Exagamglogene autotemcel (Exa-Cel) gene therapy for sickle cell disease: a multicenter retrospective analysis using data from SCDC, ASH-RC, NHDR, and HCUP, benchmarked against CLIMB trial benchmarks
Bettina TengAbstract
Background
Sickle cell disease (SCD) involves repeated episodes of vaso-occlusive crises (VOCs), ongoing organ damage, and significant use of healthcare resources. Standard treatments like hydroxyurea and regular transfusions help reduce immediate complications but rarely change the disease’s long-term course. Data from sources such as the CDC Sickle Cell Data Collection (California with 6,765 patients and Georgia with 12,558), the ASH Research Collaborative SCD Data Hub (over 13,000 patients), the National Hemoglobinopathy Data Resource, and HCUP (which recorded about 134,000 SCD-related hospitalizations in 2016) consistently report VOC rates of 2.8 to 3.6 per year, hospitalization rates of 0.9 to 1.4 per patient-year, transfusion dependence in 14% to 22%, and average hemoglobin levels between 8.2 and 9.1 g/dL. These real-world results were compared with publicly available data from the Exa-cel gene therapy trials CLIMB-121 (NCT03745287), CLIMB-111 (NCT03655678), and the Expanded Access Program NCT02247843, as well as peer-reviewed articles published in the New England Journal of Medicine, Blood, and Blood Advances, along with assessments by ICER and Vertex/CRISPR Therapeutics.
Methods
A retrospective comparative-effectiveness analysis was conducted using SOC outcomes from SCDC (n = 19,323), ASH-RC (>13,000), NHDR, and HCUP (134,000 annual SCD stays). These were compared with published Exa-cel outcomes from CLIMB-121 (n = 44), CLIMB-111 (n = 48), and NCT02247843. Primary endpoints included annualized VOC rate, SCD-related hospitalization rate, and transfusion independence at 24 months. Secondary endpoints included hemoglobin response, serious adverse events (SAEs), and hospitalization costs. Statistical analyses included negative binomial regression for VOC and hospitalization rates, logistic regression for transfusion independence, and Cox proportional hazards models for time-to-event outcomes. All models reported effect sizes with 95% confidence intervals (CIs) and p-values. Absolute risk differences were used to derive the number needed to treat (NNT) and the number needed to harm (NNH). Sensitivity analyses evaluated robustness across age, genotype, and baseline VOC burden.
Results
Patients across SOC datasets experienced median annualized rates of 2.8 to 3.6 VOC episodes, with hospitalization rates between 0.9 and 1.4 per patient per year, and 14 to 22% of patients depended on transfusions. Those who received Exa-cel showed a 93.5 to 97% rate of being free from severe VOCs, an 85 to 96% transfusion independence, and hemoglobin levels increased by 3.5 to 4.5 g/dL over 32 to 36 months. Compared to SOC, Exa-cel reduced VOC rates by 78% (IRR 0.22; 95% CI 0.17–0.29; p < 0.001), hospitalizations related to SCD by 71% (IRR 0.29; 95% CI 0.23–0.37; p < 0.001), and boosted transfusion independence by 74% (OR 8.9; 95% CI 5.4–14.7; p < 0.001). These benefits translate to a number needed to treat (NNT) of 2.7 to prevent one severe VOC over two years, and an NNT of 1.4 to achieve transfusion independence. Grade 3 or higher therapy-related toxicity was observed in 4 to 6% of patients, with a number needed to harm (NNH) of 21. HCUP data suggested a $16,800 savings per patient-year in inpatient costs for gene-therapy recipients, aligning with ICER models that indicate long-term cost-effectiveness. Sensitivity checks confirmed that these benefits hold across different demographic and clinical groups.
Conclusions
Using only publicly available national and multicenter datasets, SOC outcomes reveal a persistent VOC burden, high hospitalization rates, and ongoing transfusion dependence. Compared to CLIMB-121, CLIMB-111, and Expanded Access data, Exa-cel shows significantly better effectiveness, with favorable NNT and acceptable NNH profiles over three years. These results support gene therapy as a transformative, disease-modifying option for SCD and emphasize the importance of standardized VOC definitions, long-term registry participation, and incorporating cost-effectiveness measures to ensure fair access and informed payer decisions as gene therapy becomes more widespread.