Thymus Involution across the Human Menstrual Cycle

Structured Abstract

Importance:

Thymic involution in humans has traditionally been conceptualized as a linear, age-associated process. Animal studies suggest transient hormone-dependent fluctuations in thymus size during reproductive transitions, including across the estrous cycle, but translational evidence in humans remains limited.

Objective:

To determine whether thymus size changes across the menstrual cycle in healthy women and whether these changes are associated with fluctuations in estradiol and progesterone levels.

Design:

Observational repeated-measures study using quantitative computed tomography (qCT). Data were collected as part of an imaging study examining menstrual cycle–related physiological variation. Participants underwent assessments during menses and the early luteal phase within the same menstrual cycle. Mixed-effects models with robust variance estimation evaluated associations between cycle phase, hormonal contraception, reproductive hormone levels, and thymus size. Thymus segmentation was independently conducted by 2 blinded raters.

Setting:

Single-center university-based imaging study conducted at the University of Iowa.

Participants:

Thirty-one non-smoking women with regular menstrual cycles were included, of whom n = 16 used oral hormonal birth control and n = 15 did not use hormonal contraception. Exclusion criteria included pregnancy, breastfeeding, postmenopausal status, diabetes, body mass index greater than 30 kg/m², hysterectomy, or use of long-term noncyclic hormonal contraception. Participants tracked menstrual cycles using temperature monitoring and ovulation kits before completing visits during menses and the early luteal phase. All participants provided informed consent before study participation.

Main Outcomes and Measures:

Primary outcome was thymus size measured in mm³ using ultra-low-dose qCT imaging. Secondary outcomes included serum estradiol and progesterone levels.

Results:

Estradiol levels increased from menses to the early luteal phase independent of hormonal contraceptive status. Progesterone levels were significantly lower among women using hormonal birth control. Thymus size differed significantly by hormonal contraceptive group, with larger thymus volumes observed among women not using hormonal birth control. Among women not using hormonal contraception, 66.7% demonstrated thymic involution from menses to the early luteal phase, compared with 37.5% of women using hormonal contraception. Estradiol and progesterone significantly interacted in predicting thymus size.

Conclusions and Relevance:

These findings provide first-in-human evidence suggesting that thymus involution demonstrates short-term dynamics across the menstrual cycle and may be influenced by reproductive hormones and hormonal contraception.

Introduction

Thymus involution in humans is considered a linear process associated with aging. Thymic size has been shown to shrink at an annual rate of approximately 3% per year until 40 years of age and at an annual rate of 1% thereafter.¹ Mice show transient changes in thymus size during phases of hormonal transition. Early studies illustrated pregnancy-induced involution of the thymus in mice,² even when controlling for changes in body weight.³ Following studies identified in pregnant mice inversely correlated thymus weight with progesterone levels. Importantly, the association between thymus weight and progesterone is observed progressively over pregnancy in mice,⁴ and the pregnancy-induced decrease in thymus size is transient, recovering 2-weeks postpartum.⁵ No such association is present in non-pregnant and pseudo-pregnant mice.⁶ Animal studies across the estrous cycle further illustrate that the proestrus stage - associated with a peak in circulating estrogen – is characterized by the greatest decrease in thymus weight.⁷ Early castration in cattle, pigs and rabbits, has been shown to delay thymus involution.⁸ Chemical and surgical castration even reversed age-related thymic involution in rats.⁹ Experimental studies further illustrate dose dependent effects of sex-steroid administration on thymus size in rats and mice, with higher doses of estrogen administration resulting in a greater decrease of thymus weight.¹⁰ Data on dynamic within-subject changes in thymus size during phases of hormonal transition in humans are scarce. One study¹¹ reported on two pregnant patients with signs of pulmonary embolism who underwent repeated computer tomography (CT) imaging. Both cases showed an enlarged thymus decreasing in size following birth. Changes in thymus size across the menstrual cycle in humans have previously not been studied. We hypothesized that thymus size demonstrates dynamic within-subject variation across the menstrual cycle in humans, with greater thymic involution associated with increases in estradiol and modulation by progesterone and hormonal contraceptive use.

Methods

We drew on existing data from quantitative CT (qCT) in n = 31 non-smoking, healthy women with regular menstrual cycles (n = 16 on cyclic hormonal birth control), who underwent qCT imaging at menses (M) and the early luteal phase (ELP).¹² Subjects were recruited from campus advertisements and normal, non-smoking cohorts from other research studies examining the lungs. Exclusion criteria included subjects who were pregnant, breastfeeding, body mass index > 30 kg/cm2, post-menopausal, diabetic, were taking long-term non-cyclic birth control (i.e., implant, patch, or injected), or had a hysterectomy. All study subjects provided written informed consent to participate in the study and data was collected with approval from the University of Iowa Institutional Review Board (IRB# 201606790), in accordance with the Declaration of Helsinki.

Women were asked to track their menstrual cycle via temperature and an ovulation kit for 1–2 months, before completing two study visits (M and ELP). The two study visits were completed within the same month/cycle for most subjects and were completed over more than one cycle in some cases due to scheduling restrictions. At each visit, subjects had peripheral blood collected and underwent ultra-low-dose qCT imaging. Hormone levels of estradiol and progesterone were analyzed from peripheral blood using the Roche Diagnostics Estradiol III and Progesterone III assays processed by the Emory Warner Pathology Laboratories.¹³ Ultra-low-dose, non-contrast chest qCT imaging was performed using a SOMATOM Force CT scanner (Siemens Healthcare) with 120 kV, 36 ref mAs, CareDose on, pitch 1.2, 0.5-s rotation time, 0.5-mm slice thickness, and Qr40 (admire 3, < 5-s scan). The CTDI_vol for one scan at 36 mAs was approximately 2.4 mGy or 1.0 mSv. Thymus size in mm3 was manually traced in ITK-SNAP,¹⁴ independently by two raters (TS and KES). Raters were blinded to group and phase and showed good agreement (two-way mixed ICC 0.895; 95%: [0.823;0.936]). The average of both raters was used for all analyses.

Data were analyzed using mixed models with robust variance estimation, addressing the main effect of cycle phase (M/ELP) and group (hormonal birth control/no hormonal birth control) as well as their interaction on hormone levels and thymus size, using the subject ID as random factor over repeated measures. Contrasts were Bonferroni corrected and all analyses were adjusted for age and days between repeated assessments. The relative change in thymus size was described descriptively in change in percent from M to ELP. Additional mixed models predicted the change in thymus size based on hormone interactions. Simple linear regression was used to illustrate bivariate associations. Statistical analyses were performed with Stata/SE (version 18.0; StataCorp LP, College Station, TX, US), with a set significance level of α = 0.05.

Results

Females not on hormonal birth control (NBC) were older compared to those on oral hormonal birth control (OBC; t₍₂₉₎ = 2.779, p < .01). Full descriptive data by group and cycle phase is provided in Table 1. Estradiol showed a significant main effect of cycle phase (model: χ² _(62/31) = 19.84, p = .001; phase: χ² = 6.15, p = .013), suggesting a significant increase from M to the ELP independent of group (NBC/OBC). No significant effect of group nor a significant group by phase interaction was observed. Progesterone showed a significant main effect of group (model: χ² _(62/31) = 16.11, p = .01; group: χ² = 9.83, p < .01), suggesting lower progesterone in females OBC in contrast to those NBC. No significant effect of cycle phase nor a significant group by phase interaction was observed. Individual trajectories in estradiol and progesterone by phase of the menstrual cycle are illustrated in Figure 1A. Thymus size (mm³) showed a significant main effect of group (model: χ² _(62/31) = 19.68, p = .001; group: χ² = 6.93, p = .01), suggesting greater thymus size in NBC in contrast to OBC. No significant effect of phase nor a significant group by phase interaction was observed. Individual trajectories of thymus size by phase of the menstrual cycle are illustrated in Figure 1B. In NBC, n = 10 (66.67%) showed thymus involution from M to ELP (−0.81 to −30.32% change in relative size) while thymus size increased in n = 5 (+2.98 to +66.59% change in relative size). In OBC, n = 6 (37.50%) showed thymus involution from M to ELP (−5.81 to −28.43% change in relative size) and n = 10 showed an increase in thymus size (+9.69 to +55.34% change in relative size). Estradiol and progesterone showed a significant interaction in predicting thymus size: χ² _(62/31) = 53.76, p < .001; interaction: coef. = −6.85, p < 0.01. As illustrated in Figure 1C, greater thymus involution, independent of group, was associated with a greater increase in estradiol from M to ELP, whereas an increase in thymus size was associated with a greater increase in progesterone from M to ELP.

Table 1:

Descriptive data by group and phase of the menstrual cycle;

	no birth control (NBC)		on birth control (OBC)
	M	ELP	M	ELP
n	15		16
Age in years, mean (SD), [range]	31.93 (9.48) [21–51]		24.63 (4.44) [20–38]
Days between visits, mean (SD), [range]	24.07 (21.12) [7–77]		19.94 (11.34) [9–44]
Type/dosing of birth control, n (%)
NGM-EE	0 (0.00)		6 (37.50)
DRSP-EE	0 (0.00)		8 (50.0)
LNG-EE	0 (0.00)		1 (6.25)
DSG-EE	0 (0.00)		1 (6.25)
Dosing in mg, mean (SD), [range]*	0 (0.00) [0.00–0.00]		0.03 (0.01) [0.02–0.035]
Estradiol in pg/mL, mean (SD), [range]*	49.60 (43.09) [6–178]	97.73 (62.74) [21–207]	32.44 (26.54) [5–91]	102.50 (183.05) [5–560]
Progesterone in ng/mL, mean (SD), [range]**	1.65 (2.79) [0.20–10]	3.51 (4.09) [0.20–10.70]	0.36 (0.17) [0.20–0.90]	0.41 (0.16) [0.20–0.70]
Thymus size (mm3), mean (SD), [range]	19269.28 (9035.55) [9757.17–38414.95]	18573.06 (8745.54) [7629.9436998.55]	13428.74 (4050.06) [4424.4120231.40]	14940.71 (6392.18) [6872.6928805.90]

M: menses; ELP: early luteal phase; SD: standard deviation; DRSP: drospirenone; DSG: desogestrel; NGM: norgestimate; LNG: levonorgestrel; EE: ethinyl estradiol;

^*:n = 6 assays in the OBC group below the measurement threshold of <5 pg/ml, imputed with 5 pg/ml;

^**:n = 2 assays in the OBC group below the measurement threshold of <0.2 ng/mL, imputed with 0.20 ng/ml

Figure 1: Individual Change in Reproductive Hormones and Thymus Size by Group and Phase;

M: menses; ELP: early luteal phase;

Discussion

To our knowledge, the present data are first-in-human, showing secondary within-subject dynamics of thymus size on the level of days (not years). Thymus size shows intra-individual variance across the menstrual cycle and is further influenced by hormonal contraception. Our findings challenge established conceptions of thymus involution in humans. The functional significance of these findings, however, remain speculative. Thymic atrophy in aging has been suggested to contribute to immunosenescence and inflammaging via decreased thymic output of naïve T cells.¹⁵ Here we illustrate secondary dynamics of thymus involution as a function of fluctuating reproductive hormones and hormonal contraception, with widespread implications for female health across the lifespan.

Only recently, thymectomy has been shown to reduce new T-cell production and health in adulthood,¹⁶ suggesting that the thymus remains functionally important beyond puberty. A general decrease in T lymphocytes is observed during the menstrual period in humans.¹⁷ Specifically, CD4+ T cells are significantly lower in the midluteal phase compared to the early follicular phase.¹⁸ Studies addressing hormone-dependent changes in thymus size in association with thymocyte release and composition of recent thymic emigrants are warranted. Dynamic changes in thymus size across the menstrual cycle may explain corresponding fluctuations in pro-inflammatory markers,^19,20 via a secondary arc to the cholinergic inflammatory reflex, sensitive to changes in reproductive hormones. Efferent signaling of the vagus nerve innervates with macrophages in the periphery, releasing acetylcholine (ACh) which inhibits the synthesis of pro-inflammatory cytokines.²¹ The cholinergic anti-inflammatory pathway, modulates via α7 nicotinic acetylcholine receptor (α7nAChR) macrophage inflammation by inhibiting HMGB1 and cytokine release.²² A non-neural source of ACh are T cells, of which only CD4+ T cells express choline acetyltransferase (ChAT), responsible for the synthesis of ACh. Splenic-sympathetic fibers are thought to release norepinephrine acting on ChAT CD4+ T cells, to release ACh, inhibiting tumor necrosis factor alpha (TNFα) released in splenic macrophages.²³ ACh thereby attenuates the release of IL-6 and other cytokines.

Decreased availability of ChAT expressing CD4+ T cells following dynamic thymic involution may thus compromise the cholinergic anti-inflammatory reflex and control of low-grade inflammation during phases of hormonal transition, explaining greater risk for inflammatory disorders in females.

Challenging the dominance of a unisex view on physiological function, the suggested mechanism may explain sex disparities in the pathophysiology and response to treatment in a host of diseases. The further discovery of the proposed pathway may inform a new understanding of differential susceptibility to immune-mediated disorders (i.e., autoimmune thyroiditis, rheumatoid arthritis and multiple sclerosis), and could explain increased risk for mood and behavioral changes during phases of hormonal transition in females. Inflammation is a key pathogenetic factor in depression²⁴ with elevated CRP and IL-6 observed among depressed females (not males).²⁵ Both inflammatory markers (IL-6 and CRP) are thought to promote depression characteristic sickness behavior²⁶ via central nervous system circuitry, and are linked to specific symptom profiles, including anhedonia and reduced motivation but also a loss of appetite and decreased concentration.²⁷ Phases of hormonal transition (i.e., puberty, pregnancy, menopause) across the female lifespan are associated with an increased risk of depression onset, and females also show variance in depressed mood and suicidal ideation across the menstrual cycle,²⁸ leading to the formulation of distinct psychiatric entities, including the Premenstrual Syndrome (PMS) and Premenstrual Dysphoric Disorder (PMDD). Dynamic involution of the thymus in the presence of changing reproductive hormones, may be present as contributing pathogenic factor, triggering an inflammatory cascade in female depression.

The present study has several limitations that need to be addressed. qCT imaging is limited in the reliable assessment of thymus size. Our approach to manual tracing showed reasonable inter-rater agreement, dual energy CT imaging with iodinated contrast enhancement or magnetic resonance imaging and the use of neural network models for automated segmentation may lead to more robust results.²⁹ Here we drew on secondary data from a small sample. Future prospective studies are warranted, including more frequent assessments over the menstrual cycle (and other phases of hormonal transition) to study trajectories in the association between sex hormones and thymus size at greater temporal resolution. Further, studies in men may provide important insights to general dynamics in thymus size independent of fluctuating reproductive hormones. Respective studies should further include the assessment of pro-inflammatory markers and lymphocyte composition, to address the functional significance of changes in thymus structure. Further, studies suggest compensatory mechanisms, illustrating an enlarged spleen after thymectomy and vice versa. Assessing spleen function and size alongside dynamics of the thymus during phases of hormonal transition seems warranted.

In conclusion, we present first in-human data, suggesting secondary dynamics to thymus involution, demonstrated here across the menstrual cycle. Findings challenge the prevailing textbook view of singular aging-associated thymus atrophy. The concept of thymus involution in humans is based on cross-sectional (predominantly post-mortem) samples, supporting a static view on thymus size (and function). Advances in neuroimaging technology enable to overcome these long-held beliefs and rethink the nature of the thymus.

Key Points.

Question:

Does thymus size change dynamically across the human menstrual cycle in association with fluctuations in reproductive hormones and hormonal contraception use?

Findings:

In this repeated-measures quantitative computed tomography study of n = 31 healthy women, thymus size demonstrated within-person variation across the menstrual cycle. Greater thymic involution was associated with larger increases in estradiol, whereas increases in thymus size were associated with larger increases in progesterone; thymus size also differed significantly by hormonal contraceptive status.

Meaning:

These findings suggest that thymic involution in humans may demonstrate short-term hormone-dependent dynamics beyond age-related atrophy, with potential implications for sex differences in immune regulation and inflammatory disease.

Funding Statement

Julian Koenig and Holger Winkels acknowledge financial support for the Mapping Autonomic Neural Interaction and Control (MANIAC) Emerging Group by the University of Cologne Excellent Research Support Program. Jessica Sieren acknowledges funding from the National Institutes of Health (NIH R01HL112986 and S10OD018526).

Data Sharing Statement:

Research data are not shared due to legal and privacy issues.