Abstract 4246

Introduction

Serum ferritin (SF) is typically used as a surrogate marker for body iron load in thalassemia major (TM) patients. SF is useful as it can be regularly and inexpensively measured, allowing early trend recognition in chelation management. Various guidelines have recommended target SFs for TM patients, regardless of the prescribed chelator. The emergence of non-invasive LIC determinations has now allowed us to examine whether relationships between SF and LIC vary with chelator. Any differences in this relationship among chelators can have important implications for treatment guidelines.

Methods

We assessed whether the relationship between LIC measured by Ferriscan® (LIC-R2) and ferritin (using most recent SF within 3 months of LIC-R2 and mean SF in the year prior to LIC-R2 (SFmean)) varied according to chelators. TM patients (n=84), mean age 33 (±10) years, who were compliant and continuously on one chelation modality for > 1 year prior to LIC-R2 were retrospectively analysed. Chelation regimes included deferoxamine (DFO) (n=17), deferasirox (DFX) (n=47) or deferiprone (DFP) (n=23). The following patients were excluded: those with rheumatological diseases, HCV-RNA+ hepatitis, steatohepatitis and recent infections to minimize confounders of SF elevation; those on combination DFO and DFP due to their highly variable DFO frequencies; those with LIC >43mg/g dw as their LICs could not be reliably quantified.

Results

Patients had received a median of 10 (1-33), 1.5 (1-4) and 8 (1-19) years of DFO, DFX and DFP respectively prior to LIC-R2. Baseline characteristics were not significantly (sig) different among chelator groups (P>0.25): 51% had splenectomy; SF was determined at a median of 11(0-55) days from LIC-R2. The 3 chelator groups had similar median LICs (P=0.99) but DFP patients had sig lower median SF and SFmean than DFO (P=0.00) and DFX (P<0.01) patients (Table). This was observed across clinically relevant LIC thresholds of <3.2, 3.2–7 and >7 mg/g dw (P<0.05; DFP vs DFO and DFX). For high LICs associated with increased cardiac risk (>15mg/g dw), the median of DFP patients' SFmean was only 2529 (1776-2734) μg/L, which had a tendency to be less than that of DFX patients (median 4563 (1761-6871) μg/L, P=0.09) despite similar median LICs (P= 0.71). These were reflected by a lower median SF/LIC ratio in DFP than DFO (P=0.00) and DFX (P=0.00). Linear regressions of LIC vs SF were then compared between different chelators for SF <4000μg/L and LIC ≤ 20 mg/g dw as the best linear correlations were observed for these values across all 3 chelators. DFP's linear regression line (LIC =(0.0043 ±0.0011) × SF+ (0.74 ± 1.63), R =0.65) was sig different from DFO's (LIC =(0.0025 ±0.0010) × SF + (-0.55 ±2.28), R =0.58, P =0.00) and DFX's (LIC =(0.0022 ±0.0004) × SF +(0.55 ±0.75), R =0.69, P =0.047). DFP had higher predicted LICs at various SF compared to DFO and DFX (table). DFO was not sig different from DFX for the above parameters.

Conclusions

Once patients are stabilized on >1 year of a given chelator, the relationship between LIC and SF shows differences among chelators, especially for SF <4000μg/L and LIC ≤20 mg/g dw. Patients on DFP monotherapy will tend to have higher LICs at any given SF in this range than those on DFO or DFX. About half of our DFP patients had LIC >15mg/g dw despite SFmean <2500μg/L. As SF <4000μg/L mainly reflects reticuloendothelial (RE) rather than hepatocellular iron, these are consistent with a preferential removal of RE compared with hepatocyte iron by DFP, likely due to its rapid inactivation by glucuronidation within the hepatocyte, while DFO and DFX have a relative trophism for hepatocellular iron. These findings suggest that for chelation involving DFP, the SF trend may not reflect the LIC in the same way as with DFO or DFX, emphasizing the value of LIC monitoring in addition to SF.

Table 1.

LIC, ferritin and predicted LIC.

DFODFXDFP
Median LIC(mg/g dw) 5 (1.2–30.6) 4.8 (0.8–36.5) 5 (0.5–34.9) 
Median SF (μg/L) 1927 (1378–5182) 1713 (312–6085) 1142 (133–2897) 
Median SFmean (μg/L) 2147 (950–6063) 2006 (773–7290) 1240 (230–2734) 
Median SF/LIC (μgL–1/mgg–1) 523 (120–1562) 403 (52–1188) 181 (56–910) 
Predicted LIC (95%CI) (mg/g dw) at: 
SF 1000 μg/L 1.9 (0–4.9) 2.8 (1.9–3.7) 5.1 (3.2–6.9) 
SF 2000 μg/L 4.4 (2.8–5.9) 5 (4.2–5.8) 9.4 (6.8–12.0) 
SF 4000 μg/L 9.3 (5.1–13.5) 9.5 (7.3–11.6) 18 (11.1–24.9) 
DFODFXDFP
Median LIC(mg/g dw) 5 (1.2–30.6) 4.8 (0.8–36.5) 5 (0.5–34.9) 
Median SF (μg/L) 1927 (1378–5182) 1713 (312–6085) 1142 (133–2897) 
Median SFmean (μg/L) 2147 (950–6063) 2006 (773–7290) 1240 (230–2734) 
Median SF/LIC (μgL–1/mgg–1) 523 (120–1562) 403 (52–1188) 181 (56–910) 
Predicted LIC (95%CI) (mg/g dw) at: 
SF 1000 μg/L 1.9 (0–4.9) 2.8 (1.9–3.7) 5.1 (3.2–6.9) 
SF 2000 μg/L 4.4 (2.8–5.9) 5 (4.2–5.8) 9.4 (6.8–12.0) 
SF 4000 μg/L 9.3 (5.1–13.5) 9.5 (7.3–11.6) 18 (11.1–24.9) 
Disclosures:

Off Label Use: Deferiprone, an oral iron chelator, is an off-label medication in North America. Shah:Swedish Orphan/Apotex: Honoraria; Norvatis: Honoraria, Speakers Bureau. Porter:Norvatis: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau.

This icon denotes a clinically relevant abstract

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution