Biological Parameters Predictive Of Percent Dense Red Blood Cell Decrease Under Hydroxyurea Therapy

Introduction Dehydrated, dense red blood cells (DRBC), a subpopulation of sickle cells, are characterized by density >1.112; their increased mean corpuscular hemoglobin concentration (MCHC) leads to hemoglobin S (HbS) polymerization. The %DRBC is the biological parameter most strongly associated with some sickle-cell–disease (SCD) chronic organ damage, e.g., renal dysfunction, leg ulcers and priapism, also called hemolytic subphenotypes (Bartolucci et al. Blood 2012; Kato et al. Blood Rev 2007). Proven hydroxyurea (HU) efficacy against SCD lowers vaso-occlusive crisis, acute chest syndrome frequencies, and mortality (Charache et al. N Engl J Med 1995). The classical biological parameters indicating HU response are fetal hemoglobin (HbF) and mean corpuscular cell volume (MCV) increases (Steinberg et al. Blood 1997). However, we previously found decreased %DRBC in 33 patients after 6 months of HU (Bartolucci et al. Blood 2012) .We analyzed baseline biological parameters to identify those predictive of %DRBC decline under HU. Patients and methods We conducted a monocenter, prospective, longitudinal study on SCD patients undergoing HU therapy. Data were collected at baseline (day 0) and after 6 months of HU. Inclusion criteria were: SS and S-β0 thalassemia patients, age >18 years. Non-inclusion criteria were pregnancy, chronic blood transfusion and refused consent. Biological parameters determined were: %DRBC assessed with the phthalate density-distribution technique, D50 (defined as the density at which [(height of cells below phthalate index/sum of those above and below that index) = 0.5]), white blood-cell count (WBC count), MCV, MCHC, mean corpuscular hemoglobin content (MCH), total Hb, reticulocytes (RET), %HbF, platelet count, total bilirubin, alanine aminotransferase, aspartate aminotransferase, lactate dehydrogenase. Results are expressed as means ± SD, numbers or %, as appropriate. Quantitative parameters were compared between groups with Student’s paired t-test. Correlations were established with Pearson’s correlation coefficient. Multiple linear regressions were used to explore models that better predicted DRBC variation under HU. The final models included the variables that remained significantly associated with %DRBC decline after adjustment for the other variables in the models. R-squared (r2) were used as measures of variance explained by the models. P < 0.05 defined significance. This study was approved by the local Institutional Review Board. Results and Discussion Fifty-nine patients, mean age 35 ± 9 years, were included. Their %DRBC fell significantly by 40.7% after 6 months of HU therapy (P = 0.0003) from 10.1 ± 8% to 6 ± 4%. The %HbF rose from 7.2 ± 4% to 17.3 ± 8%. Our univariate analysis identified variables significantly correlated with %DRBC (Table). Multivariate analysis retained a significantly positive correlation between %DRBC decreases under HU and the pretreatment %DRBC on day 0. This statistical model accounted for 71.9% of the variability of %DRBC decline under HU. Pertinently, no correlation was found between %HbF and %DRBC changes, suggesting different mechanisms of action. Conclusion Our results confirmed the HU impact on %DRBC decrease, suggesting new indications to prevent or treat SCD complications associated with high %DRBC. They also showed that the major parameter predictive of DRBC decline under HU was the baseline %DRBC. %HbF and %DRBC changes under HU were not correlated. Prospective studies are needed to analyze the therapeutic effects of HU on chronic complications. Disclosures: No relevant conflicts of interest to declare.