Combining Fetal Hemoglobin and Reticulocytosis to Index Clinical Severity of Sickle Cell Disease in Children.

Abstract 2556

Poster Board II-533

Predictors of disease severity during infancy or childhood in patients with sickle cell disease (SCD) are needed to guide treatment decisions with therapies that have known toxicities [transfusion, hydroxyurea (HU), bone marrow transplant]. Erythrocyte fetal hemoglobin (HbF) expression levels above 20% reduce sickle hemoglobin (HbS) polymerization and decrease hemolysis. As a result of the decreased hemolysis, the survival of erythrocytes is prolonged, and the overall level of erythropoiesis is reduced. To determine if clinical markers of increased HbF production and decreased erythropoiesis may be combined to score disease severity, we developed a Fetal Hemoglobin-Reticulocytosis Index (FRI) defined as: [HbF (%) × non-transfused F-cells (%)] / [Absolute Reticulocyte Count (K/uL)]. For these studies, red cell lysates were analyzed by high power liquid chromatography (HPLC) to estimate HbA, HbS, and HbF fractions. F-cells were analyzed by flow cytometry using antibodies directed against HbF, while transfused cells were labeled with antibodies directed against HbA. Dual staining with both antibodies provided a method for accurately distinguishing transfused and non-transfused F-cells (NT F-cells). A minimum of 10,000 cells was analyzed in all samples. Absolute reticulocyte counts (ARC) were determined using a Sysmex XE 2100 hematology analyzer (Sysmex America, Mundelein, IL). Preliminary studies revealed FRI values near 100 at one month of age followed by a rapid drop before the age of 4 years. Blood from children between the ages of 4 and 21 years was also studied to determine if FRI correlates with therapeutic regimen. FRI values for three groups were compared: those treated with chronic transfusion (n=19, mean FRI=0.72±1.04), HU (n=19, mean FRI=5.61±6.24), versus supportive care alone that did not include recent transfusions (n=42, mean FRI=2.70 ±4.85). When the FRI values from each of these groups were placed in rank order, the slope of the line increased sharply from a linear to an exponential shape near the FRI value of 2. To determine if the FRI=2 inflection may be indicative of reduced disease severity, the number of SCD events were determined in the 42 study subjects treated with supportive care. Overall, twenty-eight (66.7%) patients had an FRI<2, and fourteen (33.3%) patients had an FRI≥2. Among those patients, SCD events were tallied (listed in descending order according to number of events): painful crises requiring hospitalization (FRI<2, n=128; FRI≥2, n=25), pneumonia /acute chest syndrome (FRI<2, n=74; FRI≥2, n=18), splenic sequestration (FRI<2, n=14; FRI≥2, n=0), conditional transcranial Doppler [(TCD), FRI<2, n=13; FRI≥2, n=1), silent stroke (FRI<2, n=4; FRI≥2, n=2), bacteremia (FRI<2, n=2; FRI≥2, n=1), cholecystectomy (FRI<2, n=3; FRI≥2, n=0), and nephropathy (FRI<2, n=1; FRI≥2, n=0). None of the supportive care group had an overt stroke, abnormal TCD, sickle cell retinopathy, or priapism. Age adjusted analysis showed that the FRI≥2 group had significantly fewer total events per year [events/year: FRI<2 (0.70±0.52) vs. FRI≥2 (0.38 ± 0.36), p=0.02]. These data suggest that combining the clinical parameters of fetal hemoglobin production and reticulocytosis provides a simple index for SCD severity. Based upon this retrospective data, prospective studies are underway to determine if the FRI decline during infancy or FRI levels in childhood are useful to predict clinical severity and treatment decisions in SCD patients.