Stress-related mental disorders : an exploration astrocytic biomarkers, comorbidities, and cognition

Background

Prolonged exposure to stressors without sufficient recovery can lead to physical and mental symptoms. In Sweden, individuals with symptoms related to chronic stress may receive diagnoses such as stress-induced exhaustion disorder (SED) or depression. As emerging research suggests that both SED and depression may be associated with subsequent cognitive impairments, a better understanding of the interplay between SED, depression and cognitive health is warranted.

The overall aim of this thesis was to deepen the understanding of potential pathophysiological effects of chronic stress on the brain, to investigate whether SED and depression are associated with cognitive impairment and dementia, and to examine differences in the comorbidity patterns of SED and depression.

Methods

Study 1 and 2 recruited participants from two psychiatric clinics in Stockholm (2014-2018). Women and men (18-65 years) diagnosed with SED (as classified in ICD-10-SE) or depression (as classified in ICD-10) within the past three months were invited if diagnosed within the past three months. Exclusion criteria included any somatic or psychiatric condition that could better explain their symptoms. If patients fulfilled both SED and depression they were excluded if the physician did not judge that the depressive symptoms were secondary to the symptoms of SED. Healthy controls were recruited via Statistics Sweden (SCB) in 2009.

Study 1 examined whether astrocyte-derived extracellular vesicles (EVs) could be detected in peripheral blood of individuals with SED, and if so, whether concentrations of these differed between individuals with SED, depression, and healthy controls. EVs were quantified using flow cytometry and labeled against glial fibrillary acidic protein (GFAP) and aquaporin 4 (AQP4) to verify astrocyte origin.

Study 2 further explored the suggested impact of stress on neurons and astrocytes, using biofluid markers ("biomarkers") of brain injury. Levels of S100B and neurofilament light (NfL) in peripheral blood were analyzed by commercially available enzyme-linked immunosorbent assays (ELISAs).

Studies 3 and 4 used ICD-10 codes from the Stockholm Regional Health Care Data Warehouse (VAL databases), covering primary care, outpatient specialist care, and hospital care in Region Stockholm. Included individuals (1.36 million) had no history of SED or depression in 2011, were aged 18-65 years in 2011, and remained in Region Stockholm during the follow up. In Study 3, individuals diagnosed with mild cognitive impairment (MCI) or dementia in 2012-2013 were excluded.

Study 3 examined the association between SED, depression, or both (2012- 2013), and later diagnoses of MCI, other forms of dementia and Alzheimer's disease (AD) (2014-2022). Odds ratios (ORs) with 99% confidence intervals (CIs) were calculated, adjusting for age, sex, neighborhood socioeconomic status, diabetes, and cardiovascular disorders.

Study 4 examined the association between SED, depression, or both (2012- 2013), and psychiatric disorders and post-viral fatigue during 2014-2022. ORs with 99% confidence intervals, were adjusted for age and neighborhood socioeconomic status.

Results Study 1 and 2 included 31 individuals with SED, 31 with depression and 61 healthy controls.

In Study 1 higher concentrations of astrocyte-derived EVs were observed in individuals with SED compared to both individuals with depression and healthy controls. Individuals with depression showed higher levels of GFAP-positive EVs and EVs co-expressing AQP4/GFAP compared to controls.

Study 2: Levels of S100B were elevated in individuals with SED compared to those with depression and controls, though only in women. Levels of S100B positively correlated with symptoms of cognitive failure, depressive symptoms and levels of astrocyte-derived EVs. Plasma levels of NfL did not differ between groups or correlate with symptom severity.

Study 3 included 4 346 individuals diagnosed with SED, 40 101 with depression, and 1 898 with both conditions (2012-2013). Individuals with SED had increased risk for AD (OR 2.45, CI 1.22-4.91) and MCI (OR 1.87, CI 1.20-2.91). Individuals with depression had similar risk for AD (OR 2.32, CI 1.85-2.90) but higher risk for MCI (OR, 2.85 CI 2.53-3.22). The risk was highest in individuals with both SED and depression, with an OR 4.00 (CI 1.67-9.58) for AD and OR 3.87 (CI 2.39-6.27) for MCI.

Study 4 included 4 347 individuals diagnosed with SED, 40 134 with depression, and 1 902 with both conditions (2012-2013). Individuals with prior SED showed stronger associations with stress-related diagnoses than those with depression. Notable associations included acute stress reaction (OR 3.08, CI 2.79-3.41), unspecified reaction to severe stress (OR 4.08, CI 3.67-4.53), post-COVID-19 (OR 2.74, CI 2.06-3.66) and post-viral fatigue syndrome (OR 5.13, CI 4.19-6.28). In contrast, depression was more strongly associated with various psychiatric disorders. The highest associations between depression and other psychiatric diagnoses (OR > 5) were found for post-traumatic stress disorder (PTSD) (OR 6.44, CI 6.08-6.82), substance use disorders (OR 5.41, CI 5.11-5.73), manic episodes, bipolar affective disorder, persistent mood disorder, (OR 7.86, CI 7.44- 8.31, neurotic disorder (OR 5.86, CI 5.70-6.02), borderline personality disorder (OR 9.72, CI 8.88-10.64), autism spectrum disorders (OR 6.69 CI 6.20- 7.22), attention deficit hyperactivity disorder (ADHD) (OR 6.39, CI 6.09- 6.70), and suicide attempts (OR 6.77, CI 6.13-7.48), alcohol-related disorders (OR 4.02, CI 3.83-4.21) and schizophrenia, schizotypal disorders, delusional disorders (OR 2.87, CI 2.63-3.13). Notably, some conditions, including PTSD, autism spectrum disorders and ADHD, also showed strong associations with SED (OR >3.5). Conclusion This thesis adds to the knowledge about possible changes in the brain in individuals diagnosed with SED and indicates prognostic and comorbid differences between SED and depression. However, due to methodological limitations, causal relationships cannot be established.

First, astrocyte-derived markers were elevated in peripheral blood in individuals with SED compared to healthy controls (Study 1 and 2). Second, SED and depression were both associated with an increased risk of MCI and AD, with the highest risk observed in individuals diagnosed with both conditions (Study 3). Third, SED and depression exhibit distinct and divergent patterns of psychiatric comorbidity (Study 4).

Overall, this thesis supports the clinical relevance of distinguishing SED and depression as separate clinical phenotypes, and the possibility that SED and depression are both related to cognitive decline. Furthermore, our results suggests that glial cells are involved, though it does not seem to affect neurons or axonal integrity based on the biomarker results, in these conditions. Given the heterogeneity of these diagnoses, their suggested heightened risk for MCI and AD as well as differing comorbid patterns, their relation should be further explored.

List of scientific papers

I. Wallensten J, Nager A, Åsberg M, Borg K, Beser A, Wilczek A, Mobarrez F. Leakage of astrocyte-derived extracellular vesicles in stress-induced exhaustion disorder: a cross-sectional study. Sci Rep. 2021 Jan 21;11(1):2009. Erratum in: Sci Rep. 2023 Jun 23;13(1):10211. https://doi.org/10.1038/s41598-021-81453-8

II. Wallensten J, Mobarrez F, Åsberg M, Borg K, Beser A, Wilczek A, Nager A. Plasma levels of S100B and neurofilament light chain protein in stress-related mental disorders. Sci Rep. 2022 May 18;12(1):8339. https://doi.org/10.1038/s41598-022-12287-1

III. Wallensten J, Ljunggren G, Nager A, Wachtler C, Bogdanovic N, Petrovic P, Carlsson AC. Stress, depression, and risk of dementia - a cohort study in the total population between 18 and 65 years old in Region Stockholm. Alzheimers Res Ther. 2023 Oct 2;15(1):161. https://doi.org/10.1186/s13195-023-01308-4

IV. Wallensten J, Ljunggren G, Nager A, Wachtler C, Petrovic P, Carlsson AC. Differences in psychiatric comorbidity patterns in patients diagnosed with chronic stress-induced exhaustion disorder and depression - A cohort study in the total population of Region Stockholm. J Affect Disord. 2024 Apr 15;351:765-773. https://doi.org/10.1016/j.jad.2024.01.273