Medical management of chronic thromboembolic pulmonary hypertension

Rationale for use of PAH therapies in CTEPH

Drugs that target key pathways involved in the pathology of PAH are now established in the treatment of this condition [18, 19]. The similarities in pathological features between the two conditions provide a rationale for evaluating PAH therapies in patients with CTEPH [14].

Endothelin-1 is a powerful vasoconstrictor and smooth muscle mitogen synthesised and secreted by vascular endothelial cells. Levels of endothelin-1 are elevated in PAH and CTEPH [20–24], and recent evidence suggests a potential role for endothelin-1 in smooth muscle cell proliferation within the chronic clot in CTEPH, as well as in the distal arteriopathy [25]. Endothelin receptor antagonists (ERAs) act by blocking the type A endothelin receptor selectively, or both the type A and type B endothelin receptors nonselectively, preventing the vasoconstrictive and proliferative actions of endothelin-1 [18, 19, 26].

Prostacyclin, a highly potent vasodilator that can inhibit platelet aggregation and smooth muscle cell proliferation, is produced by endothelial cells [27]. Levels of prostacyclin are reduced in PAH, and thus prostanoid medications have been used [18, 19, 27].

Nitric oxide (NO), an endogenous vasodilator produced by the endothelium, inhibits platelet aggregation and growth of smooth muscle cells [28, 29]. NO activates soluble guanylate cyclase (sGC) to synthesise cyclic guanosine monophosphate (cGMP), a secondary signalling molecule that ultimately leads to decreased intracellular calcium and smooth muscle relaxation. Levels of endogenous NO synthase and NO metabolites are diminished in patients with PAH [28, 29], and thus the pathway is a key target for treatment. Inhibitors of phosphodiesterase (PDE)-5 prevent degradation of cGMP, while stimulators of sGC, such as riociguat, augment cGMP synthesis [18, 19]. There are no studies of prostacyclin or NO pathways specifically in CTEPH and the absence of an animal model greatly hampers research into the physiology of this condition. The rationale for the use of prostacyclin analogues, PDE-5 inhibitors and riociguat in CTEPH depends on its pathophysiological resemblance to PAH.

Open-label studies

Small, open-label trials of PAH therapies, including bosentan [30–32], treprostinil [33], epoprostenol [34–36] and sildenafil [37] have been conducted in patients with CTEPH. These trials are limited by their small patient cohorts, lack of randomisation and blinding, and, in many studies, by the absence of a control arm.

Randomised controlled trials

Efficacy data for randomised controlled trials of targeted therapy in patients with CTEPH [38–41] are summarised in table 1; safety results are in table 2. The efficacy and safety of the dual ERA bosentan were evaluated in the randomised, double-blind, placebo-controlled BENEFIT study [39]. Eligible patients (n=157) had symptomatic CTEPH (World Health Organization functional class (WHO FC) II–IV), 6-min walking distance (6MWD) <450 m, mPAP ≥25 mmHg, PCWP <15 mmHg and pulmonary vascular resistance (PVR) at rest ≥300 dyn·s·cm⁻⁵. Patients were randomised to placebo or bosentan at an initial dose of 62.5 mg twice daily for 4 weeks, subsequently increased to the target dose of 125 mg twice daily (patients weighing <40 kg were maintained at 62.5 mg twice daily). After 16 weeks of treatment, mean PVR (co-primary end-point) decreased from baseline in bosentan-treated patients and increased in the placebo arm (figure 1) [39]. There was no statistically significant effect of bosentan on 6MWD, the other co-primary end-point (bosentan +2.9 m, placebo +0.8 m; mean treatment effect +2.2 m, 95% CI −22.5–26.8 m). Statistically significant improvements for bosentan over placebo on haemodynamic parameters are presented in table 1. There was no statistically significant decrease in time to clinical worsening with bosentan versus placebo (hazard ratio 0.63, 95% CI 0.15–2.64), with few clinical worsening events in either treatment group (bosentan n=5 (6%), placebo n=3 (4%)). The most frequent adverse events in the bosentan arm were peripheral oedema, abnormal liver function test, headache, right ventricular failure, nasopharyngitis, vertigo and palpitations (table 2). Two patients (3%) in the bosentan group and four (5%) in the placebo group withdrew as a result of adverse events, and one death occurred in each treatment arm (both judged to be unrelated to study treatment).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Percentage of baseline pulmonary vascular resistance (PVR) at week 16 in patients receiving bosentan or placebo in the BENEFIT study [39]. Reproduced and modified from [39] with permission.

The ERA macitentan (10 mg·day⁻¹) is currently under investigation in the randomised, double-blind, placebo-controlled MERIT-1 study, which plans to enrol 78 patients with surgically inoperable CTEPH and is expected to report in the near future (ClinicalTrials.gov NCT02021292). The primary outcome is PVR at week 16. The results from this trial are expected soon. The study will be followed by an open-label extension, MERIT-2, which concentrates on safety and is expected to complete in 2018 (ClinicalTrials.gov NCT02060721). The ERA ambrisentan (5 mg·day⁻¹) was to have been investigated in patients with inoperable CTEPH in the AMBER-I study (ClinicalTrials.gov NCT01884675) and its open-label follow-up, AMBER-II (ClinicalTrials.gov NCT01894022), but only 33 out of 160 planned patients were enrolled (ambrisentan n=17, placebo n=16). The available data from AMBER-I have been published on the ClinicalTrials.gov website, showing a median change in 6MWD at 16 weeks (primary end-point) of +25 m in the ambrisentan arm versus −10 m in the placebo arm (outcome 1). Median change from baseline in PVR was −130 dyn·s·cm⁻⁵ with ambrisentan and −103 dyn·s·cm⁻⁵ with placebo (outcome 2).

Sildenafil, a PDE-5 inhibitor, was evaluated in a randomised, placebo-controlled pilot study in 19 patients with inoperable or persistent CTEPH (WHO FC II/III) and a 6MWD of 100–450 m [41]. Patients were randomised to sildenafil 40 mg three times daily or placebo for 12 weeks, at which time all placebo patients were switched to sildenafil, with a repeat assessment after 1 year. At 12 weeks the change from baseline in 6MWD (primary end-point) did not differ significantly between the sildenafil and placebo arms, but an improvement in PVR was seen (table 1). At 1 year (n=17), there were significant improvements from baseline in the sildenafil arm for 6MWD (+36 m, 95% CI 8–64 m; p=0.014) and PVR (−149 dyn·s·cm⁻⁵, 95% CI −228–−71 dyn·s·cm⁻⁵; p<0.001). It should be noted that the overall severity of CTEPH at baseline was lower in the placebo arm than in the sildenafil arm.

The Aerosolized Iloprost Randomized Study included 203 patients with severe pulmonary hypertension, including 57 with CTEPH, randomised to receive inhaled iloprost or placebo [42]. The study reported an overall positive treatment effect, but the subgroup analyses described results for primary pulmonary hypertension and a “nonprimary” population that included CTEPH, collagen vascular disease and use of appetite suppressants. The implications of this trial for CTEPH are unclear, because the results for CTEPH patients alone were not reported.

Real-world data

Single centres and registries have described the management of CTEPH: examples are summarised in table 3 [10, 43–48]. The patients reported may be postoperative, technically inoperable, those who refuse surgery and those with operable disease but who are not suitable for PEA because of comorbid conditions. Drug use, effectiveness and outcomes can be expected to vary between these groups. The national centre for PEA surgery in the UK [44], an international registry [10, 11] and several national pulmonary hypertension registries [43, 45–48] indicate that targeted medical therapy is used in both nonoperated and operated patients, but more frequently in the former. ERAs (particularly bosentan) are the most frequently used targeted therapy, the exception being in Latvia, where 90% of patients receive sildenafil [48]. Most patients receiving targeted therapy are treated with a single agent, particularly when treatment is initiated, although some receive combination therapy during the course of the disease [10, 43, 45–47]. The coadministration of riociguat with PDE-5 inhibitors is contraindicated, because excessive reductions of systemic blood pressure can occur with this combination [49]. Published outcomes data from registries are limited, but the most recent results show that 1-year survival with medical treatment for nonoperated patients with CTEPH is ≥88% (table 3). Recent data from the international CTEPH registry show that nonoperated patients who received targeted therapy had more severe disease at baseline, in terms of a shorter time from symptoms to diagnosis, a higher frequency of WHO FC III/IV, shorter 6MWD and a more severe haemodynamic profile (higher mPAP and PVR; lower cardiac output and PCWP) [10]. Survival is consistently worse among nonoperated than operated patients. Interpretation of this finding should consider differences in baseline characteristics and the diverse reasons for a patient being considered inoperable (e.g. comorbidity or distal disease).

The CHEST-1 study

The efficacy and safety of the dual-action sGC stimulator riociguat were assessed in CHEST-1, a phase III, multicentre, randomised, double-blind, placebo-controlled study [38]. Patients (n=261) had inoperable CTEPH, as assessed prospectively by an adjudication committee of experienced PEA surgeons, or persistent/recurrent pulmonary hypertension after PEA, with a 6MWD of 150–450 m, PVR >300 dyn·s·cm⁻⁵ and mPAP ≥25 mmHg. Eligible patients were randomised to receive placebo or riociguat at an initial dose of 1 mg three times daily. In the first 8 weeks of the study, the dose was adjusted according to systolic systemic arterial pressure and signs/symptoms of hypotension to a final individualised dose of up to 2.5 mg three times daily. Patients then received their individualised dose for a further 8 weeks. At week 16, 77% of patients still participating in the study were taking the maximal riociguat dose of 2.5 mg three times daily. During the study, the dose of the study drug was decreased in 18 patients (10%) in the riociguat group compared with three (3%) in the placebo group. At week 16, 6MWD (the primary end-point) had increased from baseline by a mean of 39 m in riociguat-treated patients compared with a 6 m decrease in the placebo arm (least-squares mean difference 46 m, 95% CI 25–67 m; p<0.001). In addition, significantly more patients experienced improvement/stabilisation of their WHO FC in the riociguat arm (33%/62%, respectively) compared with the placebo arm (15%/78%, respectively; p=0.003), although there was no statistically significant difference in the incidence of clinical worsening events between the two arms (riociguat 2%, placebo 6%). Significant benefits for riociguat versus placebo on key efficacy parameters are shown in table 1. The most frequently occurring adverse events in the riociguat arm were headache, dizziness, dyspepsia, peripheral oedema and nasopharyngitis (table 2). The most frequently occurring serious adverse events were right ventricular failure, syncope and haemoptysis (table 2). Deaths related to adverse events occurred in two patients (1%) in the riociguat group (heart failure n=1 and acute renal failure n=1) and in three patients (3%) in the placebo group (respiratory insufficiency n=1, circulatory arrest n=1 and cardiac arrest n=1). With 89% of patients receiving the higher doses of 2.5 and 2.0 mg three times daily at week 16, dose-dependent efficacy and safety have not been analysed.

Based on the results of the CHEST-1 trial, riociguat was licensed in Europe for the treatment of adult patients with WHO FC II/III and inoperable CTEPH or persistent/recurrent pulmonary hypertension after surgical treatment [49]. The adjudication committee excluded 164 (37%) out of the 446 patients screened for CHEST-1 because they were considered “operable” in terms of disease distribution. This observation illustrates the importance of thorough evaluation of CTEPH patients and the use of pharmacological therapy only in patients not eligible for PEA.

Riociguat responder analysis

For patients with PAH, several parameters that indicate response to treatment have been identified, and threshold values for these parameters which correlate with survival have been defined. These “responder thresholds” form part of treatment goals for patients with PAH [1, 50, 51]. These values include 6MWD ≥380 m, WHO FC I/II, cardiac index ≥2.5 L·min⁻¹·m⁻², mixed venous oxygen saturation (S_vO2) ≥65% and N-terminal pro-brain natriuretic peptide (NT-proBNP) <1800 pg·mL⁻¹.

An exploratory analysis from CHEST-1 examined the proportion of patients who achieved the responder thresholds described earlier at baseline and at the end of the study [52]. In addition, the analysis examined the proportions of patients achieving PVR <500 dyn·s·cm⁻⁵ because values above this level are strongly correlated with increased mortality in patients with CTEPH [53]. Riociguat increased the proportion of patients with 6MWD ≥380 m, WHO FC I/II and PVR <500 dyn·s·cm⁻⁵ from 37%, 34% and 25%, respectively, at baseline to 58%, 57% and 50%, respectively, at week 16, whereas there was little change in placebo-treated patients. Similar changes were observed for thresholds for cardiac index, S_vO2, NT-proBNP level and right atrial pressure. Additionally, riociguat was associated with an increase in the proportion of patients meeting the combination of response criteria for 6MWD, WHO FC, cardiac index, S_vO2 and NT-proBNP level. These results suggest that riociguat increased the proportion of patients achieving criteria defining a positive response to therapy.

Riociguat long-term data

Patients who completed CHEST-1 were eligible to enter a long-term open-label extension study (CHEST-2) [40, 54]. Overall, 237 patients (98%) entered CHEST-2 and received riociguat (patients who received placebo in CHEST-1 were initiated at a dose of 1 mg three times daily and adjusted to an individual dose as in the CHEST-1 study). The primary objective of CHEST-2 was to evaluate the long-term safety of riociguat (mean treatment duration was 83 weeks, median 75 weeks). Of 157 patients treated for 1 year in CHEST-2, 12 (8%) were receiving additional PAH therapies (ERAs n=8, prostanoids n=4). The most common adverse events in the extension study were nasopharyngitis (23%), dizziness (19%), peripheral oedema (18%) and diarrhoea (14%). For adverse events of special interest, hypotension was reported in 6% of patients and syncope in 7%. Overall, exposure-adjusted rates of adverse events were lower in CHEST-2 than in CHEST-1. The most common drug-related serious adverse events were syncope (2%) and hypotension (1%), which resolved in all cases. Exploratory efficacy analyses showed that the increase in 6MWD seen in patients originally randomised to riociguat was maintained, and 6MWD also increased in patients originally randomised to placebo, to an extent comparable to the former group. The estimated rate of clinical worsening-free survival at 1 year was 88% (95% CI 83–92%), with an estimated overall survival rate at 1 year of 97% (95% CI 93–98%). 2-year data from CHEST-2 have recently been published. The mean±sd change from baseline in 6MWD was +52±66 m at 1 year (n=209) and +50±68 m at 2 years (n=162) [54]. The estimated survival rate was 93% (95% CI 89–96%) at 2 years (figure 2) [54], with no new safety signals detected with the additional duration of treatment.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2

Kaplan–Meier analysis of survival over 2 years of treatment with riociguat in the CHEST-2 study [54]. Overall survival was 97% (95% CI 93–98%) at 1 year and 93% (95% CI 89–96%) at 2 years. If the worst-case scenario was assumed, in which patients who dropped out were assumed to have died, survival was 92% (95% CI 88–95%) at 1 year and 87% (95% CI 82–91%) at 2 years. Reproduced and modified from [54] with permission.

Bridging therapy

One potential use of medical therapy in CTEPH is as a bridge to PEA [14, 55]. Registry data show that a substantial proportion of operable patients receive such therapy (table 3), but no randomised trials have been performed in this patient population. The first report of bridging therapy was from 12 patients with severe CTEPH (PVR >1200 dyn·s·cm⁻⁵) who received continuous i.v. prostacyclin before undergoing PEA [35]. Treatment was associated with a 28% decrease in mean PVR (from 1510 to 1088 dyn·s·cm⁻⁵; p<0.001) and a marked decrease in plasma BNP level (from 547 to 188 pg·mL⁻¹; p<0.01). Operative mortality was 8%. In a retrospective analysis of nine patients who received continuous i.v. epoprostenol before PEA, six patients experienced either clinical stability or improvement, with a mean reduction in PVR of 28%, and three experienced clinical deterioration [56]. In contrast, in a study of inhaled iloprost before PEA, there were no significant changes in mPAP, cardiac index or PVR; the authors concluded that bridging therapy may have detrimental effects on systemic haemodynamics [57].

In a single-blind randomised study, 25 patients with CTEPH were randomised to standard of care with or without bosentan for 16 weeks before surgery [58]. After 16 weeks, there were statistically significant benefits in favour of bosentan in total pulmonary resistance (–299 dyn·s·cm⁻⁵; p=0.004), 6MWD (+33 m; p=0.014) and mPAP (−11 mmHg; p=0.005). Postoperative mortality was four (31%) out of 13 patients who received bosentan versus three (25%) out of 12 patients who did not.

A retrospective analysis compared 111 patients with CTEPH who received targeted therapy, including bosentan, sildenafil, epoprostenol and combination therapy, before undergoing PEA with 244 patients who did not receive targeted therapy [59]. Patients receiving medical therapy had little or no improvement in haemodynamic or post surgical outcomes, and referral of operable patients for PEA was delayed compared with patients who did not receive medical therapy. In the international CTEPH registry there were no differences in PEA complications between operated patients who received bridging therapy compared with those who did not, but multivariable analysis indicated that bridging therapy was associated with increased mortality (hazard ratio 2.62, 95% CI 1.30–5.28; p=0.0072) [10].

At present there is no evidence to recommend bridging therapy before PEA, and all eligible patients should proceed to surgery without delay.

Operable patients not undergoing surgery

Medical treatment may be appropriate for patients who are technically operable in terms of disease distribution, but for whom surgery is refused or is considered to be too high risk because of comorbid conditions. Their risk–benefit ratio may be unacceptable because of an excessive surgical risk or because the anticipated benefit is small, as in patients with mild symptoms and good functional capacity: the effects of medical therapy may differ greatly between these populations. There are no robust data in this population, and treatment is based on anecdotal evidence and consensus. The 2015 European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines state that “Medical treatment of CTEPH with targeted therapy may be justified […] in the presence of an unacceptable surgical risk–benefit ratio” [1]. Other authors affirm that all operable patients should proceed to PEA without delay [55].

Patients receiving BPA

Studies, mainly undertaken in Japan, have described use of targeted medical therapy for “bridging” prior to BPA or in patients whose haemodynamic parameters did not normalise after BPA [60–68]. The implications for the use of targeted PAH therapies are difficult to assess in patients undergoing a series of BPA procedures. These studies were uncontrolled, and in European practice, patients in the Japanese cohorts would have been referred for PEA assessment first rather than medical therapy or BPA.