Protocol No: ECCT/20/11/05 Date of Protocol: 05-12-2019

Study Title:

A Phase 3, Randomized, Double-Blind, Placebo-Controlled Study of Voxelotor (GBT440) in Pediatric Participants with Sickle Cell Disease (HOPE Kids 2)

Study Objectives:

Primary Objective

The primary objective is to evaluate the effect of voxelotor compared to placebo on the TCD time-averaged mean of the maximum velocity (TAMMV) arterial cerebral blood flow at 24 weeks in SCD participants ≥ 2 to < 15 years of age with conditional (170 to < 200 cm/sec) TCD flow velocity.
 
Secondary Objectives
The secondary objectives are to evaluate the effects of voxelotor compared to placebo on:
  • ï‚·        Change in TCD flow velocity at Week 48 and Week 96
  • ï‚·        Conversion of TCD category from conditional to abnormal
  • ï‚·        Reversion of TCD category from conditional to normal
  • ï‚·        Proportion of participants with TCD response
  • ï‚·        Change in Hb over time
  • ï‚·        Change in clinical measures of hemolysis
 
Exploratory Objectives
The exploratory objectives will assess the effect of voxelotor compared to placebo on:
  • ï‚·        Growth (height, weight, and head circumference)
  • ï‚·        Neuro-cognitive function
  • ï‚·        Patient reported symptoms and quality of life
  • ï‚·        Physician assessed change in global health status
  • ï‚·        Healthcare utilization
  • ï‚·        Exploratory biomarkers

Safety Objective

The safety objective is to assess the safety and tolerability of voxelotor compared to placebo.

 
Pharmacokinetic Objective
The pharmacokinetic (PK) objective is to assess the PK of voxelotor as evaluated by PPK analysis.
Laymans Summary:

SUMMARY OF RESULTS:

Participant Disposition A total of 711 participants were screened. Of these screened participants, 475 were screen failures. The primary reason for screen failure was not meeting the inclusion criteria for a conditional TCD flow velocity.

Overall, 236 participants were randomized and treated, 120 in the voxelotor group and 116 in the placebo group. As of LPLV, 33.5% (79/236) of study participants had completed 96 weeks of treatment. A total of 66.5% (157/236) discontinued treatment early. The most frequent reason for early discontinuation of treatment was abnormal TCD (prospectively defined in the protocol as criteria for discontinuation if ≥200 cm/sec but <220 cm/sec on 1 test and ≥200 cm/sec on re-test or ≥220 cm/sec on one test), 19.9% (47/236) overall, (17.5% [21/120] voxelotor, 22.4% [26/116] placebo). The reason of ‘Study Terminated By Sponsor’ was noted for 24 (20.0%)voxelotor participants and 18 (15.5%) placebo participants; however, the majority of participants in the ‘Other’ category also discontinued treatment early due to the study termination/sponsor decision (40/41).

In total, 83.1% (196/236) of participants completed at least 48 weeks in the study.

Demographic and Other Baseline Characteristics:
Demographic characteristics were similar in the voxelotor group and the placebo group. In the study population overall, the mean (standard deviation [SD]) age was 7.2 (3.16) years, with the majority of participants in the 4- to <12-year age group (77.5% [183/236]). Similar percentages of participants were male and female (48.3% [114/236] and 51.7% [122/236], respectively). The majority of participants identified as African (63.1% [149/236]) and as Not Hispanic or Latino (96.2% [227/236]). The majority resided in sub-Saharan Africa ([SSA] 82.6% [195/236]) with an additional 15.3% (36/236) in the Middle East and North Africa (MENA).

Baseline characteristics were generally similar in the voxelotor group and the placebo group. Mean (SD) Hb at baseline was 7.7 (1.04) g/dL in the voxelotor group and 7.6 (1.03) g/dL in the placebo group. The mean (SD) TCD flow velocity was 182.6 (7.24) cm/sec and 182.8 (7.85) cm/sec, respectively and the majority of participants in both treatment groups were in the 170 cm/sec to <185 cm/sec category, 72 (60.0%) and 73 (62.9%), respectively. The majority of participants were not using HU at baseline, 79 (65.8%) in the voxelotor group and 81 (69.8%) in the placebo group.

The percentage of participants who did not experience any VOCs in the 12 months prior to screening was lower in the voxelotor group (47.5% [57/120]) than in the placebo group (58.6% [68/116]). Of the participants who had 1 or more prior VOCs, the majority reported 1 or 2 VOCs in the prior 12 months; 53/63 in the voxelotor group and 42/48 in the placebo group. The majority of participants in both treatment groups had no VOCs in the prior 12 months that required hospitalization, 71.7% and 79.3%, respectively. Note that number of prior VOCs was not a stratification factor for randomization or endpoint analysis.

Baseline characteristics in those participants who were taking HU at baseline and those who were not taking HU at baseline were generally similar; small differences observed in hematologic characteristics are consistent with the hematologic effects of HU.

Exposure
Exposure to study drug was similar in both the voxelotor and placebo groups. As of the 01 May 2024 dosing pause, the actual mean (SD) exposure to study drug was 71.9 (25.68) weeks in the voxelotor group and 70.0 (29.75) weeks in the placebo group. Few participants in either group experienced a dose reduction, 5.0% (6/120) in the voxelotor group and 3.4% (4/116) in the placebo group. The majority of participantsin both treatment groups missed at least 1 dose of study drug, 78.3% (94/120) in the voxelotor group and 72.4% (84/116) in the placebo group. Of these participants, the majority of participants missed doses due to other (non-compliance).

Adherence, defined as actual exposure divided by duration of exposure, to the daily dosing regimen was high in bothtreatment groups. Mean (SD) study drug adherence was 96.13% (7.923) in the voxelotor group and 96.91% (6.161) in the placebo group.

CONCLUSIONS: Efficacy Conclusions
Study GBT440-032 met its primary efficacy objective to demonstrate the effect of voxelotor compared to placebo on the TCD TAMMV arterial cerebral blood flow at 24 weeks in SCD participants 2 to <15 years of age with conditional (170 to <200 cm/sec) TCD flow velocity at baseline. Treatment with voxelotor resulted in a statistically significant improvement in TCD flow velocity vs. placebo at Week 24 (primary endpoint). Statistically significant improvements in TCD flow velocity observed at Week 24 were sustained at Week 48 (key secondary endpoint). Treatment with voxelotor resulted in improvements in secondary endpoints, including those related to TCD flow velocity as well as hemoglobin and clinical measures of hemolysis, specifically bilirubin levels.

Safety Conclusions
The study showed treatment with voxelotor in this pediatric population with conditional TCD flow velocity was generally well tolerated. A higher percentage of voxelotor participants experienced SCD-related TEAEs than placebo participants and there was an imbalance in the number of fatal events with more events occurring in participants receiving voxelotor than in participants receiving placebo. All treatment-emergent fatal events occurred in participants in SSA countries and malaria was associated with several of the fatal events

Abstract of Study:
Investigational Product: 
Voxelotor as 300 and 500 mg tablets or as 300, 400, 600, and 900 mg powder for oral suspension (packaged in stick pack pouches) Matching placebo
 
Number of Clinical Sites:  The study will be conducted at approximately 50 international clinical sites.
Number of Study Participants:  Approximately 224 participants with sickle cell disease (SCD), aged ≥ 2 to < 15 years.
 
Treatment:  Participants will be randomized in a 1:1 ratio to receive voxelotor or placebo. All participants younger than 12 years of age and randomized to voxelotor will receive a dose based on their body weight, to provide exposure corresponding to the adult dose of 1500 mg/day. Participants 12 years or older will take 1500 mg/day.
Participants will receive study drug as a tablet formulation or powder for oral suspension formulation (packaged in stick pack pouches).
The tablet formulation is to be administered orally, once daily. Once swallowed, the tablet should be followed by drinking a liquid such as water. The powder for oral
suspension formulation is to be dispersed in liquid and administered orally, once daily. Both dosage forms may be taken with or without food.
The participant’s weight at screening will be used to determine their initial treatment dose. Dosing should be adjusted if the participant’s weight increases or decreases over 2 consecutive clinic visits. Participant’s weight will be measured and dose adjustments made as needed, as outlined in the protocol.
 
Objectives:
Primary Objective:
The primary objective is to evaluate the effect of voxelotor compared to placebo on the transcranial Doppler (TCD) time-averaged mean of the maximum velocity
(TAMMV) arterial cerebral blood flow at 24 weeks in SCD participants ≥ 2 to < 15 years of age with conditional (170 to < 200 cm/sec) TCD flow velocity.
Secondary Objectives:
The secondary objectives are to evaluate the effects of voxelotor compared to placebo on:
  • ï‚·Change in TCD flow velocity at Week 48 and Week 96 ï‚·
  •     Conversion of TCD category from conditional to abnormal
  • ï‚· Reversion of TCD category from conditional to normal
  •     Change in hemoglobin (Hb) over time
  •     Change in clinical measures of hemolysis
Exploratory Objectives:
The exploratory objectives will assess the effect of voxelotor compared to placebo on:
  • ï‚·        Growth (height, weight, and head circumference)
  • ï‚·        Neuro-cognitive function
  • ï‚·        Patient reported symptoms and quality of life
  • ï‚·        Physician assessed change in global health status
  • ï‚·        Healthcare utilization
  • ï‚·        Exploratory biomarkers
Safety Objectives:
The safety objective is to assess the safety and tolerability of voxelotor compared to placebo.
Pharmacokinetic Objectives:
The pharmacokinetic (PK) objective is to assess the PK of voxelotor as evaluated by population pharmacokinetics (PPK) analysis.

Study Design:

The study will be conducted globally at approximately 50 clinical sites.

The study consists of 3 periods:
ï‚·        Screening: up to 35 days prior to randomization
ï‚·        Treatment: 96 weeks
ï‚·        Follow-Up: 28 days (4 weeks) after the last dose
Approximately 224 participants aged 2 to < 15 years will be randomized 1:1 to voxelotor or placebo.
Participants ≥ 12 years old will be administered a fixed dose of voxelotor 1500 mg/day. Participants younger than 12 years of age will receive voxelotor at a weight-based (1500 mg-equivalent) dose.
Following completion of study treatment, eligible participants will be given the option to enroll in an open-label extension study (under a separate protocol) to receive voxelotor for a minimum of 3 years (3 years or until commercial product becomes available, whichever is longer).
At the time of randomization, participants will be stratified by hydroxyurea (HU) use (yes/no), age (2 to ≤ 8 years; > 8 to < 15 years), Screening TCD flow velocity value (170 to < 185 cm/sec; 185 to < 200 cm/sec) and geographic region (Africa including Middle East and North Africa [MENA]; rest of world). Stratification of participants by age was chosen to evenly balance by age distribution using the expected median age of 8 years.
Dosing of participants between 2 and < 4 years of age:
At least 15 participants from 2 to < 4 years of age will be enrolled. Enrollment of participants <4 years of age will be initiated after the Data and Safety Monitoring Board (DSMB) has reviewed safety and PK data from at least 12 participants < 30 kg treated with voxelotor for at least 28 days. These data for at least 12 participants may be generated from any voxelotor clinical studies.
 
Duration of Study Participation:  The duration of study involvement for an individual participant is expected to be approximately 105 weeks.
The study will end when the last participant’s last visit occurs.
 
Statistical Methods:
Primary Endpoint
The primary efficacy endpoint is the change from Baseline at 24 weeks in TAMMV arterial cerebral blood flow, as measured by TCD. Secondary Endpoints
The secondary efficacy endpoints are as follows:
  • Change from Baseline in TCD flow velocity at Week 48 and Week 96 ï‚·       
  • Incidence of conversion to abnormal TCD flow velocity (≥ 200 cm/sec) over 24, 48, and 96 weeks
  • Incidence of reversion to normal TCD flow velocity (<170 cm/sec) over 24, 48, and 96 weeks
  • Proportion of participants with TCD flow velocity reduction ≥ 15 cm/sec at Week 24, Week 48, and Week 96
  • Change in Hb from Baseline at Week 24, Week 48, and Week 96
  • Change in clinical measures of hemolysis (unconjugated bilirubin, % reticulocyte, absolute reticulocyte, and lactate dehydrogenase [LDH]) from Baseline at Week 24, Week 48, and Week 96
Exploratory Endpoints:
The exploratory endpoints are as follows:
  • ï‚·        Growth and development as assessed by change from Baseline in height, weight, and head circumference
  • ï‚·        Neurocognitive function assessed by National Institutes of Health (NIH) toolbox for processing speed
  • ï‚·        Participant fatigue assessed by the Peds QL Multidimensional Fatigue Scale
  • ï‚·        Number of missed school days (if applicable) due to SCD
  • ï‚·        Number of missed work days (if applicable) of the primary caregiver due to participant’s SCD
  • ï‚·        Investigator assessment of change in the overall health condition of the participant assessed with Clinical Global Impression– Change (CGI-C)
  • ï‚·        Exploratory biomarkers (dense cells, creatinine:albumin ratio assessments)
  • ï‚·        Healthcare resource utilization (eg, incidence of hospitalizations, emergency room/clinic visits)
Safety Endpoints
The safety endpoints include adverse events (AEs), clinical laboratory assessments, and vital signs.
Pharmacokinetic Endpoint
The endpoint is whole blood and plasma voxelotor PK as evaluated by PPK analysis using nonlinear mixed effects modeling.
 
Sample Size
The sample size is planned to provide sufficient statistical power for the primary efficacy analysis and for a robust safety database. In addition, the number of events required to assess secondary efficacy endpoints, eg, percentage conversion to abnormal TCD category, is also considered.

For the primary endpoint (change from Baseline in TCD flow velocity [cm/sec]), assuming equal allocation across treatment groups and a common variance, an estimate of the number of participants in each of 2 treatment groups used the following assumptions in the power calculation:
  • ï‚·    Targeted treatment effect: difference in mean change from Baseline in TCD flow velocity is 15 cm/sec
  • ï‚·    Common standard deviation is 24 cm/sec
  • ï‚·    A significance level of 5% (2-sided)
Assuming a 20% drop out rate, a sample size of 112 participants per group provides > 95% power to detect the targeted treatment effect for the following hypothesis test using a Student’s t-test:
 
H012   vs.  HA:µ1µ2

where µ1 and µ2  stand for mean change from Baseline in TCD flow velocity at Week 24 in voxelotor and placebo groups, respectively.
For the secondary efficacy endpoint of percentage conversion to abnormal TCD category, assuming 35% of participant in the placebo group have experienced a TCD flow velocity > 200 cm/sec by Week 96, a hazard ratio of 0.5 between voxelotor group and placebo, a fixed follow up time of up to 96 weeks
for each participant, and an overall 20% drop out rate, a total of 49 events is expected. This provides approximately 65% power for a log-rank test at a two-sided significance level of 5% and approximately 75% power at a two-sided significance level of 10% (PASS version 11).
 
Efficacy Analyses
The primary analysis of change in TCD flow velocity from Baseline to Week 24 will be performed using a mixed model for repeated measures (MMRM). Independent variables will include treatment, study visit, treatment by study visit interaction, and randomization stratification factors (ie, HU use, age, Screening TCD flow velocity, and geographic region). Within-subject variability will be modeled using an unstructured covariance matrix.
The secondary endpoints will be analyzed as described below.
  • ï‚·    Change from Baseline endpoints:
    •   Change in TCD flow velocity from Baseline to Week 48 and Week 96 will be analyzed with a similar MMRM as described for the primary efficacy endpoint (change from Baseline in TCD flow velocity at Week 24).
    • ​Change in Hb and clinical measures of hemolysis (unconjugated bilirubin, % reticulocytes, absolute reticulocyte, and LDH) from Baseline at Week 24, Week 48, and Week 96 will be analyzed with a similar approach.
  • ​TCD responder endpoint:
    •   The TCD responder endpoint, defined as a TCD flow velocity reduction from Baseline ≥ 15 cm/sec at a given timepoint, eg, Week 24, Week 48, or Week 96, will be analyzed using an exact Cochran-Mantel-Haenszel (CMH) general association test stratified by the randomization stratification factors for the treatment difference between voxelotor and placebo.
  • ​Time to events endpoints:
    • For the incidence of conversion to abnormal TCD, the Kaplan-Meier method will be used to estimate the landmark rates and the associated 95 % confidence intervals in the voxelotor and placebo groups at Week 24, Week 48, and Week 96. A log-rank test stratified for the randomization stratification factors will be used to test the difference between voxelotor and placebo groups. A Cox proportional hazard model may be used to estimate the hazard ratio, as appropriate. Reversion from conditional to normal TCD at Week 24, Week 48, and Week 96 will be analyzed using a similar approach.
​Safety and Tolerability Analyses
Safety analysis will be performed on all participants receiving at least one dose of study drug. Adverse events will be classified according to the Medical Dictionary for Regulatory Activities (MedDRA). The incidence of treatment-emergent adverse events (TEAEs), defined as an event that occurs on or after Day 1 of study treatment or the worsening of a preexisting condition on or after Day 1 of study treatment, will be tabulated by System Organ Class (SOC), preferred term, severity,
and relationship to study drug. Changes in laboratory parameters (hematology, serum chemistry, and liver function tests) and vital signs (eg, blood pressure [BP], pulse, and body temperature) over time will be summarized descriptively. SCD-related AEs will be collected and summarized separately (including
vaso-occlusive crisis [VOC], acute chest syndrome, osteonecrosis, priapism, etc.) Pharmacokinetic Analyses PPK analysis will consist of all participants who receive active study drug and have at least one measured concentration at a scheduled PK time point after the start of dosing. If any participants are found to be noncompliant with respect to dosing or have incomplete data, protocol deviations, or clinical events that affect PK, a decision will be made on a case-by-case basis as to their inclusion in the analysis.
Participants in this population will be used for all PK summaries. PPK analyses using nonlinear mixed effects modeling will be performed to characterize voxelotor PK in plasma and whole blood. The influence of demographic covariates (such as body weight, height, age, gender) on voxelotor PK parameters (ie, clearance [CL] and volume of distribution) will be investigated. If appropriate, the voxelotor PK data may be pooled with PK data from other studies. Continuous variables will be descriptively summarized using mean, standard deviation (SD), coefficient of variation (CV%, as appropriate), median, minimum, maximum, and, as appropriate,
geometric mean. Categorical variables will be descriptively summarized by presenting the number (frequency) and percentage in each category.