Hydroxyurea Adherence By Personal Best HbF and the 2-Site “HABIT” Intervention Trial for Pediatric Sickle Cell Disease

Introduction: Hydroxyurea (HU) induction of fetal hemoglobin (HbF) is a major therapeutic effect for sickle cell disease (SCD) HbSS or HbSB⁰thalassemia. Non-adherence to a HU daily regimen is a barrier to its full effect. As no uniform level of HbF induction exists, we hypothesized that a child’s historical “personal best” (PB) HbF level at maximum clinical dose can be used as a marker for HU adherence. The NIH-funded “HABIT” study is a two-site randomized trial to improve patient-centered HU adherence in adolescents age 10-18 years. Eligibility is determined by fall-off from PB HbF over the previous year. Adherence measures include tracking HbF, pharmacy refill data and self-report.

Methods: In a cross-sectional analysis of the entire 2-site clinic-based sample of 95 youth with SCD on HU, ages 10-18 years, and from the subset enrolled in HABIT to date, we assessed demographics, HU use and HbF at pre-treatment, PB and a recent time point. Here, HU non-adherence was defined as ≥20% deviation from PB HbF. Data were analyzed using descriptive statistics and Pearson correlation; groups were stratified by site and participation in HABIT, and were compared using chi-square and student’s t-tests.

Results: Only ethnicity and mean HU dosing significantly differed by site (more Latinos/other and lower HU dosing at Columbia vs. Einstein). Only 39 (41%) youth remained within 20% of their PB HbF, with no significant differences in deviation from PB HbF by: site, gender; ethnicity; age (< or ≥14 years); higher induced PB (HbF of ≥20%); or time to PB (≤ or >1 year). For the 39 adherent to HU, mean deviation from PB HbF was 7.3±8.4% (median 7.2); for the 56 non-adherent, mean deviation was 40.5±16.4% (median 36.8; p<0.001). Greater deviation from PB was associated with: 1) Younger age at HU initiation (r = -0.24, p = 0.03); and 2) Longer duration of HU use (r = 0.28, p=0.009). For the 13 HABIT subjects, significant differences from the other 82 patients were limited to the greater proportion of Latinos/other and slightly older ages. Analyses will be performed to examine relationships between deviation from PB HbF, alterations in MCV, acute clinical events and use of urgent medical services (ER use/admissions).

Conclusion: Our definition of HU non-adherence as ≥20% deviation from Personal Best HbF identified two non-overlapping patient groups ages 10-18 years, with the majority found to be non-adherent. These findings also suggest that HU initiation at younger ages may increase risk of under-adherence during adolescence. PB HbF appears to be a useful marker for HU adherence, and may aid in clinical and patient-centered assessment and intervention. These results underscore the need to assess HU adherence in adolescent patients and for intervention trials such as HABIT.

¹Comparing HABIT subjects to the remaining 82 patients in the sample.

²Designation of “other” includes multi-racial

HABIT: Hydroxyurea Adherence for Personal Best in Sickle Cell Treatment Funding: 5R21NR013745 (PIs NSG and AMS)