Major depressive disorder is associated with changes in a cluster of serum and urine biomarkers

• Published in: Journal of psychosomatic research Vol. 125; p. 109796
• Main Authors: van Buel, Erin M., Meddens, Marcus J.M., Arnoldussen, Eduard A., van den Heuvel, Edwin R., Bohlmeijer, Willem C., den Boer, Johan A., Muller Kobold, Anna, Boonman-de Winter, Leandra J.M., van Rumpt, Dirk, Timmers, Lambertus F.J., Veerman, Mattheus F.A., Kamphuis, Johannes S., Gladkevich, Anatoliy V., Schoevers, Robert A., Luiten, Paul G.M., Eisel, Ulrich L.M., Bosker, Fokko J., Klein, Hans C.
• Format: Journal Article
• Language: English
• Published: England Elsevier Inc 01.10.2019; Elsevier Science Ltd
• Subjects: Adult; Algorithms; Area Under Curve; Biological markers; Biomarker panel; Biomarkers; Biomarkers - blood; Biomarkers - urine; Case-Control Studies; Depressive Disorder, Major - blood; Depressive Disorder, Major - urine; Depressive personality disorders; Discrimination; ELISA; Female; Humans; Major depressive disorder; Male; Mental depression; Middle Aged; Permutation analysis; Phenotypes; ROC Curve; Serum; Urine; Urine tests; Validity; Urine; Biomarkers; Permutation analysis; Biomarker panel; Major depressive disorder; ELISA
• ISSN: 0022-3999; 1879-1360
• DOI: 10.1016/j.jpsychores.2019.109796

Major Depressive Disorder (MDD) is a heterogeneous disorder with a considerable symptomatic overlap with other psychiatric and somatic disorders. This study aims at providing evidence for association of a set of serum and urine biomarkers with MDD. We analyzed urine and serum samples of 40 MDD patients and 47 age- and sex-matched controls using 40 potential MDD biomarkers (21 serum biomarkers and 19 urine biomarkers). All participants were of Caucasian origin. We developed an algorithm to combine the heterogeneity at biomarker level. This method enabled the identification of correlating biomarkers based on differences in variation and distribution between groups, combined the outcome of the selected biomarkers, and calculated depression probability scores (the “bio depression score”). Phenotype permutation analysis showed a significant discrimination between MDD and euthymic (control) subjects for biomarkers in urine (P < .001), in serum (P = .02) and in the combined serum plus urine result (P < .001). Based on this algorithm, a combination of 8 urine biomarkers and 9 serum biomarkers were identified to correlate with MDD, enabling an area under the curve (AUC) of 0.955 in a Receiver Operating Characteristic (ROC) analysis. Selection of either urine biomarkers or serum biomarkers resulted in AUC values of 0.907 and 0.853, respectively. Internal cross-validation (5-fold) confirmed the association of this set of biomarkers with MDD. •This study investigates biomarker panels for Major Depressive Disorder (MDD).•The biomarker panels were assessed in serum as well as in urine.•A new method was used to combine results of multiple biomarkers into a single score.•This scoring method is based on differences in variation and distribution.•A panel of 9 serum and 8 urine biomarkers was identified to correlate with MDD.