Aggressive combination therapy for treatment of systemic sclerosis-associated pulmonary hypertension

Abstract

Pulmonary arterial hypertension is an important complication of systemic sclerosis with high mortality but should be regarded as a treatable manifestation of the disease. Management draws on experience from other forms of pulmonary arterial hypertension and benefits from an increasing number of licenced therapies. Outcome is variable but recent clinical trials suggest that combination therapies used early in the disease may be associated with better outcomes. This is important because previous clinical trials using short-term gain in exercise capacity did not show significant benefit compared to that observed for idiopathic or heritable forms of pulmonary arterial hypertension. Thus, it is important to identify cases as early as possible and to manage cases that are in a high-risk group using early combination therapy. This review summarises the most recent analyses of clinical trial data, with a focus on those patients with SSc-associated pulmonary arterial hypertension and provides the evidence base that supports current treatment recommendations for aggressive pulmonary arterial hypertension occurring in systemic sclerosis, including the early use of combination pulmonary arterial hypertension–specific drugs in appropriate cases.

Introduction

Long-term event-driven clinical trials for patients with World Health Organization (WHO) Group I pulmonary arterial hypertension (PAH) have demonstrated benefits of targeted vasodilator therapies, including a substantial proportion of cases in which different PAH-specific drugs were used in combination. They have also re-affirmed the treatment effects that occur for key subtypes of PAH, especially that associated with connective tissue disease (PAH-CTD) and systemic sclerosis (PAH-SSc). The recent AMBITION trial has demonstrated the efficacy of upfront combination therapy in delaying clinical failure among patients with PAH (1). An identical efficacy to that observed in idiopathic PAH (iPAH) with confidence intervals below unity is evident in the subgroup of patients with SSc (2). In isolation, as a post hoc analysis of a very carefully selected population, one might not get too excited about this observation; however, similar findings in the GRIPHON trial (3) and the SERAPHIN trial (4) provide grounds for asserting that there is a genuine effect being observed here. Furthermore, there is published data demonstrating the mechanism of benefit (5), with a much greater haemodynamic response and improved right ventricular function using upfront combination therapy in the SSc population. The consequence is quite profound, given the misgivings that have been expressed in respect of the value of treating patients with CTD including those with scleroderma (6). To explore the magnitude of the shift in perspectives required, we shall set out the background reasons for concern in respect of PAH-SSc and explore the explanations proposed, dissect the recent ‘outcome’ trials and finally consider not just the aggression of therapy in terms of intensity but also timing.

PAH-SSc ‘a challenging subgroup’

While patients with SSc have a similar histological basis for PAH (7) and the initial open-label randomised trial suggested a high level of effort response in a short-term trial (8), subsequent short-term trials were disappointing (9), as the treatment effect in terms of the primary endpoint (6-min walk distance (6MWD)) appeared consistently inferior to that of the iPAH population. The only meta-analysis using individual patient data confirmed attenuated benefit in the PAH-CTD population (6) in terms of the primary endpoint in these trials (6MWD) and time to clinical worsening, while being associated with increased risk of adverse events (but not serious adverse events (SAEs)) in those exposed to active therapy (10). The authors suggested that exclusion of the PAH-CTD population might reduce the necessary study size and associated costs. Fig. 1 shows the impact on 6MWD and clinical worsening from this meta-analysis; of note, the impact on this endpoint is not significantly different between the iPAH and SSc subpopulations. The efficacy in the SSc population in terms of 6MWD persists after correction for comorbidities and background medications, which is driven by deterioration in the placebo group. However, the ‘apparent’ benefit in terms of clinical worsening becomes non-significant unless the SERAPHIN trial is censored at 18 weeks; as shown in the following, the 3 mg dosing was ineffective in the CTD population, and when all follow-up is included, this trial provides 72% of effect observed, thus potentially biasing the analysis.

Fig. 1.

Meta-analysis of pulmonary hypertension trials including PAH-CTD and PAH-SSc. Comparison of short-term gains in 6-min walk distance suggest comparable absolute treatment effect for PAH-SSc, but this does not reach statistical significance, most likely due to greater variability and small number of cases. For clinical worsening, the reduction in hazard ratio is comparable for PAH-SSc and iPAH.

Source: Modified from Rhee et al. (10)

CW = clinical worsening; Hosp = hospitalisation; Rx = treatment; Tx = transplantation; RVF = right ventricular failure.

Proposed explanations for the limited improvement on therapy include the older age of SSc (11) PAH patients, the prevalence of left heart (12), lung disease (13) and the higher prevalence of pulmonary veno-occlusive disease (PVOD) (14). The negative results in older trials were counter-intuitive given the excellent clinical response seen in registries with substantial improvements in survival (15, 16). One very recent publication has suggested no progress in SSc survival in the past decade (17); intriguingly, this article provides further support for the need for aggressive treatment of PAH-SSc patients, as maximal administered therapy (despite a 50% 3 year mortality) was monotherapy in 40%, oral combination therapy in only 37% and prostanoid usage of any type was almost 50% less frequent in the PAH-SSc patients. One thing is clear from the confusing data published on the PAH-SSc population that a more rigorous approach to diagnosis is essential in this population. To provide a secure diagnosis of ‘clean’ (WHO Group I) PAH-SSc, a rigorous assessment to exclude other forms of PH is needed. We propose such an approach in Fig. 2, recognising that most trials and registries have not mandated such rigour and therefore outcomes and analyses should take that into account.

Fig. 2.

A comprehensive diagnostic algorithm for PAH-SSc. All cases of suspected PH should be evaluated and associated lung fibrosis or cardiac causes of PH excluded. Features of other causes such as PVOD and CTEPH also require exclusion. While the principle for the evaluation of PAH-SSc is identical to that of iPAH, the likelihood of co-existent lung or heart involvement is much greater. Some abnormality of the heart or lung is present in the majority of patients. The issue is not generally exclusion of lung or heart disease, but determining whether the abnormality observed is sufficient to explain the haemodynamic abnormality found. For lung disease, <20% volume of fibrosis or less than 5% emphysema is unlikely to cause precapillary PH in SSc patients. For heart disease, even minor abnormalities should cause suspicion that demonstration of a wedge of <15 mmHg is not sufficient to make a diagnosis of precapillary PH, thus LVEDP and CMR should be considered. Finally, if more than a single CT feature of PVOD (septal lines, lymphadenopathy or centrilobular nodules) is identified, there is a significant risk of pulmonary oedema with advanced therapies.

LVEF = left ventricular ejection fraction; LA = left atrium; MAPSE = mitral annular plane systolic elevation; LVEDP = left ventricular end diastolic pressure; CMR = cardiac magnetic resonance.

It is notable that the increased rate of adverse events in CTD patients in the meta-analysis was driven by gastrointestinal (GI) bleeding and infections (10). The former may reflect the high use of anticoagulants in PAH-CTD patients (SSc being associated with angiodysplasia and gastric antral vascular ectasia) before recent study (16) showed that this did not improve outcomes in these patients and may well have been contributed by drug–drug interactions with warfarin. Increased infection rate may reflect frequent use of corticosteroids and other immunosuppression in non-SSc CTDs such as systemic lupus erythematosus (SLE) and so may not be as relevant to the PAH-SSc subgroup.

Is co-morbid cardiopulmonary disease the reason for poor outcomes in trials?

Lung fibrosis–associated pulmonary hypertension (PH-SSc) has been reported as being associated with a particularly dismal prognosis (18, 19). Most trials do not require high-resolution computed tomography (HRCT) exclusion of lung disease, relying instead on spirometric assessment alone. This allows patients with combined fibrosis and emphysema and a proportion of patients with significant fibrosis to be recruited. Fortunately, most expert centres use accepted criteria or the study by Goh et al. (2008) to exclude significant fibrosis when diagnosing SSc-PAH, thus identifying significant emphysema and fibrosis before attributing the PH to vasculopathy (20). However, one must accept that with current inclusion and exclusion criteria, a proportion of patients with significant lung disease could contaminate the results for all populations recruited. From the UK national registry, we see that this could affect around 30% of patients diagnosed as having iPAH (21) – a similar level to the prevalence of lung disease to that seen in SSc. One may conclude that parenchymal lung disease remains a confounder – but not especially so in the SSc subpopulation.

Pulmonary hypertension associated with left heart disease has been reported as having a worse prognosis than PAH-SSc (22), thus underestimating the prevalence of post-capillary PH in studies could theoretically lead to a much worse outcome in SSc PH subgroups. Fox et al. (12) found that two-third of patients with PH-SSc had post-capillary PH, most being identifiable by the presence of an elevated wedge pressure, but one in three being identified only on assessment of the left ventricular end diastolic pressure (14%) or unmasked on fluid challenge (17%). The reliability of fluid challenge in this setting remains to be established (23), and the issues in respect of the haemodynamic definition of post-capillary PH in SSc are complex (24). In the AMBITION study, the modification of the recruitment criteria would have excluded most such patients; however, recruitment in the GRIPHON and SERAPHIN studies used standard criteria. If this explanation were correct, one would expect a blunted therapeutic response in the studies that failed to rigorously exclude ‘occult’ left heart disease.

It is worth observing that in the study by Bourji et al. (22), the patients with left heart disease had higher mean pulmonary artery pressures than the PAH-SSc patients, and despite elevated wedge pressures, 75% were treated with advanced therapies, over half with ERAs. Thus, the poorer prognosis in this subgroup may have had many contributors.

Gunther et al. (14) found that over 60% of SSc-PAH patients had two or more CT features of PVOD and that this was associated with a 50% incidence of pulmonary oedema on therapy and early death. A high prevalence of PVOD as reported by Gunther et al. could readily explain a poor response to therapy and higher SAE rate on therapy. However, the population studied was highly selected (failed therapy referred for transplantation), and no other reports of a high prevalence of pulmonary oedema followed. In an analysis of 66 patients referred to our service, we observed only 11% with two or more features of PVOD and only 6% developed pulmonary oedema on therapy (25) – insufficient to explain the apparent poor response to therapy reported.

PAH-CTD in recent outcome trials

The four recent large-scale long-term outcome trials (1, 3, 4, 26) have reported almost identical benefit in the iPAH and PAH-CTD subgroups and unlike previous studies have tended to include a sufficiently large CTD population to allow subpopulation analysis. Tab. I shows the number of patients in iPAH groups, CTD and SSc in each of these trials and the proportion in the active therapy group on combination therapy. Fig. 3 shows a forest plot based on data abstracted from these studies and demonstrates the remarkable consistency between the iPAH (and various associated conditions according to study groupings) and CTD populations for the primary endpoint in these trials. In the COMPASS-2 trial, bosentan and sildenafil were ineffective in both populations, possibly reflecting the drug–drug interaction between bosentan and sildenafil, though a high drop-out rate and significant proportion with possible left heart disease may also explain the outcome. For SERAPHIN 10 mg, AMBITION and GRIPHON, almost identical outcomes are evident. In the SERAPHIN study, 64% of patients were on combination therapy; in GRIPHON, 80%; while in the study treatment arm of the AMBITION trial, 100% were on combination therapy. Numerically superior efficacy in the CTD population is associated with the most aggressive of these regimes. In the 3mg arm of the SERPHIN study, efficacy is evident in the iPAH arm but not the CTD arm, suggesting that less intense therapy is effective in the iPAH arm but not the CTD arm, although the smaller number of CTD cases in the SERAPHIN study may have been a factor and limited the power to confirm benefit. Moreover, the effect size in the 10mg arm is numerically greater than in the 3mg arm and consistent with the effect seen in the IPAH arm.

Table I.

Patient disposition and CTD subgroups in combination therapy trials for PAH

Study	iPAH group	CTD PAH	SSc-PAH	SLE PAH	Combination Rx
COMPASS-2 (18)	226	88	38	–	100% versus 0%
SERAPHIN (4)3 and 10 mg	449	225	116	66	64%
AMBITION (1)	279	187	118	17	100% versus 0%
GRIPHON (3)	712	334	170	82	80%

iPAH = idiopathic PAH; CTD = connective tissue disease; SSc = systemic sclerosis; PAH = pulmonary hypertension; SLE = systemic lupus erythematosus.

Fig. 3.

Comparison of iPAH groupings and connective tissue disease (CTD) populations. Forest plot summarising overall treatment effect on hazard ratio for event-driven long-term trials including combination PAH therapy in SSc. For the licenced dose of macitentan and in selexipag and AMBITION, study effect was similar for the PAH-CTD and iPAH subjects. The trials had different primary endpoints, slightly different definitions of the IPAH groupings, different proportions of the various CTD populations and in the case of the AMBITION trial an incident population, so the trials should not be compared to each other; however, presentation in a single forest plot allows visual appreciation of the relative impact of different aggressiveness of the treatment regime with relative outcomes between the iPAH and CTD groups. HR – hazard ratios for morbidity or mortality/clinical failure outcomes relative to placebo groups. For each study, the iPAH group is the upper of the two hazard ratios shown. Numbers of patients followed by numbers with events is shown for the active comparator group and placebo for each study.

The SSc population in the AMBITION trial

The AMBITION trial (1, 2) was almost unique in the field of pulmonary hypertension, in that patients were enrolled ‘upfront’ prior to any therapy and randomised to ambrisentan and tadalafil initial combination therapy (ambrisentan 10 mg plus tadalafil 40 mg) or to ambrisentan 10 mg (plus placebo) or tadalafil 40 mg (plus placebo) monotherapy.

Patients were aged 18–75 years, had baseline WHO functional class II or III symptoms and the average time from diagnosis to enrolment was only 22 days. Initial inclusion criteria required only standard haemodynamic criteria for precapillary PH, a total lung capacity ≥60% of predicted normal and forced expiratory volume in 1 s ≥55% of predicted normal. After 6 months, a blinded review of the participants’ baseline demographic data revealed a high prevalence of risk factors for left ventricular diastolic dysfunction and a high drop-out rate among such patients. Therefore, the eligibility criteria were amended to include more rigorous haemodynamic requirements (pulmonary vascular resistance (PVR) ≥ 300 dynes-sec cm–5 if left ventricular end diastolic pressure (LVEDP) or pulmonary capillary wedge pressure (PCW) < 12 mmHg, PCW ≥ 500 if PCW/LVEDP 12–15 mmHg) and to exclude patients with ≥3 risk factors for left ventricular diastolic dysfunction: body mass index ≥30 kg/m², history of essential hypertension, diabetes mellitus, and historical evidence of significant coronary artery disease.

Given these modified criteria, patients of advanced age and those likely to have cardiac co-morbidity are excluded from the main analysis. Potentially, this selects for a more favourable treatment cohort, where a more ‘vascular’ profile is expected. However, when comparing to registry populations as shown in Tab. II, the resultant study population is quite typical – slightly younger and less symptomatic, but very similar in terms of pulmonary and haemodynamic parameters.

Table II.

Comparison of clinical and demographic features of SSc subgroups within clinical trial and registry cohorts trial and registry cohorts

Parameter	AMBITION (22)	REVEAL (30)	UK national registry (18)	RFH registry (15)	COMPERA (16)
Age (years)	61.5	61.8 ± 11	63.9 ± 10.5	63 ± 11	67 (59.5–74)
Sex (female, %)	84	88.5	82	85	84
FC I or II/III/IV (%)	23/77/0	25/60/15	16/68/16	21/79 (III + IV)	14/72/14
TLC (% predicted)	90	82.7 ± 18.4	NR	NR (FVC 94%)	NR
FEV1 (% predicted)	86	73.4 ± 16.4%	NR	NR	NR
mPAP (mmHg)	44	44.6 ± 11.8	42 ± 17	40 ± 12	41 ± 11
PVR (dynes-sec cm–5)	670	9.6 ± 5.3	715 ± 597	607 ± 417	697 ± 385
PCW (mmHg)	8.5	9.1 ± 3.5	NR	10 ± 3	9.3 ± 3.3

TLC = total lung capacity; FEV1 = forced expiratory volume over the first second; mPAP = mean pulmonary artery pressure; PVR = pulmonary vascular resistance; PCW = pulmonary capillary wedge pressure; FC = functional class; NR = not reported.

Results of the AMBITION trial

The results of the AMBITION trial are quite striking. They suggest that with an aggressive upfront approach, equivalent outcomes are achievable in PAH-SSc and iPAH/heritable PAH (HPAH) as shown in Tab. III. The primary endpoint was time to clinical failure – a composite endpoint was defined as the first occurrence of death, hospitalisation for worsening PAH (any hospitalisation for worsening PAH, lung or heart/lung transplant, atrial septostomy or initiation of parenteral prostanoid therapy), disease progression (decrease of >15% from baseline 6MWD combined with WHO functional class III or IV symptoms at two consecutive visits separated by ≥14 days) or unsatisfactory long-term clinical response (any decrease from baseline 6MWD at two consecutive post-baseline clinic visits separated by ≥14 days and WHO functional class III symptoms assessed at two clinic visits separated by ≥6 months).

Table III.

Summary of outcome data from the AMBITION trial

Parameter	Combination iPAH/HPAH	Monotherapy iPAH/HPAH	HR	Combination PAH-SSc	Monotherapy PAH-SSc	HR
Primary EP reached	19%	32%	0.51 (0.31–0.83)	21%	40%	0.44 (0.22–0.89)
Hospitalisation for PAH	8%	18%	0.41 (0.2–0.82)	8%	19%	0.36 (0.13–1.04)
6MWD at 24 weeks (m)	+52.5	+26.6		+40.9 m	+12.2 m
NTproBNP	−71.2%	−50%	−42.5 (–54.6;–27)	−62.8%	−38.4%	−36.9 (–59.3; –10.2)
Satisfactory clinical outcome	40%	28%	1.7 (1.04–2.9)	31%	29%	1.1 (0.47–2.6)
15% fall in 6MWD	24%	32%		31%	44%
SAE rate	33%	38%		44%	49%
AE resulting in discontinuation	11%	10%		14%	13%

iPAH = idiopathic PAH; HPAH = heritable PAH; SSc = systemic sclerosis; PAH = pulmonary hypertension; NTproBNP = N-terminal prohormone of brain natriuretic peptide; 6MWD = 6-min walk distance; AE = adverse event.

The primary endpoint occurred in 32% of iPAH/HPAH patients and 40% of PAH-SSc patients receiving monotherapy. By contrast, in the initial combination therapy group, the primary endpoint occurred in 19% of the iPAH/HPAH patients (a 49% reduction) and 21% of PAH-SSc patients (a 56% reduction). Thus, with an upfront aggressive strategy, there is much less difference in outcomes between iPAH/HPAH and PAH-SSc than in previous trials.

In those receiving initial combination therapy (Tab. III), the likelihood of hospitalisation for PAH, improvement in 6-min walking distance and reduction in N-terminal prohormone of brain natriuretic peptide (NTproBNP) were very similar to the effect of combination therapy in the iPAH/HPAH groups. The impact in terms of improvement in 6MWD is particularly striking. In this study, it was not deterioration in 6MWD in the placebo group (who received monotherapy) but rather an impressive positive response to combination therapy that delivered the treatment effect. By contrast, there was a trend towards fewer achieving a satisfactory clinical outcome. Although discontinuation rates resulting from adverse events was higher in the PAH-SSc population, neither AE-related discontinuation rates nor SAE rates were increased by combination therapy when compared to monotherapy (Tab. III). Comparing the secondary endpoint outcomes in the monotherapy arms between the SSc and iPAH/HPAH populations, it is notable that monotherapy appears less effective in the SSc population in terms of 6MWD (+12 vs +26 m), reduction in NTproBNP (−38% vs −42%) and 15% fall in 6MWD (44% vs 32%); however, the magnitude of improvement in each of these in the combination group relative to the monotherapy arms is just as great. Thus, there is a synergistic effect of combination therapy in the SSc subgroup. The suggestion, therefore, from this analysis is that in the PAH-SSc population, a similar medium-term outcome to iPAH/HPAH can be achieved with a sufficiently aggressive therapeutic approach. There is, however, a residual note of caution that the degree of improvement is more likely to be insufficient to attain a satisfactory clinical outcome in nearly 70% of patients.

Mechanistic basis for efficacy of upfront combination therapy in PAH-SSc

Hassoun et al. (5) evaluated 24 patients with PAH-SSc who received upfront ambrisentan and Tadalafil following the AMBITION protocol. The co-primary endpoints were change in PVR and reduction in right ventricular (RV) mass at 36 weeks, both were significant. Upfront combination therapy reduced mean pulmonary arterial pressure (mPAP) from 42 ± 12 to 30 ± 7 mmHg and increased cardiac index from 2.6 ± 0.7 to 3.3 ± 1.2 L min/m², a significantly greater haemodynamic improvement that is recorded in previous monotherapy trials in this population.

As observed in the AMBITION trial, 6MWD increased significantly from 343 ± 131 to 395 ± 99 m. Furthermore, a clear improvement in right ventricular function was observed; right ventricular ejection fraction improved from 46 ± 10% to 57% ± 9%, while right ventricular end systolic volume fell from 82.1 (65.6–97.7) to 55.8 (49.4–79.2) and left ventricular end diastolic volume increased from 114.0 (84.8–130.2) to 135.3 (112.4–160.1). All these changes reach significance and indicate substantial off-loading with upfront combination therapy, with improved RV function and LV filling. This was an open-label trial with per-protocol analysis; that said, of the 17 patients excluded before treatment, 12 met exclusion criteria, and only 1 patient was subsequently excluded as they elected to have palliation.

The GRIPHON trial

The GRIPHON (3) is not directly comparable to the AMBITION trial – both in terms of trial construct and the details reported, thus gives different but complimentary information. GRIPHON was a larger trial that included 170 PAH-SSc patients (15% of population) (28), while the AMBITION trial included 118 PAH-SSc patients (24% of population). The mean age of the scleroderma population was marginally older (60 years old), while the overall population was younger in the GRIPHON trial; 48 versus 54.3 years in the AMBITION trial. Time from diagnosis to trial inclusion in the AMBITION trial was short (median < 1 month) and equivalent in both the PAH/HPAH and SSc-PAH populations, and GRIPHON was a prevalent population study. In the GRIPHON trial, time from PAH diagnosis to enrolment was 1.6 years in the SSc population (26), but 2.4 years in the overall population (3) (> 60% of the whole population had PAH/HPAH/drug or toxin-associated PAH). Background therapy of PAH-SSc patients in the GRIPHON trial was almost identical to the whole population; 78% SSc versus 80% whole population were on background therapy, and 36% PAH-SSc patients were on background combination therapy, compared to 33% of the whole population.

When comparing the studies, we are further limited by the lack of individual reporting of the details of the PAH/HPAH population in the GRIPHON trial, and the PAH/HPAH population may have reached a more stable phase of the disease, given the likely longer time from diagnosis to enrolment. The primary endpoint was different though similar, comprising disease progression or worsening of PAH that resulted in hospitalisation, initiation of parenteral prostanoid therapy or long-term oxygen therapy, need for lung transplantation or balloon atrial septostomy or death from any cause. Disease progression was defined as a ≥15% decrease in 6MWD from baseline, confirmed by a second test on a different day and worsening in WHO functional class (for patients in functional class II/III at baseline) or need for additional PAH therapy (for patients in functional class III/IV at baseline). The most important difference here is the absence of ‘unsatisfactory long-term clinical outcome’ as an endpoint.

Overall, this was a very encouraging study that reached its key clinical trial endpoints. In those randomised to placebo among the iPAH/HPAH/Drug group, the primary endpoint occurred in 43%, compared to 49% in the PAH-SSc group (28). In the active treatment group, the primary endpoint occurred in 28% of the iPAH/HPAH/Drug group compared to 32.5% with PAH-SSc. Again, suggesting that with aggressive management, including 80% combination and 36% triple-therapy PAH-SSc patients benefits at least as much as iPAH/HPAH patients and event rates approach those seen in the iPAH/HPAH group.

In terms of secondary endpoints, we can only compare outcomes among PAH-SSc with the outcome in the whole population. With this limitation, as shown in Tab. IV, there is remarkable consistency between the effect of selexipag in all patients and those with SSc. Death was an uncommon event as a primary endpoint; however, numbers of events were not insignificant up to the end of study (i.e. including follow-up of those who had stopped treatment after meeting a primary endpoint). Here, a trend is noted – a possible reduction in mortality in the post-selexipag group for the whole population, with no impact on deaths in the SSc cohort. It is, however, impossible to attribute a causative relationship in this setting since the management decisions taken after cessation of trial treatment are unknown.

Table IV.

Summary of outcome data from the GRIPHON clinical trial

	GRIPHON trial		PAH-SSc subgroup
	Placebo (N = 582)	Selexipag (N = 574)	Placebo (N = 93)	Selexipag (N = 77)
Primary composite endpoint of morbidity/mortality up to the end of treatment	Number of patients (%)		Number of patients (%)
All events	242 (41.6)	155 (27)	46 (49.5)	25 (32.5)
Hospitalisation for worsening of PAH	123 (21.1)	86 (15)	22 (23.7)	10 (13.0)
Disease progression	100 (17.2)	38 (6.6)	18 (19.4)	8 (10.4)
Death from any cause	18 (3.1)	28 (4.9)	1 (1.1)	4 (5.2)
Initiation of parenteral prostanoid therapy or long-term O2 therapy for worsening PAH	13 (2.2)	10 (1.7)	5 (5.4)	2 (2.6)
Need for lung transplantation or BAS for worsening of PAH	2(0.3)	1 (0.2)	0	1 (1.3)
Secondary endpoint of all-cause death up to the end of the study	Number of patients (%)	Number of patients (%)
Death from any cause	137 (23.5)	102 (17.8)	22 (23.7)	17 (22.1)

SSc = systemic sclerosis; PAH = pulmonary hypertension; BAS = balloon angioplasty septostomy.

Adverse events were common in the GRIPHON trial with over 95% reporting some adverse events. SAEs were numerically more common in the SSc subpopulation, but the same trend towards reduced events in the active treatment subgroup was seen. Discontinuation due to adverse events was also numerically more common in the SSc cohort (7.1% iPAH vs 13.2% PAH-SSc among placebo patients and 14.3% iPAH vs 19.5% PAH-SSc in the active therapy arm), again with no trend towards a relative excess of SAE due to therapy when compared to the whole population. Thus, in a population with 80% on combination therapy and 36% on triple therapy, there is no evidence that escalation of treatment is disproportionately disadvantageous in the SSc cohort (Tab. IV).

Concluding remarks: current approaches for PAH-SSc

SSc-PAH patients remain complex and we do not have all the answers yet. What is clear is that in those with true PAH-SSc, an aggressive approach with early combination therapy is the required standard of care. This is in line with current evidence-based treatment recommendations (29) and represents a meaningful advance in treatment for SSc in general that can be expected to improve disease outcome and survival. Nevertheless, a recent study suggests that median survival for incident PAH-SSc remains poor at 4 years, and it is hoped that with more aggressive evidence-based management, outlines in this review, this can be further improved (27). Less clear is how we estimate the relative contributions of cardiac, pulmonary and musculoskeletal co-morbidity. Furthermore, in managing such patients, we must always be cognisant of the role of the auto-immune system, which drives the vasculopathy and can cause deterioration or crises entirely independent of the cardiopulmonary axis.