AMBITION

In Review

Galiè N, et al. "Initial Use of Ambrisentan plus Tadalafil in Pulmonary Arterial Hypertension". The New England Journal of Medicine. 2015. 373(9):834-44.
PubMed • Full text • PDF

Clinical Question

In patients with World Health Organization (WHO) functional class II or III pulmonary arterial hypertension, is it more effective to start treatment with both ambrisentan and tadalafil simultaneously, or to start treatment with only a single agent?

Bottom Line

Starting treatment with the combination of ambrisentan and tadalafil resulted in fewer clinical failure events than when either ambrisentan or tadalafil was started as a single agent.

Major Points

Pulmonary arterial hypertension (PAH, also described as WHO group I pulmonary hypertension) is a disease with a complex pathophysiology primarily mediated by pathologic changes in the pulmonary vasculature that cause increased pulmonary vascular resistance (PVR). Increases in PVR cause increasing pressures in the right ventricle, which is particularly sensitive to afterload increases; this ultimately leads to progressive cardiopulmonary failure. Multiple vasoactive agents have been studied for use in reversing or preventing progression of these vascular changes and thereby decreasing or maintaining PVR. Ambrisentan, an endothelin receptor antagonist (ERA), was previously shown to improve exercise capacity in PAH patients in the ARIES-1 and ARIES-2 trials, with continued benefits shown over two years of treatment. Tadalafil, a phosphodiesterase type 5 inhibitor (PDE5i), had also been shown to increase exercise capacity in the PHIRST and PHIRST-2 trials, with benefits sustained after 68 total weeks of treatment. Although these agents (and others targeting different vasoactive mechanisms) showed benefits by themselves, it was unclear whether combinations of these agents would produce any additional benefits. Most studies at that time had examined sequential add-on strategies, in which additional agents were added only after progression of disease or lack of significant response to an initial agent; these studies had mixed results. The AMBITION trial was designed to determine if there was a benefit to simultaneously starting two agents with differing mechanisms of action in treatment-naïve patients.

In AMBITION, ambrisentan and tadalafil were chosen to use in combination due to their differing mechanisms of action and lack of drug interaction. Patients in WHO functional classes II or III (slight or marked limitations of physical activity) who were treatment-naïve (or had previously received <14 days of PAH therapy) were randomized into groups receiving ambrisentan monotherapy, tadalafil monotherapy, or the combination of ambrisentan and tadalafil. Efficacy was judged using a composite primary end-point containing multiple "clinical failure events," defined as death from any cause, hospitalization for worsening PAH, disease progression, or "unsatisfactory long-term clinical response" (as judged by decreasing exercise tolerance and WHO functional class III symptoms after 6 months in the trial). The trial showed a lower overall rate of clinical failure events in the combination-therapy group, with 18% of combination-therapy patients experiencing an event compared to 31% of patients in the pooled-monotherapy group (HR of 0.50; 95% CI of 0.35-0.72; P<0.001). The reduction in clinical failure events was mainly driven by a lower rate of hospitalization for worsening pulmonary arterial hypertension in the combination therapy group. Adverse effects were more frequent in the combination-therapy group, but rates of hypotension were similar.

As a result of AMBITION, initial combination-therapy with ambrisentan and tadalafil was included as a recommendation in the European Society of Cardiology and European Respiratory Society guidelines for PAH treatment in WHO functional class II and III patients.

Guidelines

ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension (2015, adapted)

• The combination of ambrisentan and tadalafil is recommend as initial therapy in WHO functional class II or III patients with PAH (class 1, level B)
• The combination of other ERAs and PDE5 inhibitors is recommended as initial therapy in WHO functional class II or III patients with PAH (class 2a, level C)
• Sequential therapy with macitentan added to sildenafil, riociguat added to bosentan, or selexipag added to an ERA and/or PDE5i is recommended in WHO functional class II or II patients with PAH (class 1, level B).
• Sequential therapy with sildenafil added to epoprostenol is recommended in WHO functional class III patients with PAH (class 1, level B).

Design

• Multi-center, double-blind, randomized, placebo-controlled trial
• N=500
  • Combination-therapy with ambrisentan and tadalafil (n=253)
  • Ambrisentan-monotherapy (n=126)
  • Tadalafil-monotherapy (n=121)
• Setting: 120 sites in 14 countries
• Enrollment: October 18, 2010 through July 31, 2014
• Analysis: Modified intention-to-treat (of the initial 610 enrolled participants, five never received a study medication and were excluded from analysis; additionally, the entry criteria for the trial were amended after randomization in an attempt to reduce the likelihood of enrolling participants with PAH due to left ventricular diastolic dysfunction, which reduced the analysis set to 500 participants)
• Co-primary endpoints:
  • Death from any cause
  • Hospitalization for worsening PAH (or for lung/heart transplantation, atrial septostomy, or initiation of parenteral prostanoid therapy)
  • Disease progression (defined as a decrease by >15% from baseline in the 6-minute walk distance in addition to having WHO functional class III or IV symptoms at two consecutive visits separated by 14 days)
  • Unsatisfactory long-term clinical response (defined as any decrease from baseline in the 6-minute walk distance at two consecutive visits separated by 14 days in addition to having WHO functional class II symptoms at two visits separated by ≥6 months; this could only be fulfilled after a participant had been in the trial for ≥6 months)

Population

Inclusion Criteria

• Age 18 to 75 years old
• Weight ≥40 kg
• Previous diagnosis of WHO group 1 pulmonary hypertension (composed of idiopathic PAH, hereditary PAH, or PAH associated with connective tissue disease, drugs or toxins, HIV, or repaired congenital heart defects)
  • Diagnosis was established by ruling out other known causes of patient's symptoms, per current PAH guidelines
  • Hemodynamic criteria consistent with PAH by right heart catheterization:
    • Mean pulmonary artery pressure ≥25 mmHg
    • Pulmonary vascular resistance ≥240 dyne⋅sec/cm⁵
    • Pulmonary capillary wedge pressure (PCWP) or left ventricular end-diastolic pressure (LVEDP) ≤12 mmHg if PVR ≥300 to <500 dyne⋅sec/cm⁵, or PCWP or LVEDP ≤15 mmHg if PVR≥500 dyne⋅sec/cm⁵.
• WHO functional class II or III symptoms
• Preivous pulmonary function tests showing total lung capacity ≥60% of predicted normal and forced expiratory volume in one second ≥55% of predicted normal
• Able to walk a distance ≥125 m and ≤500 m at screening
• Resting arterial saturation of ≥88% at screening, either with or without supplemental oxygen
• No previous treatment with a therapy approved for PAH, or treatment with such a therapy for <14 days and no use of the therapy within 7 days of enrollment

Exclusion Criteria

• Received previous PAH therapy (PDE5i, ERA, or chronic prostanoid use where chronic is defined as >7 days) within 4 weeks of screening
• Previously received ERA or PDE5i treatment and treatment was discontinued for any reason other than liver function abnormalities
• Previously received ambrisentan or tadalafil and was discontinued due to safety or tolerability reasons
• Received intravenous inotropes (e.g., dobutamine or dopamine) within 2 weeks of screening
• Uncontrolled hypertension (≥180/110 mmHg) or severe hypotension (<90/50 mmHg) at screening
• Use of calcium channel blockers or HMG-CoA reductase inhibitors (e.g., statins) on an unstable dose 4 weeks prior to screening (no dose adjustments could have been made <4 weeks prior to screening)

The exclusion criteria were amended to include the following criterion after randomization in order to reduce the likelihood of enrolling participants with PAH due to left ventricular diastolic dysfunction:

• Participants must not have ≥3 of the following heart failure with preserved ejection fraction risk factors: BMI ≥30 kg/m², history of essential hypertension, diabetes mellitus (any type), historical evidence of significant CAD (history of MI, PCI, or ≥50% stenosis in ≥1 vessel on angiography), positive stress test, previous coronary artery bypass graft, or stable angina.

A full list of inclusion/exclusion criteria can be found in the article's supplementary appendix.

Baseline Characteristics

• Mean age: 54.4 years
• Female: 78%
• White: 89%
• Mean pulmonary artery pressure: 48.7 mmHg
• Mean 6-minute walk distance: 352.6 m
• PAH classification:
  • Idiopathic: 53%
  • Connective tissue disease: 37%
  • Drug or toxin exposure: 3%
  • Hereditary: 3%
  • Congenital heart disease: 2%
  • HIV: 2%

Interventions

• Participants were first stratified according to their cause of PAH (idiopathic, hereditary, or nonidiopathic) and WHO functional class (II or III). They were then randomly assigned in a 2:1:1 ratio within each stratification level to receive ambrisentan plus tadalafil (combination-therapy), ambrisentan plus placebo (ambrisentan-monotherapy), or tadalafil plus placebo (tadalafil-monotherapy).
• Ambrisentan was uptitrated from 5 mg to 10 mg after an 8 week period. Tadalafil was uptitrated from 20 mg to 40 mg after a 4 week period.
• If a participant experienced a primary endpoint and survived, the participant could change therapies (switch from combination-therapy to one of the monotherapy options, start prostanoid therapy, or start any other available therapy), but blinding of the initial randomized study-group assignment was maintained.

Outcomes

Primary Outcomes

Experienced clinical failure event

Combination-therapy: 18%

vs. pooled-monotherapy: 31% (HR of 0.50; 95% CI of 0.35-0.72; P<0.001)

vs. ambrisentan-monotherapy: 34% (HR of 0.48; 95% CI of 0.31-0.72; P<0.001)

vs. tadalafil-monotherapy: 28% (HR of 0.53; 95% CI of 0.34-0.83; P<0.001)

Death

Combination-therapy: 4%

vs. pooled-monotherapy: 3%

vs. ambrisentan-monotherapy: 2%

vs. tadalafil-monotherapy: 5%

Hospitalization for worsening PAH

Combination-therapy: 4%

vs. pooled-monotherapy: 12%

vs. ambrisentan-monotherapy: 14%

vs. tadalafil-monotherapy: 10%

Disease progression

Combination-therapy: 4%

vs. pooled-monotherapy: 6%

vs. ambrisentan-monotherapy: 10%

vs. tadalafil-monotherapy: 3%

Unsatisfactory long-term clinical response

Combination-therapy: 7%

vs. pooled-monotherapy: 9%

vs. ambrisentan-monotherapy: 9%

vs. tadalafil-monotherapy: 10%

Secondary Outcomes

Change in geometric mean of NT-proBNP from baseline to week 24

Combination-therapy: -67.2%

vs. pooled-monotherapy: -50.4% (P<0.001)

vs. ambrisentan-monotherapy: -56.2% (P=0.01)

vs. tadalafil-monotherapy: -43.8% (P<0.001)

Patient-reported satisfactory clinical response at week 24

Combination-therapy: 39% "yes"; 61% "no"; 8% "unknown"

vs. pooled-monotherapy: 29% "yes"; 71% "no"; 9% "unknown" (OR 1.56; 95% CI of 1.05-2.32; P=0.03)

vs. ambrisentan-monotherapy: 31% "yes"; 69% "no"; 10% "unknown" (OR 1.42; 95% CI of 0.88-2.31; P=0.15)

vs. tadalafil-monotherapy: 27% "yes"; 73% "no"; 7% "unknown" (OR 1.72; 95% CI of 1.05-2.83; P=0.03)

Median (IQR) change in 6-minute walk distance from baseline to week 24

Combination-therapy: 48.98 m (4.63 to 85.75 m)

vs. pooled-monotherapy: 23.80 m (-12.25 to 64.53 m; P<0.001)

vs. ambrisentan-monotherapy: 27.00 m (-14.00 to 63.25 m; P<0.001)

vs. tadalafil-monotherapy: 22.70 m (-8.25 to 66.00 m; P=0.003)

Change in WHO functional class at week 24

Combination-therapy: 37% improved; 58% no change; 5% deteriorated

vs. pooled-monotherapy: 33% improved; 60% no change; 7% deteriorated (P=0.24)

vs. ambrisentan-monotherapy: 34% improved; 59% no change; 7% deteriorated (P=0.30)

vs. tadalafil-monotherapy: 33% improved; 62% no change; 6% deteriorated (P=0.36)

Adverse Events

Selected adverse event rates are shown below; see the article's supplementary appendix for a full listing.

Peripheral edema

45% in combination-therapy vs 33% in ambrisentan-monotherapy and 28% in tadalafil-monotherapy

Headache

42% in combination-therapy vs 33% in ambrisentan-monotherapy and 35% in tadalafil-monotherapy

Nasal congestion

21% in combination-therapy vs 15% in ambrisentan-monotherapy and 12% in tadalafil-monotherapy

Diarrhea

20% in combination-therapy vs 23% in ambrisentan-monotherapy and 19% in tadalafil-monotherapy

Dizziness

20% in combination-therapy vs 19% in ambrisentan-monotherapy and 12% in tadalafil-monotherapy

Hypotension

8% in combination-therapy vs 7% in ambrisentan-monotherapy and 7% in tadalafil-monotherapy

Syncope

5% in combination-therapy vs 7% in ambrisentan-monotherapy and 7% in tadalafil-monotherapy

Adverse effects leading to treatment discontinuation

12% in combination-therapy vs 11% in ambrisentan-monotherapy and 12% in tadalafil-monotherapy

Criticisms

• The inclusion criteria for the trial were amended after the first 6 months because a review of demographic data from patients enrolled up to that point showed a high prevalence of risk factors for left ventricular diastolic dysfunction (LVDD). This resulted in the removal of 105 participants from the analysis set. Pulmonary hypertension due to LVDD (WHO group II pulmonary hypertension) is a distinct clinical entity than PAH (WHO group I pulmonary hypertension), and treatments intended for use in PAH do not show clear benefit in pulmonary hypertension due to LVDD (and may be associated with harm). The restrictions added in an attempt to reduce the number of participants who actually had pulmonary hypertension due to LVDD were quite restrictive, which may reduce how generalizable the study is to an usual patient population.
  • Subsequent analysis of the 105 removed participants was unable to show a significant difference in effect for combination-therapy, although it was difficult to determine if this was due to a lack of response or because of the smaller sample size.
• The primary endpoint event that was most markedly different in the combination-therapy group was by far the rate of hospitalization for worsening PAH; while this event is associated with poorer prognosis, the rates of death and disease progression were similar in this trial between the combination-therapy and pooled-monotherapy groups. Additionally, retrospective analysis showed no difference in rates of hospitalization for any cause between the two groups.
• Strictly speaking, the results of the study can only be applied to the use of ambrisentan and tadalafil in combination; the substitution of other ERA or PDE5i agents is a reasonable option in certain situations per the ESC/ERS guidelines for PAH treatment but is not directly supported by this study.

Funding

• The study was co-sponsored by Gilead Sciences (manufacturer of the Letaris brand of ambrisentan) and GlaxoSmithKline (distributor of the Volibris brand of ambrisentan outside of the US and distributor of the Adcirca brand of tadalafil in Europe), both of which provided funding and ambrisentan drug supply
• Eli Lilly and Company (manufacturer of the Adcirca brand of tadalafil) provided additional funding and tadalafil drug supply