A linear, physiologically based pharmacokinetic model has been used to study drug delivery characteristics of anticancer drugs after both iv and intra-arterial infusion. It confirms the general belief that intra-arterial infusion produces an increase in local tissue levels and a reduction in systemic drug availability. The increase in local drug concentration depends largely on the blood flow rate of the infusing artery and the rate of drug elimination by the rest of the body. A low arterial blood flow rate and a high drug elimination rate will ensure a high local drug level. On the other hand, the reduction of systemic drug delivery depends on the ability of the infused region to eliminate the drug. Lower systemic availability will be obtained if a substantial amount of the drug is removed during the first pass through the infused region. A review of the clinical response rates of five cytotoxic drugs (Adriamycin, bleomycin, 5-fluorouracil, melphalan, and methotrexate) infused intra-arterially as single agents indicates that, despite a general belief in the advantages of intra-arterial infusion, only 5-fluorouracil shows a significant increase in response rate in the treatment of cancer when compared to systemic routes of administration. There appears to be no justification for the higher costs and substantial risks of intra-arterial infusion for the other drugs due to the lack of controlled studies and the lack of a substantial increase in the response rates. It is suggested that a future controlled study be required to validate the claims of superiority of intra-arterial infusion therapy.