University of Groningen
Meaningful and feasible composite clinical worsening definitions in paediatric pulmonary
arterial hypertension
TOPP Investigators; Beghetti, Maurice; Brand, Monika; Berger, Rolf M. F.; Humpl, Tilman;
Wheeler, Jeremy G.; Ivy, D. Dunbar; Bonnet, Damien
Published in:
International Journal of Cardiology
DOI:
10.1016/j.ijcard.2019.04.062
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
TOPP Investigators, Beghetti, M., Brand, M., Berger, R. M. F., Humpl, T., Wheeler, J. G., Ivy, D. D., &
Bonnet, D. (2019). Meaningful and feasible composite clinical worsening definitions in paediatric pulmonary
arterial hypertension: An analysis of the TOPP registry. International Journal of Cardiology, 289, 110-115.
https://doi.org/10.1016/j.ijcard.2019.04.062
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Meaningful and feasible composite clinical worsening de
finitions in
paediatric pulmonary arterial hypertension: An analysis of the
TOPP registry
☆
Maurice Beghetti
a,⁎
,1, Monika Brand
b,1, Rolf M.F. Berger
c,1, Tilman Humpl
d, Jeremy G. Wheeler
e,1,
D. Dunbar Ivy
f,1, Damien Bonnet
g,1, on behalf of theTOPP investigators
a
Paediatric Cardiology Unit, and Centre Universitaire Romand de Cardiologie et Chirurgie Cardiaque Pédiatrique, University of Lausanne and Geneva, Switzerland
b
Global Medical Science and Epidemiology, Actelion Pharmaceuticals Ltd, Allschwil, Switzerland
c
Center for Congenital Heart Diseases, Department of Paediatric Cardiology, Beatrix Children's Hospital, University Medical Center Groningen, University of Groningen, Groningen, Netherlands
dDivision of Cardiology, Hospital for Sick Children, University of Toronto, Toronto, Canada e
Quanticate Ltd, London, UK
f
Department of Pediatrics, Children's Hospital Colorado, University of Colorado School of Medicine, Aurora, CO, USA
g
M3C-Necker, Paediatric Cardiology, Reference Centre for Complex Congenital Heart Diseases, Hôpital Necker Enfants Malades, Université Paris Descartes, Paris, France
a b s t r a c t
a r t i c l e i n f o
Article history:
Received 27 September 2018 Received in revised form 25 March 2019 Accepted 19 April 2019
Available online 25 April 2019
Background: Composite clinical worsening (cCW) outcomes might allow measurement of disease progression in paediatric pulmonary arterial hypertension (PAH). This TOPP registry analysis investigated three cCW outcomes and their predictive strength for lung transplantation/death.
Methods: Patients≤17 years with idiopathic/familial PAH or PAH-associated congenital heart disease diagnosed ≤3 months before enrolment were included. cCW outcomes included the following variables at enrolment and/or follow-up: all-cause death, PAH-related hospitalisation, lung transplantation, atrial septostomy (cCW1, 2 and 3), WHO FC deterioration, intravenous/subcutaneous prostanoids initiation, syncope (cCW2,3) and occur-rence/worsening of≥2 PAH symptoms (cCW3). The predictive value of CW (excluding transplantation and death) to transplantation or death was assessed. Predictive values of each cCW for lung transplantation/death were analysed by Cox proportional hazards models.
Results: From 255 patients,first-event rate/100 person-years (95% CI) were cCW1: 23.1(19.3,27.6), cCW2: 43.6 (37.6,50.6), and cCW3: 46.3(40.0,53.7) with PAH-related hospitalisation as the most frequentfirst event in each. The cCW definitions comprised from endpoints (excluding transplantation and death), were associated with higher risk [hazard ratio (95% CI)] for lung transplantation/death [4.23(2.27,7.91), 3.25(1.65,6.39), 2.74 (1.41,5.34), respectively]; individual parameters with higher risks were WHO FC deterioration [3.49 (1.47,8.29)], PAH-related hospitalisation [2.62(1.32,5.20)] and occurrence/worsening of≥2 PAH symptoms [2.13(1.02,4.45)].
Conclusions: These data support the use of cCW outcomes in paediatric PAH research. WHO FC deterioration, PAH-related hospitalisation, occurrence/worsening of≥2 PAH symptoms may be important for risk assessment during clinical management.
© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Keywords: Clinical worsening Paediatric PAH Clinical endpoints Disease worsening 1. Introduction
Paediatric pulmonary arterial hypertension (PAH) is a serious condi-tion with significant morbidity and mortality, however prognosis has improved with the use of PAH-specific therapies approved for use in adults [1]. As there are multiple similarities between paediatric and adult patients with PAH [1–3], diagnostic and therapeutic algorithms used in adult PAH have been adopted, with some adaptations, for dis-ease management in children, based on small-scale studies and expert opinion [3,4]. However, there are also important differences between
☆ Data sharing statement: The data and details of the analytic methods that support the findings of this analysis are available from the TOPP Executive Board on reasonable request.
⁎ Corresponding author at: Paediatric Cardiology Unit, Children's University Hospital, 6 rue Willy Donzé, 1211 Geneva 14, Switzerland.
E-mail address:maurice.beghetti@hcuge.ch(M. Beghetti).
1
This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
https://doi.org/10.1016/j.ijcard.2019.04.062
0167-5273/© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Contents lists available atScienceDirect
International Journal of Cardiology
children and adults with PAH [1,3], therefore specific data for paediatric patients are required from real-world evidence and clinical trials.
To assess disease progression, or clinical worsening (CW), in the paediatric PAH population with the same outcome measure across all age ranges is more difficult than in adults. For example, 6-minute-walk distance (6MWD) that is widely used in adult PAH studies to mea-sure exercise capacity is not a reliable outcome meamea-sure in paediatric studies of young children [3]. Right-sided heart catheterisation (RHC) is key at diagnosis and beneficial during follow-up in clinical practice for indications such as clinical deterioration, assessment of treatment effects, and listing for lung transplantation [1]; however, its use in pae-diatric clinical trials for the sole purpose of evaluating the efficacy of a drug is not supported given the risk involved with this invasive proce-dure [5]. These limitations of applying outcomes appropriate for adults to the paediatric population need to be overcome. Non-invasive clinical outcomes tailored for use in paediatric PAH are required in order to as-sess the efficacy of therapies in a clinical trial setting [6].
The use of composite CW outcomes is increasingly encouraged in PAH clinical research [7–9]. An evaluation of the Registry to Evaluate Early and Long-term PAH Disease Management (REVEAL) found that a composite definition of time to clinical worsening, and its individual components, was highly predictive of death in adults [10]. Recent clini-cal trials of adult PAH therapies have used morbidity/mortality compos-ite endpoints to demonstrate efficacy [11–13]. An observational study from a Dutch national referral centre including 70 patients reported on the predictive value of various composite CW definitions in paediat-ric PAH [14]. Therefore, we aimed to provide robust evidence that the use of composite endpoints might also be appropriate in paediatric PAH research if soft endpoints are revealed to be predictive of the hard endpoints lung transplantation and death.
Tracking Outcomes and Practice in Pediatric Pulmonary Hyperten-sion (TOPP) is a global, prospective, non-interventional registry conducted between January 2008 and July 2015. The registry was de-signed to provide information about demographics, disease course, treat-ment, and outcomes in paediatric pulmonary hypertension (PH) [15]. The objective of the current analysis of the TOPP registry was to evaluate the frequency of CW events using three different composite CW de fini-tions in newly diagnosed (incident) children with PAH, and to determine the predictive values of these CW definitions as well as their individual components for lung transplantation or death in this population.
2. Methods
2.1. Study design and patient population
The design of TOPP has been reported in detail previously [15]. In brief, TOPP enrolled patients with RHC-confirmed PAH (World Symposium on Pulmonary Hypertension [WSPH] group 1) or (World Health Organization) WHO group 3–5 PH, aged from 3 months to 18 years, with increased pulmonary vascular resistance and normal left-sidedfilling pressures. The TOPP protocol was approved by the institutional review boards and ethics committees of the centres contributing to the registry. Patient and/or parental consent was obtained for all patients. The authors had full access to all data and agree to be accountable for its integrity and analysis.
The current analysis included solely incident (diagnosis≤3 months before enrolment) patients aged≤17 years at diagnosis and enrolment, with idiopathic PAH (IPAH), familial PAH (FPAH) or PAH associated with congenital heart disease (APAH-CHD). Patients with APAH-CHD included those with an open, clinically significant, congenital systemic-to-pulmonary shunt (unrepaired, or repaired but with a substantial residual shunt), a corrected (closed) congenital systemic-to-pulmonary shunt, or with CHD that had never been associated with a congenital systemic-to-pulmonary shunt (i.e. transposition of the great arteriesflowing arterial switch) [15]. In addition, patients had to have at least one follow-up visit at any time after enrolment to be included in the analysis.
2.2. Variables
The following variables selected for the presented research were collected in the TOPP registry at enrolment and/or during follow-up: demographics (age, sex); disease charac-teristics: WHO functional class (WHO FC), presence/history, occurrence or progression of symptoms (dyspnoea, cyanosis, cough, fatigue, chest pain/discomfort, dizziness, near-syncope); PAH-related hospitalisations (hospitalisations because of increased right-sided heart failure or haemoptysis); occurrence of syncope; initiation of intravenous/
subcutaneous prostanoids; atrial septostomy; lung transplantation; and death. The com-posite CW definitions included the disease variables summarised inTable 1and named CW 1, 2, and 3. WHO FC deterioration was defined as a numerical increase of ≥1 FC, and, under this definition, patients classified as FC IV were not able to deteriorate further.
2.3. Statistical analysis
Descriptive statistics were used, including mean, SD, median, lower (Q1) and upper (Q3) quartiles, minimum and maximum for continuous variables, and counts and percent-ages for categorical variables. Rates forfirst events occurring after enrolment were calcu-lated, where the denominator was time at risk, defined as person-years observation time until thefirst event. For the composite CW definition, the event rate per 100 person-years was calculated using thefirst event to occur among all the components. Kaplan-Meier es-timates were conducted for time tofirst CW1, CW2, and CW3 events.
Factors predictive of lung transplantation or death were identified by Cox propor-tional hazards models. In order to evaluate the predictive value of the three CW definitions for lung transplantation or death, these two components were removed from the compos-ite and these CW definitions were named CW1′, CW2′ and CW3′.
First, the individual component variables at diagnosis were assessed (Table 1: E, F, G, and H) in univariate models.
Second, univariate and multivariate models of thefirst event of individual components (Table 1: C, D, E, F, G, H) were used. The survival time was determined in a time-dependent manner, starting at the time of thefirst event of the CW′ component. Multivariate models included only variables with pb 0.15 in the univariate models. A subgroup univariate anal-ysis was conducted according to PAH aetiology (IPAH/FPAH and APAH-CHD).
Statistical analyses were carried out using SAS v9.4 (SAS Institute, Cary, NC, USA).
3. Results
The analysis included 255 incident patients with PAH - 159 patients with IPAH/FPAH and 96 patients with APAH-CHD - from 34 centres in 20 countries (Supplemental material: Appendix). Patients were mostly female (60%) and had a median (Q1, Q3) age of 7 (3, 13) years of age at diagnosis. The WHO FC distribution at diagnosis was: FC I 33 (13%), FC II 109 (43%), FC III 87 (34%), and FC IV 24 (9%); two patients had missing WHO FC. The median (Q1, Q3) time between diagnosis and enrolment into TOPP was 0 (0, 1) months, and the median (Q1, Q3) observation time in TOPP was 33 (12, 55) months. At enrolment, 30 (12%) patients initiated intravenous/subcutaneous prostanoids. Presence/history of syncope was reported in 22 (9%) patients once, in 34 (13%) patients sev-eral times, in four (2%) patients regularly, and the information was
missing in two (1%) patients. Presence/history of≥2 PAH symptoms
was reported in 51 (20%) patients. Patient demographics and disease characteristics at enrolment for the two subgroups IPAH/FPAH and APAH-CHD are presented inTable 2.
Table 1
Composite clinical worsening definitions.
Clinical worsening 1 Clinical worsening 2 Clinical worsening 3 (A) Death (all-cause) (A) Death (all-cause) (A) Death (all-cause) (B) Lung
transplantation
(B) Lung transplantation (B) Lung transplantation (C) PAH-related hospitalisationa (C) PAH-related hospitalisationa (C) PAH-related hospitalisationa
(D) Atrial septostomy (D) Atrial septostomy (D) Atrial septostomy (E) WHO FC deteriorationb
(E) WHO FC deteriorationb
(F) Initiation of i.v./s.c. prostanoids (onlyfirst event) (F) Initiation of i.v./s.c. prostanoids (onlyfirst event) (G) Syncope (G) Syncope (H) Occurrence/worsening ≥2 PAH symptomsc
i.v., intravenous; PAH, pulmonary arterial hypertension; s.c., subcutaneous; WHO FC, World Health Organization functional class.
a
Increased right-heart failure, haemoptysis.
b
Increase≥1 WHO FC.
c
Progression (in patients with symptom at enrolment) and occurrence/progression (patients without condition at enrolment) of≥2 symptoms: dyspnoea (includes occur-rence/progression at rest and progression with exertion in patients with condition at en-rolment; and occurrence/progression of all dyspnoea in patients without symptom at enrolment), cyanosis, cough, fatigue, chest pain/discomfort, dizziness and near-syncope (occurrence only).
The event rates per 100 person-years are summarised inTable 3. The first-event rate (95% confidence interval [CI]) for CW1 was 23.1
(19.3–27.6) per 100 years (121 events over 524.7
person-years). For CW2, the first-event rate almost doubled to 43.6
(37.6–50.6) per 100 years (173 events over 396.5
person-years) with the additional components E, F, G (WHO FC deterioration, initiation of intravenous/subcutaneous prostanoids and syncope;
Table 1) added. Thefirst-event rate for CW3 was 46.3 (40.0–56.7) per 100 person-years (175 events over 396.5 person-years) with thefinal
component H (PAH symptoms; Table 1) added. The respective
Kaplan-Meier estimate analysis showed that at 6, 12, and 18 months a first CW1 event (95% CI) was experienced by 23.6% (18.8–29.4), 31.5% (26.0–37.7), and 36.9% (31.1–43.5) patients, respectively; a first CW2 event was experienced by 30.1% (24.8–36.2), 41.7% (35.8–48.2), and 54.9% (48.6–61.5) patients, respectively; and a first CW3 event was
ex-perienced by 33.3% (27.8–39.5), 44.5% (38.5–51.0), and 57.3%
(50.9–63.7) patients, respectively.
PAH-related hospitalisation was the most frequentfirst event in
CW1, with afirst-event rate of 17.9 (14.6–21.9) per 100 person-years,
followed by all-cause death (3.0 (1.9–5.0) per 100 person-years) and PAH-related death (1.7 (0.8–3.3) per 100 person-years). PAH-related hospitalisation and WHO FC deterioration were the most frequentfirst
events in CW2, withfirst-event rates of 17.9 (14.2–22.6) and 16.6
(13.1–21.2) per 100 person-years, respectively, and in CW3, with first event rates of 16.2 (12.6–20.8) and 15.4 (11.9–19.9) per 100 person-years, respectively. Atrial septostomy occurred as thefirst event in only one patient.
When evaluating factors predictive of lung transplantation or death in a univariate analysis, patients with WHO FC III at diagnosis were at significantly higher risk than patients with WHO FC I/II; hazard ratio (HR) 1.93 (95% CI, 1.07–3.49); p = 0.03 (Table 4).
Table 2
Patient demographics and disease characteristics at enrolment. IPAH/FPAH n = 159
APAH-CHD n = 96
Female, n (%) 103 (65) 50 (52)
Median age at diagnosis (Q1, Q3), years 8 (3, 13) 5 (2, 12) WHO FC, n (%) I 17 (11) 16 (17) II 66 (42) 43 (45) III 55 (35) 32 (33) IV 19 (12) 5 (5) Missing 2 0
Median observation period (Q1, Q3), months 33 (12, 50) 34 (14, 58) Median time from diagnosis to enrolment
(Q1, Q3), months
0 (0, 1) 0 (0, 1) Initiation of i.v./s.c. prostanoids, n (%) 29 (18) 1 (1) Syncope Never 104 (66) 89 (93) Once 17 (11) 5 (5) Several times 32 (20) 2 (2) Regularly 4 (3) 0 (0) Missing 2 0
Patients with presence/history≥2 PAH symptoms, n (%) 39 (25) 12 (13) Percentages are based on the number of patients (n) in the analysis set and groups/sub-groups with non-missing data. APAH-CHD, pulmonary arterial hypertension associated with congenital heart disease; IPAH/FPAH, idiopathic pulmonary arterial hypertension/fa-milial pulmonary arterial hypertension; i.v., intravenous; PAH, pulmonary arterial hyperten-sion; Q, quartile; s.c., subcutaneous; WHO FC, World Health Organization functional class.
Table 3
Summary of composite clinical worsening definitions.
Clinical worsening 1 (N = 255) Clinical worsening 2 (N = 255) Clinical worsening 3 (N = 255) Number of events 121 173 175 Person-years 524.7 396.5 377.6 First-event ratea (95% CI) 23.1 (19.3–27.6) 43.6 (37.6–50.6) 46.3 (40.0–53.7)
First event in the respective composite definition n Event ratea
95% CI n Event ratea 95% CI n Event ratea 95% CI Death All-cause 16 3.0 (1.9–5.0) 14 3.5 (2.1–6.0) 14 3.7 (2.2–6.3) PAH-related 9 1.7 (0.8–3.3) 8 2.0 (0.9–4.0) 8 2.1 (0.9–4.2) PAH-related hospitalisation 94 17.9 (14.6–21.9) 71 17.9 (14.2–22.6) 61 16.2 (12.6–20.8) Transplantation 1 0.2 (0.0–1.1) 1 0.3 (0.0–1.4) 1 0.3 (0.0–1.5) Atrial septostomy 1 0.2 (0.0–1.1) 1 0.3 (0.0–1.4) 1 0.3 (0.0–1.5) Deterioration in WHO FC – – – 66 16.6 (13.1–21.2) 58 15.4 (11.9–19.9)
Initiation of i.v./s.c. prostanoids – – – 4 1.0 (0.3–2.6) 4 1.1 (0.3–2.7)
Syncope – – – 8 2.0 (0.9–4.0) 7 1.9 (0.7–3.8)
Occurrence/worsening≥2 PAH symptoms – – – – – – 21 5.6 (3.6–8.5)
CI, confidence interval; i.v., intravenous; PAH, pulmonary arterial hypertension; s.c., subcutaneous; WHO FC, World Health Organization functional class.
a
Rate per 100 person-years.
Table 4
Predictive factors for lung transplantation or death in the overall population. Hazard ratio 95% CI p-Value Prognostic factors at diagnosis for lung transplantation or death (univariate) WHO FC (III vs. I/II)a
1.93 1.07–3.49 0.03 WHO FC (IV vs. I/II)a
1.33 0.45–3.89 0.60 Presence/history of≥2 PAH symptoms 1.22 0.69–2.18 0.50 Syncope (regularly vs. never)a 1.16 0.16–8.47 0.88
Syncope (once vs. never)a
0.92 0.33–2.57 0.87 Syncope (several times vs. never)a
0.90 0.38–2.12 0.80 Initiation of i.v./s.c. prostanoids 0.69 0.25–1.91 0.47 Prognostic value of component'sfirst events for lung transplantation or death:
individual disease variables (univariate)
WHO FC deterioration 6.70 3.12–14.36 b0.0001 Occurrence/worsening of≥2 PAH symptoms 4.77 2.54–8.97 b0.0001 Initiation of i.v./s.c. prostanoids 4.59 2.13–9.87 b0.0001 PAH-related hospitalisation 4.22 2.26–7.89 b0.0001 Occurrence of syncope 2.34 0.94–5.81 0.07 Prognostic value of component'sfirst events for lung transplantation or death:
individual disease variables (multivariate)b
WHO FC deterioration 3.49 1.47–8.29 b0.01 PAH-related hospitalisation 2.62 1.32–5.20 b0.01 Occurrence/worsening≥2 PAH symptoms 2.13 1.02–4.45 0.045 Prognostic value of composite clinical worsening outcomes for lung
transplantation or death (univariate)c
Clinical worsening 1 4.23 2.27–7.91 b0.0001 Clinical worsening 2 3.25 1.65–6.39 b0.001 Clinical worsening 3 2.74 1.41–5.34 b0.01 CI, confidence interval; i.v., intravenous; PAH, pulmonary arterial hypertension; s.c., sub-cutaneous; WHO FC, World Health Organization functional class.
aData are missing for two patients. b
Multivariate analysis conducted on variables with pb 0.15 in univariate models.
c
Lung transplantation and death were removed from the analysis of the clinical wors-ening composite endpoints.
In the remaining analyses the survival time was defined in a time-dependent manner. In the univariate analysis of thefirst event of indi-vidual components (Table 1: C, D, E, F, G, H), WHO FC deterioration had the strongest predictive value for lung transplantation or death (HR, 6.70; 95% CI, 3.12–14.36), followed by occurrence/worsening of ≥2 PAH symptoms (HR, 4.77; 95% CI, 2.54–8.97), and initiation of intra-venous/subcutaneous prostanoids (HR, 4.59; 95% CI, 2.13–9.87); all p b 0.0001 (Table 4). In multivariate analysis, afirst event of WHO FC dete-rioration (HR, 3.49; 95% CI, 1.47–8.29), PAH-related hospitalisation (HR, 2.62; 95% CI, 1.32–5.20), and occurrence or worsening of ≥2 PAH symp-toms (HR, 2.13; 95% CI, 1.02–4.45) were significant independent risk factors for lung transplantation or death (Table 4).
All three CW′ definitions (excluding the components lung transplan-tation and death) were associated with a significantly higher risk for lung transplantation or death (Table 4).
Predictive models for lung transplantation or death according to aetiology are shown in Supplementary Table 1. For the subgroup of pa-tients with IPAH/FPAH (n = 159), the univariate and multivariate anal-yses found results similar to those in the overall population. However, for the subgroup of patients with APAH-CHD (n = 96), none of the
CW definitions or individual components were significant predictive
factors for lung transplantation or death (Supplementary Table 1). 4. Discussion
The current analysis, using data from the TOPP registry, a large co-hort of paediatric patients with PAH, primarily aimed to create and de-termine the predictive value of a CW endpoint (including death and transplantation) that could be used in clinical trials. The analysis also describes the relevance of composite CW definitions (excluding death and transplantation) to predict lung transplantation or death, which could be used for future prognostic studies and to identify patients with a higher risk of disease progression. The CW definitions comprised a combination of disease variables. All-cause death, PAH-related hospitalisation (owing to increased right-sided heart failure or haemoptysis), lung transplantation, and atrial septostomy were in-cluded in all three CW definitions, with the addition of WHO FC deteri-oration, initiation of intravenous/subcutaneous prostanoids and
syncope to CW2 and CW3, and occurrence or worsening of≥2 PAH
symptoms to CW3. These CW definitions were designed to evaluate
the differences in the proportion of events during follow-up when more subjective clinical outcomes were introduced into the definition. Furthermore, this analysis assessed the predictive value of the individ-ual disease variables and the CW′ definition (excluding lung transplan-tation and death) for lung transplantransplan-tation or death.
Ourfirst interesting result is that the number of events in a paediat-ric cohort of PAH is high regardless of the CW definition used. As ex-pected, the number of events increases when more subjective clinical outcomes are included in the CW definition. The observed event rate, ranging from 23.1 to 46.3 per 100 person-years is consistent with an earlier study of 70 patients with paediatric PAH that showed the num-ber of clinical events was high in the study population, when consider-ing a composite definition of death, lung transplantation, PAH-related hospitalisation, initiation of intravenous prostanoids and functional de-terioration (WHO FC and/or 6MWD) [14].
The second interesting result is the frequency offirst CW event, in this population of incident patients. This high frequency of events and their precocity are valuable information for designing future trials as paediatric PAH is a rare and heterogeneous disease, and reducing the need for large cohorts of patients to evaluate the efficacy of a new com-pound or treatment strategy is a crucial issue. Knowing the frequency and timing of CW events is important information for sample size calcu-lation of future trials.
The third notablefinding generated by the current study is the pre-dictive value of individual disease variables WHO FC deterioration,
PAH-related hospitalisation, and the occurrence or worsening of≥2 PAH
symptoms. These three variables were predictive of lung transplanta-tion or death in both the univariate and multivariate models. These find-ings are consistent with and expand upon the results of the national Dutch cohort study that found FC deterioration to be predictive of lung transplantation or death in paediatric PAH [14,16]. They are also in agreement with an analysis of adult patients with PAH that found re-duced survival in patients with unchanged or worsened WHO-FC
com-pared with patients whose WHO FC improved [17]. Furthermore, the
results from the current analysis provide evidence for thefirst time that the occurrence or progression of≥2 PAH symptoms is predictive of lung transplantation or death in paediatric PAH. We also show here that PAH-related hospitalisation was the most frequentfirst event in the CW definitions, similar to the findings in adult PAH trials that have used composite morbidity/mortality endpoints including PAH-related hospitalisation. Hospitalisation for PAH deterioration is certainly an im-portant factor as mortality is higher in paediatric patients who are hospitalised for PH compared with general paediatric hospitalisations
[18]. An analysis of the REVEAL cohort found first PAH-related
hospitalisations to be more frequent in adult than paediatric PAH patients, and predictive of subsequent rehospitalisation and poorer 3-year survival [19]. It is of note that the paediatric REVEAL analysis in-cluded mainly prevalent patients diagnosed during childhood. The cur-rent study only included incident patients; therefore, ourfindings are potentially more robust for predicting long-term outcomes, as the im-pact of a survival bias is minimised compared with prevalent cohorts.
Syncope was not found to be predictive of lung transplantation or death in this cohort, which is in line with results of previous cohort studies [20,21]. Recently, in a small study of 55 paediatric patients with PAH, although the absence of syncope was linked to improved prognosis, the presence of syncope was not considered to be a risk factor for lung transplantation or death [22]. We believe that this information has a clinical importance for the decision to escalate therapy. Indeed, the practice of prescribing an additional drug or recommending a paediatric patient for lung transplantation when syncope recurs should certainly be reviewed in the light of thesefindings.
Composite CW endpoints have been adopted in recent clinical trials of adult PAH therapies as morbidity/mortality endpoints [11–13], and their use is being encouraged in paediatric PAH clinical trials [2,5,23]. Analysis of composite CW endpoints has always been a challenge be-cause of the lack of a standardised definition [24,25]. The current study evaluated the relevance of three different definitions. Interest-ingly, all three CW composite outcomes were significantly predictive of lung transplantation or death. This was in agreement with the find-ings of a recent cohort study of paediatric patients in the Netherlands [14]. These current analyses may inform the design and sample size cal-culation of future paediatric PAH trials and support the use of a
compos-ite CW endpoint in those trials. Indeed, the first paediatric trial
incorporating these composite endpoints in the design was initiated in 2017 [26].
Compared with results in the overall population, similar results were observed in the subgroup of patients with IPAH/FPAH, with an even greater predictive value of lung transplantation or death for several of the individual disease variables (Supplementary Table 1). Conversely, no significant predictors of lung transplantation or death could be iden-tified in the APAH-CHD subgroup. However, this subgroup included het-erogeneous underlying CHD physiologies and the number of outcome events was small, which contributed to the wide CIs. Indeed, patients with Eisenmenger physiology have different survival outcomes than those with a closed shunt and PAH [27–29]. Identifying specific predic-tors in subsets with APAH-CHD would require large patient numbers and, potentially, other factors in the predictive model, such as the type of shunt (pre- or post-tricuspid), and the type of underlying CHD.
Thefindings of the current study define a clinical endpoint for fur-ther use in clinical trials. A PAH worsening definition based on occur-rence or progression of symptoms may be predictive for death and lung transplantation, and therefore could be used for prognostic studies
and in clinical practice to identify patients at higher risk of disease pro-gression and poorer outcomes. The current study identified the occur-rence of symptoms as a reliable predictor of outcomes in paediatric PAH. The different CW definitions may be useful for risk stratification of a patient during follow-up.
4.1. Limitations
There are certain limitations to this study inherent to observational studies. Further analyses are warranted including analysis for compet-ing risks, adjustment for treatment patterns and other potential con-founders. Such additional analysis could be considered based on an updated cohort from the TOPP registry as well as from controlled clini-cal trials. The study only defined PAH worsening and not PAH improve-ment. This information would have been important to further guide the management of these patients in clinical practice, and to facilitate the design of future paediatric PAH trials. Similarly, therapeutic targets were not confirmed in this cohort. These targets are scarcely identified in paediatric PAH [16]. The Dutch cohort found that improvements in WHO FC, N-terminal brain pronatriuretic peptide, and tricuspid annular plane systolic excursion after treatment are associated with better sur-vival [16]. The learnings from the TOPP-1 registry have informed the on-going TOPP-2 registry [30], which aims to further develop treatment goals that were proposed at the 2013 World Symposium on PH [1], as well as to identify new treatment goals. This database will focus more on the collection of follow-up variables, such as NT-proBNP, which have become important for risk assessment in recent years and may provide further analysis of the associations between these variables and clinical endpoints.
5. Conclusions
The results from this analysis of the TOPP registry suggest that differ-ent CW definitions could be used as primary endpoints in future paedi-atric PAH trials and prognostic studies. Indeed, the number of events according to the different definitions is high and the events occur early in the course of the disease, suggesting that the number of patients needed to be included in randomised control trials could be reduced compared with the event-driven trials in adult PAH [12,31]. In addition, WHO FC deterioration, PAH-related hospitalisation, and occurrence or worsening of≥2 PAH symptoms are predictive of lung transplantation or death. Thesefindings are of importance for individual monitoring as well as for the individual risk assessment and therapeutic decisions during follow-up in paediatric PAH patients.
Funding
This was a collaborative study between the Association for Paediatric Pulmonary Hypertension (PePH) and Actelion Pharmaceuticals Ltd. This work was supported by Actelion Pharmaceuticals Ltd.
Disclosures
M. Beghetti has received grant/research support from Actelion Phar-maceuticals Ltd. and Bayer HealthCare; has been a consultant for: Actelion Pharmaceuticals Ltd., Bayer HealthCare, Eli Lilly & Co., and GlaxoSmithKline; and has attended speakers bureau(s) for Actelion Pharmaceuticals Ltd. M. Brand is an employee of Actelion Pharmaceuti-cals Ltd. and holds modest stock in Actelion. The University Medical Center Groningen has received research support for R.M.F. Berger from Actelion Pharmaceuticals Ltd. The University Medical Centre Groningen has contracted with Actelion Pharmaceuticals Ltd., GlaxoSmithKline, Eli Lilly & Co., Bayer HealthCare for consultancy activ-ities of R.M.F. Berger; T. Humpl is a consultant for Actelion Pharmaceu-ticals Ltd. J.G. Wheeler has nothing to disclosure. The University of Colorado has received research support for D.D. Ivy from Actelion
Pharmaceuticals Ltd., BayerHealthCare, Eli Lilly & Co., United Therapeu-tics, and GlaxoSmithKline; the University of Colorado contracts with Actelion Pharmaceuticals Ltd., Bayer HealthCare, Eli Lilly & Co., and United Therapeutics for D.D. Ivy to be a consultant. D.D. Ivy receives grant funding from the National Institutes of Health and the Food and Drug Administration. D. Bonnet has received grant/research support from Actelion Pharmaceuticals Ltd; and has been a consultant for Actelion Pharmaceuticals Ltd, Bayer HealthCare, and Eli Lilly & Co.
Acknowledgements
The authors thank the TOPP investigators for their participation. The authors also thank Peter Cornelisse, employed by Actelion Pharmaceu-ticals Ltd., for providing statistical support. Actelion PharmaceuPharmaceu-ticals Ltd. funded editorial writing support provided by Lynda McEvoy, PhD, ApotheCom, London, UK, and editorial/submission support provided by Richard McDonald, Watermeadow Medical, Oxfordshire, UK, an Ashfield company, part of UDG Healthcare plc.
Appendix A. Supplementary data
Supplementary data to this article can be found online athttps://doi. org/10.1016/j.ijcard.2019.04.062.
References
[1] D.D. Ivy, S.H. Abman, R.J. Barst, R.M. Berger, D. Bonnet, T.R. Fleming, et al., Pediatric pulmonary hypertension, J. Am. Coll. Cardiol. 62 (2013) D117–D126.
[2] S.G. Haworth, M. Beghetti, Assessment of endpoints in the pediatric population: congenital heart disease and idiopathic pulmonary arterial hypertension, Curr. Opin. Pulm. Med. 16 (Suppl. 1) (2010) S35–S41.
[3] R.J. Barst, S.I. Ertel, M. Beghetti, D.D. Ivy, Pulmonary arterial hypertension: a compar-ison between children and adults, Eur. Respir. J. 37 (2011) 665–677.
[4] S.H. Abman, G. Hansmann, S.L. Archer, D.D. Ivy, I. Adatia, W.K. Chung, et al., Pediatric pulmonary hypertension: guidelines from the American Heart Association and American Thoracic Society, Circulation. 132 (2015) 2037–2099.
[5] H. Sun, N. Stockbridge, R.L. Ariagno, D. Murphy, R.M. Nelson, W. Rodriguez, Reliable and developmentally appropriate study end points are needed to achieve drug de-velopment for treatment of pediatric pulmonary arterial hypertension, J Perinatol: official journal of the California Perinatal Association 36 (2016) 1029–1033.
[6] R.M. Berger, Pulmonary hypertension: smaller kids, smaller steps, Lancet Respir. Med. 2 (2014) 348–350.
[7] M.M. Chakinala, R. Barst, From short-term benefits to long-term outcomes: the evo-lution of clinical trials in pulmonary arterial hypertension, Pulm. Circ. 3 (2013) 507–522.
[8] I.R. Preston, S. Suissa, M. Humbert, New perspectives in long-term outcomes in clin-ical trials of pulmonary arterial hypertension, Eur. Respir. Rev. 22 (2013) 495–502.
[9] Committe for Medicinal Products for Human Use (CHMP), Guidelines on the Clinical Investigations of Medicinal Products for the Treatment of Pulmonary Arterial Hyper-tension, http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_ guideline/2009/12/WC500016686.pdfOctober 2009.
[10] A.E. Frost, D.B. Badesch, D.P. Miller, R.L. Benza, L.A. Meltzer, M.D. McGoon, Evaluation of the predictive value of a clinical worsening definition using 2-year outcomes in patients with pulmonary arterial hypertension: a REVEAL Registry analysis, Chest. 144 (2013) 1521–1529.
[11]T. Pulido, I. Adzerikho, R.N. Channick, M. Delcroix, N. Galie, H.A. Ghofrani, et al., Macitentan and morbidity and mortality in pulmonary arterial hypertension, N. Engl. J. Med. 369 (2013) 809–818.
[12] O. Sitbon, R. Channick, K.M. Chin, A. Frey, S. Gaine, N. Galie, et al., Selexipag for the treatment of pulmonary arterial hypertension, N. Engl. J. Med. 373 (2015) 2522–2533.
[13] N. Galie, J.A. Barbera, A.E. Frost, H.A. Ghofrani, M.M. Hoeper, V.V. McLaughlin, et al., Initial use of ambrisentan plus tadalafil in pulmonary arterial hypertension, N. Engl. J. Med. 373 (2015) 834–844.
[14] M.J. Ploegstra, S. Arjaans, W.M. Zijlstra, J.M. Douwes, T.R. Vissia-Kazemier, M.T. Roofthooft, et al., Clinical worsening as composite study end point in pediatric pul-monary arterial hypertension, Chest. 148 (2015) 655–666.
[15]R.M. Berger, M. Beghetti, T. Humpl, G.E. Raskob, D.D. Ivy, Z.C. Jing, et al., Clinical features of paediatric pulmonary hypertension: a registry study, Lancet (London, England) 379 (2012) 537–546.
[16] M.J. Ploegstra, J.M. Douwes, M.T. Roofthooft, W.M. Zijlstra, H.L. Hillege, R.M. Berger, Identification of treatment goals in paediatric pulmonary arterial hypertension, Eur. Respir. J. 44 (2014) 1616–1626.
[17]R.J. Barst, L. Chung, R.T. Zamanian, M. Turner, M.D. McGoon, Functional class im-provement and 3-year survival outcomes in patients with pulmonary arterial hy-pertension in the REVEAL Registry, Chest. 144 (2013) 160–168.
[18]D.B. Frank, M.A. Crystal, D.L. Morales, K. Gerald, B.D. Hanna, G.B. Mallory Jr., et al., Trends in pediatric pulmonary hypertension-related hospitalizations in the United States from 2000-2009, Pulmonary circulation. 5 (2015) 339–348.
[19] C.D. Burger, P.K. Long, M.R. Shah, M.D. McGoon, D.P. Miller, A.J. Romero, et al., Char-acterization offirst-time hospitalizations in patients with newly diagnosed pulmo-nary arterial hypertension in the REVEAL Registry, Chest. 146 (2014) 1263–1273.
[20]S. Moledina, A.A. Hislop, H. Foster, I. Schulze-Neick, S.G. Haworth, Childhood idio-pathic pulmonary arterial hypertension: a national cohort study, Heart 96 (2010) 1401–1406.
[21]R.L. van Loon, M.T. Roofthooft, M. van Osch-Gevers, T. Delhaas, J.L. Strengers, N.A. Blom, et al., Clinical characterization of pediatric pulmonary hypertension: complex presentation and diagnosis, J. Pediatr. 155 (2009) 176–182 (e1).
[22] M. Zuk, K. Mazurkiewicz-Anton, A. Migdal, D. Jagiellowicz-Kowalska, A. Turska-Kmiec, L. Ziolkowska, et al., Prognosis in children with pulmonary arterial hyperten-sion: 10-year single-centre experience, Kardiol. Pol. 74 (2016) 159–167.
[23] R.D. Torok, J.S. Li, P.J. Kannankeril, A.M. Atz, R. Bishai, E. Bolotin, et al., Recommenda-tions to enhance pediatric cardiovascular drug development: report of a multi-stakeholder think tank, J. Am. Heart Assoc. 7 (2018).
[24] N. McGlinchey, A.J. Peacock, Endpoints in PAH clinical trials in the era of combina-tion therapy: how do we decide whether something is working without going bankrupt? Drug Discov. Today 19 (2014) 1236–1240.
[25] N. Galie, G. Simonneau, R.J. Barst, D. Badesch, L. Rubin, Clinical worsening in trials of pulmonary arterial hypertension: results and implications, Curr. Opin. Pulm. Med. 16 (Suppl. 1) (2010) S11–S19.
[26] Actelion Pharmaceuticals Ltd. A Study to Find Out Whether the Medicine Macitentan Works in Children With Pulmonary Arterial Hypertension (PAH) (TOMORROW),https://clinicaltrials.gov/ct2/show/NCT029324102016.
[27] S.G. Haworth, A.A. Hislop, Treatment and survival in children with pulmonary arte-rial hypertension: the UK Pulmonary Hypertension Service for Children 2001-2006, Heart 95 (2009) 312–317.
[28] R.L. van Loon, M.T. Roofthooft, T. Delhaas, M. van Osch-Gevers, A.D. ten Harkel, J.L. Strengers, et al., Outcome of pediatric patients with pulmonary arterial hyperten-sion in the era of new medical therapies, Am. J. Cardiol. 106 (2010) 117–124.
[29] A. Manes, M. Palazzini, E. Leci, M.L. Bacchi Reggiani, A. Branzi, N. Galie, Current era sur-vival of patients with pulmonary arterial hypertension associated with congenital heart disease: a comparison between clinical subgroups, Eur. Heart J. 35 (2014) 716–724.
[30] Association for Pediatric Pulmonary Hypertension. Tracking Outcomes and Practice in Pediatric Pulmonary Hypertension (TOPP-2).https://clinicaltrials.gov/ct2/show/ NCT026106602015.
[31] V.V. McLaughlin, M.M. Hoeper, R.N. Channick, K.M. Chin, M. Delcroix, S. Gaine, et al., Pulmonary arterial hypertension-related morbidity is prognostic for mortality, J. Am. Coll. Cardiol. 71 (2018) 752–763.
