Background: Acute Lymphoblastic Leukaemia (ALL) , most common malignancy amongst children with front-line treatment considered major success, 20% of children predicted to either relapse or show resistance to treatment with reported dismal outcomes. Aim: To evaluate clinical characteristics of children diagnosed with refractory / relapsed ALL and to determine 3-year overall survival ouctomes. Method: Retrospective review of patients (aged 1 -14 years) diagnosed with ALL during the period January 2002 to December 2018, data extracted for details of baseline characteristics at diagnosis and at relapse . Results: Total of 347 newly diagnosed children with ALL identified, three induction failures and 28 relapses, with total 31 patients a cohort relapse rate of 9% observed. The male-to-female ratio observed is 4.16:1 and mean duration of CR1 was 26 months : 15 (48%) relapsing ≤ 18 months,seven (23%) during 18 to 36 months and nine (29%) relapsed > 36 months of IF or CR1.Eighteen patients (62%) had isolated BM relapse, six (20%) patients experience isolated Extra-medullary relapse and five (17%) patients experienced BM with other sites. Three-year Overall Survival (OS) of the cohort was 62.3%, while of those patients who achieved CR post first-salvage therapy 3-year OS of 79.5% observed with a statistically significant difference, p value <0.05 comparing to patients who did not achieve remission post first-salvage therapy (3-year OS: 46.4%). The same statistical difference observed in 3-years OS observed comparing duration of remission of CR prior to relapse: ≤ 18 months: 33.2%; 18 – 36 months: 66.7% and > 36 months: 87.5%; the same trend continued when comparing 3-years OS based on risk stratification at relapse: LR: 83.3%; IR: 80% and HR: 44.8%. Conclusion: Incidence and outcomes reported on this study is comparable to internationally reported data with duration of CR1, risk-stratification at relapse and remission status post-salvage therapy determined as significant prognostic factors for survival. No survivial difference amongst patients who received HSCT after induction to those who received chemotherapy, could be attributed to a smaller sample size warranting a multi-institutional observational study. The findings corroborates with the need for novel therapies and treatment approaches in these group of patients