Sex differences of brain cortical structure in major depressive disorder

Abstract Background Major depressive disorder (MDD) has different clinical presentations in males and females. However, the neuroanatomical mechanisms underlying these sex differences are not fully understood. Objective The purpose of present study was to explore the sex differences in brain cortical thickness (CT) and surface area (SA) of MDD and the relationship between these differences and clinical manifestations in different gender. Methods High-resolution T1-weighted images were acquired from 61 patients with MDD and 61 healthy controls (36 females and 25 males, both). The sex differences in CT and SA were obtained using the FreeSurfer software and compared between every two groups by post hoc test. Spearman correlation analysis was also performed to explore the relationships between these regions and clinical characteristics. Results In male patients with MDD, the CT of the right precentral was thinner compared to female patients, although this did not survive Bonferroni correction. The SA of several regions, including right superior frontal, medial orbitofrontal gyrus, inferior frontal gyrus triangle, superior temporal, middle temporal, lateral occipital gyrus, and inferior parietal lobule in female patients with MDD was smaller than that in male patients (P < 0.01 after Bonferroni correction). In female patients, the SA of the right superior temporal (r = 0.438, P = 0.008), middle temporal (r = 0.340, P = 0.043), and lateral occipital gyrus (r = 0.372, P = 0.025) were positively correlated with illness duration. Conclusion The current study provides evidence of sex differences in CT and SA in patients with MDD, which may improve our understanding of the sex-specific neuroanatomical changes in the development of MDD.


Introduction
Major de pressi ve disorder (MDD) is a significant threat to people's health, c har acterized by symptoms such as de pression, cogniti ve impairment, loss of interest, and low energy (Song et al., 2018 ).It is predicted to become the leading cause of global disease burden by 2030 (Malhi and Mann, 2018 ).MDD is more common in women, with twice the pr e v alence compar ed to men fr om adolescence to adulthood (Nolen-Hoeksema, 2016 ).Hormonal changes, such as estrogen and progesterone imbalance in females and low testoster one le v els in males , ha ve been implicated in the sex differences in depression (Kundakovic and Rocks, 2022 ).Additionally, biological, psychological, and social-cultural factors contribute to the sex differences in depression (Hyde and Mezulis, 2020 ).
Clinical c har acteristics of MDD differ between males and females.Males with MDD often exhibit se v er e symptoms suc h as impulsivity , irritability , and insomnia, while females with MDD ar e mor e likel y to pr esent with somatic symptoms and atypical depression symptoms including increased appetite, weight gain, fatigue, and difficulty sleeping (Kim et al., 2015 ;Yang et al., 2017 ).Females also tend to have a better response to serotonergic antidepr essants compar ed to males (Sr amek et al., 2016 ).
Neur oima ging studies have shown differences in brain activity between female and male patients with MDD.Male patients with MDD exhibit lo w er amplitude of lo w-frequenc y fluctuation values in the left superior/middle frontal gyrus, and higher amplitude of lo w-frequenc y fluctuation values in the left postcentral gyrus when compared to female patients (Yao et al., 2014 ).Structural brain studies have also identified sex differences in gray matter volume (GMV) changes in individuals with MDD (Taki et al., 2005 ;Yang et al., 2017 ).GMV is determined by cortical thickness (CT) and surface area (SA), both of which have genetic and phenotypic independence, and reflect the different properties of gray matter structur es (Gr asby et al. , 2020 ;W inkler et al. , 2010 ).CT reflects the number of cells within a column, while SA reflects the number of cortical columns (Lee et al., 2021 ).Ther efor e, CT and SA measur ements should be taken into consideration separ atel y and prioritized over GMVs (Hanford et al., 2016 ;Winkler et al., 2010 ).
Most studies on sex-specific brain structure alterations in depr ession hav e focused on GMV (Kong et al. , 2013 ;Yang et al. , 2017 ), with limited r esearc h on CT and SA (Hu et al. , 2022 ;Li et al. , 2020 ).T hus , our study aimed to investigate the sex differences in CT and SA in individuals with MDD and explore the associations between these differences and clinical manifestations in different sexes.

Participants
T his study in volv ed patients dia gnosed with MDD obtained fr om the Department of Psychiatry Clinic, West China Hospital of Sic huan Univ ersity.All patients underwent a compr ehensiv e interview conducted by two seasoned psychiatrists, following the diagnostic criteria for MDD as per the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV).Additionally, their mental condition was assessed using the 17-item Hamilton Rating Scale for Depression (Hamilton Depression Rating Scale, HAMD-17).We included only those patients who were in depressi ve e pisodes and scored a minimum of 18 on the HAMD-17, had not r eceiv ed an y antidepr essant tr eatment for at least 1 year pr eceding the imaging examination, and did not have any other neurological or affective disorders.
We recruited the healthy control (HC) group through poster advertising.All volunteer participants underwent the SCID-I Non-Patient Edition scale to rule out the possibility of neur opsyc hiatric disorders .T he exclusion criteria for all participants were as follows: (i) history of traumatic brain damage; (ii) organic brain lesions, suc h as br ain tumors; (iii) alcohol or drug abuse (as per DSM-IV diagnostic criteria); (iv) pregnancy; and (v) neurological diseases such as e pile psy and multiple sclerosis.
Based on these inclusion and exclusion criteria, we recruited 61 patients with untreated MDD (25 male, 36 female) and 61 healthy volunteers (25 male, 36 female) in the study.An experienced radiologist confirmed the absence of abnormalities on conventional ma gnetic r esonance ima ging (MRI) scans for all participants .T he Medical Ethics Committee of the Second People's Hospital of Yibin a ppr ov ed this study (no. 2014-056-01), and all participants provided written informed consent.

MRI Data Acquisition
High-resolution anatomical images of the whole brain were obtained using a 3 Tesla MRI system (EXCITE, General Electric), equipped with an eight-c hannel phased-arr ay head coil.We emplo y ed a 3D, sa gittal, ma gnetization-pr epar ed r a pid gr adient ec ho (MPRAGE) sequence to acquir e thr ee-dimensional T1-weighted images .T he parameters used were: 156 axial slices; slice thickness, 1 mm; TR, 1900 ms; TE, 2.26 ms; flip angle, 12 • ; FOV, 240 × 240 mm; and data matrix, 256 × 256.All participants were fitted with foam padding and earplugs, and instructed to remain still during scanning (Yang et al., 2017 ).

Data Preprocessing
The cortical surface of the 3D T1 image was constructed using the F reeSurfer softw are ( http://surfer .nmr.mgh.har vard.edu/,v.5.3.0).This software measures the CT and SA of the entire cortex by automatically performing surface reconstruction, transformation, and high-resolution inter-individual calibration steps (Han et al., 2014 ;Muschelli et al., 2018 ).First, the orig inal Dig ital Imag ing and Communications in Medicine images for T1-weighted data are converted to the Neuroimaging Informatics Technology Initiative format, then to the MGZ format, and head movement correction is applied.Next, completing the affine transformation from the original volume to the Montreal Neurological Institute (MNI) 305 atlas and carrying out the Talair ac h coordinate system transformation.Then standardizing the signal strength of the original volume, performing the strength correction, and removing the deviation of the signal strength, and generating the original curved surface and performing automatic local anatomical correction, and expanding the generated cortical image and converting it to the spherical distribution template .T he ima ges wer e then smoothed using a 25-mm, full-width at half-maximum Gaussian kernel.The r esults fr om the automatic ima ge pr ocessing for eac h subject wer e c hec k ed man uall y to assess whether the br ain surface r econstruction was consistent with the grey matter boundary.If inconsistent, manual editing is performed.We obtained CT by calculating the shortest distance from the gray/white boundary to the gr ay/cer ebr ospinal fluid boundary at each vertex.The SA of eac h hemispher e is calculated by adding up the ar ea of all tessellations on the gray matter surface (Deng et al., 2019 ;Qiu et al., 2014 ;Xiao et al., 2023 ).

Sta tistical Anal ysis
We compared demographic and clinical data across the four groups (male MDD, female MDD, male HC, and female HC) using SPSS (v.25.0).Age and education years among groups were compared using a tw o-w ay analysis of co variance .T he illness duration and HAMD score in male and female groups with MDD did not exhibit normal distributions and were analyzed as nonparametric using the Mann-Whitney U -test.CT and SA among the four groups were analyzed using a tw o-w ay analysis of covariance , taking sex (male , female) and diagnosis (MDD, HC) as between-participant factors, and intr acr anial volume as a covariate in MATLAB.The statistical results were adjusted using Bonferr oni corr ection with a significance le v el of P < 0.01.We identified br ain r egions with sex differ ences in CT and SA among the four gr oups, whic h wer e then compar ed using a post hoc test (Student-Ne wman-K euls method) between e v ery two gr oups .T he results wer e corr ected using Bonferr oni corr ection.The statistical thr eshold was set at P < 0.05.Spearman correlation analysis was performed to investigate the correlation of brain regions with sex differences in MDD and their clinical characteristics (illness duration, HAMD score), with the statistical significance threshold set at P < 0.05.

Participants' Demographic and Clinical Characteristics
P articipants' demogr a phic and clinical c har acteristics ar e pr esented in Table 1 .No significant differences were observed between the four groups in terms of age or years of education ( P > 0.05).Similarl y, ther e was no significant difference in HAMD score between male and female patients with MDD ( P > 0.05).Ho w e v er, male patients with MDD had a longer illness duration than female patients ( P < 0.05).

The Main Effect of Sex on CT and SA
The effect of sex on CT was observed in the right pr ecentr al gyrus ( P < 0.01 after FDR correction), but this was not maintained after a ppl ying the Bonferr oni corr ection (a statistical thr eshold of P < 0.01).Refer to the Supplementary Material for further discussion and descriptions of the results.A significant effect of sex on SA was found in bilateral inferior frontal gyrus triangle, middle tempor al, later al occipital, left postcentr al, r ostr al anterior cingulate, supr amar ginal gyrus, insula, right superior frontal, medial orbitofr ontal, superior tempor al gyrus, and inferior parietal lobule (Bonferr oni corr ection, P < 0.01) (Fig. 1 ).On examining the effect of diagnosis and sex-diagnosis interactions on CT and SA, respectiv el y, no cluster surviv es Bonferr oni corr ection or FDR corr ection at a threshold of P < 0.01.

Post Hoc Comparison Results
Post hoc analysis revealed that the CT of the right pr ecentr al gyrus in the male MDD group was thinner than that of the female MDD group ( Supplementary Table S1 ).The SA of the right superior fr ontal, medial orbitofr ontal gyrus, inferior fr ontal gyrus triangle, superior temporal, middle temporal, lateral occipital gyrus, and inferior parietal lobule in the female MDD group was smaller than that of the male MDD group (Table 2 ).The detailed results of pairwise comparisons of other groups are shown in the Supplementar y Mater ial .

Correla tion Anal ysis
In female patients, a positiv e corr elation was found between the SA of the right superior temporal ( r = 0.438, P = 0.008, Fig. 2 A), middle temporal ( r = 0.340, P = 0.043, Fig. 2 B), and lateral occipital gyrus ( r = 0.372, P = 0.025, Fig. 2 C) and illness dur ation.Ther e were no significant correlations between the SA and HAMD score or illness duration in male patients with MDD.

Discussion
Our study's findings indicate sex differences in the cortical structure of male and female patients with MDD.This may have implications for the clinical c har acteristics of depression in males and females.Specifically, we found thinner CT of the right precentral gyrus in males compared to females with MDD, although this difference did not survive Bonferroni's correction.Additionally, the SA of multiple brain regions was larger in males than in females with MDD, particularly in the fr ontotempor al and parietal cortices .T he alter ations of SA ar e mor e significant than CT, possibly due to the genetic and phenotypic independence of these measures.Our findings indicate that sex differences in CT and SA might be associated with the neurobiological processes underlying the different clinical characteristics of male and female patients with MDD.

Sex Differences in Surface Area
We found significant differences in SA between male and female groups of MDD.These regions primarily involve the default mode network (DMN), frontoparietal network (right inferior frontal gyrus triangle, superior frontal, medial orbitofrontal gyrus, superior temporal, middle temporal gyrus, and inferior parietal lobule), and visual network (right lateral occipital gyrus), which are thought to be essential in the disease process of depression (Yao et al., 2014 ).Decreased DMN connection is related to social dysfunction in people with se v er e depr ession (Vanderhasselt et al., 2013 ).The right v entr olater al pr efr ontal cortex (VLPFC), the superior temporal and middle temporal gyrus are important regions of the DMN, and play a role in rumination, self-referential processing, and emotional a ppr aisal (Song et al., 2022 ;Yang et al., 2021 ;Zhou et al., 2020 ).It was found that GMV and SA decreased in the right VLPFC in patients with MDD, and the reduced GMV of the right VLPFC related to negative emotions such as sorrow, anxiety, and fatigue (Lener et al., 2016 ).Dysfunction of the temporal lobe cannot suppress the generation of negative emotions by the prefrontal limbic system, whic h ultimatel y manifests as persistent negative emotional experiences (Yu et al., 2017 ).In self-r elated negativ e e v ents, patients with depression exhibit lo w er brain activity in the medial temporal lobe subsystem and incr eased br ain activity in the dorsal medial pr efr ontal cortex subsystem of the DMN (Wang et al., 2022 ).In MDD, females pay greater attention to suicide, while males have a higher risk of successful suicide as they are more likely to succeed when committing killing themselves (Cavanagh et al., 2017 ).Cao et al. found that suicide attempters with MDD hav e alter ations in intr a-and inter-netw ork connectivity betw een the DMN and right frontal-parietal network (Cao et al., 2020 ).Depressed patients with suicidality had abnormal cortical morphology in some brain regions within the DMN, frontolimbic circuitry, and tempor al r egions (Li et al., 2021a ;Li et al., 2021b ).When compared to suicide non-attempters, suicide attempters with MDD had smaller SA in the left superior frontal gyrus and larger SA in the left lateral occipital gyrus (Kang et al., 2020 ).
The right dorsolateral prefrontal cortex (DLPFC), and right inferior parietal lobule ar e ac knowledged as important frontoparietal network-r elated ar eas and ar e link ed to cogniti v e contr ol, attention, and decision-making processes (Yao et al., 2014 ).Reduced functional connectivity (FC) in MDD has been observed between the DLPFC and the other areas of frontoparietal network (Ma et al., 2020 ;Yun and Kim, 2021 ).In MDD, antidepr essant ther a p y w as significantly associated with DLPFC abnormalities (Nestor et al., 2022 ;Zhukovsky et al., 2021 ).As networks implicated in the cognitive regulation of emotion, the frontoparietal network and subcortical r egions, including the bilater al fusiform gyrus, ar e r elated to the state-dependent reconstruction of emotion regulation networks in individuals with MDD due to antidepressant treatment (Zhao et al., 2021 ).Males with MDD sho w ed enhanced neural responses to acute psychosocial stress in the DLPFC and right frontoparietal network compared with females (Dong et al., 2022 ).The ENIGMA MDD W orking Group' s findings revealed that patients with more se v er e insomnia in MDD had smaller SA in part areas of the frontoparietal lobe and that only SA was predictive of the severity of insomnia in MDD, whereas GMV and CT had no pr edictiv e v alue (Leerssen et al., 2020 ).As evidenced by an increasing number of Figur e 1: T he main effect of sex differ ence on the SA of gr ay matter among four gr oups (L, left hemispher e; R, right hemispher e).The numbers on the color bars indicate −log P values .T he warm color (red) indicates decreased SA in the female group (female MDD patients and female HCs) compared with the male group (male MDD patients and male HCs).
Ta ble 2: P ost hoc test results showing the differences in SA in right hemisphere between the male and female MDD groups (mm 2 , x ± s).x , y , and z are the coordinates of the primary peak locations in the MNI space; M, male MDD group; F, female MDD gr oup.Bonferr oni corr ection; P < 0.05 is consider ed statistically significant.

Region
r esearc h, MDD-r elated suicide attempts are linked to the reduced SA and FC of the inferior parietal cortex (Campos et al., 2021 ;Chen et al., 2015 ;Li et al., 2022 ).The sex differences in SA we observed in the DMN and frontal-parietal network in individuals with MDD may associate with the different clinical symptoms (such as insomnia) and suicide risk between male and female patients with MDD.
The right lateral occipital gyrus is a region of the visual network, which is important in facial perception, expression, and emotional processing (Lee et al., 2021 ;Moreno-Ortega et al., 2019 ).
Incr eased BOLD r esponses to sad stim uli in visual cortices may indicate effecti ve antide pressant treatment (Furey et al., 2013 ;Keedwell et al., 2010 ;Moreno-Ortega et al., 2019 ).A recent article about the sex-specific SA and CT c har acteristics in MDD discov er ed that the alterations in the SA of the prefrontal cortex and the local gyrification index of the visual cortex were reversed in male and female MDD when compared to gender-matched HCs (Hu et al., 2022 ).In the visual networks, the occipital gyrus's diminished functional connectivity was connected to impaired visual processing in MDD (Lu et al., 2020 ).It has been discov er ed that the abnormal FC in the visual network was associated with the clinical symptoms of MDD (Wu et al., 2023 ).The alteration of visual cortical excitability was connected to the psychopathological characteristics of MDD (Du et al., 2022 ).

Correlations with Illness Duration and HAMD Score
In female individuals with MDD, we observed a positive correlation between illness duration and the SA of the right superior tempor al, middle tempor al, and later al occipital gyrus .T he duration of depression is related to the degr ee of cognitiv e impairment (Pabel et al., 2018 ).MDD has been associated with cortical transcriptomic changes and there are sex differences in the effects of cortical gene expression on brain morphology (Miles et al., 2021 ).This study suggests that transcriptome-based polygenic score was associated with smaller amygdala volume and lo w er pr efr ontal gyrification across sexes in female patients, and related to hypergyrification in temporal and occipital areas in male patients.Another stud y re ported that the decreased GMV in the right temporal gyrus was correlated with longer illness duration, and long-term sick patients had reduced GMV in the right superior and middle temporal gyrus compared to the early course cohort (Jiang et al., 2023 ).The temporal lobe is crucial for cognitive processing, including language, memory, and object vision processing (Davey et al., 2016 ).It has been reported that the thinner occipital cortex may be an endophenotype for MDD, and the response time to emotionally disturbing tasks in adolescents with depression was negativ el y corr elated with the activ ation of the lateral occipital cortex (Colich et al., 2017 ).Hu et al .found that the higher local gyrification index in the left visual cortex was correlated with higher HAMD score in female patients with MDD, while this correlation was not observed in males with MDD (Hu et al., 2022 ).They also conducted a corr elation anal ysis between the structur al c haracteristics and the subscales of the HAMD scale, founding that higher local gyrification index in the left visual cortex was correlated with higher somatization score in female patients with MDD.Considering the limited sample size, we did not analyze the correlation between cortical structure with sex differences and the HAMD subscor e gr oup.These findings highlight the pr esence of sex-specific brain structural changes in MDD and provide neur oanatomical mec hanisms for gender differ ences in clinical manifestations in MDD patients.Zac k ov á et al. demonstr ated that the reduced volume of the superior temporal gyrus may be related to communication deficits and infrequent participation in socially stimulating activities in MDD (Zacková et al., 2021 ).

Limitations
Despite the strengths of our study, several limitations should be noted.First, the sample size was r elativ el y small, whic h limits the further investigation of the relationship between sex-specific cortical changes and specific clinical manifestation.Second, since the incidence rate for women is higher than that of men, the number of included female patients was greater than male patients in our present study, and the imbalance of sample size between men and women may affect the statistical results.Third, the longer illness duration in the male MDD group compared to the female MDD group may have confounded our results, as illness duration has been shown to be associated with brain alterations in MDD.Finall y, the study's cr oss-sectional design pr ecluded us fr om examining changes in brain structure o ver time , highlighting the need for longitudinal studies with larger sample sizes.

Conclusion
Ov er all, our study r e v ealed sex differ ences in SA in patients with MDD.These findings are instrumental in exploring the sex-specific neur oanatomical mec hanisms of clinical manifestations in patients with MDD.

Figure 2 :
Figure 2: Correlation analysis in female MDD patients.(A) Correlation between the SA of the right superior temporal gyrus and illness duration.(B) Correlation between the SA of the right middle temporal gyrus and illness duration.(C) Correlation between the SA of the right lateral occipital gyrus and illness duration.
bThe P values were obtained by Mann-Whitney U -test.