Iron Unloading By Therapeutic Phlebotomy in Previously Transfused Children with Sickle Cell Anemia: The Twitch Experience

Background: TCD With Transfusions Changing to Hydroxyurea (TWiTCH, ClinicalTrials.gov NCT01425307), an NHLBI-sponsored Phase III multicenter trial, compared transfusions to hydroxyurea for maintaining TCD velocities in children with sickle cell anemia who previously received transfusions for abnormal TCD velocities. Iron overload was treated with serial phlebotomy in children randomized to hydroxyurea. At the first scheduled interim analysis, non-inferiority of hydroxyurea was demonstrated and the study was terminated prematurely.

Methods: Participants randomized to hydroxyurea received decreasing volumes of monthly transfusions during hydroxyurea dose escalation to maximum tolerated dose (MTD), averaging 6-7 months. During this transfusion overlap period, no chelation therapy was given. After hydroxyurea MTD was reached, transfusions were discontinued and children started monthly phlebotomy if their entry liver iron concentration (LIC) by MRI-R2 (FerriScan®) was ≥2 mg Fe/g dry weight liver (DWL). The prescribed phlebotomy volume was 10 mL/kg (maximum 500 mL) with adjustments for anemia (5 mL/kg for Hb 8.0-8.5 g/dL and held if Hb <8.0 g/dL). Phlebotomy was performed over 30 minutes with immediate equal volume normal saline replacement, typically using peripheral venous access. LIC was assessed at study entry, midpoint (12 months), and exit (24 months/early closure). Ferritin was monitored monthly using a centralized laboratory. Iron loading calculations were based on actual transfusion and phlebotomy volumes.

Results: Sixty children (mean age 9.7±3.2 years; range 5.2-19.0 years; 48% male) were randomized to the Hydroxyurea Treatment Arm. The average duration of previous transfusions was 4.5±2.8 years. Almost all (51/60, 85%) had previously received chelation, primarily deferasirox, and 48 (80%) were on chelation therapy at study enrollment.

Hydroxyurea MTD was achieved in 57 children (95%), and 54 commenced phlebotomy (two had low iron burden with LIC <2 and one had Hb <8.0 g/dL). A total of 914 phlebotomy procedures were scheduled per protocol for these 54 children and 756 (83%) were fully completed. There were 77 procedures cancelled due to anemia and another 81 procedures cancelled due to planned anesthesia (16), provider preference (14), hydroxyurea-related cytopenia (13), intercurrent illness (11), inadequate iv access (9), family request (5) or other (13). In 94% of phlebotomy procedures that were initiated, the full volume was removed; for the remaining 6% (47 procedures), a reduced volume was removed due to loss of venous access (37), symptoms such as headache or lightheadedness (7), or other reasons (3).

A total of 18 Adverse Events (17 Grade 2 and one Grade 3) occurred in 14 participants in association with phlebotomy (2.3% prevalence). The most common complication was light headedness/near-syncope (6) followed by anemia (4), hypotension (3), headache (3), and pain at the venous access site (1). One subject had a syncopal episode followed by transient weakness, which was centrally adjudicated as TIA.

An average of 53.6±21.8 mL/kg blood was administered in the hydroxyurea-treated arm, which calculates to an average iron loading of 40.1±16.3 mg Fe/kg, while an average of 112 mL/kg of venous blood was removed by phlebotomy, which calculates to an average iron unloading of 36.1±15.7 mg Fe/kg. For the 54 children who received phlebotomy, the average LIC was 12.0± 9.7 mg/g at study entry, 13.4±10.3 at midpoint reflecting overlap transfusions without chelation, and 9.7±8.9 at study exit reflecting serial phlebotomy, for an average net LIC decrease of 2.3±4.1 mg/g. Average serum ferritin at study entry was 3105±741 ng/mL and 1392±1542 ng/mL at study exit. For 39 children who completed all 24 months of treatment before study closure, the overall average LIC decrease was 3.2±3.8 mg/gram DWL and 10 had final LIC measurements <3 mg Fe/g. Calculated net iron loading was not significantly associated with measured changes in LIC or ferritin.

Conclusions: In the TWiTCH trial, phlebotomy was a feasible, safe, well-tolerated, and effective treatment for transfusional iron overload in children with sickle cell anemia. Although initial overlap transfusions without chelation limited the phlebotomy effects, in children who reached hydroxyurea MTD and discontinued chronic transfusions, monthly phlebotomy led to net iron unloading and lower LIC, and significantly reduced iron burden.