Skip to main content
NCBI home page
Brain, Behavior, & Immunity - Health logoLink to Brain, Behavior, & Immunity - Health
. 2022 Nov 17;26:100561. doi: 10.1016/j.bbih.2022.100561

Sex differences in a double-blind randomized clinical trial with minocycline in treatment-resistant depressed patients: CRP and IL-6 as sex-specific predictors of treatment response

Giulia Lombardo a,∗, Maria Antonietta Nettis a,b, Caitlin Hastings a, Zuzanna Zajkowska a, Nicole Mariani a, Naghmeh Nikkheslat a, Courtney Worrell a, Daniela Enache a,c, Anna McLaughlin a, Melisa Kose a, Anna Bogdanova a, Luca Sforzini a, Anthony J Cleare a,b, Allan H Young a,b, Paola Dazzan a,b, Valeria Mondelli a,b, Carmine M Pariante a,b
PMCID: PMC9712814  PMID: 36467125

Abstract

Background

Inflammation is a well-known risk factor for depression. Specifically, patients who do not respond to antidepressant treatment show higher levels of inflammatory biomarkers compared with responders. Thus, several studies have investigated the efficacy of anti-inflammatory add-on treatment in this population. However, major depressive disorder is more prevalent in females than in males, with sex differences present in antidepressant treatment response and in immune system regulation. To explore sex differences in inflammatory profiles and treatment responses, we investigated a cohort of patients with treatment resistant depression (TRD), for which they received an adjunctive, anti-inflammatory treatment with minocycline – the Minocycline in Depression (MINDEP) study.

Methods

The MINDEP study is a 4-week double-blind, randomised, placebo-controlled clinical trial (stratified by sex) with 39 TRD participants, which demonstrated the efficacy of minocycline, an antibiotic with anti-inflammatory properties, in TRD patients with major depressive disorder (MDD) and evidence of low-grade inflammation measured with C-reactive protein (CRP) ≥ 3 mg/L. In these secondary analyses, we investigated the differential effects of minocycline in females (N = 22, 10 randomised to minocycline and 12 randomised to placebo) and in males (N = 17, 8 randomised to minocycline and 9 randomised to placebo) on changes in depressive symptoms (Δ- Hamilton Rating Scale for Depression (HAMD)-17), taking also into consideration CRP levels (CRP ≥3 mg/L vs. CRP <3 mg/L). Additionally, we investigated the role of serum IL-6 in predicting treatment response to minocycline, using sex-specific medians of IL-6, in novel exploratory analyses.

Results

Sex differences in Δ-HAMD-17 indicate that only females (F = 10.49, p = 0.005), but not males (F = 1.64, p = 0.22), presented an effect of CRP levels on the response to minocycline. Also, we detected sex differences in the relationship between serum CRP and IL-6 levels: CRP was strongly correlated with IL-6 in females (Spearman's ρ = 0.658, P < 0.001) but not in males (ρ = 0.007, p = 0.979). Exploratory analyses found that IL-6 was indeed a better predictor of response than minocycline than CRP, as we found an interaction between study arms and IL-6 groups (above and below the IL-6 sex-specific median) in females (F = 4.435 p = 0.050) and, at trend statistical level, in males (F = 4.258 p = 0.060). Moreover, Δ-HAMD-17 was numerically comparable in the two high-IL-6 group taking minocycline (females, mean 9.20 ± SD 7.80; males, mean 8.80 ± SD 5.97), confirming that high IL-6, differently from high CRP, identified responders to minocycline both in males and females.

Conclusion

Our findings highlight the need of sex-specific inflammatory biomarkers in predicting antidepressant response to anti-inflammatories in TRD patients, with the possibility of CRP being a relevant predictor of treatment response only for females, and IL-6 being relevant for both sexes.

Keywords: Minocycline, Major depressive disorder, Treatment resistant depression, Sex differences, C-Reactive protein, Interleukin 6

1. Introduction

Major depressive disorder (MDD) shows sex differences, not only in incidence but also in symptomatology and antidepressant response, but whether such differences can be exploited for sex-specific pharmacological interventions is still unclear. Women are almost twice as likely as men to develop depression (Kuehner 2017), with female patients having younger age of MDD onset and more frequent atypical depression in comparison with men (Schuch et al., 2014). Furthermore, female patients tend to respond better to selective serotonin reuptake inhibitors (SSRIs), whereas male patients respond better to tricyclic antidepressants (TCAs) (Young et al., 2009; Keers and Aitchison 2010). Notably, men have increased risk to attempted and completed suicides, even though suicidal ideation and suicidal behaviour are more associated with female sex (Barrigon and Cegla-Schvartzman, 2020; Cantor 2000; Oquendo et al., 2001; Skogman et al., 2004).

One of the most relevant findings in depression is the key role of inflammation. Specifically, depressed patients show higher levels of inflammatory biomarkers in comparison with controls (Osimo et al., 2020; Pitharouli et al., 2021), with patients who do not respond to the common antidepressant treatments having even higher levels of inflammatory biomarkers in comparison with responders (Cattaneo et al., 2013, 2015, 2020). Interestingly, the immune system differs between males and females. For instance, in the general population, females show a stronger immune response in comparison with males (Klein and Flanagan 2016), with higher rates of auto-immune diseases (Quintero et al., 2012). The immune system might also play a relevant role in women with depression. For instance, in one of the few studies testing sex-differences in this context, depressed females show increased interleukin (IL)-8, interferon (INF)-γ, and leptin, and decreased IL-5 and adiponectin, compared with female controls, while there are no significant differences in inflammatory markers between depressed males and male controls (Birur et al., 2017). However, our recent study from the UK Biobank finds that CRP is elevated in both males and females depressed participants vs. controls, and even after adjusting for age, body mass index (BMI), early life stressors and medical conditions (Pitharouli et al., 2021).

Based on this notion of increased inflammation in depression, anti-inflammatory add-on treatment has been tested as a new therapeutic strategy for patients with MDD. Several studies have described the antidepressant efficacy of a broad-spectrum of anti-inflammatory medications (Köhler et al., 2014). Minocycline, a tetracycline antibiotic, recently attracted interest for its capacity to provide neuroprotection (Garrido-Mesa et al., 2013) and to decrease inflammatory biomarkers (Cheng et al., 2015), with an inhibitory effect on IL-6 expression in central nervous system (CNS) (Campbell et al., 2011; Libbey et al., 2011). Several studies have investigated minocycline efficacy in depression and found that this compound shows efficacy in comparison with placebo in decreasing depressive symptoms, and that it is relatively well-tolerated (Rosenblat and McIntyre 2018; Nettis 2021). While a cross-sectional study based on health records found that Ibuprofen use is associated with significantly less depression in women but not men (Lehrer and Rheinstein 2019), to our knowledge sex differences in the antidepressant efficacy of minocycline have been not evaluated. Of note, however, in patients with stroke, minocycline reduces ischemic damage only in men but not in women, and only male patients exhibit clinical improvement (Amiri-Nikpour et al., 2015). Furthermore, in a middle cerebral artery occlusion animal model, minocycline reduces infarct volume only in male mice whereas female mice do not show significant changes (Li and McCullough 2009). Interestingly, only male rats show a reduction of depressive-like symptom and microglia activation in response to minocycline (Schmidtner et al., 2019). These results highlight possible sex differences in the neuroprotective and anti-depressant effects of minocycline, with male-sex being possibly associated a better outcome than female-sex. A recent, large clinical trials by Hellmann-Regen et al. (2022) found no effects of minocycline on depressive symptoms; although they stratified their sample for median CRP values, these were indicative of very low inflammation, with median values equivalent to approximately 1.2 mg/L in the minocycline arm and 0.6 mg/L in the placebo arm; moreover, they stratified the sample by sex but did not look at sex by CRP groups interaction (Hellmann-Regen et al., 2022).

This is the first report, to our knowledge, to analyse not only sex differences in the response to minocycline treatment but also to investigate the effect of baseline inflammatory levels in the clinical outcome of male and female patients with depression following an anti-inflammatory treatment. We investigated a cohort of patients with treatment resistant depression (TRD) and C-reactive protein (CRP) > 1 mg/L that received an adjunctive treatment with minocycline for four weeks - the Minocycline in Depression (MINDEP) study. We have recently published in the overall sample (n = 39) that only patients with low-grade inflammation, as indicated by levels of the pro-inflammatory biomarker, CRP equal to or above 3 mg/L, show a clinical improvement compared with placebo (Nettis et al., 2021). Here we test sex differences in these effects, and we compare the predictive abilities of serum CRP and serum IL-6.

2. Methods

The main methods and results are described in the recently published paper (Nettis et al., 2021). Briefly, the MINDEP study was a 4-week single centre, double-blind, randomised (1:1 minocycline/placebo), placebo-controlled, parallel group trial with add-on minocycline treatment (200 mg/day), investigating the efficacy of minocycline in TRD patients with major depressive disorder and increased peripheral inflammation detected by CRP levels equal to or more than 1 mg/L (Nettis et al., 2021). Patients were randomised (1:1 minocycline/placebo) by the method of block randomisation, stratified by sex, via a web-based randomisation system at the Clinical Trials Unit at the IoPPN. Treatment resistant depression was defined as non-response to the current antidepressant treatment taken for 6 ≥ weeks at therapeutic dose and Hamilton Depression Rating Scale (HAMD)-17 score ≥14.

2.1. Depressive symptoms

We used HAMD-17 and Beck Depression Inventory-II (BDI-II) clinical assessments. We calculated two total sub-scores based on the BDI-II items identified by Thombs et al. (2010). The following BDI-II items evaluated cognitive/affective and somatic symptoms, respectively: 1) sadness, pessimism, past failure, loss of pleasure, guilty feelings, punishment feelings, self-dislike, self-criticalness, suicidal ideation, crying, agitation, loss of interest, indecisiveness, worthlessness (items 1–14); 2) loss of energy, sleep problems, irritability, appetite problems, concentration, fatigue, loss of interest in sex (items 15–21).

2.2. Clinical outcome

We investigated the clinical improvement using the same criterion and by calculating Δ-HAMD-17 score (baseline HAMD-17 – week-4 HAMD-17) between baseline and end of treatment.

2.3. CRP groups

Nettis et al. (2021), highlighted that the presence of low-grade inflammation at the baseline (high-sensitivity (hs-) CRP ≥3 mg/L) in the minocycline study arm was associated with greater clinical improvement. This finding was also corroborated by the hypothesis-free ROC analysis which identified the cut-off ≥ 2.8 mg/L as best discerning responders vs. non-responders to minocycline. Of course, the authors performed the analysis without considering differences between males and females. Therefore, we investigated sex differences in treatment response by focussing on the original cut-off equal to 3 mg/L, that is by analysing male (n = 17) and female (n = 22) patients, separately with hs-CRP ≥ 3 mg/L (CRP>3 group, 8 males and 10 females) and hs-CRP <3 mg/L (CRP<3 group, 9 males and 12 females; all had CRP>1 mg/L).

2.4. Inflammatory biomarkers

As Nettis et al. (2021) described, the whole inflammatory panel included the following serum pro- and anti-inflammatory biomarkers: hs-CRP and IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL12p70, IL-13, TNF-α, INF-γ, at baseline and week-4, using MSD V-PLEX sandwich immunoassays, MSD Pro-inflammatory Panel 1 (human) kit (Dabitao et al., 2011; King et al., 2019) and plates read on an MSD QuickPlex SQ 120, as we had extensively used before (Hepgul et al., 2012; Russell et al., 2019). The inter-assay coefficient of variations was <10%. A large number of samples for IL-1β (35%), IL-4 (82%) and IL12p70 (25%) levels were non detected or below the lowest detectable values for baseline and week 4 time points, and we excluded these cytokines from the statistical analyses. Most of the investigated inflammatory biomarkers exhibited a non-normal distribution; thus, we applied the natural logarithmic transformation.

2.5. Statistical analyses

Sex differences at the baseline: The analyses included One-Way ANOVA, independent samples t-test, Chi-square test and bivariate Spearman's correlation to investigate differences in demographic data (age, BMI, ethnicity, self-report medications, current smoker users and alcohol units per week), inflammatory profile, childhood trauma questionnaire (CTQ) and depressive symptoms between male and female TRD patients at the baseline. We performed Spearman's bivariate correlation analyses for evaluations of the association between two continuous variables within sexes.

Treatment response: Univariate general linear model (GLM) analyses and independent samples t-tests were used to investigate the effects of study arms and CRP groups on Δ-HAMD-17. We performed Spearman's bivariate correlational analyses between baseline hs-CRP and Δ-HAMD-17, and between baseline IL-6 and Δ-HAMD-17, to investigate their potential role as predictors of treatment response to minocycline, separately in males and females. We evaluated the difference between pair of correlations using Fisher z statistics in the cocor package in R http://comparingcorrelations.org/(Diedenhofen and Musch, 2015).

As exploratory analyses, we performed GLM analyses separately in males (N = 17) and females (N = 22) to further clarify sex specific role of study arms and CRP groups. The GLM analysed three models for CRP stratification: 1) the first model had as fixed factors study arms and sex (overall sample); 2) the second model had as fixed factors study arms, sex and CRP groups (overall sample); and 3) the third model had as fixed factors study arms and CRP groups (within sex analyses).

We calculated sex-specific medians of IL-6 (females: 0.640 pg/ml; males: IL-6 0.960 pg/ml) to split the female and male samples as above and below the respective IL-6 medians, and investigate the role of baseline IL-6 levels in treatment response. Therefore, similarly to the previous analyses, we performed the GLM analyses investigating sex-specific IL-6 groups and study arms as fixed factors, in the overall sample and separately in females and males.

All statistical analyses were performed using IBM SPSS Statistics Version 27.0 (IBM Ltd, UK).

3. Results

3.1. There are minimal sex differences at baseline

Males were older than females (males, 48.3 ± 7.0 years; females, 39.90 ± 10.60 years, t 2.98 p = 0.005), and they exhibited higher consumption of alcohol than women (males 14.28 ± 12.65; females 3.43 ± 4.21, t = 3.425 p = 0.003). There was no significant difference in BMI between the groups (means were around 31–32 kg/m2 for both sexes t = 0.26 p = 0.80); see Table 1 for all results. Of note, 10 out of 22 female patients (45%) self-reported the use of oral contraceptive/hormonal therapy.

Table 1.

Demographics characteristics of the MINDEP sample.

Demographics
Males (n = 17) Females (n = 22) Statistics
BMI (kg/m2) mean (SD) 31.65 (5.69) 31.10 (7.05) t = 0.262 p = 0.795
Age (years) mean (SD) 48.35 (7.05) 39.90 (10.60) t = 2.98 p = 0.005*
Current smoker users (yes) N (% within sex) 5 (29.4%) 6 (27.3%) χ2 = 0.22 p = 0.88
Alcohol units per week, mean (SD) 14.28 (12.65) 3.43 (4.21) t = 3.425 p = 0.003
Class of current antidepressant medication (N, % within sex)
  • 1)

    SSRI (n = 10, 58.8%)
  • 1)

    SSRI (n = 11, 50%)

  • χ2 = 1.43 p = 0.70
  • 2)

    Other AD (n = 2 11.8%)
  • 2)

    Other AD (n = 6, 27.3%)
  • 3)

    AD + AP (n = 2, 11.8%)
  • 3)

    AD + AP (n = 2, 9.1%)
  • 4)

    two or more AD (n = 3, 17.6%)
  • 4)

    two or more AD (n = 3, 13.6%)
Months on current medication (N, % within sex)
  • 1)

    ≤ 6 months (n = 4, 26.7%)
  • 1)

    ≤ 6 months (n = 6, 27.3%)

  • χ2 = 1.49 p = 0.47
  • 2)

    6< >12 months (n = 0, 0%)
  • 2)

    6< >12 months (n = 2, 9.1%)
  • 3)

    ≥ 12 months (n = 11, 73.3%)
  • 3)

    ≥ 12 months (n = 14, 63.6%)
Oral contraceptive/hormonal therapy (N, %) n/a 10 (45.45%) n/a
HAMD-17, mean (SD) 18.71 (4.51) 17.36 (2.32) t = 1.12 p = 0.275
HAMD-17, mean (SD) 18.71 (4.51) 17.36 (2.32) t = 1.12 p = 0.275
BDI-II total score, mean (SD) 25.59 (9.82) 25.59 (9.32) t = 0.067 p = 0.947
BDI-C/A, mean (SD) 17.59 (6.65) 19.14 (7.66) t = 0.662 p = 0.512
BDI-S, mean (SD) 8.00 (4.09) 6.45 (2.30) t = 1.395 p = 0.176
Open in a new tab

Note: AD (antidepressants); AP (antipsychotics); BMI (body mass index); N (number). BDI (Beck's Depression Inventory scale); BDI-C/A (BDI cognitive affective subscale); BDI-S (BDI somatic subscale); HAMD-17 (Hamilton Depression Rating Scale 17 items).

*Bold means that the results are statistically significant.

*Italicized text that the results are different at statistical trend.

We did not detect significant differences between sexes in HAMD-17 total score (males, mean 18.71 ± SD 4.51; females, mean 17.36 ± SD 2.32; t = 1.12 p = 0.27) or BDI total score (males, mean 25.59 ± SD 9.82; females, mean 25.59 ± SD 9.32; t = 0.07 p = 0.95), or the sub-scores (somatic and cognitive/affective sub-scales) at the baseline (see Table 1).

Chi-square test did not detect significant difference CRP groups distribution between sexes, with 9 males and 12 females in CRP<3 group, and 8 males and 10 females in the CRP>3 group (χ2 = 0.010 p = 0.92). However, on average females exhibited numerically higher hs-CRP levels in comparison with males (males, mean 2.994 ± SD 1.60 mg/L; females, mean 4.53 ± SD 5.36 mg/L) even though the difference was not significant (t = 0.524 p = 0.60). Interestingly, male patients had numerically higher IL-6 levels than females (males 0.956 ± 0.344 pg/ml; females 0.777 ± 0.410 pg/ml), but again the difference was not significant (t = 1.561 p = 0.13). There were no significant differences in other immune biomarkers between males and females at baseline (see Table 2).

Table 2.

Sex differences in the inflammatory profile at the baseline in TRD patients.

Inflammatory biomarkers
CRP ≥3 (mg/L), n (%) Males
 SD Females
 SD Statistics
8 (44.4%)
 10 (55.6%)
 χ2 = 0.10 p = 0.921
mean mean
CRP (mg/L) 2.994 1.598 4.532 5.364 t = 0.524 p = 0.604
IL-2 (pg/ml) 0.175 0.132 0.152 0.13 t = 0.395 p = 0.695
IL-6 (pg/ml) 0.956 0.344 0.777 0.41 t = 1.561 p = 0.127
IL-8 (pg/ml) 10.785 4.16 9.526 2.023 t = 0.951 p = 0.348
IL-10 (pg/ml) 0.333 0.332 0.365 0.269 t = 0.390 p = 0.699
IL-13 (pg/ml) 0.564 0.461 0.524 0.376 t = 0.293 p = 0.771
TNF-α (pg/ml) 3.362 0.789 3.13 0.616 t = 0.800 p = 0.429
INF-γ (pg/ml) 3.158 2.702 2.387 1.419 t = 0.211 p = 0.834
Open in a new tab

Note: CRP (C-reactive protein); IL- (interleukin); IFN- (interferon); TNF (tumour necrosis factor).

3.2. Sex differences in Δ-HAMD-17 indicate that only females, but not males, show an effect of CRP levels on the response to minocycline

The GLM analyses revealed sex differences in the interaction between treatment response and baseline inflammatory status (CRP groups) (see Table 3). The GLM model 1 is the same as published by Nettis et al. (2021) and is presented here as background information. This model shows an effect at statistical trend for study arms (placebo vs. minocycline) on Δ-HAMD-17 in the overall samples (F = 3.10 p = 0.087), describing numerical superiority in the improvement following minocycline when compared with placebo in the overall samples. GLM model 2 shows a significant effect of study arms (F = 12.17 p = 0.001), and of their interaction with CRP groups (CRP < and CRP ≥3 mg/L) (F = 9.18 p = 0.005), again describing the previously reported, and statistically significant, improvement in depressive symptoms only in those taking minocycline and with CRP ≥3 mg/L (Nettis et al., 2021). Interestingly for the specific hypotheses of the present report, we also found an interaction, at statistical-trend level, between sex and CRP groups (F = 3.80 p = 0.060) on Δ-HAMD-17, suggesting that the effects of the CRP groups on the response to minocycline was different in males and females.

Table 3.

General Linear Model analysis: Model 1 (not including CRP groups as a factor); Model 2 (including CRP groups as a factor); Model 3 (study arms and CRP groups within sexes).

GLM in the MINDEP cohort
Model 1 Model 2 Model 3
Males Females
Study arms F = 3.10 p = 0.087 F = 12.17 p = 0.001 F = 6.17 p = 0.027 F = 5.53 p = 0.030
Sex F=0.23 p = 0.637 F = 1.51 p = 0.229 n/a n/a
CRP groups n/a F = 14.88 p = 0.001 F = 1.32 p = 0.271 F = 22.90 P<0.001
Study arms * sex F = 1.34 p = 0.254 F = 0.40 p = 0.531 n/a n/a
Study arms * CRP groups n/a F = 9.18 p = 0.005 F = 1.64 p = 0.222 F = 10.49 P = 0.005
Sex * CRP groups n/a F = 3.80 p = 0.060 n/a n/a
Study arms * sex * CRP groups n/a F = 8.12 p = 0.375 n/a n/a
Open in a new tab

Note: Model 1. Fixed factor study arms and sex (overall sample). Model 2. fixed factors study arms, sex and CRP groups (overall sample). Model 3. fixed factors study arms and CRP groups (in males and females separately). N/A: not applicable as the specific factor, or the specific interaction is not part of the model.

*Bold means that the results are statistically significant.

*Italicized text that the results are different at statistical trend.

The GLM univariate model 3 further clarifies this (trend-level) interaction between sex and CRP groups (see Table 3). Specifically, in females we detected a significant effect of study arms (F = 5.53 p = 0.030), and of their interaction with CRP groups (F = 10.49, p= 0.005), that is, the same results as the main sample: females with CRP ≥ 3 mg/L show a greater improvement with minocycline compared with females with CRP < 3 mg/L. Specifically, t-test and post-hoc analyses with Bonferroni's correction show that female patients overall did not show a difference in Δ-HAMD-17 comparing minocycline (mean 5.10 ± SD 6.95) with placebo (mean 4.08 ± SD 4.12; t = 0.426 p = 0.675), but CRP ≥3 females receiving minocycline showed a greater clinical improvement (Δ-HAMD-17, N = 3, mean 14.33 ± SD 4.72) than CRP<3 females receiving minocycline (N = 7, mean 1.14 ± SD 2.03, p < 0.001), CRP ≥3 females receiving placebo (N = 7, mean 5.14 ± SD 4.37, p < 0.05) and CRP<3 females receiving placebo (N = 5, mean 2.60 ± SD 3.64; p < 0.05) (see Table 4).

Table 4.

Differences in treatment response between study arms X CRP groups in female sample.

FEMALES
Study arms X CRP groups
CRP-/M Group 1, (n = 7) CRP+/M Group 2, (n = 3) CRP-/PL Group 3, (n = 5) CRP+/PL Group 4, (n = 7) Statistics Bonferroni post-hoc Multiple Comparisons
Δ-HAMD-17 mean (SD) 1.14 (2.03) 14.44 (4.72) 2.60 (3.64) 5.14 (4.37) F=9.85 p=<0.001 Group 2 vs.
Group 1: p<0.001
Group 3: p = 0.002
Group 4: p = 0.011
Open in a new tab

Note: Δ-HAMD-17 (clinical improvement: HAMD-17 total score at the baseline – HAMD-17 total score at week-4); CRP-/M (Group 1; patients with CRP <3 mg/L and taking minocycline); CRP+/M (Group 2; patients with CRP equal to or more than 3 mg/L and taking minocycline); CRP-/PL (Group 3; patients with CRP <3 mg/L and taking placebo); CRP+/PL (Group 4; patients with CRP equal to or more than 3 mg/L and taking placebo). *Bold means that the results are statistically significant.

In contrast, in males we detected a significant effect of study arms (F = 6.17 p = 0.027) but no interaction between study arms and CRP groups (F = 1.64, p = 0.22) (see Table 3). This suggests that all male subjects, both CRP ≥ 3 and CRP<3, improve with minocycline. Specifically, in the whole sample of males, patients taking minocycline exhibited greater clinical improvement (Δ-HAMD-17, mean 6.25 ± SD 5.95) than patients taking placebo (Δ-HAMD-17, mean 1.33 ± SD 3.08; t = 2.179 p = 0.046). Additionally, we did not detect significant differences in Δ-HAMD-17 between the four sub-groups based on study arm and CRP (CRP ≥3 minocycline, CRP ≥3 placebo, CRP<3 minocycline and CRP<3 placebo) (F = 2.572 p = 0.99) in the male sample.

3.3. Serum CRP correlates with IL-6 only in females

The finding that only female patients showed an interaction between CRP groups and study arms, while male patients did not, might be indicative of a different type of immunological profile in females vs. males with CRP below and above 3 mg/L. To test this hypothesis, we conducted explorative Spearman's correlations between CRP and the cytokines, separately in males and females within sexes. Results are displayed in Table 5.

Table 5.

Correlation analyses between baseline CRP and cytokines in male patients and female patients.

Bivariate correlations between CRP and cytokine baseline levels within sexes
IL-2 IL-6 IL-8 IL-10 IL-13 TNF-α IFN-γ
CRP (mg/L) Males (n = 17) −0.251 0.007 −0.125 −0.363 −0.455 0.093 0.185
Females (n = 22) 0.199 .658** −0.256 0.384 0.022 0.317 −0.407
Open in a new tab

**. Correlation is significant at the 0.01 level (2-tailed).

*. Correlation is significant at the 0.05 level (2-tailed).

Note: CRP (C-reactive protein); IL- (interleukin); IFN- (interferon); TNF (tumour necrosis factor).

Interestingly, we found that CRP was strongly correlated with IL-6 in females (Spearman's ρ = 0.658, p < 0.001) but was not correlated in males (Spearman's ρ = 0.007, p = 0.979), and the two pairs of correlations were significantly different (z = −2.2210, p = 0.0264).

Consistent with these findings, CRP ≥3 females had higher IL-6 levels than CRP<3 females, even if only at statistical trend (t = 2.065 p = 0.063), whereas there was not difference in IL-6 levels between CRP ≥3 and CRP<3 males (t = 0.264 p = 0.79). This sex difference in the association between CRP and IL-6 was also consistent with our aforementioned finding that females had numerically higher CRP than males, and males had numerically higher IL-6 than females (see Table 2).

3.4. Serum IL-6 is a better predictor of response to minocycline than CRP

Our findings so far suggested that high IL-6 could be a better predictor of response to minocycline than high CRP: all males responded to minocycline regardless of CRP levels, and as a group males had (numerically) higher IL-6 than females, while only CRP ≥3 female, who also had higher IL-6 levels, responded to minocycline. It is also of note that Nettis et al. (2021) had already shown, in this same sample, that partial responders to minocycline (showing at least 25% symptoms reduction) not only have higher levels of baseline hs-CRP, but also higher levels of baseline IL-6 (Nettis et al., 2021). We thus decided to test if IL-6 at the baseline is a better predictor of treatment response to minocycline than increased CRP, and if, differently from CRP, is equally predictive both in males and females.

To investigate this hypothesis, we split both males and females in those above and below the IL-6 levels sex-specific medians (females = 0.640 pg/ml; males = 0.960 pg/ml), and then we repeated the GLM analysis on the response (Δ-HAMD-17) in the whole sample, and separately in males and females. See Table 6.

Table 6.

General Linear Model analysis: including sex-specific IL-6 groups and study arms as fixed factors, in the whole cohort and within sexes.

GLM – Sex-specific IL-6 groups and study arms on Δ-HAMD-17
Whole cohort (A) Females (B) Males (C)
Study arms F = 3.027 p = 0.092 F = 0.225 p = 0.641 F = 4.258 p = 0.060
Sex-specific IL-6 groups F = 1.744 p = 0.178 F = 2.949 p = 0.103 F = 1.267 p = 0.281
Study arms * Sex-specific IL-6 groups F = 3.189 p = 0.037 F = 4.435 p = 0.050 F = 4.258 p = 0.060
Open in a new tab

Note: Fixed factors sex-specific IL-6 groups and study arms in analyses performed in the whole cohort (A), and separately in females (B) and males (C). *Bold means that the results are statistically significant.

*Italicized text that the results are different at statistical trend.

In the whole sample, we found an interaction between study arms and IL-6 groups (F = 3.189 p = 0.037), which was also present as statistically significant in females (F = 4.435 p = 0.050) and at statistical trend level in males (F = 4.258 p = 0.060). Most importantly, Δ-HAMD-17 was numerically comparable in the two high-IL-6 group taking minocycline (females, mean 9.20 ± SD 7.80; males, mean 8.80 ± SD 5.97), indicating that high IL-6, differently from high CRP, identifies responders to minocycline both in males and females.

To validate that these findings were due to using IL-6 as a biomarker rather than a sex-specific threshold, we replicated the GLM analyses considering sex-specific CRP medians (CRP = 2.90 mg/L for males; CRP = 2.50 mg/L for females). We found, again, that only females presented an interaction between CRP levels and study arm (F = 10.487 p = 0.005), but not males (F = 1.645 p = 0.22).

4. Discussion

To our knowledge, this is the first report investigating sex differences in the efficacy of add-on anti-inflammatories in TRD patients with increased inflammation in the context of a clinical trial. We conducted secondary analyses in our MINDEP study, comparing minocycline and placebo in TRD patients selected for hs-CRP ≥ 1 mg/L. Our published findings show that those with high inflammatory level (CRP ≥3 mg/L) at the baseline have a greater clinical improvement after minocycline add-on treatment (Nettis et al., 2021). The present paper, based on secondary and exploratory analyses in undoubtedly a small sample, indicates that these effects are different in males and females. Specifically, TRD male patients benefit from minocycline treatment independently of their baseline CRP levels, while TRD female patients benefit from minocycline only if they have CRP ≥3 mg/L at baseline. We finally tested if high IL-6 could be a better predictor of response than CRP and found that both male and female patients with increased baseline IL-6 levels showed similar improvement. If replicated in larger samples, this evidence indicates that there is the need to consider sex-specific biomarkers (and biomarkers levels) in stratifying TRD patients to access adjuvant anti-inflammatory strategy.

4.1. Is there a sex-specific relationship between CRP and IL-6?

Our preferred interpretation of these findings is that they suggest that the relationship between CRP levels and the overall immune profiles is different in males and females, as indicated also by the numerical evidence of higher CRP in females (vs. males) and higher IL-6 in males (vs. females). This suggestive evidence contradicts the commonly-held assumption that CRP is primarily the end-product (and a direct representation) of IL-6 signal in the blood (Kim and Bajaj 2014; El Ayadi et al., 2018) and thus usually considered as reflecting IL-6 levels. Indeed, we found a clear correlation between CRP and IL-6 levels only in females, but not in males. This is interesting in the view of the evidence that low-grade inflammation detected by CRP above 3 mg/L mirrors peripheral and central inflammation measured through other cytokines in both plasma and cerebrospinal fluid (CSF) (Felger et al., 2020); however, the authors of this study also found some important sex-specific differences: for example, high anhedonia was predicted by CSF CRP only in females.

Previous studies also indicate the presence of sex-differences in the relationship between CRP and IL-6. Lockwood et al. (2016) found that females produce more IL-6 response than males in response to acute laboratory stress, and Ye et al. (2021) found, in the UK Biobank Cohort, that females showed a stronger association than males between CRP and depressive symptoms (Ye et al., 2021). This suggests that females are more sensitive to stress-induced inflammation. However, Lockwood et al. also found that only males, and not females, show a correlation between baseline CRP levels and the IL-6 increase, that is, the opposite correlational pattern that we found (Lockwood et al., 2016). This suggests that sex-specific effect may vary based on the functional status of the immune system and the stress response. Indeed, Kudielka and Kirschbaum propose that the higher immune sensitivity to stress in females might be due to males being able to mount a stronger hypothalamic-pituitary-adrenal (HPA) axis response than females, leading to suppression of the immune response (Kudielka and Kirschbaum, 2005). We have extensively discussed before the effects of sex hormones on the immune system (Lombardo et al., 2021): while the picture is complex, male sex hormones (androgens) have mainly anti-inflammatory proprieties (Gilliver, 2010), whereas female sex hormones have both pro- and anti-inflammatory proprieties (Bereshchenko et al., 2018). We can speculate that the anti-inflammatory action of androgens might limit the increase of baseline CRP in response to baseline IL-6 in males, thus leading to the lack of correlation that we found – a model that is similar to that proposed by Lockwood et al. in females to explain the lack of correlation between baseline CRP and stress-evoked IL-6. Additionally, there is the need to consider female-specific factors, as both CRP and IL-6 vary in their levels throughout the menstrual cycle (Gaskins et al., 2012; Eagan et al., 2021). Moreover, the use of oral contraceptives (in 10 out of the 22 females in our study) can influence both CRP and IL-6 levels (Pradhan et al., 2002), and indeed in one study IL-6 shows a significant association with CRP levels only in females using oral contraceptives but not in those using vaginal contraceptives (Divani et al., 2015).

4.2. Is IL-6 a better predictor of response to minocycline than CRP?

We found that subjects with high IL-6 levels (measured as above the sex-specific median of levels) show better response to minocycline in both females and males, with similar Δ-HAMD-17 in males and females (Δ-HAMD-17 mean 8.8 vs 9.2). The ability of increased levels of IL-6 at the baseline to predict the response to minocycline could be due the fact that minocycline shows inhibitory proprieties on IL-6 levels. In fact, minocycline can lower IL-6 concentrations in macrophage and microglial cells in animal model (male rats) (Moini-Zanjani et al., 2016), and patients with schizophrenia and taking minocycline treatment exhibit a reduction in IL-6 (and IL-1β) levels, and these decreases correlate with improvement in cognitive symptoms (Zhang et al., 2019). In our sample, males have numerically higher mean IL-6 than females, and this could explain why, as a group, they respond better to minocycline. It is possible that the better response previously described in male patients with stroke (Amiri-Nikpour et al., 2015) could also be described by higher levels of IL-6, but the authors did not test this.

4.3. Limitations

Three major limitations need to be considered for our study. Firstly, while the main GLM and correlational analyses are in the complete sample (n = 39), the division in CRP/IL-6- and sex-based groups meant that some of the secondary and exploratory analyses compared very small groups, some single-figured. There is no doubt that these findings should be considered as a pilot data, useful for guiding new directions in research, and not as established findings. Secondly, we did not have information of the menstrual phase of female patients, and approximately half of women were on oral contraceptives. This is important as higher number of follicular waves are associated with higher inflammatory levels (Clancy et al., 2013), and hormonal changes during menstrual cycle are associated with variations in CRP levels (Wander et al., 2008; Gaskins et al., 2012).

5. Conclusion

In conclusion, our pilot findings confirms that TRD patients of both sexes with evidence of increased peripheral inflammation – above sex-specific thresholds for serum IL-6 – show greater clinical improvement after minocycline treatment. In this study, IL-6 performs better than CRP as predictor of response, as high CRP only predicts response in females. We thus want to draw attention to the need of sex-specific inflammatory biomarkers in predicting response to anti-inflammatory medications in TRD patients, in order to exploit the full potential for developing sex-specific tailored medicine.

Declaration of competing interest

CMP, GL, LS, CW, and MK, are members of EU-PEARL project and report grant from European Commission (Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No 853966-2). This Joint Undertaking receives support from the European Union’s Horizon 2020 research and innovation programme and EFPIA. CMP is also funded by a Senior Investigator award from the National Institute for Health Research (NIHR); the Medical Research Council (grants MR/L014815/1, MR/J002739/1 and MR/N029488/1); the European Commission (EARLYCAUSE grant SC1–BHC-01-2019); the NARSAD; the Psychiatry Research Trust; and the Wellcome Trust (SHAPER, Scaling-up Health-Arts Programme to scale up arts interventions, grant 219425/Z/19/Z). Less than 10% of his support in the last 10 years derives from commercial collaborations, including consultation and speakers fees from Boehringer Ingelheim, Eli Lilly, Compass, Eleusis, GH Research, Lundbeck, and Värde Partners. VM is also funded by MQ: Transforming Mental Health (Grant: MQBF/1 and MQBF/4) and the Medical Research Foundation (Grant: MRF-160-0005-ELP-MONDE). PD has received speaker's fees from Lundbeck and Janssen and is supported by the Medical Research Council (MR/S003444/1). AJC has in the last three years received honoraria for educational activities from Janssen; honoraria for consulting from Allergan and Janssen; sponsorship for conference attendance from Janssen; and research grant support from Protexin Probiotics International Ltd.

Acknowledgments

This work was funded by the Psychiatry Research Trust through funding collected in memory of Claire Nacamuli. We thank the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London and the National Institute for Health Research NIHR/Wellcome King's Clinical Research Facility. All visits took place at the Clinical Research Facility of King's College Hospital. The team of nurses has to be thanked for providing their valuable expertise to the study. The views expressed are those of the authors and not necessarily those of the Nacamuli family and the Psychiatry Research Trust, the NHS, the NIHR, or the Department of Health and Social Care.

Data availability

Data will be made available on request.

References

  1. Amiri-Nikpour M.R., Nazarbaghi S., Hamdi-Holasou M., Rezaei Y. An open-label evaluator-blinded clinical study of minocycline neuroprotection in ischemic stroke: gender-dependent effect. Acta Neurol. Scand. 2015;131(1):45–50. doi: 10.1111/ane.12296. [DOI] [PubMed] [Google Scholar]
  2. Barrigon M.L., Cegla-Schvartzman F. Sex, gender, and suicidal behavior. Curr. topics in behav. neurosci. 2020;46:89–115. doi: 10.1007/7854_2020_165. [DOI] [PubMed] [Google Scholar]
  3. Bereshchenko O., Bruscoli S., Riccardi C. Glucocorticoids, sex hormones, and immunity. Front. Immunol. 2018;9:1332. doi: 10.3389/fimmu.2018.01332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Birur B., Amrock E.M., Shelton R.C., Li L. Sex differences in the peripheral immune system in patients with depression. Front. Psychiatr. 2017;8:108. doi: 10.3389/fpsyt.2017.00108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Campbell J.H., Burdo T.H., Autissier P., Bombardier J.P., Westmoreland S.V., Soulas C., González R.G., Ratai E.M., Williams K.C. Minocycline inhibition of monocyte activation correlates with neuronal protection in SIV neuroAIDS. PLoS One. 2011;6(4) doi: 10.1371/journal.pone.0018688. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cantor C.H. 2000. Suicide in the Western World; pp. 9–28. The international handbook of suicide and attempted suicide. [DOI] [Google Scholar]
  7. Cattaneo A., Ferrari C., Turner L., Mariani N., Enache D., Hastings C., Kose M., Lombardo G., McLaughlin A.P., Nettis M.A., Nikkheslat N., Sforzini L., Worrell C., Zajkowska Z., Cattane N., Lopizzo N., Mazzelli M., Pointon L., Cowen P.J., Cavanagh J., et al. Whole-blood expression of inflammasome- and glucocorticoid-related mRNAs correctly separates treatment-resistant depressed patients from drug-free and responsive patients in the BIODEP study. Transl. Psychiatry. 2020;10(1):232. doi: 10.1038/s41398-020-00874-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cattaneo A., Gennarelli M., Uher R., Breen G., Farmer A., Aitchison K.J., Craig I.W., Anacker C., Zunsztain P.A., McGuffin P., Pariante C.M. vol. 38. 2013. Candidate genes expression profile associated with antidepressants response in the GENDEP study: differentiating between baseline 'predictors' and longitudinal 'targets; pp. 377–385. (Neuropsychopharmacology: Official Publication of the American College of Neuropsychopharmacology). 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cattaneo A., Macchi F., Plazzotta G., Veronica B., Bocchio-Chiavetto L., Riva M.A., Pariante C.M. Inflammation and neuronal plasticity: a link between childhood trauma and depression pathogenesis. Front. Cell. Neurosci. 2015;9:40. doi: 10.3389/fncel.2015.00040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cheng S., Hou J., Zhang C., Xu C., Wang L., Zou X., Yu H., Shi Y., Yin Z., Chen G. Minocycline reduces neuroinflammation but does not ameliorate neuron loss in a mouse model of neurodegeneration. Sci. Rep. 2015;5 doi: 10.1038/srep10535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Clancy K.B., Baerwald A.R., Pierson R.A. Systemic inflammation is associated with ovarian follicular dynamics during the human menstrual cycle. PLoS One. 2013;8(5) doi: 10.1371/journal.pone.0064807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dabitao D., Margolick J.B., Lopez J., Bream J.H. Multiplex measurement of proinflammatory cytokines in human serum: comparison of the Meso Scale Discovery electrochemiluminescence assay and the Cytometric Bead Array. J. Immunol. Methods. 2011;372(1–2):71–77. doi: 10.1016/j.jim.2011.06.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Diedenhofen B., Musch J. cocor: a comprehensive solution for the statistical comparison of correlations. PLoS One. 2015;10(3) doi: 10.1371/journal.pone.0121945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Divani A.A., Luo X., Datta Y.H., Flaherty J.D., Panoskaltsis-Mortari A. Effect of oral and vaginal hormonal contraceptives on inflammatory blood biomarkers. Mediat. Inflamm. 2015 doi: 10.1155/2015/379501. 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Eagan L.E., Chesney C.A., Mascone S.E., Shenouda N., Ranadive S.M. Interleukin-6 is higher in naturally menstruating women compared with oral contraceptive pill users during the low-hormone phase. J. Appl. Physiol. 2021;131(2):544–552. doi: 10.1152/japplphysiol.00921.2020. Aug 1. Epub 2021 Jun 17. PMID: 34138651. [DOI] [PubMed] [Google Scholar]
  16. El Ayadi A., Herndon D.N., Finnerty C.C. Total Burn Care. Elsevier; 2018. Biomarkers in burn patient care; pp. 232–235. [Google Scholar]
  17. Felger J.C., Haroon E., Patel T.A., Goldsmith D.R., Wommack E.C., Woolwine B.J., Le N.A., Feinberg R., Tansey M.G., Miller A.H. What does plasma CRP tell us about peripheral and central inflammation in depression? Mol. Psychiatr. 2020;25(6):1301–1311. doi: 10.1038/s41380-018-0096-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Garrido-Mesa N., Zarzuelo A., Gálvez J. Minocycline: far beyond an antibiotic. Br. J. Pharmacol. 2013;169(2):337–352. doi: 10.1111/bph.12139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gaskins A.J., Wilchesky M., Mumford S.L., Whitcomb B.W., Browne R.W., Wactawski-Wende J., Perkins N.J., Schisterman E.F. Endogenous reproductive hormones and C-reactive protein across the menstrual cycle: the BioCycle Study. Am. J. Epidemiol. 2012;175(5):423–431. doi: 10.1093/aje/kwr343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gilliver S.C. Sex steroids as inflammatory regulators. J. Steroid Biochem. Mol. Biol. 2010;120(2–3):105–115. doi: 10.1016/j.jsbmb.2009.12.015. [DOI] [PubMed] [Google Scholar]
  21. Hellmann-Regen J., Clemens V., Grözinger M., Kornhuber J., Reif A., Prvulovic D., Goya-Maldonado R., Wiltfang J., Gruber O., Schüle C., Padberg F., Ising M., Uhr M., Friede T., Huber C., Manook A., Baghai T.C., Rupprecht R., Heuser I. Effect of minocycline on depressive symptoms in patients with treatment-resistant depression: a randomized clinical trial. JAMA Netw. Open. 2022;5(9) doi: 10.1001/jamanetworkopen.2022.30367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hepgul N., Pariante C.M., Dipasquale S., DiForti M., Taylor H., Marques T.R., Morgan C., Dazzan P., Murray R.M., Mondelli V. Childhood maltreatment is associated with increased body mass index and increased C-reactive protein levels in first-episode psychosis patients. Psychol. Med. 2012;42(9):1893–1901. doi: 10.1017/S0033291711002947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Keers R., Aitchison K.J. Gender differences in antidepressant drug response. Int. Rev. Psychiatr. 2010;22(5):485–500. doi: 10.3109/09540261.2010.496448. [DOI] [PubMed] [Google Scholar]
  24. Kim J., Bajaj M. In: Normal Adipose Tissue Biology: Adipocytokines and Inflammation. Mitchell R., editor. Elsevier Science; Amsterdam, Netherlands: 2014. pp. 488–497. (Pathobiology of Human Disease: a Dynamic Encyclopedia of Disease Mechanisms). 2014. [Google Scholar]
  25. King E., O'Brien J., Donaghy P., Williams-Gray C.H., Lawson R.A., Morris C.M., Barnett N., Olsen K., Martin-Ruiz C., Burn D., Yarnall A.J., Taylor J.P., Duncan G., Khoo T.K., Thomas A. Inflammation in mild cognitive impairment due to Parkinson's disease, Lewy body disease, and Alzheimer's disease. Int. J. Geriatr. Psychiatr. 2019;34(8):1244–1250. doi: 10.1002/gps.5124. [DOI] [PubMed] [Google Scholar]
  26. Klein S.L., Flanagan K.L. Sex differences in immune responses. Nat. Rev. Immunol. 2016;16(10):626–638. doi: 10.1038/nri.2016.90. [DOI] [PubMed] [Google Scholar]
  27. Köhler O., Benros M.E., Nordentoft M., Farkouh M.E., Iyengar R.L., Mors O., Krogh J. Effect of anti-inflammatory treatment on depression, depressive symptoms, and adverse effects: a systematic review and meta-analysis of randomized clinical trials. JAMA Psychiatr. 2014;71(12):1381–1391. doi: 10.1001/jamapsychiatry.2014.1611. [DOI] [PubMed] [Google Scholar]
  28. Kudielka B.M., Kirschbaum C. Sex differences in HPA axis responses to stress: a review. Biol. Psychol. 2005;69(1):113–132. doi: 10.1016/j.biopsycho.2004.11.009. [DOI] [PubMed] [Google Scholar]
  29. Kuehner C. Why is depression more common among women than among men?. The lancet. Psychiatry. 2017;4(2):146–158. doi: 10.1016/S2215-0366(16)30263-2. [DOI] [PubMed] [Google Scholar]
  30. Lehrer S., Rheinstein P.H. Nonsteroidal anti-inflammatory drugs (NSAIDs) reduce suicidal ideation and depression. Discov. Med. 2019;28(154):205–212. [PubMed] [Google Scholar]
  31. Li J., McCullough L.D. Sex differences in minocycline-induced neuroprotection after experimental stroke. J. Cerebr. Blood Flow Metabol. : off. J. Int. Soc. Cereb. Blood Flow and Metab. 2009;29(4):670–674. doi: 10.1038/jcbfm.2009.3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Libbey J.E., Kennett N.J., Wilcox K.S., White H.S., Fujinami R.S. Interleukin-6, produced by resident cells of the central nervous system and infiltrating cells, contributes to the development of seizures following viral infection. J. Virol. 2011;85(14):6913–6922. doi: 10.1128/JVI.00458-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lockwood K.G., Marsland A.L., Cohen S., Gianaros P.J. Sex differences in the association between stressor-evoked interleukin-6 reactivity and C-reactive protein. Brain Behav. Immun. 2016;58:173–180. doi: 10.1016/j.bbi.2016.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lombardo G., Mondelli V., Dazzan P., Pariante C.M. Sex hormones and immune system: a possible interplay in affective disorders? A systematic review. J. Affect. Disord. 2021;290:1–14. doi: 10.1016/j.jad.2021.04.035. [DOI] [PubMed] [Google Scholar]
  35. Moini-Zanjani T., Ostad S.N., Labibi F., Ameli H., Mosaffa N., Sabetkasaei M. Minocycline effects on IL-6 concentration in macrophage and microglial cells in a rat model of neuropathic pain. Iran. Biomed. J. 2016;20(5):273–279. doi: 10.22045/ibj.2016.04. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Nettis M.A. Minocycline in Major Depressive Disorder: and overview with considerations on treatment-resistance and comparisons with other psychiatric disorders. Brain, behavior, & immun. - health. 2021;17 doi: 10.1016/j.bbih.2021.100335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nettis M.A., Lombardo G., Hastings C., Zajkowska Z., Mariani N., Nikkheslat N., Worrell C., Enache D., McLaughlin A., Kose M., Sforzini L., Bogdanova A., Cleare A., Young A.H., Pariante C.M., Mondelli V. Augmentation therapy with minocycline in treatment-resistant depression patients with low-grade peripheral inflammation: results from a double-blind randomised clinical trial. Neuropsychopharmacology : off. publ. Am. College of Neuropsych. 2021;46(5):939–948. doi: 10.1038/s41386-020-00948-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Oquendo M.A., Ellis S.P., Greenwald S., Malone K.M., Weissman M.M., Mann J.J. Ethnic and sex differences in suicide rates relative to major depression in the United States. Am. J. Psychiatr. 2001;158(10):1652–1658. doi: 10.1176/appi.ajp.158.10.1652. [DOI] [PubMed] [Google Scholar]
  39. Osimo E.F., Pillinger T., Rodriguez I.M., Khandaker G.M., Pariante C.M., Howes O.D. Inflammatory markers in depression: a meta-analysis of mean differences and variability in 5,166 patients and 5,083 controls. Brain Behav. Immun. 2020;87:901–909. doi: 10.1016/j.bbi.2020.02.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Pitharouli M.C., Hagenaars S.P., Glanville K.P., Coleman J., Hotopf M., Lewis C.M., Pariante C.M. Elevated C-reactive protein in patients with depression, independent of genetic, health, and psychosocial factors: results from the UK Biobank. Am. J. Psychiatr. 2021;178(6):522–529. doi: 10.1176/appi.ajp.2020.20060947. [DOI] [PubMed] [Google Scholar]
  41. Pradhan A.D., Manson J.E., Rossouw J.E., Siscovick D.S., Mouton C.P., Rifai N., Wallace R.B., Jackson R.D., Pettinger M.B., Ridker P.M. Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease: prospective analysis from the Women's Health Initiative observational study. JAMA. 2002;288(8):980–987. doi: 10.1001/jama.288.8.980. [DOI] [PubMed] [Google Scholar]
  42. Quintero O.L., Amador-Patarroyo M.J., Montoya-Ortiz G., Rojas-Villarraga A., Anaya J.M. Autoimmune disease and gender: plausible mechanisms for the female predominance of autoimmunity. J. Autoimmun. 2012;38(2–3) doi: 10.1016/j.jaut.2011.10.003. J109–J119. [DOI] [PubMed] [Google Scholar]
  43. Rosenblat J.D., McIntyre R.S. Efficacy and tolerability of minocycline for depression: a systematic review and meta-analysis of clinical trials. J. Affect. Disord. 2018;227:219–225. doi: 10.1016/j.jad.2017.10.042. [DOI] [PubMed] [Google Scholar]
  44. Russell A., Hepgul N., Nikkheslat N., Borsini A., Zajkowska Z., Moll N., Forton D., Agarwal K., Chalder T., Mondelli V., Hotopf M., Cleare A., Murphy G., Foster G., Wong T., Schütze G.A., Schwarz M.J., Harrison N., Zunszain P.A., Pariante C.M. Persistent fatigue induced by interferon-alpha: a novel, inflammation-based, proxy model of chronic fatigue syndrome. Psychoneuroendocrinology. 2019;100:276–285. doi: 10.1016/j.psyneuen.2018.11.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Schmidtner A.K., Slattery D.A., Gläsner J., Hiergeist A., Gryksa K., Malik V.A., Hellmann-Regen J., Heuser I., Baghai T.C., Gessner A., Rupprecht R., Di Benedetto B., Neumann I.D. Minocycline alters behavior, microglia and the gut microbiome in a trait-anxiety-dependent manner. Transl. Psychiatry. 2019;9(1):223. doi: 10.1038/s41398-019-0556-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schuch J.J., Roest A.M., Nolen W.A., Penninx B.W., de Jonge P. Gender differences in major depressive disorder: results from The Netherlands study of depression and anxiety. J. Affect. Disord. 2014;156:156–163. doi: 10.1016/j.jad.2013.12.011. [DOI] [PubMed] [Google Scholar]
  47. Skogman K., Alsén M., Öjehagen A. Sex differences in risk factors for suicide after attempted suicide. Soc. Psychiatr. Psychiatr. Epidemiol. 2004;39:113–120. doi: 10.1007/s00127-004-0709-9. [DOI] [PubMed] [Google Scholar]
  48. Thombs B.D., Ziegelstein R.C., Pilote L., Dozois D.J., Beck A.T., Dobson K.S., Fuss S., de Jonge P., Grace S.L., Stewart D.E., Ormel J., Abbey S.E. Somatic symptom overlap in Beck Depression Inventory-II scores following myocardial infarction. Br. J. Psychiatr. : J. Ment. Sci. 2010;197(1):61–66. doi: 10.1192/bjp.bp.109.076596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wander K., Brindle E., O'Connor K.A. C-reactive protein across the menstrual cycle. Am. J. Phys. Anthropol. 2008;136(2):138–146. doi: 10.1002/ajpa.20785. [DOI] [PubMed] [Google Scholar]
  50. Ye Z., Kappelmann N., Moser S., Davey Smith G., Burgess S., Jones P.B., Khandaker G.M. Role of inflammation in depression and anxiety: tests for disorder specificity, linearity and potential causality of association in the UK Biobank. EClinic. Med. 2021;38 doi: 10.1016/j.eclinm.2021.100992. Jun 26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Young E.A., Kornstein S.G., Marcus S.M., Harvey A.T., Warden D., Wisniewski S.R., Balasubramani G.K., Fava M., Trivedi M.H., John Rush A. Sex differences in response to citalopram: a STAR*D report. J. Psychiatr. Res. 2009;43(5):503–511. doi: 10.1016/j.jpsychires.2008.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Zhang L., Zheng H., Wu R., Kosten T.R., Zhang X.Y., Zhao J. The effect of minocycline on amelioration of cognitive deficits and pro-inflammatory cytokines levels in patients with schizophrenia. Schizophr. Res. 2019;212:92–98. doi: 10.1016/j.schres.2019.08.005. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data will be made available on request.

Articles from Brain, Behavior, & Immunity - Health are provided here courtesy of Elsevier

RESOURCES