeletal	70 (1.6)	49 (1.2)	1.44 (1.00 to 2.07)	0.004 (−0.001 to 0.009)	241 (–1283 to 110)
Infectious	740 (17)	751 (18)	0.98 (0.84 to 1.15)	-0.011 (−0.027 to 0.005)	–90 (–37 to 216)
Nervous system	189 (4.2)	79 (1.9)	2.43 (1.88 to 3.15)	0.024 (0.017 to 0.031)	42 (32 to 60)
Psychiatric	53 (1.2)	22 (0.5)	2.43 (1.48 to 3.98)	0.007 (0.003 to 0.011)	149 (93 to 351)
Weight loss	329 (7.4)	90 (2.1)	4.22 (2.70 to 6.60)	0.05 (0.04 to 0.06)	19 (16 to 23)
AE leading to discontinuation	669 (15)	392 (9.2)	1.63 (1.45 to 1.84)	0.06 (0.04 to 0.07)	17 (14 to 23)

AE, adverse event; CI, confidence interval; NNH, number needed to harm; RR, relative risk.

The all-cause mortality rate was reported in four trials including 5014 patients [Calverley et al. 2007, 2009; Lee et al. 2011]. None of the studies were sufficiently powered to detect a difference in mortality. The mortality rate did not differ between the roflumilast and placebo groups (2.2% versus 2.5 %; RR 0.90; 95% CI 0.63–1.29; p = 0.56) (Figure 4). In the COPD safety pool, 84 deaths (1.5%) were reported in the 500 μg roflumilast group and 86 (1.6%) in the placebo group (p = 0.69) [Forest Research Institute, 2010]. COPD (25%) and pneumonia (11%) were the most common causes of death. None of the deaths were considered likely or definitely related to the study medication [Forest Research Institute, 2010].

Figure 4.

Summary effects of roflumilast on mortality.

The incidence of serious AEs (SAEs) was reported in six trials comprising 7022 patients [Calverley et al. 2007, 2009; Lee et al. 2011; Rabe et al. 2005; Rennard et al. 2011]. In our pooled analysis, SAEs occurred in 17.5% and 18.4% with roflumilast and placebo. The difference between the groups was not statistically significant (RR 1.06; 95% CI 0.86–1.33; p = 0.57; I² = 55%). In the COPD safety pool including the total of 11,257 patients (500 μg roflumilast = 5766; placebo = 5491), COPD-related SAEs were significantly less frequent with roflumilast (5.8% versus 7.1%; p = 0.008), but atrial fibrillation was more common with roflumilast compared with placebo (0.4% versus 0.2%; p = 0.02) [Forest Research Institute, 2010]. Respiratory failure, myocardial infarction and chest pain were reported more frequently with placebo but the differences were not statistically significant. Overall incidence of SAEs was also similar between the roflumilast and placebo groups (13.5% versus 14.2%; p = 0.3) (Figure 5). Rare but serious psychiatric SAEs were also reported in the COPD safety pool [Forest Research Institute, 2010]. There were three completed suicides (0.05%) (one while on the medication and two > 3 weeks after discontinuation) and two suicide attempts (0.03%) in 6563 patients treated with roflumilast (500 μg = 5766; 250 μg = 797) and no completed or attempted suicides in 5491 patients on placebo [Forest Research Institute, 2010].

Figure 5.

Summary of serious adverse events from COPD Safety Pool. A-fib, atrial fibrillation; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; SAE, serious adverse event.

Sensitivity analyses

Sensitivity analyses using odds ratio and risk difference did not affect the results. Excluding the clinical trials one by one from the pooled analyses did not affect the results. Funnel plots as well as the Begg-Mazumdar (Kendall’s tau) and Egger tests showed no evidence of publication bias for the all efficacy and safety endpoints except for the pooled analysis for SAEs (Egger test, p = 0.004). The univariate and multivariate regression meta-analyses failed to identify any subgroup in which roflumilast would be more efficacious to reduce moderate to severe exacerbations. None of the variables, including mean age, FEV₁, duration of the clinical trials, concomitant use of tiotropium bromide or salmeterol and history of prior exacerbations significantly affected the study results.

Discussion

Our meta-analysis and systematic review confirmed that roflumilast improved lung function compared with placebo. However, no significant difference was noted in the mortality rate and health-related quality of life between groups. The rate of moderate to severe exacerbations was significantly lower with roflumilast but the rate of severe exacerbations did not differ between groups. Discontinuation of treatment due to AEs was significantly more frequent with roflumilast and overall incidence of AEs was also significantly increased with roflumilast. The overall incidence of SAEs did not differ between groups in our pooled analysis as well as in the COPD safety pool. COPD-related SAEs were less frequent but the incidence of atrial fibrillation was significantly more common with roflumilast compared with placebo. In the COPD safety pool, suicidality was more common with roflumilast (0.08 versus 0%).

Considering all the clinically pertinent and statistically significant findings, if 1000 patients with moderate to severe COPD are treated with roflumilast for 1 year, 140 moderate to severe exacerbations (primarily moderate ones that would require a short course of therapy with antibiotics or prednisone) would be avoided, however 200 patients would experience at least one AE, of which 60 would discontinue the medication, 2.5 patients would develop atrial fibrillation and 0.8 patients would attempt suicide, of which 0.5 patients would succeed. Overall quality of life would not be affected with the additional cost of $207 a month [Anonymous, 2011].

Although roflumilast is approved for patients with COPD with severe and very severe airflow limitation and a history of exacerbations, its place in the current armamentarium of COPD treatments and the risk–benefit ratio are still at the center of debate [Puhan, 2011]. It is concerning that another PDE4 inhibitor, cilomilast, was rejected by the FDA advisory panel in September 2003, based on concerns over the efficacy of the agent, as well as gastrointestinal side effects. However, the FDA temporarily approved cilomilast in October 2003 but the final approval is contingent on the outcome of additional efficacy and safety studies [Giembycz, 2006], which are still ongoing. The risk–benefit ratio of the entire class of orally active PDE4 inhibitors appears to be in question.

Some patients continue to have symptoms or repeated exacerbations despite combination therapy with a long-acting β agonist (LABA) and ICS. Additional pharmacological treatment options in such a refractory disease may include a long-acting muscarinic antagonist (LAMA), such as tiotropium bromide, theophylline or roflumilast. Triple inhaled therapy (LAMA plus LABA plus ICS) has been shown to reduce COPD exacerbations as well as hospitalizations and improve lung function, respiratory symptoms and quality of life without much safety concern [Aaron et al. 2007; Welte et al. 2009]. A randomized, controlled trial of roflumilast with the background therapy of an ICS plus LABA combination is currently ongoing [Calverley et al. 2012].

Theophylline, a nonselective phosphodiesterase inhibitor, could be compared with roflumilast. Although theophylline can be toxic and close monitoring of drug levels is necessary, it has been shown to improve respiratory symptoms, gas exchange, lung function and respiratory muscle function in patients with COPD [Murciano et al. 1989]. In a meta-analysis of 20 randomized, controlled trials, the mean improvement in FEV₁ with theophylline was 100 ml compared with placebo [Ram et al. 2002], which is much greater than that of roflumilast (52 ml) noted in this meta-analysis. Although comparing the efficacy of medications across studies is hampered by biases as studies commonly use different selections criteria, the above comparison may still be valid given comparable baseline characteristics between the two meta-analyses. The mean age and baseline FEV₁ ranged from 58 to 69 years and 0.96 to 1.15 liters in the theophylline meta-analysis [Ram et al. 2002], which are quite comparable to those of the current meta-analysis.

In regards to other clinical outcomes for theophylline, it did improve a quality life measure in one study [Guyatt et al. 1987], which has not been shown with roflumilast. The use of theophylline was also associated with a small reduction in COPD exacerbations in a large retrospective study [Cyr et al. 2008]. The incidence of gastrointestinal AEs was more frequent with theophylline than with placebo (RR 7.67; 95% CI 1.47–39.94) but the dropout rate did not differ between groups (RR 0.80; 95% CI 0.23–2.76; p = 0.72) in the meta-analysis [Ram et al. 2002], which suggests most of the AEs associated with theophylline might have been mild. Clinical studies directly comparing theophylline and roflumilast are warranted to determine which medication would be preferred in cases with a refractory disease.

The recommended dosage of roflumilast is 500 μg once daily and the cost of 30 days’ maintenance treatment according to the average wholesale price listings is $207 [Anonymous, 2011], which is comparable to those of maintenance inhaled therapies. It is very unlikely that roflumilast would be more cost effective than other maintenance inhaled therapies such as a LAMA or a LABA plus ICS combination because roflumilast has not been shown to reduce hospitalizations which are known to have the greatest impact on the cost effectiveness in COPD management besides the acquisition costs of medications [Hilleman et al. 2000]. Hertel and colleagues concluded that roflumilast would be cost effective in the UK [Hertel et al. 2012]. However, the cost effectiveness was probably overestimated since they did not incorporate the costs and loss of quality of life related to AEs associated with roflumilast into their study.

Our meta-analysis has limitations. First, a combination therapy of an ICS and a LABA was not allowed in any of the included trials, which is often used in patients with severe COPD with recurrent exacerbations. The risk–benefit ratio of roflumilast could change in relation to concomitant therapies. In fact, the benefit of roflumilast was limited to the subgroup of patients who were taking concomitant ICS [Rennard et al. 2011]. A further study would be needed involving the patients who are receiving LABA/ICS combination therapy.

Second, roflumilast will likely be used in patients with recurrent exacerbations despite other therapies. Such patients would likely have other cardiopulmonary abnormalities and may be on long-term oxygen therapy (LTOT). However, most of the included studies excluded patients with significant cardiopulmonary abnormalities except for COPD. In addition, patients on LTOT were excluded in half of the clinical studies, as shown in Table 1. Therefore, the results from this analysis may not be generalizable in real-world practice.

Third, rare SAEs, such as atrial fibrillation and suicidality, were significantly more common with roflumilast but such events may have happened just by accident. It is unknown if this is a chance observation or if roflumilast is arrhythmogenic or suicidogenic at this time. The risk–benefit ratio of roflumilast could change when it is widely used in a real-world practice since patients with significant cardiopulmonary abnormalities were excluded in most of the included studies (Table 1). A long-term safety profile of roflumilast needs to be established with postmarketing surveillance since such information is typically unknown for an entirely new class of medicines and their long-term side effects could be revealed only in a postmarketing study as happened with rofecoxib [Cahana and Mauron, 2006].

In summary, the efficacy of roflumilast appeared relatively modest compared with other pharmacologic therapies for COPD. AEs leading to discontinuation of therapy were not uncommon with roflumilast. SAEs were rare but the healthcare providers should be aware of the potential for suicidality and atrial fibrillation. Postmarketing safety surveillance, an efficacy trial with the background therapy of an ICS plus LABA combination, and a head-to-head trial with theophylline are warranted to further investigate the risk– benefit ratio and the place of roflumilast in the current armamentarium of COPD therapies.

Footnotes

Acknowledgements

The author would like to thank Drs Dennis Suich and John Onofrio for valuable comments on an earlier draft of this paper.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement

The authors declare no conflicts of interest in preparing this article.

References

Aaron

Vandemheen

Fergusson

Maltais

Bourbeau

Goldstein

. (2007) Tiotropium in combination with placebo, salmeterol, or fluticasone-salmeterol for treatment of chronic obstructive pulmonary disease: a randomized trial. Ann Intern Med 146: 545–555.

Anonymous (2011) Roflumilast (Daliresp) for COPD. Med Lett Drugs Ther 53: 59–60.

Barnes

(2008) Emerging pharmacotherapies for COPD. Chest 134: 1278–1286.

Bateman

Rabe

Calverley

Goehring

Brose

Bredenbroker

. (2011) Roflumilast with long-acting beta2-agonists for COPD: influence of exacerbation history. Eur Respir J 38: 553–560.

Cahana

Mauron

(2006) The story of Vioxx-no pain and a lot of gain: ethical concerns regarding conduct of the pharmaceutical industry. J Anesth 20: 348–351.

Calverley

Martinez

Fabbri

Goehring

Rabe

(2012) Does roflumilast decrease exacerbations in severe COPD patients not controlled by inhaled combination therapy? the REACT study protocol. Int J Chron Obstruct Pulmon Dis 7: 375–382.

Calverley

Rabe

Goehring

Kristiansen

Fabbri

Martinez

(2009) Roflumilast in symptomatic chronic obstructive pulmonary disease: two randomised clinical trials. Lancet 374: 685–694.

Calverley

Sanchez-Toril

McIvor

Teichmann

Bredenbroeker

Fabbri

(2007) Effect of 1-year treatment with roflumilast in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 176: 154–161.

Cyr

Beauchesne

Lemiere

Blais

(2008) Effect of theophylline on the rate of moderate to severe exacerbations among patients with chronic obstructive pulmonary disease. Br J Clin Pharmacol 65: 40–50.

10.

Fabbri

Calverley

Izquierdo-Alonso

Bundschuh

Brose

Martinez

. (2009) Roflumilast in moderate-to-severe chronic obstructive pulmonary disease treated with longacting bronchodilators: two randomised clinical trials. Lancet 374: 695–703.

11.

FDA (Food and Drug Administration) (2010) Pulmonary-Allergy Drugs Advisory Committee (PADAC). Available at http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/Pulmonary-AllergyDrugsAdvisoryCommittee/UCM212109.pdf (accessed 27 August 2012).

12.

Forest Research Institute (2010) DAXAS® (roflumilast) in chronic obstructive pulmonary disease. Available at http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/Pulmonary-AllergyDrugsAdvisoryCommittee/UCM207379.pdf (accessed 27 August 2012).

13.

Giembycz

(2006) An update and appraisal of the cilomilast phase III clinical development programme for chronic obstructive pulmonary disease. Br J Clin Pharmacol 62: 138–152.

14.

Grootendorst

Gauw

Verhoosel

Sterk

Hospers

Bredenbroker

. (2007) Reduction in sputum neutrophil and eosinophil numbers by the PDE4 inhibitor roflumilast in patients with COPD. Thorax 62: 1081–1087.

15.

Guyatt

Townsend

Pugsley

Keller

Short

Taylor

. (1987) Bronchodilators in chronic air-flow limitation. Effects on airway function, exercise capacity, and quality of life. Am Rev Respir Dis 135: 1069–1074.

16.

Halbert

Natoli

Gano

Badamgarav

Buist

Mannino

(2006) Global burden of COPD: systematic review and meta-analysis. Eur Respir J 28: 523–532.

17.

Hertel

Kotchie

Samyshkin

Radford

Humphreys

Jameson

(2012) Cost-effectiveness of available treatment options for patients suffering from severe COPD in the UK: a fully incremental analysis. Int J Chron Obstruct Pulmon Dis 7: 183–199.

18.

Hilleman

Dewan

Malesker

Friedman

(2000) Pharmacoeconomic evaluation of COPD. Chest 118: 1278–1285.

19.

Kelly Freeman

(2012) Clinical considerations for roflumilast: a new treatment for COPD. Consult Pharm 27: 189–193.

20.

Kochanek

Murphy

Miniño

Kung

(2011) US Department of Health and Human Services, CDC, National Center for Health Statistics. National Vital Statistics Reports. Deaths: preliminary data for 2009. Available at http://www.cdc.gov/nchs/data/nvsr/nvsr59/nvsr59_04.pdf (accessed 27 August 2012).

21.

Lee

Hui

Mahayiddin

Roa

Jr Kwa

Goehring

. (2011) Roflumilast in Asian patients with COPD: a randomized placebo-controlled trial. Respirology 16: 1249–1257.

22.

Moher

Cook

Eastwood

Olkin

Rennie

Stroup

(1999) Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of reporting of meta-analyses. Lancet 354: 1896–1900.

23.

Moher

Hopewell

Schulz

Montori

Gotzsche

Devereaux

. (2010) CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. J Clin Epidemiol 63: e1–e37.

24.

Murciano

Auclair

Pariente

Aubier

(1989) A randomized, controlled trial of theophylline in patients with severe chronic obstructive pulmonary disease. N Engl J Med 320: 1521–1525.

25.

Puhan

(2011) Phosphodiesterase 4 inhibitors for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 8: ED000028.

26.

Rabe

Bateman

O’Donnell

Witte

Bredenbroker

Bethke

(2005) Roflumilast – an oral anti-inflammatory treatment for chronic obstructive pulmonary disease: a randomised controlled trial. Lancet 366: 563–571.

27.

Ram

Jones

Castro

De Brito

Atallah

Lacasse

. (2002) Oral theophylline for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 4: CD003902.

28.

Rennard

Calverley

Goehring

Bredenbroker

Martinez

(2011) Reduction of exacerbations by the PDE4 inhibitor roflumilast-the importance of defining different subsets of patients with COPD. Respir Res 12: 18.

29.

Rutten-van Molken

van Nooten

Lindemann

Caeser

Calverley

(2007) A 1-year prospective cost-effectiveness analysis of roflumilast for the treatment of patients with severe chronic obstructive pulmonary disease. Pharmacoeconomics 25: 695–711.

30.

Welte

Miravitlles

Hernandez

Eriksson

Peterson

Polanowski

. (2009) Efficacy and tolerability of budesonide/formoterol added to tiotropium in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 180: 741–750.

Efficacy and safety of roflumilast in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis

Abstract

Background:

Methods:

Results:

Conclusions:

Keywords

Introduction

Methods

Identification of trials

Study selection and data extraction

Data analysis

Sensitivity analyses

Results

Study selection

Study characteristics and patients

Exacerbations

Health status

Lung function

Safety

Sensitivity analyses

Discussion

Footnotes

Acknowledgements

Funding

Conflict of interest statement

References