AUTHORS
Matthew L Repp, MD, Grace Bosma, MS2, Diana Abbott, PhD3, Alisha Chitrakar4, Cassie Kemmler5, Tamir Kanias, PhD6, Keith Neeves, PhD7, Gemlyn George, MD8, Lisa Pei Chu, MD9, Kimberly A Thomas, PhD10, Brandon McMahon, MD11
AFFILIATIONS
1 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
2 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
3 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
4 Vitalant Research Institute, Denver, Colorado, United States
5 Vitalant Research Institute, Denver, Colorado, United States
6 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States, Vitalant Research Institute, Denver, Colorado, United States
7 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
8 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
9 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
10 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States, Vitalant Research Institute, Denver, Colorado, United States
11 University of Colorado Anschutz Medical Campus, Aurora, Colorado, United States
ABSTRACT
Background
High-oxygen-affinity hemoglobinopathies are a rare cause of congenital erythrocytosis, reducing tissue oxygen delivery and driving compensatory erythropoiesis. Unlike JAK2 V617F-associated erythrocytosis, where thrombotic risk and risk-directed interventions are established, the thrombotic potential of erythrocytosis from high-oxygen-affinity variants such as Hemoglobin San Diego (HbSD) is undefined. Data as to whether these patients harbor intrinsic erythrocyte membrane or hemostatic abnormalities are lacking.
Objectives
To better characterize the erythrocyte membrane and hemostatic phenotype associated with HbSD by evaluating RBC membrane biology, deformability, and hemostatic capacity.
Methods
Peripheral blood was collected from a healthy adult male patient with genetically confirmed HbSD [β109 (GTG→ATG)] and 12 healthy control donors per approved IRB protocols. Red cell membrane phenotype was assessed by flow cytometry for adhesion receptors (ICAM-4, CD36, Lu/BCAM, CD49d) and for phosphatidylserine (PS) exposure by lactadherin binding; deformability and osmotic fragility measured by osmotic-gradient ektacytometry; thrombin generation assayed by a calibrated automated thrombogram adapted for whole blood assessment (untriggered and tissue factor [TF]–triggered); and viscoelastic testing performed using thromboelastography (TEG; kaolin and native). The single case was compared with the control distribution using leave-one-out analysis. Controls are summarized as median [Q1, Q3].
Results
The HbSD sample showed erythrocytosis (hemoglobin 19.4 g/dL, hematocrit 56.1%) with thrombocytopenia (platelets 97 ×109/L) and normal red-cell indices. Membrane phenotyping revealed a selective adhesion-receptor abnormality: ICAM-4 was reduced in both frequency (77.1% vs 96.0% [95.3, 96.7]) and surface receptor density (445 vs 1015 [900, 1105]) compared to controls, while Lu/BCAM surface receptor density was increased (866 vs 442 [311, 642]). CD36 and CD49d expression and frequency were similar between HbSD and controls. PS exposure on both RBCs and platelets was also similar. Ektacytometry showed reduced deformability (EImax 0.492 vs 0.533 [0.526, 0.567], curve area 126 vs 154 [152, 156]) with a left-shifted hyperosmotic point (Ohyper 414 vs 472 [458, 482] mOsm/kg) and increased osmotic fragility (osmotic hemolysis 27.0% vs 12.6% [12.3, 14.5]), whereas oxidative fragility was not increased (oxidative hemolysis 11% vs 13% [7, 22]). Whole blood thrombin generation demonstrated prolonged lag time and time to peak thrombin under both untriggered and TF-triggered conditions (TF lag 15.9 vs 4.35 [3.74, 4.70] min; TF time to peak 33.0 vs 10.0 [9.2, 12.2] min) with preserved peak thrombin levels. Endogenous thrombin potential (ETP) was elevated with TF triggering (1976 vs 915 [779, 1030] nM·min) but normal in untriggered samples. Kaolin TEG demonstrated prolonged reaction time (9.6 vs 6.4 [5.8, 6.8] min), reduced α-angle (39 vs 68 [66, 71]°), and reduced maximum amplitude (MA; 42.7 vs 65 [63, 72] mm), with a concordant pattern on native tracings. Using leave-one-out analysis, case omission shifted the cohort estimate beyond that of any individual control: hematocrit −11.8 vs −0.1 [−2.5, 5.0]; ICAM-4 frequency 18.8 vs −3.3 [−4.1, −2.5]; osmotic fragility −13.7 vs 3.1 [0.9, 3.5]; TF-triggered ETP −1043 vs 107 [−19, 254]; and MA 23 vs −1 [−8, 1].
Conclusions
To our knowledge, this is the first functional characterization of the erythrocyte and hemostatic phenotype associated with HbSD. HbSD demonstrated intrinsic erythrocyte membrane abnormalities characterized by impaired deformability, increased osmotic fragility, and altered adhesion receptor expression (reduced ICAM-4 and increased Lu/BCAM). Reduced deformability and increased osmotic fragility are consistent with altered erythrocyte membrane mechanics, potentially reflecting a reduced membrane surface-area-to-volume ratio. Increased thrombin generation was accompanied by reduced clot strength, the latter likely reflecting the patient's thrombocytopenia and elevated hematocrit; whether these findings translate to clinically meaningful changes in hemostatic capacity remains to be evaluated. Although limited to a single case, leave-one-out analysis identified the HbSD sample as a consistent outlier. These findings suggest that high-oxygen-affinity hemoglobinopathies such as HbSD may harbor intrinsic erythrocyte and procoagulant abnormalities that contribute to thrombotic risk.
DISCLOSURES
Matthew L Repp, MD: No relevant conflicts of interest to declare.