ALBERT

Description: Sickle cell disease (SCD) is a clinical syndrome that encompasses several different genotypes, the 3 most common being homozygosity for the bS allele (HbSS), compound heterozygosity of HbS and HbC (HbSC), and compound heterozygosity of HbS and HbSb thalassemia (HbSb+ or HbSb0 thalassemia). Generally, patients with HbSS and HbSb0 thalassemia genotypes have the most severe clinical manifestations, while patients with HbSC and HbSβ+-thalassemia are thought to be less severe. Within each of these genotypic groups, however, there are also substantial phenotypic differences. This heterogeneity makes it difficult to quantify the severity of the disease process and to guide therapeutics. As more intensive, high risk and costly treatments such as hematopoietic stem cell transplant and gene therapy are developing, the ability to assess patients at highest risk of early mortality becomes increasingly important. Integrating varied clinical, laboratory, and imaging markers for personalized risk prediction has been difficult, however, newer machine learning methods for outcome prediction take a more agnostic approach than traditional statistical methods and can detect complex, non-linear relationships in the data. In this study, we sought to apply machine learning methods to a well-characterized cohort of SCD patients followed at the National Institutes of Health in order to identify clinically meaningful subgroups of patients at highest risk of mortality. Between 2006 and 2017, 601 patients (age 35±13 years, 51% female) underwent echocardiogram, standard laboratory markers and hemoglobin electrophoresis resulting in 61 candidate variables. Among these patients, 488 had HbSS, 12 HbSb0 thalassemia, 80 HbSC, 20 HbSb+ thalassemia. All-cause mortality was ascertained by proxy interview, through medical records, and through the CDC National Death Index. Average follow-up time was 5 years and 130 patients were deceased. A random survival forest (RSF) algorithm followed by nested model selection and AIC Cox regression analysis identified 13 predictors of mortality (estimated right ventricular systolic pressure, peak tricuspid regurgitant (TR) velocity, mitral E velocity, septal and posterior wall thickness, IVC diameter, right atrial area, BUN, alkaline phosphatase, N-terminal-pro brain natriuretic peptide (BNP), creatinine, potassium and bicarbonate). This model performed better than individual clinical and laboratory variables with a C-statistic of 0.822 (genotype 0.524, eGFR 0.624, NT-proBNP 0.686, TR velocity 0.703). K-means clustering grouped all patients into 3 main clusters with significant survival differences. Survival at 8 years for the entire group was 70%; for individual clusters, survival was 43% for cluster 1, 72% for cluster 2, and 88% for cluster 3 (Figure 1A). Since TR velocity is recognized as one of the most specific independent predictors of mortality, we compared our results with this parameter. There was a better stratification of mortality risk using the 7 strongest parameters from RSF compared with TR velocity alone (Figure 1B), particularly for longer term outcomes. In this cohort of 601 patients with SCD, machine learning methods were used to show the heterogeneity of this disorder and the ability to detect phenotypic clusters with different mortality profiles. Although there are many individual predictors of mortality, few methods other than assessment by an expert clinician can integrate all known variables in deeply phenotyped patients. RSF and cluster analysis was used in this cohort to analyze a large amount of data in order to identify seven variables that could stratify patients into groups with significantly different outcomes. The specificity of this approach was high (c-statistic 0.822) and better than that of individual markers of end-organ involvement. Disclosures No relevant conflicts of interest to declare.

Description: Abstract 3240 The role of pulmonary hypertension as a common and attributable cause of mortality in patients with sickle cell disease remains controversial. To assess this question and explore risk factors for death in patients with sickle cell disease we evaluated 632 patients in the Walk-PHASST pulmonary hypertension screening cohort, recruited from nine different study sites in the United States and one site in the United Kingdom. Methods: Patient characteristics and their associations with mortality were analyzed with Cox proportional hazards regression analysis. Based on data from three right heart catheterization screenings studies that have recently been published, we defined the presence of pulmonary hypertension for this analysis by a Doppler-echocardiographic measurement of the tricuspid regurgitant jet velocity (TRV) ≥ 3.0 m/s, which has a 67–75% positive predictive value for a mean pulmonary artery pressure ≥ 25 mm Hg by right heart catheterization. This therefore represents a very conservative threshold for a large population screening study. Among subjects with a measurable TRV (n=572), 64 (11.2%) had measurements of ≥ 3.0 m/sec. Among those with measurable NT-proBNP (n=582), 140 (24.1%) had measurements ≥160 pg/mL, a value associated with both pulmonary hypertension and mortality. A total of 39 (7.4%) had both high TRV (≥3.0 m/sec) and high NT-proBNP (≥160 pg/mL). Results: Over a median follow-up time of 29 months, we observed 22 deaths. 50% (N=11) of these patients had a TRV≥ 3.0 m/sec. At 24 months the cumulative survival was 83% for patients with TRV ≥ 3.0 m/sec and 98% for patients with TRV 〈 3.0 m/sec (p

Description: Abstract 2658 Introduction: Six minute walk distance (6MWD), is a measure of exercise capacity commonly used as an endpoint in pulmonary hypertension (PH) clinical trials. Many patients with sickle cell disease (SCD) have acute pain crises or chronic pain syndromes that impair their quality of life. While patients with SCD who are undergoing screening for PH are generally screened in steady state, i.e., when they have not had a recent pain crisis, the impact of chronic pain on exercise capacity in this group of patients has not been previously evaluated. Methods: walk-PHaSST was a multi-center screening study designed to identify subjects with SCD at increased risk for symptomatic PH, defined by a tricuspid regurgitant velocity (TRV) ≥ 2.7 m/sec and 6MWD between 150–500 meters, for enrollment in a double-blind placebo controlled trial of sildenafil. The primary endpoint was the change in 6MWD after 16 weeks of treatment. We examined the relationship between subjects' self-reported acute and chronic pain and baseline 6MWD in the screened SCD patients in walk-PHaSST. Results: For 90% of subjects, the information about pain was reported by the patient or parent/family member. Documentation of pain management and utilization of services was verified from medical records in 10% of subjects. Ninety four percent of all subjects reported having a history of acute sickle cell pain crises; 6% reported never having had an acute pain crisis. For the subjects who reported a history of acute pain crises, the ‘typical’ acute pain rating on a scale of 0 to 10 was ≥ 7 (maximum 10) for 77% of this subset of subjects. A total of 342 (50%) subjects reported not having had any pain crises in the preceding week. Of 720 subjects screened medical history and 6 MWD was available in 673 patients. Of these 633 (94%) subjects did not report having had a pain crisis requiring an emergency department visit or hospitalization in the preceding week. A total of 39% of subjects reported chronic sickle cell related pain; no rating was reported for chronic pain. 88% of patients reported using medications for pain control while 15% reported using non-drug therapy including physical therapy in 3%, alternative therapy in 2%, acupuncture in 2% and hypnosis in 〈 1% of patients. The mean 6MWD for the screened population was 439 meters (median 438 m, range 123–713 m). A total of 171/673 (26%) subjects had a 6MWD 〉500 meters, which was above the screening cut-off for enrollment in the main interventional trial. By univariate analysis, subjects reporting chronic pain had a significant lower odds ratio for walking 〉 500 meters (OR 0.637, 95% C.I 0.44–0.99); a similar observation was seen with those subjects with a history of acute pain crises (OR 0.47, 95% C.I 0.24–0.91). Multivariable logistic regression analysis revealed a significant inverse relationship between chronic pain but not acute pain and 6MWD after adjusting for age, TRV, gender, hematocrit and smoking history (See Table 1). The mean 6MWD decreased by 27 meters with self reported chronic pain after adjusting for TRV, age, gender, hematocrit and 6MWD. Conclusions: TRV is a known predictor of 6MWD. However, these data suggest that patient self reported sickle cell related chronic pain is also an independent predictor of 6MWD. This relationship raises interesting questions about the potentially confounding effects of pain on exercise capacity as assessed by the 6MW test. Further study is warranted to investigate an association between chronic pain and exercise capacity in SCD as well as exploration of appropriate endpoints for future clinical trials in patients with SCD and suspected symptomatic PH. Disclosures: Barst: Pfizer: Consultancy, Research Funding. Rosenzweig:Pfizer: Research Funding. Badesch:Pfizer: Honoraria, Research Funding. Hassell:Novartis: Research Funding.

Description: Abstract 4804 Introduction: Smoking is known to promote vascular inflammation, in-vitro platelet aggregation and expression of endothelial adhesion molecules, processes that contribute to vasculopathy. Inflammation, abnormal platelet activation with thrombus formation and endothelial cell activation also play a role in vaso-occlusion in sickle cell disease (SCD). These overlapping pathobiological mechanisms suggest the possibility of a relationship between smoking and SCD vaso-occlusive pain. While small single center studies have suggested a link between environmental smoke exposure and hospitalization rate for acute chest syndrome and SCD pain (West et al 2003, Cohen et al 2010) there is a paucity of data derived from large multicenter studies about the interplay between smoking and pain phenotype in SCD. Aims: To determine the relationship between patient self reported chronic pain and history of current or former cigarette smoking in the SCD subjects screened in the walk-PHaSST study. Methods: Walk-PHaSST was a multi-center, placebo-controlled, double-blind 16-week trial designed to evaluate the safety and efficacy of sildenafil in patients with tricuspid regurgitant velocity [TRV] ≥2.7m/s and decreased exercise capacity as assessed by the six-minute walk distance (6MWD). We analyzed the data from all subjects screened for the walk-PHaSST trial. In the screening trial, subjects were evaluated by medical history, physical examination, laboratory screening, echocardiography and 6MWD testing. Univariate and stepwise multivariable logistic regression was used for this analysis. P value of

Description: In a prospective study started in 2001 echocardiography (ECHO) was used to screen 195 unselected adult sickle cell patients for pulmonary hypertension (PHTN). The prevalence of PHTN [tricuspid regurgitant jet velocity (TRV) of 2.5 m/sec or higher, corresponding to a pulmonary artery systolic pressure (PAs) of at least 35 mm Hg] was 32 %. Markers of hemolysis (low Hb, and high reticulocytes and LDH) were associated with PHTN risk. Although the pulmonary pressures in sickle-related PHTN were not as high as in other forms of PHTN, they were predictive of short survival (NEJM2004;350:886). We now report the TRV results in 113 patients who had repeat ECHO exams at 2.1 years (range 1.2–4.1 y) after study enrollment. Their mean age was 36.2 ± 10.3 y and 59 % were female. Their hemoglobin types were: SS, 77.7 %; SC, 14.3 %; Sickle-thal.+, 6.3 %; and Sickle-thal.0, 1.8 %. At baseline, 40 patients had PHTN (mean TRV 2.75 m/sec) and their mean TRV at the 2 y follow up was 2.71 m/sec (p=0.45). In 6 subjects the TRV decreased to normal values from baselines of 2.5 – 3 m/sec. Nevertheless, the mortality of the patients diagnosed with PHTN at baseline currently stands at 40% at 40 months of follow-up. Seventy-three patients did not have PHTN at enrollment (mean TRV 2.0 m/sec) and as a group their TRV at the 2 y follow up had not changed significantly (mean 2.1, p=0.1). However, 11 of the 73 patients (15.1%) with initially normal TRV developed PHTN at 2 years (mean TRV 2.70, range 2.5–3.5 m/sec). The table shows that age, sex, Hb type, Hb F %, and hydroxyurea treatment of the patients who developed PHTN at the 2 y follow up were not significantly different from those who did not. However, patients who developed PHTN were significantly more anemic, had higher reticulocyte counts, and higher serum ferritin values. A trend toward an increased bilirubin and LDH was observed. Hence, hemolysis probably played a pathogenetic role in these patients’ increased pulmonary pressures, as is also postulated for sickle patients who had PHTN at baseline. These preliminary results suggest that the incidence of PHTN in adult sickle patients could be as high as 7 % per year and that ECHO screening of these patients every two years is probably indicated. No PHTN PHTN P value N 62 11 Mean age, y (±SD) 34.1 ± 8.9 34.6 ± 13.4 0.85 Sex (% female) 56 64 0.75 Non-SS (%) 27 18 0.5 On hydroxyurea (%) 41 25 0.5 Hb (g/dl, mean±SD) 9.98 ± 1.74 8.68 ± 1.91 0.04 Hb F (%,mean±SD) 7.9 ± 6.8 7.5 ± 7.4 0.869 Abs. Retics (x1000/mm3, mean±SD) 205 ± 107 288 ± 151 0.045 LDH (U/L, mean±SD) 286 ± 107 359 ± 179 0.09 Bilirubin (mg/dl, mean±SD) 2.5 ± 1.88 3.7 ± 2.58 0.10 Ferritin (mg/L, mean±SD) 632 ± 925 1439 ± 1849 0.046 Creatinine (mg/dl, mean±SD) 0.76 ± 0.23 0.69 ± 0.25 0.36

Description: Abstract 1023 Introduction. Pulmonary hypertension (PH) is a complication associated with thalassemia syndromes, particularly thalassemia intermedia. There are limited data on the safety and efficacy of selective pulmonary vasodilators in this at-risk population. Methods. We evaluated the safety and efficacy of a 12-week prospective, phase 1/2 pilot scale study of sildenafil (100 mg TID) in ten b-thalassemia patients with elevated TRV on Doppler-echocardiography ≥ 2.5 m/s suggestive of PH risk. Patients were evaluated at baseline and at weeks 2, 4, 8 and 12 of sildenafil therapy, and six minute walk distance (6MWD), biomarkers of hemolysis, coagulation, inflammation and adhesion were assessed. Results. Our study population had an average age of 37±12.3 years, 8/10 were male, and 50% were thalassemia intermedia. Splenectomy prevalence was high (90%), and only 30% of patients were transfused since infancy. The mean pre-transfusion hemoglobin was 10.4±1.5 g/dL. A 12-week open-label trial resulted in a significant decrease in TRV by 13.3% (3.0±0.7 vs. 2.6±0.5m/s, p=0.04, Figure 1), improvements in left ventricular end systolic/diastolic volume (p≤0.02), diffusion capacity for carbon monoxide (DLCO, p=0.003) and a trend towards a reduced Borg Dyspnea Score and improved NYHA Functional class. No significant change in 6MWD was noted, although 6MWD correlated strongly with DLCO (ρ=0.72, p=0.03) suggesting that oxygen diffusion across the alveolar-capillary membrane was an important determinant of exercise capacity. Sildenafil was generally well tolerated, but most patients experienced anticipated transient headaches and visual/color disturbances associated with sildenafil use. One patient withdrew from the study due to worsening dyspnea. No other serious adverse events were reported. A strong direct correlation between total dose of sildenafil (mg) taken and % change in plasma NO metabolite concentration was observed (ρ =0.80, p=0.01). A significant increase in plasma and erythrocyte arginine concentration occurred, without an associated change in plasma arginase activity/concentration, nitric oxide metabolites or vascular endothelial growth factor. However arginase concentration was elevated in this cohort similar to prior reports, and correlated inversely to hemoglobin (ρ=-0.41, p=0.01), and directly to ALT (ρ=0.40, p=0.004), AST (ρ=0.38, p=0.04), left ventricular end systolic volume (ρ=0.77, p=0.001), and end-diastolic volume (ρ=0.79, p=0.001). Conclusion. Our study suggests that sildenafil is safe and may improve cardiopulmonary hemodynamics in patients at risk for PH, however improved exercise capacity as reflected by an improved 6MWD was not observed. The reduction in left ventricular dimensions is promising, and could reflective of either increased inotropy or chronotropy, or decreased systemic afterload. This is also the first report of an influence of sildenafil on diffusion capacity of the lungs in patients with thalassemia and the first description of increased plasma and erythrocyte arginine concentration after sildenafil therapy. Given the association of arginine bioavailability with long-term survival in cardiovascular disease, this is an unexpected effect of sildenafil that warrants further investigation. These data support the need for further clinical trials evaluating the use of sildenafil in thalassemia. Disclosures: Taher: Novartis: Research Funding, Speakers Bureau.

Description: In adults with sickle cell disease (SCD), an increased tricuspid regurgitation velocity (TRV) by Doppler echocardiography is associated with increased morbidity and mortality. Although sildenafil has been shown to improve exercise capacity in patients with pulmonary arterial hypertension, it has not been evaluated in SCD. We therefore sought to determine whether sildenafil could improve exercise capacity in SCD patients with increased TRV and a low exercise capacity. A TRV ≥ 2.7 m/s and a 6-minute walk distance (6MWD) between 150 and 500 m were required for enrollment in this 16-week, double-blind, placebo-controlled sildenafil trial. After 74 of the screened subjects were randomized, the study was stopped early due to a higher percentage of subjects experiencing serious adverse events in the sildenafil arm (45% of sildenafil, 22% of placebo, P = .022). Subject hospitalization for pain was the predominant cause for this difference: 35% with sildenafil compared with 14% with placebo (P = .029). There was no evidence of a treatment effect on 6MWD (placebo-corrected effect −9 m; 95% confidence interval [95% CI] −56-38; P = .703), TRV (P = .503), or N-terminal pro-brain natriuretic peptide (P = .410). Sildenafil appeared to increase hospitalization rates for pain in patients with SCD. This study is registered at www.clinicaltrials.gov as NCT00492531.

Description: Abstract 1074 Non-Cardiopulmonary Factors Affecting the Six-Minute Walk Distance in Patients with Sickle Cell Disease: Results from the Walk-PHaSST Study. INTRODUCTION: The six-minute walk (6MW) test is frequently used to assess exercise capacity. Patients with sickle cell disease (SCD) can have decreased 6MW distance (6MWD) compared to controls. The 6MWD in conjunction with the TR-jet velocity (TRV) and NT-proBNP have recently been proposed to have a greater predictive value for screening SCD patients suspected of having pulmonary hypertension (PH) than TRV alone. (Parent et al, NEJM, 365; 1, 2011 365 (1):44–53). The American Thoracic Society guidelines recommend caution in controlling for sources of variability in the 6MWD (Am J Respir Crit Care Med 166. 111–117, 2002). Age and height are known confounders of the 6MWD. However, non-cardiopulmonary factors including skeletal-mechanics and pain may also impact the 6MWD. AIM: This study explores whether non-cardiopulmonary factors affect the 6MWD in SCD patients. METHODS: We analyzed data from subjects screened for the walk-PHaSST trial. Walk-PHaSST was a multi-center, placebo-controlled, double-blind, 16-week trial evaluating the safety and efficacy of oral sildenafil for the treatment of Doppler-defined PH (TRV '2.7m/s) in subjects with SCD aged 〉12 years. The primary endpoint in the trial was change in 6MWD. During screening, subjects were evaluated by self-reported medical history, physical examination, blood sampling, echocardiography and 6MWD. Univariate and multivariable linear regression was performed. A two sided p value

Description: Individuals with SCD have recurrent episodes of ischemia-reperfusion in vital organs, and abnormal microcirculation is considered a major determinant of end organ damage. Hydroxyurea (HU) treatment has been shown to produce ATP- and nitric oxide-mediated vasodilation, and to improve red cells deformability Few tools are available to quantitatively measure microvascular blood flow in vivo. The purpose of this study was to assess the ability of a novel ECHO based imaging modality to assess the effect of hydroxyurea on cardiac and skeletal muscle perfusion in patients with sickle cell anemia. Methods and Results: Twenty-one HBSS/B0 patients, of whom 15 were treated with HU(average daily dose of 18 ± 8 mg/kg/day), underwent brachial artery ultrasound, echocardiography and contrast enhanced ultrasound (CEU) perfusion imaging of deep flexor muscles of the forearm as well as the myocardium at rest and during vasodilator stress with regadenoson ( ClinicalTrials.gov NCT016028090). Quantitative image analysis was performed to obtain microvascular blood volume and flow velocity measurements. Patients on HU had a lower white blood cell count (6.7±2.1 K/uL vs 10.0±2.1 K/uL, p

Description: Abstract 89 Background: Patients with sickle cell disease (SCD) and pulmonary hypertension (PH) have increased mortality. Whether SCD-associated PH (SCD-PH) is predominantly a pulmonary pre-capillary (PAH) vs. post-capillary (PVH) process is an area of ongoing investigation. SCD-PH is often complicated by high cardiac output (CO) related to anemia. The transpulmonary gradient (TPG) represents a pressure differential across the pulmonary vascular bed that the right ventricle (RV) has to overcome regardless of the total volume being ejected and has been employed to determine eligibility for cardiac transplantation because it eliminates the confounding effect of CO (PVR=TPG/CO). Typically, a TPG ≥ 12 mmHg indicates significant PAH with increased risk of acute RV failure post-transplantation. With significant PAH, there is often morphologic adaptation by the RV. MRI enables accurate quantification of RV function and structure. In idiopathic PAH, RV dilation and decreased function have been correlated with poor prognosis. Thus, we hypothesize that patients with SCD and a TPG ≥ 12 mmHg would have lower functional capacity, increased mortality, and MRI evidence of RV dysfunction. Materials & Methods: Five hundred and thirty one consecutive patients (age 35.6 ± 12.6, 54% (n=284) female, 73% (n=387) HbSS phenotype) with SCD were prospectively screened for PH using echocardiography (tricuspid regurgitant jet ≥ 2.5 m/s) without any exclusion criteria. Eighty four patients (age 41 ± 13, 55% (n=46) female, 82% (n=69) HbSS phenotype, mean Hb 8.8 ± 1.7) underwent RHC, and forty-one patients (age 42 ± 15, 54% (n=22) female, 80% (n=33) HbSS phenotype, mean Hb 8.9 ± 1.8) underwent cardiac MRI (CMR) within one week of RHC. CMR sequences consisted of cine imaging and late gadolinium enhancement imaging. Results: Of the 84 catheterized patients, forty-six had a TPG ≥ 12 mmHg and 38 had a TPG 〈 12 mmHg. Of the 41 patients who had both RHC and CMR, twenty-one had a TPG ≥ 12 mmHg and 20 had a TPG 〈 12 mmHg. Those with a TPG ≥ 12 mmHg had higher mortality (median years from enrollment to death were 4.0 vs. 4.7 years with 21 (46%) deaths in the high TPG group vs. 6 (16%) deaths in the lower TPG group, p=0.008), poorer functional class (p=0.007), shorter 6-minute walk distance (p=0.003), higher pulmonary vascular resistance index (p