We evaluated the impact of a prospective audit and feedback antimicrobial stewardship program (ASP) on antibiotic prescription and resistance trends in a hematology-oncology unit in a university hospital (National University Cancer Institute, Singapore [NCIS]). A prospective interrupted time-series study comprising 11-month pre-intervention (PIP) and intervention evaluation phases (IEP) flanking a one-month implementation phase was carried out. Outcome measures included defined daily dose per 100 (DDD/100) inpatient-days of ASP-audited and all antibiotics (encompassing audited and non-audited antibiotics), and the incidence-density of antibiotic-resistant microorganisms at the NCIS. Internal and external controls were DDD/100 inpatient-days of paracetamol at the NCIS and DDD/100 inpatient-days of antibiotics prescribed in the rest of the hospital. There were 580 ASP recommendations from 1,276 audits, with a mean monthly compliance of 86.9%. Significant reversal of prescription trends towards reduced prescription of audited (coefficient = -2.621; 95% confidence interval [CI]: -4.923, -0.319; p = 0.026) and all evaluated antibiotics (coefficient = -4.069; 95% CI: -8.075, -0.063; p = 0.046) was observed. No changes were seen for both internal and external controls, except for the reversal of prescription trends for cephalosporins hospital-wide. Antimicrobial resistance did not change over the time period of the study. Adverse outcomes-the majority unavoidable-occurred following 5.5% of accepted ASP recommendations. Safe and effective ASPs can be implemented in the complex setting of hematology-oncology inpatients.