Abstract
Purpose
Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a debilitating disease characterized by persistent fatigue and decreased daily activity following physical and/or cognitive exertion. While ME/CFS affects both sexes, there is a higher prevalence in women. However, studies evaluating this sex-related bias are limited.
Methods
Circulating steroid hormones, including mineralocorticoids (aldosterone), glucocorticoids (cortisol, corticosterone, 11-deoxycortisol, cortisone), androgens (androstenedione, testosterone), and progestins (progesterone, 17α-hydroxyprogesterone), were measured in plasma samples using ultra-high performance liquid chromatography–tandem mass spectrometry (UHPLC–MS/MS). Samples were obtained from mild/moderate (ME/CFSmm; females, n=20; males, n=8), severely affected patients (ME/CFSsa; females, n=24; males, n=6), and healthy controls (HC, females, n=12; males, n=17).
Results
After correction for multiple testing, we observed that circulating levels of 11-deoxycortisol, 17α-hydroxyprogesterone in females, and progesterone in males were significantly different between HC, ME/CFSmm, and ME/CFSsa. Comparing two independent groups, we found that female ME/CFSsa had higher levels of 11-deoxycortisol (vs. HC and ME/CFSmm) and 17α-hydroxyprogesterone (vs. HC). In addition, female ME/CFSmm showed a significant increase in progesterone levels compared to HC. In contrast, our study found that male ME/CFSmm had lower circulating levels of cortisol and corticosterone, while progesterone levels were elevated compared to HC. In addition to these univariate analyses, our correlational and multivariate approaches identified differential associations between our study groups. Also, using two-component partial least squares discriminant analysis (PLS-DA), we were able to discriminate ME/CFS from HC with an accuracy of 0.712 and 0.846 for females and males, respectively.
Conclusion
Our findings suggest the potential value of including steroid hormones in future studies aimed at improving stratification in ME/CFS. Additionally, our results provide new perspectives to explore the clinical relevance of these differences within specific patient subgroups.
Graphical abstract
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a debilitating multisystem disease that primarily manifests as the inability to perform or participate in routine activities that were possible before the onset of the disease [1]. The decrease in activity level is accompanied by core symptoms, such as unrefreshing sleep and fatigue at rest, which persists for more than six months. Post-exertional malaise (PEM), defined as the worsening of symptoms after minimal physical or mental effort, is widely accepted as the hallmark feature for diagnosing ME/CFS [1, 2]. The clinical manifestation of ME/CFS varies from a mild form, in which patients are still able to participate in social life (e.g., work, school), to moderate and severe forms, characterized by individuals being primarily homebound and bedridden. The most extreme manifestation of the disease occurs in people who are completely dependent on assistance for basic daily needs, such as eating and repositioning in bed [1, 3]. As a result, ME/CFS not only negatively impacts the quality of life of individuals and caregivers but also places a substantial and often silent burden on the healthcare system. This is due to affected individuals often remaining undiagnosed due to a lack of standardized clinical assessment and diagnostic markers [2]. Although the etiology remains elusive, up to 75% of ME/CFS cases report an infectious episode preceding the onset of the disease [1, 2]. Current evidence strongly suggests that ME/CFS is associated with immune, metabolic, and vascular abnormalities [4]. However, despite the higher prevalence in women than in men [1], endocrinological studies focusing on the steroid hormones that may underlie this sex bias are scarce.
Steroid hormones are synthesized from cholesterol primarily in the adrenal gland, gonads, and placenta, playing crucial regulatory roles in various physiological and pathological processes [5,6,7]. The adrenal gland produces mineralocorticoids to regulate blood volume and pressure, and glucocorticoids to modulate metabolism and the immune system [8, 9]. The gonads synthesize sex steroid hormones in both females and males. These hormones not only modulate sexual function but also immune responses, which can influence sex-specific susceptibility to infectious and autoimmune diseases [10]. Previous studies have reported differences in steroid hormones levels between ME/CFS patients and healthy controls (HC) using various experimental approaches [11,12,13,14,15,16]. However, there has been a lack of research aimed at analyzing potential differential patterns based on sex and disease severity. Therefore, our experimental design focused on measuring the circulating levels of nine steroid hormones using highly specific quantitative measurements by ultra-high performance liquid chromatography coupled to triple quadrupole tandem mass spectrometry (UHPLC-MS/MS). These nine steroid hormones are crucial in biosynthetic and metabolic pathways related to steroidogenesis [5]. The set of steroid hormones was classified into four categories: mineralocorticoids (aldosterone), glucocorticoids (cortisol, corticosterone, 11-deoxycortisol, cortisone), androgens (androstenedione, testosterone), and progestins (progesterone, 17α-hydroxyprogesterone). Plasma samples from female and male ME/CFS patients with mild/moderate (ME/CFSmm) and severe (ME/CFSsa) symptoms were compared to their respective healthy controls (HC).
Materials and methods
Plasma samples and clinical data
This retrospective study utilized plasma samples collected before 10am from both female and male individuals provided by the UK ME/CFS Biobank (UKMEB) [17, 18]. The female cohort comprised 56 individuals, categorized into HC (n = 12; 21%), ME/CFSmm (n = 20; 36%), and ME/CFSsa patients (n = 24; 43%). The male cohort included 31 individuals, distributed into HC (n = 17; 55%), ME/CFSmm (n = 8; 26%), and ME/CFSsa patients (n = 6; 19%) (Table 1).
Individuals diagnosed with ME/CFS were evaluated by a clinician based on the Canadian Consensus [19] and/or CDC-1994 (‘Fukuda’) [20] criteria. The diagnosis was validated through responses on the Symptoms Assessment form, ensuring compliance with the case definition and study eligibility. Participants in the study completed a set of questionnaires aimed at evaluating disability levels, including the Fatigue Severity Scale, measuring the severity of fatigue symptoms to provide insights into their impact. Additionally, the Pain and Fatigue Analog Scale assessed the subjective experience of pain and fatigue on a visual analog scale, allowing participants to express the intensity of their symptoms. The Medical Outcomes Survey Short Form (SF-36v2) was employed as a comprehensive multidimensional instrument to evaluate various aspects of participants' physical and mental well-being, offering a holistic view of health-related quality of life. These instruments collectively contributed to a thorough assessment of participants' health status and functional components [21, 22].
Exclusion criteria for participants included individuals who, within the preceding three months, (1) had used drugs known to modify immune function (e.g., azathioprine, cyclosporine, methotrexate, steroids) or had taken antiviral medications; (2) had received any vaccinations; (3) had a history of acute or chronic infectious diseases such as hepatitis B and C, tuberculosis, or HIV (excluding infections by herpes virus or other retroviruses); (4) had another severe medical condition such as cancer, coronary heart disease, or uncontrolled diabetes; (5) had a severe mood disorder; (6) had been pregnant or breastfeeding in the prior 12 months; or (7) presented with morbid obesity (BMI ≥ 40). Home visits were conducted to recruit patients with mobility restrictions (severely affected), while healthy subjects and mild/moderate patients were invited to a recruiting center for clinical assessment and blood sampling [21, 22].
Standards and solvents
Certified reference standards of corticosterone solution (C-117-1ML; CAS: 50-22-6), cortisol solution (C-106-1ML; CAS: 50-23-7), cortisone solution (C-130-1ML; CAS: 53-06-5), and 11-deoxycortisol solution (D-061-1ML; CAS: 152-58-9) were purchased from Merck (KgaA, Darmstadt, Germany). Stable isotope-labeled standards, namely aldosterone-D7 (CAS: 1261254-31-2), androstene-3,17-dione-13C3 (CAS: 327048-86-2), corticosterone-D4 (CAS: 2243253-91-8), cortisol-D4 (CAS: 73565-87-4), cortisone-13C3 (CAS: 2350278-95-2), 11-deoxycortisol-D5 (CAS: 1258063-56-7), 17α-hydroxyprogesterone-D8 (CAS: 850023-80-2), progesterone-D9 (CAS: 15775-74-3), and testosterone-13C3 (CAS: 327048-83-9) were also obtained from Merck (KgaA, Darmstadt, Germany). Purified water was prepared in-house using a Milli-Q water system from Millipore (Bedford, MA, USA). HPLC-grade methanol (MeOH; 34885-1L-M), ethyl acetate (EtOAc; 1.00868), methyl-tert-butyl ether (MTBE; 34875-1L-M), LC–MS-grade acetonitrile (ACN; 1.00029. 1000), 2-propanol (IPA; 1.02781.1000), formic acid (FA; 1002531000), and ammonium fluoride (338869-25G) were purchased from Merck (KGaA, Darmstadt, Germany).
Preparation of standard solutions
To prepare the stock solution of the internal standard (IS), the stable isotope-labeled compounds were combined in ACN to achieve final concentrations of 2000 ng/mL for aldosterone-D7, cortisone-13C3 and cortisol-D4, 400 ng/mL for corticosterone-D4 and 11-deoxycortisol-D5, 200 ng/mL androstene-3,17-dione-13C3, testosterone-13C3, 17α -hydroxyprogesterone-D8, and 100 ng/mL for progesterone-D9. The working solution (WSL) was obtained by diluting the stock 1:10 in MeOH. The standard stock solution mixture in MeOH with a concentration of 1000 ng/mL was prepared using the nine individual steroid hormones. A 500 μL aliquot of the stock solution mixture was transferred to a 5 mL volumetric flask and brought up to the highest calibration point with a 50:50 (v:v) mixture of MeOH and H2O. Nineteen additional dilutions were made using a 50:50 (v:v) MeOH:H2O mixture, and the calibration curve samples were prepared by adding 10 μL of the IS WSL solution to 100 μL of the mixed standard solution to cover a calibration range from 0.00019 ng/mL to 100 ng/mL. Stock solutions were stored at –20°C and allowed to reach room temperature before use.
Sample preparation
Supported liquid extraction (SLE) was employed to mitigate the impact of the sample matrix. The SLEs were purchased from Agilent Technologies (5610-2005). Briefly, 100 µL of plasma were combined with 100 µL of H2O and 10 µL of IS WSL. The resultant mixture was transferred to the SLE tube, placed onto the sorbent bed with gentle pressure (2–3 psi), and allowed to equilibrate for 5 min. Subsequently, 400 µL of a 1:1 mixture of methyl-tert-butyl ether (MTBE) and ethyl acetate (EtOAc) were introduced into each tube, followed by elution at a rate of 1 drop per second (2 psi). This elution process was iterated three times, and a final application of 6 psi was employed to desiccate the sorbent. The entire eluent was dried with nitrogen (N2) flow at 40°C in a TurboVap water bath (Biotage, Sweden) and then reconstituted with 100 µL of MeOH. Samples were stored at 4 °C overnight, subjected to centrifugation at 2500×g for 5 min, and subsequently loaded into the autosampler.
UHPLC-MS/MS measurements and conditions
All separations were performed on a 1290 Infinity UHPLC system (Agilent Technologies) equipped with an Agilent ZORBAX RRHD Eclipse Plus C18 column (2.1 × 100 mm; 1.8 µm; 821725-902) and a ZORBAX RRHD C18 guard column (2.1 × 5 mm; 1.8 µm; 821725-901). The mobile phases comprised 0.2 mM ammonium fluoride in H2O as mobile phase A and 0.2 mM ammonium fluoride in MeOH as mobile phase B. The gradient conditions were as follows: 0–3.0 min; 50–60% B; 3.0–7.0 min, 60–86% B; 7.0–7.1 min; 86-100% B, followed by a return to the initial conditions. The total chromatographic run time was 8.5 min. The flow rate was set to 0.4 mL/min, and the column temperature was maintained at 40 °C. The injection volume was 3 µL, and a needle wash with 1:1:1:1 ACN/MeOH/IPA/H2O with 0.2% FA was utilized. Mass detection was carried out in dynamic multiple reaction monitoring (dMRM) mode on an Agilent 6460 triple quadrupole system using positive electrospray ionization (ESI) mode. Specific settings can be found in Tables S1 and S2, while Figure S1 provides a representative UHPLC-dMRM chromatogram.
Statistical analysis
For comparing the means of clinical variables among the three groups (HC, ME/CFSmm, and ME/CFSsa), one-way ANOVA was employed when the respective data followed a normal distribution. In cases where normal distribution was not met, the non-parametric Kruskal–Wallis test was used. The significance level for these tests was set at 5%. Central tendency and variability in the dataset were estimated using the median and interquartile range (IQR).
To assess differences in the circulating levels of steroid hormones in female and male ME/CFS patients compared to HC, ANOVA, and Kruskal–Wallis tests were utilized to determine statistical significance. Unadjusted p values were computed for each test. To address multiple testing, we controlled the false discovery rate (FDR) at a 5% level using the Benjamini–Hochberg (BH) and Benjamini–Yekutieli (BY) procedures.
We also conducted a predictive analysis to assess the disease status (HC versus ME/CFS) of each participant using the following classifiers and the nine steroid hormones as respective predictors: (i) linear discriminant analysis (LDA), (ii) random forest (RF), and (iii) partial least square discriminant analysis (PLS-DA). This analysis was carried out separately for male and female datasets. In PLS-DA, the analysis was performed with one and two latent components. Using a higher number of components in PLS-DA, while potentially enhancing predictive performance, was prone to overfitting the data. For LDA and PLS-DA, the probability of an individual being an ME/CFS patient was estimated based on a leave-one-out procedure. For RF, the same probability was estimated using Bootstrap with 10,000 simulated decision trees.
After obtaining the classification probability for each participant, we constructed a Receiver Operating Characteristic (ROC) curve, where 1-specificity (x-axis) was plotted against sensitivity (y-axis). The corresponding area under the curve (AUC) and the point on the curve closest to the pair (0,1) (representing perfect classification) were determined. This point served as the optimal cutoff to predict the health status of each individual. Finally, we utilized the predicted health status to estimate accuracy (proportion of individuals with correctly predicted health status), sensitivity (Se, proportion of cases with correctly predicted health status), and specificity (Sp, proportion of controls with correctly predicted health status) associated with each classifier. This analysis was carried out in the R software using the following packages: MASS (for LDA) [23], randomForest (for RF) [24], caret (for PLS-DA) [25], pROC (for ROC curve and AUC calculation) [26], OptimalCutpoints (for optimal cutoff estimation, accuracy, sensitivity, and specificity) [27].
Analyses utilizing Spearman's correlation coefficient (Rsp) were conducted to examine the statistical relationship between data from two steroid hormones. Each steroid hormone was ranked from lowest to highest, and the correlation coefficient along with corresponding p values were computed using GraphPad Prism. Next, to compare Spearman's correlation matrices between HC and individuals with ME/CFS (or its subgroups), with or without the inclusion of data on aldosterone (which contains missing data), a permutation variant of Jennrich's test was implemented. This test was originally designed for comparing two Pearson's correlation matrices [28]. The computational procedure involved the following steps: (i) fixing the values of the group variable; (ii) permutating the original data set; (iii) computing the Jennrich's test statistic in the permutated data set; (iv) repeating steps (ii) and (iii) until obtaining 1000 values of the test statistic; (v) estimating the p value by determining the proportion of times the observed test statistic in the original dataset was higher than the values of the test statistic based on the permutated datasets.
Principal component analysis (PCA) of the steroid hormones data was employed to evaluate the similarity of study participants concerning sex and disease severity. In this analysis, we utilized the first and second principal components, as these components explained over 90% of the data variability. Heatmaps were generated for the entire dataset and separately for the female and male cohorts using the MetaboAnalyst R package [29] to conduct a thorough analysis of the data. Clustering of the data was performed using Euclidean distance, aiming to unveil underlying patterns and relationships. To identify distinct subgroups, the resulting clusters were further segmented into multiple groups. Data visualization was executed using the pheatmap R package [30].
Results
Baseline demographics
Table 1 provides an overview of the demographic and clinical characteristics of the female and male participants. Noteworthy clinical differences were primarily identified in waist circumference for the female cohort and body fat for the male cohort when comparing the three independent groups. Applying the same statistical criteria, median values of blood-related tests, such as inorganic phosphate, bilirubin, creatine phosphokinase, and vitamin B12 in the female cohort, and glomerular filtration rate in the male cohort, showed differences among the three groups. Further distinctions were identified in the female cohort when comparing the aforementioned parameters between two independent groups using the non-parametric Mann–Whitney test. For instance, statistically significant differences were observed in body mass index (ME/CFSmm vs. ME/CFSsa; p = 0.0238), body muscle (HC vs. ME/CFSsa; p = 0.0249), and C-reactive protein (HC vs. ME/CFSsa; p = 0.0161). Conversely, pulse oximetry (HC vs. ME/CFSmm; p = 0.0226), albumin (HC vs. ME/CFSmm; p = 0.0135), and thyroid stimulating hormone (HC vs. ME/CFSmm; p = 0.0439) demonstrated differences in the male cohort using the same criteria outlined above. Moreover, the median scores for various functional components, as assessed by the questionnaires Fatigue Severity Scale, Pain and Fatigue Analog Scale, and SF-36, revealed significant reductions in both female and male subgroups of ME/CFS patients when compared with their respective healthy counterparts (Table 1).
Univariate analysis of steroid hormone data
The concentrations of a panel of steroid hormones, including cortisone, cortisol, corticosterone, 11-deoxycortisol, aldosterone, androstenedione, testosterone, 17α-hydroxyprogesterone, and progesterone, were measured in plasma samples by UHPLC-MS/MS (Table 2). We employed ANOVA and Kruskal–Wallis tests to assess differences in steroid hormone levels between female and male ME/CFS patients compared to HC (Fig. 1A). For each test, unadjusted, BH-adjusted, and BY-adjusted p values were calculated (Table S3). Following correction for multiple testing, only circulating levels of 11-deoxycortisol (p = 0.049) and 17α-hydroxyprogesterone (p = 0.049) in females and progesterone (p = 0.005) in males showed significant differences among our three groups (Fig. 1B–D; Table S3). Additionally, the BH procedure was utilized to determine the statistical significance of potential differences when comparing two independent groups (Table S4). In our female cohort, ME/CFSsa exhibited elevated circulating levels of 11-deoxycortisol compared to both HC (p = 0.0276) and ME/CFSmm (p = 0.0269), as well as higher 17α-hydroxyprogesterone versus HC (p = 0.0129). Furthermore, plasma progesterone levels were higher in ME/CFSmm than in HC (p = 0.0406). Using the same statistical approach in our male cohort, ME/CFSmm displayed lower circulating levels of cortisol (p = 0.0260) and corticosterone (p = 0.0109) compared to HC. Conversely, ME/CFSmm exhibited higher progesterone levels than HC (p = 0.0004). No statistical differences were found in the set of measured steroid hormones in the male cohort when comparing HC versus ME/CFSsa and ME/CFSmm versus ME/CFSsa (Table S4).
Correlation analysis of steroid hormone data
We performed non-parametric Spearman's correlation coefficient (Rsp) analysis to evaluate the degree of association between specific steroid hormones in both the female and male cohorts (Fig. 2). In the female HC group, we identified eight significant positive correlations: cortisol versus corticosterone (Rsp = 0.89, p = 0.003); androstenedione versus 17-hydroxyprogesterone (Rsp = 0.79, p = 0.003); testosterone versus androstenedione (Rsp = 0.78, p = 0.004); testosterone versus 17-hydroxyprogesterone (Rsp = 0.78, p = 0.004); cortisol versus cortisone (Rsp = 0.69, p = 0.017); androstenedione versus cortisone (Rsp = 0.66, p = 0.022); testosterone versus corticosterone (Rsp = 0.62, p = 0.037); progesterone versus 17-hydroxyprogesterone (Rsp = 0.62, p = 0.037). Comparing the female HC to ME/CFSmm, five of the eight positive associations were also found: cortisol versus corticosterone (Rsp = 0.83, p = 0.001); cortisol versus cortisone (Rsp = 0.78, p = 0.001); progesterone versus 17-hydroxyprogesterone (Rsp = 0.77, p = 0.001); testosterone versus androstenedione (Rsp = 0.67, p = 0.001); androstenedione versus cortisone (Rsp = 0.47, p = 0.036). Additionally, three positive significant correlations were observed for corticosterone versus 11-deoxycortisol (Rsp = 0.69, p = 0.001); corticosterone versus cortisone (Rsp = 0.54, p = 0.013); cortisol versus 11-deoxycortisol (Rsp = 0.52, p = 0.021); and one negative correlation for 17-hydroxyprogesterone versus aldosterone (Rsp = -0.58, p = 0.016). In the female ME/CFSsa group, only 4 positive significant correlations were identified, with two overlapping with the HC group: cortisol versus corticosterone (Rsp = 0.84, p = 0.001); progesterone versus 17-hydroxyprogesterone (Rsp = 0.83, p = 0.001) and two overlapping with the ME/CFSmm group: cortisol versus 11-deoxycortisol (Rsp = 0.84, p = 0.001); corticosterone versus 11-deoxycortisol (Rsp = 0.67, p = 0.001) (Fig. 2A).
On the contrary, we identified six significantly positive significant correlations in our male HC group: progesterone versus 17-hydroxyprogesterone (Rsp = 0.81, p = 0.001); cortisol versus corticosterone (Rsp = 0.79, p = 0.003); testosterone versus progesterone (Rsp = 0.79, p = 0.001); testosterone versus 17-hydroxyprogesterone (Rsp = 0.71, p = 0.002); corticosterone versus 11-deoxycortisol (Rsp = 0.60, p = 0.012); androstenedione versus 17-hydroxyprogesterone (Rsp = 0.57, p = 0.018). Compared to the male HC group, only one of these six positive associations was found in the male ME/CFSmm counterparts: cortisol versus corticosterone (Rsp = 0.83, p = 0.015). However, we additionally observed six significantly positive correlations for cortisol versus 11-deoxycortisol (Rsp = 0.88, p = 0.007); androstenedione versus cortisol (Rsp = 0.88, p = 0.007); corticosterone versus cortisone (Rsp = 0.81, p = 0.022); androstenedione versus 11-deoxycortisol (Rsp = 0.81, p = 0.022); androstenedione versus cortisone (Rsp = 0.79, p = 0.028); androstenedione versus corticosterone (Rsp = 0.76, p = 0.037). Finally, we identified four significantly positive correlations in male ME/CFSsa, one of which was also found in HC: corticosterone versus 11-deoxycortisol (Rsp = 0.89, p = 0.033), and two in ME/CFSmm: androstenedione versus corticosterone (Rsp = 0.94, p = 0.017); androstenedione versus 11-deoxycortisol (Rsp = 0.94, p = 0.017). Interestingly, a positive significant association was observed exclusively in male ME/CFSsa compared to either ME/CFSmm or HC, progesterone versus cortisol (Rsp = 0.94, p = 0.017) (Fig. 2B). To complement our analysis, we also calculated Rsp-based correlation matrices for the nine measured steroid hormones in each study group. Our results showed no statistical difference between different pairs of correlation matrices associated with the study groups (Figure S2; Table S5).
Multivariate analysis of steroid hormone data
We conducted principal component analysis (PCA) to identify potential clusters within our study groups (Figure S3). This analysis aimed to visualize our cohort of individuals based on steroid hormones, sex, and disease severity in a two-dimensional plot. Using the first two principal components did not reveal any distinct clusters based on severity, sex, or a combination of both (Figure S3A). However, plotting the second and third principal components allowed us to differentiate between female and male individuals, though not individuals with varying disease severity (Figure S3B). Consequently, the steroid hormone data seemed to provide limited information about the study groups.
We further employed various statistical approaches, including linear discriminant analysis (LDA), random forest (RF), and partial least squares discriminant analysis (PLS-DA), to assess the predictive and discriminative capacities of the nine steroid hormones in distinguishing between HC and ME/CFS (Figure S4; Table 3). The two-component PLD-DA demonstrated the best predictive performance for the female cohort, with an estimated AUC of 0.712 and accuracy of 0.702. Notably, this classifier performed well in both cases and controls (Sensitivity = 0.714 vs. Specificity = 0.667). Concerning the male cohort, three out of four classifiers achieved the same level of predictive accuracy (0.818). However, the two-component PLD-DA emerged as the best classifier based on AUC. Similar to the female cohort, this classifier achieved a reasonable balance between sensitivity and specificity (0.778 vs. 0.846, respectively).
To complement our findings, we conducted a clustering analysis to categorize our study cohorts based on the similarity of their circulating steroid hormone profiles. Within the female cohort, three clusters were observed primarily driven by the concentrations of 17α-hydroxyprogesterone, progesterone, and 11-deoxycortisol in ME/CFS patients, with no clear differentiation between subgroups. While two additional patterns emerged related to aldosterone and cortisone, individuals from HC, ME/CFSmm, and ME/CFSsa were clustered together (Fig. 3A). The same analysis was applied to the male dataset, and despite a prominent cluster related to testosterone, no differences were observed between our study groups (Fig. 3B).
Discussion
Despite evidence indicating a higher prevalence of ME/CFS in women compared to men (3:1 ratio) [1], there remains a significant research gap concerning potential differences in circulating levels of steroid hormones in patients stratified by both sex and disease severity. Steroid hormones are synthesized by both males and females, displaying variations in circulating levels and effects on target tissues [5,6,7,8,9]. While testosterone is predominantly produced in males, and estrogen and progesterone are primarily synthesized in females [31], these differences have historically limited research on the role of testosterone in women and estrogen and progesterone in men [32]. Notably, our study revealed elevated progesterone levels in both female and male ME/CFSmm patients compared to their healthy counterparts. While the biological effects of progesterone in men are presently not fully understood, there is evidence suggesting an immunosuppressive effect of progesterone in women [33]. Our findings indicate that basal plasma progesterone is elevated in male ME/CFSmm patients, similar to previous reports in female patients [34], proving an interesting avenue for further investigation. Importantly, progesterone has been associated with a hypercoagulable state [35]. This is particularly noteworthy considering recent reports of fibrinoid microclots in ME/CFS patients [36]. However, additional research is required to establish a potential link between progesterone and these immunologic and hemostasis-related effects in both female and male ME/CFS patients.
Sexual dimorphism linked to glucocorticoids significantly influences various physiological processes, encompassing both the immune response [37, 38] and metabolism [39], both of which undergo alterations in ME/CFS [40,41,42,43]. Cortisol, the principal glucocorticoid governing these processes under stress, is subject to modulation interactions with sex steroid hormones [39]. Interestingly, ME/CFS patients have been reported to exhibit reduced cortisol levels [15, 44], which is associated with documented dysfunction of the hypothalamic-pituitary-adrenal (HPA) axis in a substantial proportion of ME/CFS cases [45]. However, existing evidence suggests that this dysfunction occurs at a central level, likely due to corticotropin-releasing hormone (CRH) deficiency rather than a primary adrenal insufficiency [46]. Consequently, our findings of elevated levels of 11-deoxycortisol in female ME/CFSsa patients and decreased levels of both cortisol and corticosterone in male ME/CFSmm patients offer novel perspectives for further investigating the intricate interplay between glucocorticoids in different ME/CFS subtypes.
In response to stress, the adrenal glands secrete both cortisol and progesterone in women and men [47]. Stress-induced progesterone plays a role in increasing the bioavailability of cortisol by competitively binding to corticosteroid-binding globulin (CBG), thereby reducing the amount of cortisol that can bind to CBG [48]. Additionally, it participates in the negative feedback loop of the HPA axis to regulate the magnitude of the stress response [49]. Progesterone metabolites, particularly allopregnanolone, primarily mediate this inhibitory effect. Allopregnanolone serves as a potent neuroactive steroid that inhibits hypothalamic CRH production [50]. The observed alterations in glucocorticoid and progesterone levels in male ME/CFSmm patients suggest the presence of a negative feedback mechanism. Similar patterns have been proposed in women during the luteal phase of the menstrual cycle, where elevated progesterone and decreased cortisol levels are noted, contrasting with women during the follicular phase who exhibit the opposite pattern [51]. Glucocorticoids play a central role in a broad range of normal and stress-related responses [52]. Reduced glucocorticoid production, relative to progesterone and allopregnanolone levels, may help to elucidate the characteristic dysregulation of physiological functions in ME/CFS, such as diminished physical performance leading to fatigue, pain, and cognitive impairment [45]. Nevertheless, it remains unknown whether this alteration causes the symptoms or if it represents a compensatory mechanism observed in a subset of ME/CFS patients.
The observed elevation in the progesterone/glucocorticoid ratio in ME/CFSmm patients suggests a potential dysregulation in adrenal corticosteroid pathways, considering progesterone's role as a precursor in corticosteroid biosynthesis [5,6,7]. Our findings point towards the adrenal enzyme P450c11β (11β-hydroxylase; CYP11B1 gene) as a potential site of dysregulation. This enzyme is responsible for converting 11-deoxycorticosterone to corticosterone and 11-deoxycortisol to cortisol [5,6,7]. A deficiency in P450c11β activity could contribute to the observed reduction in corticosterone and cortisol levels in male ME/CFSmm patients, as well as the accumulation of 11-deoxycortisol and 17α-hydroxyprogesterone in female ME/CFSsa patients. However, it is important to note that the levels of steroid hormones and other parameters identified in the blood of patients in our cohort do not meet the criteria for classical P450c11β deficiency [52]. Moreover, our findings may be influenced by sex bias due to the presence of extra-adrenal sources of progesterone and 11-deoxycorticosterone in women, specifically the ovaries [53]. While men also produce progesterone from their testicles, it occurs at lower levels [32]. This complexity suggests that the interaction between steroid hormones is more intricate than previously assumed, and consequently, no single hormone may serve as an adequate biomarker. Consistent with these considerations, we achieved a discriminative accuracy of 0.712 and 0.846 for females and males, respectively, in distinguishing between ME/CFS and healthy controls using a two-component PLS-DA that incorporated our set of nine steroid hormones. Further studies should prioritize assessing the potential predictive value of these and other steroid hormones in stratifying ME/CFS patients based on sex and disease severity.
While we observed differences in steroid hormone levels within our ME/CFS patient cohort, stratified by both sex and severity, it is imperative to confirm our findings with a larger sample size. Although our study excluded pregnant women and those within 12 months postpartum or lactating, we recommend incorporating clinical data related to menstrual cycle, menopausal status, pregnancy history, and contraceptive use. This inclusion will enhance our understanding of the impact of ME/CFS on female reproductive health. In agreement with Pollack et al. [54], we certainly consider that this information proves value as it enables the reduction of potential selection bias and data heterogeneity, thereby positively influencing data analysis and interpretability in female ME/CFS patients. Addressing these pertinent aspects will significantly advance the comprehension of ME/CFS pathophysiology and may contribute to the development of personalized therapeutic approaches tailored to specific patient subgroups. Our findings underscore the importance of investigating variations in pathological mechanisms related to severity, with particular attention to severely affected individuals. This subgroup is often underrepresented in ME/CFS research due to limited healthcare access.
Data availability
The dataset used in this study is available upon reasonable request to the corresponding author.
References
Bateman L, Bested AC, Bonilla HF et al (2021) Myalgic encephalomyelitis/chronic fatigue syndrome: essentials of diagnosis and management. Mayo Clin Proc 96:2861–2878. https://doi.org/10.1016/j.mayocp.2021.07.004
Choutka J, Jansari V, Hornig M, Iwasaki A (2022) Unexplained post-acute infection syndromes. Nat Med 28:911–923. https://doi.org/10.1038/s41591-022-01810-6
Dafoe W (2021) Extremely severe ME/CFS—a personal account. Healthcare 9:504. https://doi.org/10.3390/healthcare9050504
Fluge Ø, Tronstad KJ, Mella O (2021) Pathomechanisms and possible interventions in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). J Clin Investig. https://doi.org/10.1172/JCI150377
Miller WL (2017) Steroidogenesis: unanswered questions. Trends Endocrinol Metab 28:771–793. https://doi.org/10.1016/j.tem.2017.09.002
Miller WL, Auchus RJ (2019) The “backdoor pathway” of androgen synthesis in human male sexual development. PLoS Biol 17:e3000198. https://doi.org/10.1371/journal.pbio.3000198
Miller WL, Auchus RJ (2011) The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders. Endocr Rev 32:81–151. https://doi.org/10.1210/er.2010-0013
Hahner S, Ross RJ, Arlt W et al (2021) Adrenal insufficiency. Nat Rev Dis Primers 7:19. https://doi.org/10.1038/s41572-021-00252-7
Chakraborty S, Pramanik J, Mahata B (2021) Revisiting steroidogenesis and its role in immune regulation with the advanced tools and technologies. Genes Immun 22:125–140. https://doi.org/10.1038/s41435-021-00139-3
Klein SL, Flanagan KL (2016) Sex differences in immune responses. Nat Rev Immunol 16:626–638
Miwa K (2017) Down-regulation of renin–aldosterone and antidiuretic hormone systems in patients with myalgic encephalomyelitis/chronic fatigue syndrome. J Cardiol 69:684–688. https://doi.org/10.1016/j.jjcc.2016.06.003
Boneva RS, Decker MJ, Maloney EM et al (2007) Higher heart rate and reduced heart rate variability persist during sleep in chronic fatigue syndrome: a population-based study. Auton Neurosci 137:94–101. https://doi.org/10.1016/j.autneu.2007.08.002
Miwa K, Fujita M (2014) Renin–aldosterone paradox in patients with myalgic encephalomyelitis and orthostatic intolerance. Int J Cardiol. https://doi.org/10.1016/j.ijcard.2014.01.043
Germain A, Barupal DK, Levine SM, Hanson MR (2020) Comprehensive circulatory metabolomics in ME/CFS reveals disrupted metabolism of acyl lipids and steroids. Metabolites 10:34. https://doi.org/10.3390/metabo10010034
Chang C-J, Hung L-Y, Kogelnik AM et al (2021) A comprehensive examination of severely ill ME/CFS patients. Healthcare 9:1290. https://doi.org/10.3390/healthcare9101290
Germain A, Ruppert D, Levine S, Hanson M (2018) Prospective biomarkers from plasma metabolomics of myalgic encephalomyelitis/chronic fatigue syndrome implicate redox imbalance in disease symptomatology. Metabolites 8:90. https://doi.org/10.3390/metabo8040090
Blauensteiner J, Bertinat R, León LE et al (2021) Altered endothelial dysfunction-related miRs in plasma from ME/CFS patients. Sci Rep 11:10604. https://doi.org/10.1038/s41598-021-89834-9
Bertinat R, Villalobos-Labra R, Hofmann L et al (2022) Decreased NO production in endothelial cells exposed to plasma from ME/CFS patients. Vascul Pharmacol 143:106953. https://doi.org/10.1016/j.vph.2022.106953
Carruthers BM, Jain AK, De Meirleir KL et al (2003) Myalgic encephalomyelitis/chronic fatigue syndrome. J Chronic Fatigue Syndr 11:7–115. https://doi.org/10.1300/J092v11n01_02
Fukuda K, Straus SE, Hickie I et al (1994) The chronic fatigue syndrome: a comprehensive approach to its definition and study. Ann Intern Med 121:953–959. https://doi.org/10.7326/0003-4819-121-12-199412150-00009
Lacerda EM, Mudie K, Kingdon CC et al (2018) The UK ME/CFS Biobank: a disease-specific biobank for advancing clinical research into myalgic encephalomyelitis/chronic fatigue syndrome. Front Neurol 9:1026. https://doi.org/10.3389/fneur.2018.01026
Lacerda EM, Bowman EW, Cliff JM et al (2017) The UK ME/CFS Biobank for biomedical research on myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and multiple sclerosis. Open J Bioresour. https://doi.org/10.5334/ojb.28
Venables WN, Ripley BD (2002) Modern applied statistics with S. Springer, New York
Liaw A, Wiener M (2002) Classification and regression by randomForest. R News 2:18–22
Kuhn M (2008) Building predictive models in R using the caret package. J Stat Softw. https://doi.org/10.18637/jss.v028.i05
Robin X, Turck N, Hainard A et al (2011) pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinform 12:77. https://doi.org/10.1186/1471-2105-12-77
López-Ratón M, Rodríguez-Álvarez MX, Suárez CC, Sampedro FG (2014) OptimalCutpoints: an R package for selecting optimal cutpoints in diagnostic tests. J Stat Softw. https://doi.org/10.18637/jss.v061.i08
Jennrich RI (1970) An asymptotic χ2 test for the equality of two correlation matrices. J Am Stat Assoc 65:904–912. https://doi.org/10.1080/01621459.1970.10481133
Xia J, Sinelnikov IV, Han B, Wishart DS (2015) MetaboAnalyst 3.0—making metabolomics more meaningful. Nucleic Acids Res 43:W251–W257. https://doi.org/10.1093/nar/gkv380
Kolde R, Kolde M (2015) Package “pheatmap.” R Package 1:790
Hammes SR, Levin ER (2019) Impact of estrogens in males and androgens in females. J Clin Investig 129:1818–1826. https://doi.org/10.1172/JCI125755
Oettel M, Mukhopadhyay A (2004) Progesterone: the forgotten hormone in men? Aging Male 7:236–257. https://doi.org/10.1080/13685530400004199
Kolatorova L, Vitku J, Suchopar J et al (2022) Progesterone: a steroid with wide range of effects in physiology as well as human medicine. Int J Mol Sci 23:7989. https://doi.org/10.3390/ijms23147989
Murphy B (2004) Elevated levels of some neuroactive progesterone metabolites, particularly isopregnanolone, in women with chronic fatigue syndrome. Psychoneuroendocrinology 29:245–268. https://doi.org/10.1016/S0306-4530(03)00026-X
Swanepoel AC, Visagie A, de Lange Z et al (2016) The clinical relevance of altered fibrinogen packaging in the presence of 17β-estradiol and progesterone. Thromb Res 146:23–34. https://doi.org/10.1016/j.thromres.2016.08.022
Nunes J, Kruger A, Proal A et al (2022) The occurrence of hyperactivated platelets and fibrinaloid microclots in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). Pharmaceuticals 15:931. https://doi.org/10.3390/ph15080931
Cain DW, Cidlowski JA (2017) Immune regulation by glucocorticoids. Nat Rev Immunol 17:233–247. https://doi.org/10.1038/nri.2017.1
Taves MD, Ashwell JD (2021) Glucocorticoids in T cell development, differentiation and function. Nat Rev Immunol 21:233–243. https://doi.org/10.1038/s41577-020-00464-0
Kroon J, Pereira AM, Meijer OC (2020) Glucocorticoid sexual dimorphism in metabolism: dissecting the role of sex hormones. Trends Endocrinol Metab 31:357–367. https://doi.org/10.1016/j.tem.2020.01.010
Hoel F, Hoel A, Pettersen IKN et al (2021) A map of metabolic phenotypes in patients with myalgic encephalomyelitis/chronic fatigue syndrome. JCI Insight. https://doi.org/10.1172/jci.insight.149217
Naviaux RK, Naviaux JC, Li K et al (2016) Metabolic features of chronic fatigue syndrome. Proc Natl Acad Sci 113:E5472–E5480. https://doi.org/10.1073/pnas.1607571113
Mandarano AH, Maya J, Giloteaux L et al (2020) Myalgic encephalomyelitis/chronic fatigue syndrome patients exhibit altered T cell metabolism and cytokine associations. J Clin Investig. https://doi.org/10.1172/jci132185
Uhde M, Indart AC, Green PHR et al (2023) Suppressed immune and metabolic responses to intestinal damage-associated microbial translocation in myalgic encephalomyelitis/chronic fatigue syndrome. Brain Behav Immun Health 30:100627. https://doi.org/10.1016/j.bbih.2023.100627
de Vega WC, Herrera S, Vernon SD, McGowan PO (2017) Epigenetic modifications and glucocorticoid sensitivity in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). BMC Med Genomics 10:11. https://doi.org/10.1186/s12920-017-0248-3
Papadopoulos AS, Cleare AJ (2012) Hypothalamic–pituitary–adrenal axis dysfunction in chronic fatigue syndrome. Nat Rev Endocrinol 8:22–32. https://doi.org/10.1038/nrendo.2011.153
Demitrack MA, Dale JK, Straus SE et al (1991) Evidence for impaired activation of the hypothalamic–pituitary–adrenal axis in patients with chronic fatigue syndrome. J Clin Endocrinol Metab 73:1224–1234. https://doi.org/10.1210/jcem-73-6-1224
Herrera AY, Nielsen SE, Mather M (2016) Stress-induced increases in progesterone and cortisol in naturally cycling women. Neurobiol Stress 3:96–104. https://doi.org/10.1016/j.ynstr.2016.02.006
Cameron A, Henley D, Carrell R et al (2010) Temperature-responsive release of cortisol from its binding globulin: a protein thermocouple. J Clin Endocrinol Metab 95:4689–4695. https://doi.org/10.1210/jc.2010-0942
Klusmann H, Schulze L, Engel S et al (2022) HPA axis activity across the menstrual cycle—a systematic review and meta-analysis of longitudinal studies. Front Neuroendocrinol 66:100998. https://doi.org/10.1016/j.yfrne.2022.100998
Crowley SK, Girdler SS (2014) Neurosteroid, GABAergic and hypothalamic pituitary adrenal (HPA) axis regulation: what is the current state of knowledge in humans? Psychopharmacology 231:3619–3634. https://doi.org/10.1007/s00213-014-3572-8
Hamidovic A, Karapetyan K, Serdarevic F et al (2020) Higher circulating cortisol in the follicular vs. luteal phase of the menstrual cycle: a meta-analysis. Front Endocrinol (Lausanne). https://doi.org/10.3389/fendo.2020.00311
White PC (2023) Steroid 11β-hydroxylase deficiency and related disorders. In: Genetic steroid disorders. Elsevier, pp 63–79
Nahoul K, Dehennin L, Salat-Baroux J, Scholler R (1988) Deoxycorticosterone secretion by the human ovary. J Steroid Biochem 31:111–117. https://doi.org/10.1016/0022-4731(88)90213-0
Pollack B, von Saltza E, McCorkell L et al (2023) Female reproductive health impacts of Long COVID and associated illnesses including ME/CFS, POTS, and connective tissue disorders: a literature review. Front Rehabil Sci. https://doi.org/10.3389/fresc.2023.1122673
Acknowledgements
This study received support from ME Research UK (SCIO charity number SC036942, awarded to FW) and utilized samples from the UKMEB, which were processed and stored at the UCL-RFH BioBank. The authors express gratitude to all the study participants who donated their blood to the UKMEB, the ME/CFS community who raised funds to facilitate the UKMEB projects, and private donations received from Austria at FH JOANNEUM. FW acknowledges the support of the Österreichische Gesellschaft für ME/CFS and the FH JOANNEUM University of Applied Sciences in grant applications. NS was partially financed by national funds through FCT – Fundação para a Ciência e a Tecnologia under the project UIDB/00006/2020. DOI: 10.54499/UIDB/00006/2020 (https://doi.org/10.54499/UIDB/00006/2020). DM acknowledges funding from ANID (Beca de Doctorado Nacional PFCHA-21190736) and UDEC/MINEDUC: Proyecto UCO 1866. FW also thanks Valentina B. Cohen and María Richardson for their valuable and insightful feedback. As a research team dedicated to the study of ME/CFS, we strongly believe that conducting patient-centered research will ultimately lead to improved patient outcomes. FW extends special thanks to Elisa Lafuente for her support and encouragement throughout this study. Also, FW expresses his gratitude to Margareta Rodler and Fritz Stuparitz for their constant motivation.
Funding
Open access funding provided by FH Joanneum - University of Applied Sciences.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Disclosure of potential conflicts of interest
The authors declare no pertinent financial or non-financial interests to disclose.
Research involving human participants and/or animals
This study, conducted in adherence to the principles outlined in the Declaration of Helsinki, received approval from two independent research ethics committees: the Medical University of Graz, Austria (reference number: 31-059 ex 18/19), and the UCL Biobank Ethical Review Committee-Royal Free (B-ERC-RF) London NHS Foundation Trust (reference number: EC2018.005).
Informed consent
Informed consent was secured from all individuals participating in this study.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Figure S1.
Representative ultra-high performance liquid chromatography-dynamic multiple reaction monitoring (UHPLC-dMRM) chromatograms. A Representative UHPLC-dMRM chromatogram of the target steroid hormones and the corresponding internal standards in a plasma sample prepared using supported liquid extraction (SLE). B Representative UHPLC-dMRM chromatogram of the targeted steroids and their corresponding internal standards in a calibration-mix with a concentration of 100 ng/ml for the targeted steroid hormones. Figure S2. Spearman's correlation coefficient derived from circulating levels of steroid hormones. A matrix encompassing the circulating levels of the nine steroid hormones was employed to calculate the Spearman's correlation coefficient, stratified by sex and disease severity. Participants were categorized into four groups: HC, representing healthy controls; ME/CFSmm, covering mild/moderate cases; and ME/CFSsa, denoting severe ME/CFS patients. Statistical analysis was carried out using R software version 4.0.2. Figure S3. Principal component analysis derived from circulating levels of steroid hormones. The scatterplots illustrate the distribution of each study participant in the first/second (A) and second/third (B) principal components. Participants were classified into three groups: HC, covering healthy controls; ME/CFSmm, representing mild/moderate ME/CFS patients; and ME/CFSsa, indicating severe ME/CFS patients. Statistical analysis was conducted using R software version 4.0.2. Figure S4. Predictive analysis for disease status estimation in female and male cohorts. Predictive analysis was conducted to estimate the disease status of each participant in both the female and male cohorts. Different classifiers, including linear discriminant analysis (LDA), random forest (RF), and partial least squares discriminant analysis (PLS-DA), were employed. The set of measured steroid hormones served as respective predictors. Receiver Operating Characteristic (ROC) curves and area under the curve (AUC) were derived to predict the health status of each participant (HC vs. ME/CFS) using the nine steroid hormones as predictors in both the female (A) and male (B) cohorts. (PDF 383 KB)
Table S1.
Analytical parameters for UHPLC-dMRM of steroid hormones and internal standards. The table presents analytical parameters utilized for UHPLC-dMRM of steroid hormones and internal standards. Details on the employed dMRM transitions, collision energies (CE), fragmentor voltages (F), cell accelerator voltages (CAV), retention times (RT), delta retention times (Δ RT), and polarity modes of the target steroids hormones and corresponding internal standards are provided. Table S2. Source parameters for electrospray ionization of targeted steroid hormones and corresponding internal standards. The table provides source parameters employed for electrospray ionization in UHPLC-MS/MS analysis of steroid hormones and internal standards. It includes values for various parameters, encompassing drying gas temperature and flow, nebulizer pressure, sheath gas temperature and flow, and capillary voltage in positive ( +) mode. Table S3. Corrected p values from multiple testing derived for plasma levels of steroid hormones in female and male ME/CFS patients compared to healthy controls. The table displays p values corrected by multiple testing, derived from ANOVA and Kruskal–Wallis tests, for plasma levels of steroid hormones in female and male ME/CFS patients compared to healthy controls. For each test, unadjusted, false discovery rate (FDR)-adjusted, and Benjamini–Yekutieli (BY)-adjusted p values were calculated. Table S4. p values for comparisons of circulating levels of steroid hormones between independent groups in female and male cohorts. The table presents p values obtained by comparing two independent groups within both female and male cohorts concerning their circulating levels of steroid hormones. Participants were classified into three groups: HC, denoting healthy controls; ME/CFSmm, representing mild/moderate ME/CFS patients; and ME/CFSsa, indicating severe ME/CFS patients. The displayed p values have been adjusted using the Benjamini–Hochberg test. Table S5. Estimated p values of Jennrich's permutation test for equality of two Spearman's correlation matrices. The table displays p values obtained from Jennrich’s permutation test for the equality of two Spearman’s correlation matrices. The presented p values encompass matrices 1 and 2, with and without aldosterone data, respectively. This statistical approach was adopted to address missing information in the aldosterone data(PDF 393 KB)
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Pipper, C., Bliem, L., León, L.E. et al. Sex and disease severity-based analysis of steroid hormones in ME/CFS. J Endocrinol Invest 47, 2235–2248 (2024). https://doi.org/10.1007/s40618-024-02334-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40618-024-02334-1