Reduced peripheral natural killer cell counts may be a predictive marker for cesarean scar pregnancy Authors: Ye Shen, Jinqiu Zhang, Hua Fang, Li Zhang, Qi Jiang, Hua Gong, Qi Chen, Min Zhao Published: 2026 ABSTRACT AbstractCesarean scar pregnancy (CSP), a condition characterised by inflammation, occurs in 0.2% of women with a history of cesarean section(s), and is clinically categorised by type 1 and type 2 according to the ultrasound findings. Recent studies reported that systemic inflammation indices derived from routine whole-blood tests may serve as additional clinical predictive markers for early CSP detection, with lymphocyte counts showing a reduction. However, the specific lymphocyte populations contributing to this reduction remain unclear. This study included 72 women diagnosed with CSP and 50 gestation-matched controls. The subsets of peripheral lymphocytes, including CD45 lymphocytes/monocytes, CD3 T cells, CD4 T helper cells, CD8 cytotoxic T cells, CD19 B cells, total NK cells and NKT-like cells, were measured. We found a significant reduction in peripheral total NK cells and CD8 cytotoxic T cells in CSP women compared to controls. However, there were no differences in peripheral immune cell profiles between type 1 and type 2 CSP. Using total NK cell counts with a cut-off value of 210/µl, the area under the curve (AUC) was 0.698, with a sensitivity of 64%. Our findings suggest that the reduction in peripheral total NK cells and CD8 T cytotoxic cells may play a role in the development of CSP. Additionally, peripheral total NK cell counts, derived from routine whole-blood tests, could serve as a predictive marker for early CSP diagnosis.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12865-026-00841-9.Keywords: Cesarean scar pregnancy, Total NK cells, CD8 cytotoxic T cells, Routine blood test Introduction Cesarean scar pregnancy (CSP) occurs when an embryo partially or fully implants in the scar tissue from a previous cesarean section, affecting 1 in 530 women with a history of cesarean Sect [1]. CSP is associated with significant risks, including severe haemorrhage and uterine rupture if the pregnancy continues [2, 3]. While the mechanism of underlying CSP currently remains unclear, cesarean scar tissue frequently presents with inflammation, suggesting that CSP is characterized by an inflammatory process [4]. Increased levels of IL-1α, IL-1β, IL-6, IL-8, and TNF-α are reported in women diagnosed with cesarean scar pregnancy [5]. Clinically, CSP is traditionally classified into two subtypes, type 1 and type 2, based on ultrasound findings of the site of the gestation sac [6–8]. The two subtypes of CSP could manifest in different inflammation conditions. Additionally, multiple cesarean sections may exacerbate the process of inflammation by creating scar defects, potentially leading to a chronic inflammatory condition [9]. Recent increasing evidence has highlighted that, in addition to the gold standard diagnostic modalities of ultrasound and magnetic resonance imaging, systemic inflammation indices generated from routine whole blood tests could serve as potential additional clinical predictive markers for the early detection of CSP [10, 11]. These studies have consistently reported lower lymphocyte counts in routine peripheral blood tests in women diagnosed with CSP, compared to those with normal pregnancies [10, 11]. Lymphocytes primarily include T cells, B cells, and natural killer (NK) cells. However, these studies did not examine specific lymphocyte subsets, leaving the question of whether the observed reductions in lymphocyte counts are attributable to changes in specific lymphocyte populations. Lymphocytes are key regulators of inflammation [12], and reduced lymphocyte counts may reflect an increased risk of infection or immune dysregulation [13, 14]. Lymphocytes facilitate adaptive immunity, and each lymphocyte subtype contributes uniquely to immune responses. T cells mediate cellular immune responses and are key regulators of inflammation by releasing cytokines that activate other immune cells. The key subsets of T cells are CD4 + helper cells, which initiate the response, and CD8 + cytotoxic cells, which are responsible for identifying and eliminating infected cells [15]. B cells function in both initiating and regulating inflammation through antibody production, the release of pro- or anti-inflammatory cytokines, and the activation of T cells, such as CD4 + T cells [16]. NK cells participate in innate immunity, acting as important and dual-functional regulators of inflammation and promoting and resolving immune responses. In general, NK cells are rapidly recruited to inflammatory sites, then release pro-inflammatory cytokines such as TNF-alpha, and exert cytotoxicity [17]. On the other hand, NK cells can also produce anti-inflammatory cytokines, such as IL-10, to prevent excessive immune reactions [17]. Additionally, both NK cells and T cells are crucial for maintaining immune balance to establish maternal-fetal immune tolerance during early pregnancy, particularly during implantation. Given the inflammation characteristics reported in CSP and the unique inflammatory environment of early pregnancy, it remains unclear whether the reduced peripheral lymphocyte counts observed in CSP reflect changes in specific lymphocyte subsets. Therefore, we undertook this study to investigate further the association of lymphocyte subsets, including T cells and their subsets, B cells, and NK cells, with CSP. The findings may contribute to understanding the early prediction and pathogenesis of CSP. Methods and materials The Ethics Committee of Wuxi Maternity and Child Health Hospital approved this study (reference number: 2024-06-1024-55). Participants provided broad informed consent for the use of blood and placental tissue samples for immunological analyses, as approved by the hospital ethics committee. This study was performed in accordance with the Medical Association Declaration of Helsinki. Study population This study involved 72 women diagnosed with CSP at 7–9 weeks of gestation who received optional treatments at our Family Planning Clinic at Wuxi Maternity and Child Health Hospital between December 2023 and August 2024. The diagnosis of CSP was based on the transvaginal ultrasound findings, including the presence of a gestational sac in the area of the scar tissue, with or without fetal cardiac activity on Doppler flow ultrasonogram (Pillips Affiniti-70 W), in addition to a history of prior cesarean section and a positive pregnancy test [7, 18]. A control group of 50 gestation-matched pregnant women with no history of complicated pregnancies was included as a control group. All participants in this study were Han Chinese. None of the participants smoked or consumed alcohol during conception preparation or confirmed pregnancy. Data on the pregnancy history, body mass index (BMI), and CSP subsets were collected from the hospital’s electronic databases. Type 1 CSP is defined as the embryo implanted in the myometrium and grows toward the cervico-isthmic space or uterine cavity. Type 2 CSP is defined as the embryo implanted deeply within the scar tissue and penetrates the uterine wall, sometimes reaching the uterine blood vessels, as determined by ultrasound [6–8]. Peripheral blood tests of the immune cells Whole blood samples (3 ml) were collected from a vein at the time of CSP diagnosis, prior to treatment, using an EDTA tube. The subsets of immune cells were measured using the NovoCyte D2060R Flow Cytometer System (Agilent, California, USA), with 50 µL of whole blood used for each measurement. The subsets of immune cells included CD45 + lymphocytes, CD3 + T cells, CD3 + CD4+ T helper cells, CD3 + CD8+ cytotoxic T cells, CD3-CD19 + B cells, total NK cells (CD3-CD16/CD56+), and NKT-like cells (CD3 + CD16/56+). A commercial available kit for multicolour flow cytometry with a fixed antibody panel was purchased from Agilent (California, USA; catalogue number 8920251). Details of antibodies are summarised in Supplementary Table 1. Data analysis, including all gating procedures and reporting of immune cell subsets, was performed according to the manufacturer’s instructions. The results of these tests were reported as percentages and counts per microliter (µL) of the sample. CD45 + expression was used to confirm leukocyte purity. Lymphocytes were then identified within the CD45 + population based on low side‑scatter characteristics and gated accordingly. All subset percentages were calculated within this gated lymphocyte population (lymphocyte gate = 100%). Within this lymphocyte gate, CD3 + T cells and CD19 + B cells were expressed as percentages of total lymphocytes. Total NK cells and NKT-like cells were also expressed as a percentage of the gated lymphocytes. CD4 + T helper cells and CD8 + cytotoxic T cells were expressed as a percentage of the total CD3 + T cell population. Absolute counts of CD3 + T cells, CD4 + T helper cells, CD8 + cytotoxic T cells, total NK cells, NKT-like cells and CD19 + B cells were calculated by multiplying the absolute count of CD45 + lymphocytes by the percentage of each corresponding subset obtained from flow cytometry. Representative flow cytometry plots are presented in Fig. 1. Representative flow cytometry plots Statistical analysis Absolute counts or percentages of immune cells measured in peripheral blood tests were reported as median and interquartile range (IQR), or as mean and standard deviation (SD). The statistical differences in maternal age, gestational age, and BMI between CSP and controls were assessed using a t-test (Mann-Whitney), and the statistical differences in parity and gravidity between CSP and controls were assessed using a chi-square test. Additionally, the statistical differences in the percentages or absolute counts of CD19 + B cells, CD45 + lymphocytes, total NK cells, CD4 + T helper cells, CD8 + cytotoxic T cells, NKT-like cells, and CD3 + T cells between CSP and controls, or between type 1 and type 2 CSP, were assessed using a t-test (Mann-Whitney). Additionally, Pearson correlation analysis was performed. All statistical analyses were performed by GraphPad Prism (version 10.1.2). The area under the curve (AUC) was calculated for predictive analysis using total NK cell and CD8 + cytotoxic T cell counts on CSP using IBM SPSS (version 29.0). using A p-value less than 0.05 was considered statistically significant. Results The clinical parameters of women included in this study are summarized in Table 1. The mean maternal age at the time of CSP diagnosis was 36 years (± 4), significantly higher than that of the controls (29 ± 4, p < 0.001). There were no statistical differences in the gestational age and BMI between the two groups (p > 0.05). In contrast, higher parity was observed in CSP women. Among the 72 women with CSP, 42 (58%) were diagnosed with type 1 CSP. Table 1Clinical characteristics of the study cohort. Baseline demographics, including maternal age, gestational age, parity, and BMI and the subsets of CSPCSP (n = 72)Controls (n = 50)P value Maternal age (years, mean/SD)36 ± 429 ± 6< 0.001Gestational age (days, mean/SD)50 ± 1253 ± 18p > 0.05BMI (Kg/m2, mean/SD)23.15 ± 322.25 ± 4p > 0.05Parity 0 (n, %)0 (0%)41 (83%)< 0.0001 1 (n, %)37 (51%)9 (17%) ≥ 2 (n, %)35 (49%)0 (0%)Gravidity 1 (n, %)10 (14%)38 (76%)< 0.0001 2 (n, %)11 (15%)8 (16%) ≥ 3 (n, %)51 (71%)4 (8%)Subsets of CSP Type 142 (58%)N/AN/A Type 230 (42%)Abbreviations: BMI Body mass index, CSP Cesarean scar pregnancyStatistical analysis: P values were calculated using the Mann–Whitney test for continuous variables (maternal age, gestational age, and BMI) and the chi‑square test for categorical variables (parity, gravidity, and CSP subsets) Clinical characteristics of the study cohort. Baseline demographics, including maternal age, gestational age, parity, and BMI and the subsets of CSP Abbreviations: BMI Body mass index, CSP Cesarean scar pregnancy Statistical analysis: P values were calculated using the Mann–Whitney test for continuous variables (maternal age, gestational age, and BMI) and the chi‑square test for categorical variables (parity, gravidity, and CSP subsets) The comparison of peripheral immune cell profiles between women diagnosed with CSP and controls is presented in Table 2. There was no statistical difference in the percentage of CD19 + B cells, CD45 + lymphocytes, CD8 + cytotoxic T cells, or CD3 + T cells or NKT-like cells between the two groups. However, there was a significant decrease in the percentage of total NK cells in CSP women, compared to controls (Table 2, p = 0.0001), as well as a significant increase in the percentage of CD4 + T helper cells (Table 2, p = 0.0004). Table 2The comparison of the profiles of immune cells, including the percentage and total counts of CD19 + B cells, total NK cells, NKT-like cells, CD4 + T helper cells, CD45 + lymphocytes, CD8 + cytotoxic T cells and CD3 + T cells, between women with and without CSPBlood test parametersCSP Women (n = 72)Controls (n = 50)P valuePercentage of immune cells CD 19 + B cells (%, mean/SD)13 ± 3.4%13 ± 2.9%0.906 Total NK cells (%, mean/SD)12 ± 6.7%15 ± 5.5% 0.0001 NKT-like cells (%, mean/SD)6.4 ± 3.4%7.0 ± 4.1%0.4023 CD4 + T helper cells (%, mean/SD)39 ± 6.6%34 ± 6.7% 0.0004 CD45 + lymphocytes (%, mean/SD)25 ± 9.6%25 ± 6.3%0.5690 CD8 + cytotoxic T cells (%, mean/SD)25 ± 6.8%27 ± 5 0.6%0.1042 CD3 + T cells (%, mean/SD)69 ± 7.4%68 ± 6.1%0.1933Counts of immune cells CD19 + B cells (/µl, median/IQR)186 (147, 259)213 (176, 271)0.0787 Total NK cells (/µl, median/IQR)175 (120, 240)245 (186, 346) 0.0002 NKT-like cells (/µl, median/IQR)93 (61, 166)101 (71, 157)0.6821 CD4 + T helper cells (/µl, median/IQR)609 (429, 772)591 (475, 683)0.6431 CD45 + lymphocytes (/µl, median/IQR)1557 (1241, 1950)1680 (1487, 1946)0.0796 CD8 + cytotoxic T cells (/µl, median/IQR)371 (272, 559)459 (369, 557) 0.0203 CD3 + T cells (/µl, median/IQR)1042 (831, 1440)1156 (1007, 1287)0.1362Abbreviations: IQR Interquartile range The comparison of the profiles of immune cells, including the percentage and total counts of CD19 + B cells, total NK cells, NKT-like cells, CD4 + T helper cells, CD45 + lymphocytes, CD8 + cytotoxic T cells and CD3 + T cells, between women with and without CSP Abbreviations: IQR Interquartile range Additionally, there was no statistical difference in total counts of CD19 + B cells, CD45 lymphocytes, CD4 + T helper cells, NKT-like cells and CD3 + T cells between the two groups. In contrast, total NK cells and CD8 + cytotoxic T cells were significantly decreased in CSP women compared to controls (Table 2, p = 0.0002, or p = 0.0203). The AUC was 0.698 (95% CI: 0.605, 0.791, p < 0.001) when using the peripheral counts of total NK cells as a predictor (Fig. 2). The cut-off count for total NK cells was 210/µl with 64% sensitivity and 66% specificity. Additionally, the AUC was 0.623 (95% CI: 0.524, 0.723; p = 0.015) when peripheral CD8 + cytotoxic T cell counts were used as a predictor. The cut-off count for cytotoxic CD8 + cytotoxic T cells was 454/µl, with 54% sensitivity and 60% specificity. Receiver operating characteristic (ROC) curve showing the performance of peripheral CD16 + CD56 natural killer (NK) cell counts (absolute counts) for predicting cesarean scar pregnancy (CSP). The curve displays sensitivity versus 1- specificity across all possible thresholds. The area under the curve (AUC) was 0.698, indicating moderate discrimination. Data represent absolute cell counts measured per microliter of peripheral blood CSP is clinically divided into two subsets: type 1 and type 2 [6–8]. The comparison of the immune cell profiles between women diagnosed with type 1 and type 2 CSP is summarised in Table 3. There was no statistical difference in the percentages of CD19 + B cells, CD45 + lymphocytes, total NK cells, NKT-like cells, CD4 + T helper cells, CD8 + cytotoxic T cells, and CD3 + T cells between the two CSP types. Additionally, there were no statistical differences in the counts of CD19 + B cells, CD45 + lymphocytes, total NK cells, NKT-like cells, CD4 + T helper cells, CD8 + cytotoxic T cells, and CD3 + T cells between the two CSP types. Table 3The comparison of the counts of peripheral CD19 + B cells, total NK cells, NKT-like cells, CD4 + T helper cells, CD45 + lymphocytes, CD8 + cytotoxic T cells and CD3 + T cells between women with type 1 and type 2 CSPBlood test parametersType 1 CSP (n = 42)Type 2 CSP (n = 30)P valuePercentage of immune cells CD 19 + B cells (%, mean/SD)13 ± 4.9%13 ± 3.8%0.428 Total NK cells (%, mean/SD)12 ± 5.7%12 ± 8.0%0.329 NKT-like cells (%, mean/SD)5.7 ± 3.0%7.3 ± 3.8%0.083 CD4 + T helper cells (%, mean/SD)38 ± 6.9%38 ± 6.1%0.693 CD45 + lymphocytes (%, mean/SD)25 ± 10%25 ± 8.6%0.778 CD8 + cytotoxic T cells (%, mean/SD)25 ± 7.1%25 ± 6.7%0.705 CD3 + T cells (%, mean/SD)69 ± 7.2%70 ± 7.7%0.276Counts of immune cells CD19 + B cells (/µl, median/IQR)186 (147, 259)213 (176, 271)0.2836 Total NK cells (/µl, median/IQR)175 (120, 240)245 (186, 346)0.8403 NKT-like cells (/µl, median/IQR)89 (68, 353)110 (78, 297)0.118 CD4 + T helper cells (/µl, median/IQR)609 (429, 772)591 (475, 683)0.2587 CD45 + lymphocytes (/µl, median/IQR)1557 (1241, 1950)1680 (1487, 1946)0.2875 CD8 + cytotoxic T cells (/µl, median/IQR)371 (272, 559)459 (369, 557)0.5118 CD3 + T cells (/µl, median/IQR)1042 (831, 1440)1156 (1007, 1287)0.277Abbreviations: IQR: Interquartile range The comparison of the counts of peripheral CD19 + B cells, total NK cells, NKT-like cells, CD4 + T helper cells, CD45 + lymphocytes, CD8 + cytotoxic T cells and CD3 + T cells between women with type 1 and type 2 CSP Abbreviations: IQR: Interquartile range Discussion In this observational discovery study, we found a significant reduction in peripheral total NK cells, both as a percentage and in absolute counts, in women diagnosed with CSP compared with controls. The AUC for predicting the probability of CSP was 0.698, with 64% sensitivity and 66% specificity. Additionally, a significant reduction of absolute counts of peripheral CD8 + cytotoxic T cells was observed in women diagnosed with CSP. There was no difference in peripheral immune cell counts measured in this study between type 1 and type 2 CSP. Early detection and prediction are essential for the appropriate management of women diagnosed with CSP. Although CSP can present with vaginal bleeding and abdominal pain, up to 38% CSP women are asymptomatic [19, 20]. Ultrasound or magnetic resonance imaging (MRI) images are the gold standard for visualizing the location of a gestational sac in CSP diagnosis. Interestingly, recent studies have reported that systemic inflammation indices generated from routine whole blood tests could serve as potential additional clinical predictive markers for early CSP diagnosis [10, 11]. Both studies found significantly reduced lymphocyte counts. However, which specific lymphocyte subsets, such as NK cells, T cells, or B cells, contribute to the reduction of peripheral lymphocytes has not been investigated. NK cells, T cells, and B cells play an essential role in innate and adaptive immune responses (reviewed in [21]). CSP is characterised by inflammation (reviewed in [18]), suggesting a systemic or local immune response and an increase in total immune cell counts. In our current study, we found significantly reduced counts of peripheral total NK cells in CSP women at the time of diagnosis compared with healthy pregnancies. Lower peripheral NK cell counts suggest that the maternal body may not produce sufficient critical immune cells to protect against infections and diseases. This is because NK cells are innate lymphocytes with multiple immunoregulatory functions. NK cells can modulate immune responses by producing anti-inflammatory cytokines, such as IL-10, to control overactive immune responses, regulate T cell responses to prevent excessive inflammation, and promote antibody-dependent cytotoxicity [22]. CD56 is a protein commonly used as a marker to identify NK cells on the surface of immune cells. CD56 + NK cells are part of the innate immune system and control inflammation (reviewed in [23]). The functions of CD56 + NK cells include producing high levels of cytokines and chemokines with immunoregulatory properties [24, 25]. Low levels of CD56 + NK cells can increase susceptibility to infections and contribute to inflammation by disrupting regulatory control, allowing an unchecked immune response [26]. Our finding of lower counts of peripheral total NK cells, which are the majority of NK cells in peripheral blood, could indicate impaired immune surveillance or an altered inflammatory state, potentially contributing to inadequate modulation of local tissue inflammation in women diagnosed with CSP. However, there is no difference in the counts of NKT-like cells, suggesting that this specific immune subset remains stable across and may not play a primary role in the differential immune profiles of CSP. Using the area under the receiver operating characteristic (ROC) curve (AUC) to assess diagnostic accuracy is common in clinical practice [17]. When the AUC is greater than 0.5 generally indicates acceptable diagnostic accuracy; it depends on the disease [27]. In our current study, with an optimal cut-off count of 210/µl for peripheral NK cells, the AUC was 0.698, exceeding the 0.5 threshold and acceptable for a diagnostic test for CSP. Therefore, our findings suggest that peripheral total NK cell counts may serve as a predictive marker for diagnosing CSP. It is well known that NK cells are crucial for early pregnancy processes, such as vascular remodeling [28] and regulating local inflammation to support proper implantation (reviewed in [29]). After the embryo is implanted, NK cells in the uterus help regulate inflammation to support ongoing pregnancy. Age is known to influence the distribution of peripheral NK cell subsets, such as CD56 + NK cells, which decline with age. In our study, the age at CSP was significantly older than that of controls. This age difference partially explains the observed reduction in peripheral total NK cells in the CSP group, and age may be a potential confounder. However, correlation analysis showed no significant association between maternal age and any of the immune cell counts measured in our study, suggesting that the immunological changes observed are specific to the CSP pathology rather than age-related. B cells and T cells are the most critical components of adaptive immune responses (reviewed in [30]. In our current study, we found no changes in the total counts of peripheral CD19 + B cells, CD3 + T cells, and CD4 + T helper cells in women with CSP compared to normal pregnancies. In contrast, a significant reduction in peripheral CD8 + cytotoxic T cells was observed in women diagnosed with CSP. As an immunoregulator, CD8 + cytotoxic T cells produce cytokines that activate other immune cells to regulate inflammation, modulate or suppress or enhance immune response to maintain homeostasis [31, 32]. CD8 + cytotoxic T cells also influence the differentiation of CD4 + T helper cells, directing the immune response towards an inflammatory state. Our findings could suggest that the systemic adaptive immune response is also significantly altered in CSP conditions. This finding is consistent with the observed changes in peripheral NK cells, part of the innate immune system. Taken together, our findings suggest a dysregulated immune response in both innate and adaptive systems, without broadly affecting B cells or some T cell subsets in this inflammatory condition. Further investigation into the functional activity of NK cells and T cytotoxic cells, as well as their interactions with other immune cells in CSP, could provide additional insights. Although CSP is traditionally divided into two subtypes based on the location of the gestational sac, our study found no statistical differences in immune cell counts measured between the two subtypes. Various anatomical, physiological, and potentially genetic risk factors may influence the development of these subtypes. For instance, multiple cesarean sections may result in more extensive scar tissue, and this effect could increase the risk of developing type 1 CSP [33]. Our current findings are consistent with our previous findings, which indicated that both share many clinical characteristics [34]. Together, these data suggest that both CSP subtypes are under the same underlying inflammation conditions. There are some limitations to this study. Firstly, because of the nature of pregnancy, early sampling may not reflect normal conditions. Clinical symptoms of several pregnancy complications are present at a late gestational age. Therefore, selecting controls from first-trimester terminations may introduce bias. Secondly, higher parity was observed among CSP women. Live births or incomplete pregnancies could impact the body’s immune system. Epidemiology studies showed a negative association between the incidence of ovarian and endometrial cancer and parity [35–39]. At least in part, this negative association is due to the maternal immune response during pregnancy [40, 41]. While live birth(s) may enhance immunity against inflammation, this is not seen in CSP women. In this study, we observed reduced numbers of peripheral total NK cells and CD8 + cytotoxic T cells in women with CSP. As samples were collected after diagnosis, it remains unclear whether these systemic changes represent a pre-existing immunological environment that predisposes to CSP (cause) or a secondary response to the abnormal implantation and chronic inflammation (consequence). Additionally, given that these immune cells are key regulators of trophoblast invasion, their reduction may reflect a systemic immune response of the impaired uterine environment in CSP. Future prospective studies are needed to confirm whether the reduction in peripheral total NK cells and CD8 + cytotoxic T cells is a cause or a consequence of CSP. In conclusion, our study reported reduced peripheral counts of total NK cells and CD8 + cytotoxic T cells in women with CSP. Our findings suggest that reductions in peripheral NK cells and T cytotoxic cells may contribute to the development of CSP. Additionally, using NK cell counts with a cut-off of 210/µL as a predictive measure yielded 64% sensitivity. However, CSP prediction still relies on a combination of clinical presentation and imaging, rather than a definitive predictive marker. Supplementary Information Supplementary Material 1. Acknowledgements Not applicable. Clinical trial number Not applicable. Authors’ contributions All authors were involved in the drafting, editing, and approval of the manuscript for publication. In addition to this, each author contributed to the following work. YS, and QC: Study design and conception. YS, and HF: Sample collection. HG, LZ, and QJ: immune cell profile measurements. JZ: Pathology performance. HG and QC: Data analysis and interpretation. QC: manuscript draft and finalise. MZ: manuscript revision. Funding This study did not receive any financial support. Data availability The datasets used and analyzed included in the current study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate The Ethics Committee of Wuxi Maternity and Child Health Hospital approved this study (reference number: 2024-06-1024-55). This study was performed in accordance with the Medical Association Declaration of Helsinki. Blood samples were collected from participants who provided signed consent forms. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes The original online version of this article was revised: The Figure 1 legend was corrected. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Ye Shen and Jinqiu Zhang equally contributed to this study. Change history 7/12/2026 The Figure 1 legend was corrected. Change history 7/14/2026 A Correction to this paper has been published: 10.1186/s12865-026-00876-y Contributor Information Qi Chen, Email: q.chen@auckland.ac.nz. Min Zhao, Email: lwan480@aucklanduni.ac.nz. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The datasets used and analyzed included in the current study are available from the corresponding author upon reasonable request.