← Back
NeuroTrials.ai
Neurology Clinical Trial Database

TWiTCH

TCD With Transfusions Changing to Hydroxyurea (TWiTCH): a multicentre, randomised controlled trial

Share Share Copied! Compare

Year of Publication: 2016

Authors: Russell E. Ware, Barry R. Davis, William H. Schultz, ..., Robert J. Adams

Journal: Lancet

Citation: Lancet. 2016 February 13; 387(10019): 661–670

Link: https://doi.org/10.1016/S0140-6736(15)01041-7

Clinical Question

Can hydroxyurea substitute for chronic transfusions to maintain TCD velocities and prevent primary stroke in children with sickle cell anemia?

Bottom Line

Hydroxyurea therapy was non-inferior to continued transfusions for maintaining TCD velocities in children with sickle cell anemia after a period of transfusions, with no strokes in either group and superior iron unloading.

Major Points

  • Multicentre Phase III non-inferiority trial comparing hydroxyurea to continued transfusions
  • 121 children with sickle cell anemia and abnormal TCD velocities randomized 1:1
  • Primary endpoint was 24-month TCD velocity using mixed model analysis
  • Study terminated early after first interim analysis showed non-inferiority
  • No strokes occurred in either arm; 3 TIAs per arm
  • Hydroxyurea group achieved mean 27% HbF at maximum tolerated dose
  • Iron overload improved significantly more with hydroxyurea plus phlebotomy

Design

Study Type: Multicentre, randomized, open-label, non-inferiority trial

Randomization: 1

Blinding: TCD readers and stroke adjudicators were masked to treatment assignment; investigators masked to TCD results

Follow-up Duration: 24 months treatment period with 6-month post-treatment visit

Centers: 26

Countries: United States, Canada

Sample Size: 121

Analysis: Intention-to-treat using general linear mixed model; STATA 14.0 and SAS 9.4

Inclusion Criteria

  • Age 4-16 years with sickle cell anemia
  • Abnormal TCD velocities (originally ≥200 cm/sec)
  • ≥12 months of chronic transfusions
  • Verified original abnormal TCD
  • Adequate TCD windows and blood flow

Exclusion Criteria

  • Documented clinical stroke or transient ischemic attack
  • Severe vasculopathy (Grade 4 or higher on brain MRA)
  • Inadequate TCD velocities from poor blood flow or bone windows
  • Inadequate baseline brain MRI/MRA

Arms

FieldControlAlternative Treatment
InterventionContinued monthly transfusions to maintain ≤30% HbS with deferasirox chelation for iron overload (10-40 mg/kg/day)Hydroxyurea initiated at 20 mg/kg/day and escalated to maximum tolerated dose (mean 26.9 mg/kg/day) with overlap transfusions for 4-9 months, followed by serial phlebotomy (10 mL/kg monthly)
Duration24 months24 months

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
24-month TCD time-averaged mean velocity on index side calculated from general linear mixed modelPrimary143 ± 1.6 cm/sec138 ± 1.6 cm/sec8.82×10^-16 (non-inferiority), 0.023 (superiority)
New stroke eventsSecondary00
Transient ischemic attacksSecondary33
Change in serum ferritinSecondary-38 ± 2095 ng/mL-1805 ± 1651 ng/mL<0.001
Change in liver iron concentrationSecondary2.4 ± 8.7 mg Fe/gm dry wt-1.9 ± 4.2 mg Fe/gm dry wt0.001
Final HbF levelSecondary10.3 ± 7.3%24.3 ± 7.9%<0.0001
Sickle-related serious adverse eventsAdverse10 events (6 participants)23 events (9 participants)
Deferasirox-related adverse eventsAdverse19 events (9 participants)0
Phlebotomy-related adverse eventsAdverse018 events (14 participants)

Subgroup Analysis

Children with persistently elevated TCD velocities despite transfusions were eligible, unlike STOP2, broadening applicability

Criticisms

  • Open-label design with no placebo control
  • Study terminated early after first interim analysis
  • Many participants were older than peak stroke incidence age
  • Relatively short follow-up period for hydroxyurea without transfusions
  • Children with severe vasculopathy were excluded (12% of screened patients)
  • Duration of prior transfusions before transition to hydroxyurea not standardized

Funding

National Heart, Lung, and Blood Institute (NHLBI)

Based on: TWiTCH (Lancet, 2016)

Authors: Russell E. Ware, Barry R. Davis, William H. Schultz, ..., Robert J. Adams

Citation: Lancet. 2016 February 13; 387(10019): 661–670

Content summarized and formatted by NeuroTrials.ai.