Background: Currently, the treatment decisions in elderly patients with acute myeloid leukemia (AML) are difficult and remain controversial. This study aims to evaluate the effect of thalidomide plus chemotherapy on elderly patients with AML. Methods: 70 elderly AML patients (median age 71 years) were enrolled into this prospective study and randomly assigned to either the control arm (PC, n = 35) or the investigational arm (TPC, n = 35). Patients in the PC arm received a non-intensive regimen composed of cytarabine, aclarubicin and G-CSF (CAG) chemotherapy for induction of remission, and patients in the TPC arm received in addition thalidomide at a maximum dose of 200 mg/day. Results: After 2 courses of induction therapy, complete response rate of TPC and PC arms was 54.3% and 57.1%, respectively (p = 0.810). At the last follow-up, the Kaplan-Meier estimate showed that the median overall survival (OS) and event-free survival (EFS) in patients in the PC arm was inferior to those of patients in the TPC arm. Using a stratified Cox model adjusted for randomized treatment, patients receiving thalidomide plus chemotherapy were shown to derive some survival benefit in both OS and EFS. Overall, the hematological and non-hematological toxicity were similar between the 2 treatment arms. Conclusions: Thalidomide in combination with chemotherapy is an alternative treatment option for elderly patients with AML.