Associations among serum markers of inflammation, life stress and suicide risk in patients with major depressive disorder

Highlights

• Adulthood adversity was associated with suicide risk in MDD patients.

• Patients with suicide risk had higher level of inflammatory markers.

• The association between adulthood adversity and suicide risk partly depended on inflammatory activation.

Abstract

Background

Patients with major depressive disorder (MDD) are at high risk for suicide. As the worst outcome of MDD and common self-concealment in patients with suicide risk, studies of biomarkers may provide useful tools for suicide prevention and treatment.

Methods

This study recruited 168 patients with MDD from the Objective Diagnostic Markers and Personalized Intervention in MDD patients (ODMPIM), including 50 patients with suicide risk. Based on previous evidence and hypothesis, 23 targeted serum biomarkers involving immune-inflammation, neurotrophins, hypothalamic–pituitary–adrenal (HPA) axis and metabolism, were measured. We used path analysis and principal components analysis (PCA) to clarify the associations among serum biomarkers, childhood adversities, adulthood life events, severity of depression and suicide risk.

Results

We identified that patients with suicide risk had a higher level of inflammatory markers in serum than patients without suicide risk (P < 0.001), especially chemokine (C-X-C motif) ligand 1 (CXCL-1). After using the Bonferroni correction, there were no differences in biomarkers related to neurotrophins, HPA-axis and metabolism. In addition, a higher proportion of patients with suicide risk had adulthood adversity (assessed by Life Events Scale) (P = 0.003). Intriguingly, path analysis demonstrated that the association between adulthood adversity and suicide risk mainly depended on severity of depression and inflammatory index.

Conclusion

This study highlights the possible role of inflammation involved in suicide risk of MDD patients. Inflammatory markers have the potential for early identification and then reducing suicidal behaviors or becoming novel treatment targets in suicide risk management.

Introduction

Suicide - a fatal act of intentionally causing one's own death - is the worst outcome of major depressive disorder (MDD) (Turecki and Brent, 2016). The clinical spectrum of suicidal behavior is heterogeneous and varies in the degree of suicidal intent, severity and adversely impacts clinical outcome. Noteworthy, troubling evidence suggests that a considerable percentage of suicidal individuals hide their suicidal intent from others (Friedlander et al., 2012). Reliable predictors of completed suicides have included prior suicide attempts (Da Cruz et al., 2011) and life-time or current history of psychiatric disorders, including MDD (Arsenault-Lapierre et al., 2004). In parallel with international literature, a comprehensive meta-analysis of Chinese MDD patients indicated worrisome trends with pooled lifetime prevalence rates of 53.1% (from 11 studies), 17.5% (from 4 studies) and 23.7% (from 18 studies) for suicidal ideation, plan and attempts, respectively (Dong et al., 2018).

Despite considerable linking suicide with MDD, identifying individuals at risk has remained a challenge (Hawton et al., 2013). Younger age (Chen et al., 2014; Holma et al., 2010), male gender (Chen et al., 2014; Hawton et al., 2013), cigarette smoking (Oquendo et al., 2007), unemployment (Chen et al., 2014), aggressive traits (Keilp et al., 2006), hopelessness (Hawton et al., 2013), low social support (Holma et al., 2010; Lin et al., 2020), exposure to early-life adversities (Brodsky et al., 2001) and stressful life events (Lin et al., 2018; Oquendo et al., 2013; Park et al., 2015) have all been implicated as risk factors. In addition to the diagnosis of depression and presence of melancholia (Hawton et al., 2013) and psychotic symptoms (Chen et al., 2014), other psychiatric disorders (psychosis, anxiety (Hawton et al., 2013), borderline personality disorder (Soloff et al., 2000), substance abuse disorders (Dalca et al., 2013; Hawton et al., 2013)) and medical comorbidities (such as diabetes) (Goodwin et al., 2003; O'Connor and Nock, 2014) have also been identified as key risk factors. Key questions still remain unanswered including how long does this risk last, how it evolves over the entire period of a depressive episode and at what point can we say when such risk has abated.

It is likely that biomarkers reflective of pathophysiological processes leading to the expression of suicidal behavior in MDD patients would be of great help. Several strategies have been applied to investigate the biomarker profile of suicide. Peripheral biomarker studies have attracted considerable interest due to advantages, such as availability, lower costs and lower invasiveness of the procedures. Findings from peripheral biomarkers have highlighted that serum and plasma biomarkers may help recognize suicide risks in MDD patients. Recent data of peripheral biomarkers for suicide in MDD have implicated enhanced inflammatory activity (Grudet et al., 2014; Janelidze et al., 2011; O'Donovan et al., 2013), altered energy metabolism (Baek et al., 2014; Park et al., 2014; Ruljancic et al., 2011; Segoviano-Mendoza et al., 2018), dysfunctional HPA axis activity (Brunner et al., 2002; McGirr et al., 2011), decreased neurotrophic support (Deveci et al., 2008; Kim et al., 2007; Lee et al., 2007) and reduced monoamine levels (Ruljancic et al., 2011; Spreux-Varoquaux et al., 2001).

However, results from peripheral studies are mixed and do not provide a coherent view. Firstly, a possible explanation for mixed findings could be the biological heterogeneity that underlies MDD and suicidal behavior. Probably, dysfunction of specific biological mechanism is more prominent in distinct subsets of patients, while other mechanisms may be more significant in others. Therefore, single biomarker may have weak predictive ability of suicidality in MDD patients. Secondly, confounding variables such as the severity of depression (Hawton et al., 2013), psychiatric (Dalca et al., 2013; Hawton et al., 2013; Soloff et al., 2000) and physical (Goodwin et al., 2003) comorbidities, medications (Sen et al., 2008), and childhood (Brodsky et al., 2001) and adulthood (Oquendo et al., 2013) adversity that may affect both the risk of suicide and the peripheral levels of biomarkers in MDD patients may need to be accounted for.

In summary, the pathogenesis of suicide is complex, involving clinical, psychological, sociological and biological factors, and these factors interact with each other (Turecki and Brent, 2016). Therefore, this study investigated a comprehensive range of serum markers that were sampled concurrently and have been previously linked to suicide (23 targeted biomarkers reflecting immune, neurotrophic, HPA-axis and metabolic parameters), psychological stress (life adversities in the childhood and adulthood) and clinical characteristics (psychiatric and physical comorbidities, severity of depression) in MDD patients. The primary aim was to test the hypothesis that MDD patients who had suicide risk will demonstrate combinatorial profile of serum biomarkers that discriminate them from those without suicide risk. A secondary aim was to validate the hypothesis that the effects of psychological stress and adversity on increasing the acute/proximal risk of suicide will be mediated by above mentioned serum biomarker profile.