Abstract
Background
Recently studied therapies for pulmonary arterial hypertension (PAH) have improved outcomes among populations of patients, but little is known about which patients are most likely to respond to specific treatments. Differences in endothelin-1 biology between sexes and between whites and blacks may lead to differences in patients' responses to treatment with endothelin receptor antagonists (ERAs).
Methods
We conducted pooled analyses of deidentified, patient-level data from six randomized placebo-controlled trials of ERAs submitted to the US Food and Drug Administration to elucidate heterogeneity in treatment response. We estimated the interaction between treatment assignment (ERA vs placebo) and sex and between treatment and white or black race in terms of the change in 6-min walk distance from baseline to 12 weeks.
Results
Trials included 1,130 participants with a mean age of 49 years; 21% were men, 74% were white, and 6% were black. The placebo-adjusted response to ERAs was 29.7 m (95% CI, 3.7-55.7 m) greater in women than in men (P = .03). The placebo-adjusted response was 42.2 m for whites and −1.4 m for blacks, a difference of 43.6 m (95% CI, −3.5-90.7 m) (P = .07). Similar results were found in sensitivity analyses and in secondary analyses using the outcome of absolute distance walked.
Conclusions
Women with PAH obtain greater responses to ERAs than do men, and whites may experience a greater treatment benefit than do blacks. This heterogeneity in treatment-response may reflect pathophysiologic differences between sexes and races or distinct disease phenotypes.
Abbreviations
- 6MWD
6-min walk distance
- ERA
endothelin receptor antagonist
- ET
endothelin
- FDA
US Food and Drug Administration
- PAH
pulmonary arterial hypertension
- RCT
randomized controlled trial
- RV
right ventricular
Pulmonary arterial hypertension (PAH) leads to decreased functional status, right-sided heart failure, and death. 1 Recently studied therapies have improved outcomes, yet substantial heterogeneity remains in the risk of adverse clinical outcomes and the magnitude of the treatment response.2, 3, 4
Plasma levels of endothelin (ET)-1 are elevated among patients with PAH, contributing to pulmonary vascular remodeling.5, 6, 7 Endothelin receptor antagonists (ERAs) were the first oral therapies approved by the US Food and Drug Administration (FDA) for use in PAH and remain among the most commonly used drugs to treat it. Despite their effectiveness, experience has shown that certain individuals may have a dramatic clinical response to ERAs, whereas others may have no response or may worsen.
There are known biologic differences between sexes and among different races in the production and handling of ET-1, a potent systemic and pulmonary vasoconstrictor. Men have higher concentrations of circulating ET-1 than do women,8, 9 and men exhibit greater ET-mediated coronary vasoconstriction. 10 Blacks with systemic hypertension are known to have higher plasma levels of ET-1 than do whites and a greater increase in ET-1 in response to stress. 11 Further, blacks with systemic hypertension have a greater reduction in BP with ERAs compared with that of whites. 12 The low enrollment of men and blacks and the relatively small sizes of phase 2 and 3 trials of ERAs in PAH have until now precluded study-wide tests of heterogeneity in treatment response. We, therefore, sought to assess whether sex and race modified the effects of ERA therapy in placebo-controlled trials of ERAs.
Materials and Methods
Study Population
We obtained data on all participants included in the seven randomized placebo-controlled trials of ERAs submitted to the FDA. These seven clinical trials (Ambrisentan in Pulmonary Arterial Hypertension, Randomized, Double-Blind, Placebo-Controlled, Multicenter, Efficacy Studies [ARIES]-1; ARIES-2; Bosentan: Randomized Trial of Endothelin Receptor Antagonist Therapy [BREATHE]-1; BREATHE-2; Sitaxsentan To Relieve Impaired Exercise [STRIDE]-1; STRIDE-2; and STRIDE-4) examined three agents (ambrisentan, bosentan, and sitaxsentan). We subsequently excluded BREATHE-2 because it included only 33 patients, did not collect data on several potential confounders, and was not a phase 3 trial. Details of the included trials are provided elsewhere.13, 14, 15, 16, 17
Deidentified individual patient data for all participants were provided by the FDA. The included trials reported similar inclusion criteria, variable collection, and outcome assessment at 12-week follow-up.
Exposure
Our primary exposure variable was treatment assignment (ERA or placebo). Across trials, different agents and doses yielded similar improvements in 6-min walk distance (6MWD) relative to placebo and were, therefore, analyzed together. Covariates included age, sex, PAH diagnosis (idiopathic, connective tissue disease, HIV infection/anorexigen use, or congenital heart disease), height, weight, baseline laboratory and hemodynamic measures, and concurrent medication use.
Outcomes
The primary outcome of interest was the change in 6MWD from baseline to 12 weeks; 6MWD was the primary end point in all but one of the trials and was selected for this reason. Baseline 6MWD was obtained at the time of, or within 2 weeks prior to, randomization.
In the primary analyses, patients who were missing a 12-week 6MWD because of study-ending clinical events (n = 43) were assigned a value of 0 m for their 12-week 6MWD assessment. This choice was made a priori based on the assumption that patients who were too sick to complete the study would also be too sick to complete a 6MWD assessment. We also performed sensitivity analyses using 50, 100, 150, or 275 m (the mean 12-week 6MWD among patients who sustained a clinical event but still completed a 12-week 6MWD assessment) as the 12-week 6MWD.
We conducted four secondary analyses. First, we explored differences between baseline and the final recorded 6MWD, accounting for the fact that some trials lasted longer than 12 weeks. These analyses produced results similar to those of the primary analyses and are not reported. Second, we examined differences in the absolute distance walked at 12 weeks, which some have considered a better predictor of clinical outcomes. 18 Third, we repeated our primary analysis in the subset of patients with idiopathic PAH. Finally, we compared the proportions of patients in each treatment-by-sex and treatment-by-race stratum who sustained clinical events (death, lung transplant, atrial septostomy, hospitalization due to worsening PAH, withdrawal for worsening right-sided heart failure, or addition of prostacyclin analogs or phosphodiesterase inhibitors for PAH). We did not consider deterioration in 6MWD in the clinical event definition because this represented our primary outcome variable.
Statistical Analysis
Participant characteristics were summarized using medians and interquartile ranges for continuous variables and percentages for categorical variables. Primary and secondary outcomes were analyzed using linear or logistic regression for continuous or binary outcomes, respectively. We restricted the sample to blacks and whites when analyzing the interaction of treatment assignment with race. Separate models including all patients were used to estimate all other effects.
For the primary outcome, our most basic model included treatment (drug or placebo), sex, race, and the treatment-by-sex or treatment-by-race interaction terms. This model also adjusted for PAH diagnosis (idiopathic, connective tissue disease, HIV/anorexigen use, or congenital heart disease, entered as indicator variables), baseline 6MWD, patient height, and study. By including sex and height in the model, we effectively modeled 6MWD % predicted without violating statistical assumptions regarding ratio values. 19 In addition, a partially adjusted model additionally included diagnostic category, age, race, and sex. All other variables were examined as potential confounders in a third model that included any variables that altered the coefficient for the interaction term by ≥ 15%. A similar approach was used to model absolute distance walked without adjustment for baseline walk distance. Achievement of a certain threshold absolute walk distance, if prognostically significant, should be independent of baseline walk distance.
Multiple imputation was used to impute 12-week 6MWD for study participants who did not have a study-ending clinical event and were missing a 12-week 6MWD value (n = 45). 20 The imputation model included treatment, sex, and race, in addition to variables associated with missing 12-week 6MWD: study, baseline walk distance, and diagnosis category. Twenty imputed data sets were generated to estimate the variability among imputed values. All analyses were completed in SAS, version 9.2 (SAS Institute).
The study was approved by the institutional review board of the University of Pennsylvania (Review Board No. 4, approval 811814). All coauthors had access to the study data, take responsibility for the analysis, and had authority over manuscript preparation and the decision to submit for publication.
Results
The six trials included 1,130 patients. The participants' mean age was 49 years (range, 12–82), 21% were men, 836 (74%) were white, and 65 (6%) were black (Table 1). Seven hundred seventy-three patients (68%) were randomized to an ERA and 357 (32%) to placebo. Forty-three patients (4%) experienced study-ending clinical events; an additional 10 patients (1%) sustained clinical events but completed their 12-week 6MWD assessment. Mean baseline walk distance was 349 m (SD 87.0 m). Similar hemodynamic, laboratory, and severity of illness values were observed for men and women, and for whites and blacks (e-Tables 1, 2).
Table 1.
Characteristics of Study Participants
| Characteristic | Active Treatment (n = 773) | Placebo (n = 357) |
|---|---|---|
| Age, y | 50 (38–61) | 49 (36–60) |
| Male sex, No. (%) | 158 (20) | 78 (22) |
| Race, No. (%) | ||
| Black | 45 (6) | 20 (6) |
| White | 577 (75) | 259 (73) |
| Othera | 151 (20) | 78 (22) |
| Height, cm | 162 (157–168) | 163 (157–168) |
| Weight, kg | 69 (59–82) | 70 (59–85) |
| BMI, kg/m2 | 25.6 (22.7–30.2) | 26.2 (22.9–30.6) |
| PAH diagnosis, No. (%) | ||
| Idiopathica | 478 (62) | 229 (64) |
| Connective tissue disease | 216 (28) | 86 (24) |
| HIV infection/anorexigen use | 23 (3) | 11 (3) |
| Congenital heart disease | 54 (7) | 30 (8) |
| New York Heart Association functional class, No. (%) | ||
| I-IIa | 256 (33) | 114 (32) |
| III-IV | 516 (67) | 243 (68) |
| Baseline hemodynamics | ||
| Mean right atrial pressure, mm Hg | 7.5 (5.0–11.5) | 7.5 (4.0–12.0) |
| Mean pulmonary arterial pressure, mm Hg | 50 (40–60) | 51 (42–62) |
| Cardiac output, L/min | 4.2 (3.3–5.2) | 4.2 (3.4–5.3) |
| Cardiac index, L/min/m2 | 2.3 (1.9–2.9) | 2.3 (1.9–2.8) |
| Pulmonary capillary wedge pressure, mm Hg | 9 (7–12) | 10 (7–12) |
| Pulmonary vascular resistance, Wood units | 10.2 (6.7–14.4) | 10.0 (7.0–15.0) |
| Baseline laboratory values | ||
| Albumin, g/dL | 4.2 (3.9–4.5) | 4.2 (3.9–4.5) |
| Creatinine, mg/dL | 0.80 (0.71–0.97) | 0.80 (0.68–0.97) |
| Hemoglobin, g/dL | 14.8 (13.4–16.0) | 14.5 (13.0–15.8) |
| Sodium, mEq/L | 140 (138–142) | 140 (138–142) |
| Warfarin use, No. (%) | 466 (60) | 230 (64) |
| Baseline 6MWD, m | 360 (289–414) | 366 (288–416) |
| Study, No. (%) | ||
| ARIES-1 | 134 (17) | 67 (19) |
| ARIES-2 | 127 (16) | 65 (18) |
| BREATHE-1 | 145 (19) | 69 (19) |
| STRIDE-1 | 118 (15) | 60 (17) |
| STRIDE-2 | 185 (24) | 62 (17) |
| STRIDE-4a | 64 (8) | 34 (10) |
Data are presented as median (interquartile range) unless indicated otherwise. 6MWD = 6-min walk distance; ARIES = Ambrisentan in Pulmonary Arterial Hypertension, Randomized, Double-Blind, Placebo-Controlled, Multicenter, Efficacy Studies; BREATHE = Bosentan: Randomized Trial of Endothelin receptor Antagonist Therapy; PAH = pulmonary arterial hypertension; STRIDE = Sitaxsentan To Relieve Impaired Exercise.
Reference category for analysis.
Treatment Response by Sex
The placebo-adjusted treatment response (mean improvement in those who received ERAs vs placebo) was 44.1 m among women and 16.7 m among men in the base model (Table 2). Thus, treatment response was 27.4 m (95% CI, 1.3-53.5 m) greater in women than in men. Women allocated to placebo showed a decrement in 6MWD that was prevented by assignment to ERAs. Men assigned to placebo did not show a decrement in 6MWD, but also did not show the benefit observed in women with assignment to ERAs (e-Fig 1). As shown in the partially adjusted model in Table 2, adjustment for age, race, and diagnosis category did not change these results. Further adjustment for potential confounding variables, such as laboratory and hemodynamic values, did not change the results (data not shown).
Table 2.
Change in 6MWD From Baseline to 12-wk Follow-up by Sex
| Model | Women (n = 894) | Men (n = 236) | Difference-in-Difference (95% CI) | P Value for Interaction |
|---|---|---|---|---|
| Base modela | ||||
| Treatment | 33.4 | 18.8 | … | … |
| Placebo | −10.7 | 2.1 | … | … |
| Δ (treatment-placebo) | 44.1 | 16.7 | 27.4 (1.3–53.5) | .04 |
| Partially adjusted modelb | ||||
| Treatment | 35.6 | 20.3 | … | … |
| Placebo | −9.7 | 4.7 | … | … |
| Δ (treatment-placebo) | 45.3 | 15.6 | 29.7 (3.7–55.7) | .03 |
See Table 1 legend for expansion of abbreviation.
Adjusted for study, height, and baseline walk distance.
Adjusted for study, height, baseline walk distance, age, race, and diagnosis category; additional regression output provided in e-Table 9.
Women also demonstrated greater absolute walk distances at 12 weeks, with a placebo-adjusted 6MWD that was 27.5 m (95% CI, −8.3-63.3 m) greater than that for men (Table 3). After adjustment for confounders, the placebo-adjusted improvement for women was 44.9 m (95% CI, 9.2-80.6 m) greater than that for men.
Table 3.
Absolute Distance Walked at 12-wk Follow-up by Sex
| Model | Women (n = 894) | Men (n = 236) | Difference-in-Difference (95% CI) | P Value for Interaction |
|---|---|---|---|---|
| Base modela | ||||
| Treatment | 381.6 | 378.0 | … | … |
| Placebo | 337.6 | 361.5 | … | … |
| Δ (treatment-placebo) | 44.0 | 16.5 | 27.5 (−8.3–63.3) | .13 |
| Partially adjusted modelb | ||||
| Treatment | 373.1 | 370.7 | … | … |
| Placebo | 326.0 | 357.5 | … | … |
| Δ (treatment-placebo) | 47.1 | 13.2 | 33.9 (−1.0–68.6) | .06 |
| Fully adjusted modelc | ||||
| Treatment | 369.4 | 362.5 | … | … |
| Placebo | 319.5 | 357.5 | … | … |
| Δ (treatment-placebo) | 49.9 | 5.0 | 44.9 (9.2–80.6) | .01 |
Adjusted for study and height.
Adjusted for study, height, age, race, and diagnosis category.
Adjusted for study, height, age, race, diagnosis category, cardiac index, and pulmonary capillary wedge pressure; additional regression output provided in e-Table 10.
Results were similar in sensitivity analyses in which higher values were used for the 12-week 6MWD among patients experiencing interval clinical events (e-Tables 3, 4), but the magnitudes of effect sizes diminished when larger values were entered for these 12-week walk distances. Likewise, analysis of only patients with a diagnosis of idiopathic PAH yielded similar overall results (e-Table 5), making it less likely that unmeasured differences in disease subtypes between women and men explained the results. Treatment with ERAs (vs placebo) did not influence the risk of clinical events among men after adjusting for study and baseline walk distance (OR = 1.28; 95% CI, 0.22-7.49), but significantly reduced the risk of clinical events among women (OR = 0.27; 95% CI, 0.14-0.55) (P value for interaction = .11).
Treatment Response by Race
The placebo-adjusted treatment response was 41.5 m among whites and −3.5 m among blacks in the base model (Table 4), a treatment response difference of 45.0 m (95% CI, −2.4-92.4 m) (e-Fig 2). Results were similar in the models adjusting for diagnosis category, age, and sex. Further adjustment for potential confounding variables did not change these results.
Table 4.
Change in 6MWD From Baseline to 12-wk Follow-up by Race
| Model | White (n = 836) | Black (n = 65) | Difference-in-Difference (95% CI) | P Value for Interaction |
|---|---|---|---|---|
| Base modela | ||||
| Treatment | 31.7 | 14.4 | … | … |
| Placebo | −9.8 | 17.9 | … | … |
| Δ (treatment-placebo) | 41.5 | −3.5 | 45.0 (−2.4–92.4) | .06 |
| Partially adjusted modelb | ||||
| Treatment | 32.5 | 12.0 | … | … |
| Placebo | −9.7 | 13.4 | … | … |
| Δ (treatment-placebo) | 42.2 | −1.4 | 43.6 (−3.5–90.7) | .07 |
See Table 1 legend for expansion of abbreviations.
Adjusted for study, height, and baseline walk distance.
Adjusted for study, height, baseline walk distance, age, sex, and diagnosis category; additional regression output provided in e-Table 11.
Placebo-adjusted absolute walk distances at 12 weeks were 41.6 m for whites and 11.1 m for blacks, a difference of 30.5 m (95% CI, −35.8-96.8 m) (Table 5). Similar results were found in partially adjusted models including age, sex, and diagnostic category, but an attenuation of effects was observed when including New York Heart Association class in fully adjusted models. Results were similar in sensitivity analyses (e-Tables 6, 7) and in analyses restricted to patients with idiopathic PAH (e-Table 8). There were no clinical events among blacks in the placebo groups precluding calculation of a race-by-treatment interaction.
Table 5.
Absolute Distance Walked at 12-wk Follow-up by Race
| Model | White (n = 836) | Black (n = 65) | Difference-in-Difference (95% CI) | P Value for Interaction |
|---|---|---|---|---|
| Base modela | ||||
| Treatment | 384.1 | 368.0 | … | … |
| Placebo | 342.5 | 356.9 | … | … |
| Δ (treatment-placebo) | 41.6 | 11.1 | 30.5 (−35.8–96.8) | .37 |
| Partially adjusted modelb | ||||
| Treatment | 387.0 | 360.0 | … | … |
| Placebo | 344.2 | 344.9 | … | … |
| Δ (treatment-placebo) | 42.8 | 15.1 | 27.7 (−36.0–91.4) | .39 |
| Fully adjusted modelc | ||||
| Treatment | 398.7 | 379.5 | … | … |
| Placebo | 357.1 | 351.9 | … | … |
| Δ (treatment-placebo) | 41.6 | 27.6 | 14.0 (−47.8–75.8) | .66 |
Adjusted for study and height.
Adjusted for study, height, age, sex, and diagnosis category.
Adjusted for study, height, age, sex, diagnosis category, and New York Heart Association class; additional regression output provided in e-Table 12.
Discussion
This study suggests that sex and race may influence the response to ERA treatment of PAH. Specifically, we found that women had significantly greater responses in terms of change in 6MWD than did men. The differences between races did not meet the criteria for significance at the 0.05 level, although they had large effect estimates and CIs that just included the null value. These findings were consistent across different parameterizations of 6MWD, and were unlikely to be due to difference in illness severity because of comparable baseline hemodynamic and laboratory measures. We also found an indication of greater reductions in clinical events with ERAs among women and whites, but these differential effects were either not statistically significant or were unable to be quantified because of the low numbers of events.
Interestingly, women and whites experienced deteriorations in 6MWD when assigned to placebo, whereas men and blacks did not and even showed some improvement on placebo. Women experienced greater benefits from ERAs compared with men, and whites may have benefitted more than blacks. Based on these observations, future studies are needed to explore the mechanisms accounting for why women and whites may show a greater response to ERA treatment. For example, such differences might be explained, in part, by known differences in estrogen concentrations or signaling21, 22 or in right ventricular (RV) contractility 23 between men and women. Women with idiopathic PAH have a higher RV ejection fraction at baseline, independent of a variety of confounders, including pulmonary vascular resistance. 23 Indeed, studies have shown improved survival among women with PAH in long-term treated cohorts. 24 It is possible that RV adaptation could allow women to have a greater or more rapid response to PAH treatment, while having a more severe trajectory without targeted therapy. Alternatively, off-target (ie, mechanistically unpredicted) drug effects could account for some of the observed variation between sexes. Finally, it is possible that these observed differences are due to variations in disease expression on 6MWD between men and women, and whites and blacks.
Blacks have increased levels of circulating ET-111, 25 and greater increases in ET-1 in response to stress. 12 It is possible that these higher ET-1 levels are not sufficiently inhibited by currently available ERAs. A possibly reduced treatment response in blacks is consistent with literature reporting higher mortality among blacks with idiopathic PAH, 26 and with studies suggesting that blacks with cardiovascular and pulmonary illnesses often have worse outcomes than do whites.27, 28, 29, 30
The identification of treatment-response heterogeneity can improve health care. First, pinpointing patient characteristics that modify baseline risks for adverse events and/or favorable therapeutic response can inform individual treatment decisions.2, 31 Second, isolating sources of treatment-response heterogeneity can enhance knowledge of the biologic mechanisms of disease pathogenesis in PAH. Third, recognizing differences in subpopulations with PAH could maximize the efficiency of future randomized controlled trials (RCTs) by providing metrics upon which to stratify patients at enrollment. Therefore, extrapolating the summary result of an RCT to all patients may be inappropriate because the risks of adverse events may exceed the probabilities of clinical benefits for certain patients.2, 32, 33 Quantifying a treatment's risk-benefit ratio for an individual with a given baseline risk may promote safer, more rational, and cost-effective drug use. 34
Several limitations of this analysis were also limitations of the trials in our study. These include the use of 6MWD as the primary end point, the inclusion of mostly women and whites, and the short exposure times to active drug and placebo. Whether 6MWD should be viewed as an intermediate end point of clinical import (as it is by the FDA) or as a surrogate requires further research. 35 Acceptable type 1 error rates for interaction terms are controversial and a full picture of effect modification involves more than just the interaction P value. 36 Our analyses showed large point estimates of the interactions between treatment assignment and both sex and race; however, the results for race were not statistically significant at the P < .05 level. The identification of “white” and “black” was not uniform in the RCTs included in our study, and such misclassification would tend to bias results toward the null. We were unable to control for site differences within each study because of the low number of participants at each site. Finally, although our study was limited to the trials submitted to the FDA for drug approval, these are the precise studies upon which ERA treatment recommendations are made by the FDA.
Conclusions
In conclusion, our results indicate that women with PAH derive greater treatment responses to ERAs than do men. Whites with PAH may also have better responses than do blacks. Further studies are needed to determine whether these differences are attributable to heterogeneous responses to ERAs specifically or to more generalized differences in baseline risks for clinical deterioration among groups defined by race and sex.
Supplementary Material
Acknowledgments
Author contributions: Dr Gabler: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr French: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr Strom: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr Liu: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr Palevsky: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr Taichman: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr Kawut: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Dr Halpern: contributed to data analysis, manuscript preparation, and decision to submit for publication.
Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Kawut has received consulting fees, advisory board fees, speaking fees, unrestricted educational grants, and/or research funding from Pfizer, Actelion, Bayer, Novartis, Merck, Gilead, United Therapeutics, and Lung Rx. Dr Gabler has participated in research projects funded by Pfizer. Dr Strom has served in both research and consultant roles for Pfizer and has received gifts from Pfizer for his pharmacoepidemiology training program. Dr Taichman has received research grant support to the University of Pennsylvania from Actelion, Inc, for participation in the REVEAL Registry. Dr Palevsky has served as a speaker and/or consultant to Actelion Pharmaceuticals, Gilead, and Pfizer, the companies with the rights to the three endothelin receptor antagonists analyzed in this study. These roles in no way impacted on this analysis of the results of previously published studies. Dr Halpern has received research grants related to this work from the American Thoracic Society, Pfizer, and Actelion. Drs French and Liu 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: Neither the FDA nor the funding source had a role in the design of this study nor in the decision to submit it for publication. The FDA did review the study prior to submission as a condition of the original contract, but did not request any changes. Drs Norman Stockbridge and Salma Lemtouni of the FDA have provided written permission for us to acknowledge them as we have.
Other contributions: We are grateful to Maximilian Herlim for invaluable help in preparing the data for analysis and to Norman Stockbridge, MD, PhD, and Salma Lemtouni, MD, MPH, at the US FDA for providing us with the data to conduct this study.
Additional information: The e-Tables and e-Figures can be found in the Online Supplement at http://chestjournal.chestpubs.org/content/141/1/20/suppl/DC1.
Footnotes
Funding/Support: This work was supported by an American Thoracic Society Fellows Career Development Award and an Actelion Pharmaceuticals Entelligence Young Investigator Award (Dr Halpern).
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/site/misc/reprints.xhtml).
References
- 1.Taichman DB, Mandel J. Epidemiology of pulmonary arterial hypertension. Clin Chest Med. 2007;28(1):1–22. doi: 10.1016/j.ccm.2006.11.012. [DOI] [PubMed] [Google Scholar]
- 2.Kravitz RL, Duan N, Braslow J. Evidence-based medicine, heterogeneity of treatment effects, and the trouble with averages. Milbank Q. 2004;82(4):661–687. doi: 10.1111/j.0887-378X.2004.00327.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Harch S, Whitford H, McLean C. Failure of medical therapy in pulmonary arterial hypertension. Is there an alternative diagnosis? Chest. 2009;135(6):1462–1469. doi: 10.1378/chest.08-2006. [DOI] [PubMed] [Google Scholar]
- 4.McLaughlin VV, Archer SL, Badesch DB. ACCF/AHA 2009 expert consensus document on pulmonary hypertension: a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association developed in collaboration with the American College of Chest Physicians, American Thoracic Society, Inc, and the Pulmonary Hypertension Association. Circulation. 2009;119(16):2250–2294. doi: 10.1161/CIRCULATIONAHA.109.192230. [DOI] [PubMed] [Google Scholar]
- 5.Benigni A, Remuzzi G. Endothelin antagonists. Lancet. 1999;353(9147):133–138. doi: 10.1016/S0140-6736(98)09423-9. [DOI] [PubMed] [Google Scholar]
- 6.Stewart DJ, Levy RD, Cernacek P, Langleben D. Increased plasma endothelin-1 in pulmonary hypertension: marker or mediator of disease? Ann Intern Med. 1991;114(6):464–469. doi: 10.7326/0003-4819-114-6-464. [DOI] [PubMed] [Google Scholar]
- 7.Rubens C, Ewert R, Halank M. Big endothelin-1 and endothelin-1 plasma levels are correlated with the severity of primary pulmonary hypertension. Chest. 2001;120(5):1562–1569. doi: 10.1378/chest.120.5.1562. [DOI] [PubMed] [Google Scholar]
- 8.Miyauchi T, Yanagisawa M, Iida K. Age- and sex-related variation of plasma endothelin-1 concentration in normal and hypertensive subjects. Am Heart J. 1992;123(4 Pt 1):1092–1093. doi: 10.1016/0002-8703(92)90734-d. [DOI] [PubMed] [Google Scholar]
- 9.Polderman KH, Stehouwer CD, van Kamp GJ, Dekker GA, Verheugt FW, Gooren LJ. Influence of sex hormones on plasma endothelin levels. Ann Intern Med. 1993;118(6):429–432. doi: 10.7326/0003-4819-118-6-199303150-00006. [DOI] [PubMed] [Google Scholar]
- 10.Stauffer BL, Westby CM, Greiner JJ, Van Guilder GP, Desouza CA. Sex differences in endothelin-1-mediated vasoconstrictor tone in middle-aged and older adults. Am J Physiol Regul Integr Comp Physiol. 2010;298(2):R261–R265. doi: 10.1152/ajpregu.00626.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Treiber FA, Kapuku GK, Davis H, Pollock JS, Pollock DM. Plasma endothelin-1 release during acute stress: role of ethnicity and sex. Psychosom Med. 2002;64(5):707–713. doi: 10.1097/01.psy.0000021952.59258.1c. [DOI] [PubMed] [Google Scholar]
- 12.Campia U, Cardillo C, Panza JA. Ethnic differences in the vasoconstrictor activity of endogenous endothelin-1 in hypertensive patients. Circulation. 2004;109(25):3191–3195. doi: 10.1161/01.CIR.0000130590.24107.D3. [DOI] [PubMed] [Google Scholar]
- 13.Rubin LJ, Badesch DB, Barst RJ. Bosentan therapy for pulmonary arterial hypertension. N Engl J Med. 2002;346(12):896–903. doi: 10.1056/NEJMoa012212. [DOI] [PubMed] [Google Scholar]
- 14.Channick RN, Simonneau G, Sitbon O. Effects of the dual endothelin-receptor antagonist bosentan in patients with pulmonary hypertension: a randomised placebo-controlled study. Lancet. 2001;358(9288):1119–1123. doi: 10.1016/S0140-6736(01)06250-X. [DOI] [PubMed] [Google Scholar]
- 15.Galiè N, Olschewski H, Oudiz RJ, 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. 2008;117(23):3010–3019. doi: 10.1161/CIRCULATIONAHA.107.742510. [DOI] [PubMed] [Google Scholar]
- 16.Barst RJ, Langleben D, Frost A, STRIDE-1 Study Group Sitaxsentan therapy for pulmonary arterial hypertension. Am J Respir Crit Care Med. 2004;169(4):441–447. doi: 10.1164/rccm.200307-957OC. [DOI] [PubMed] [Google Scholar]
- 17.Barst RJ, Langleben D, Badesch D, STRIDE-2 Study Group Treatment of pulmonary arterial hypertension with the selective endothelin-A receptor antagonist sitaxsentan. J Am Coll Cardiol. 2006;47(10):2049–2056. doi: 10.1016/j.jacc.2006.01.057. [DOI] [PubMed] [Google Scholar]
- 18.Sitbon O, Humbert M, Nunes H. Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: prognostic factors and survival. J Am Coll Cardiol. 2002;40(4):780–788. doi: 10.1016/s0735-1097(02)02012-0. [DOI] [PubMed] [Google Scholar]
- 19.Dewey FE, Rosenthal D, Murphy DJ, Jr, Froelicher VF, Ashley EA. Does size matter? Clinical applications of scaling cardiac size and function for body size. Circulation. 2008;117(17):2279–2287. doi: 10.1161/CIRCULATIONAHA.107.736785. [DOI] [PubMed] [Google Scholar]
- 20.Klebanoff MA, Cole SR. Use of multiple imputation in the epidemiologic literature. Am J Epidemiol. 2008;168(4):355–357. doi: 10.1093/aje/kwn071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sakao S, Tanabe N, Tatsumi K. The estrogen paradox in pulmonary arterial hypertension [published retraction appears in Am J Physiol Lung Cell Mol Physiol. 2011;300(3):L508] Am J Physiol Lung Cell Mol Physiol. 2010;299(4):L435–L438. doi: 10.1152/ajplung.00057.2010. [DOI] [PubMed] [Google Scholar]
- 22.Sweeney L, Voelkel NF. Estrogen exposure, obesity and thyroid disease in women with severe pulmonary hypertension. Eur J Med Res. 2009;14(10):433–442. doi: 10.1186/2047-783X-14-10-433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kawut SM, Al-Naamani N, Agerstrand C. Determinants of right ventricular ejection fraction in pulmonary arterial hypertension. Chest. 2009;135(3):752–759. doi: 10.1378/chest.08-1758. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Humbert M, Sitbon O, Chaouat A. Survival in patients with idiopathic, familial, and anorexigen-associated pulmonary arterial hypertension in the modern management era. Circulation. 2010;122(2):156–163. doi: 10.1161/CIRCULATIONAHA.109.911818. [DOI] [PubMed] [Google Scholar]
- 25.Ergul S, Parish DC, Puett D, Ergul A. Racial differences in plasma endothelin-1 concentrations in individuals with essential hypertension. Hypertension. 1996;28(4):652–655. doi: 10.1161/01.hyp.28.4.652. [DOI] [PubMed] [Google Scholar]
- 26.Davis KK, Lilienfeld DE, Doyle RL. Increased mortality in African Americans with idiopathic pulmonary arterial hypertension. J Natl Med Assoc. 2008;100(1):69–72. doi: 10.1016/s0027-9684(15)31177-9. [DOI] [PubMed] [Google Scholar]
- 27.Kawut SM, Horn EM, Berekashvili KK. New predictors of outcome in idiopathic pulmonary arterial hypertension. Am J Cardiol. 2005;95(2):199–203. doi: 10.1016/j.amjcard.2004.09.006. [DOI] [PubMed] [Google Scholar]
- 28.Lederer DJ, Arcasoy SM, Barr RG. Racial and ethnic disparities in idiopathic pulmonary fibrosis: A UNOS/OPTN database analysis. Am J Transplant. 2006;6(10):2436–2442. doi: 10.1111/j.1600-6143.2006.01480.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lederer DJ, Caplan-Shaw CE, O'Shea MK. Racial and ethnic disparities in survival in lung transplant candidates with idiopathic pulmonary fibrosis. Am J Transplant. 2006;6(2):398–403. doi: 10.1111/j.1600-6143.2005.01205.x. [DOI] [PubMed] [Google Scholar]
- 30.Spertus JA, Jones PG, Masoudi FA, Rumsfeld JS, Krumholz HM. Factors associated with racial differences in myocardial infarction outcomes. Ann Intern Med. 2009;150(5):314–324. doi: 10.7326/0003-4819-150-5-200903030-00007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rothwell PM, Mehta Z, Howard SC, Gutnikov SA, Warlow CP. Treating individuals 3: from subgroups to individuals: general principles and the example of carotid endarterectomy. Lancet. 2005;365(9455):256–265. doi: 10.1016/S0140-6736(05)17746-0. [DOI] [PubMed] [Google Scholar]
- 32.Ioannidis JPA, Lau J. Heterogeneity of the baseline risk within patient populations of clinical trials: a proposed evaluation algorithm. Am J Epidemiol. 1998;148(11):1117–1126. doi: 10.1093/oxfordjournals.aje.a009590. [DOI] [PubMed] [Google Scholar]
- 33.Kent DM, Hayward RA. Limitations of applying summary results of clinical trials to individual patients: the need for risk stratification. JAMA. 2007;298(10):1209–1212. doi: 10.1001/jama.298.10.1209. [DOI] [PubMed] [Google Scholar]
- 34.Hayward RA, Kent DM, Vijan S, Hofer TP. Reporting clinical trial results to inform providers, payers, and consumers. Health Aff (Millwood) 2005;24(6):1571–1581. doi: 10.1377/hlthaff.24.6.1571. [DOI] [PubMed] [Google Scholar]
- 35.Snow JL, Kawut SM. Surrogate end points in pulmonary arterial hypertension: assessing the response to therapy. Clin Chest Med. 2007;28(1):75–89. doi: 10.1016/j.ccm.2006.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Marshall SW. Power for tests of interaction: effect of raising the type I error rate. Epidemiol Perspect Innov. 2007;4:4. doi: 10.1186/1742-5573-4-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.