The aim of this study was to determine the steroidogenic endocrine disrupting effect of the three most widely used serotonin-noradrenaline reuptake inhibitors duloxetine, venlafaxine and tramadol, using two in vitro models, the H295R assay and a recombinant CYP17 enzyme assay. Steroid hormones were quantified using LC-MS/MS. Duloxetine showed endocrine disrupting effects at 5-20μM with CYP17 being the main target. Venlafaxine also affected the steroidogenesis, mainly by affecting the CYP17 lyase reaction, although at much higher concentrations i.e. 100μM. Tramadol only exerted minor effects on the steroidogenesis with the lowest observed effect at 314μM. Based on the H295R results, the inhibition of CYP17 by duloxetine and venlafaxine was investigated in a recombinant CYP17 assay with the use of the 4 major CYP17 substrates pregnenolone, progesterone, 17α-hydroxypregnenolone and 17α-hydroxyprogesterone. Both duloxetine and venlafaxine inhibited CYP17 enzyme activity, but duloxetine was most potent. IC50-values were in the range 5.3-21μM for duloxetine and 1318-2750μM for venlafaxine. Overall, results from the recombinant CYP17 assay confirmed the results from the H295R cell assay. Using testosterone as end point, the margin of safety (defined as NOAEL/Cmax) for duloxetine was 1.6 indicating that duloxetine may have endocrine disrupting effects. In contrast, venlafaxine and tramadol showed higher margins of safety (venlafaxine: 24; tramadol: 157) indicating a lower potential to disrupt the human steroidogenesis.