BACKGROUND: Influenza is a health risk to children receiving chemotherapy for cancer. An absolute lymphocyte count (ALC) >1,000 cells/mm(3) has been associated with the ability to produce an immune response to influenza vaccine during chemotherapy. However, clinical efficacy of influenza vaccination during chemotherapy remains unclear. PROCEDURE: We conducted a prospective cohort study in children receiving chemotherapy for cancer during two consecutive influenza seasons. Assessments of immune cells and serologic response were measured immediately before and after receiving influenza vaccine. Patients were monitored for influenza or influenza-like illness (ILI). RESULTS: Two hundred fifty-nine patients were studied over 2 years. The seroresponse rate was 62% (98/157). The median ALC at vaccination was higher in seroresponders than nonresponders, 854 cells/mm(3) versus 602 cells/mm(3) , respectively (P < 0.036). Univariate analysis showed that patients with an ALC <1,000 cells/mm(3) at the time of vaccination were twice as likely to be sero-nonresponders (P < 0.02, OR = 2.4, 95% CI: 1.1-5.0). Twelve percent (31/259) of patients developed influenza, of whom all had fever at presentation, 26% (8/31) required hospitalization, and 81% (25/31) had chemotherapy delays. No deaths were associated with influenza infection. The proportion of patients with influenza was not different between seroresponders and nonresponders. CONCLUSIONS: Influenza infection following immunization remains a source of morbidity in children undergoing chemotherapy. Lymphopenia at vaccination predicted sero-nonresponse. Seroresponse was not associated with a decreased frequency of influenza infection or ILI when compared to sero-nonresponders, suggesting clinical effectiveness of vaccination is likely multifactorial. Further investigation into the efficacy of the influenza vaccine is needed to refine immunization recommendations.