Elsevier

Psychoneuroendocrinology

Volume 78, April 2017, Pages 105-113
Psychoneuroendocrinology

Can C-reactive protein inform antidepressant medication selection in depressed outpatients? Findings from the CO-MED trial

https://doi.org/10.1016/j.psyneuen.2017.01.023Get rights and content

Highlights

  • Pre-treatment C-reactive protein (CRP) levels predict differential response to currently available antidepressant treatments.

  • Depressed patients with low CRP level (<1 mg/L) respond better to SSRI monotherapy whereas those with higher levels respond better to combination of bupropion and SSRI.

  • A CRP threshold (< or ≥1 mg/L) based treatment assignment, as compared to random treatment allocation, will require treatment of 8.6 depressed patients for 1 additional remission.

Abstract

Objective

Currently, no valid measures inform treatment selection for depressed patients. Whether C-reactive protein (CRP) in particular and two other acute phase reactants (inflammatory markers) could differentiate between patients responding to either of two treatments with different mechanisms of action was assessed.

Method

Subjects included Combining Medications to Enhance Depression Outcomes (CO-MED) trial participants randomly assigned to either escitalopram plus placebo (SSRI monotherapy, n = 51) or bupropion plus escitalopram combination (bupropion-SSRI combination, n = 55) with baseline plasma samples. CRP, serum amyloid P component, and alpha-2-macroglobulin were measured using the Bioplex Pro™ human acute-phase 4-plex panel. We conducted mixed model analyses of depressive symptom (Quick Inventory of Depressive Symptomatology Self-Report) and side-effect burden (Frequency, Intensity, and Burden of Side-Effects Rating Scale) obtained weekly or every other week over the 12-week acute-phase of CO-MED trial to evaluate the relationship between these outcomes and baseline CRP and other acute-phase reactants.

Results

The treatment arms did not differ in depressive symptom or side effect outcomes. Most participants (69.8%, 74/106) had baseline CRP levels greater than 1 mg/L (indicative of systemic inflammatory activity). Higher baseline CRP levels were associated lower depression severity (correlation coefficient = −0.63) with bupropion-SSRI combination but not with SSRI monotherapy (correlation coefficient = 0.40). The overall remission rate was 41.5%. The estimated remission rate with CRP threshold based assignment (SSRI monotherapy for <1 mg/L and Bupropion-SSRI for ≥1 mg/L) was 53.1%, with a number needed to treat of 8.6. Side effect burden was unrelated to any baseline inflammatory marker.

Conclusions

Baseline CRP levels relate differentially to antidepressant treatment outcomes in persons with major depressive disorder.

Clinicaltrials.gov identifier: NCT00590863

Introduction

The current trial-and-error approach of antidepressant medication selection, based on subjective factors such as cost or patients and/or provider preference often necessitates multiple attempts to attain adequate symptomatic control (Rush et al., 2006b). While the search of clinical markers has yet to produce clinically useful results (Arnow et al., 2015), inflammatory biomarkers present a promising avenue to inform antidepressant medication selection. Recently, Uher et al. found that depressed outpatients with levels of C-reactive protein (CRP) below 1 mg/L had greater reduction in depression severity with escitalopram, a selective serotonin reuptake inhibitor (SSRI) medication, whereas those with levels of at least 1 mg/L had better response to nortriptyline, a tricyclic antidepressant (TCA) medication (Uher et al., 2014).

CRP, a plasma protein, synthesized by the liver, is sensitive to inflammatory cytokines and increases markedly (up to 10,000-fold) in acute response to serious infection or tissue injury (hence the label acute-phase reactant) (Pepys and Hirschfield, 2003). The 1 mg/L CRP threshold is accepted as a marker of systemic inflammation that is associated with both vascular and non-vascular mortality (Kaptoge et al., 2010). CRP is a clinically pragmatic measure of inflammation as it is 1) readily available; 2) inexpensive; 3) relatively stable in stored biological specimens; 4) unaffected by time of day or meal intake; and 5) stable year-to-year in individual subjects absent acute factors (Brown et al., 2014, Danesh et al., 2004, Emberson et al., 2004, Pepys and Hirschfield, 2003, Ridker, 2003).

The potential role of systemic inflammation in antidepressant treatment response is suggested by the lower likelihood of response to commonly used antidepressant treatments (Strawbridge et al., 2015) such as psychotherapy (Harley et al., 2010) and SSRI medications (Eller et al., 2008, O’Brien et al., 2007, Yoshimura et al., 2009) in patients with elevated levels of inflammatory biomarkers. Additionally, serotonergic versus noradrenergic antidepressants have been postulated to differentially affect the balance of immune response mediated by type 1 (Th1) versus type 2 (Th2) T-helper cells. Serotonergic drugs promote immune response mediated by Th1 cells whereas noradrenergic antidepressant promote immune response mediated by Th2 cells (Martino et al., 2012). SSRI medications 1) suppress the expression of cluster of differentiation 4 (CD4) glycoprotein on macrophages (Greeson et al.); 2) decrease the production of interleukin (IL) 4, a key cytokine that promotes differentiation of naïve Th cells to Th2 cells (Kubera et al., 2000, Shenoy et al., 2013); 3) suppress the secretion of IL-6 which is produced by Th2 cells (Hannestad et al., 2011); and 4) increase the level of pro-inflammatory cytokines that promote Th1 cell-mediated immune response (IL-1 beta, interferon gamma [IFNγ], and tumor necrosis factor alpha [TNFα]) in the frontal cortex of rodents (Warner-Schmidt et al., 2011). Notably, Warner-Schmidt et al. found that co-administration of non-steroidal inflammatory drugs (NSAIDs) not only blocked the increase in levels of inflammatory cytokines with SSRIs but also antagonized their behavioral effects in animal models of forced swim and tail suspension tests. In the same study, NSAIDs did not inhibit the behavioral effects of bupropion (Warner-Schmidt et al., 2011). Additionally, predominantly noradrenergic antidepressant, such as nortriptyline, but not SSRIs, inhibit immune response mediated by Th1 cells (Sacerdote et al., 1994).

Inflammation is also associated with changes in brain dopamine metabolism (Felger, 2016); dopamine in turn affects immune system, especially immune response mediated by interleukin 17 producing Th cells (labelled Th17) that are distinct from Th1 and Th2 cells (Lieberknecht et al., 2016, Melnikov et al., 2016). Bupropion, a non-serotonergic antidepressant, inhibits dopamine reuptake, increases brain extracellular dopamine concentration (Ascher et al., 1995, Cremers et al., 2016), and has been shown to reduce inflammation mediated by both Th1 (by reducing IL-1 beta, IFNγ, and TNFα) and Th17 cells (Brustolim et al., 2006, Ebbinghaus et al., 2012, Warner-Schmidt et al., 2011). As some depressed patients have low-grade systemic inflammation as evidenced by elevated levels of pro-inflammatory cytokines (Dantzer et al., 2008, Haapakoski et al., 2015) such as IL-6 (Dowlati et al., 2010, Haapakoski et al., 2015) and acute-phase reactants, such as CRP (Wium-Andersen et al., 2013) and haptoglobin (Maes et al., 1994); those with elevated systemic markers of inflammation may respond better to non-serotonergic antidepressants such as bupropion as compared to SSRIs.

This secondary data analysis used the clinical trial and biological data from the CO-MED trial to further evaluate whether pre-treatment CRP level, at a predefined threshold of 1 mg/L, differentially predicts clinically meaningful treatment outcomes (depression severity and side-effects) to SSRI monotherapy versus bupropion-SSRI combination which has noradrenergic, dopaminergic, and serotonergic effects, analogous to nortriptyline used by Uher et al. (2014). We conducted other secondary analyses to determine whether acute-phase reactants- specifically serum amyloid P component, alpha-2-macroglobulin and haptoglobin might also predict differential treatment outcomes.

Section snippets

Study overview and participants

This report relies on data from the Combining Medications to Depression Outcomes trial CO-MED trial (Rush et al., 2011) which recruited 665 treatment-seeking depressed participants who were randomly assigned after stratification for site to three treatment arms: SSRI monotherapy, bupropion-SSRI combination, and venlafaxine-mirtazapine combination (Rush et al., 2011). The present analytic sample included a sub-set of participants with baseline plasma samples who were treated with SSRI

Results

Of the 665 participants in CO-MED trial, 106 participants who were treated with SSRI monotherapy or bupropion-SSRI combination provided plasma and constitute the analytic sample for this report. Table 1 presents the baseline mean concentrations of CRP, log of CRP, alpha-2-macroglobulin, log of alpha-2-macroglobulin, and serum amyloid P component along with select demographic variables; there was no significant difference between the two treatment arms in baseline clinical and biological

Discussion

At baseline, levels of CRP, but NOT alpha-2-macroglobulin or serum amyloid P component levels, were differentially associated with reduction in depression severity between SSRI monotherapy versus bupropion-SSRI combination. Lower levels of CRP were associated with greater depressive symptom reduction with SSRI monotherapy whereas higher baseline CRP levels were associated with better outcomes with bupropion-SSRI combination. We estimated that if CRP threshold based treatment assignment is

Conclusion

Baseline CRP levels can predict differential treatment outcomes to SSRI monotherapy versus bupropion-SSRI combination. Low CRP levels are associated with greater reductions in depression severity with SSRI monotherapy, while with bupropion-SSRI combination, those with higher levels (≥1 mg/L) had a better response. Inflammatory markers such as CRP have a role in clinical practice.

Previous presentation: The findings of this report were presented at the 2016 Annual Meeting of American College of

Acknowledgements

CO-MED trial was funded by National Institute of Mental Health under contract N01 MH-90003 to the University of Texas Southwestern Medical Center at Dallas (Principal Investigators, M. H. Trivedi and A.J. Rush). This work was also supported in part through the Center for Depression Research and Clinical Care at UT Southwestern (Principal Investigator: Madhukar H. Trivedi, MD) and Hersh Foundation. The authors thank the clinical staff at each clinical site for their assistance with this project;

References (49)

  • A.J. Rush et al.

    An evaluation of the quick inventory of depressive symptomatology and the hamilton rating scale for depression: a sequenced treatment alternatives to relieve depression trial report

    Biol. Psychiatry

    (2006)
  • P. Sacerdote et al.

    Chlorimipramine and nortriptyline but not fluoxetine and fluvoxamine inhibit human polymorphonuclear cell chemotaxis In vitro

    Gen. Pharmacol.

    (1994)
  • R.C. Shelton et al.

    The pro-inflammatory profile of depressed patients is (partly) related to obesity

    J. Psychiatr. Res.

    (2015)
  • A.R. Shenoy et al.

    Citalopram suppresses thymocyte cytokine production

    J. Neuroimmunol.

    (2013)
  • R. Strawbridge et al.

    Inflammation and clinical response to treatment in depression: a meta-analysis

    Eur. Neuropsychopharmacol.

    (2015)
  • R. Yoshimura et al.

    Higher plasma interleukin-6 (IL-6) level is associated with SSRI- or SNRI-refractory depression

    Prog. Neuropsychopharmacol. Biol. Psychiatry.

    (2009)
  • B.A. Arnow et al.

    Depression subtypes in predicting antidepressant response: a report from the iSPOT-D trial

    Am. J. Psychiatry

    (2015)
  • J.A. Ascher et al.

    Bupropion: a review of its mechanism of antidepressant activity

    J. Clin. Psychiatry

    (1995)
  • A.S. Brown et al.

    Elevated maternal C-reactive protein and autism in a national birth cohort

    Mol. Psychiatry

    (2014)
  • T.I. Cremers et al.

    Development of a rat plasma and brain extracellular fluid pharmacokinetic model for bupropion and hydroxybupropion based on microdialysis sampling, and application to predict human brain concentrations

    Drug Metab. Dispos.

    (2016)
  • J. Danesh et al.

    C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease

    N. Engl. J. Med.

    (2004)
  • R. Dantzer et al.

    From inflammation to sickness and depression: when the immune system subjugates the brain

    Nat. Rev. Neurosci.

    (2008)
  • M. Ebbinghaus et al.

    The anti-inflammatory effects of sympathectomy in murine antigen-induced arthritis are associated with a reduction of Th1 and Th17 responses

    Ann. Rheum. Dis.

    (2012)
  • J.R. Emberson et al.

    Extent of regression dilution for established and novel coronary risk factors: results from the British Regional Heart Study

    Eur. J. Cardiovasc. Prev. Rehab.

    (2004)
  • Cited by (143)

    View all citing articles on Scopus
    View full text