Endocrine Noise: Sex-Specific Disruption of Hypothalamic–Pituitary–Adrenal (HPA) Axis by Endocrine-Disrupting Chemicals

Abstract

Environmental chemicals are rarely considered stressors in the way that psychological or physical stressors are. Yet many endocrine-disrupting chemicals (EDCs) interact with the body’s core stress response system. This review examines how EDCs alter hypothalamic–pituitary–adrenal (HPA) regulation and how biological sex influences those responses. Drawing on human epidemiological data and experimental models, we describe how EDC exposure affects cortisol dynamics, feedback sensitivity, and adrenal signaling, with a particular focus on sex-dependent outcomes. We propose the concept of endocrine noise to describe how low-dose, often mixed EDC exposures introduce persistent interference into hormone signaling without necessarily causing overt endocrine deficiency or excess. In this framework, EDCs act as chronic, low-grade stressors that reset the timing, feedback precision, and rhythmic organization of the HPA axis rather than as isolated reproductive toxicants. We argue that EDCs should be understood as chronic, context-dependent stress modifiers that reshape sex-specific “risk architectures” for affective, metabolic, and immune disorders. Recognizing sex-specific HPA architecture and endocrine noise has immediate implications for study design and regulation, including the need for sex-stratified analyses, circadian-sensitive sampling of cortisol, and risk assessments that consider how the same exposure can push female and male stress systems in divergent directions.

1. Introduction

Endocrine-disrupting chemicals (EDCs) are a diverse group of compounds present in plastics, pesticides, industrial effluents, and many consumer products, resulting in widespread human exposure. For decades, research on EDCs has focused on reproductive and thyroid systems, where their effects are well documented [1,2,3]. However, growing evidence suggests that another important target is the body’s stress response system.

The hypothalamic–pituitary–adrenal (HPA) axis regulates physiological responses to stress through tightly coordinated glucocorticoid signaling. Cortisol helps the body respond and adapt to changing internal and external demands, but its effects depend not only on hormone concentration, but also on timing, rhythmicity, feedback sensitivity, and recovery after challenge. When these regulatory features are chronically altered, the risk of stress-related dysfunction increases, including metabolic, cardiovascular, and psychiatric disorders [4].

Importantly, HPA-axis regulation is not uniform across individuals. Biological sex is a major determinant of how this system is organized and functions [5,6,7]. Females, on average, show stronger HPA responses and slower recovery times than males. These differences are shaped by sex hormones as well as receptor-level mechanisms, including weaker glucocorticoid negative feedback and sex differences in glucocorticoid and corticotropin-releasing hormone (CRH) receptor function in stress-regulatory brain regions. Such features are clinically relevant, as many stress-related disorders, including depression and autoimmune disease, occur more frequently in women [8].

The effects of EDC exposure on the HPA axis are also likely to depend on developmental timing. The stress-response system is not static across the lifespan, and periods such as prenatal development, early life, and puberty represent sensitive windows during which environmental exposures may recalibrate neuroendocrine function in lasting ways [7,9]. In this context, EDCs should not be viewed solely as disruptors of reproductive endocrinology, but more broadly as modulators of stress-system regulation across a sexually dimorphic developmental landscape.

Consistent with this framework, both human and animal studies have linked EDC exposure to altered glucocorticoid levels and disrupted HPA reactivity [10,11,12,13]. Experimental studies further suggest that these effects may differ in direction between males and females and, when exposure occurs during sensitive developmental windows, may induce lasting neuroendocrine reprogramming [14,15,16]. Yet many toxicological and epidemiological studies still pool males and females, or examine only one sex, thereby obscuring biologically meaningful differences. Without sex-disaggregated data, important patterns of vulnerability may remain undetected.

In parallel, most reviews treat “stress” as something external such as psychosocial strain, trauma, or adversity rather than as a physiological system whose regulation may itself be modified by chemical exposure. This has left environmental chemical stress insufficiently integrated into most models of stress biology y [17,18].

In this review, we address these gaps by reframing EDCs as modulators of stress-system response and regulation within a sexually dimorphic HPA architecture. We further introduce the concept of endocrine noise, defined here as persistent, low-grade interference with the fidelity of hormonal signaling and feedback. We do not use this term to describe all forms of endocrine disruption. Rather, we distinguish endocrine noise from developmental resetting: during sensitive windows, EDC exposure may recalibrate HPA-axis function and shift long-term trajectories, whereas in other contexts persistent exposure may impair the precision, timing, and coordination of endocrine signaling [6,7,9]. Such recalibration may be adaptive in the short term under environmental constraint, yet become maladaptive over time by altering responsivity, recovery, and vulnerability to later physiological or psychosocial challenge.

We therefore (i) summarize how sex shapes HPA organization before any chemical exposure; (ii) examine evidence for sex-specific HPA outcomes following EDC exposure across development; (iii) outline mechanistic intersections of stress and sex-steroid signaling; and (iv) discuss how these interactions create sex-dependent vulnerability patterns for affective, metabolic, and immune disorders. Finally, we highlight methodological and regulatory implications, including recommendations for sex- conscious study design and cortisol phenotyping.

2. Current Understanding of Sex-Specific HPA-Axis Organization Across the Lifespan Chemical Exposure

Before examining how EDCs might alter stress biology, it is necessary to establish that the hypothalamic–pituitary–adrenal (HPA) axis is not organized identically in females and males. Even in the absence of chemical exposure, sex differences are evident in developmental programming, cortisol rhythmicity, feedback sensitivity, and hormone bioavailability [19,20]. These baseline differences shape how subsequent perturbations are translated into endocrine signals and therefore provide the biological context in which EDC effects must be interpreted.

2.1. Developmental and Genetic Foundations

Sex differences in HPA-axis regulation do not begin at puberty. They begin at conception with different XX and XY biology, including dosage effects from X-linked genes and male-specific expression of Y-linked genes. Importantly, not all X-linked genes are fully silenced in females; some escape X inactivation, especially during development, creating plausible routes to sex-differentiated brain development in regions that regulate glucocorticoids [19,21,22]. These early genetic differences influence the maturation of stress-regulatory circuits, including the hypothalamus, hippocampus, and prefrontal cortex, which in turn calibrate glucocorticoid secretion and feedback sensitivity across the life course [23].

Puberty then represents a second major phase of HPA axis remodeling. Rising estrogens and androgens alter receptor expression, synaptic plasticity, and the coupling between circadian systems and adrenal output. During and after puberty, the HPA axis is recalibrated in a sex-specific manner: androgens in males generally impose an inhibitory tone on stress-induced HPA activity, whereas estrogens in females tend to support greater HPA responsiveness and less efficient glucocorticoid feedback [6,24].

Beyond single-genes effects, polygenic interactions further amplify this dimorphism. Genome-wide analyses indicate that common variants can exert sex-discordant effects on stress responsivity, metabolism, and immunity [25], while transcriptomic studies show sex-dependent blood transcriptome responses to glucocorticoid receptor activation [26]. Postnatal animal models also support this principle, showing that liver–HPA crosstalk contributes to sex differences in glucocorticoid binding and secretion [27]. Taken together, these observations indicate that adult females and males do not arrive at stress regulation from the same physiological starting point; they enter later in life with HPA systems that are already tuned differently [8,19,28].

2.2. Cortisol Dynamics: Rhythm and Bioavailability

Sex differences in HPA-axis function are often expressed more clearly in temporal patterns than in a single static cortisol value. The most informative contrasts frequently appear in daily rhythm, adaptation across repeated challenges, and the degree to which circulating hormone measurements reflect biologically active cortisol at the tissue level.

In rodents exposed to repeated restraint, females sometimes show less reduction in stress-hormone output across sessions, with a more persistently elevated HPA response than males. In a 2017 study by Grafe et al., this pattern was associated with greater hypothalamic orexin activity, and dampening orexin signaling during repeated stress largely removed the sex difference in HPA output [29]. In contrast, Goel et al. in 2022 found that males and females both showed declines in adrenocorticotropic hormone (ACTH) and corticosterone across repeated restraint, but the serotonergic adaptations underlying this change differed by sex [30].

Human studies support the same broader principle. In ambulatory datasets, men on average may show higher daytime cortisol and a flatter decline across the day, although these effects vary with waking time, sleep, age, and daily context [31]. Around awakening, the relevant phenotype is not simply peak concentration but the structure of the curve itself. The cortisol awakening response (CAR) reflects the rapid rise in cortisol during the first 30–45 min after waking and represents a regulated component of morning secretion rather than random fluctuation [32]. In the classic study of CAR norms, women showed a similar initial rise to men but a delayed early-morning decrease, indicating that post-awakening secretion can differ even when peak height appears comparable [33].

Interpretation is further complicated by reproductive state and hormone binding. Menstrual phase and oral contraceptive use can alter apparent sex contrasts depending on the sampling matrix and timing of measurement. Most circulating cortisol is protein-bound, largely to corticosteroid-binding globulin (CBG), which is estrogen-sensitive. Thus, total serum cortisol may increase without a corresponding rise in free cortisol, particularly in high-estrogen states [34,35,36]. These considerations are also relevant when translating rodent findings to human, as species differences in CBG regulation may complicate direct interpretation of free versus total glucocorticoid measures [37,38].

2.3. Feedback Sensitivity and Receptor Control

A defining feature of the HPA axis is that cortisol limits its own production through negative feedback. This feedback is mediated through two complementary receptor system: mineralocorticoid receptors (MR) and glucocorticoid receptors (GR), which are expressed in stress-regulatory brain regions (such as the hippocampus, hypothalamus, and prefrontal cortex) and in the pituitary. MR has a higher affinity for endogenous glucocorticoids and is engaged even at low basal hormone concentrations whereas GR is progressively recruited as cortisol rises during the morning peak or in response to acute stress [39,40]. Together, these receptor systems determine both the gain of HPA activation and the speed of recovery [41].

Accordingly, HPA regulation cannot be reduced to peak hormone output alone. Alterations in MR–GR balance within forebrain and limbic system influence not only the magnitude of HPA activation but also the temporal organization of stress response and the efficiency of recovery [42]. Feedback sensitivity therefore depends on receptor context—including co-regulators, intracellular chaperones, regional circuitry, and upstream corticotropin-releasing hormone drive—rather than on circulating cortisol concentration alone.

Within this framework, sex differences are especially evident in recovery dynamics. In females, GR-mediated negative feedback is often less effective, reflecting both estrogenic modulation and differences in receptor signaling within hypothalamic and pituitary regions. As a result, glucocorticoid levels may remain elevated for longer after stress exposure. In contrast, males more often show tighter feedback inhibition, a pattern supported in part by androgen signaling [19]. Emerging evidence also suggests that interactions between MR and GR contribute to these sex-specific differences, including MR-dependent regulation of the GR co-chaperone FKBP5 [8,43].

Gonadal steroids are major modifiers of this feedback regime and help explain why sex effects are often state-dependent rather than fixed. In rodent models, androgens generally dampen HPA responsivity, whereas estradiol more commonly enhances it, in part through receptor-mediated effects within stress circuitry and hypothalamic control of HPA output [44]. More specifically, estradiol can weaken GR-dependent negative feedback via estrogen receptor signaling in hypothalamic regions that regulate HPA drive [45]. Thu, some of the most informative sex differences in stress biology are expressed not as stable differences in mean cortisol levels, but as differences in recovery, habituation, and regulatory stability [6].

Glucocorticoid signaling is also temporally structured. Cortisol is secreted in ultradian pulses superimposed on circadian rhythms, and this temporal organization carries biological significance for receptor activation, transcriptional regulation, and downstream metabolic and immune function. Sex differences extend to this domain as well, with females tending to show more labile rhythmicity and males showing tighter coupling between ACTH pulses and cortisol release [46,47,48,49].

Because these temporal and feedback properties differ before any EDC exposure occurs, they provide the mechanistic substrate on which later chemical disruption is superimposed. This baseline organization is therefore essential for understanding why similar exposures may later yield sex-divergent endocrine outcomes.

3. Endocrine Noise as a Framework for Chemical Modulation of Stress Regulation

3.1. What Is Endocrine Noise?

Stress biology has traditionally conceptualized threats to homeostasis as psychosocial or physical in nature (trauma, deprivation, infection, and adversity), transduced into neuroendocrine responses via the HPA axis [7,50]. Within this framework, the stress response is usually understood as an adaptive reaction to challenge: the brain detects a perturbation, the HPA axis mobilizes cortisol, and physiological systems adjust in order to restore stability.

Environmental chemicals have not usually been discussed in these same terms, even though many of them can interfere with hormone signaling. EDCs have historically been studied mainly for their effects on reproduction and thyroid function [51,52]. Growing evidence challenges this separation. Many EDCs, widespread in plastics, pesticides, industrial by-products and personal care products, interact directly with hormone receptors and regulatory pathways that are integral to stress-system function [53,54,55]. Rather than acting as classical acute stressors, these compounds may introduce persistent, low-grade interference into endocrine signaling systems.

We conceptualize this form of interference as endocrine noise (Box 1). Unlike classical stressors, endocrine noise does not overwhelm the system through intensity. Instead, it disrupts regulatory precision by desynchronizing hormonal rhythms, altering receptor-level signaling, and weakening feedback control while leaving hormone production largely intact. In this sense, endocrine noise resembles static superimposed on an otherwise intelligible signal: the signal remains present, but its temporal precision and interpretability are progressively reduced. This framework helps explain why endocrine disruption frequently manifests as subtle dysregulation rather than overt endocrine failure.

Although endocrine noise is introduced here as a conceptual framework, several established analytical approaches allow the phenomenon to be examined empirically. In physiological terms, endocrine noise refers to a reduction in the fidelity of hormone signaling across time, manifested as increased variability, reduced rhythmic structure, or reduced predictability of endocrine output rather than simple shifts in mean hormone concentrations. Unlike a classical stress response, which is typically characterized by context-linked HPA activation followed by expected recovery, endocrine noise reflects degradation in the temporal organization and regulatory precision of endocrine signaling, even when average hormone concentrations remain within the expected range.

In the context of HPA-axis regulation, this loss of signaling fidelity can be operationalized using several complementary metrics Circadian organization is commonly characterized using features such as the CAR and the diurnal cortisol slope derived from repeated sampling protocols [56]. Dispersion-based indices—including within-day variance or coefficients of variation—capture instability in hormone levels across sampling intervals. Temporal irregularity in hormone secretion can be quantified using information-theoretic metrics such as approximate entropy or sample entropy, which estimate the unpredictability of physiological time series [57,58,59]. Chronobiological modeling such as cosinor analysis provides complementary estimates of circadian rhythm parameters including MESOR (Midline Estimating Statistic Of Rhythm), amplitude, and acrophase [60], while pulse-detection or deconvolution methods can characterize ultradian structure when high-frequency plasma cortisol sampling is available [46,61].

We propose interpreting endocrine noise only when convergent changes appear across these dimensions—for example, reduced circadian amplitude, increased temporal irregularity, and elevated within-day variability—rather than from a single isolated metric. This approach emphasizes degradation of the temporal organization of endocrine signaling rather than simple changes in mean cortisol levels. Figure 1 illustrates how endocrine noise distorts sex-specific HPA architectures, transforming baseline sex differences into divergent vulnerability trajectories.

Figure 1. Endocrine Noise and Sex-Specific Disruption of HPA Axis Regulation. Schematic showing how sex differences in HPA axis regulation shape divergent responses to EDCs. (Left) Females exhibit higher HPA reactivity and slower recovery; males show more efficient feedback. (Center) Chronic EDC exposure introduces “endocrine noise”, a disruption of timing and feedback precision illustrated here as conceptual oscillations representing cortisol rhythmicity. (Right) Estrogenic EDCs may amplify stress reactivity in females; anti-androgenic EDCs may impair HPA restraint in males, increasing sex-specific risk for emotional, immune, and metabolic dysfunction. Created with BioRender.com.

Box 1. Endocrine noise: conceptual definition and candidate empirical indicators.

Definition. Endocrine noise refers to reduced fidelity of endocrine signaling across time. In the HPA axis, it is expressed less by a simple shift in mean cortisol concentration than by disruption of temporal organization, increased irregularity, and reduced precision of feedback control.

Candidate empirical indicators in HPA-axis studies

• Rhythm structure: CAR, diurnal slope, circadian amplitude.
• Dispersion metrics: Within-day variance, coefficient of variation (CV).
• Temporal irregularity: Approximate entropy (ApEn), sample entropy (SampEn),
• Chronobiological organization: MESOR, amplitude, and acrophase from cosinor analysis
• Feedback regulation: Sensitivity to glucocorticoid suppression (e.g., dexamethasone suppression paradigms), Recovery dynamics following stress challenge
• Ultradian organization: Pulse frequency, pulse amplitude, ACTH–cortisol coupling (when high-frequency sampling is available)

Epidemiological evidence is consistent with this view. Higher body burdens of persistent EDCs, such as per- and polyfluoroalkyl substances (PFAS) and polychlorinated biphenyls (PCBs), have been associated with elevated perceived stress and increased inflammatory markers—hallmarks of chronic stress physiology—despite the absence of an identifiable acute stressor [12,62], Experimental studies further show that prenatal or early-life exposure to EDC mixtures can mimic some neuroendocrine effects of psychosocial stress during the same developmental windows, including shifts in cortisol output and HPA axis feedback [9,63].

Importantly, these effects are not sex-neutral [8]. The HPA axis is structurally and functionally sexually dimorphic: females exhibit greater or more sustained glucocorticoid responses, slower recovery in certain contexts, and weaker negative feedback, shaped by differences in sex hormone signaling and stress-related receptor expression [5,6,7,8]. As a result, identical sources of endocrine noise are unlikely to be interpreted equivalently in females and males. This conceptual framing shifts attention away from chemical exposure as a uniform stressor and toward sex-specific modulation of stress-system architecture.

3.2. Endocrine Noise in Feedback-Regulated Systems (HPA Axis)

The consequences of endocrine noise become particularly evident in feedback-regulated endocrine systems such as the HPA axis, where stability depends on precise timing, proportionality, and signal integration. The HPA axis operates as a dynamic loop in which hypothalamic corticotropin-releasing hormone (CRH) drives pituitary ACTH secretion in pulsatile fashion, stimulating adrenal cortisol release. Cortisol, in turn, feeds back rapidly at multiple levels of the axis to suppress its own production. Because this system depends not only on hormone concentration, but also on the timing and coordination of signaling, even modest disruption can alter regulatory precision.

In this context, endocrine disruption does not need to abolish hormone production to produce biologically meaningful effects. Persistent low-grade interference at the level of receptor binding, receptor crosstalk, feedback sensing, or signal integration may distort how endocrine information is encoded and interpreted. In the HPA axis, such disruption may be expressed as blunted circadian amplitude, altered pulsatility, impaired post-stress recovery, or weaker coupling between upstream ACTH drive and downstream cortisol output. These alterations are relevant because pulsatile and circadian glucocorticoid signaling influence receptor activation, transcriptional responses, and downstream metabolic and immune regulation [46,47].

EDCs may be particularly difficult to detect within this framework because their effects are often subtle, cumulative, and context dependent. Rather than producing overt endocrine failure, they may gradually reduce the fidelity of feedback-regulated signaling, leaving mean hormone levels apparently intact while altering the temporal structure of endocrine output. This may help explain why EDC-related HPA dysregulation is not always captured by conventional single-time-point endocrine measures. It also provides a mechanistic basis for considering why low-dose effects may be nonlinear and why similar exposures may not be interpreted identically in females and males.

3.3. Non-Monotonic Dose–Response Relationships and Sex-Specific Sensitivity

EDCs frequently exhibit non-monotonic dose–response (NMDR) relationships, in which effects observed at low, environmentally relevant exposures differ qualitatively from those seen at higher doses and therefore cannot be predicted from conventional monotonic toxicological models [64,65]. Such patterns are particularly relevant in endocrine systems because biological output depends not only on dose, but also on receptor affinity, receptor desensitization, competing receptor pathways, feedback regulation, and dose-dependent metabolism [64,65]. Recent mechanistic work further suggests that interference with systemic endocrine negative feedback may itself be sufficient to generate non-monotonic responses, even in the absence of direct cytotoxicity [66].

The intersection of NMDR with biological sex is especially relevant in the HPA axis, because females and males do not interpret endocrine signals within the same physiological context. Sex differences in glucocorticoid regulation are shaped by gonadal steroid signaling, receptor expression, and feedback sensitivity, such that a weak endocrine perturbation may be amplified, buffered, or redirected differently across sexes. In this framework, the dose range at which an EDC begins to alter HPA function may differ between females and males because the same ligand is acting within distinct receptor and transcriptional environments.

Several mechanisms may contribute to this sex-dependent dose sensitivity. First, steroid receptor signaling is highly context dependent and influenced by ligand concentration, receptor abundance, and the conformational changes that determine coregulator recruitment. Second, glucocorticoid, estrogen, and androgen receptors participate in transcriptional crosstalk at the chromatin level, where activation of one receptor can alter the genomic binding and transcriptional output of another [67]. Such interactions provide a biologically plausible basis for low-dose effects that differ by sex, particularly in stress-regulatory tissues in which glucocorticoid and sex-steroid pathways are already tightly integrated. From this perspective, sex-specific NMDR behavior does not require different exposure levels alone; it can emerge because identical doses are interpreted differently by sexually dimorphic endocrine architectures.

Direct evidence for fully dose-resolved, sex-specific NMDR curves in HPA-axis outcomes remains limited. However, the available literature supports the broader principle that low-dose EDC exposure can produce sex-divergent neuroendocrine effects. In rats, perinatal low-dose bisphenol A exposure has been shown to perturb basal and stress-induced HPA activity in a sexually dimorphic manner, with females showing increased basal corticosterone and altered hypothalamic glucocorticoid receptor regulation, whereas males exhibit exaggerated corticosterone responses to stress [68]. Similarly, perinatal bisphenol A (BPA) exposure has been associated with increased basal corticosterone, ACTH, and CRH-related measures in males but different glucocorticoid receptor and mineralocorticoid receptor responses in females, consistent with sex-divergent disruption of feedback regulation [14]. Human cohort data point in the same direction: in the APrON study, prenatal BPA exposure was associated with elevated baseline cortisol in infant girls but lower baseline cortisol in boys, with opposite patterns in stress reactivity, supporting the view that the same exposure may shift HPA phenotype in different directions across sexes [69].

These observations suggest that non-monotonicity and sex-specificity are likely to intersect in stress-regulatory systems, even though direct sex-stratified NMDR analyses remain uncommon in the HPA literature. Low-dose endocrine effects should therefore not be assumed to be biologically equivalent across sexes, and future studies should explicitly test dose-by-sex interactions using time-resolved endocrine phenotyping and repeated exposure assessment.

4. Life-Course Patterns of Sex-Specific HPA Disruption Following EDC Exposure

Building on the conceptual framework of endocrine noise and the sexually dimorphic organization of the hypothalamic–pituitary–adrenal axis described above, this section examines how EDC exposure alters stress regulation across the lifespan in a sex-specific manner. Rather than producing a uniform stress phenotype, EDCs are associated with divergent cortisol outcomes in females and males that depend on developmental timing and physiological context. A life-course perspective is essential here because the HPA axis is developmentally plastic and differentially sensitive across the lifespan. In the sections that follow, exposures occurring during sensitive developmental windows are interpreted primarily as forms of developmental resetting rather than as endocrine noise in the narrower sense defined in Section 3. During sensitive periods, EDC exposure may recalibrate neuroendocrine set points, feedback sensitivity, and stress responsivity in ways that are not immediately pathological and may even appear adaptive in the short term. However, such adjustments may later become maladaptive if the recalibrated system encounters subsequent chemical or psychosocial challenges in a different physiological context. In this sense, early exposure may contribute to maladaptive plasticity, whereby neuroendocrine adaptation reshapes later vulnerability to affective, metabolic, and immune disease in a sex-specific manner. Section 4.1 summarizes empirical evidence for sex-divergent patterns of HPA disruption across developmental windows, while Section 4.2 addresses the biological integration of stress and sex-steroid signaling that may account for these divergences.

4.1. Sex-Divergent Cortisol Outcomes Across Development

A defining feature of the HPA-axis response to EDCs is that biological sex often determines the directionality, rather than the magnitude, of the disruption. Both human and animal studies suggest that the same chemical signal is integrated differently within male and female endocrine contexts, leading to divergent cortisol trajectories.

Human cohort data provide striking evidence of this divergence. A 2017 study by Giesbrecht et al. found that prenatal BPA exposure was associated with elevated basal cortisol in 3-month-old girls but suppressed levels in boys; conversely, stress-induced reactivity was heightened in boys but blunted in girls following identical prenatal exposure [69]. Importantly, these associations emerged despite comparable exposure levels within the same intrauterine environment, indicating that sex differences were not driven by dose but by differential integration of the chemical signal.

These findings are mirrored in rodent models, where perinatal BPA exposure elevates basal glucocorticoids in females while leaving males unaffected, yet selectively exaggerates the male HPA response to acute stressors [68,70]. These effects are often subtle under baseline conditions but become pronounced when the system is challenged, supporting the idea that EDCs may recalibrate stress responsivity rather than simply shifting resting hormone levels.

The disruption of HPA dynamics is even more complex when considering real-world chemical mixtures. Recent research utilizing a developmentally relevant EDC mixture (“NeuroMix” containing BPA, phthalates, and vinclozolin) further confirms that sex modifies the temporal logic of the stress response. In rat models, prenatal exposure to these mixtures resulted in a significant delay in corticosterone recovery following restraint stress, specifically in males, a phenotype not observed in females [71]. This demonstrates that sex may affect dynamic regulation, or the capacity to stop a stress response, even when baseline levels seem normal.

Systematic syntheses reinforce the consistency of these observations. A recent review of 22 cohort studies concluded that prenatal EDC exposure consistently yields sex-specific associations with HPA-axis hormones, particularly regarding phenols and phthalates [11]. These disruptions persist into critical developmental transitions such as puberty, a period characterized by profound recalibration of both stress and sex-steroid hormones. For example, in peripubertal boys, higher urinary BPA concentrations have been associated with a significant reduction (approximately 20–25%) in serum cortisol [72]. This cortisol suppression was accompanied by elevated testosterone levels, suggesting that EDCs may disrupt the inhibitory crosstalk that typically exists between the hypothalamic-pituitary-gonadal (HPG) axis and the HPA axis.

Taken together, these findings indicate that EDC exposure does not impose a uniform stress phenotype. Instead, it interacts with sex-specific developmental trajectories of HPA regulation, producing divergent cortisol outcomes that depend on timing, context, and challenge state. The key empirical pattern is therefore not simply whether cortisol increases or decreases, but that comparable exposures may shift males and females toward different regulatory states. Recognizing this principle is essential for interpreting inconsistencies in the literature and for designing studies capable of detecting biologically meaningful, sex-specific effects of environmental chemical exposure.

4.2. Integration of Stress and Sex-Steroid Signaling as a Source of Divergence

EDCs do not act on the HPA axis in isolation. Instead, their effects emerge at the intersection of glucocorticoid signaling with sex steroid pathways that already differ fundamentally between females and males. Estrogen receptors (ERα/ERβ), androgen receptors (AR), and glucocorticoid receptors (GR) belong to the same nuclear receptor superfamily and share co-regulators, chromatin access points, and downstream transcriptional machinery [67]. This overlap allows estrogenic or anti-androgenic EDCs to alter glucocorticoid signaling indirectly, without displacing endogenous glucocorticoids. Rather than acting through receptor presence alone, these effects depend on contextual factors such as co-regulator recruitment, chromatin accessibility, and timing of receptor activation, which differ between females and males. As a result, identical chemical exposures can produce sex-biased glucocorticoid responses through altered transcriptional integration rather than overt receptor antagonism [73].

Estrogen signaling is a key modulator of stress regulation and provides an important mechanistic context for female-biased HPA responses. Estradiol has been shown to potentiate stress responsivity and weaken glucocorticoid negative feedback, particularly through ERα-mediated effects within hypothalamic and limbic stress circuits [44,45]. Estrogenic EDCs, including bisphenols and certain organochlorines, may therefore amplify this permissive context by augmenting estrogenic tone or competing for shared transcriptional coactivators. In such a system, even weak estrogenic signaling can shift the balance of receptor integration, favoring prolonged or exaggerated stress responses without altering baseline glucocorticoid synthesis.

In males, androgen signaling plays a largely inhibitory role in stress regulation, contributing to tighter feedback restraint and more efficient recovery following stress exposure. Anti-androgenic EDCs, such as certain phthalates, vinclozolin, p,p′-dichlorodiphenyldichloroethylene (DDE), may disrupt this inhibitory tone by antagonizing AR signaling or reducing androgen availability [74,75,76,77]. Rather than producing a uniform increase in cortisol, such interference effectively “releases the brake” on stress-axis drive, biasing the system toward heightened reactivity or delayed termination of responses. This mechanism provides a coherent explanation for male-specific vulnerability patterns observed under anti-androgenic exposure, independent of exposure dose [78,79].

Beyond receptor presence alone, the integration of stress and sex-steroid signaling is shaped by context-dependent factors including co-regulator recruitment, chromatin accessibility, and timing of receptor activation. Transcriptomic studies demonstrate that identical endocrine signals can elicit sex-dependent gene expression profiles in stress-relevant tissues, reflecting differences in transcriptional context rather than differences in circulating hormone concentrations [73,80]. These findings support the view that EDCs can alter regulatory integration, without necessarily producing overt endocrine abnormalities at rest.

Developmental timing further constrains these interactions. During sensitive windows such as perinatal life and puberty, coordination between the HPA and HPG axes is actively recalibrated. Chemical interference during these periods may therefore bias the maturation of stress–sex steroid integration in sex-specific ways, establishing regulatory trajectories that persist into later life [11,81]. Importantly, such effects may remain latent under basal conditions and emerge primarily when regulatory demand increases.

Taken together, these mechanisms indicate that sex-divergent HPA outcomes following EDC exposure do not require sex differences in exposure intensity or toxicokinetics. Instead, they arise from differential integration of chemical signals into stress–sex steroid networks that are already organized differently in females and males. This framework explains why identical exposures can produce divergent stress phenotypes and underscores the importance of mechanistic models that move beyond single-receptor or single-hormone explanations.

5. Chemical Modulation of the Temporal Architecture of the HPA Axis

Building on the feedback and timing mechanisms described above, this section considers how these sex-specific properties shape long-term adaptation, including recovery, habituation, and vulnerability to stress-related disease.

Cortisol secretion reflects the circadian output of the HPA axis and is characterized by a rapid increase after awakening (the cortisol awakening response, CAR), a progressive decline across the day, and superimposed ultradian pulses. These temporal features are not ancillary to stress regulation; they determine receptor engagement, transcriptional dynamics, and downstream physiological effects. Because these rhythmic properties differ by sex even in the absence of chemical exposure, similar EDC exposures are superimposed on non-equivalent baseline regulatory architectures [32,82].

At the molecular level, circadian HPA activity is coordinated with core clock machinery, as glucocorticoid signaling interacts with canonical clock components such as PER1, PER2, and BMAL1, allowing cortisol rhythms to function both as outputs of the circadian system and as chronoregulatory signals that organize gene regulation across the day [32].

Across human and animal studies, females generally exhibit a more pronounced morning cortisol dynamic and steeper diurnal decline than males, whereas males more often show flatter profiles with reduced amplitude. These differences are shaped by sex-steroid modulation of central stress circuits, circadian coupling, and cortisol bioavailability, including estrogen-sensitive regulation of corticosteroid-binding globulin. Accordingly, comparable perturbations of cortisol timing or amplitude may be more readily detected in females, whereas subtler disruption in males may require denser, time-resolved sampling for detection [32,82,83].

Human evidence increasingly suggests that EDCs preferentially disrupt the temporal organization of cortisol rather than uniformly altering mean hormone levels. In the APrON cohort, higher maternal urinary bisphenol A concentrations were associated with lower waking cortisol and a flatter diurnal slope, consistent with compression of the normal circadian cortisol signal [69]. Similar patterns have been reported in occupational settings: pesticide-exposed farmers exhibited suppressed morning cortisol and attenuated daytime decline compared with organic controls, indicating loss of rhythmic amplitude rather than isolated hypo- or hypercortisolemia [84]. Systematic reviews of epidemiological studies further conclude that associations between prenatal synthetic chemical exposure and HPA-axis outcomes are heterogeneous but repeatedly implicate rhythm-related measures, including CAR and diurnal regulation, underscoring the relevance of circadian phenotyping [11,85].

Sex differences in the circadian organization of the HPA axis offer a coherent explanation for why findings in the EDC literature are often inconsistent and sex-specific. The same degree of disruption in cortisol rhythm—such as a reduction in amplitude or a flattening of the diurnal slope—may be readily detectable in individuals with higher baseline rhythmicity, whereas it appears attenuated or clinically subtle in those with flatter profiles. When males and females are analyzed together, or when cortisol timing is not taken into account, these unequal effects can offset one another, leading to null or inconsistent associations despite biologically meaningful, sex-stratified patterns. In this context, apparent variability across studies may reflect systematic differences in how circadian disruption is expressed and measured across sexes, rather than true inconsistency in underlying biological effects.

From a clinical and translational perspective, this means that cortisol measurements taken at a single, untimed point are unlikely to capture biologically meaningful disruption. Instead, studies should prioritize repeated, time-anchored sampling across the day, report sex-stratified measures of circadian rhythm such as the cortisol awakening response and diurnal slope, and interpret HPA-axis outcomes within a circadian framework to more accurately identify sex-specific vulnerability to endocrine disruption.

6. Sex-Specific Stress Risk Architecture in the Context of EDC Exposure

6.1. From Diagnosis to Risk Architecture

Persistent dysregulation of the HPA axis, particularly in the context of environmental EDC exposure, is more accurately understood as a risk architecture than as a discrete clinical diagnosis. By risk architecture, we refer to the sex-specific configuration of endocrine response traits, developmental exposures, and contextual stressors that shifts probabilistic vulnerability across multiple health domains. Flattened diurnal cortisol slopes, blunted morning surges, and elevated evening cortisol have been associated with increased vulnerability to a wide range of clinical conditions, including mood disorders, metabolic dysfunction, and immune dysregulation [86,87]. These features reflect a persistent alteration in stress-system regulation that increases disease susceptibility across systems, rather than a disease-specific endocrinopathy.

In this framework, EDCs do not necessarily introduce a novel pathological entity; rather, they can act as risk modifiers by degrading the precision of endocrine signaling. Through weak or partial agonist/antagonist activity at glucocorticoid and sex-steroid receptors, EDCs can impair the timing, amplitude, and feedback regulation of cortisol signaling [1,65]. This form of endocrine interference is particularly relevant in a system that is already sexually dimorphic in baseline regulation and stress responsivity [8,20]. The resulting phenotype is not overt endocrine failure, but reduced regulatory fidelity that increases susceptibility to downstream stress-related pathology.

At present, this framework should be regarded as hypothesis-generating rather than clinically predictive. Robust estimates of risk associated with a specific combination of HPA phenotype and EDC exposure profile are not yet available, because most studies do not combine high-resolution endocrine phenotyping with repeated exposure assessment across development.

6.2. Divergent Risk Pathways: Female vs. Male Vulnerability

The sexually dimorphic features of the HPA axis described earlier translate into distinct clinical vulnerability profiles when disrupted by EDC exposure. Rather than inducing novel disease entities, EDCs modulate existing sex-specific stress architectures, shifting risk trajectories based on underlying endocrine tone and life-course regulatory history [8,20].

In females, such dysregulation often manifests as increased risk for affective disorders, fatigue, sleep disruption, and autoimmune dysregulation [86,88]. Estrogenic EDCs may further destabilize feedback efficiency or amplify stress reactivity, particularly when exposure coincides with developmental sensitivity or psychosocial adversity. In contrast, male risk profiles are typically skewed toward metabolic, cardiovascular, and externalizing outcomes. This aligns with evidence that androgens normally support tighter HPA feedback and more rapid post-stress recovery [8,76,78]. Anti-androgenic EDCs may erode this regulatory inhibition, thereby unmasking stress hyper-reactivity and shifting the system toward cardiometabolic instability, even without altering baseline cortisol levels.

Importantly, these trajectories are probabilistic, not deterministic. EDCs function as risk modifiers, interacting with sex-specific baselines and life-course regulatory histories to shape vulnerability.

Developmental exposure studies also suggest that estrogenic EDCs can reduce or blur normally sex-differentiated traits, particularly in neurobehavioral domains, reflecting a reorganization of sex-specific regulatory architecture rather than a uniform exaggeration of dimorphism [89]. Thus, similar alterations in cortisol dynamics, such as changes in the CAR or diurnal slope, may carry different implications across sexes, aligning with distinct patterns of vulnerability in females and males depending on physiological context, timing of exposure, and broader environmental conditions.

6.3. Domains of Consequence Without Disease Cataloging

A transdiagnostic perspective is more clinically meaningful than cataloging diseases. HPA dysregulation contributes to multiple domains of risk—affective, metabolic and immune—regardless International Classification of Diseases. For instance, cortisol rhythm flattening and CAR blunting predict worse health outcomes across domains [86]. Chronic glucocorticoid imprecision may lead to immune resistance and pro-inflammatory drift, again modulated by sex-specific immune responsiveness [90].

These domains should be understood as overlapping analytic categories rather than mutually exclusive disease classes. Affective risk includes vulnerability related to mood, anxiety, sleep, and stress sensitivity; metabolic risk includes adiposity, insulin resistance, glucose dysregulation, and cardiometabolic strain; immune risk includes altered inflammatory balance, glucocorticoid sensitivity, and immune responsivity. The same endocrine phenotype may contribute to more than one domain, but this structure helps organize the downstream consequences of sex-specific HPA disruption.

This approach foregrounds vulnerability rather than diagnosis, making space for early intervention. Time-resolved cortisol metrics (such as CAR, diurnal slope, evening nadir), sex-stratified analyses, and dynamic assessments of bioavailable cortisol (e.g., free cortisol/CBG ratios) are more likely to detect meaningful EDC effects than static, single-sample assays [86].

Importantly, EDCs should be understood not as primary causes of stress-related disease, but as amplifiers of vulnerability. By converging with psychosocial stressors on the same sex-biased HPA architecture, chemical exposures can accelerate regulatory drift even when each exposure alone would produce modest effects [11,12,91]. This “double exposure” framework explains why risk may emerge cumulatively and heterogeneously across populations.

6.4. Clinical Implications and Preventive Perspectives

Clinicians should interpret endocrine instability in the context of both sex and stress history. Flattened cortisol rhythms, delayed recovery, or blunted CAR may serve as early biomarkers of stress-system vulnerability, especially in individuals with cumulative stress exposure or elevated likelihood of chemical exposure. These patterns are not deterministic, but they may signal increased sensitivity to future challenges and reduced physiological flexibility.

Prevention is therefore likely to benefit from dual attention to modifiable endocrine stability and modifiable exposure. Behavioral strategies that reinforce circadian integrity—sleep regularity, consistent light exposure, and stress regulation—may mitigate HPA lability. At the population level, public health measures that reduce EDC burdens are especially warranted in groups facing converging social and chemical adversity. Such efforts do not eliminate risk but may recalibrate the stress-response system toward greater resilience.

7. Methodological Blind Spots in Human Studies on EDC–HPA Interactions

Despite increasing interest in EDCs as modulators of stress regulation, human studies in this area remain fragmented by persistent methodological challenges that obscure biologically meaningful effects—especially those related to sex-specific HPA dysregulation.

7.1. Ignoring Sex as a Biological Modifier

An important limitation of the current literature is that sex is often acknowledged but not fully leveraged in the analysis of HPA axis outcomes. While most studies report the sex of participants or adjust for it statistically, far fewer treat sex as a meaningful biological context that shapes stress regulation. This is a relevant concern given that the HPA axis differs between males and females in its baseline activity, feedback regulation, and receptor dynamics [7,20]. When these differences are not explicitly examined, environmentally driven effects may vary by sex in ways that are difficult to detect in pooled analyses. As a result, combining male and female data can reduce biological clarity and contribute to inconsistent or attenuated findings, even when underlying endocrine disruption is present.

7.2. Misaligned Sampling with Circadian Rhythms and Reliance on Single Time-Point Measures Timing

Hormone levels governed by the HPA axis exhibit strong circadian rhythms. Cortisol in particular follows a robust 24 h cycle, peaking shortly after awakening and reaching a nadir around midnight [92]. Afternoon sampling, inconsistent timing across participants, or reliance on single time-point measurements render results highly vulnerable to confounding by circadian variability [56,61].

Moreover, many EDCs appear to target the structure of the rhythm itself (flattening the slope, blunting the morning peak, or disrupting feedback regulation) effects that cannot be captured in single-point designs [1,86]. More reliable indicators, such as CAR, diurnal slope, or area under the curve (AUC), are rarely used, despite offering better biological resolution.

7.3. Inadequate Characterization of Female Hormonal Status

For studies including women, hormonal status is often neglected. Menstrual cycle phase and hormonal contraceptive use are key modulators of cortisol binding and receptor sensitivity [93,94], meaning that the same cortisol level may have different implications depending on estrogen status. Without this context, analyses risk conflating normal endocrine variation with EDC effects, or missing disruptions that are phase-specific.

7.4. Narrow Focus on Cortisol Alone

In addition, the predominant reliance on cortisol as a single outcome limits interpretation of sex-specific adrenal disruption. Cortisol reflects HPA output but does not capture broader adrenal steroid balance or peripheral glucocorticoid metabolism, both of which differ between females and males [19,20]. Experimental and mechanistic studies indicate that EDCs can interfere with adrenal steroidogenesis and with enzymes regulating cortisol activation and inactivation, including 11β-hydroxysteroid dehydrogenases [1,95]. However, most human studies do not assess complementary adrenal hormones such as DHEA or cortisone, limiting the ability to detect sex-divergent patterns of anabolic–catabolic balance or tissue-level glucocorticoid exposure. As a result, multidimensional adrenal effects of EDC exposure may remain undetected when analyses rely exclusively on circulating cortisol.

7.5. Heterogeneity in Protocols and Measurement Tools

Finally, methodological heterogeneity remains a barrier to synthesis. Studies differ widely in hormone matrices (saliva, serum, hair), sampling protocols, assay sensitivity, and analytic thresholds. Very few include challenge paradigms or longitudinal measures, limiting the ability to distinguish between basal regulation and dynamic reactivity [11]. Even within saliva-based designs, nonadherence to sampling timing and differences in protocol implementation can materially bias cortisol features, adding noise and reducing reproducibility [56]. Advancing the field will require biologically attuned designs that reflect the known features of the HPA axis: its sexual dimorphism, circadian structure, and adaptive flexibility. This means moving beyond one-size-fits-all models toward sex-stratified, timing-sensitive (CAR, diurnal slope), and multi-hormonal approaches, with careful attention to hormonal status and rhythm dynamics [8]. Without these refinements, important patterns of endocrine disruption, especially those relevant to sex-specific health risk, will continue to be missed.

8. Future Directions: Toward Sex-Conscious Environmental Endocrinology

This review synthesizes evidence that EDCs perturb HPA-axis function not uniformly but in sex-specific ways, introducing endocrine noise that distorts cortisol rhythms, feedback efficiency, and responsivity to later stressors. The challenge now is to translate these insights into more precise research designs, regulatory frameworks, and preventive strategies. A sex-aware lens, treating biological sex as a fundamental modulator of chemical stress, will sharpen our ability to detect, interpret, and mitigate these effects.

8.1. Enhanced Exposure Assessment

To capture the sex-specific impact of EDCs, exposure assessment must move beyond pooled, single-sample snapshots. Women, for example, typically show higher phthalate or paraben biomarkers due to differential use of personal care products, while men may have greater pesticide exposures in occupational settings [96,97]. These external disparities are compounded by biological differences in metabolism, body composition, and hormone-driven clearance. Future biomonitoring should incorporate sex-stratified, multi-matrix sampling (urine, hair, plasma), repeated across critical windows such as pregnancy or puberty. Combining these data with dynamic cortisol metrics like the diurnal slope or CAR, will clarify how exposure patterns translate into HPA disruption across sexes [11].

8.2. Regulatory and Toxicological Refinement

Although funders and some regulators now require consideration of sex as a biological variable, historical male bias in toxicology has left many chemical safety thresholds based on sex-aggregated or male-dominant data [98,99]. Because HPA feedback, circadian rhythm metrics, and steroid ratios are sexually dimorphic and mechanistically relevant to EDC effects, regulatory testing should require sex-stratified analyses and incorporate time-resolved endocrine endpoints (e.g., CAR, diurnal slope, cortisol, DHEA, cortisone/cortisol) to detect dimorphic levels and to justify sex-specific uncertainty factors where appropriate [100].

8.3. Preventive and Clinical Applications

Endocrinologists and primary care clinicians are increasingly well positioned to integrate environmental exposures into patient care, especially when endocrine instability is suspected. For example, in women with disrupted cortisol recovery or sleep disturbances during pregnancy, personal care product substitution could serve as a feasible, exposure-reducing intervention [101,102]. For men who are regularly exposed to anti-androgenic pesticides at work, reducing chemical exposure and supporting normal androgen-dependent physiology may help prevent HPA overactivation [103,104].

Public health strategies should also adopt a sex-informed lens, targeting chemical burden reductions during sensitive windows and designing outreach campaigns that acknowledge sex-specific vulnerabilities.

8.4. Research Priorities

Moving from concept to application requires a coordinated research agenda. Longitudinal human cohorts should integrate sex-stratified EDC exposure metrics with dynamic HPA phenotyping across development. Future work should also examine broader pathways that may modify sex-specific HPA-axis responses to EDC exposure. These include transgenerational epigenetic mechanisms and microbiome-related pathways, both of which are biologically plausible but remain insufficiently characterized in studies directly addressing sex-specific HPA outcomes [105,106,107].

Future prenatal cohort studies should also assess maternal HPA activity and psychosocial stress in parallel with repeated EDC exposure measures in order to better distinguish confounding, mediation, and potential synergistic effects.

Advanced in vitro platforms, including endocrine organoids and sex-informed three-dimensional models [108,109] can facilitate mechanistic dissection of receptor crosstalk and sexually dimorphic transcriptional responses that are difficult to resolve in traditional animal or two-dimensional cell cultures. Finally, clinical trials of preventive interventions—such as circadian-aligned nutrition or stress inoculation—should test whether modulating the endocrine environment can buffer HPA axis sensitivity, especially in high-EDC contexts.

9. Conclusions

A sex-aware approach to environmental endocrinology does more than correct for oversight; it proposes reframing EDCs from general disruptors to context-dependent modulators of stress-system architecture. Incorporating sex into study design, regulation, and prevention enables more accurate risk identification and equitable protection. This framework holds promise not only for resolving inconsistencies in the EDC literature but also for developing more precise strategies that strengthen stress resilience across the lifespan.