BACKGROUND: Realizing Effectiveness Across Continents with Hydroxyurea (REACH) is an open-label non-randomised trial of hydroxyurea (hydroxycarbamide) in children with sickle cell anaemia in sub-Saharan Africa. The short-term results of REACH on safety, feasibility, and effectiveness of hydroxyurea were published previously. In this paper we report results from extended hydroxyurea treatment in the REACH cohort up to 8 years. METHODS: In this open-label, non-randomised, phase 1/2 trial, participants were recruited from four clinical sites in Kilifi, Kenya; Mbale, Uganda; Luanda, Angola; and Kinshasa, Democratic Republic of Congo. Eligible children were 1-10 years old with documented haemoglobin SS or haemoglobin Sβ zero thalassaemia, weighing at least 10 kg. Participants received fixed-dose hydroxyurea of 17.5 (±2.5) mg/kg per day for 6 months (fixed-dose phase), followed by 6 months of dose escalation (2·5-5·0 mg/kg increments every 8 weeks) as tolerated, up to 20-35 mg/kg per day (maximum tolerated dose; MTD), defined as mild myelosuppression. After the MTD was reached, hydroxyurea dosing was optimised for each participant on the basis of changes in bodyweight and laboratory values over time (MTD with optimisation phase). After completion of the first 12 months, children with an acceptable toxicity profile and favourable responses were given the opportunity to continue hydroxyurea until the age of 18 years. The safety and feasibility results after 3 years has been reported previously. Here, haematological responses, clinical events, and toxicity rates were compared across the dosing phases (fixed-dose hydroxyurea vs MTD with optimisation phase) as protocol-specified outcomes. REACH is registered on ClinicalTrials.gov (NCT01966731) and is ongoing. FINDINGS: We enrolled 635 children between July 4, 2014, and Nov 11, 2016. 606 children were given hydroxyurea and 522 (86%; 266 [51%] boys and 256 [49%] girls) received treatment for a median of 93 months (IQR 84-97) with 4340 patient-years of treatment. The current (Oct 5, 2023) mean dose is 28·2 (SD 5·2) mg/kg per day with an increased mean haemoglobin concentration (7·3 [SD 1·1] g/dL at baseline to 8·5 [1·5] g/dL) and mean fetal haemoglobin level (10·9% [SD 6·8] to 23·3% [9·5]) and decreased absolute neutrophil count (6·8 [3·0] × 109 cells per L to 3·6 [2·2] × 109 cells per L). Incidence rate ratios (IRR) comparing MTD with fixed-dose hydroxyurea indicate decreased vaso-occlusive episodes (0·60; 95% CI 0·52-0·70; p<0·0001), acute chest syndrome events (0·21; 0·13-0·33; p<0·0001), recurrent stroke events (0·27; 0·07-1·06; p=0·061), malaria infections (0·58; 0·46-0·72; p<0·0001), non-malarial infections (0·52; 0·46-0·58; p<0·0001), serious adverse events (0·42; 0·27-0·67; p<0·0001), and death (0·70; 0·25-1·97; p=0·50). Dose-limiting toxicity rates were similar between the fixed-dose (24·1 per 100 patient-years) and MTD phases (23·2 per 100 patient-years; 0·97; 0·70-1·35; p=0·86). Grade 3 and 4 adverse events were infrequent (18·5 per 100 patient-years) and included malaria infection, non-malarial infections, vaso-occlusive pain, and acute chest syndrome. Serious adverse events were uncommon (3·6 per 100 patient-years) and included malaria infections, parvovirus-associated anaemia, sepsis, and stroke, with no treatment-related deaths. INTERPRETATION: Hydroxyurea dose escalation to MTD with dose optimisation significantly improved clinical responses and treatment outcomes, without increasing toxicities in children with sickle cell anaemia in sub-Saharan Africa. FUNDING: US National Heart, Lung, and Blood Institute and Cincinnati Children's Research Foundation.