To evaluate a method using a multifunctional clean-up column coupled with high performance liquid chromatography as an official analytical method for the determination of deoxynivalenol in wheat used as food or feed, an inter-laboratory study was performed in 12 laboratories using four naturally contaminated wheat samples and one spiked sample. The relative standard deviations for repeatability (RSDr) and reproducibility (RSDR) of naturally contaminated wheat were in the range 5.8-11.3% and 12.0-20.7%, respectively. The HORRAT was less than 1.0 in each sample. From the spiking test, the recovery rate, RSDr, RSDR and HORRAT value were 100.0%, 11.2%, 10.3% and 0.5, respectively. The limit of quantification is 0.10 mg/kg from the range obtained in a linear calibration. Thus, it should be useful as a sensitive and validated analytical method for the determination of deoxynivalenol in wheat intended for use in food and feed.