Investigating markers of accelerated aging in patients with Treatment Resistant Depression

Mood disorders affect 5.4% of the general population, exert a substantial socioeconomic burden and significantly impact the patients’ quality of life. Mood disorders are associated with a reduced life expectancy compared to the general population (up to 20 years), and this is largely accounted for by a higher incidence of age-related disorders, in particular metabolic and cardiovascular illnesses, and suicide. This evidence supported the hypothesis that accelerated biological aging might be a feature of mood disorders, but findings have been so far controversial, and less is known about the role and clinical utility of aging markers in the management of treatment response. Pharmacological treatments represent the main approach in mood disorders, but treatment resistance concerns 10–30% of patients with major depressive disorder (MDD) and 75% of cases of unresolved morbidity of bipolar disorder (BD) and, as such, the dentification of reliable response biomarkers represent an unmet need. One of the most studied and effective approaches to manage TRD patients is the electroconvulsive therapy (ECT), but its use is hampered by its invasive nature and side effects. Moreover, the underlying biology of its mechanism of action has yet to be understood. The general aim of this thesis was to explore markers of accelerated aging, namely leucocyte telomere length (LTL) and mitochondrial DNA copy numbers (mtDNAcn), in patients with TRD and in response to ECT. To achieve this aim, we performed two studies. The first study involved a sample of 148 patients with TRD (125 with MDD, and 23 with BD) treated with ECT, where we aimed at: a) exploring if LTL was a marker of TRD by comparing patients with TRD with non-psychiatric controls (NPC); b) investigating if baseline LTL could predict response to ECT; c) exploring the role of genetic variance in this association using genome wide genotyping data from a sub-group of 107 TRD patients. Findings from this study showed that LTL was negatively correlated with age (ρ=-0.25, p<0.0001) and significantly shorter in TRD patients with either MDD (F=35.18, p<0.0001) or BD (F=20.84, p<0.0001) compared to NPCs. Conversely, baseline LTL was not associated with response to ECT or remission. Moreover, we did not detect any significant overlap between genetic variants or genes associated with LTL and response to ECT. The second study was performed on a sub-sample of 31 TRD patients from the previous study who had been treated with ECT and followed prospectively for one month after the end of the ECT session. Blood samples were collected at two different timepoints, before ECT and one month after the end of ECT session. Here we also tested mtDNAcn. A sample of 65 NPCs was also included. The longitudinal nature of the study allowed us to explore if the changes in depression severity observed after ECT were correlated with changes in LTL and mtDNAcn. We showed that TRD patients had significantly shorter LTL and higher mtDNAcn compared to NPCs (respectively: t=-2.94, p=0.002; t=7.36, p<0.001). In the TRD sample, LTL was inversely correlated with MADRS scores at baseline. Difference in MADRS scores between baseline and T2 was significantly inversely correlated with difference in LTL (r=-0.40; p=0.036), suggesting that patients with worse improvement in symptoms also show larger shortening in LTL. Changes in mtDNAcn were not associated with response to ECT and were not correlated with LTL. In conclusion, results from these studies suggest premature cell senescence in patients with severe depression, and that LTL and mtDNAcn may constitute disease biomarker for TRD. A potential implication of telomere dynamics in the ECT response can be suggested. Nevertheless, our limited understanding of the biological significance of mtDNAcn and the limited sample size of our studies require further investigation to better delineate the involvement of aging markers in TRD and response to ECT.