Neurocognitive Improvement with Hydroxyurea Therapy in Children with Sickle Cell Anemia in Uganda: Analysis from the Completed 30-Month BRAIN Safe II Trial

Abstract

Introduction: Children with sickle cell anemia (SCA) are at risk of neurocognitive impairment. The primary outcome of our open label hydroxyurea (HU) trial for children with SCA in Uganda, “BRAIN SAFE II,” was neurocognitive function at month 30 trial completion compared to enrolment. We previously reported significantly improved function at the month 18 interim assessment.1 Here we present neurocognitive results at trial completion. Methods: A random sample of 265 children ages 3-9 years with confirmed SCA were enrolled from the Mulago Hospital Sickle Cell Clinic in Kampala if: a) HU use was <6 months; and b) without evidence of prior stroke by stroke-focused neurologic exam. HU dose was escalated to maximum tolerated dose (MTD). Three key neurocognitive domains were assessed at trial months 0, 18 and 30: cognition, attention and executive function. Transcranial Doppler ultrasound (TCD) was performed in parallel by trained study medical staff. Participants' age-aligned siblings (N=110), all confirmed as being without SCA, underwent neurocognitive testing to establish age-normalized Z-scores for the SCA sample. Results: At enrolment, mean age of SCA participants was 5.6±1.7 years and fetal Hb (HbF) was 11.9±8.1%. Mean hemoglobin (Hb) was 7.8±1.2 vs. 12.5±1.1g/dL (p<.001) in the SCA and control groups, respectively. At enrolment the SCA sample scored significantly lower than controls in the 3 neurocognitive domains. At the month 18 interim assessment, re-testing of the 254 active SCA participants had demonstrated significantly improved mean Z-scores in all 3 tested areas. At month 30, the 252 active participants were mean age 8.1±1.7 years. Mean HU dose was 25.9±2.7mg/kg, Hb 8.8±1.5g/dL and HbF 21.6±10.9%. Neurocognitive Z-scores were significantly higher in cognition and attention compared to both enrolment and month 18 (p<.001). In contrast, executive function at month 30 was no longer significantly improved vs. enrolment. Anemia, HbF, blood oxygen saturation each significantly improved at months 18 and 30 vs. enrolment. As reported at month 18,1 improved scores were associated with younger age and higher Hb, as well as lower TCD velocity vs. enrolment. Conclusions: These findings confirm our prior month-18 interim findings that HU therapy at MTD improved neurocognitive function in Ugandan children with SCA. Additional findings at trial completion were: 1) Cognition and attention further improved at month 30 vs. month 18; 2) Better function found at month 18 was not sustained at month 30; and 3) Predictors of good response included higher HbF and lower TCD velocity vs. enrolment. Additional analyses now underway will more precisely identify key clinical and demographic variables for predicting the neurocognitive impact of HU. Our results demonstrate that neurocognitive impairment in children with SCA is reversible with HU therapy and that most improvements seen at month 18 were sustained at month 30. Our findings suggest that early, continuous HU therapy may have long-term benefits on preserving brain function in children with SCA. References: 1S. Kasule Naggayi, et al., Blood 142 (2023) 275-276