Red Blood Cell Deformability and Morphological Dynamic Sickling As a Metric to Monitor Pharmacodynamic Response to Voxelotor

The recent introduction of new therapies, such as hemoglobin (Hb) oxygen affinity modifiers, fetal Hb (HbF) inducers, and gene therapy (replacement of HbS with non-sickling Hb), that affect the Hb polymerization rate for sickle cell disease (SCD), highlight the need for new metrics to assess changes in sickling characteristics as therapeutic response. Primary outcomes in clinical trials often focus on indirect clinical improvement i.e. decrease in vaso-occlusive events (VOC), Hb increase, fatigue and pain score. There is currently a lack of direct biomarkers to assess the pharmacodynamic impact of these therapies. Two methods have been developed to assess downstream Hb polymerization, LoRRca (oxygenscan), a commercial oxygen gradient ektacytometry that measures deformability as a function of oxygen tension (pO ₂), and dynamic sickling assay (DSA), an enzymatic oxygen scavenging system that directly assesses morphological sickling. In the present study, we investigate the relationship between red blood cell (RBC) deformability and kinetics of sickling morphology after a treatment with different doses of voxelotor (GBT440, Global Blood Therapeutics), the only FDA approved Hb oxygen modifier.

Blood samples from eight SCD patients were collected following the IRB #041718MP2E & IRB #Pro00066220 protocols. RBCs were incubated with voxelotor at five different doses based on sample hematocrit (Hct): 25% (0.32 ± 0.06 mM), 50% (0.65 ± 0.11 mM), 75% (0.97 ± 0.17 mM), 100% (1.29 ± 0.23 mM) and 125% (1.62 ± 0.28 mM), or with an equal volume of DMSO-containing buffer (vehicle).

Oxygenscan is performed (LoRRca, RR Mechatronics, The Netherlands) as per manufacturer's instructions. The following parameters are calculated: i) elongation index (EImin), measured at minimum RBC deformability, ii) point of sickling (PoS), the pO ₂ at which RBC deformability decreases below 5% of EImax during deoxygenation and iii) ΔEI, the difference between EImin and EImax.

DSA uses an enzymatic reaction that allows for tight control of the rate and depth of hypoxia, resulting in RBC morphological changes observed through time-lapse photography. Sickling profiles were constructed to represent: i) morphological points of sickling at 50% of induced fraction (mPoS _i@50%), ii) rate of sickling at 50% of induced fraction (rate _i@50%) and iii) area under the curve at 15 minutes (AUC ₁₅).

Paired t-test analysis of the LoRRca elongation curves show a significant dose response in Elmin (Vehicle: 0.16 ± 0.16; 25%: 0.20 ± 0.16, p=0.0033; 50%: 0.23 ± 0.15, p=0.0001; 75%: 0.27 ± 0.15, p=0.0004; 100%: 0.30 ± 0.14, p=0.0009; 125%: 0.35 ± 0.13, p=0.0015), ΔEl (Vehicle: 0.26 ± 0.11; 25%: 0.22 ± 0.11, p=0.0016; 50%: 0.18 ± 0.11, p=0.0003; 75%: 0.15 ± 0.08, p=0.0002; 100%: 0.11 ± 0.06, p=0.0004; 125%: 0.09 ± 0.07, p=0.0011) and PoS (Figure 1; Vehicle: 35 ± 11 mmHg; 25%: 32 ± 6, p=0.0092; 50%: 28 ± 7, p=0.0038; 75%: 25 ± 6, p<0.0001; 100%: 17 ± 5, p=0.0007; 125%: 14 ± 4, p=0.0015). These findings indicate improvements in RBC deformability under hypoxic conditions when the drug is present.

Furthermore, paired t-test of DSA data showed a significant dose response with AUC ₁₅ (Vehicle: 958 ± 104; 25%: 909 ± 105, p=0.0061; 50%: 845 ± 146, p=0.0004; 75%: 768 ± 194, p=0.0014; 100%: 672 ± 179, p<0.0001; 125%: 623 ± 192, p=0.0002), rate _i@50% (Vehicle: 0.87 ± 0.26; 25%: 0.77 ± 0.30, p=0.1236; 50%: 0.61 ± 0.29, p=0.0003; 75%: 0.33 ± 0.13, p<0.0001; 100%: 0.28 ± 0.10, p<0.0001; 125%: 0.25 ± 0.09, p=0.0004) and mPoS _i@50% (Figure 2; Vehicle: 4.4 ± 0.7 minutes; 25%: 4.7 ± 0.9, p=0.0260; 50%: 5.0 ± 1.2, p=0.0407; 75%: 5.9 ± 1.7, p=0.0082; 100%: 7.0 ± 1.6, p=0.0002; 125%: 7.8 ± 1.8, p=0.0004). These results indicate a delay in morphological sickling due to drug treatment.

Based on the results of our study there is a dose-dependent decrease in RBC deformability and in the kinetics of morphologic sickling based of DSA following voxelotor treatment. These two methods have demonstrated a comparable ability to assess downstream therapeutic effects on delaying/inhibiting HbS polymerization. The combination of these two unique properties, RBC deformability and morphology change, can possibly provide a comprehensive approach to describe the cellular level of phenotypes, which may predict SCD severity. With further development these complementary assays have the potential to become important tools for drug development and monitoring response to SCD therapies.