*4.4. Biomarkers*

Biomarkers are quantifiable characteristics of biological processes, which could prove helpful in improving diagnostic objectivity of MDD and TRD as well as help in personalizing its treatment. For MDD, no specific biomarkers have ye<sup>t</sup> been found, though several markers have been shown to be potential candidates, such as brain-derived neurotrophic factor (BDNF), interleukins (IL) 1 and 6, tumor necrosis factor (TNF), malondialdehyde (MDA), hypothalamic-pituitary-adrenal (HPA) activity, and cortisol responses [86,87]. Every biomarker as a standalone shows a low sensitivity and specificity, partly explained by the heterogeneity of MDD. To overcome this shortcoming, either examining a biological panel of several markers [88] or phenotyping MDD and TRD into distinct subtypes could be considered. However, a recent meta-analysis showed that only cortisol has a predictive e ffect on onset/relapse and recurrence of MDD making the integration of biomarkers for personalizing TRD treatment a futuristic milestone ye<sup>t</sup> to be discovered [89].

### *4.5. Insights into Symptomatic Improvement after Deep Brain Stimulation*

For TRD, di fferent regions in the mood circuit can be stimulated with DBS (Table 2), although it is still unclear which depressive-symptoms respond to the stimulation of a specific target. More research into the mood circuit is needed to untangle which emotions arise from specific brain regions. This may vary from basic animal research, disentangling neuronal function per brain region, and ultra-high field MR studies in humans, all of which could shed light on the dysfunctional brain circuits in TRD. In contrast to the motor system that is studied thoroughly [90,91], emotional circuitry is far less understood. One reason for this is that animal research into mood circuitry remains complicated as

there is considerable heterogeneity between species [92]. Modeling depression in animals is complex as there are several depressive-like behavior models, such as the chronic unpredictable stress paradigm (CUS), which give insight into depression pathology [93]. DBS is investigated within these models to unravel behavioral and cellular changes following DBS [94].

Alongside the standard clinical rating scales, the use of momentary assessment techniques, such as the experience sampling method (ESM), could enhance the documentation of the momentary mood states [95]. The ESM includes short repeated assessments of experiences and behaviors, as well as moment-to-moment changes in mental states in the context of daily life. Research has shown that depressed patients can improve their depressive symptomology while using weekly ESM for six weeks, and add-on ESM derived feedback resulted in a significant decrease in HDRS scores compared to controls (*p* < 0.01; −5.5 point reduction in HDRS at 6 months) [96]. In add-on-derived feedback, a psychologist or psychiatrist gives feedback on the association between the participants momentary affective states and specific daily life contexts [97]. ESM-derived feedback could further improve treatment by showing within-subject changes in a heterogeneous TRD population and contribute to clinical decision-making [97]. In the case of DBS, the use of ESM may reveal specific response patterns depending on the brain region that is stimulated, which can provide valuable information about emotional circuitry. This can be done using well-evaluated day-to-day scores, including questionnaires that go into detail on current mood and adaptive functioning.
