Abstract
Introduction
Transfusion-dependent thalassemia (TDT) is the prototype of ineffective erythropoiesis (IE)-associated anemias, which are known as iron-loading anemias. Patients with TDT receive chronic transfusions to both alleviate anemia and suppress IE. Iron overload arises from both transfusions and ongoing IE-mediated gut iron absorption. The interplay between iron metabolism and IE is complex, and the optimal transfusion threshold that balances anemia management, IE, and iron overload remains unclear. Current international guidelines recommend maintaining pretransfusion hemoglobin (Hb) levels between 9.0-10.5 g/dL in TDT. A critical knowledge gap exists, however, in balancing transfusion-related iron overload with IE-related iron overload, hindering the optimization of transfusion regimens for patients with TDT. While several erythropoiesis and iron metabolism biomarkers have been studied in TDT with strong pre-clinical and translational data, their clinical application remains quite limited.
In IE, the pathological overproduction of erythroferrone (ERFE) by an expanded population of erythroblasts suppresses hepcidin leading to iron overload. In addition to ERFE, growth differentiation factor-15 (GDF-15) is elevated in TDT and contributes to iron overload by suppressing hepcidin. The potential of these markers in TDT to guide transfusion strategies remains poorly investigated. We hypothesized that the optimal transfusion threshold for patients with TDT can be personalized based on their individual erythropoietic response to anemia.
Methods
Pre-transfusion samples were collected longitudinally in a prospective study from TDT over a median duration of 7 months (Range: 3-7 months). ERFE, GDF-15 and hepcidin were measured by ELISA. Clinically measured hemoglobin and ferritin levels were extracted from medical records. Spearman correlation coefficients (r) were used to assess relationships between biomarkers. We used linear mixed-effects model with ERFE as the outcome with Hb as a fixed predictor, and individual patient as a random intercept, to test ERFE variation with Hb across the whole cohort of samples. In addition, we calculated the interclass correlation (ICC) to quantify ERFE value clustering among patients (Nakagawa et al, J R Soc Interface, 2017).
Results
We analyzed a total of 42 samples from five patients with TDT each providing 6–12 pre-transfusion time points. The median age was 14 years (Range: 6-19 years); 3 males and 2 females. A significant negative correlation was observed between Hb and ERFE (r=-0.36, 95% CI: -0.61 to -0.06, P=0.02). ERFE levels clustered individually by patient despite variation in Hb; the random- intercept variance exceeded the residual variance resulting an ICC of 0.9, indicating that 90% of the ERFE variability is explained by differences between patients indicating strong inter-subject ERFE variation.
A strong negative correlation was observed between ERFE and the hepcidin/ferritin ratio (r = -0.90; 95% CI: -0.95 to -0.80; P < 0.001). A negative correlation was also observed between Hb and GDF-15 (r=-0.43, 95% CI: -0.65 to -0.14, P =0.005) and between GDF-15 and the hepcidin/ferritin ratio (r=-0.87, 95% CI: -0.09 to -0.08, P<0.001). The correlation between Hb and hepcidin/ferritin ratio was weaker and did not reach statistical significance in our cohort (r=0.31, 95% CI: -0.03-0.58, P=0.06). These findings indicate a stronger association between hepcidin suppression and both ERFE and GDF-15 compared to Hb, suggesting that ERFE and GDF-15 may be better indicators of hepcidin suppression than Hb alone in TDT patients.
Conclusions
The pre-transfusion hemoglobin threshold required to suppress IE varies among patients with TDT, indicating that a uniform target pre-transfusion Hb level may not be sufficient to fully suppress IE and prevent iron overload in all patients. Our data indicate that ERFE and GDF-15 are superior to Hb alone in assessing IE suppression in TDT. Further studies are needed to evaluate these biomarkers in a larger sample of patients and assess the feasibility of using these tests clinically to define individualized, optimal pre-transfusion Hb targets in TDT.
