Apheresis blood processing with heparin improves stem cell collection outcomes for sickle cell gene therapy

Abstract

Abstract The expansion of gene therapy (GT) for sickle cell disease (SCD) has underscored the critical need to collect large numbers of stem cells within a short timeframe for cell product manufacturing. Limited mobilization and impaired apheresis collection efficiency contribute to delays and frequent manufacturing failures, prolonging the pre-transplant phase and limiting access for additional patients. Hematopoietic progenitor cell (HPC) collections in SCD are frequently complicated by flow disruptions within the apheresis circuit, including catheter pressure alarms, buffy coat interface instability, and partial or complete occlusion of the tubing set, events rarely seen in non-SCD patients or during automated red cell exchanges in SCD. The mechanism behind these complications remains unclear but may involve microaggregate formation of cellular blood components within the buffy coat and apheresis circuit, potentially driven by SCD-associated hypercoagulability. The absence of such issues during red cell exchange may reflect the targeting of red blood cells rather than leukocytes and shorter procedure duration, both of which may reduce cellular activation and aggregation. Anticoagulant citrate dextrose solution A (ACD-A) is the only anticoagulant validated for use with the widely used Spectra Optia® Apheresis System (Terumo BCT). Heparin, historically used as a circuit anticoagulant in apheresis HPC collections, was replaced by ACD-A due to its shorter duration of action. We compared collection outcomes with and without heparin by evaluating stem cell extraction efficiency from peripheral blood (PB) to the final product. All patients received standard of care ACD-A anticoagulation, and some received additional heparin in two loading doses, without maintenance dosing or intraprocedural anticoagulation monitoring. This approach was readily integrated into existing apheresis collection protocols. We hypothesized that the addition of a second anticoagulant would reduce intraprocedural circuit complications and improve HPC collection outcomes compared to ACD-A alone. A total of 91 HPC apheresis procedures were performed in 26 patients with SCD. Patients were mobilized with single-agent plerixafor, and the target procedure duration was ≥4 total blood volumes (TBV), up to a maximum of 8 hours. Heparin was added to 53% of procedures (n=49), administered as two IV boluses of 75 units/kg (maximum 5000 units) at 10 minutes prior to apheresis initiation and again at 3 hours. Prothrombin and partial thromboplastin times were required for patients receiving heparin and were normal in all cases. The ACD-A inlet ratio most commonly ranged from 1:8 to 1:10. With heparin, collection efficiency (CD34+ cells collected ÷ [pre-apheresis CD34+ count × total blood volume processed] or CE2 formula) significantly improved (mean 50% vs 39%, p=0.02), and CD34+ cell yields were higher (mean 5.5 × 10⁶ vs 3.4 × 10⁶ CD34+ cells/kg, p=0.0007). Access pressure alarms occurred less frequently (2%, 1/49 vs 21%, 9/42), and fewer procedures ended prematurely due to circuit obstruction (defined as <3 TBV processed: 2%, 1/49 vs 17%, 7/42). The frequency of collection interface instability occurred at similar rates (67% with heparin vs 64% without). TBV processed was comparable (4.1 vs 3.8), though procedure duration was longer with heparin (406 vs 357 minutes, p<0.05). No significant differences were observed in baseline predictive factors for apheresis outcomes, including patient age, PB white blood cell count, PB platelet count, and pre-mobilization PB CD34+ count. No adverse events including bleeding or VOEs were reported in this patient group. Adverse events related to hypocalcemia caused by ACD-A were unchanged with concomitant heparin use. There were no manufacturing failures or procedural adjustments required for downstream cell processing. These findings support the use of systemic heparin anticoagulation to reduce circuit-related complications during HPC collection in SCD, enabling longer procedures, improved collection efficiency and, potentially, greater CD34+ cell yield. Further investigation into the underlying mechanisms, including real-time sampling of the buffy coat and cellular activation markers, may clarify the frequently observed interface instability and guide future protocol optimization. Our strategy has the potential to reduce the number of HPC collections needed and improve the likelihood of achieving target cell doses for GT manufacture.