We attempted to determine the effects of prior antineoplastic chemotherapy and age on gentamicin pharmacokinetics in children (age 1–18 yrs) with cancer and in controls, and to establish a protocol for gentamicin dosing and monitoring to ensure rapid attainment of therapeutic serum concentrations in these patients. In a prospective controlled study, patients with fever who were receiving empiric gentamicin for confirmed or suspected infections were separated into three groups: 29 with cancer who were receiving a continuing chemotherapy protocol with nonnephrotoxic antineoplastic agents; 23 with cancer who were receiving a continuing chemotherapy protocol with nephrotoxic antineoplastic agents; and 25 control patients who did not have cancer. Three blood samples (one predose, two postdose concentrations), collected between the third and sixth gentamicin doses from each patient, were analyzed by the Emit assay. Pharmacokinetic parameters were calculated and gentamicin dosages recommended based on the Sawchuk‐Zaske method of serum level interpretation. When normalized by body weight, there was no significant difference in clearance, volume of distribution, and half‐life between the control group and either group of patients with cancer. However, when normalized by body surface area, patients receiving prior nephrotoxic chemotherapy appeared to have a lower mean clearance (98.2 ml/min/1.73 m2) than those exposed to nonnephrotoxic chemotherapy (117.4 ml/min/1.73 m2) and controls (113.3 ml/min/1.73 m2; ANCOVA p=0.033). When kinetic parameters were normalized by body weight, the effect of advancing age yielded a decrease in both clearance (p<0.001) and volume of distribution (p=0.02), and an increase in gentamicin half‐life (p<0.001). When normalized by body surface area, age had no significant effect on clearance (p=0.579). There was no significant difference in gentamicin daily dose requirements (mg/kg) between the chemotherapy groups, which may be due to the lack of significant effects of chemotherapy on gentamicin's volume of distribution and clearance normalized by body weight. The final maintenance doses (mg/kg/day mean ± SD) for patients with cancer were 10.8 ± 1.8 for those age 1–5 years, 8.9 ± 1.1 for those age 6–12 years, and 7.9 ± 1.9 for those age 13–18 years. However, when normalized by body surface area, the age‐dependent doses became remarkably similar for children in all three age groups (ANOVA p=0.932), approximately 250 mg/m2/day. We recommend that pediatric patients with cancer who require treatment for fever and neutropenia be given higher than standard gentamicin dosages to achieve therapeutic serum concentrations promptly. In particular, initial empiric doses of 10 mg/kg/day are appropriate for those age 1–5 years.