ORIGINAL RESEARCH published: 15 March 2022 doi: 10.3389/fpsyt.2022.836771 BDNF and Cortisol in the Diagnosis of Cocaine-Induced Depression Francina Fonseca 1,2,3† , Joan Ignasi Mestre-Pinto 1,3*† , Rocío Rodríguez-Minguela 1 , Esther Papaseit 4,5 , Clara Pérez-Mañá 4,5 , Klaus Langohr 6,7 , Margherita Barbuti 8 , Magí Farré 4,5 , Marta Torrens 1,2,9* and NEURODEP GROUP ‡ 1 Addiction Research Group (GRAd), Neuroscience Research Program, Hospital del Mar Medical Research Institute (IMIM), Barcelona, Spain, 2 Institut de Neuropsiquiatria i Addiccions, Hospital del Mar, Barcelona, Spain, 3 Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, Barcelona, Spain, 4 Clinical Pharmacology Department, Hospital Universitari Germans Trias i Pujol (IGTP), Badalona, Spain, 5 Department of Pharmacology, Therapeutics and Toxicology, Universitat Autònoma de Barcelona (UAB), Cerdanyola del Vallès, Spain, 6 Department of Statistics and Operations Edited by: Research, Universitat Politècnica de Catalunya - BarcelonaTech, Barcelona, Spain, 7 Integrative Pharmacology and Systems Florence Vorspan, Neuroscience Research Group, Neuroscience Research Programme, Hospital del Mar Medical Research Institute (IMIM), Université de Paris, France Barcelona, Spain, 8 Psychiatry 2 Unit, Department of Clinical and Experimental Medicine, University Hospital of Pisa, Pisa, Italy, 9 Department of Psychiatry and Legal Medicine, Universitat Autònoma de Barcelona (UAB), Cerdanyola del Vallès, Spain Reviewed by: Marco Di Nicola, Agostino Gemelli University Polyclinic Background: Major depressive disorder (MDD) and cocaine use disorder (CUD) are (IRCCS), Italy Gabriel Rubio, related with disability and high mortality rates. The assessment and treatment of University Hospital October 12, Spain psychiatric comorbidity is challenging due to its high prevalence and its clinical severity, *Correspondence: mostly due to suicide rates and the presence of medical comorbidities. The aim of this Joan Ignasi Mestre-Pinto study is to investigate differences in brain derived neurotrophic factor (BDNF) and cortisol jmestre@imim.es Marta Torrens plasmatic levels in patients diagnosed with CUD-primary-MDD and CUD-induced-MDD mtorrens@psmar.cat and also to compare them to a sample of MDD patients (without cocaine use), a sample † These authors have contributed of CUD (without MDD), and a group of healthy controls (HC) after a stress challenge. equally to this work Methods: A total of 46 subjects were included: MDD (n = 6), CUD (n = 15), ‡ NEURODEP GROUP: Josep Martí CUD-primary-MDD (n = 16), CUD-induced-MDD (n = 9), and 21 HC. Psychiatric Bonany (J.M.B.), Julián Mateus (J.M.), comorbidity was assessed with the Spanish version of the Psychiatric Research Paola Rossi (P.R.), Claudio Tamarit (C.T.), Gabriel Vallecillo (G.V.) Interview for Substance and Mental Disorders IV (PRISM-IV), and depression severity was measured with the Hamilton Depression Rating Scale (HDRS). Patients were Specialty section: administered the Trier Social Stress Test (TSST) before and after the biological measures, This article was submitted to including BDNF, and cortisol levels were obtained. Addictive Disorders, a section of the journal Results: After the TSST, Cohen’s d values between CUD-primary-MDD and Frontiers in Psychiatry CUD-induced-MDD increased in each assessment from 0.19 post-TSST to 2.04 Received: 15 December 2021 Accepted: 02 February 2022 post-90-TSST. Pairwise differences among CUD-induced-MDD and both MDD and Published: 15 March 2022 HC groups had also a large effect size value in post-30-TSST and post-90-TSST. In Citation: the case of the BDNF concentrations, CUD-primary-MDD and CUD-induced-MDD in Fonseca F, Mestre-Pinto JI, post-90-TSST (12,627.27 ± 5488.09 vs.17,144.84 ± 6581.06, respectively) had a large Rodríguez-Minguela R, Papaseit E, Pérez-Mañá C, Langohr K, Barbuti M, effect size (0.77). Farré M, Torrens M and NEURODEP GROUP (2022) BDNF and Cortisol in Conclusion: Results suggest a different pathogenesis for CUD-induced-MDD with the Diagnosis of Cocaine-Induced higher levels of cortisol and BDNF compared with CUD-primary-MDD. Such variations Depression. should imply different approaches in treatment. Front. Psychiatry 13:836771. doi: 10.3389/fpsyt.2022.836771 Keywords: depression, cocaine use disorder, cortisol, dual diagnosis, brain derived neurotrophic factor (BDNF) Frontiers in Psychiatry | www.frontiersin.org 1 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression INTRODUCTION identification of outcome predictors in MDD not associated with substance use (19). Mental and substance use disorders (SUD) are related with 7% The hypothalamic-pituitary-adrenal (HPA) axis is involved of the global burden of disease as measured in disability-adjusted in the pathogenesis of MDD (20). Traditional studies observed life years (DALYs) and increasing significant mortality rates (1). a blunted stress response in MDD following a stress challenge, In women, DALYs are generally associated with major depression such as the Trier Social Stress Test (TSST) (21). Other studies, disorder (MDD) while, in men, addictive disorders are more however, found a hyper-response but only in patients with severe prevalent (1). depression (22). The HPA axis could also play a role in both The assessment and treatment of psychiatric comorbidity is CUD and MDD. In a study performed in cocaine-dependent challenging because of its high prevalence (2, 3) and clinical patients, an infusion of intravenous cocaine was associated with severity, mostly due to suicide rates and the presence of medical adrenocorticotropic hormone (ACTH) and cortisol levels and comorbidities (4–6). Moreover, it is essential to distinguish depressive symptoms measured with the Hamilton Depression between primary and induced MDD as they vary with respect Rating Scale (HDRS) (23). to prognosis, relapse risk (7), and response to antidepressants The BDNF belongs to the peptide family involved in (8). Traditionally, the implications of induced depression have neural plasticity, neurogenesis, and neural survival (24). It been minimized. Some clinicians believe that it involved a mild also has a key role in acute and chronic responses to syndrome that could revert with substance abstinence. As a substances of abuse. For instance, in a prospective study, BDNF consequence, induced MDD was not treated unless symptoms plasma concentrations were associated in cocaine addiction persisted (9). Subsequent research, however, demonstrates that with relapse risk in early recovery (25). Another study relapse risk is even greater in the case of induced MDD than demonstrated that plasma concentrations of BDNF during early primary (7). Moreover, some longitudinal studies demonstrate cocaine abstinence correlated with withdrawal syndrome and that patients with an initial diagnosis of induced MDD after craving (26). Moreover, BDNF is associated with a number some years developed a primary one (10). With respect to of psychiatric syndromes, including depression (27). The antidepressants, a number of studies and reviews indicate neurotrophic hypothesis of depression postulates that low levels differences in response depending on the type of depression with of BDNF could induce atrophy at limbic structures and prefrontal worse response to serotonin selective reuptake inhibitors (SSRI) cortex (28), whereas antidepressant treatment increases BDNF for induced MDD (11, 12). levels in depressed patients (29). With respect to dual diagnosis At the international level, cocaine is one of the most patients, BDNF levels are shown to present differences in samples widely used illicit drugs. The 2021 United Nations Office on from cocaine addicts with and without depression. Those with Drugs and Crime World Report estimated that around 20 a comorbid diagnosis of cocaine addiction and depression, million individuals aged 15–64 years (0.4%) had consumed irrespective of being primary or induced, show lower BDNF cocaine during 2019 (13). Its use is associated with medical levels (30). and psychopathological comorbidities, for example, increased Both MDD and SUD are complex diseases that result from risk of blood-borne infections (such as HIV and hepatitis C); changes in differing physiological systems. Thus, to better elevated rates of mortality; and increased prevalence of mental understand their pathophysiologies, the combined study of health disorders, mainly depression, psychotic episodes, and different systems and networks is required. In this regard, a recent suicide attempts (14). There are no approved pharmacological paper by Chen et al. finds that combining the results of serum treatments for cocaine use disorder (CUD), and only some weak BDNF, cortisol, and interferon-gamma could help in making an effects from psychotherapy are described (14, 15). accurate diagnosis of MDD (31). Individuals taking cocaine are reported to present a high risk At present, it is crucial to perform accurate diagnoses of CUD- of depression (16). The euphoria induced by acute cocaine use primary-MDD and CUD-induced-MDD. As the monoamine can induce a cycle of self-treatment of depressive symptoms, hypothesis of depression is proven insufficient to explain leading to a severe presentation of both CUD and MDD. differences between both types of MDD (32), it is essential Due to the reasons mentioned above, clinicians are faced to investigate the involvement of different systems in dual with having to distinguish between CUD-primary-MDD depression. We, therefore, carried out this study aimed at and CUD-induced-MDD in cocaine consumers. In a similar investigating differences in BDNF and cortisol plasmatic levels manner to psychiatric disorders, there are, however, no valid in patients diagnosed with CUD-primary-MDD and CUD- biomarkers for their correct identification. The diagnosis of induced-MDD and also to compare them with a sample of MDD (induced/primary) is based on the subjective identification MDD patients (without cocaine use), a sample of CUD (without of clinical symptoms, and there are no clear standards for MDD), and a group of healthy controls (HC) before and after a differential diagnosis. A recent study comparing DSM-5 criteria stress challenge. only found that “changes in weight or appetite” had a differing prevalence among the two disorders (17). The identification METHODS of biological markers in depressive disorders and SUD could help in the process of accurate diagnosis. MDD and CUD Subjects and Study Design share some neurobiological pathways (18), for example, brain In this cohort study, the sample included subjects with (i) derived neurotrophic factor (BDNF) and cortisol levels. They MDD (n = 6), (ii) CUD (n = 15), (iii) CUD and primary are described as being able to assist in the diagnosis and MDD (n = 16), and (iv) CUD and induced MDD (n = 9). All Frontiers in Psychiatry | www.frontiersin.org 2 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression patients were recruited at the addiction treatment facilities of the TSST Parc Salut Mar Institute of Neuropsychiatry and Addiction in The TSST is an acute stress test that consists of two tasks: public Barcelona, Spain. speaking and a mathematical task (39). Participants were asked Inclusion criteria were aged >18 years, Caucasian origin, to deliver a speech about their holidays or favorite book/film to body mass index 19–29 Kg/m2 , and the absence of any other a group of experts in nonverbal communication. After 5 min, psychiatric disorder and/or SUD other than MDD and/or CUD. three individuals (the audience) unfamiliar to the participant In patients with primary/induced MDD the most recent episode entered the room. The participant was instructed by one audience had to be in remission, and the 17-item HDRS (33, 34) score member (the spokesperson) to begin his/her prepared speech <6. In the CUD groups, subjects had to have maintained at least (without notes) for 5 min. If the individual paused, he/she was 4 weeks of substance abstinence prior to the trial as confirmed instructed by the spokesperson to continue. At the end of the by random urine controls. Cognitive or language limitations that speaking task, the individual was instructed to serially subtract precluded assessments, pregnancy or breast-feeding, use of anti- 17 from 3,164 or 2,043 (randomly) as quickly and accurately inflammatory drugs or monoamine oxidase inhibitors, and any as possible. The mental mathematic recitation continued for medical problem that might interfere in the study procedures 5 min, at the end of which the spokesperson instructed the were considered exclusion criteria. individual to stop, and the audience left the procedure room. HCs (n = 21) were included from a database of subjects willing Both tests were videorecorded. The experimental assessment to participate in medical research projects at the Pharmacology was conducted before the test (pre-TSST); immediately after Unit of the Hospital del Mar Institute of Medical Research (post-TSST); and after 30 (post30-TSST), 60 (post60-TSST), and (IMIM), Barcelona, Spain. In the HC group, the exclusion criteria 90 min (post90-TSST). At the same time points, physiological were any Axis I psychiatric disorder, family history of depression, and biochemical data were obtained. The TSST is proven useful and any SUD except nicotine. in inducing acute stress response even in patients with CUD (40). After basal clinical and psychiatric assessment, both patients and HCs participated in the stress experimental sessions with Biological Measurements the TSST. Heart rate (HR), systolic blood pressure (SBP), diastolic blood Subjects were admitted to the IMIM Clinical Research Unit pressure (DBP), respiratory rate (RR), and temperature were facilities at 08.00. Those presenting nicotine addiction were monitored by Dash 3,000 monitor (GE, Wisconsin, USA) at treated during the experimental session with patches according different times: before the test (pre-TSST); immediately after to their nicotine daily dose. A urine sample was collected for (post-TSST); and after 30 (post30-TSST), 60 (post60-TSST), and drug testing (Instant-View R , Multipanel 10 Test Drug Screen, 90 min (post90-TSST). At the same time points, blood samples Alfa Scientific Designs Inc., Poway, CA, USA). Participants were were collected from the subjects. required to be drug-free before the experimental session. The subjects remained sitting/lying in a calm laboratory environment Cortisol during the session with restricted social interactions. The TSST To assess cortisol levels, 5 ml of peripheral blood sample was was performed at 13.00 hours. This was carried out to (i) assure centrifuged at 4,000 rpm for 10 min. The serum obtained a similar waking time for all participants the day of the test, was frozen at −20◦ C until analysis was conducted by (ii) control activities that could affect HPA axis functioning, electrochemiluminescence, using an Immulite-2000 XPi (iii) avoid heterogeneity of the cortisol response, and (iv) analyzer (Siemens). assure a period of rest before the protocol was administered (35, 36). BDNF BDNF was obtained before (pre-TSST), immediately after (post- TSST), and at 90 min (post90-TSST). Five milliliters of peripheral Clinical and Psychiatric Assessments blood sample was centrifuged at 4,000 rpm for 10 min. The At baseline assessment, a closed-ended questionnaire was serum obtained was frozen at −20◦ C until analysis, which was used to record participants’ sociodemographic characteristics, performed with 500 microliters of serum by ELISA and the kit family history, medical assessment, history of substance use, Human BDNF Quantikine ELISA Kit of R&D-Vitro SA and and previous psychiatric treatment. Psychiatric diagnoses were polyclonal antibodies. performed according to DSM-IV-TR criteria with the Spanish version of the Psychiatric Research Interview for Substance and Statistical Analysis Mental Disorders IV (PRISM-IV) (37). PRISM was specifically A descriptive analysis of all variables of interest was carried out designed to deal with the issues of psychiatric diagnosis in separately in each of the study groups. For this purpose, the SUD patients. It helps differentiate primary disorders, SUD, and mean, median, standard deviation, and range were calculated. the expected effects of intoxication and withdrawal. Diagnoses Repeated-measure ANOVA models were used to analyze the obtained through the PRISM interview are demonstrated to have intragroup and intergroup changes of both the cortisol and good-to-excellent validity and test–retest reliability for primary- BDNF concentrations. The models included group condition as MDD and substance-induced MDD (38). In the MDD patients, a main factor in addition to all two- and three-way interactions. depression severity was evaluated with the Spanish version of the The computation of simultaneous confidence intervals and HDRS (34). adjusted p-values to guarantee a family-wise error rate of 0.05 Frontiers in Psychiatry | www.frontiersin.org 3 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression TABLE 1 | Sociodemographic and clinical characteristics of the sample at baseline (n = 67). HC (N = 21) MDD (N = 6) CUD-induced-MDD CUD-primary-MDD CUD (N = 9) (N = 16) (N = 15) Sex (Male) N (%) 14 (66.7) 5 (83.3) 7 (77.8) 13 (81.3) 12 (80) Age (Mean ± SD) 32.6 ± 4.8 45.7 ± 13.2 37.7 ± 11.4 44.8 ± 7.8 38.0 ± 9.5 Civil status (% Single) 10 (47.6) 2 (33.3) 6 (66.7) 6 (37.5) 10 (66.7) Work status (% Employed) 10 (47.6) 2 (33.3) 3 (33.3) 7 (43.8) 4 (26.7) Depression (MDD) HDRS (Mean ± SD) 0.57 ± 1.21 1.17 ± 1.83 0.56 ± 0.73 1.38 ± 1.09 0.73 ± 1.33 Age of onset first induced-MDD (Mean ± SD) - - 33.3 ± 11.8 - Age of onset first primary-MDD (Mean ± SD) - 36.2 ± 10.9 - 37.5 ± 6.4 - Number of episodes (Mean ± SD) - 1.8 ± 1.0 3.2 ± 2.1 2.5 ± 1.9 - Family history of depression (%) - 5 (83.3) 4 (44.4) 11 (68.8) 5 (33.3) Current antidepressant treatment (%) - 5 (83.3) 4 (44.4) 10 (62.5) 2 (13.3) Age of cocaine problematic use - - 26.3 ± 9.0 29.3 ± 6.8 26.3 ± 7.1 Nicotine use disorder (%) - - 7 (77.8) 12 (75) 11 (73.3) HC, healthy controls; MDD, major depression disorder; CUD, cocaine use disorder; SD, standard deviation. was based on the multivariate t distribution of the vector of main sociodemographic and clinical characteristics of the sample test statistics. are described in Table 1. The final groups were 21 HC, 6 MDD, 9 Next, one-way ANOVA models were fitted to compare CUD-induced-MDD, 16 CUD-primary-MDD, and 15 CUD. the study groups with the mean of the variables. The More than 76% of the total sample were single men aged model assumptions (homoscedasticity and normally distributed >32 years. All groups had low HDRS scores, and the depressed residuals) were checked with residual plots and the Levene groups had more than one MDD episode with a similar age of (homoscedasticity) and Kolmogorov-Smirnov tests, respectively. onset. Family history of depression and current treatment with If assumptions held and group differences were statistically antidepressants were also more prevalent in these groups. In the significant, the Bonferroni test was applied for the post hoc CUD groups, the age of onset of problematic cocaine use was very pairwise comparisons. Cohen’s d was used to quantify the effect similar, and current nicotine use was >73%. size of the pairwise differences among study groups (small: Five out of six subjects in the MDD group were on d ≤ 0.20; medium: d ≥ 0.50; large: d ≥ 0.80; very large: d varying types of antidepressants. In the CUD-induced and ≥ 1.30). Cohen’s d is a standardized score, analogous to a z primary-MDD groups, the majority of patients were also on score. Following Cohen’s effect size conventions, only differences antidepressant treatment. The CUD-primary-MDD patients higher than a medium effect size (d ≥ 0.50) were considered were treated with SSRIs although in the induced-MDD of relevance. group other types of antidepressants were prescribed. Types All data were analyzed using the IBM Corp. Released 2013 of antidepressants in the study groups are described in IBM SPSS Statistics for Windows, Version 22.0 (Armonk, NY: (Supplementary Table S1). IBM Corp.). In the case of the group comparisons, statistical significance was set at 0.05 (to protect against type I errors), Trier Social Stress Test (TSST) and for model assumption tests at 0.1 (to protect against type Biological Measures II errors). Changes in HR, SBP, DBP, and RR before and after the TSST are depicted in Supplementary Table S2. CUD and HC presented Ethics Statement significant changes over time in HR and DBP without differences The clinical protocol was approved by the local Research in the rest of the groups. Ethical Committee CEIC-Parc de Salut Mar, Barcelona, Spain (2009/3494/I and 2012/4751/I), and the study was conducted Cortisol in accordance with the Declaration of Helsinki and Spanish All groups showed a similar response pattern during the laws concerning clinical research. Volunteers were financially TSST follow-ups with the CUD-induced-MDD presenting the compensated. All subjects gave written informed consent prior highest cortisol concentrations in post-60 TSST and the CUD- to their participation in the study. primary-MDD the lowest cortisol concentrations in post-90 TSST (Figure 1). RESULTS One-way ANOVA tests yielded p-values >0.117 when comparing groups in each assessment (Table 2). A paired Demographic and Clinical Characteristics T-test showed significant within-group changes in cortisol A total of 67 subjects were included in the study to assess possible concentrations over time except in the MDD group with the differences in BDNF and cortisol levels during the TSST. The lowest range of change [9.48 ± 2.7–12.7 ± 3.47]. Frontiers in Psychiatry | www.frontiersin.org 4 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression FIGURE 1 | Mean cortisol concentrations during Trier social stress test (TSST). HC, healthy control (n = 21); MDD, major depression disorder (n = 6); CUD, cocaine use disorder (n = 15), CUD-primary-MDD, (n = 16), CUD-induced-MDD (n = 9). TABLE 2 | Mean cortisol concentrations during Trier social stress test (TSST). Cortisol mcg/dL Pre-TSST Mean ± SD Post-TSST Mean ± SD Post-30-TSST Mean ± SD Post-60-TSST Post-90-TSST Mean ± SD Mean ± SD HC (N = 21) 9.06 ± 3.67 12.24 ± 4.52* 10.27 ± 3.95* 11.45 ± 3.36 10.87 ± 3.72* MDD (N = 6) 9.48 ± 2.7 10.62 ± 1.35 9.63 ± 3.55 12.7 ± 3.47 11.05 ± 2.55 CUD-induced-MDD (N = 9) 9.09 ± 5 11.4 ± 6.95* 11.1 ± 5.15* 15.18 ± 2.17* 12.64 ± 1.55 CUD-primary-MDD (N = 16) 8.63 ± 2.17 10.48 ± 3.09 9.11 ± 2.6* 12.25 ± 4.3 8.69 ± 2.11* CUD (N = 15) 7.16 ± 3.03 9.03 ± 3.06* 8.6 ± 2.99 11.27 ± 4.25* 11.01 ± 6.12 One-way ANOVA p = 0.453 p = 0.253 p = 0.461 p = 0.117 p = 0.160 *Paired T-Test <0.05. TSST, Trier social stress test; HC, healthy controls; MDD, major depression disorder; CUD, cocaine use disorder; SD, standard deviation. After the TSST, Cohen’s d values between CUD-primary- paired T-test showed significant within-group changes in BDNF MDD and CUD-induced-MDD increased in each assessment concentrations over time except in the MDD group. from 0.19 post-TSST to 2.04 post-90-TSST. Pairwise differences Table 5 depicts the pairwise differences among the CUD- among CUD-induced-MDD and both MDD and HC groups had induced-MDD and HC groups that had the largest effect also a large effect size value post-30-TSST and post-90-TSST size value in the three assessments with values >0.94. The (Table 3). difference in BDNF concentrations between CUD-primary- MDD and CUD-induced-MDD post-90-TSST (12,627.27 ± 5488.09 vs.17,144.84 ± 6581.06, respectively) also had a large BDNF effect size (0.77). All groups showed a similar decreasing pattern in BDNF concentrations during TSST follow-ups. The CUD-induced- MDD group had the highest BDNF concentrations in each DISCUSSION assessment (Figure 2). One-way ANOVA demonstrated statistical differences post- The most important finding of this study is the different 90-TSST (p = 0.032); no differences, however, were reported response observed after a stress challenge (TSST) in the levels when performing Bonferroni’s post-hoc comparisons (Table 4). A of cortisol and BDNF in primary and induced depression. Frontiers in Psychiatry | www.frontiersin.org 5 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression As the diagnosis of depression is based on clinical criteria, detection of measurable biomarkers has implications in the study sometimes with suboptimal rates of validity and accuracy, the of the pathophysiology of depression and the introduction of effective treatments. As previously reported, the monoamine theory of depression cannot completely explain the pathogenesis of induced TABLE 3 | Effect size coefficient (Cohen’s d) pairwise comparisons of cortisol depressions (32), and other physiological systems should be concentrations in each assessment. studied. Stress is related to both MDD (41, 42) and SUD (43, 44), MDD CUD-induced CUD-primary CUD and in turn, cortisol is associated with stress. In our study, -MDD -MDD (N = 15) patients diagnosed with CUD-induced-MDD showed higher levels of cortisol after an acute stress challenge compared with Pre-TSST HC (N = 21) 0.12 0.01 0.14 0.56 CUD-primary-MDD ones. Such differences could indicate MDD (N = 6) 0.09 0.37 0.79 that varying mechanisms are involved in these two types CUD-induced-MDD 0.13 0.50 (N = 9) of depression. CUD-primary-MDD 0.56 Moreover, when analyzing BDNF plasma levels, we observed (N = 16) similar differences with higher concentrations of BDNF at Post-TSST HC 0.40 0.16 0.44 0.81 90 min after the TSST in the CUD-induced-MDD compared MDD 0.14 0.05 0.59 with the CUD-primary-MDD and MDD without cocaine CUD induced MDD 0.19 0.49 use. Traditional research describes lower levels of BDNF in CUD primary MDD 0.47 depressive patients (28). In our sample, higher levels of BDNF Post-30- HC 0.17 0.19 0.34 0.47 at 90 min were observed in the CUD and the CUD-induced- TSST MDD 0.32 0.18 0.33 MDD groups; surprisingly, the HC group showed the lowest CUD induced MDD 0.54 0.64 BDNF concentrations. Varying concentrations and level changes CUD primary MDD 0.18 depending on the type of depression could explain differences Post-60- HC 0.37 1.22 0.21 0.05 in therapeutical response to treatment between induced and TSST MDD 0.90 0.11 0.35 primary depressive disorders and the lack of response of some CUD induced MDD 0.79 1.08 of them. CUD primary MDD 0.23 The use of cortisol and BDNF levels as markers to differentiate Post-90- HC 0.05 0.54 0.70 0.03 cocaine-induced and primary depressions could help in the TSST MDD 0.80 1.06 0.01 design of personalized treatments. They would permit the CUD induced MDD 2.04 0.33 correct selection of antidepressant, thus avoiding prolonged CUD primary MDD 0.51 periods before patient response to treatment (45). Nevertheless, Cohen’s effect size: small (d > 0.20), medium (d > 0.50), large (d > 0.80), and very large reviews and meta-analysis have not clearly defined whether (d > 1.30). HC, healthy controls; MDD, major depression disorder; CUD, cocaine use there are differences in BDNF level increases depending on the disorder; TSST, Trier social stress test. antidepressant evaluated (29). With respect to BDNF levels in FIGURE 2 | Mean BDNF concentrations during Trier social stress test (TSST). HC, healthy control (n = 21); MDD, major depression disorder (n = 6); CUD, cocaine use disorder (n = 15), CUD-primary-MDD, (n = 16), CUD-induced-MDD (n = 9). Frontiers in Psychiatry | www.frontiersin.org 6 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression TABLE 4 | Mean BDNF concentrations during Trier social stress test (TSST). BDNF Pre-TSST Post-TSST Post-90-TSST HC (N = 21) 23,053.15 ± 4,818.28 14,993.06 ± 3,882.56* 11,782.55 ± 4,138.4* MDD (N = 6) 22,510.41 ± 5619.91 15,912.88 ± 3,456.21 14,872.87 ± 2,471.91 CUD-induced-MDD (N = 9) 28,040.91 ± 5,287.74 18,854.29 ± 4,662.07* 17,144.84 ± 6,581.06* CUD-primary-MDD (N = 16) 27,076.84 ± 8,457.33 18,735.27 ± 6,045* 12,627.27 ± 5,488.09* CUD (N = 15) 22,739.07 ± 5,673.17 16,423.37 ± 4,950.92* 15,888.35 ± 5,050.1* One-way ANOVA p = 0.082 p = 0.125 p = 0.032 *Paired T-Test <0.05. HC, healthy controls; MDD, major depression disorder; CUD, cocaine use disorder; TSST, Trier social stress test. additionally, such individuals presented increased vulnerability TABLE 5 | Effect size coefficient (Cohen’s d) pairwise comparisons of BDNF concentrations in each assessment. to depression and substance use (46). In this regard, another study evaluated salivary cortisol and hemodynamic data (BP BDNF MDD CUD-induced CUD-primary CUD and HR) response to the TSST in subjects prenatally exposed to -MDD -MDD (N = 15) cocaine. When comparing these subjects to nonexposed ones, it Pre-TSST HC (N = 21) 0.11 1.01 0.61 0.06 was observed that the former presented higher rates of cortisol MDD (N = 6) 1.02 0.58 0.04 levels before and after the TSST, a finding that suggests an CUD-induced-MDD 0.13 0.96 impaired response to stress in subjects prenatally exposed to (N = 9) cocaine (47). Another explanation could be the presence of CUD-primary-MDD 0.60 untreated depression as a risk factor for CUD. In a study (N = 16) evaluating an animal model, depressed rats administered more Post-TSST HC 0.24 0.94 0.76 0.33 cocaine than nondepressed ones showed higher concentrations MDD 0.69 0.51 0.11 of BDNF at the prelimbic cortex (48). In this regard, it should CUD-induced-MDD 0.02 0.50 be noted that our participants were either abstinent or in CUD-primary-MDD 0.42 remission from the last depressive episode. Previous research has Post-90- HC 0.80 1.08 0.18 0.91 not described differences depending on the time patients were TSST MDD 0.42 0.46 0.22 abstinent from psychostimulant drugs and TSST response (40). CUD-induced-MDD 0.77 0.22 For instance, the fact that patients did not present depressive CUD-primary-MDD 0.62 symptoms (HDRS < 6) hindered results being presented as a state marker although they could be interpreted as a trait maker. Cohen’s effect size: small (d > 0.20), medium (d > 0.50), large (d > 0.80), and very large In other words, CUD-induced and CUD-primary-MDD should (d > 1.30). HC, healthy controls; MDD, major depression disorder; CUD, cocaine use disorder; TSST, Trier social stress test. have different stress responses, and levels of BDNF and cortisol correlate with impaired stress responsiveness in these types of patients. The TSST has been useful in other kinds of research CUD, one study evaluated changes in plasma concentrations to discriminate different stress responses, for example, in young during detoxification. It was observed that chronic cocaine people exposed to prenatal cocaine (47). use was associated with lower levels of BDNF, and during More comprehensive knowledge regarding dual depression detoxification, the levels increased, correlating with cocaine biomarkers and the differences between primary and induced craving (26). In our sample, at baseline, subjects with higher depression are essential to introduce effective treatment, concentrations of BDNF were those with CUD and MDD, either particularly as the improvement of depressive symptoms requires primary or induced, although findings were nonsignificant. In at least 4 weeks after commencement of antidepressant therapy. addition, after the stress challenge, BDNF levels decreased in Moreover, previous research, in accordance with the present all groups although maintaining the higher levels those of the results, demonstrates differential neurobiological processes CUD-induced-MDD group. underlying induced and primary MDD (32, 49) suggesting poor Our findings do not signify a causal model, and it was outcomes with SSRI antidepressants (8, 12). For instance, in not possible to clarify whether the differences in cortisol and the case of depression with lower levels of BDNF, the first line BDNF levels were primary or secondary to induced/primary treatment should be those antidepressants that are shown to depression. Could a previously disrupted HPA axis be a marker raise BDNF levels, such as agomelatine (50). Martinotti reports of depression risk? Another study reported that patients with that, in patients with MDD, a correlation between depressive CUD presented a previous childhood history (parent neglect), symptom improvement and BDNF serum concentrations was higher scores in depression severity (measured by the SCL- observed after 2 weeks of agomelatine treatment. A recent review 90), and greater levels of ACTH and cortisol in plasma than also (29) focused on the effects of antidepressants in BDNF HCs. The authors concluded that early life events (neglect and levels and found that, in general, antidepressants increased the poor attachment to parents) influenced HPA axis function, and levels of BDNF. It was not possible, however, to identify the Frontiers in Psychiatry | www.frontiersin.org 7 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression differential effects by type of antidepressant; a better description provided their written informed consent to participate in of depression phenotypes is probably called for. this study. There are several limitations to this study. First, the sample size was small for all groups. Indeed, the strict inclusion criteria AUTHOR CONTRIBUTIONS made it difficult to find pure cocaine/depressed-only patients. For this reason, although our results suggest biochemical MT was the principal investigator of the grants supporting the differences between CUD-primary-MDD and CUD-induced- research. MT and MF were responsible for the study concept and MDD, such findings should be confirmed by the analysis design. MT, MF, JM-P, and FF designed the protocol. FF, RR-M, of a larger set of samples. A second limitation is that the CP-M, and EP selected the participants. JM-P, CP-M, EP, and MDD patients were under remission, and differences could, MF conducted the TSS test sessions. JM-P and KL performed the therefore, be underestimated. The decision to include patients statistical analysis and FF, JM-P, MF, and MT interpreted findings. in remission was made due to ethical reasons in order not FF, JM-P, and MB wrote the initial draft of the manuscript. MT to expose individuals with depression to a stress situation and MF provided critical revision of the manuscript for key (even in a controlled environment). This means that the intellectual content. All authors have read and agreed to the differences observed should be considered as trait markers or published version of the manuscript. risk factors to develop depression and not clinical depression itself. In previous research, the TSST is useful to discriminate FUNDING risk factors in stress response for healthy controls (51), the general population (52), and patients with active depression This work was supported by grants from the Instituto de (53). In addition, most of the depressed group participants Salud Carlos III–ISCIII Red de Trastornos Adictivos 2016 were under antidepressant treatment, which could influence (RD16/0017/0010 and RD16/0017/0003); Fondo de Investigación BDNF levels as previously described (29). Finally, the small Sanitaria (FIS) (PI09/02121, PI12/01838, PI16/00603); National sample size also hindered a proper evaluation of the effect R + D+I and funded by the Instituto de Salud Carlos of gender on the results. Previous authors observe differences III (ISCIII) and the European Regional Development Fund in the cortisol response to acute stress between genders after (FEDER) grant Juan Rodes (JR 16/00020); Ministerio de an acute stress challenge (54, 55). Studies evaluating the Sanidad, Política Social e Igualdad, Plan Nacional Sobre Drogas results between men and women are therefore essential to (PNSD) (2012I054); Suport Grups de Recerca AGAUR-Gencat adapt interventions. (2017 SGR 316, 2017 SGR 530); Ministerio de Economía Reliable biomarkers are needed to detect and diagnose y Competitividad (MTM2015-64465-C2-1-R); ISCIII-Redes de depression subtypes. One strength of the study is that Investigación Cooperativa Orientadas a Resultados en Salud these molecules could be analyzed routinely. Future studies (RICORS); Red de Investigación en Atención Primaria de investigating their involvement in the outcome and response to Adicciones (RIAPAd), grant number RD21/0009/0001. The treatment are warranted. funding agencies had no role in study design, data collection, interpretation, and no influence on the writing. DATA AVAILABILITY STATEMENT ACKNOWLEDGMENTS The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. We would like to thank the Pharmacology Department and the CAS Barceloneta nursing team for their valuable assistance ETHICS STATEMENT throughout the clinical part of the study. The authors are grateful to the participants and the psychiatry research support The studies involving human participants were reviewed and staff for their generosity and interest which made this the clinical protocol was approved by the local Research study possible. Ethical Committee CEIC-Parc de Salut Mar, Barcelona, Spain (2009/3494/I and 2012/4751/I) and the study was conducted SUPPLEMENTARY MATERIAL in accordance with the Declaration of Helsinki and Spanish laws concerning clinical research. Volunteers were financially The Supplementary Material for this article can be found compensated. All subjects gave written informed consent prior online at: https://www.frontiersin.org/articles/10.3389/fpsyt. to their participation in the study. The patients/participants 2022.836771/full#supplementary-material REFERENCES 2. Torrens M, Mestre-Pintó J-I, Domingo-Salvany A. Comorbidity of Substance Use and Mental Disorders in Europe. Luxembourg: Publications Office of the 1. Rehm J, Shield KD. Global Burden of Disease and the Impact of European Union (2015). Mental and Addictive Disorders. Curr Psychiatry Rep. (2019) 21:10. 3. Lai HMX, Cleary M, Sitharthan T, Hunt GE. Prevalence of comorbid doi: 10.1007/s11920-019-0997-0 substance use, anxiety and mood disorders in epidemiological surveys, Frontiers in Psychiatry | www.frontiersin.org 8 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression 1990-2014: a systematic review and meta-analysis. Drug Alcohol Depend. 21. Burke HM, Davis MC, Otte C, Mohr DC. Depression and cortisol responses (2015) 154:1–13. doi: 10.1016/j.drugalcdep.2015.05.031 to psychological stress: a meta-analysis. Psychoneuroendocrinology. (2005) 4. McCabe SE, West BT, Strobbe S, Boyd CJ. Persistence/recurrence of and 30:846–56. doi: 10.1016/j.psyneuen.2005.02.010 remission from DSM-5 substance use disorders in the United States: 22. Deuschle M, Schweiger U, Weber B, Gotthardt U, Körner A, Schmider J, et al. Substance-specific and substance-aggregated correlates. J Subst Abuse Treat. Diurnal activity and pulsatility of the hypothalamus-pituitary-adrenal system (2018) 93:38–48. doi: 10.1016/j.jsat.2018.07.012 in male depressed patients and healthy controls. J Clin Endocrinol Metab. 5. Davis L, Uezato A, Newell JM, Frazier E. Major depression and (1997) 82:234–8. doi: 10.1210/jcem.82.1.3689 comorbid substance use disorders. Curr Opin Psychiatry. (2008) 21:14–8. 23. Elman I, Breiter HC, Gollub RL, Krause S, Kantor HL, Baumgartner WA, doi: 10.1097/YCO.0b013e3282f32408 et al. Depressive symptomatology and cocaine-induced pituitary-adrenal axis 6. Martín-Santos R, Fonseca F, Domingo-Salavany A, Ginés JM, Ímaz ML, activation in individuals with cocaine dependence. Drug Alcohol Depend. Navinés R, et al. Dual diagnosis in the psychiatric emergency room in Spain. (1999) 56:39–45. doi: 10.1016/S0376-8716(99)00009-5 Eur J Psychiatry. (2006) 20:147–56. doi: 10.4321/S0213-61632006000300002 24. Rauti R, Cellot G, D’Andrea P, Colliva A, Scaini D, Tongiorgi E, et al. 7. Samet S, Fenton MC, Nunes E, Greenstein E, Aharonovich E, Hasin D. BDNF impact on synaptic dynamics: extra or intracellular long-term release Effects of independent and substance-induced major depressive disorder on differently regulates cultured hippocampal synapses. Mol Brain. (2020) 13:43. remission and relapse of alcohol, cocaine and heroin dependence. Addiction. doi: 10.1186/s13041-020-00582-9 (2013) 108:115–23. doi: 10.1111/j.1360-0443.2012.04010.x 25. Ornell F, Hansen F, Schuch FB, Pezzini Rebelatto F, Tavares AL, Scherer 8. Torrens M, Fonseca F, Mateu G, Farré M. Efficacy of antidepressants JN, et al. Brain-derived neurotrophic factor in substance use disorders: a in substance use disorders with and without comorbid depression. A systematic review and meta-analysis. Drug Alcohol Depend. (2018) 193:91– systematic review and meta-analysis. Drug Alcohol Depend. (2005) 78:1–22. 103. doi: 10.1016/j.drugalcdep.2018.08.036 doi: 10.1016/j.drugalcdep.2004.09.004 26. Corominas-Roso M, Roncero C, Eiroa-Orosa FJ, Gonzalvo B, Grau-Lopez L, 9. Blanco C, Alegría AA, Liu SM, Secades-Villa R, Sugaya L, Davies C, et al. Ribases M, et al. Brain-derived neurotrophic factor serum levels in cocaine- Differences among major depressive disorder with and without co-occurring dependent patients during early abstinence. Eur Neuropsychopharmacol. substance use disorders and substance-induced depressive disorder: Results (2013) 23:1078–84. doi: 10.1016/j.euroneuro.2012.08.016 from the National Epidemiologic Survey on Alcohol and Related Conditions. 27. Lin CC, Huang TL. Brain-derived neurotrophic factor and mental disorders. J Clin Psychiatry. (2012) 73:865–73. doi: 10.4088/JCP.10m06673 Biomed J. (2020) 43:134–42. doi: 10.1016/j.bj.2020.01.001 10. Magidson JF, Wang S, Lejuez CW, Iza M, Blanco C. Prospective study 28. Molendijk ML, Spinhoven P, Polak M, Bus BAA, Penninx BWJH, Elzinga of substance-induced and independent major depressive disorder among BM. Serum BDNF concentrations as peripheral manifestations of depression: individuals with substance use disorders in a nationally representative sample. evidence from a systematic review and meta-analyses on 179 associations Depress Anxiety. (2013) 30:538–45. doi: 10.1002/da.22122 (N=9484). Mol Psychiatry. (2014) 19:791–800. doi: 10.1038/mp.2013.105 11. Foulds JA, Douglas Sellman J, Adamson SJ, Boden JM, Mulder RT, Joyce PR. 29. Mosiołek A, Mosiołek J, Jakima S, Pieta A, Szulc A. Effects of Depression outcome in alcohol dependent patients: an evaluation of the role of antidepressant treatment on neurotrophic factors (Bdnf and igf-1) in independent and substance-induced depression and other predictors. J Affect patients with major depressive disorder (mdd). J Clin Med. (2021) 10:3377. Disord. (2015) 174:503–10. doi: 10.1016/j.jad.2014.11.052 doi: 10.3390/jcm10153377 12. Torrens M, Tirado-Muñoz J, Fonseca F, Farré M, Gonzalez-Pinto A, Arrojo 30. Pedraz M, Martín-Velasco AI, García-Marchena N, Araos P, Serrano A, M, et al. Clinical practice guideline on pharmacological and psychological Romero-Sanchiz P, et al. Plasma concentrations of BDNF and IGF-1 in management of adult patients with depression and a comorbid substance use abstinent cocaine users with high prevalence of substance use disorders: disorder. Adicciones. (2021) 1559. doi: 10.20882/adicciones.1559. [Epub ahead relationship to psychiatric comorbidity. PLoS ONE. (2015) 10:e0118610. of print]. doi: 10.1371/journal.pone.0118610 13. United Nations Office on Drugs and Crime. World Drug Report. (2021) (Sales 31. Chen S, Zhang Y, Yuan Y. The Combination of Serum BDNF, Cortisol No. E.21.XI.8). Geneva (2021). Available online at: https://www.unodc.org/ and IFN-Gamma Can Assist the Diagnosis of Major Depressive Disorder. res/wdr2021/field/WDR21_Booklet_4.pdf (accessed December 15, 2021). Neuropsychiatr Dis Treat. (2021) 17:2819–29. doi: 10.2147/NDT.S322078 14. Farrell M, Martin NK, Stockings E, Bórquez A, Cepeda JA, Degenhardt L, et al. 32. Fonseca F, Mestre-Pintó JI, Gómez-Gómez À, Martinez-Sanvisens D, Responding to global stimulant use: challenges and opportunities. Lancet. Rodríguez-Minguela R, Papaseit E, et al. The tryptophan system in cocaine- (2019) 394:1652–67. doi: 10.1016/S0140-6736(19)32230-5 induced depression. J Clin Med. (2020) 9:1–14. doi: 10.3390/jcm9124103 15. Brandt L, Chao T, Comer SD, Levin FR. Pharmacotherapeutic strategies for 33. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. treating cocaine use disorder-what do we have to offer? Addiction. (2021) (1960) 23:56–62. doi: 10.1136/jnnp.23.1.56 116:694–710. doi: 10.1111/add.15242 34. Bobes J, Bulbena A, Luque A, Dal-Ré R, Ballesteros J, Ibarra N, 16. Butler AJ, Rehm J, Fischer B. Health outcomes associated with crack-cocaine et al. Evaluación psicométrica comparativa de las versiones en use: Systematic review and meta-analyses. Drug Alcohol Depend. (2017) español de 6, 17 y 21 ítems de la Escala de valoración de Hamilton 180:401–16. doi: 10.1016/j.drugalcdep.2017.08.036 para la evaluación de la depresión. Med Clin. (2003) 120:693–700. 17. Alías-Ferri M, García-Marchena N, Mestre-Pintó JI, Araos P, Vergara- doi: 10.1016/S0025-7753(03)73814-7 Moragues E, Fonseca F, et al. Cocaine and depressive disorders: When 35. Allen AP, Kennedy PJ, Dockray S, Cryan JF, Dinan TG, Clarke G. The Trier standard clinical diagnosis is insufficient. Adicciones. (2021) 33:193–200. Social Stress Test: Principles and practice. Neurobiol Stress. (2016) 6:113–26. doi: 10.20882/adicciones.1321 doi: 10.1016/j.ynstr.2016.11.001 18. Gómez-Coronado N, Sethi R, Bortolasci CC, Arancini L, Berk M, 36. Narvaez Linares NF, Charron V, Ouimet AJ, Labelle PR, Plamondon H. Dodd S, et al. review of the neurobiological underpinning of comorbid A systematic review of the Trier Social Stress Test methodology: Issues in substance use and mood disorders. J Affect Disord. (2018) 241:388–401. promoting study comparison and replicable research. Neurobiol Stress. (2020) doi: 10.1016/j.jad.2018.08.041 13:100235. doi: 10.1016/j.ynstr.2020.100235 19. de Menezes Galvão AC, Almeida RN, de Sousa Junior GM, Leocadio- 37. Torrens M, Serrano D, Astals M, Pérez-Domínguez G, Martín-Santos R. Miguel MA, Palhano-Fontes F, de Araujo DB, et al. Potential biomarkers of Diagnosing Comorbid Psychiatric Disorders in Substance Abusers: Validity major depression diagnosis and chronicity. PloS ONE. (2021) 16:e0257251. of the Spanish Versions of the Psychiatric Research Interview for Substance doi: 10.1371/journal.pone.0257251 and Mental Disorders and the Structured Clinical Interview for DSM-IV. Am 20. Wichmann S, Kirschbaum C, Böhme C, Petrowski K. J Psychiatry. (2004) 161:1231–7. doi: 10.1176/appi.ajp.161.7.1231 Cortisol stress response in post-traumatic stress disorder, 38. Hasin D, Samet S, Nunes E, Meydan J, Matseoane K, Waxman R. Diagnosis panic disorder, and major depressive disorder patients. of Comorbid Psychiatric Disorders in Substance Users Assessed With the Psychoneuroendocrinology. (2017) 83:135–41. doi: 10.1016/j.psyneuen.2017. Psychiatric Research Interview for Substance and Mental Disorders for DSM- 06.005 IV. Am J Psychiatry. (2006) 163:689–96. doi: 10.1176/ajp.2006.163.4.689 Frontiers in Psychiatry | www.frontiersin.org 9 March 2022 | Volume 13 | Article 836771 Fonseca et al. BDNF and Cortisol in Dual Depression 39. Kirschbaum C, Pirke KM, Hellhammer DH. The ’Trier Social Stress Test’–a of Psychosocial Stress on Cognitive Performance: An Exploratory Pilot Study. tool for investigating psychobiological stress responses in a laboratory setting. Cogn Behav Neurol. (2018) 31:79–85. doi: 10.1097/WNN.0000000000000153 Neuropsychobiology. (1993) 28:76–81. doi: 10.1159/000119004 52. McLaughlin C, Schutze R, Pennell C, Henley D, Robinson M, Straker L. 40. Harris DS, Reus VI, Wolkowitz OM, Mendelson JE, Jones RT. et al. The anticipatory response to stress and symptoms of depression and Repeated psychological stress testing in stimulant-dependent patients. anxiety in early adulthood. Psychoneuroendocrinology. (2022) 136:105605. Prog Neuropsychopharmacol Biol Psychiatry. (2005) 29:669–77. doi: 10.1016/j.psyneuen.2021.105605 doi: 10.1016/j.pnpbp.2005.04.012 53. Young EA, Lopez JF, Murphy-Weinberg V, Watson SJ, Akil H. 41. Sanacora G, Yan Z, Popoli M. The stressed synapse 2.0: pathophysiological Hormonal evidence for altered responsiveness to social stress mechanisms in stress-related neuropsychiatric disorders. Nat Rev Neurosci. in major depression. Neuropsychopharmacology. (2000) 23:411–8, (2022) 23:86–103. doi: 10.1038/s41583-021-00540-x doi: 10.1016/S0893-133X(00)00129-9 42. Zhao MZ, Song XS, Ma JS. Gene × environment interaction in 54. Waldrop AE, Price KL, DeSantis SM, Simpson AN, Back SE, McRae AL, major depressive disorder. World J Clin Cases. (2021) 9:9368–75. et al. Community-dwelling cocaine-dependent men and women respond doi: 10.12998/wjcc.v9.i31.9368 differently to social stressors versus cocaine cues. Psychoneuroendocrinology. 43. Caccamise A, van Newenhizen E, Mantsch JR. Neurochemical mechanisms (2010) 35:798–806. doi: 10.1016/j.psyneuen.2009.11.005 and neurocircuitry underlying the contribution of stress to cocaine seeking. J 55. Wisniewski AB, Brown TT, John M, Cofranceso J, Golub ET, Ricketts Neurochem. (2021) 157:1697–713. doi: 10.1111/jnc.15340 EP, et al. Cortisol levels and depression in men and women using 44. McGrath AG, Briand LA. A potential role for microglia in stress- and heroin and cocaine. Psychoneuroendocrinology. (2006) 31:250–5. drug-induced plasticity in the nucleus accumbens: A mechanism for stress- doi: 10.1016/j.psyneuen.2005.08.002 induced vulnerability to substance use disorder. Neurosci Biobehav Rev. (2019) 107:360–9. doi: 10.1016/j.neubiorev.2019.09.007 Conflict of Interest: FF has received travel grants during the last 3 years 45. Rush AJ, Trivedi MH, Wisniewski SR, Nierenberg AA, Stewart JW, Warden from Lundbeck, Otsuka, Indivior, Pfizer, Gilead and Servier; she has also received D, et al. Acute and longer-term outcomes in depressed outpatients requiring grant/research support from Indivior and Servier. MT has been consultant/advisor one or several treatment steps: a STAR∗ D report. Am J Psychiatry. (2006) and/or speaker for Gilead Sciences, Merck Sharp & Dohme Corp, Indivior, 163:1905–17. doi: 10.1176/ajp.2006.163.11.1905 Mundipharma Pharmaceutics, Servier and Adamed. 46. Gerra G, Leonardi C, Cortese E, Zaimovic A. Dell’Agnello G, Manfredini M, et al. Adrenocorticotropic hormone and cortisol plasma levels directly The remaining authors declare that the research was conducted in the absence of correlate with childhood neglect and depression measures in addicted any commercial or financial relationships that could be construed as a potential patients. Addict Biol. (2008) 13:95–104. doi: 10.1111/j.1369-1600.2007. conflict of interest. 00086.x 47. Chaplin TM, Freiburger MB, Mayes LC, Sinha R. Prenatal cocaine exposure, Publisher’s Note: All claims expressed in this article are solely those of the authors gender, and adolescent stress response: a prospective longitudinal study. and do not necessarily represent those of their affiliated organizations, or those of Neurotoxicol Teratol. (2010) 32:595–604. doi: 10.1016/j.ntt.2010.08.007 the publisher, the editors and the reviewers. Any product that may be evaluated in 48. Zilkha N, Barnea-Ygael N, Keidar L, Zangen A. Increased relapse to cocaine- this article, or claim that may be made by its manufacturer, is not guaranteed or seeking in a genetic model for depression. Addict Biol. (2020) 25:e12756. doi: 10.1111/adb.12756 endorsed by the publisher. 49. Farré A, Tirado J, Spataro N, Alías-Ferri M, Torrens M, Fonseca F. Alcohol Induced Depression: Clinical, Biological and Genetic Features. J Clin Med. Copyright © 2022 Fonseca, Mestre-Pinto, Rodríguez-Minguela, Papaseit, Pérez- (2020) 9:2668. doi: 10.3390/jcm9082668 Mañá, Langohr, Barbuti, Farré, Torrens andNEURODEP GROUP. This is an open- 50. Martinotti G, Pettorruso M, De Berardis D, Varasano PA, Pressanti GL, De access article distributed under the terms of the Creative Commons Attribution Remigis V, et al. Agomelatine Increases BDNF Serum Levels in Depressed License (CC BY). The use, distribution or reproduction in other forums is permitted, Patients in Correlation with the Improvement of Depressive Symptoms. Int provided the original author(s) and the copyright owner(s) are credited and that the J Neuropsychopharmacol. (2016) 19:pyw003. doi: 10.1093/ijnp/pyw003 original publication in this journal is cited, in accordance with accepted academic 51. Beversdorf DQ, Carpenter AL, Alexander JK, Jenkins NT, Tilley MR, White practice. No use, distribution or reproduction is permitted which does not comply CA, et al. Influence of Serotonin Transporter SLC6A4 Genotype on the Effect with these terms. Frontiers in Psychiatry | www.frontiersin.org 10 March 2022 | Volume 13 | Article 836771