Topic: 26. Sickle cell disease Background: Sickle cell disease (SCD) is a chronic hemolytic anaemia characterized by inflammation and vaso-occlusive crises (VOC). Hemolysis results in the release of hemoglobin (Hb) into the circulation, which is then oxidized into cell-free heme. Heme can damage endothelial cells and promote complement activation, contributing to vasculopathy (Merle et al., 2019; Belcher et al.,2014; Ghosh et al.,2013) and sickle cell nephropathy (SCN). To date, methods to dose plasmatic heme are mostly inaccurate, due to the Hb signal leading to overestimation. As both an actor of the pathophysiology and a potent biomarker of hemolysis, accurate plasmatic heme dosage is of great interest for SCD. Aims: Investigate relationships between heme concentration measured using our new method and other routinely used biomarkers of anemia, hemolysis and SCN severity in SCD patients. Methods: 102 steady state SCD patients (> 1 month after VOC, > 3 months after red blood cell transfusion) treated (n=60) or not treated (n=42) with hydroxycarbamide (HU) and who gave their written informed consent were included in the study. Plasmatic heme and hemopexin (HPX) were measured by spectrophotometry followed by mathematical conversion of the signal by use of reference spectra of the different species and chemical modifications of the iron redox and ligation states (Kiger et al.,2019). SCN severity was assessed by urine microalbumin to creatinine ratio (ACR). Patients were separated into three groups according to ACR value: normal (ACR≤3 mg/mmol); microalbuminuria (330 mg/mmol). Correlations were calculated using Spearman test, comparisons between HU-treated, or non-treated patients with Mann-Whitney test and comparisons between ACR groups with Kruskal-Wallis test. Results: Heme was associated with anemia biomarkers. We calculated significant negative correlations between heme and Hb (r = -0.36; p < 0.005) and hematocrit (r = -0.5; p < 0.001). As a modulator of SCD clinical severity, foetal Hb percentage was found to be negatively correlated with heme (r = -0.30; p < 0.01). We found significant and positive correlations between heme concentration and lactate dehydrogenase (r = 0.65; p < 0.001), total bilirubin (r = 0.41; p < 0.001), aspartate aminotransferase (r = 0.38; p < 0.001) and reticulocyte % (r = 0.44; p < 0.001) (Figure 1A). In this cohort, a significant positive correlation was found between heme and ACR (r = 0.34; p = 0.001) and the difference in heme concentrations between the three groups was statistically significant (p < 0.009, median: 1.5, 3.5 and 4.5 µM, for normal, micro- and macroalbuminuria groups, respectively) as for HPX (p < 0.001, median HPX concentration: 0.31, 0.12 and 0.11 g/L). LDH was significantly different (p<0.015) between groups but less discriminatory between micro and macro albuminuria groups (median: 365, 493 and 447 U/l) (Figure 1B). Finally, we measured significantly lower heme levels in patients treated with HU compared to non-treated patients (median: 1.5 µM vs. 3.4 µM; p < 0.005), as well as higher HPX levels (median: 0.25 vs. 0.14 g/L; p <0.005). Summary/Conclusion: Our results indicate the complexity of determining a precise threshold of hemolysis and its impact on organ damage for clinical use in SCD, based on the measurement of the routinely used biomarkers. Thus, as a reliable biomarker and a mediator of both hemolysis and SCN, accurate quantification of cell-free heme should be assessed for all patients suffering from hemolytic anemia. Our observations also highlight the importance of therapies aiming at decreasing cell-free heme in SCD patients to prevent kidney damage.Keywords: Hemoglobin, Hemolysis, Sickle cell disease, Heme

P1431: ACCURATE CELL-FREE HEME DOSAGE IN PLASMA CORRELATES WITH ANEMIA, HEMOLYSIS AND SICKLE CELL NEPHROPATHY SEVERITY