Gonadal hormone abnormalities in young patients with first-episode schizophrenia Authors: Qiang Hu, Jindong Wang, Jing Liang, Meihong Xiu, Shuangli Zhang, Fengchun Wu Published: 2024 ABSTRACT AbstractBackgroundGonadal hormones have been reported to be involved in the molecular mechanisms of schizophrenia (SCH). However, only a few studies have examined the gonadal hormone dysfunctions in first-episode schizophrenia (FES) patients, in particular in young patients with SCH. This study was designed to investigate the sex differences in gonadal hormones in young and antipsychotic-naïve FES patients.MethodsOne hundred and sixty-two patients with SCH and 74 healthy controls were recruited, and blood gonadal hormones, including estradiol (E2), follicular-stimulating hormone (FSH), progesterone (PROG), luteinizing hormone (LH), and testosterone (TESTO), were measured in young FES patients and controls.ResultsWe found that both male and female young FES patients showed gonadal hormone disturbances at the onset of psychosis. Male patients exhibited a significantly higher rate of abnormal E2 (25.6% vs 3.9%), while female patients had higher rates of abnormal FSH (0% vs 5.3%), PROG (0% vs 21.1%), LH (3.5% vs 17.1%), and TESTO (3.5% vs 13.2%) (all P < .05). Multivariate logistic regression analysis further identified that specific gonadal hormone indices, including E2, LH, and TESTO, were factors associated with sex differences in young FES patients, after controlling for age, smoking status, and body mass index.ConclusionsOur study reveals an overall gonadal hormone imbalance in young antipsychotic-naïve FES patients, highlighting sex differences at the onset of psychosis. Our study provides a foundation for further research into the role of gonadal hormones in the pathophysiology of SCH and the potential for personalized medicine approaches based on hormonal balance. Future studies were warranted to explore these differences and their implications for clinical practice to improve the treatment outcomes for individuals suffering from SCH. Significance Statement This work analyses the sex differences in sex hormones in young patients with drug-naïve first-episode schizophrenia (FES), which is often the initial debut of schizophrenia. Previous studies reported abnormal sex hormones in FES patients. Considering that hormones were influenced by age, the current research reexamined sex hormones in young and drug-naïve FES patients and assessed their sex differences. The results showed that young FES patients had abnormal hormone levels. Male and female patients had different abnormalities in sex hormones. Male patients had abnormal estradiol, and female patients had abnormal follicular-stimulating hormone, progesterone, luteinizing hormone, and testosterone. This study shows an overall sex hormone imbalance in young FES patients, highlighting sex differences at the onset of psychosis. This study provides a foundation for further research into the role of sex hormones in the pathophysiology of schizophrenia and the potential for personalized medicine approaches based on hormonal balance. INTRODUCTION Schizophrenia (SCH) is a chronic brain disorder that affects approximately 1% of the global population.1 The onset of SCH occurs most commonly in late adolescence or early adulthood.2,3 The disorder leads to severe impairments in various areas of life, including personal relationships, social interactions, and occupational functioning.4 Sex differences have long been observed in psychiatric diseases, including SCH, bipolar disorder, major depressive disorder, and autism spectrum disorder.5 These disorders are characterized by sexually dimorphic incidence rate, clinical symptoms, or prognosis.6-10 It is evident that in clinical practice the onset of illness is usually 3-5 years earlier in men than women. It is now generally accepted that there is one peak age of onset in male patients, between 21 and 25 years, and 2 peaks in female patients, 1 between 25 and 30 years and the other after 45 years.11,12 Studies have reported that male patients not only have an earlier age of onset, but also have a higher prevalence, more severe clinical symptoms, a higher propensity to have negative symptoms, different pharmacokinetics of antipsychotic drugs, lower rates of adverse effects, lower social functioning, and worse disease outcomes than female patients.5,13-15 There are also differences between the sexes in the relationship between specific disease risk variants and SCH (ie, “sex-specific” effects), or variants may have different effect sizes in each sex (ie, “sex-dependent” effects).16 The significant sex difference has suggested a possible implication of gonadal hormones in the pathophysiology of SCH.17 There is evidence that gonadal dysfunction alters neuronal and synapse numbers and synaptic morphology.18,19 Markham et al. reported that the development of central nervious system (CNS) in the embryo is strongly influenced by sex steroids,20 which affect brain function and lead to alterations in the brain during puberty, pregnancy, adulthood, menopause, and aging.20 In adolescence, gonadal hormones play an important role in influencing normal neurodevelopment, such as increased connectivity and gray matter reduction.21 Increased gonadal hormones during adolescence are associated with mood, cognition, memory, and various pathophysiological decreases in prefrontal, parietal, and temporal gray matter and with the increase in amygdala and hippocampal volumes, therefore, contributing to progressive changes in the brain areas.22,23 There is evidence that the brain areas most affected by sex hormones are closely linked to the pathophysiology of SCH.24 The estrogen (E2) hypothesis of SCH is supported by the late onset and the second peak of onset after 45 years in females. Across the lifespan, females experience a variety of endogenous fluctuations in hormone concentrations, including oscillations of E2 and progesterone (PROG) throughout the menstrual cycle, dramatic changes throughout pregnancy and the postpartum period, and an overall reduction in circulating hormone concentrations correlated with aging (eg, menopause).25 These fluctuations can influence the clinical symptoms of psychiatric disorders. Indeed, E2 deficiency has been shown to be highly correlated with the severity of psychiatric symptoms in female patients during menopause.26 On the other hand, E2 levels have been reported to be negatively correlated with symptoms in male patients with SCH.27 In addition, other sex hormones, such as testosterone (TESTO) and estradiol (E2), have also been implicated in the pathophysiology of SCH.28-32 Considering that gonadal hormone dysfunction is involved in SCH and affects its course, it is necessary to examine its role in the onset stage of psychosis. Previous studies have examined changes in the levels of gonadal hormones in first-episode schizophrenia (FES) patients.28-31,33,34 Notably, bioavailable gonadal hormone concentrations undergo a gradual decrease during aging in humans. Young patients (excluding menopausal patients) are better able to determine the potential role of gonadal hormones in sex differences in patients with SCH. Thus, young antipsychotic-naïve FES patients (aged from 18 to 30 years old) were recruited in this study to answer these questions: (1) What is the prevalence of gonadal steroid dysfunction in young FES patients? and (2) Is there a significant sex difference in the prevalence of gonadal hormone abnormalities in males and females, after adjusting for age, onset age, duration of illness, and BMI? METHODS Samples One hundred and sixty-two participants (male/female = 86/76) diagnosed with SCH were recruited using the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) criteria. The diagnosis was confirmed by 3 experienced psychiatrists using the Structured Clinical Interview for DSM-IV. They were experiencing their first episode of psychosis and were antipsychotic-naïve or had not previously received antipsychotics for more than 2 weeks. We defined them as antipsychotic-naive FES patients as described in a previous study.35 Other inclusion criteria were: (1) 16-35 years; (2) disease duration <60 months; and (3) male or female. The exclusion criteria were: (1) diagnosable organic brain disease; (2) mental retardation or autism spectrum disorders; (3) abuse or drug dependence; (4) major medical comorbidities, including hypertension, type 2 diabetes mellitus, etc.; (5) pregnancy or breast-feeding; (6) history of irregular menses or endocrine disease; (7) signs of hyperandrogenism; and (8) currently receiving hormone replacement therapy or hormonal contraceptives. None of the female patients reported using hormonal contraceptives. Seventy-four (44 males and 30 females) healthy controls (matched to the patient group subjects for age, sex, and smoking status) were recruited. Only persons without a history of mental disorders were enrolled in the study. All control subjects were physically healthy and were not taking any medications. Subjects suffering from diabetes mellitus, hypertension, cardiac, or thyroid disease were excluded from the study, as well as those who met the DSM-IV criteria for alcohol or substance abuse. The Ethics Committee of the hospital approved this study, and written informed consent was obtained from each patient. Determination of Gonadal Hormone Concentrations Gonadal hormone concentrations, including E2, follicular-stimulating hormone (FSH), PROG, luteinizing hormone (LH), and TESTO, were measured using the chemiluminescence immunoassays by an automated Beckman UniCel DxI800A analyzer (Beckman Coulter). In this study, the Chinese Endocrine Society and medical guidelines provided normal ranges of gonadal hormones for the general population aged from 18 to 50 years, which are derived from statistically analyzing a large number of healthy individuals to establish a normal range. The reference values for the male healthy population were as follows: E2 (55.07-142.99 pmol/L), FSH (1.27-19.26 mIU/mL), PROG (0.45-6.55 nmol/L), LH (1.24-8.62 mIU/mL), and TESTO (6.07-27.1 nmol/L). The reference values for the female healthy population were as follows: E2 (follicular phase: 55.65-543.79 pmol/L; mid-luteal phase:111.38-1006.74 pmol/L; ovulation phase:108-1624.86 pmol/L), FSH (follicular phase: 3.85-8.78 mIU/mL; mid-luteal phase:1.79-5.12 mIU/mL; ovulation phase:4.54-22.51 mIU/mL), PROG (follicular phase: 0.98-4.83 nmol/L; mid-luteal phase:16.41-59.02 nmol/L), LH (follicular phase: 2.12-10.89 mIU/mL; mid-luteal phase:1.2-12.86 mIU/mL; ovulation phase:19.18-103.03 mIU/mL), and TESTO (0.35-2.60 nmol/L). Statistical Analysis All data were analyzed using SPSS, version 22.0. Statistical significance was set at P < .05. The normally distributed numerical variable was reported as mean ± SD. Nonnormally distributed numerical variables were reported as median and IQRs. Chi-square tests were used to compare differences in the prevalence of abnormal gonadal hormones between male and female FES patients. Since the levels of most sex hormones were not normally distributed in male and female patients (Shapiro-Wilk test), these variables were reported with median and IQRs. Spearman’s rank correlation analyses were conducted to calculate the correlations between demographic and clinical characteristics and gonadal hormones for all patients or male or female FES patients. Further logistic regression analysis controlling for potential confounding variables that significantly correlated with sex in the univariate analysis was performed to identify gonadal hormones and other characteristics related to sex differences. In the regression model, the confounding factors included age, smoking status, duration of illness, and BMI. RESULTS Demographic and Clinical Characteristics in Males and Females The average ages of all recruited young FES patients and healthy controls were 24.2 ± 6.1 and 24.9 ± 5.9 years. The mean BMI was 22.3 ± 3.4 for patients and 21.5 ± 2.4 for controls. Smoking rates were 26/162, 16.0% in patients and 8/74, 10.8% in controls. We found no significant differences in age, sex, BMI, and smoking between patients and controls (all P > .05). Table 1 shows the demographics of male and female patients. We found no significant differences in age and disease duration between men and women (all P > .05). Although male and female patients had higher rates of abnormal BMI values, there was no significant difference in the rate of BMI abnormality between men and women (43% vs 43.1%, X2 = 0.15, P = .75). Table 1.The demographic and sex hormone levels in young patients with schizophrenia.Male patientsn = 86Female patientsn = 76Age (ys)23.5 ± 5.324.9 ± 6.8Disease duration (ms)26.5 ± 21.130.8 ± 19.4Onset age (ys)21.0 ± 5.022.9 ± 6.8Smoke status Yes23 (26.7%)3 (3.9%) No63 (73.3%)73 (96.1%)Drinking Yes8 (9.3%)0 (0%) No78 (91.7%)76 (100%)Education (ys)12.0 ± 3.511.8 ± 3.7BMI (kg/m2)22.9 ± 3.421.7 ± 3.2E2 (pmol/L), median (IQR)103.00 (78.75,124.00)238.00 (147.00,420.00)FSH (mIU/mL), median (IQR)3.55 (2.75,5.00)5.75 (4.13,7.08)PROG (nmol/L), median (IQR)1.80 (1.11,2.39)2.84 (1.61,13.97)LH (mIU/mL), median (IQR)3.58 (2.54,4.81)5.59 (3.28,9.55)TESTO (nmol/L), median (IQR)13.85 (10.83,17.03)1.80 (1.40,2.40)Abbreviations: BMI, body mass index; E2, estradiol; FSH, follicular-stimulating hormone; LH, luteinizing hormone; ms, months; PROG, progesterone; TESTO, testosterone; ys years. The demographic and sex hormone levels in young patients with schizophrenia. Abbreviations: BMI, body mass index; E2, estradiol; FSH, follicular-stimulating hormone; LH, luteinizing hormone; ms, months; PROG, progesterone; TESTO, testosterone; ys years. Correlation of Gonadal Hormones With Clinical Characteristics The median values of E2, FSH, PROG, LH, and TESTO of all patients or the male or female patient groups are shown in Table 1. Spearman’s rank correlation analysis showed a significant association between age and TESTO concentration (r = −0.17, P = .04) and between BMI and the concentrations of E2 (r = −0.18, P = .02) and PROG (r = −0.27, P < .001). Additionally, onset age correlated with FSH (r = 0.17, P = .03) and TESTO (r = −0.19, P = .01). Further correlation analysis for males and females respectively found that these associations were only present in male patients, but not in female patients (Table 2). Table 2.Associations between demographic and clinical variables and gonadal hormones in male and female FES patients.Statistic (P-value)aE2FSHPROGLHTESTO Male FES patients (n = 86) Age r −0.090.12−0.21−0.16−0.36 P .40.28.05.13.001Onset age r −0.190.14−0.23−0.20−0.38 P .07.19.03.06<.001Disease duration r 0.23−0.020.110.150.11 P .03.84.30.16.30BMI r −0.120.22−0.380.01−0.28 P .26.046<.001.95.009 Female FES patients (n = 76) Age r 0.120.150.010.10-0.03 P .29.18.93.41.80Onset age r 0.050.17-0.020.08-0.04 P .66.14.89.47.76Disease duration r 0.10−0.080.040.01−0.01 P .39.48.72.90.93BMI r −0.050.04−0.110.04−0.09 P .69.71.37.71.47 aIndicates Spearman’s rank correlation analysis.Abbreviations: BMI, body mass index; E2, estradiol; FSH, follicular-stimulating hormone; LH, luteinizing hormone; P, progesterone; T, testosterone. Associations between demographic and clinical variables and gonadal hormones in male and female FES patients. aIndicates Spearman’s rank correlation analysis. Abbreviations: BMI, body mass index; E2, estradiol; FSH, follicular-stimulating hormone; LH, luteinizing hormone; P, progesterone; T, testosterone. We found no significant associations between gonadal hormones and BMI, age, and sex in control subjects (all P > .05). Sex Differences in the Rates of Gonadal Hormone Abnormalities in Patients We calculated the prevalence of abnormal gonadal hormones in young FES patients and control subjects according to the normal reference values for gonadal hormones. No gonadal hormone abnormalities were observed in either male or female control subjects. Of all the patients, we found that 25 (15.4%) had abnormal E2 concentration (male/female = 22/3, X2 = 14.4, P < .001), 4 (2.5%) had abnormal FSH (male/female = 0/4, X2 = 4.6, P = .03), 16 (9.9%) had abnormal PROG (male/female = 0/16, X2 = 20.0, P < .001), 16 (9.9%) had LH abnormalities (male/female = 3/13, X2 = 8.4, P = .004), and 13 (8.0%) had TESTO abnormalities (male/female = 3/10, X2 = 5.1, P = .02). In addition, the abnormalities in hormone levels were also identified as being increased or decreased in Table 3. Table 3.The rates of abnormal gonadal hormones in male and female patients with schizophrenia.Gonadal hormoneTotalsampleMale patientsFemale patients X 2(P)E214.4 (< .001)Increase660Normal1376473Decrease19163FSH4.6 (.03)Increase000Normal1588672Decrease404PROG20.0 (< .001)Increase909Normal1468660Decrease707LH8.4 (.004)Increase734Normal1478364Decrease808TESTO5.1 (.02)Increase10010Normal1498366Decrease330Abbreviations: E2, estradiol; FSH, follicular-stimulating hormone; LH, luteinizing hormone; PROG, progesterone; TESTO, testosterone. The rates of abnormal gonadal hormones in male and female patients with schizophrenia. Abbreviations: E2, estradiol; FSH, follicular-stimulating hormone; LH, luteinizing hormone; PROG, progesterone; TESTO, testosterone. Compared to female FES patients, male patients had a higher rate of E2 abnormalities (25.6% vs 3.9%) and lower rates of FSH abnormalities (0% vs 5.3%), PROG abnormalities (0% vs 21.1%), LH (3.5% vs 17.1%), and TESTO abnormalities (3.5% vs 13.2%). The differences in the rates of abnormal gonadal hormones between male and female patients were statistically significant (all P < .05). Univariate logistic regression analysis was used to analyze sex differences in gonadal hormones in all patients, after controlling for age, BMI, duration of disease, and smoking status. Given the strong relationship between age and onset age (r = 0.96), we included only age in the logistic regression model. We found a significant difference in the prevalence of abnormal E2 concentration between male and female patients (odd ratio: 7.936, 95% CI, 2.081-30.266), as well as significant differences in the prevalence of abnormal LH concentration (odd ratio: 7.411, 95% CI, 1.781-30.834) and abnormal TESTO concentration (odd ratio: 7.450, 95% CI, 1.436-38.660). Multivariate logistic regression analysis revealed that 3 gonadal hormone indices, including E2, LH, and TESTO, as well as age and BMI were associated with sex differences in FES patients (Table 4). Table 4.Logistic regression of the gonadal hormone concentrations related to sex difference in patients with schizophrenia.BStd.ErrorWalsSig.Exp(B)(95% CI)(Constant)1.5361.3311.3320.2484.646Age (ys)0.1030.0368.2480.0041.108 (1.033, 1.189)Disease duration (ms)0.0090.0100.9040.3421.009 (0.990, 1.028)BMI (kg/m2)−0.1880.0658.4060.0360.829 (0.730, 0.941)Smoking status−2.5500.74911.5920.0010.078 (0.018, 0.339) Gonadal hormones E2−1.7640.6946.4660.0110.171 (0.044,0.667) TESTO1.7730.8454.4060.0365.891 (1.125, 30.861) LH1.8280.7416.0930.0146.224 (1.457, 26.581)Abbreviations: BMI, body mass index; E2, estradiol; LH luteinizing hormone, ms, months; TESTO, testosterone; ys, years. Logistic regression of the gonadal hormone concentrations related to sex difference in patients with schizophrenia. Abbreviations: BMI, body mass index; E2, estradiol; LH luteinizing hormone, ms, months; TESTO, testosterone; ys, years. DISCUSSION In the present study, the main findings were as follows: (1) FES patients showed abnormal gonadal hormone concentrations both in males and females; (2) men had a higher rate of abnormal E2, and women had higher rates of abnormal FSH, PROG, LH, and TESTO; and (3) after controlling for confounding factors, E2, LH, and TESTO are key factors associated with sex difference in young antipsychotic-naive FES patients. We found abnormal gonadal hormone levels and rates of hormone abnormalities compared with the reference hormone values in young antipsychotic-naïve FES patients with SCH. Men had abnormal levels of E2, LH, and TESTO, and women showed abnormal levels of all 5 gonadal hormones measured in this study. Accumulating evidence reports a potential role of gonadal hormones in the regulation of certain CNS processes, such as cognitive functioning and mood. Previous studies in FES patients with SCH have also reported alterations in gonadal hormones.28,34,36-39 This study is the first to reveal abnormal gonadal hormones in young antipsychotic-naive FES patients with SCH. We also found significant sex differences in gonadal hormone concentrations in young FES patients, but not in control subjects. There is generally accepted evidence that SCH is expressed differently in male and female patients,40-43 much of which has been attributed to E2, PROG, and TESTO.44 In this study, the rate of abnormal E2 concentrations was higher in male patients than in female patients. Other studies have also indicated that compared to healthy men, men with SCH had lower levels of E2.45 Estrogens, specifically 17β-E2, play important roles in the CNS, influencing various neuronal functions and participating in the modulation of synaptic plasticity, memory, and neuroprotection.46-50 Neuron-derived E2 is essential for astrocyte activation and the production of neuroprotective factors, such as brain-derived neurotrophic factor and insulin-like growth factor 1, as well as the glutamate transporter 1.47 Psychosis occurs more frequently in women during periods of E2 decline, such as the menstrual phase of the menstrual cycle, postpartum, and postmenopausal, which may be related to hormonal fluctuations (eg, changes in E2 levels). In contrast, when E2 concentrations are high during pregnancy, relapse rates are reduced.20 Animal studies have found that E2 modulates multiple neurotransmitter systems in our brain and influences the activity of neural circuits,51,52 which is implicated in the cognition and mood of SCH. On the other hand, we found that female patients had higher rates of abnormal FSH, PROG, LH, and TESTO concentrations than male patients. E2 plays a more critical role in men and FSH, PROG, LH, and TESTO play a more important role in women. We speculate that the significant sex differences in onset age, clinical symptoms, and treatment outcomes reported in previous literature may be due to abnormal regulation of gonadal hormones in males and females in SCH. Namely, abnormal E2 in male patients affects the disease more severely relative to FSH, PROG, LH, and TESTO in female patients. Indeed, compared to other gonadal hormones, E2 has been widely investigated in individuals suffering from SCH, particularly with the formulation and endorsement of the E2 hypothesis of SCH.49,53 When E2 levels are abnormal, it may lead to alterations in synaptic function, synaptic plasticity, and neurogenesis.34,47 In contrast, FSH, PROG, LH, and TESTO do not appear to be as important for neurodevelopment and neuronal functions. However, the relationship between hormones and SCH is complex, and our speculations may oversimplify this association. Other factors, such as the balance of multiple hormones, their receptors, interactions with other neurotransmitters, and environmental risk factors, also play a role. In particular, we found that the male and female patients were well-matched in terms of a number of demographic features, including age of onset. However, differences in age of onset have been reported to be among the most prominent sex differences in patients with SCH. In our study, although the onset age was higher in female patients than in male patients (22.9±6.8 years vs 21.0±5.0 years), the difference between the 2 groups was insignificant, making the cohort somewhat atypical. This may limit the generalizability of our findings. Another major concern is that this study found high rates of abnormal BMI in both male and female patients. Obesity has been reported to be associated with chronic low-grade inflammation with changed pro-inflammatory cytokines and hypothalamic hormones, which may have an impact on the gonadal hormones.54,55 Indeed, studies have reported decreased TESTO levels in obese men.56 On the other hand, weight loss interventions have been reported to affect the activity of the hypothalamic-pituitary-adrenal-thyroid axis, which in turn affects the release of hormones.57-59 Despite these findings highlighting the effects of obesity or weight loss on the neuroendocrine axis and indirectly on gonadal hormone levels, we found no significant difference in BMI values between men and women and no correlation between BMI and hormone levels. Limitations First, examining gonadal hormones’ specific effects in humans (in a naturalistic setting) is challenging. Different approaches to patient selection and measurement methods contribute to inconsistencies among the studies. For instance, the effects of different reproductive life stages in female patients may lead to diverse results. Second, it is unclear whether blood hormone levels correspond to CNS hormone levels. Thus, alterations in blood levels as evidence for gonadal hormones mediating sex differences in SCH should be interpreted cautiously. Third, the severity of diseases was not assessed using a rating scale and we were unable to compare sex differences in symptom severity and their relationship with gonadal hormones in patients. A future longitudinal follow-up study is needed to validate the association between gonadal hormones and disease severity in patients with SCH. Fourth, participants were antipsychotic-naïve or had been treated with antipsychotics for less than 14 days. However, we cannot provide detailed information on how many patients were on medications or adjust for this in the analysis. It is known that dopamine-receptor antagonists, commonly used in the treatment of psychotic disorders, may have an impact on gonadal hormones, which should be considered as a limitation. CONCLUSIONS This study found that hormonal imbalances may be a common feature at the onset of SCH. Hormonal imbalances differed between young male and female patients, with men exhibiting more pronounced disruptions in E2 levels and women showing a wider range of hormonal abnormalities. In this study, the confounding factors (eg, age, antipsychotic medication, BMI, and duration of disease) were controlled for, and our findings suggest that the observed hormonal differences are not solely due to external factors but may be intrinsic to the disease itself and may vary by sex. The findings underscore the importance of considering hormonal factors in the comprehensive management of SCH, particularly in young FES patients. Further research is needed to explore the underlying mechanisms by which hormonal imbalances are associated with SCH and to determine how these might affect the manifestation and treatment of this disorder in both men and women.