Background
Pulmonary hypertension (PH) is a life-threatening disorder with a variety of aetiologies [
1]. Because PH is a multifactorial condition, monotherapy focused on a single pathological pathway may be insufficient to halt disease progression. By acting on two or more biological pathways, combination therapies have the potential for increased efficacy over monotherapies. In patients with PH, two main approaches for combining treatments may be followed, with therapies introduced sequentially or concomitantly as ‘upfront’ combination therapy. Monotherapy is normally used initially, with additional therapy introduced if clinical deterioration occurs. Less frequently, combination treatment is used as first-line therapy to exploit the ‘hit hard and early’ model, which aims to use early and aggressive treatment to halt disease progression [
2]. This approach is also recommended in international guidelines for patients with PH, for those with severe disease (defined as class IV according to the World Health Organisation functional class system) [
3].
Treatment guidelines also suggest combining established pharmacotherapies for patients with PH who do not respond adequately to monotherapy, but do not recommend particular combinations or regimens [
3]. During a 3 year study employing pre-defined treatment goals to guide therapeutic decisions, combination therapy was eventually required by almost half of patients initially prescribed monotherapy [
4]. Several studies have examined the combination of the prostanoid iloprost and the phosphodiesterase type 5 (PDE-5) inhibitor sildenafil in the treatment of patients with PH. In acute haemodynamic testing, combining these drugs led to a greater reduction in pulmonary vascular resistance (PVR) than each agent alone [
5]. Furthermore, patients with pulmonary arterial hypertension (PAH) showed improved exercise capacity and haemodynamics when given sildenafil as an add-on to existing iloprost therapy [
6]. Randomized controlled trials directly comparing the efficacy of iloprost and sildenafil have not been undertaken, although a meta-analysis found no significant difference in efficacy between these therapies [
7]. The aim of this study was to examine the long-term effect of different combination regimens of inhaled iloprost and oral sildenafil on the survival and disease progression of patients with PH.
Methods
Study design
This was an observational study [
8] of patients in the Giessen Pulmonary Hypertension Registry, a single-centre registry including more than 2500 patients with newly diagnosed disease. The registry started in 1993. For this study, the patients who met the eligibility criteria had been enrolled from 1993 to 2013. Adult patients who received a combination of inhaled iloprost and oral sildenafil were eligible for inclusion. Patients who received intravenous iloprost or sildenafil, or who had begun treatment with therapies other than iloprost or sildenafil, were excluded. Each patient gave informed consent to participate. The study was approved by the University of Giessen Institutional Review Board (reference number 266/11) and followed the principles of the Declaration of Helsinki.
Three treatment regimens were studied: iloprost monotherapy followed by addition of sildenafil (iloprost/sildenafil); sildenafil monotherapy followed by addition of iloprost (sildenafil/iloprost); and upfront combination therapy of iloprost and sildenafil (iloprost + sildenafil). No pre-defined protocol was followed; treatment and doses were tailored to the individual patient’s needs and optimized by dose-titration.
Outcome measures
The primary outcome measure was transplant-free survival, as calculated by Kaplan–Meier analysis. As this was a retrospective study of patient records, complete information could not be obtained in all cases. When available, 6-minute-walk distance (6MWD), New York Heart Association (NYHA) functional class, pulmonary arterial pressure (PAP), PVR, and cardiac output were analysed. Changes were compared using intra-individual paired analysis (i.e. including only patients with both baseline and post-treatment data available). Values were determined pre-treatment (baseline), 3 months after monotherapy initiation, before combination therapy initiation (post-monotherapy baseline), and 3 months after combination therapy initiation. Patients lost to follow-up were classified as having withdrawn alive on the date of last contact.
Statistical methods
Data are presented as mean (standard deviation) or median (interquartile range), as applicable. The log-rank test was used to analyse differences in cumulative transplant-free survival; analysis of variance was applied to test for differences between groups; and the paired t-test (two-tailed) was used to examine changes in response to therapy. Cox regression, defining iloprost as the reference, was applied to control for possible confounders in survival analysis, correcting for NYHA functional class, 6MWD, and cardiac output. The Kruskal–Wallis test was performed to test for differences in parameters with skewed distributions.
Discussion
In the treatment of patients with PH, clinical studies have evaluated combinations of major pharmacological classes of medical therapies, i.e. endothelin receptor antagonists and prostanoids [
9‐
12], endothelin receptor antagonists and PDE-5 inhibitors [
13,
14], and prostanoids and PDE-5 inhibitors [
5,
6,
15‐
19]. However, only one study, of administration of the prostanoid treprostinil for up to 2 years in patients receiving oral background PAH therapy, examined long-term outcomes (survival and clinical worsening [defined as addition of a new PAH therapy, discontinuation due to disease progression, or death]) [
20]. A meta-analysis of randomized controlled studies in patients with PAH found that, compared with monotherapy, combination therapy significantly reduced clinical deterioration, increased 6MWD, and improved haemodynamics [
21]. However, no significant difference in mortality was observed between patients treated with monotherapy and those receiving combination therapy.
Iloprost and sildenafil act via different pathways (stimulating cyclic adenosine monophosphate [cAMP] production and preventing cyclic guanosine monophosphate [cGMP] breakdown, respectively), but there is evidence of cross-talk between these pathways. Raised cGMP levels inhibit cAMP breakdown, and pre-treatment of erythrocytes in vitro with PDE-5 inhibitors potentiated cAMP release in response to treprostinil [
22]. In acute haemodynamic testing in patients with PH, the combination of sildenafil and iloprost produced a greater vasodilatory response than either agent alone [
5]. There are limited data showing the long-term benefits of combining sildenafil and iloprost, but in a 16 week study the addition of sildenafil to long-term treatment with the prostacyclin epoprostenol improved exercise capacity and haemodynamics among patients with PAH compared with those receiving placebo [
23].
In our study, cumulative transplant-free survival was lower for patients who received upfront combination therapy than for those treated with initial monotherapy. At 1 year, the survival rate was 62.9 % for those who received combination therapy, compared with 95.1 % and 91.8 % for those first treated with inhaled iloprost or oral sildenafil monotherapy, respectively. However, before therapy, patients treated with iloprost + sildenafil had higher mean PVR and mean PAP than those who began monotherapy. Therefore, patients with the most severe disease had been assigned to receive upfront combination therapy, as is recommended in current international treatment guidelines for patients with PH [
3]. This approach, of treating patients with severe PAH with upfront inhaled iloprost and oral sildenafil therapy, was taken in a separate study of eight patients of NYHA functional class IV who were unable to perform a 6MWD test. Following treatment, all patients had an improvement in their functional class and were able to complete a 6MWD test, though one patient later underwent lung transplantation and subsequently died [
24]. Similarly, for the small number of patients for whom measurements were recorded in our study (
n = 16), 6MWD significantly increased following combination treatment.
Among patients treated initially with monotherapy, transplant-free survival was higher for those receiving iloprost/sildenafil than for those treated with sildenafil/iloprost. Patients treated with inhaled iloprost also remained on monotherapy longer than patients beginning oral sildenafil monotherapy. When paired recordings were available, the benefit of sequential therapy on exercise capacity was also observed for both drug regimens, with 6MWD significantly higher than pre-treatment values after 3 months of combination therapy. The results from this study suggest that when combining iloprost and sildenafil in a step-wise manner the optimal treatment regimen may be initial monotherapy with iloprost followed by add-on sildenafil if clinical deterioration occurs.
This study has limitations. Owing to the retrospective design, patients were not randomly assigned to each treatment, as highlighted by significant differences in baseline characteristics between the patient groups. Furthermore, complete data were not available for functional class, haemodynamic parameters, and exercise capacity, and there was the potential for selection bias. Unlike treatment with iloprost monotherapy, sildenafil monotherapy resulted in significant improvements in haemodynamics compared with pre-treatment values. Despite this, transplant-free survival was shorter in patients initially treated with sildenafil than among those who received iloprost therapy first. This is difficult to explain, and may reflect differences in the baseline characteristics between these patient groups or be the result of an unrecognized confounding factor. For these reasons, caution is needed when comparing the effectiveness of these two monotherapies. These outcomes also need to be viewed in the context of treatment practices over the course of this study, because patients who received initial iloprost therapy were admitted to the study centre before those first treated with sildenafil. In Europe, iloprost was approved for the treatment of PAH 2 years before sildenafil. Thus, in the early years of the study, inhaled iloprost was the only treatment available and patients may have remained on monotherapy because no other treatment options were available. Furthermore, this study included patients from a period of 20 years, over which time there were significant changes in diagnostic practices and in the clinical management of PH, which may have influenced the outcomes of patients.
Competing interests
HG has received support and/or honoraria from Actelion, AstraZeneca, Bayer Pharma AG, GlaxoSmithKline, Janssen Cilag, Lilly, Pfizer, and United Therapeutics/OMT. RV has been an investigator and consultant for LungRx. WS has received speaker/consultancy fees from Bayer Pharma AG and Pfizer. FG has received research grants from Bayer Pharma AG, Ergonex, Encysive and Pfizer; honoraria from Actelion, Bayer Pharma AG, Encysive, Novartis, and Pfizer; and has been a paid consultant for Nycomed (Altana Pharma). HAG has received fees from Actelion, Bayer Pharma AG, Gilead, GSK, Lilly, LungRx, Novartis, and Pfizer. The other authors declare that they have no competing interests.
Authors’ contributions
HG contributed to the conception and design of the study, analysis of the data, drafting the manuscript and revising the article for intellectual content. NS contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. KM contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. MR contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. RV contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. WS contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. FG contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. H-AG contributed to the conception and design of the study, analysis of the data, and revising the article for intellectual content. DB contributed significantly to the design of the study and the interpretation of the data. All the authors gave final approval of the published article and are accountable for all aspects of the work.