AUTHORS
Andrew Wilks, MD, MS, Jason Lofters, MB,BS2, Chunyan Liu, MS3, Shiran Chen, MS4, Nicole Herrera, MPH5, Elizabeth Klings, MD6, Martin Steinberg, MD7
AFFILIATIONS
1 Boston University Chobanian & Avedisian School of Medicine, Boston Medical Center, Boston, Massachusetts, United States
2 Boston University Chobanian & Avedisian School of Medicine, Boston Medical Center, Boston, Massachusetts, United States
3 Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, United States
4 Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, United States
5 ASH Research Collaborative, Washington, District of Columbia, United States
6 Boston University Chobanian & Avedisian School of Medicine, Boston Medical Center, Boston, Massachusetts, United States
7 Boston University Chobanian & Avedisian School of Medicine, Boston Medical Center, Boston, Massachusetts, United States
ABSTRACT
Background: HbF prevents the polymerization of deoxy sickle hemoglobin (HbS), which leads to the RBC injury that triggers the pathophysiology of sickle cell disease (SCD). %HbF is inversely associated with overall disease severity, but it is an imperfect predictor. Sickle RBCs appear to be fully protected from HbS polymer-induced injury when their HbF:HbS ratio is ~0.30. When mean corpuscular hemoglobin (MCH) is normal (~30 pg), ~10 pg of HbF/RBC achieves this protective ratio. We hypothesized that at a given %HbF, patients with lower MCH should have fewer disease complications because their RBCs have a more favorable HbF:HbS ratio. In two historical cohorts, the Cooperative Study of SCD (CSSCD, pre-hydroxyurea (HU) era) and Multicenter Study of HU (MSH) we examined the relationship of HbF/RBC (HbF/RBC = MCH × [% HbF] /100) on selected SCD-related outcomes. In the CSSCD cohort, every 1pg increase in HbF/RBC correlated with a 5.8% and 6.0% reduction in vasoocclusive events (VOEs) and acute chest syndrome (ACS) events, respectively. In the MSH cohort, VOEs decreased by 11.5% and ACS by 13.5% with each 1pg increase of HbF/RBC after treatment. To validate and extend these results to more SCD phenotypes, we examined this relationship using data from the ASH Research Collaborative Data Hub, a larger contemporary cohort.
Methods: Patients included were aged ≥5 years with HbS/F genotypes, did not have an active VOE or ACS, had no transfusions within 6 weeks of data collection, and had at least one steady state outpatient visit (SSOP) no later than 2024. Their initial visit was the index visit. Data collected included demographics, complete blood counts, and hemoglobin fractionation by measured at SSOP. The primary outcomes were annualized VOE and ACS rate (by ICD-10 billing codes). Secondary outcomes included diagnosis of stroke, venous thromboembolism (VTE), microalbuminuria (MA), priapism and chronic kidney disease (CKD). Chi-square or Fisher’s exact tests were used to analyze the associations with categorical characteristics, and for continuous variables, non-parametric Kruskal-Wallis tests were used. Mixed-effect negative binomial regression was used to examine the effect of HbF/RBC or MCH on annualized count outcomes; mixed-effect logistic regression was used for binary outcomes with age and sex as covariates in the model. Random intercepts were assumed for each patient.
Results: 1,771 patients met the entry criteria; their mean age was 19.3 ±13.8 yrs, 53.9% were female. Males had lower %HbF and HbF/RBC than females (p=0.005, p=<0.001). Males had higher odds of ACS than females (p=0.04). Older patients tended to have higher incidence of VOE, VTE and CKD, but lower rates of ACS. The median age at index visit decreased as the %HbF increased (p=<0.001). Each 1pg increase in HbF/RBC was associated with a 3.2% reduction in annual VOE incidence (IRR 0.97, 95% CI 0.96–0.98), and the odds of a stroke in the next calendar year was decreased by 10% (OR 0.9, 95% CI: 0.81 - 1.00; p=0.04). A 1pg increase in MCH led to 5.6% increase in the odds of having VTE in the next year (OR 1.06, 95%CI: 1.00 - 1.11; p=0.04) and trended towards a 2% increase in the risk of ACS (IRR 1.02, 95% CI 0.99-1.041; p=0.06). There was no association between HbF/RBC and rate of priapism, MA or CKD.
Conclusions: In this contemporary, real-world cohort of patients with HbS/F genotypes and treated ad libitum with disease modifying therapies including HU, increased HbF/RBC is predictive of a reduced annualized rate of acute complications from SCD such as VOE and stroke. Lower MCH, indicating a more favorable HbS:HbF ratio, may protect from VTE and ACS. More chronic SCD phenotypes often associated with hemolysis like priapism, MA and CKD, as opposed to acute SCD complications such as ACS and VOE, were unaffected by increasing HbF/RBC or lower MCH. When measured in steady-state, HbF/RBC, as opposed to HbF and MCH individually, is a more useful and physiologic biomarker, and could be used to guide treatment decisions, such as HU dosing and risk of acute SCD-related events. Lower MCH suggests a favorable HbS:HbF ratio in RBCs, but HbF may better explain the improved outcomes associated with higher HbF/RBC. These findings should be confirmed prospectively but suggest that HbF/RBC may be a useful biomarker for HbF-inducing clinical trial endpoints, and importantly, the clinical management of acute SCD subphenotypes like ACS, VTE and stroke.
DISCLOSURES
Andrew Wilks, MD, MS: No relevant conflicts of interest to declare.