Pregnancy represents a unique immunological landscape characterized by complex and dynamic adaptations in the maternal immune system, essential to sustain maternal health and support fetal development. These adaptations involve intricate changes in both innate and adaptive immune responses to achieve an optimal balance between immune tolerance and defense mechanisms, including alterations in cytokine production, immune cell trafficking, and modulation of immune responses (1–3). The current paradigm supports a pro-inflammatory environment during early and late pregnancy, promoting blastocyst adhesion and tissue remodeling in the uterine wall during implantation, and facilitating labor-associated processes at term. In contrast, mid-pregnancy is regulated by anti-inflammatory mechanisms that are critical for fetal growth and development. Maintaining this delicate equilibrium is of paramount importance, as disruptions in these adaptations are increasingly associated with adverse pregnancy outcomes, including preterm labor, preeclampsia, and intrauterine growth restriction (4, 5). Furthermore, pregnant women may become more susceptible to infection during the anti-inflammatory phase, with potentially devastating consequences for both the mother and fetus (6, 7). Conversely, the course of pre-existing maternal autoimmune or allergic diseases may improve, worsen, or remain unchanged depending on the specific condition, reflecting the complexity of maternal immune adaptations (8, 9).

Studies investigating immune adaptations at the maternal-fetal interface have been particularly valuable in enhancing our understanding of the cellular mechanisms that control maternal immune tolerance (10–13). However, due to ethical considerations and strong evidence of both local and systemic effects of maternal-fetal cross-talk, research has increasingly focused on the systemic immune adaptations experienced by women during pregnancy (3, 14, 15). For instance, although the extent and dynamics of the Th1/Th2 balance remain context-dependent and not reproducible across studies (14), the shifts observed in these subsets are pivotal in orchestrating gestational immune adaptations and are modulated from early pregnancy onward (16). Similarly, the role of regulatory T cells (Tregs) is well established, although variations in their levels and functions remain an active area of research (17).

A reduction in circulating B cells and specific B-cell subsets during pregnancy has also been reported in several studies (15, 18–21), often accompanied by decreased responsiveness to mitogens and infectious agents (22), although data from early pregnancy remains limited. As for B cells with regulatory functions (Bregs), studies have proposed a significant role of these cells in maintaining fetal tolerance during the initial stages of pregnancy. Nevertheless, the absence of specific Breg markers and the phenotypic heterogeneity used to define these cells across studies hinder the ability to draw definitive conclusions (23–25).

Coordinated interactions among diverse immune cell types are essential for the development of protective immunity. Follicular helper T cells (Tfh) are cornerstone in this process as they deliver cognate and soluble signals that drive B cell proliferation, survival, affinity maturation, and differentiation into antibody-producing cells and long-lived memory B cells (26). While Tfh cells are classically found in secondary lymphoid organs, a smaller population of phenotypically similar cells can also be found in the periphery (circulating Tfh, cTfh). These cells are thought to represent a memory pool of Tfh cells in humans, expressing CXCR5 and sharing functional similarities to their lymphoid-resident counterparts (27). Given their role in regulating B cell responses, cTfh cells have been widely studied in autoimmune diseases (28). However, their role in pregnancy and the regulation of B cell responses in this context remains poorly understood. Further characterization of follicular CD4 and CD8 T cell subsets, using additional activation and maturation markers, is necessary to clarify the dynamics of follicular T cells during pregnancy.

Despite the growing body of data, much remains to be understood regarding changes in adaptive immune responses occurring throughout normal physiological pregnancy, particularly concerning less-studied subsets. Establishing appropriate reference ranges for these immunological fluctuations in pregnancy is crucial for distinguishing physiological adaptations from pathological deviations. A precise understanding of these changes is an essential first step toward identifying immunological deviations associated with pregnancy-related complications and enhancing maternal care and monitoring strategies.

To facilitate the establishment of such benchmarks and enable the early identification of immune-related pregnancy complications, in this study, we utilized flow cytometry to provide a comprehensive longitudinal analysis of less-studied immune cell subsets. Our focus was on B cells and both follicular and non-follicular activated T cell subsets across trimesters in healthy pregnant and non-pregnant women.

Advances in cytomics and single-cell transcriptomics have significantly enhanced the ability to capture the complexities of systemic immunological adaptations during pregnancy (15, 37, 38). However, there remains a critical need to monitor immune dynamics longitudinally within the same individual using accessible biological compartments. Such approaches must yield rapid, interpretable, and clinically actionable insights, capabilities well supported by flow cytometry-based routine immunophenotyping.

This study presents a high-dimensional analysis characterizing the phenotypes and functions of peripheral immune T and B cell populations in maternal peripheral blood samples assessed longitudinally throughout normal pregnancy and compared to non-pregnant women. We validate previously reported immunological changes during pregnancy and provide novel insights into alterations in maternal peripheral immunity. Additionally, we propose preliminary reference intervals for key immune cell subsets, offering a potential framework for monitoring pregnancy progression and identifying early deviations that may precede pregnancy complications.

Significant changes in leukocyte subpopulations were observed throughout pregnancy, with a progressive increase in neutrophil and monocyte concentrations beginning in the first trimester, while lymphocyte levels remained stable and consistently lower than those in non-pregnant women. These findings are consistent with the well-established phenomenon of physiological leukocytosis during pregnancy (14, 39). The early post-implantation of the first trimester is characterized by sustained levels of inflammatory cytokines and growth factors, such as Granulocyte and Granulocyte-Macrophage Colony Stimulating Factors (G-CSF and GM-CSF, respectively), which drive neutrophil and monocyte production in the bone marrow (3, 40, 41). Additionally, pregnancy-associated hormones such as estrogen and progesterone may also contribute to these alterations by extending the neutrophil lifespan through inhibition of apoptosis (42, 43). This innate-driven inflammation plays a key role in supporting wound healing, post-implantation recovery, placentation, and uterine remodeling (44). Importantly, although peripheral neutrophils and monocytes increase dramatically in numbers, they are functionally reprogrammed to promote immune tolerance (45, 46). Neutrophils, while more abundant, show reduced pro-inflammatory activity, produce immunoregulatory mediators that inhibit T cell responses, while releasing pro-angiogenic factors at the decidua to support placental development. Pregnancy hormones and placental factors further skew neutrophils toward an anti-inflammatory state and enable crosstalk with adaptive immunity, promoting the induction of Treg cells secreting IL-10 and VEGF (47). Circulating monocytes are similarly reprogrammed by placental signals (46). First-trimester placental exosomes convert monocytes into an M2-like state with high PD-L1 expression, leading to suppressed CD4⁺/CD8⁺ T-cell proliferation and expanded Tregs (48). In parallel, trophoblast microparticles elicit moderate monocyte secretion of TNF-α, IL-6, and IL-8, contributing to a controlled inflammatory milieu (46). Throughout gestation, monocytes undergo epigenetic remodeling and subset shifts, more significant in intermediate CD14⁺CD16⁺ cells, that heighten their activation potential by term (46). Together, these innate adaptations help establish maternal-fetal tolerance, while their dysregulation is linked to complications such as preeclampsia and miscarriage.

Observed alterations in erythrogram components and the progressive decline in platelet counts are well-documented physiological adaptations in pregnancy, often attributed to increased platelet turnover and hemodilution (49).

Of the 74 lymphocyte subsets analyzed in this study, 24 subsets exhibited significant changes across pregnancy timepoints, while 23 subsets differed significantly when compared to the non-pregnant group. Peripheral blood B cell levels varied throughout normal pregnancy and differed from those in non-pregnant women. A notable decrease in total circulating B cells was particularly evident during the third trimester, a finding that has been widely reported in previous studies (19, 50, 51).

Previous studies have linked these changes in the B cell compartment to endocrine factors such as human chorionic gonadotropin (hCG), which modifies B cell differentiation and function (52, 53). Muzzio et al. also demonstrated that B lymphopoiesis in mice decreases in late pregnancy, coinciding with elevated estradiol levels (54). Another likely mechanism contributing to reduced peripheral B cell levels is their migration and retention at the maternal-fetal interface, a hypothesis supported by studies highlighting the critical role of B cells in this region (55).

Moreover, most B cell subsets exhibit decreased absolute counts during pregnancy (14), with transitional B cells showing particularly pronounced reductions in late pregnancy (21, 51, 56). However, no significant differences were observed in the distribution of naïve and memory B cell subsets across study visits. Nonetheless, variations in the proportions of major B cell subsets have been inconsistently reported in studies, likely due to distinct phenotypic and methodological approaches.

Plasmablasts decreased from the first to the second trimester but did not differ significantly from levels in non-pregnant women, contrary to previous reports (15, 51). However, naïve and memory B cell populations showed distinct activation dynamics: anergic naïve B cell proportions showed a continuous decline throughout pregnancy, while activated naïve and switched memory B cells progressively increased from early to late gestation. A study by Demery-Poulos et al. found a greater proportion of B cells in pregnant women displaying memory-like and activated phenotypes (57) whereas Apps et al. reported opposite trends in these subsets (21). Additionally, a recent study reported downregulation of genes associated with signal transduction pathways, B cell activation, and immune responses in B cells (memory, naïve, and plasmablasts) throughout gestation and in comparison to the non-pregnant state (15). Similar findings were observed in total splenic B cells from pregnant and non-pregnant mice (58). Conversely, Demery-Poulos et al. found a significantly greater fold change in B cell activation following anti-human IgM/IgG stimulation in cells isolated from pregnant compared to non-pregnant women, suggesting pregnancy-enhanced circulating B cell responsiveness (57). Supporting this idea, higher levels of serum B-cell activating factor (BAFF) have been reported in pregnant women compared to non-pregnant, suggesting that BAFF may prime B cells and contribute to the pregnancy-specific increase in activation observed in these cells (59). In contrast to these studies, we found no significant changes in the B-cell activation status, as assessed by CD69 expression, at any gestational age, either at baseline or after stimulation. Still, our assessment only focused on the total B cell compartment and did not discriminate between different subsets within it.

PD-L1 expression on B cells defines a mechanism for potent suppression of humoral immunity by attenuating T cell activation and antibody production (60). In our study, PD-L1⁺ B cell levels decreased from early to mid-pregnancy and rose again in late pregnancy, with levels remaining elevated during the early and late stages compared to non-pregnant individuals. These dynamic fluctuations are consistent with existing literature on immune modulation during pregnancy and suggest a temporally regulated role for PD-L1⁺ B cells in maintaining immune balance across gestation (61, 62). For instance, a study by Mach et al. investigated soluble PD-L1 (sPD-L1) serum levels in uncomplicated pregnancies and found a significant increase throughout gestation, with the highest concentrations in the third trimester (63). Similar results were reported by Hedman et al. (64) and Okuyama et al. (65). Since PD-L1 is highly expressed by syncytiotrophoblast and its expression increases with gestational age, the placenta may serve as an important source of circulating PD-L1 (66). While these studies primarily focused on sPD-L1, they support the concept of dynamic PD-L1 expression during pregnancy, which may reflect changes in basal PD-L1⁺ B cell populations.

A recent study reported negative correlations between PD-L1⁺ B cells and various subsets of Tregs, i.e., subsets expressing markers that influence immunosuppressive function in early pregnant women (62). In fact, elevated PD-L1 expression in regulatory B cells may suppress inflammation by interacting with PD-1 expressed on T cells and Treg cells, thus delivering inhibitory signals that mediate immune suppression (67, 68). The increased levels observed during early pregnancy may facilitate the expansion of Tregs and a shift toward an immunotolerance state during mid-pregnancy. In the third trimester, increased PD-L1 expression on B cells can modulate immune responses, to ensure they do not adversely affect the fetus while facilitating the onset of labor. Interestingly, a study demonstrated that CpG-induced PD-L1 expression on human B cells can suppress Th2 cytokine production by CD4 T cells stimulated with pollen antigens, while increasing INF-γ and IL-12 secretion (69). In our study, the lower levels of PD-L1⁺ B cells observed during the second trimester might be related to the well-documented immune shift towards a Th2-dominant state during mid-pregnancy (14). Conversely, the increased levels of PD-L1+ B cells in late pregnancy may afterwards facilitate Th1 cytokine responses in this later stage.

However, it is important to note that immune associations observed in peripheral blood and circulating subsets may not directly reflect those in decidual tissues. Further research is, thus, needed to better characterize the PD-L1 expression on B cells throughout the full course of pregnancy, also locally, given the importance of the PD-1/PD-L1 axis at the maternal-fetal interface (61).

Previous evidence suggests that human circulating transitional B cells are enriched in regulatory B cells (70), and somehow exhibit regulatory properties (71). In our study, we observed a decline in CD24hiCD38^hi T1/T2 transitional B cells beginning in the first trimester, consistent with prior reports (21, 51, 56). This reduction may be partly explained by maternal circulating hCG levels, which have been implicated in regulating both the number and function of Bregs (72). However, no definitive set of surface markers currently exists to identify Bregs, and we align with recommendations suggesting that accurate enumeration of Bregs should rely on functional characterization, specifically, single-cell cytokine production measurements, through intracellular staining for IL-10 and IL-35 (73, 74). This functional characterization is often overlooked in many studies. We aim to complement and validate the present findings through additional functional studies, including the evaluation of cytokine secretion by both B and T cells.

T cell responses are generally believed to be suppressed during pregnancy, although the exact mechanism remains unclear. We observed several dynamic changes across various T cell subsets, including Tfh cells, which are known to mediate adaptive immune responses in various human autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus, and multiple sclerosis (28). In contrast, the role of Tfh cells during pregnancy remains poorly defined, with most existing studies being conducted in mice. Limited research suggests that Tfh cells may support successful allogeneic pregnancy in mice (75), while in humans, a study by Monteiro and collaborators reported that human pregnancy favors Tfh cell expansion, potentially driven by elevated estrogen levels, which enhance Tfh cell differentiation and promote humoral immunity while balancing Th1/Th2 responses (76).

In our study, although no significant changes in cTfh cell levels were observed along pregnancy, the pregnant group consistently showed lower cTfh cell counts in all trimesters compared to non-pregnant women. Additionally, a significant decrease in cTfh cell percentages was noted in the third trimester. We also provide further insight into Tfh cell dynamics during pregnancy, demonstrating that both the frequency and absolute counts of activated cTfh cells expressing PD-1, CD38, and HLA-DR were reduced in pregnant women compared to non-pregnant controls. Similar findings were observed by Fröhlich et al., who characterized circulating mice CXCR5⁺PD-1⁺ICOS⁺ Tfh cells and found a decrease during the first trimester (77). They also reported lower levels of CXCR5⁺Bcl-6⁺ Tfh cells in normal pregnant mice compared to both non-pregnant and mice with disturbed pregnancies (77). Interestingly, a study by Zeng et al. using an allogeneic pregnancy mouse model found that uterine and placental CD4 T cells exhibited a Tfh-like phenotype. Specifically, CXCR5^hiPD-1^hi and CXCR5hiICOS^hi Tfh cells with a memory/activation phenotype peaked at mid-pregnancy and were abundantly located in the uterus, with a marked increase in the placenta by late pregnancy (75). This tissue-specific redistribution of Tfh-like cells may help explain the lower circulating Tfh cell levels observed in our study. In addition, the hormonal effect in these populations still needs more consistent functional approaches to fully detail and ascertain the real impact and how the follicular functions can be modulated. In future studies, we aim to evaluate cytokine secretion by Tfh cells after stimulation to complete the current analysis.

Follicular CD8 T cells (cTfc) are a more controversial subset, poorly described, particularly in peripheric compartments, but they might play a critical role in cytokine signaling and cell-cell interactions, modulating both Tfh and B cells in humoral responses (78). To the best of our knowledge, there are no reports on this population during pregnancy. Thus, for the first time, we report the progressive decline in cTfc levels in pregnancy, with consistently lower levels across all trimesters compared to non-pregnant women. These findings suggest that both Tfh and, particularly, Tfc populations are dynamically altered during pregnancy, providing new insights into immune adaptations beyond the traditionally studied Th cell responses.

In contrast, we identified increasing proportions of non-follicular CD4 and CD8 T cell subsets expressing both activation and maturation markers throughout pregnancy, particularly PD-1⁺CD38^hiHLA-DR⁺ and PD-1-CD38^hiHLA-DR⁺ non-Tfc subsets. In accordance with these activated phenotypes, we observed a pregnancy-specific increase in the basal proportion of CD69⁺PD-L1⁻ and CD69⁺PD-L1⁺activated peripheral CD4 and CD8 T cells. A study in C57BL/6 mice reported elevated baseline levels of CD69⁺ CD4 T cells (but not CD8) in late pregnancy compared to non-pregnant mice (79). Similarly, Demery-Poulos et al. observed increased proportions of activated CD69⁺ peripheral CD4 T cells in third-trimester pregnant women (57). Given the presence of fetal antigens in maternal circulation, it is plausible that the increased activation of T cells in pregnancy results from repeated antigenic exposure (80, 81). Moreover, cell-free fetal DNA levels in maternal circulation rise in late pregnancy, coinciding with a pro-inflammatory shift before labor (82). Phenotyping and omics studies have provided further evidence of T cell activation during labor (83, 84), with increased activation marker expression linked to T cell responses in late pregnancy (85).

Taken together, these findings suggest that pregnancy is characterized by a gradual increase in basal activation of peripheral CD4⁺ and CD8⁺ T cells, which intensifies as labor approaches. Although no studies have directly examined these subsets in pregnancy-related pathologies, an increased CD69 expression on peripheral T cells has been reported in patients with a history of recurrent spontaneous miscarriage (86, 87). Both basal and stimulated CD69 expression were higher in women who experienced miscarriage compared to those with normal pregnancies (87). Additionally, in vivo activation of T cells with an anti-CD3ϵ antibody during mid-to-late pregnancy has been shown to trigger systemic inflammation, preterm labor, and birth (88), suggesting that excessive maternal T cell activation can be detrimental to fetal health. The increased proportions of activated CD8 T cells observed in our clinical dropouts may reflect these consequences. Therefore, this study establishes an important reference for future research into immune patterns during healthy pregnancy and underscores the need for further investigation in pathological cohorts.

Indeed, regular assessment of hematological and immunological profiles is a vital and standard medical practice for evaluating health status. Normal reference intervals (RIs) for these parameters in healthy populations serve as critical cutoff points for clinical decision-making. Deviations from these reference ranges can indicate disease and are essential for diagnosis and patient monitoring, particularly in the era of evidence-based medicine. Establishing baseline immunological reference ranges in healthy pregnant women is crucial for predicting and improving pregnancy outcomes during antenatal care.

In this study, we present preliminary RIs for key lymphocyte subsets, particularly those that exhibited significant changes throughout gestation compared to the non-pregnant group. To our knowledge, this is the first report to provide reference ranges for these specific immune populations throughout the entire course of pregnancy. While previous studies have established reference values for major leukocyte populations, including lymphocytes and some subsets (89, 90), more specific immune populations have remained largely uncharacterized, leaving a critical gap in the field. Given the growing body of evidence linking immune dysregulation during pregnancy with adverse outcomes (91), monitoring immune subsets whose deviations may signal impending complications is of clear clinical relevance. Although this study was not initially designed to assess the applicability of these RIs to pregnancy complications, we did observe immune deviations in participants who discontinued the study due to pregnancy-related complications. Notably, each of these individuals exhibited at least one deviation from the established RIs, suggesting a potential foundation for future evaluation of these populations in the context of pregnancy complications. Interestingly, deviations in the second-trimester evaluations were characterized by elevated levels, whereas those in the first-trimester evaluation showed reduced levels. While these findings are noteworthy, they are not statistically significant and should not be generalized to the broader population. Further validation of these reference intervals in future larger cohort studies is necessary to confirm their clinical relevance.

Additionally, we found evidence that certain immune populations altered during gestation also varied significantly with maternal age, parity, history of miscarriage, and BMI. In contrast, a previous study with a smaller sample size found no association between maternal characteristics and immune dynamics (21). Due to the limited number of participants in our study, we are unable to establish robust reference intervals stratified by demographic and clinical variables. Nonetheless, our findings highlight the importance of larger cohort studies to refine these reference intervals and to assess their clinical relevance in pregnant populations.

We acknowledge some limitations in our study. First, while our total sample size (n=50) is comparable to that of other studies in literature, it limited our ability to conduct detailed subgroup analyses, thereby reducing the statistical robustness of some stratified comparisons. Additionally, incomplete data collection during the second and third trimesters may have impacted the completeness of the longitudinal dataset. While our cohort included only women who delivered at term, participants varied in demographic and clinical characteristics, which may have influenced immune characteristics and contributed to interindividual variability. Nonetheless, we successfully captured chronological immune adaptations throughout term pregnancies. Another limitation is the recruitment of participants from a single private healthcare center, which may restrict the generalizability of our findings. Future research should investigate how ethnic, racial, and socioeconomic factors influence immune adaptations during pregnancy. Regarding the establishment of reference interval, although we did not meet the recommended minimum sample size (n=120), we employed the robust statistical method recommended by the CLSI for smaller datasets. While validated, this approach would benefit from replication in larger cohorts to enhance the clinical utility of the proposed reference intervals. Finally, future studies incorporating more representative cohorts and postpartum follow-up will be essential to confirm the reproducibility of our findings and to further elucidate immune adaptations during and after pregnancy.

Despite these limitations, our study presents several strengths. Unlike many similar studies, we conducted a longitudinal follow-up of the same participants across defined gestational timepoints, thereby reducing cohort heterogeneity and enhancing internal validity. In addition to analyzing immune cell population frequencies and absolute cell counts, we also assessed cellular functional states and responsiveness, an aspect often overlooked in comparable research. Furthermore, our study is among the few to perform a longitudinal analysis of immune adaptations during pregnancy using a well-defined and pre-established timeline and a standardized, optimized methodology. Another key strength is our inclusion of effect size calculation, which enhances both the interpretability and reliability of our findings. Unlike p-values, which only indicate statistical significance, effect sizes quantify the magnitude of differences or associations, offering deeper insight into their biological and clinical relevance (35, 92). This is particularly important in studies with smaller sample sizes, where statistically significant results may not always reflect meaningful effects. Additionally, effect size calculations facilitate comparisons with previous research, allowing for better contextualization within the broader scientific literature. By incorporating these analyses, we strengthen the interpretative value of our results and increase the study’s potential impact in the field.

This study provides a detailed characterization of immune changes throughout gestation, offering novel insights into the immune subsets influenced by pregnancy and their functional capacities. These findings contribute to a more comprehensive understanding of systemic maternal immune adaptations during pregnancy. In addition, we present pregnancy-specific reference intervals for key immune cell subsets, which may serve as a valuable resource for both future research and potential clinical applications. Notably, the immune deviations observed in participants who experienced pregnancy complications underscore the importance of establishing additional reference values and highlight the need for validation in larger, independent cohorts.