Objective of the study To identify the prevalence of potential drug-drug interactions between hospital pharmacy dispensed anti-cancer agents and community pharmacy dispensed drugs. Setting A retrospective cohort study was conducted on the haematology/oncology department of the internal medicine ward in a large teaching hospital in Amsterdam, the Netherlands. Method Prescription data from the last 100 patients treated with anti-cancer agents were obtained from Paracelsus, the chemotherapy prescribing system in the hospital. The community pharmacy dispensed drugs of these patients were obtained by using OZIS, a system that allows regionally linked pharmacies to call up active medication on any patient. Both medication lists were manually screened for potential drug-drug interactions by using several information sources on interactions, e.g. Pubmed, the Flockhart P450 table, Micromedex and Dutch reference books. Main outcome measure Prevalence of potential drug-drug interactions between anti-cancer agents provided by the hospital pharmacy and drugs dispensed by the community pharmacy. Results Ninety-one patients were included in the study. A total of 31 potential drug-drug interactions were found in 16 patients, of which 15 interactions were clinically relevant and would have required an intervention. Of these interactions 1 had a level of severity ≥D, meaning the potential drug-drug interaction could lead to long lasting or permanent damage, or even death. The majority of the interactions requiring an intervention (67%) had a considerable level of evidence (≥2) and were based on well-documented case reports or controlled interaction studies. Most of the potential drug-drug interactions involved the antiretroviral drugs (40%), proton pump inhibitors (20%) and antibiotics (20%). The anti-cancer drug most involved in the drug-drug interactions is methotrexate (33%). Conclusion This study reveals a high prevalence of potential drug-drug interactions between anti-cancer agents provided by the hospital pharmacy and drugs dispensed by the community pharmacy. It shows us there is need for an optimal medication surveillance mechanism to detect potential drug-drug interactions between these two groups of medication, especially because of the high toxicity of anticancer drugs and thus the severe consequences these interactions can have for the patient.