Background/aim: A proliferation-inducing ligand (APRIL) has been investigated as a prognostic marker in chronic lymphocytic leukemia (CLL) patients. However, there is no cut-off level for serum APRIL (sAPRIL) levels that predict time to treatment in CLL patients. Materials and methods: Between May and December 2012, 94 consecutive CLL patients and 25 healthy controls were assessed. sAPRIL levels were measured by ELISA. Demographic data and prognostic markers were obtained from the patients’ files. Treatment-naïve patients were followed up for 6.5 years for any treatment need. Results: Patients were divided into 3 groups: Treatment-naïve (n = 47), chemotherapy receiving (n = 25), and those who had received chemotherapy previously (n = 22). There was no difference in median sAPRIL levels of patients who were receiving chemotherapy at the sampling time and the healthy controls, which indicates that sAPRIL levels might be influenced by treatment. For treatment-naïve patients, the best cut-off in predicting time to treatment was found at the sAPRIL level of 2.04 ng/mL, with 78% sensitivity and 63% specificity. Time to treatment was significantly earlier in the APRIL high group (n = 27) than in the APRIL low group (n = 20) (P = 0.010, log-rank test). Conclusion: sAPRIL, a simple, promising blood test which can be measured by ELISA, will likely obtain a place in the wide range of prognostic markers in CLL. Prospective large-scale studies are required to validate and confirm the feasibility of the proposed cut-off level of 2.04 ng/mL as a predictor of time to treatment in treatment-naïve CLL patients.Key words: Chronic lymphocytic leukemia, treatment, survival, prognosis, chemotherapy