Fetal Hemoglobin Decrease during Voxelotor Treatment

Sickle cell disease (SCD) is an inherited disorder caused by a point mutation in the β-globin gene leading to the formation of hemoglobin S (HbS). The polymerization of deoxygenated HbS is the first event in the molecular pathogenesis of SCD resulting in sickling of red blood cells (RBC), and hemolytic anemia which has been shown to be associated with organ damage and early death. Hemolytic anemia is defined by an accelerated rate of destruction RBC resulting in a lowered Hb level and a reduced gas transport capacity. Because Hb S in its oxy (R) conformation do not polymerize, drugs modifying hemoglobin (Hb) to increase the proportion of oxy-HbS have been propose to inhibit HbS polymerization. Voxelotor is such a molecule, which allosterically modifies Hb-oxygen affinity improving anemia and reducing hemolysis. However, the impact of Voxelotor on HbF levels is unknown. We describe here variations of %HbF and Mean corpuscular HbF (MCHbF) in a cohort of homozygous SCD patients treated with Voxelotor in our Sickle Cell Referral center.

We conducted a prospective register study to analyze clinical and biological evolution of SCD patients treated with Voxelotor at 1500mg taken orally once daily. From December 2021 through December 2022, 25 SCD adult patients were followed during a median of 7 months [1-12]. Among them, HbF values were analyzable for 15 patients (SS gentype n=14; women in 73%; median age of 50 years [24-61]) who completed 6 months of Voxelotor with no transfusion during the same period; and with a stable dose of hydroxyurea (HU), for at least 3 months before inclusion and during the study in case of patients treated by HU (n=10). The dose of Voxelotor was stable for 13/15 patients and was decreased in 2 patients due to skin allergy (n=1) and moderate digestive intolerance. One patient died after 6 months of treatment in a context of end-stage renal failure and severe pulmonary hypertension not related to Voxelotor treatment. Mean corpuscular HbF (MCHbF) was calculated (MCHbF (pg) = MCH x %HbF). HbF percentage was measured by high performance liquid chromatography (Biorad, Variant II, beta-thal dual program). All participants provided written informed consent (CPP Erythropedie).

From baseline to 6 months of follow up mean Hb (g/dL) concentration increased from 7.4 [5.4-8.5] to 8 [5-11.9] (p=0,0003; Figure A); absolute reticulocyte count (ARC ; Giga/L) decreased from 158 to 145 (p=0,005), %HbF decreased by 63% in all patients from 7,1% to 3,3% (p=0,008; Figure B) and MCHbF (pg) decreased from 3,36 to 1,03 (p=0,001; Figure D). A decrease in % HbF was observed from the first 15 days of treatment in both groups of patients treated or not with HU (Figure C) ; no changes in mean corpuscular volume (p=0.15) was observed in patients treated with HU during follow up.

The HbF decrease in SCD patients treated with Voxelotor could be explained by different mechanisms:

- Reduction of stress erythropoiesis which is due to an increase in erythroid output in response to anemic stress associated with a reduced expression of BCL11A, a transcriptional repressor that inhibits adult-stage HbF expression. This hypothesis is supported by the reduction in reticulocytes in patients treated with Voxelotor

- The loss of an HbF-related survival advantage for SCD RBCs, while the normal RBC lifespan is approximatively 120 days, in SCD RBCs it is only a sixth of it.

HbF is a major modulator of SCD and prolonge SCD RBCs survival. Its protective effect relies on its quantity and distribution among the RBC population, which is often heterogeneous in SCD patients. Protection against polymerisation of Voxelotor is probably more homogeneous among RBCs. As a result, RBCs containing low levels of HbF, which would normally disappear, have a prolonged lifespan. Similarly, it was presumed that sickling of erythroblasts could lead to ineffective terminal erythroid differentiation with an anti-apoptotic role for HbF during the terminal stages of erythroid differentiation. However we found no significant difference for index of ineffective erythropoiesis (soluble transferrin receptor / absolute reticulocyte count) before and after 6 months of treatment.

The question raised is therefore that of the risk of a rebound effect upon discontinuation of Voxelotor, which has a short half-life, during the time it takes for HbF to increase. To better study this hypothesis, distribution of HbF/RBC will be studied in the HEMOPROVE trial (NCT05199766) currently in progress.