Abstract Background Androstenedione and 17-hydroxyprogesterone are measured via immunoassays or high-performance liquid chromatography tandem mass spectrometry (LC-MS/MS) in clinical laboratories. Immunoassays have limited sensitivity and specificity for certain steroid hormones testing. We aimed to develop and validate a sensitive, accurate, and specific in-house LC-MS/MS assay to simultaneously measure both 17-hydroxyprogesterone and androstenedione in serum. Methods 13C-labeled androstenedione and 17-hydroxyprogesterone internal standards were added to calibrator, control, and patient samples. A hexane/ethyl acetate solution was used in a liquid-liquid extraction procedure. The organic layer was then removed and dried followed by reconstituting the extracts in 2.5 mM ammonium formate in 50% methanol. The HPLC mobile phase gradient began with 47%, 2.5 mM ammonium formate in water and 53%, 2.5 mM ammonium formate in methanol. This progressed to 56% of the methanol phase over three minutes. The mixtures were separated though an Agilent 1260 Infinity HPLC system with a Poroshell 120 EC-C18 analytical column (2.1 × 50 mm × 2.7 µm). The separated HPLC eluents then entered an Agilent 6460C QQQ mass spectrometer in positive ion mode through an electrospray ionization source where protonated androstenedione (m/z = 287.2) with the protonated androstenedione-[2, 3, 4-13C3] internal standard (m/z = 290.2) and protonated 17-hydroxyprogesterone (m/z = 331.2) with the protonated 17-hydroxyprogesterone-[2, 3, 4-13C3] (m/z = 334.2) were mass-selected for collision-induced dissociation. For androstenedione, respective quantifier and qualifier m/z values of 287.2 > 97.1 and 287.2 > 109.1 were used while respective quantifier and qualifier m/z values of 331.2 > 97.1 and 331.2 > 109.1 were used for 17-hydroxyprogesterone. Validation studies were performed including imprecision, lower limit of quantitation (LLOQ), analytical measurement range (AMR), interference, specificity, carryover, extraction recovery, matrix effect, and comparison with a reference laboratory LC-MS/MS method. The total allowable error limits used in this study were 23.5% for androstenedione and 30.2% for 17-hydroxyprogesterone. Results Total imprecision results across three concentrations yielded a maximum CV of 4.6 % for androstenedione and 6.5% for 17-hydroxyprogesteron. The AMRs of androstenedione and 17-hydroxyprogesterone were 0.2 to 13.4 and 0.1 to 22.0 ng/mL without significant carryovers. No significant interferences were observed from 11-deoxycorticosterone (700 ng/dL), testosterone (2500 ng/dL), DHEA (1000 ng/mL), bilirubin (30.1 mg/dL), hemoglobin (1032 mg/dL), or triglycerides (1858 mg/dL). The extraction recoveries of androstenedione and 17-hydroxyprogesterone were 98% and 81%, respectively. No significant matrix effects were identified. In comparison to the reference lab LC-MS/MS method, the in-house LC-MS/MS assay demonstrated a 0.523% bias for androstenedione and a -3.77% bias for 17-hydroxyprogesterone with both exhibiting excellent correlations with the reference lab measurements. More specifically, our results (as a function of values obtained from the reference lab) followed the relationship of y = 1.039x - 0.092 (R = 0.9951) for androstenedione and y = 0.957x + 0.01 (R = 0.9974) for 17-hydroxyprogesterone. Conclusions An LC-MS/MS assay is developed for simultaneous detection of serum androstenedione and 17-hydroxyprogesterone. The assay demonstrates acceptable imprecision, AMR, and accuracy. No significant carryover and interference from other structurally similar steroid hormones and other common interferents are identified.