Clinical Worsening as Composite Study End Point in Pediatric Pulmonary Arterial Hypertension

Transcription

1 [ Original Research Pulmonary Vascular Disease ] Clinical Worsening as Composite Study End Point in Pediatric Pulmonary Arterial Hypertension Mark-Jan Ploegstra, MD ; Sanne Arjaans, BSc ; Willemijn M. H. Zijlstra, BSc ; Johannes M. Douwes, MD ; Theresia R. Vissia-Kazemier, RN, MANP ; Marcus T. R. Roofthooft, MD, PhD ; Hans L. Hillege, MD, PhD ; and Rolf M. F. Berger, MD, PhD BACKGROUND: Clinical worsening (CW), an increasingly used composite end point in adult pulmonary arterial hypertension (PAH), has not yet been evaluated in pediatric PAH. This study aims to evaluate the usefulness of CW in pediatric PAH by assessing the event incidence and prognostic value of each separate component of CW and of the composite CW end point. METHODS: Seventy pediatric patients with PAH from the Dutch National Network for Pediatric Pulmonary Hypertension, who started PAH-targeted therapy between January 2000 and January 2014, were included in the study and underwent standardized follow-up. The following CW components were prospectively registered: death, lung transplantation (LTx), PAH-related hospitalizations, initiation of IV prostanoids, and functional deterioration (World Health Organization functional-class deterioration, 15% decrease in 6-min walk distance, or both). The longitudinal event incidence and prognostic value were assessed for each separate component and their combination. RESULTS: The end-point components of death, LTx, hospitalizations, initiation of IV prostanoids, and functional deterioration occurred with a longitudinal event rate of 10.1, 2.5, 21.4, 9.4 and 48.1 events per 100 person-years, respectively. The composite CW end point occurred 91.5 times per 100 person-years. The occurrences of either hospitalization, initiation of IV prostanoids, or functional deterioration were predictive of death or LTx ( P,.001 for each component). In this cohort, 1-, 3-, and 5-year transplant-free survival was 76%, 64%, and 56%, respectively. Freedom from CW at 1, 3, and 5 years was 43%, 22%, and 17%, respectively. CONCLUSIONS: CW occurred with a high event incidence and each of the soft end-point components was predictive of death or LTx. This supports the usefulness of CW as a study end point in clinical trials in pediatric PAH. CHEST 2015; 148 ( 3 ): Manuscript received December 8, 2014; revision accepted February 2, 2015; originally published Online First March 5, ABBREVIATIONS: 6MWD 5 6-min walk distance; APAH-CHD 5 pulmonary arterial hypertension associated with congenital heart disease; APAH-non-CHD 5 pulmonary arterial hypertension associated with conditions other than congenital heart disease; CW 5 clinical worsening; HPAH 5 heritable pulmonary arterial hypertension; IPAH 5 idiopathic pulmonary arterial hypertension; IQR 5 interquartile range; LTx 5 lung transplantation; PAH 5 pulmonary arterial hypertension; REVEAL Registry 5 Registry to Evaluate Early and Long-term PAH Disease Management; WHO-FC 5 World Health Organization functional class AFFILIATIONS : From the Center for Congenital Heart Diseases (Drs Ploegstra, Douwes, Roofthooft, and Berger and Mss Arjaans, Zijlstra, and Vissia-Kazemier), Department of Pediatric Cardiology, Beatrix Children s Hospital, and the Department of Epidemiology (Dr Hillege), University Medical Center Groningen, University of Groningen, Groningen, The Netherlands. Dr Ploegstra and Ms Arjaans contributed equally to this manuscript. Results from this study were presented at the European Society of Cardiology Congress 2014, August 30, 2014, Barcelona, Spain. FUNDING/SUPPORT : This study was supported by the Sebald Foundation. CORRESPONDENCE TO: Mark-Jan Ploegstra, MD, Center for Congenital Heart Diseases, University Medical Center Groningen, Hanzeplein 1, PO Box , 9700 RB Groningen, The Netherlands; m.ploegstra@ umcg.nl 2015 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: /chest journal.publications.chestnet.org 655

2 Pediatric pulmonary arterial hypertension (PAH) is a severe, progressive disease of the pulmonary vasculature and has an unsatisfactory prognosis despite the introduction of PAH-targeted therapies. 1-3 Most drugs currently used in the treatment of PAH have not been evaluated in pediatric clinical trials. 4,5 This is largely explained by the rarity and heterogeneity of pediatric PAH, leading to small study cohorts, but is also due to the lack of appropriate outcome parameters to evaluate drug efficacy. 6-8 Time to death would seem the most robust trial end point, as improving survival is the first priority in treating pediatric PAH. 3 However, death as an end point would require long-duration clinical trials in a very vulnerable group of pediatric patients unable to give consent, thereby challenging study ethics and leading to high costs. 7,8 Short-duration clinical trials with lower numbers of patients required are preferable but need an alternative end point to obtain sufficient statistical power. Such an end point should be either a direct or surrogate measure of how a patient feels, functions, or survives 9 and would ideally be able to be measured earlier and more frequently than the final end point of interest. 10 Such an end point would lead to increased statistical power, reduction of required study participants, shorter study periods, and lower costs. 11 Th e 6-min walk distance (6MWD) has been the most commonly used primary end point in the pivotal trials in adult PAH. 12,13 The absolute value of 6MWD is regarded as a clinically meaningful end point, measuring how a patient functions. Moreover, 6MWD has been demonstrated to be an independent predictor of mortality in adults and in children. 7 years old. 14,15 However, in the current era with accumulating treatment modalities, more ambitious treatment effects such as improved morbidity and mortality are desired. Evidence suggests that changes in 6MWD are not accurate surrogates for disease progression or sur- vival in adults or children. 13,16,17 Th is challenges the usefulness of 6MWD as an end point and has led to a call for alternative, more clinically meaningful end points. Clinical worsening (CW) has been suggested as an alternative end point in PAH CW consists of a combination of hard unambiguous events such as death and lung transplantation (LTx), and softer events, including hospitalizations, need for additional therapy, and worsening FOR EDITORIAL COMMENT SEE PAGE 576 of function. CW has been used for some time as a primary or secondary end point in adult trials, and its validity has been evaluated in adults. 40 Using 2-year outcome data from the Registry to Evaluate Early and Long-term PAH Disease Management (REVEAL Registry), it was shown that the soft CW end-point components were highly predictive of subsequent mortality. As 6MWD is not reliable in young children, this end point is not feasible for the pediatric age group. Although not yet evaluated, CW might be an appealing clinical end point in pediatric PAH, since it provides a patientcentered composite end point that decreases the required study participants and it would be applicable in different age groups. Moreover, it would account for the risk of rapid clinical deterioration in children. 8 However, before CW can be used in clinical trials, essential evaluation steps are required that would include a description of how frequent the end-point components of CW occur, how the soft end-point components relate to mortality, and what the timing of CW is compared with mortality. 10 Therefore, the primary aim of this study was to evaluate the usefulness of CW in pediatric PAH by assessing event incidence and prognostic value of each separate component end point and of the composite CW end point. The secondary aim was to describe the timing of CW compared with death or LTx in pediatric PAH. Materials and Methods Study Design and Population Th is study is a retrospective analysis of data from a prospective clinical registry. In The Netherlands, all children with PAH are referred to the University Medical Center Groningen, which serves as the national referral center of the Dutch National Network for Pediatric Pulmonary Hypertension. 41 Children are followed and registered prospectively according to a standardized protocol. Ethical approval for this ongoing registry was obtained from the institutional review board (medical ethics review board of the University Medical Center Groningen, approval number M ) and the subjects and/or their guardians provided written informed consent at enrollment. All treatment-naive patients in whom PAH-targeted therapy was initiated between January 1, 2000, and January 1, 2014, were included in this study. End-Point Definition and Data Collection The definition of CW included the following end-point components: (1) death; (2) LTx; (3) nonelective PAH-related hospitalizations, including hospitalizations for atrial septostomies; (4) initiation of IV prostanoids; and (5) functional deterioration, defined as either worsening of World Health Organization functional class (WHO-FC), 15% decrease in 6MWD, or both. This CW definition is in-line with various CW end points used in adult PAH trials and as proposed in consensus statements. 42,43 As the change in 6MWD as an end point has been challenged, a sensitivity analysis was performed with defining functional deterioration as worsening of WHO-FC only. The CW components 656 Original Research [ 148 # 3 CHEST SEPTEMBER 2015 ]

3 were longitudinally registered from initiation of PAH-targeted therapy until the last follow-up visit before January 1, Data Analysis Data are presented as mean SD, median (interquartile range [IQR]), or frequencies (percentage). Statistical analysis was performed using IBM SPSS version 22.0 (IBM Corporation). All statistical tests were two-sided and P values,.05 were considered statistically significant. For the first functional deterioration event, follow-up WHO-FC and 6MWD were compared with the best achieved WHO-FC or 6MWD in the first year after treatment initiation. For consecutive events, WHO-FC and 6MWD were compared with the preceding follow-up visit. As WHO-FC IV indicates a functional status where further deterioration is not possible due to a ceiling effect, WHO-FC IV was always regarded as a functional deterioration event, also when it had been present at baseline. To describe the longitudinal event incidence of the end points, the longitudinal event rate per 100 person-years was calculated for each component and for the composite of these using the following formula: event rate per 100 person-years 5 total number events/(total observation time/100). In this analysis, patients could experience multiple events and were censored at death or end of follow-up. A separate eventrate analysis was performed involving first events only, with censoring at time of the event or end of follow-up. Cumulative event incidence curves were depicted for the separate end-point components. To assess the relationship between the soft and hard end-point components, the prognostic value of the first occurring soft CW components 3, 4, and 5 were assessed using time-dependent Cox regression analysis with the hard end point of death or LTx as the dependent analysis outcome (component 1 1 2). To describe the timing of CW in relation to death or LTx, both survival and event-free survival were reported and compared using Kaplan-Meier analysis. To align the study sample with the specific setting of a clinical trial, all analyses were repeated for a subgroup of what we called trial-eligible patients. These patients were defined as not hospitalized, not in WHO-FC IV, and/or not immediately started on IV prostanoids at baseline. The results from these analyses are presented as supplemental material. Results Patient Characteristics In total, 70 patients were included in this study. The expected nationwide representation is confirmed by the fact that the current sample size over a 14-year period is in-line with several reports on annual PAH incidence (reports ranging from one to three cases per million; the Dutch pediatric population is approximately 3 million). 2,44,45 Patient characteristics at treatment initiation (baseline) are shown in Table 1. Median age was 8.0 years (IQR, years). In total, 37 patients were diagnosed with idiopathic PAH (IPAH) or heritable PAH (HPAH), 25 with PAH associated with congenital heart disease (APAH-CHD), and eight with PAH associated with conditions other than congenital heart disease (APAH-non-CHD). Of the 25 patients with APAH-CHD, 16 (64%) had Eisenmenger physiology. Patients were followed for a median of 39 months (IQR, months), for a total observation time of person-years. Event Incidence During the observation time, 28 patients (40%) died and seven (10%) underwent LTx ( Table 2 ). The soft end-point components hospitalizations, initiation of IV prostanoids, and functional deterioration occurred in 38 (54%), 26 (37%), and 50 (71%) patients, respectively. The longitudinal event rates of the separate components were as follows: 10.1 deaths, 2.5 times LTx, 21.4 hospitalizations, 9.4 initiations of IV prostanoids, and 48.1 functional deteriorations per 100 person-years ( Table 2 ). The corresponding cumulative event-incidence curves are depicted in Figure 1 and show that a substantial proportion of the patients experienced components 3 and 5 more than once. The composite CW end point occurred in 59 of 70 patients, with an event rate of 91.5 events per 100 person-years ( Table 2 ). Defining the functional deterioration component alternatively as worsening WHO-FC only, yielded a composite event rate of Stratification by diagnostic groups showed a lower event rate of the composite CW end point in APAH-CHD and a higher event rate in APAHnon-CHD. Table 3 shows the event rates when only each first event of an individual is taken into account. This yielded an event rate of 55.5 CW events per 100 person-years. The event incidence analyses were repeated in the trialeligible subgroup, consisting of 49 patients who were not hospitalized, not in WHO-FC IV, and/or not immediately started on IV prostanoids at baseline. The results are presented as supplementary material (event rates in e-tables 1 and 2, cumulative event incidence curves in e-fig 1 ) and show that the event incidence of the separate components and the end-point combinations were slightly lower in this subgroup. Prognostic Value of Soft Components The occurrences of either PAH-related hospitalization, initiation of IV prostanoids, or functional deterioration were significantly associated with death or LTx, also after adjusting for diagnosis ( P,.001 for all models) ( Table 4 ). A merge of these three soft components into a soft composite end point, in which the first occurrence of one of the three components was regarded as the event of interest, was also significantly associated with death or LTx (hazard ratio, 19.1; P,.001). Analysis of alternative combinations of these soft end points, such journal.publications.chestnet.org 657

4 TABLE 1 ] Baseline Characteristics Stratified by Diagnosis Total IPAH/HPAH APAH-CHD APAH-non-CHD Characteristic No. N 5 70 No. n 5 37 No. n 5 25 No. n 5 8 P Value Age, y ( ) ( ) ( ) ( ).967 Male sex (34) (35) 25 5 (20) 8 6 (75).018 a WHO-FC a I 4 (6) 4 (11) 0 (0) 0 (0) II 25 (36) 12 (32) 10 (40) 3 (38) III 28 (40) 16 (43) 10 (40) 2 (25) IV 13 (19) 5 (14) 5 (20) 3 (38) 6MWD, m NT-proBNP, ng/l (113-1,612) (190-3,164) (93-752) (67-6,800).287 Hemodynamics b mrap, mm Hg mpap, mm Hg PVRi, WU 3 m Cardiac index, L/min/m Treatment a Upfront monotherapy 66 (94) 34 (92) 24 (96) 8 (100) Upfront dual therapy 3 (4) 2 (5) 1 (4) 0 (0) Upfront triple therapy 1 (1) 1 (3) 0 (0) 0 (0) IV prostanoids at baseline (19) 37 8 (22) 25 2 (8) 8 3 (38).108 a Hospitalized at baseline c (14) 37 6 (16) 25 1 (4) 8 3 (38).052 a Observation time, py NA Data are presented as, median (interquartile range), or mean SD, unless otherwise indicated. 6MWD 5 6-min walk distance; APAH-CHD 5 pulmonary arterial hypertension associated with congenital heart disease; APAH-non-CHD 5 pulmonary arterial hypertension associated with conditions other than congenital heart disease; HPAH 5 hereditary pulmonary arterial hypertension; IPAH 5 idiopathic pulmonary arterial hypertension; mpap 5 mean pulmonary arterial pressure; mrap 5 mean right atrial pressure; NA 5 not applicable; NT-proBNP 5 N-terminal B-type natriuretic peptide; PAH 5 pulmonary arterial hypertension; PVRI 5 indexed pulmonary vascular resistance; py 5 person-y; WHO-FC 5 World Health Organization functional class; WU 5 Wood units. a Fisher exact test. b Right-sided heart catheterization prior to treatment initiation (the median time from catheterization to treatment initiation was 1 mo). c Nonelective PAH-related hospitalization. 658 Original Research [ 148 # 3 CHEST SEPTEMBER 2015 ]

5 TABLE 2 ] Event Rate of Separate and Combined End-Point Components Stratified by Diagnosis: All Events Total (N 5 70) IPAH/HPAH (n 5 37) APAH-CHD (n 5 25) APAH-non-CHD (n 5 8) End Point (1) Death 28 (40) (38) (36) (63) 42.5 (2) Lung transplantation 7 (10) (11) (12) (0) 0.0 (3) Hospitalization a 38 (54) (57) (44) (75) 76.5 (4) Initiation of IV prostanoids 26 (37) (49) (16) (50) 34.0 (5) Functional deterioration (5A) Worsening WHO-FC b 44 (63) (73) (44) (75) 76.5 (5AB) Worsening WHO-FC b or decreased 6MWD c 50 (71) (81) (56) (75) 93.5 End-point combinations (1)(2)(3)(4) 51 (73) (78) (60) (88) (1)(2)(3)(4)(5A) 56 (80) (89) (64) (88) (1)(2)(3)(4)(5AB) (CW composite) 59 (84) (95) (68) (88) CW 5 clinical worsening. See Table 1 legend for expansion of other abbreviations. a Nonelective PAH-related hospitalization. b WHO-FC IV was always regarded as a functional deterioration event. c Decrease of 15% in 6MWD. journal.publications.chestnet.org 659

6 Figure 1 A-F, Cumulative event incidence of separate end-point components. A, Component 1, death. B, Component 2, lung transplantation. C, Component 3, nonelective pulmonary arterial hypertension (PAH)-related hospitalization. D, Component 4, initiation of IV prostanoids. E, Component 5A, functional deterioration (defined as worsening of World Health Organization functional class [WHO-FC] only). F, Component 5AB, functional deterioration (defined as worsening of WHO-FC, 15% decrease in 6-min walk distance, or both). Since the end-point components 3 (C) and 5 (E and F) could occur repetitively, the occurrence of every second, third, fourth, and fifth event is also depicted. 660 Original Research [ 148 # 3 CHEST SEPTEMBER 2015 ]

7 TABLE 3 ] Event Rate of Separate and Combined End-Point Components Stratified by Diagnosis: First Events Total (N 5 70) IPAH/HPAH (n 5 37) APAH-CHD (n 5 25) APAH-non-CHD (n 5 8) End Point (1) Death 28 (40) (38) (36) (63) 42.5 (2) Lung transplantation 7 (10) (11) (12) (0) 0.0 (3) Hospitalization a 38 (54) (57) (44) (75) 74.0 (4) Initiation of IV prostanoids 26 (37) (49) (16) (50) 38.0 (5) Functional deterioration (5A) Worsening WHO-FC b 44 (63) (73) (44) (75) 77.7 (5AB) Worsening WHO-FC b or decreased 6MWD c 50 (71) (81) (56) (75) 77.7 End-point combinations (1)(2)(3)(4) 51 (73) (78) (60) (88) (1)(2)(3)(4)(5A) 56 (80) (89) (64) (88) (1)(2)(3)(4)(5AB) (CW composite) 59 (84) (95) (68) (88) See Table 1 and 2 legend for expansion of abbreviations. a Nonelective PAH-related hospitalization. b WHO-FC IV was always regarded as a functional deterioration event. c Decrease of 15% in 6MWD. journal.publications.chestnet.org 661

8 TABLE 4 ] Association of Soft End-Point Components With Death or Lung Transplantation Time-Dependent Model 1 Univariable Analysis Time-Dependent Model 2 Adjusted for Diagnosis End Point HR (95% CI) P Value HR (95% CI) P Value Interaction Analysis a Comparison With IPAH/HPAH APAH-CHD P Value APAH-non-CHD P Value (3) Hospitalization b 9.4 ( ), ( ), (4) Initiation of IV prostanoids 6.2 ( ), ( ), (5) Functional deterioration (5A) Worsening WHO-FC c 14.3 ( ), ( ), (5AB) Worsening WHO-FC c or 6MWD d 14.4 ( ), ( ), Soft end-point combinations (3)(4) 12.3 ( ), ( ), (3)(4)(5A) 14.1 ( ), ( ), (3)(4)(5AB) 19.1 ( ), ( ), End-point components death (1) and lung transplantation (2) were used as combined analysis end point. HR 5 hazard ratio. See Table 1 legend for expansion of other abbreviations. a P values,.05 indicate significant differences in the associations across the diagnostic groups. b Nonelective PAH-related hospitalizations. c WHO-FC IV was always regarded as a functional deterioration event. d Decrease of 15% in 6MWD. as a combination without functional deterioration, yielded significant associations as well ( P,.001 for all analyzed combinations). An interaction analysis revealed that the effect size of the found associations did not significantly differ between the diagnostic groups. Similar associations with death or LTx were found in the trial-eligible subgroup ( e-table 3 ). CW Compared With Death or LTx Figure 2 shows event-free survival curves for six end-point combinations. The composite CW end point occurred early and in a higher proportion of patients, compared with the other end-point combinations. The event-free survival curves were similar in the trial-eligible subgroup (e-fig 2 ). Discussion In this study, a proposed definition of CW was evaluated as a potential end point in pediatric PAH, which consisted of the following end-point components: death, LTx, nonelective PAH-related hospitalization, initiation of IV prostanoids, and functional deterioration. The results show that CW occurs with a high event incidence and that all soft end-point components are highly predictive of death or LTx. The first CW events occurred early in the disease course, supporting the usefulness of CW as a study end point and also as an early clinical warning sign. A study end point should be a clinically meaningful outcome, defined as a direct or surrogate measure of how a patient feels, functions, or survives. 9 To date, no validated surrogates for survival are available in PAH and multiple well-established predictors of survival in adults have been shown to fail to comply with criteria for surrogacy. 13,16,46-48 CW consists of components that are all clinically meaningful outcomes in themselves. LTx, hospitalization, initiation of IV prostanoids, and functional deterioration are all undesirable events associated with major impairment in daily life (ie, how a patient functions ), thereby making CW a valid and useful composite end point. An important criticism of CW is that the definitions used in clinical trials have not been completely consistent. 18,49 Death and LTx have been included in most definitions. The same holds true for hospitalizations, although definitions may vary. For example, in a REVEAL Registry substudy 40 in which CW was evaluated in adults, all-cause hospitalizations were included, whereas, in many trials, nonelective PAH-related hospitalizations have been used. It might be debated whether the REVEAL Registry definition is best for future trial designs, as hospitalizations due to comorbidities or social reasons should not be regarded as CW events. 49 Th e need for additional PAH therapy has been a common CW component throughout the trials, but has 662 Original Research [ 148 # 3 CHEST SEPTEMBER 2015 ]

9 Figure 2 Event-free survival of six end-point combinations. Only the first occurrences of end-point components are incorporated as events. Component 1 5 death; component 2 5 lung transplantation; component 3 5 nonelective PAH-related hospitalization; component 4 5 initiation of IV prostanoids; component 5A 5 functional deterioration (defined as worsening of WHO-FC only); component 5AB 5 functional deterioration (defined as worsening of WHO-FC, 15% decrease in 6-min walk distance, or both); CW end point 5 full composite CW end point consisting of death, lung transplantation, nonelective PAH-related hospitalization, initiation of IV prostanoids and functional deterioration. CW 5 clinical worsening. See Figure 1 legend for expansion of other abbreviations. not always exclusively been defined as initiation of IV prostanoids. In this respect, it is important to realize that dose increments and addition of oral drugs might be part of currently changing, more aggressive, and goal-oriented treatment strategies, and are not necessarily induced by actual worsening of the disease. Moreover, in contrast to the initiation of continuous IV therapy, the addition of an oral therapy does not necessarily affect a patient s ability to function in daily life. Therefore, inclusion of these events may distort the validity of CW as a clinically meaningful end point. Various definitions of functional deterioration have been used, including decrease in 6MWD only, worsening of WHO-FC only, either worsening of WHO-FC or decrease in 6MWD, and decrease in 6MWD with concurrent worsening of WHO-FC. The latter definition was proposed by the (adult) task forces on end points and clinical trial design that met at the fourth and fifth World Symposium on Pulmonary Hypertension. 42,43 However, as 6MWD is not feasible in the youngest children, 7 the either/or definition seems most preferable in pediatric PAH. Last, symptomatic progression has been proposed and occasionally used as a separate CW component, 43 but we argue that this is likely to be captured in WHO-FC. Time to clinical failure has been introduced as primary end point in a clinical trial evaluating combination therapy in adult PAH. 50 An important difference with the CW end point is the inclusion of the component unsatisfactory clinical response. As evidence from adults and children suggests that there are prognostic implications associated with failure to achieve predefined treatment goals, this might be an interesting concept for future end-point definitions. 17,51,52 CW as defined and evaluated in this study seems to provide a feasible and valid alternative end point for future pediatric studies. Notwithstanding, the authors feel the following remarks should be taken into account regarding the design of future studies. To allow comparison of pediatric studies using CW as an end point, definitions of CW should be consistent among different studies. The currently evaluated definition appears suitable. A prerequisite for a consistent implementation is a broad consensus within the pediatric field regarding the exact definition of CW. 18,42 Each of the components might need to be defined as journal.publications.chestnet.org 663

10 objectively and unequivocally as possible. 19 Accuracy and consistency of reporting CW events are of utmost importance. Therefore, to guarantee the integrity of studies using CW as an end point, the use of blinded, independent adjudication committees is recommended. 19,43,49 For example, it should be ensured that hospitalizations and functional deteriorations are indeed caused by worsening PAH and not by comorbidities or other causes. Patients with unstable disease usually cannot be included in clinical trials, which might lead to discrepancies in included patients and event rates between registries and trials. This is illustrated by a comparison of the REVEAL Registry and the event-driven Study With an Endothelin Receptor Antagonist in Pulmonary Arterial Hypertension to Improve Clinical Outcome (SERAPHIN) trial that used CW as a primary end point. 37,40 In REVEAL Registry, 64% of the patients experienced an event within 2 years, whereas in the SERAPHIN trial, only 39% had an event over a median period of 115 weeks. However, the results from the current study show only minor differences between the full cohort and the trialeligible subgroup: 84% of the full cohort experienced CW ( Table 2 ), compared with 78% of the trial-eligible subgroup ( e-table 1 ). This confirms that PAH can progress more quickly in children compared with adults, 8 and suggests that the use of CW allows for pediatric study designs with an achievable number of children. Study Limitations Evaluation of alternative end points is important though hampered by the extremely limited data on pediatric PAH. The Dutch National Network for Pediatric Pulmonary Hypertension is a national prospective registry with complete follow-up that encompasses all diagnosed children with PAH in The Netherlands. 53 Despite the relatively small number of patients in our study, limiting our analyses, we found strong associations between the soft end-point components and death or LTx. The highly standardized and complete longitudinal follow-up allowed for time-dependent Cox regression, which is an appropriate method for evaluating the prognostic value of time-varying variables such as the CW end-point components.54 Inherent to the analytic approach, censoring of deceased patients was taken into account rather than the competing risk of death. Although all subsequent CW events were captured in the cumulative event-incidence curves and longitudinal event rates, only the soft CW events that occurred first could be analyzed in time-dependent Cox regression analysis. The sample size and analytic approach did not allow for an identification of which CW component carried the most prognostic value. Conclusions CW occurs early and frequently in the follow-up of children with PAH. Each of the soft end-point components was highly predictive of death or LTx, as was the CW composite. This strongly supports the usefulness of CW as a patient-centered, composite study end point and allows for pediatric study designs with an achievable number of required children. Regarding its prognostic value, CW could also serve as an early clinical warning sign. 664 Original Research [ 148 # 3 CHEST SEPTEMBER 2015 ]

11 Acknowledgments Author contributions: M.-J. P. and S. A. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. M.-J. P., S. A., W. M. H. Z., J. M. D., T. R. V.-K., M. T. R. R., H. L. H., and R. M. F. B. contributed substantially to the study design, data analysis and interpretation, and the writing of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Berger reports that the University Medical Center Groningen received fees for his advisory board activities, outside the submitted work, from Actelion Pharmaceuticals Ltd, Pfizer Inc, GlaxoSmithKline plc, Eli Lilly and Co, Novartis AG, and Bayer AG. Drs Ploegstra, Douwes, Roofthooft, and Hillege and Mss Arjaans, Zijlstra, and Vissia-Kazemier have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript. Additional information: The e-figures and e-tables can be found in the Supplemental Materials section of the online article. References 1. Schermuly RT, Ghofrani HA, Wilkins MR, Grimminger F. Mechanisms of disease: pulmonary arterial hypertension. Nat Rev Cardiol ;8(8): van Loon RL, Roofthooft MT, Hillege HL, et al. Pediatric pulmonary hypertension in the Netherlands: epidemiology and characterization during the period 1991 to Circulation ;124(16): Zijlstra WM, Douwes JM, Rosenzweig EB, et al. Survival differences in pediatric pulmonary arterial hypertension: clues to a better understanding of outcome and optimal treatment strategies. J Am Coll Cardiol ;63(20): Barst RJ, Ivy DD, Gaitan G, et al. A randomized, double-blind, placebocontrolled, dose-ranging study of oral sildenafil citrate in treatment-naive children with pulmonary arterial hypertension. Circulation ;125(2): Barst RJ, Beghetti M, Pulido T, et al ; STARTS-2 Investigators. STARTS-2: long-term survival with oral sildenafil monotherapy in treatment-naive pediatric pulmonary arterial hypertension. Circulation ; 129 ( 19 ): Berger RM. Pulmonary hypertension: smaller kids, smaller steps. Lancet Respir Med ;2(5): Adatia I, Haworth SG, Wegner M, et al. Clinical trials in neonates and children: report of the pulmonary hypertension academic research consortium pediatric advisory committee. Pulm Circ ;3(1): Haworth SG, Beghetti M. Assessment of endpoints in the pediatric population: congenital heart disease and idiopathic pulmonary arterial hypertension. Curr Opin Pulm Med ;16(suppl 1 ): S35-S Temple R. Clinical measurement in drug evaluation. In: Nimmo W, Tucker G, eds. A Regulatory Authority s Opinion About Surrogate Endpoints. New York, NY : John Wiley ; 1995 : Fleming TR, DeMets DL. Surrogate end points in clinical trials: are we being misled? Ann Intern Med ;125(7): Fleming TR, Powers JH. Biomarkers and surrogate endpoints in clinical trials. Stat Med ;31(25): Gaine S, Simonneau G. The need to move from 6-minute walk distance to outcome trials in pulmonary arterial hypertension. Eur Respir Rev ;22(130): Gabler NB, French B, Strom BL, et al. Validation of 6-minute walk distance as a surrogate end point in pulmonary arterial hypertension trials. Circulation ;126(3): Benza RL, Miller DP, Gomberg-Maitland M, et al. Predicting survival in pulmonary arterial hypertension: insights from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL). Circulation ;122(2): Douwes M, Hegema AK, Van Der Krieke M, et al. Six-minute walk-test in childhood pulmonary arterial hypertension: walking distance and decrease in oxygen saturation provide additional prognostic information [abstract]. Eur Heart J ;35(suppl 1 ): Savarese G, Paolillo S, Costanzo P, et al. Do changes of 6-minute walk distance predict clinical events in patients with pulmonary arterial hypertension? A meta-analysis of 22 randomized trials. J Am Coll Cardiol ; 60 ( 13 ): Ploegstra MJ, Douwes JM, Roofthooft MT, Zijlstra WM, Hillege HL, Berger RM. Identification of treatment goals in paediatric pulmonary arterial hypertension. Eur Respir J ;44(6): Peacock A, Keogh A, Humbert M. Endpoints in pulmonary arterial hypertension: the role of clinical worsening. Curr Opin Pulm Med ;16(suppl 1 ): S1-S Galiè N, Simonneau G, Barst RJ, Badesch D, Rubin L. Clinical worsening in trials of pulmonary arterial hypertension: results and implications. Curr Opin Pulm Med ;16(suppl 1 ):S11-S Hoeper MM, Oudiz RJ, Peacock A, et al. End points and clinical trial designs in pulmonary arterial hypertension: clinical and regulatory perspectives. J Am Coll Cardiol ;43(12)(suppl S ):48S-55S. 21. Kawut SM, Palevsky HI. Surrogate end points for pulmonary arterial hypertension. Am Heart J ;148(4): Rubin LJ, Badesch DB, Barst RJ, et al. Bosentan therapy for pulmonary arterial hypertension. N Engl J Med ; 346 (12): Barst RJ, Langleben D, Frost A, et al ; STRIDE-1 Study Group. Sitaxsentan therapy for pulmonary arterial hypertension. Am J Respir Crit Care Med ;169(4): Galiè N, Ghofrani HA, Torbicki A, et al ; Sildenafil Use in Pulmonary Arterial Hypertension (SUPER) Study Group. Sildenafil citrate therapy for pulmonary arterial hypertension. N Engl J Med ;353(20): Barst RJ, Langleben D, Badesch D, et al ; STRIDE-2 Study Group. Treatment of pulmonary arterial hypertension with the selective endothelin-a receptor antagonist sitaxsentan. J Am Coll Cardiol ; 47 (10): McLaughlin VV, Oudiz RJ, Frost A, et al. Randomized study of adding inhaled iloprost to existing bosentan in pulmonary arterial hypertension. Am J Respir Crit Care Med ;174(11): Galiè N, Rubin LJ, Hoeper M, et al. Treatment of patients with mildly symptomatic pulmonary arterial hypertension with bosentan (EARLY study): a doubleblind, randomised controlled trial. Lancet ;371(9630): Simonneau G, Rubin LJ, Galiè N, et al ; PACES Study Group. Addition of sildenafil to long-term intravenous epoprostenol therapy in patients with pulmonary arterial hypertension: a randomized trial. Ann Intern Med ; 149 ( 8 ): Galiè N, Brundage BH, Ghofrani HA, et al ; Pulmonary Arterial Hypertension and Response to Tadalafil (PHIRST) Study Group. Tadalafil therapy for pulmonary arterial hypertension. Circulation ;119(22): Galiè N, Olschewski H, Oudiz RJ, et al ; Ambrisentan in Pulmonary Arterial Hypertension, Randomized, Double- Blind, Placebo-Controlled, Multicenter, Efficacy Studies (ARIES) Group. Ambrisentan for the treatment of pulmonary arterial hypertension: results of the ambrisentan in pulmonary arterial hypertension, randomized, double-blind, placebo-controlled, multicenter, efficacy (ARIES) study 1 and 2. Circulation ;117(23): Benza RL, Seeger W, McLaughlin VV, et al. Long-term effects of inhaled treprostinil in patients with pulmonary arterial hypertension: the Treprostinil Sodium Inhalation Used in the Management of Pulmonary Arterial Hypertension (TRIUMPH) study openlabel extension. J Heart Lung Transplant ;30(12): Sandoval J, Torbicki A, Souza R, et al ; STRIDE-4 investigators. Safety and efficacy of sitaxsentan 50 and 100 mg in patients with pulmonary arterial hypertension. Pulm Pharmacol Ther ;25(1): Oudiz RJ, Brundage BH, Galiè N, et al ; PHIRST Study Group. Tadalafil for the treatment of pulmonary arterial hypertension: a double-blind 52-week journal.publications.chestnet.org 665

12 uncontrolled extension study. J Am Coll Cardiol ;60 (8 ): Tapson VF, Torres F, Kermeen F, et al. Oral treprostinil for the treatment of pulmonary arterial hypertension in patients on background endothelin receptor antagonist and/or phosphodiesterase type 5 inhibitor therapy (the FREEDOM-C study): a randomized controlled trial. Chest ;142(6): Hoeper MM, Barst RJ, Bourge RC, et al. Imatinib mesylate as add-on therapy for pulmonary arterial hypertension: results of the randomized IMPRES study. Circulation ;127(10): Ghofrani HA, Galiè N, Grimminger F, et al ; PATENT-1 Study Group. Riociguat for the treatment of pulmonary arterial hypertension. N Engl J Med ;369(4): Pulido T, Adzerikho I, Channick RN, et al ; SERAPHIN Investigators. Macitentan and morbidity and mortality in pulmonary arterial hypertension. N Engl J Med ;369(9): Tapson VF, Jing ZC, Xu KF, et al ; FREEDOM-C2 Study Team. Oral treprostinil for the treatment of pulmonary arterial hypertension in patients receiving background endothelin receptor antagonist and phosphodiesterase type 5 inhibitor therapy (the FREEDOM-C2 study): a randomized controlled trial. Chest ;144(3): Jing ZC, Parikh K, Pulido T, et al. Efficacy and safety of oral treprostinil monotherapy for the treatment of pulmonary arterial hypertension: a randomized, controlled trial. Circulation ;127(5): Frost AE, Badesch DB, Miller DP, Benza RL, Meltzer LA, McGoon MD. 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(5): van Loon RL, Roofthooft MT, Delhaas T, et al. Outcome of pediatric patients with pulmonary arterial hypertension in the era of new medical therapies. Am J Cardiol ;106(1): McLaughlin VV, Badesch DB, Delcroix M, et al. End points and clinical trial design in pulmonary arterial hypertension. J Am Coll Cardiol ; 54 ( suppl 1 ): S97-S Gomberg-Maitland M, Bull TM, Saggar R, et al. New trial designs and potential therapies for pulmonary artery hypertension. J Am Coll Cardiol ;62(suppl 25 ): D82-D Rich S, Dantzker DR, Ayres SM, et al. Primary pulmonary hypertension. A national prospective study. Ann Intern Med ;107(2): Humbert M, Sitbon O, Chaouat A, et al. Pulmonary arterial hypertension in France: results from a national registry. Am J Respir Crit Care Med ;173(9): Savarese G, Musella F, D Amore C, et al. Haemodynamics, exercise capacity and clinical events in pulmonary arterial hypertension. Eur Respir J ;42(2): Ventetuolo CE, Gabler NB, Fritz JS, et al. Are hemodynamics surrogate end points in pulmonary arterial hypertension? Circulation ;130(9): Freedman LS, Graubard BI, Schatzkin A. Statistical validation of intermediate endpoints for chronic diseases. Stat Med ;11(2): McGlinchey N, Peacock AJ. Endpoints in PAH clinical trials in the era of combination therapy: how do we decide whether something is working without going bankrupt? Drug Discov Today ;19(8): Galie N. The AMBITION study: design and result s [abstract]. Eur Respir J ;44(suppl 58 ):A Nickel N, Golpon H, Greer M, et al. The prognostic impact of follow-up assessments in patients with idiopathic pulmonary arterial hypertension. Eur Respir J ;39(3): Barst RJ, Chung L, Zamanian RT, Turner M, McGoon MD. Functional class improvement and 3-year survival outcomes in patients with pulmonary arterial hypertension in the REVEAL Registry. Chest ;144(1): van Loon RL, Roofthooft MT, van Osch- Gevers M et al. Clinical characterization of pediatric pulmonary hypertension: complex presentation and diagnosis. J Pediatr ;155(2): Beyersmann J, Schumacher M. Timedependent covariates in the proportional subdistribution hazards model for competing risks. Biostatistics ;9(4): Original Research [ 148 # 3 CHEST SEPTEMBER 2015 ]