Differential biomarker signatures in unipolar and bipolar depression: A machine learning approach

Abstract

Objective: This study used machine learning techniques combined with peripheral biomarker measurements to build signatures to help differentiating (1) patients with bipolar depression from patients with unipolar depression, and (2) patients with bipolar depression or unipolar depression from healthy controls. Methods: We assessed serum levels of interleukin-2, interleukin-4, interleukin-6, interleukin-10, tumor necrosis factor-α, interferon-γ, interleukin-17A, brain-derived neurotrophic factor, lipid peroxidation and oxidative protein damage in 54 outpatients with bipolar depression, 54 outpatients with unipolar depression and 54 healthy controls, matched by sex and age. Depressive symptoms were assessed using the Hamilton Depression Rating Scale. Variable selection was performed with recursive feature elimination with a linear support vector machine kernel, and the leave-one-out cross-validation method was used to test and validate our model. Results: Bipolar vs unipolar depression classification achieved an area under the receiver operating characteristics (ROC) curve (AUC) of 0.69, with 0.62 sensitivity and 0.66 specificity using three selected biomarkers (interleukin-4, thiobarbituric acid reactive substances and interleukin-10). For the comparison of bipolar depression vs healthy controls, the model retained five variables (interleukin-6, interleukin-4, thiobarbituric acid reactive substances, carbonyl and interleukin-17A), with an AUC of 0.70, 0.62 sensitivity and 0.7 specificity. Finally, unipolar depression vs healthy controls comparison retained seven variables (interleukin-6, Carbonyl, brain-derived neurotrophic factor, interleukin-10, interleukin-17A, interleukin-4 and tumor necrosis factor-α), with an AUC of 0.74, a sensitivity of 0.68 and 0.70 specificity. Conclusion: Our findings show the potential of machine learning models to aid in clinical practice, leading to more objective assessment. Future studies will examine the possibility of combining peripheral blood biomarker data with other biological data to develop more accurate signatures.

Neutrophil-to-Lymphocyte Ratio, a Novel Inflammatory Marker, as a Predictor of Bipolar Type in Depressed Patients: A Quest for Biological Markers

“…After remission of depression, patients with BD had elevated IL-4 and TNF [ 74 ]. Wollenhaupt-Aguiar et al (2020) showed that the most relevant predictor markers to differentiate bipolar from unipolar depression were IL-10, thiobarbituric acid reactive substances, and IL-4 [ 75 ].…”

Biological Pathways Associated with Neuroprogression in Bipolar Disorder

“…Recently, a composite biomarker panel was able to discriminate between patients with rapid-cycling BD and healthy individuals and between manic and depressive states [22]. It was also shown that machine learning techniques, coupled with peripheral biomarkers, provided a potential diagnostic tool to aid in distinguishing depressed patients with BD from major depressive disorder [23]. The field has been investigating several biological pathways, mostly associated with inflammation, oxidative stress, neurotrophic factors, and cellular resilience.…”

Personalized Medicine Using Neuroimmunological Biomarkers in Depressive Disorders

“…In addition, patients with unipolar depression and healthy controls also showed an AUC of 0.74, with 0.69 sensitivity and 0.70 specificity using seven variables (IL-6, carbonyl, BDNF, IL-10, IL-17A, IL-4, and TNF-α). [94]. However, all the patients in this study were under medication, and drug-free patients were not included, although there is a possible association between peripheral markers and pharmacotherapy.…”

“…Although both BD and MDD are associated with an overall increase in cytokine and chemokine levels, these research findings show that BD and MDD patients have different immune profiles and that these profiles can be used for differential diagnosis with high accuracy. Poletti et al used a k-fold nested cross-validation procedure [95] when compared with the leave-one-out cross-validation in a previous study [94] to improve the model accuracy.…”