Previous Presentations: A preliminary version of the data included in this manuscript was presented at the 50th Annual Meeting of the American Society of Hematology, San Francisco, December 2008.
Effect of eculizumab on haemolysis-associated nitric oxide depletion, dyspnoea, and measures of pulmonary hypertension in patients with paroxysmal nocturnal haemoglobinuria
Article first published online: 8 MAR 2010
© 2010 Blackwell Publishing Ltd
British Journal of Haematology
Volume 149, Issue 3, pages 414–425, May 2010
How to Cite
Hill, A., Rother, R. P., Wang, X., Morris, Jr, S. M., Quinn-Senger, K., Kelly, R., Richards, S. J., Bessler, M., Bell, L., Hillmen, P. and Gladwin, M. T. (2010), Effect of eculizumab on haemolysis-associated nitric oxide depletion, dyspnoea, and measures of pulmonary hypertension in patients with paroxysmal nocturnal haemoglobinuria. British Journal of Haematology, 149: 414–425. doi: 10.1111/j.1365-2141.2010.08096.x
The study protocol was approved by an institutional ethical committee and all patients gave written informed consent.
Trial Registration: ClinicalTrials.gov number, NCT00122330.
- Issue published online: 12 APR 2010
- Article first published online: 8 MAR 2010
- Received 3 August 2009; accepted for publication 19 November 2009
- paroxysmal nocturnal haemoglobinuria;
- nitric oxide;
- pulmonary hypertension;
- Top of page
- Financial Disclosures
Pulmonary hypertension (PH) is a common complication of haemolytic anaemia. Intravascular haemolysis leads to nitric oxide (NO) depletion, endothelial and smooth muscle dysregulation, and vasculopathy, characterized by progressive hypertension. PH has been reported in patients with paroxysmal nocturnal haemoglobinuria (PNH), a life-threatening haemolytic disease. We explored the relationship between haemolysis, systemic NO, arginine catabolism and measures of PH in 73 PNH patients enrolled in the placebo-controlled TRIUMPH (Transfusion Reduction Efficacy and Safety Clinical Investigation Using Eculizumab in Paroxysmal Nocturnal Haemoglobinuria) study. At baseline, intravascular haemolysis was associated with elevated NO consumption (P < 0·0001) and arginase-1 release (P < 0·0001). Almost half of the patients in the trial had elevated levels (≥160 pg/ml) of N-terminal pro-brain natriuretic peptide (NT-proBNP), a marker of pulmonary vascular resistance and right ventricular dysfunction previously shown to indicate PH. Eculizumab treatment significantly reduced haemolysis (P < 0·001), NO depletion (P < 0·001), vasomotor tone (P < 0·05), dyspnoea (P = 0·006) and resulted in a 50% reduction in the proportion of patients with elevated NT-proBNP (P < 0·001) within 2 weeks of treatment. Importantly, the significant improvements in dyspnoea and NT-proBNP levels occurred without significant changes in anaemia. These data demonstrated that intravascular haemolysis in PNH produces a state of NO catabolism leading to signs of PH, including elevated NT pro-BNP and dyspnoea that are significantly improved by treatment with eculizumab.
Mild-to-moderate pulmonary hypertension (PH) occurs in up to 30% of adult patients with sickle cell disease and has been implicated as a complication in various other hereditary haemolytic anaemias, including thalassaemia and hereditary spherocytosis (Rother et al, 2005). It is believed to be linked to intravascular haemolysis, leading to the development of the term ‘haemolysis-associated PH’ (Collins & Orringer, 1982; Aessopos et al, 2001; Gladwin et al, 2004). The release of excessive red cell haemoglobin during intravascular haemolysis can exceed the capacity of the haemoglobin scavenging molecule, haptoglobin, resulting in high levels of cell-free haemoglobin that ultimately result in the consumption of endogenous NO (Tabbara, 1992; Rother et al, 2005). Haemolysis also releases erythrocyte arginase 1, an enzyme that converts L-arginine, the substrate for NO synthesis, to ornithine, thereby depleting the plasma pool of arginine and further reducing the systemic availability of NO (Azizi et al, 1970; Morris et al, 2003, 2004; Schnog et al, 2004). Arginase 1 activity is particularly abundant in immature or young red blood cells and reticulocytes (Azizi et al, 1970), both of which are plentiful in sickle cell disease and other conditions of rapid erythrocyte turnover. Plasma arginase 1 levels are highest in those patients with markers of accelerated haemolytic rate (Morris et al, 2005) and therefore worthy of exploration in patients with PNH.
Depletion of systemic NO has been linked to the development of systemic and pulmonary vascular resistance, PH and multiple other sequelae by dysregulation of endothelial function and smooth muscle tone (Rother et al, 2005). Cell-free plasma haemoglobin has been shown to augment hypoxic pulmonary vasoconstriction and systemic vasoconstriction by scavenging NO (Deem et al, 2002; Minneci et al, 2005). The role of haemolysis-produced cell-free plasma haemoglobin and NO depletion on systemic and PH has been explored in sickle cell disease (Reiter et al, 2002) and malaria (Aliyu et al, 2008) but has not been examined to date in other haemolytic diseases.
Elevated levels of the brain natriuretic peptide (BNP) hormone reflect cardiac chamber (particularly ventricular) volume and pressure overload (Levin et al, 1998; Hall, 2005). Elevated BNP levels strongly indicate increased pulmonary vascular resistance and right ventricular (RV) dysfunction (Nagaya et al, 1998; Leuchte et al, 2004; Andreassen et al, 2006; Machado et al, 2006; Voskaridou et al, 2007). The prognostic importance of BNP has been demonstrated in several cardiovascular disorders. In patients with pulmonary arterial hypertension, BNP levels correlate with the severity of pulmonary artery pressure elevation and with RV dysfunction. Moreover, in patients with haemolytic anaemia (sickle cell disease) an N-terminal pro-brain natriuretic peptide (NT-proBNP) level ≥160 pg/ml (normal range 16–49 pg/ml) showed a highly positive predictive value for the diagnosis of PH, demonstrating NT-proBNP as a non-invasive marker for PH (Machado et al, 2006).
In paroxysmal nocturnal haemoglobinuria (PNH), an acquired deficiency of the complement inhibitors CD55 and CD59 from the cell surface of blood cells results in the unopposed assembly of the terminal complement complex on red blood cells (RBCs), resulting in intravascular destruction and subsequent release of high levels of cell-free haemoglobin (Yamashina et al, 1990). The degree of haemolysis typically occurring in PNH surpasses that of other hereditary, acquired and iatrogenic haemolytic conditions with levels of lactate dehydrogenase (LDH; a biochemical marker of haemolysis) exceeding 20 times that of normal (Tabbara, 1992; Paquette et al, 1997; Hillmen et al, 2004). Further, studies in two different cohorts of patients with PNH have been shown to have evidence of PH, including elevated regurgitant tricuspid valve velocity and NT-proBNP levels (Hill et al, 2005a, 2006). Therefore it is reasonable to important to evaluate whether the intravascular haemolysis in patients with PNH leads to significant NO consumption and subsequently signs and symptoms of PH.
The drug eculizumab (Soliris™; Alexion Pharmaceuticals, Inc., Cheshire, CT, USA), a humanized monoclonal antibody targeting complement C5, substantially blocks C5a and complement-mediated haemolysis in PNH (Hillmen et al, 2004, 2006; Hill et al, 2005b). Patients treated with eculizumab had significant decreases in haemolysis and showed a significant improvement in fatigue (independent of changes in total haemoglobin level and other variables), quality of life, and kidney function, as well as a reduction in thrombotic events and pain, indicating the presence of significant haemolysis-dependent morbidities (Hillmen et al, 2004, 2006, 2007; Hill et al, 2005b). This study investigated whether intravascular haemolysis in patients with PNH is associated with NO depletion, systemic arterial pressures as a measure of vasomotor tone, and elevated NT-proBNP levels as a biomarker that characterises left or right ventricular pressure overload, that have previously been shown to indicate PH and RV dysfunction in hemolytic diseases. We further evaluated the influence of eculizumab therapy on these measures and associated morbidities.
- Top of page
- Financial Disclosures
TRIUMPH (Transfusion Reduction Efficacy and Safety Clinical Investigation Using Eculizumab in Paroxysmal Nocturnal Hemoglobinuria) was an international phase 3 randomized, placebo-controlled trial in which 87 patients (35 men and 52 women) were randomly assigned to eculizumab (43 patients) or placebo (44 patients) between October 2004 and June 2005. Written consent was obtained prior to entry into the trial. There were no significant differences in the baseline characteristics of the patients in the two groups (Hillmen et al, 2006). Exploratory endpoints in the TRIUMPH study included measures of cell-free plasma haemoglobin and NO depletion by patient plasma.
Biochemical measures of intravascular haemolysis
Red cell lysis was assessed by measuring serum LDH levels, a measure of haemolysis. To determine levels of cell-free plasma haemoglobin during eculizumab treatment, blood samples were collected at baseline and at scheduled visits in the double-blind, randomized, placebo-controlled, multinational phase III TRIUMPH study (Hillmen et al, 2006). All samples were assayed for levels of cell-free plasma haemoglobin by Quest Diagnostics, Chantilly, VA, USA.
The NO consumption assay was carried out according to previously described methods (Wang et al, 2004). Briefly, a 50 ml solution of 40 μmol/l DETANONOate in phosphate-buffered saline, pH 7·4, was prepared in a glass vessel actively purged with helium in-line with an NO chemiluminescence analyser. This solution produced a steady-state NO signal of approximately 50–70 mV, which was generated by the decay of DETANONOate and the release of NO. When the signal became stable, 50 μl samples of standards or PNH plasma fractions were injected into the NONOate solution. Known concentrations of haemoglobin were used to create standard curves and produced concentration-dependent, linear, and reproducible decreases in the baseline NO signal (NO consumption, mV). Data were transferred to the software programme origin Version 6.1 (OriginLab, Northampton, MA, USA) for analysis of the area under the curve of decreasing NO signal over time. The amount of NO consumed by samples was quantified by comparison of the area under the curve with that of NO consumption standard (produced from injections of haemoglobin). Samples with higher concentration of haemoglobin were diluted before the injection.
Arginase 1 in plasma was measured at baseline using an arginase 1 enzyme-linked immunosorbent assay kit from Cell Sciences (Canton, MA, USA). The assay was performed according to the manufacturer. Arginase activity in plasma samples was assayed as described previously (Morris et al, 2005). Briefly, arginase activity was measured as the conversion of [14C-guanidino]-l-arginine (American Radiolabeled Chemicals, St. Louis, MO, USA) to [14C]urea, which was converted to 14CO2 by urease and trapped as Na214CO3 and quantified by scintillation counting. Arginase activities were expressed as nmoles/ml/min.
Analysis of NT-proBNP
Levels of NT-proBNP were determined on the Elecsys 170 immunoassay system per the manufacturer’s protocol (Roche Diagnostics, Lewes, UK). The interassay % coefficient of variation (CV) was 5·0 at 380 ng/l, 4·4 at 8700 ng/l, 5·0 at 13 000 ng/l, with detection limit 20 ng/l and upper measuring limit 25 000 ng/l.
Assessment of dyspnoea
Assessment of dyspnoea was determined using the validated European Organization for Research and Treatment of Cancer Quality of Life Core 30 (EORTC-QLQ-C30) instrument. This analysis was a pre-specified exploratory analysis in TRIUMPH and patient-reported outcome questionnaires were administered at baseline before any study procedure and at weeks 1, 2, 3, 4, 12, 20 and 26 for TRIUMPH. The EORTC-QLQ-C30 instrument is a self-report questionnaire that incorporates 5 functional scales, 3 symptom scales, several single-item measures and a global health scale. Shortness of breath was captured by patient self-reported responses to frequency, severity and distress of their shortness of breath on a reported scale. Scores can range from 0 to 100, and lower scores on the symptom scale and single-item measures indicating improvement. To better describe the magnitude of any clinical impact on the pre-specified health related quality of life (QoL) parameters using both validated QoL instruments, the standard effect size (SES) was to determine the magnitude of the clinical benefits assessed by QoL instruments. The SES was calculated as a ratio of the mean change to the standard deviation of that score. A standard effect of >0·8 was one that had a large impact, whereas <0·5 had a small clinical impact.
For comparisons of categorical variables, Fisher exact test or Chi-square test was used, and for continuous variables, Wilcoxon’s rank-sum test was used. Correlation between continuous variables was assessed via regression analysis. A P-value ≤0·05 was considered to be statistically significant. Mean values are shown ± standard error (SE).
To assess the relative change in cell-free plasma haemoglobin and NO consumption during eculizumab treatment (Hillmen et al, 2006), levels of cell-free plasma haemoglobin and NO consumption were compared between eculizumab and placebo-treated patients using a Wilcoxon’s rank sum test.
- Top of page
- Financial Disclosures
Baseline measures of haemolysis in PNH patients
Levels of LDH were highly elevated in PNH patients entered into the placebo-controlled TRIUMPH study (Table I) Baseline levels of both cell-free plasma haemoglobin and arginase 1 were highly elevated (Table I). In addition, there was a significant correlation between levels of LDH and cell-free plasma haemoglobin, confirming that LDH is a good marker for haemolysis in patients with PNH (Fig 1A; R = 0·5094, P < 0·001).
|Haemolytic measures at baseline||PNH patients mean ± SD (N = 73)*||Normal range|
|Lactate dehydrogenase, (u/l)||2229 ± 1025||103–223|
|Cell-free plasma haemoglobin, (μmol/l)||27·8 ± 65·5||<2|
|Arginase 1, (ng/ml)||115·1 ± 305·9||1·8–29·5 ng/ml (Ikemoto et al, 2001)|
|Arginase 1 enzyme activity, (nmol/ml/min)||9·3 ± 26·9||0·2–3·8 nmoles/ml/min†|
|NO consumption, (μmol/l)||24·2 ± 46·8||<2|
Correlation between cell-free plasma haemoglobin from red cells in patients with PNH and NO consumption
The level of NO consumption by patient plasma was markedly increased over normal (24·2 vs. <2·0 μmol/l, respectively) (Table I). Further, a significant correlation was demonstrated between cell-free plasma haemoglobin and NO consumption (Fig 1B; R = 0·9529, P < 0·0001), indicating that plasma haemoglobin remained in the reduced ferrous oxyhaemoglobin state (Fe2+-O2), competent to react with NO. Plasma from a PNH patient consumed more NO than plasma from a normal control (Fig 1E).
Release of red cell arginase 1 during haemolysis and catabolism of arginine
At baseline, levels of arginase 1 were highly elevated in patients with PNH (115·1 ± 305·9 ng/ml compared with normal range of 1·8–29·5 ng/ml) (Table I). A significant correlation between arginase 1 and cell-free plasma haemoglobin was demonstrated (Fig 1C; R = 0·9367, P < 0·0001). In addition, arginase 1 in the plasma of PNH patients was capable of arginine catabolism (Table I) and a significant correlation between levels of arginase 1 and enzyme activity was observed (Fig 1D, R = 0·9081, P < 0·0001).
Effect of eculizumab treatment on levels of haemolysis, cell-free plasma haemoglobin and NO consumption
Measurements of LDH, cell-free haemoglobin and NO consumption were conducted at baseline and at scheduled visits during the 26-week study. Eculizumab-treated patients showed a significant reduction in haemolysis from a mean LDH level of 2200 ± 158 u/l at baseline to 327 ± 68 u/l at week 26 while levels of LDH in placebo patients remained highly elevated (2258 ± 155 u/l at baseline to 2419 ± 140 u/l at week 26: P < 0·001 for comparison between eculizumab- and placebo-treated patients). Eculizumab treatment also resulted in a concomitant decrease in levels of cell-free plasma haemoglobin from a mean of 988 ± 232·4 mg/l at baseline to 152 ± 50·5 mg/l at 26 weeks while free haemoglobin levels remained unchanged in placebo-treated patients (679 ± 119·3 to 877 ± 124·6 mg/l; Fig 2A; P < 0·001). The significant reduction in levels of free haemoglobin was demonstrated within 1 week of eculizumab therapy and maintained through the 26-week study. Similarly, the consumption of NO by plasma from eculizumab-treated patients was significantly reduced compared to plasma from placebo-treated patients (Fig 2B; P ≤ 0·03). As compared to baseline NO consumption, eculizumab treatment was associated with a 67·1% reduction while placebo-treated patients demonstrated a 14·9% increase. The reduction in NO depletion was apparent within 2 weeks of eculizumab treatment and the effect was maintained throughout the treatment period.
Levels of NT-proBNP in patients with PNH and response to eculizumab treatment
Patients in both the eculizumab- and placebo-treated groups had elevated NT-proBNP levels above normal at baseline. The mean (±SE) and median NT-proBNP levels at baseline in the eculizumab-treated group was 364 (±98·7) and 182 pg/ml, respectively, and 299·9 (±104·0) and 118·0 pg/ml in the placebo-treated group, respectively (Table II). This study did not include normal control individuals; however, previous studies have demonstrated that NT-proBNP levels in normal volunteers range from 16–49 pg/ml using the same methods as were employed in the current study. Treatment with eculizumab reduced NT-proBNP levels from baseline, a median reduction of 560 pg/ml during treatment as measured by the area under the curve (AUC), compared to a median increase of 135 pg/ml NT-proBNP AUC in the placebo group (Table II; P < 0·05 eculizumab versus placebo).
|Baseline||Week 2||Week 14||Week 26|
|P-value (eculizumab vs. Placebo)||P = 0·04*|
In the current study, 46·6% (34/73) of PNH patients had baseline levels of NT-proBNP ≥160 pg/ml (a level considered indicative of PH in another haemolytic disease, sickle cell disease). Overall, eculizumab-treated patients showed a 50% reduction in the incidence of patients with levels of NT-proBNP ≥160 pg/ml over the course of the 26-week treatment period (from 52·5% to 26·3%), while the incidence did not significantly change with placebo (from 39·4% to 43·8%; P < 0·005, eculizumab vs. placebo: Fig 3A). The improvement in BNP levels in eculizumab-treated patients was rapid as it was apparent after 2 weeks of treatment (52·5% to 20%) and was maintained through 26 weeks (Fig 4B).
Dyspnoea in patients with PNH and response to eculizumab treatment
Dyspnoea, a symptom of PH or heart failure, also improved during eculizumab treatment relative to placebo during the study (mixed model analysis; P < 0·001). The median scores for dyspnoea were 33·3 at baseline in both eculizumab- and placebo-treated patients (Fig 3B). The dyspnoea score in eculizumab-treated patients significantly decreased to a median of 16·7 (P = 0·006) (lower score indicates improvement) while the score in placebo-treated patients remained unchanged (33·3; P = 0·005 eculizumab- vs. placebo-treated groups). The SES (which denotes the magnitude of clinical improvement) for the change in dyspnoea between eculizumab and placebo treatment was moderate to large (SES 0·69; P < 0·001). Moreover, eculizumab treatment was associated with a significant improvement in dyspnea at an early time point in which anaemia was not significantly changed. The improvement in dyspnoea in eculizumab-treated patients was rapid, occurring within 2 weeks of treatment. The mean changes in dyspnoea scores from baseline were significantly improved early and at the end of the observation in the eculizumab-treated group, −10·9 (P < 0·005), and −7·9 (P = 0·031) at 2 and 26 weeks, respectively (Figs 3B and 4A), and were significantly improved vs. placebo (P = 0·002 at week 26). The mean changes in dyspnoea scores in the placebo group did not show significant improvement from baseline at week 2 (−2·3; P = 0·58) and showed worsening of dyspnoea at week 26 (+8·9; P = 0·033). The mean changes in haemoglobin levels from baseline were not significantly changed early or at the end of observation in eculizumab treated patients, +2·2 g/dl (P = 0·23) at 2 weeks and −2·0 g/dl (P = 0·95) at 26 weeks, respectively (Figs 3C and 4A). In contrast, the haemoglobin levels worsened from baseline in the placebo group, a significant decrease of 66 g/l (P = 0·0005) at 2 weeks and a decrease of 81 g/l (P = 0·0008) at 26 weeks.
Due to the early improvement in dyspnoea and NT-proBNP levels during treatment with eculizumab (Fig 4A, B), a regression analysis was performed and indicated a positive correlation between the improvement in dyspnoea and decrease NT-proBNP levels (R2 = 0·0273 P = 0·015 (coef = 0·00765).
Systemic hypertension in patients with PNH and response to eculizumab treatment
As the study did not permit measurement of pulmonary arterial pressures, the effect of eculizumab on vasomotor tone was instead measured by sequential examinations of systemic arterial pressures at baseline and at 1, 2 and 4 weeks. Eculizumab-treated patients showed a rapid decrease in both systolic (−10 mmHg P = 0·01 vs. baseline; Fig 5A) and diastolic (−4·0 mmHg, P < 0·001; Fig 5B) pressures within the first four weeks of therapy as compared to placebo-treated patients, consistent with significant reductions in vasomotor tone.
- Top of page
- Financial Disclosures
The pathophysiology of haemolytic syndromes is being increasingly understood in an integrated manner (Rother et al, 2005). Intravascular haemolysis is associated with many common signs and symptoms in PNH and other haemolytic syndromes including thrombosis, fatigue, impaired quality of life, haemoglobinuria, dyspnoea, dysphagia, abdominal pain, anaemia, and erectile dysfunction (Hillmen et al, 2004, 2006; Hill et al, 2005c; Rother et al, 2005; Schubert et al, 2008). These signs and symptoms in PNH, and in patients with other haemolytic syndromes, are tied to catastrophic outcomes, including acute renal failure, chronic kidney disease, stroke/transient ischaemic attack, ischaemic bowel, hepatic failure, and deep vein thrombosis (Clark et al, 1981; Vellenga et al, 1982; Doukas et al, 1984; Hillmen et al, 1995, 2007; Adams et al, 2002; Hill et al, 2005a,c). Many of these symptoms have been under-recognised as significant contributors to morbidity and mortality in patients with PNH and other haemolytic diseases. Of these, all but anaemia can be directly related to the release of cell-free haemoglobin and subsequent NO depletion (Rosse, 2000; Hill et al, 2005c, 2008; Rother et al, 2005).
NO is consumed by cell-free plasma oxyhaemoglobin during excessive intravascular haemolysis (Rother et al, 2005), and levels of haemolysis correlate with NO consumption in sickle cell disease (Kato et al, 2006). Here we confirm an association between intravascular haemolysis and NO depletion through direct binding to free haemoglobin and reduction of NO production through reduced arginase 1 availability in patients with PNH. The high correlation between cell-free haemoglobin and NO consumption and arginase 1 supports the role intravascular haemolysis plays in the morbidities of PNH. Blocking intravascular haemolysis with eculizumab resulted in a dramatic reduction in haemolysis, free haemoglobin and NO consumption. Further, there was subsequent change in vasomotor tone, a functional output of NO function, as measured by the change in systolic and diastolic blood pressures.
NO plays a major role in vascular homeostasis and has been shown to be a critical regulator of basal and stress-mediated smooth muscle relaxation and vasomotor tone. PH has been demonstrated in previous cohorts of patients with PNH (Hill et al, 2005a,b,c; Hill et al, 2006). Doppler echocardiography was performed in 28 haemolytic PNH patients to estimate pulmonary artery pressures as defined by a tricuspid regurgitant jet velocity (TRV) of at least 2·5 m/s at rest. Fourteen of 20 evaluable patients (70%) demonstrated elevated pulmonary artery systolic pressures. Twelve (60%) had mild-to-moderate PHT while two (10%) had moderate-to-severe pressures (mean TRV 3·7 ± 0·2 m/s). These PNH patients also demonstrated high NO consumption levels. Plasma from patients with PNH (n = 32) consumed 34·6 ± 8·3 micromolar NO while plasma from normal subjects (n = 9) consumed 2·2 ± 0·6 micromolar NO (P < 0·0001). Although our current study did not allow direct measurements of pulmonary arterial pressure, we showed a strong association between intravascular haemolysis and vasomotor tone during eculizumab treatment with a significant decrease in systemic blood pressure at weeks 1 to 4 weeks of the study. The median change of systolic pressure (−10 mmHg) with eculizumab was clinically significant. For example, it has been estimated that an intervention in a population that results in a 5 mmHg reduction of systolic blood pressure in the population would result in a 14% overall reduction in mortality due to stroke, a 9% reduction in mortality due to chronic heart disease and a 7% decrease in all-cause mortality (Chobanian et al, 2003). While the current study indicated a statistically and clinically significant improvement in vasomotor tone in PNH patients treated with eculizumab, it was not designed to evaluate whether this would be associated with any improvement in major adverse cardiovascular events in treated PNH patients. Taken together, these data suggest that patients with PNH have high levels of NO consumption and increased vasomotor tone that can manifest as PH or systemic hypertension through elevation of NO.
Elevated levels of NT-proBNP strongly indicate elevated pulmonary vascular resistance and subsequent RV dysfunction in patients with isolated right-sided disease (Nagaya et al, 1998; Leuchte et al, 2004; Andreassen et al, 2006; Fijalkowska et al, 2006; Machado et al, 2006; Voskaridou et al, 2007). Our study found high level NT-proBNP and dyspnea, consistent with the development of PH and RV stress. We demonstrated elevated NT pro-BNP levels in our PNH cohort (median 182 and 118 pg/ml for eculizumab- and placebo-treated groups, respectively) and almost half the patients (47%) had NT-proBNP levels ≥ 160 pg/ml, supporting previous studies that demonstrated PH as measured by TRV and BNP levels in PNH patients (Hill et al, 2006). The prognostic importance of BNP and NT-proBNP has been demonstrated in patients with left-sided cardiac failure as well as unstable and stable coronary artery disease (de Lemos et al, 2001; Gardner et al, 2003; Bibbins-Domingo et al, 2007). In patients with sickle cell disease, NT-proBNP levels of ≥160 pg/ml or greater showed a 78% predictive value for the presence of PH and also independently predicted patient mortality (Machado et al, 2006). In a study of a separate group of haemolytic patients, heterozygote HbS/Β-Thalassaemia patients, NT-proBNP levels were highly correlated with tricuspid valve velocity and an NT-proBNP ≥153·6 pg/ml had 86% sensitivity and 95% specificity for PH (Voskaridou et al, 2007). This further supports the level of approximately 160 pg/ml as a threshold for definition of clinically significant PH in haemolytic patients. Levels of NT-proBNP were also associated with increased levels of LDH, aspartate aminotransferase and total bilirubin and low haemoglobin levels, all of which would be expected from chronic haemolysis (Machado et al, 2006; Voskaridou et al, 2007) In our study 47% of PNH patients had elevated NT-proBNP levels (≥160 pg/ml) and treatment with eculizumab was associated with a significant reduction in haemolysis and subsequently a reduction in NT-proBNP levels in 50% of these high risk patients. These data suggest an increased pulmonary arterial resistance in PNH patients. Further, in the current study, intervention with eculizumab was associated with an improvement in NT-proBNP before and without a significant change in anaemia compared to baseline.
Dyspnoea is a clinical symptom of increased pulmonary pressure and PH. Eculizumab-treated patients showed a statistically significant (P < 0·001) and clinically meaningful (SES 0·69) improvement in dyspnoea compared to placebo during the 26-week treatment period using the EORTC QLQ-C30 instrument. Improvement in dyspnoea occurred within weeks of eculizumab treatment and was sustained throughout the study. While the anaemia itself and other non-cardiac/pulmonary causes may lead to dyspnoea, it is interesting to note that the improvement in dyspnoea during eculizumab treatment was not associated with a change in haemoglobin level, as haemoglobin levels fluctuated and ultimately showed no changed at week 26. Therefore, it appears that the significant improvement in dyspnoea with eculizumab treatment is not meaningfully related to any change in haemoglobin. Similar to the course for dyspnoea, and in contrast to haemoglobin levels, vasomotor tone (systemic blood pressure) and NT-proBNP levels, haemolysis and NO consumption all improved within 2 weeks of treatment and were sustained throughout the course of the study (Figs 3 and 4). Thrombosis is a common symptom in patients with PNH and evidence of pulmonary embolism and micro thrombi may contribute to PH in patients with PNH and in other clinical settings. Of the 34 patients with elevated NT-proBNP levels in the study, only two patients had documented pulmonary embolism, at 1 and 3·5 years, respectively, prior to study entry. It is possible that some patients may have sub-clinical embolisms similar to that detected in a separate cohort of PNH patients (Hill et al, 2006) that could be contributing to the elevated arterial pulmonary pressure. However, treatment with eculizumab resulted in a rapid reversal, within 2 weeks, of elevated systemic blood pressure, dyspnea and NT-proBNP levels. We consider it more likely that the improvement in measures of dyspnoea and PH were due to associated rapid reductions in haemolysis, NO consumption, and pulmonary vascular tone vs. rapid dissolution of diffuse pulmonary microthrombi. Moreover, the strong association of the rapid improvement in dyspnea with eculizumab, and the parallel observation of the rapid improvements in NT-proBNP, together with the lack of association of the dyspnea improvement with significant changes in anaemia, provides independent and strong clinical support that the improvements in NT-proBNP indicate substantial improvements in right-sided stress and pulmonary vascular hypertension with eculizumab treatment.
PH is a common morbidity in haemolytic diseases, in which it has been reported as an independent risk factor for death (Gladwin et al, 2004). Indeed, PH has been previously identified in a separate cohort of PNH patients using cardiac doppler measurements (Hill et al, 2005a). Further, clinical symptoms of PNH, an acquired haemolytic anaemia with excessive levels of intravascular haemolysis, are also consistent with PH and include severe fatigue, severe dyspnoea. Importantly, we have previously shown that the significant improvements in fatigue in PNH patients treated with eculizumab were unrelated to changes in haemoglobin. In the current study, we extended those observations and showed that the significant improvements in dyspnoea with eculizumab are also unrelated to changes in haemoglobin. Blocking haemolysis with the terminal complement inhibitor eculizumab significantly reduced the incidence of measures of PH in patients with PNH. The consequences of NO consumption on pulmonary vascular tone are likely to be contributory in PNH. Similar effects on systemic vasculature have been described in association with symptoms in patients with PNH (Rother et al, 2005). The systemic removal of NO as a consequence of chronic haemolysis has been linked to smooth muscle constriction and shown to contribute to clinical morbidities including severe oesophageal spasm and dysphagia, abdominal pain, erectile dysfunction and thrombosis (Radomski et al, 1987; Murray et al, 1995; Olsen et al, 1996; Przybelski et al, 1996; Moyo et al, 2004; Schafer et al, 2004). Others have previously shown that administration of exogenous free haemoglobin causes pulmonary vasoconstriction in animal models (Mazmanian et al, 1989; Hasunuma et al, 1991; Voelkel et al, 1995; Deem et al, 1998, 2000, 2001, 2002; Heller et al, 1998; Rother et al, 2005).
In conclusion, we demonstrated that multiple measures of haemolysis were highly elevated in patients with PNH including LDH, cell-free plasma haemoglobin, arginase 1 and NO consumption. Further, treatment of patients with the terminal complement inhibitor eculizumab resulted in a dramatic decrease in all subsequently measured parameters of haemolysis. Similarly, high levels of NT-proBNP rapidly declined as the haemolysis resolved with treatment. These data support a causal relationship between haemoglobinaemia, NO depletion and elevated levels of NT-proBNP. Specifically, haemolysis and subsequent generation of cell-free haemoglobin and arginase, leading to NO consumption and reduced synthesis, respectively, would be expected to result in disruption of homeostatic vascular function, including control of basal and shear stress-mediated vasodilator tone (Rother et al, 2005). Moreover, the current study showed a tight linkage between improvements in haemolysis and decreases in NO consumption, systemic vasomotor tone, and NT-proBNP levels. The current study suggests that chronic haemolysis in PNH contributes to elevated vascular tone and indications of PH which may be reduced by inhibition of haemolysis with the terminal complement inhibitor eculizumab.
- Top of page
- Financial Disclosures
The authors would like to thank Louise Arnold, Carol Bilbrough, David Buchanan, Sue Davison, and Catherine Norton from the research team in the Department of Haematology, Leeds General Infirmary, Leeds, U.K. We would also like to acknowledge Jason Chan, Alexion Pharmaceuticals, Inc. for performing statistical analyses and Dr. Gus Khursigara, Alexion Pharmaceuticals, Inc. for critical review of the manuscript. Supported by an educational grant from Alexion Pharmaceuticals, Inc.
- Top of page
- Financial Disclosures
R.P.R. and L.B. have financial interest in and are employed by Alexion Pharmaceuticals Inc.; R.P.R. and L.B. hold patents related to the work that is described in the present study and receive no royalties for these patents; A.H., R.K., S.J.R. M.B. and P.H. have received honoraria and served on Scientific Advisory Boards for Alexion Pharmaceuticals, Inc.; M.B. and P.H. have received research grants from Alexion Pharmaceuticals, Inc.; X.W., S.M., K.Q. and MTG have no relevant conflicts of interest.
- Top of page
- Financial Disclosures
This study was funded by a research grant from Alexion Pharmaceuticals, Inc. Dr Gladwin’s work was funded by the Intramural Research Division of the NHLBI, NIH and by the Institute of Transfusion Medicine and the Hemophilia Center of Western Pennsylvania, Pittsburgh, PA.
- Top of page
- Financial Disclosures
- 2002) Gastrointestinal involvement in paroxysmal nocturnal hemoglobinuria: first report of electron microscopic findings. Digestive Diseases and Sciences, 47, 58–64. , , , & (
- 2001) Cardiac involvement in thalassemia intermedia: a multicenter study. Blood, 97, 3411–3416. , , , , , , , & (
- 2008) Sickle cell disease and pulmonary hypertension in Africa: a global perspective and review of epidemiology, pathophysiology, and management. American Journal of Hematology, 83, 63–70. , , , , , & (
- 2006) N-terminal pro-B-type natriuretic peptide as an indicator of disease severity in a heterogeneous group of patients with chronic precapillary pulmonary hypertension. American Journal of Cardiology, 98, 525–529. , , , , & (
- 1970) Arginase activity in erythrocytes of healthy and ill children. Clinical Chimica Acta, 28, 391–396. , & (
- 2007) N-terminal fragment of the prohormone brain-type natriuretic peptide (NT-proBNP), cardiovascular events, and mortality in patients with stable coronary heart disease. Journal of the American Medical Association, 297, 169–176. , , , , & (
- 2003) Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension, 42, 1206–1252. , , , , , , , , , & (
- 1981) The kidneys in paroxysmal nocturnal hemoglobinuria. Blood, 57, 83–89. , , , & . (
- 1982) Pulmonary hypertension and cor pulmonale in the sickle hemoglobinopathies. American Journal of Medicine, 73, 814–821. & (
- 1998) Red-blood-cell augmentation of hypoxic pulmonary vasoconstriction: hematocrit dependence and the importance of nitric oxide. American Journal of Respiratory and Critical Care Medicine, 157, 1181–1186. , , , & (
- 2000) Effects of the RBC membrane and increased perfusate viscosity on hypoxic pulmonary vasoconstriction. Journal of Applied Physiology, 88, 1520–1528. , , & (
- 2001) Effects of S-nitrosation of hemoglobin on hypoxic pulmonary vasoconstriction and nitric oxide flux. American Journal of Respiratory and Critical Care Medicine, 163, 1164–1170. , , , & (
- 2002) Effects of S-nitrosation and cross-linking of hemoglobin on hypoxic pulmonary vasoconstriction in isolated rat lungs. Circulation Research, 91, 626–632. , , , , , , & (
- 1984) Intestinal infarction caused by paroxysmal nocturnal hemoglobinuria. American Journal of Hematology, 16, 75–81. , & (
- 2006) Serum N-terminal brain natriuretic peptide as a prognostic parameter in patients with pulmonary hypertension. Chest, 129, 1313–1321. , , , , , & (
- 2003) N-terminal pro-brain natriuretic peptide. A new gold standard in predicting mortality in patients with advanced heart failure. European Heart Journal, 24, 1735–1743. , , , & (
- 2004) Pulmonary hypertension as a risk factor for death in patients with sickle cell disease. The New England Journal of Medicine, 350, 886–895. , , , , , , , , , , , , , , & (
- 2005) NT-ProBNP: the mechanism behind the marker. Journal of Cardiac Failure, 11, S81–S83. (
- 1991) Effects of inhibitors of EDRF and EDHF on vasoreactivity of perfused rat lungs. American Journal of Physiology, 260, L97–L104. , , , & (
- 1998) Role of NO and endothelin in hemoglobin-induced pulmonary vasoconstriction. Shock, 10, 401–406. , , , , , & (
- 2005a) Nitric Oxide Consumption and Pulmonary Hypertension in Patients with Paroxysmal Nocturnal Hemoglobinuria. (Abstract). Blood, 106, 305a. , , , , , , , , , , , , , & (
- 2005b) Sustained response and long-term safety of eculizumab in paroxysmal nocturnal hemoglobinuria. Blood, 106, 2559–2565. , , , , , , & (
- 2005c) Improvement in the symptoms of smooth muscle dystonia during eculizumab therapy in paroxysmal nocturnal hemoglobinuria. Haematologica, 90, ECR40. , & (
- 2006) High Definition Contrast-Enhanced MR Imaging in Paroxysmal Nocturnal Hemoglobinuria (PNH) Suggests a High Frequency of Subclinical Thrombosis. (Abstract).Blood, 108, 292. , , , , , , , , & (
- 2008) Improvement in fatigue with eculizumab treatment of patients with paroxysmal nocturnal hemoglobinuria (PNH) occurs independent of changes in anemia. (Abstract).Haematologica, 93, 359. , , , , , , , , , , & (
- 1995) Natural history of paroxysmal nocturnal hemoglobinuria. The New England Journal of Medicine, 333, 1253–1258. , , , & (
- 2004) Effect of eculizumab on hemolysis and transfusion requirements in patients with paroxysmal nocturnal hemoglobinuria. The New England Journal of Medicine, 350, 552–559. , , , , , , , , , & (
- 2006) The complement inhibitor eculizumab in paroxysmal nocturnal hemoglobinuria. The New England Journal of Medicine, 355, 1233–1243. , , , , , , , , , , , , , , , , , , & (
- 2007) High Incidence of Progression to Chronic Renal Insufficiency in Patients with Paroxysmal Nocturnal Hemoglobinuria (PNH). (Abstract).Blood, 110, 1074. , , , , , & (
- 2001) A useful ELISA system for human liver-type arginase, and its utility in diagnosis of liver diseases. Clinical Biochemistry, 34, 455–461. , , , , , , , & (
- 2006) Lactate dehydrogenase as a biomarker of hemolysis-associated nitric oxide resistance, priapism, leg ulceration, pulmonary hypertension, and death in patients with sickle cell disease. Blood, 107, 2279–2285. , , , , , , , , , & (
- 2001) The prognostic value of B-type natriuretic peptide in patients with acute coronary syndromes. The New England Journal of Medicine, 345, 1014–1021. , , , , , , , & (
- 2004) Clinical significance of brain natriuretic peptide in primary pulmonary hypertension. Journal of the American College of Cardiology, 43, 764–770. , , , , , , , & (
- 1998) Natriuretic peptides. The New England Journal of Medicine, 339, 321–328. , & (
- 2006) N-terminal pro-brain natriuretic peptide levels and risk of death in sickle cell disease. The Journal of the American Medical Association, 296, 310–318. , , , , , , , , , , , , , & (
- 1989) Methylene blue potentiates vascular reactivity in isolated rat lungs. Journal of Applied Physiology, 66, 1040–1045. , , , , & (
- 2005) Hemolysis-associated endothelial dysfunction mediated by accelerated NO inactivation by decompartmentalized oxyhemoglobin. Journal of Clinical Investigation, 115, 3409–3417. , , , , , , , , & (
- 2003) Arginine therapy: a new treatment for pulmonary hypertension in sickle cell disease? American Journal of Respiratory and Critical Care Medicine, 168, 63–69. , , , , , , , & (
- 2004) Decreased arginine bioavailability and increased serum arginase activity in asthma. American Journal of Respiratory and Critical Care Medicine, 170, 148–153. , , , , & (
- 2005) Dysregulated arginine metabolism, hemolysis-associated pulmonary hypertension, and mortality in sickle cell disease. The Journal of the American Medical Association, 294, 81–90. , , , , , , , , & (
- 2004) Natural history of paroxysmal nocturnal haemoglobinuria using modern diagnostic assays. British Journal of Haematology, 126, 133–138. , , & (
- 1995) The effects of recombinant human hemoglobin on esophageal motor functions in humans. Gastroenterology, 109, 1241–1248. , , , , & (
- 1998) Plasma brain natriuretic peptide levels increase in proportion to the extent of right ventricular dysfunction in pulmonary hypertension. Journal of the American College of Cardiology, 31, 202–208. , , , , , , , , , , , , & (
- 1996) Enhancement of platelet deposition by cross-linked hemoglobin in a rat carotid endarterectomy model. Circulation, 93, 327–332. , , , , & (
- 1997) Clinical characteristics predict response to antithymocyte globulin in paroxysmal nocturnal haemoglobinuria. British Journal of Haematology, 96, 92–97. , , , , & (
- 1996) Phase I study of the safety and pharmacologic effects of diaspirin cross-linked hemoglobin solution. Critical Care Medicine, 24, 1993–2000. , , , , & (
- 1987) Endogenous nitric oxide inhibits human platelet adhesion to vascular endothelium. Lancet, 2, 1057–1058. , & (
- 2002) Cell-free hemoglobin limits nitric oxide bioavailability in sickle-cell disease. Nature Medicine, 8, 1383–1389. , , , , , & (
- 2000) Paroxysmal Nocturnal Hemoglobinuria. Hoffman, Churchill Livingstone, Philadelphia, PA, USA, pp. 331–342. (
- 2005) The clinical sequelae of intravascular hemolysis and extracellular plasma hemoglobin: a novel mechanism of human disease. The Journal of the American Medical Association, 293, 1653–1662. , , & (
- 2004) Rapid regulation of platelet activation in vivo by nitric oxide. Circulation, 109, 1819–1822. , , , , & (
- 2004) Evidence for a metabolic shift of arginine metabolism in sickle cell disease. Annals of Hematology, 83, 371–375. , , , , , & (
- 2008) Eculizumab, a terminal complement inhibitor, improves anaemia in patients with paroxysmal nocturnal haemoglobinuria. British Journal of Haematology, 142, 263–272. , , , , , , , , , , & (
- 1992) Hemolytic anemias. Diagnosis and management. The Medical Clinics of North America, 76, 649–668. (
- 1982) Vascular problems in paroxysmal nocturnal haemoglobinuria. A report of two cases. Blut, 45, 261–265. , , & (
- 1995) Nitric oxide-related vasoconstriction in lungs perfused with red cell lysate. The Journal of the Federation of American Societies for Experimental Biology, 9, 379–386. , , & (
- 2007) Pulmonary hypertension in patients with sickle cell/beta thalassemia: incidence and correlation with serum N-terminal pro-brain natriuretic peptide concentrations. Haematologica, 92, 738–743. , , , , & (
- 2004) Biological activity of nitric oxide in the plasmatic compartment. Proceedings of the National Academy of Sciences of the United States of America, 101, 11477–11482. , , , , , , & (
- 1990) Inherited complete deficiency of 20-kilodalton homologous restriction factor (CD59) as a cause of paroxysmal nocturnal hemoglobinuria. The New England Journal of Medicine, 323, 1184–1189. , , , , , , , , , , , and (