Pomalidomide Augments Erythropoiesis and Fetal Hemoglobin Production in a Humanized Mouse Model of Sickle Cell Disease

Introduction: Pomalidomide (PL) is an IMiD^™ immunomodulatory compound. Thalidomide and lenalidomide have previously been shown to mitigate blood transfusion dependency in patients with thalassemia major and myelodysplastic syndromes, respectively.^1,2 In a recent in vitro study, PL, which is currently being evaluated for the treatment of hematological cancers, stimulated erythropoiesis, F-cell production, total hemoglobin (Hb) and fetal hemoglobin (HbF) synthesis in human CD34⁺ cells.³ Furthermore, combined treatment of CD34⁺ cells with PL and Hydroxyurea (HU) synergistically upregulated HbF expression. These results suggest a promising role for PL in the treatment of β-hemoglobinopathies. To evaluate the in vivo efficacy of PL in sickle cell disease (SCD), we conducted an eight week study in a relevant knockout-transgenic (KT) mouse model. ⁴ 

Methods:Animals. Six week old KT homozygous sickle mice were treated daily (Mon–Fri; i.p. injections) for eight weeks with the following compounds: Grp 1. Vehicle (n=8); Grp 2. PL (10 mg/kg; n=9); Grp 3. HU (100 mg/kg; n=7); Grp 4. PL+ HU (n=8). Mice were maintained in an accredited pathogen-free animal facility according to NIH and institutional guidelines. Mice were anesthetized with Ketamine/Xylazine and blood collected by intracardiac puncture into 0.5 ml vacutainer EDTA tubes. Complete blood count (CBC) was analyzed with the CBC-Diff^™ Veterinary Hematology System (Heska Inc., Loveland, CO). Reticulocyte counts were determined by supravital staining with methylene blue. HbF analysis of mouse hemolysates was done by HPLC using a weak cation-exchange column SynChropak CM-300 (Elchrom INC, Darien, IL). Organ analysis (liver, spleen, and femur marrow) included organ weights (%bw) and histology of H&E paraffin sections.

Statistical analysis. One-Way ANOVA followed by Student-Newman-Kuels (Sigma Stat). Data are reported as the mean ± SE. A P-value of < 0.05 was considered significant.

Results:

1. Activity level and social behavior were unaffected by the treatment groups. There were no physical signs of drug toxicity and weight gains were comparable for all groups.

2. PL significantly augmented HbF expression comparable to HU (HbF [peak%]: Veh: 6.24±0.35; PL: 9.51±0.83, P<0.01; HU: 10.54±0.77, P<0.01/total Hb [gm/dl]: Veh: 7.16±0.38; PL: 7.67±0.33; HU: 6.09±0.79). Surprisingly, PL-induced HbF production returned to control values after combinatory therapy with HU (HbF [peak%]: Veh: 6.24±0.35; PL+HU: 7.14±0.32). In contrast to HU, PL enhanced erythropoiesis as evidenced by increases in total RBCs, reticulocytes, and spleen weight (RBCs [×10⁶/μl]: Veh: 4.84±0.14; PL: 5.49±0.19, P<0.05; HU: 4.47±0.51/Retics [%]: Veh: 36.81±5.78; PL: 40.47±4.73; HU: 16.41±3.94, P=0.01/Spleen weight [%bw]: Veh: 4.07±0.27; PL: 4.92±0.2, P=0.02; HU: 2.75±0.25, P<0.01). The total WBC count was largely unaffected by PL, but decreased significantly with HU (WBC [×10³/μl]: Veh: 14.43±2.74; PL: 11.62±2.62; HU: 6.85±0.82, P<0.05).

3. The PL group showed a trend toward increased marrow hyperplasia. A detailed analysis of cell density, erythroid and myeloid cells is underway. Liver histology revealed decreased tissue inflammation and focal necrosis in ~50% of PL-treated animals.

Summary & Conclusions: 1. PL caused a robust induction of HbF in sickle mice. These results extend earlier in vitro findings of PL’s bioactivity in human erythroid progenitor cells. The HbF response to PL was similar to the established HbF-inducer HU, but surprisingly was lost in combination with HU. This inhibitory effect suggests that PL and HU stimulate HbF by distinct mechanisms. 2. Additionally, PL augmented erythropoiesis as demonstrated by increases in total RBCs, reticulocytes, spleen weight, and marrow hyperplasia. In contrast, HU showed the expected findings of generalized myelosuppression. 3. PL reduced liver injury and inflammation in ~50% of treated animals. Liver preservation tended to be associated with higher HbF values, though PL’s immunomodulatory properties may have contributed to this outcome. 4. In summary, PL exhibits a highly favorable hematological profile in a humanized mouse model of SCD. These results warrant further investigation in a Phase I trial of patients with SCD.