Clinical significance of CD27⁺CD28⁺ effector memory T cells as potential prognostic biomarkers in ovarian cancer

Abstract

Objective

Ovarian cancer remains the most lethal gynaecological malignancy, and the identification of reliable biomarkers for determining treatment responses and prognosis is needed. This study investigated the clinical significance of helper and cytotoxic effector memory T-cell (T_EM) subsets, defined by CD27 and CD28 coexpression, circulation and local compartments.

Methods

Multi-parametric flow cytometry was used to analyse peripheral blood (n = 30), ascites (n = 14) and tumour tissue (n = 31) from patients with high-grade ovarian cancer (FIGO III-C, n = 31) and age-matched healthy controls (n = 16). Subsets were stratified by histopathology, neoadjuvant chemotherapy status, and clinical outcomes.

Results

T_EM cells were identified as the predominant T-cell population across all the compartments, with significant enrichment within the tumour microenvironment (reaching > 90%). In peripheral blood from healthy controls, T_EM cells likewise represented the dominant memory T-cell subset, with CD27⁺CD28⁺ cells constituting the most abundant T_EM cell fraction. The CD27⁺CD28⁺ T_EM cell subset emerged as the unequivocally dominant T_EM fraction in both circulation and tumour tissue. Notably, a high frequency (> 63%) of circulating CD27⁺CD28⁺ helper or cytotoxic T_EM cells significantly correlated with prolonged progression-free survival (15–16 months vs. 9–10.5 months), an increased incidence of reactive lymph nodes, and favourable CA125 kinetics. Conversely, neoadjuvant chemotherapy was associated with a significant reduction in tumour-infiltrating CD27⁺CD28⁺ T_EM cells.

Conclusion

The preservation of CD27 and CD28 on circulating T_EM cells was associated with favourable clinical outcomes and markers of active immunosurveillance, whereas the loss of these molecules may indicate terminal differentiation and impaired antitumour immunity in patients with ovarian cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13048-026-02077-y.

Introduction

Ovarian cancer represents a clinical challenge, ranking as the most lethal gynaecological malignancy and the second leading cause of cancer-related death in women [1]. Current treatment protocols for this highly heterogeneous disease include optimal cytoreductive surgery followed by platinum-based chemotherapy; however, despite initial success, many patients experience recurrence marked by the development of chemotherapy resistance, significantly decreasing overall survival and highlighting the urgency for innovative therapeutic approaches, particularly those targeting the immune system [2, 3]. The tumour microenvironment (TME) in ovarian carcinoma is a key determinant of therapeutic resistance, and patients frequently exhibit an immunosuppressive phenotype that severely limits the efficacy of immunotherapies, such as immune checkpoint inhibitors (ICIs). The ovarian cancer TME is enriched in immunosuppressive cells, including myeloid-derived suppressor cells (MDSCs) and tumour-associated macrophages (TAMs), which actively suppress the function of antitumour subsets of tumour-infiltrating lymphocytes (TILs), such as CD8⁺ cytotoxic T cells and type-1 helper T (Th1) cells. A comprehensive investigation of alterations in immune cell populations, particularly the functional phenotypes of T lymphocytes, is necessary for optimizing and advancing effective immune intervention strategies in ovarian cancer [4].

Long-term antitumour immunity is predicated on the differentiation of naïve T (T_N) cells following antigen encounter into diverse memory T-cell populations [5]. Central memory T (T_CM) cells reside mainly in lymphoid tissues, where they retain a high proliferative capacity upon antigen restimulation but are relatively inert in terms of immediate effector molecule secretion [6, 7]. Conversely, effector memory T (T_EM) cells and terminal effector memory (T_EMRA) cells preferentially migrate to peripheral inflamed sites, including the TME. T_EM cells rapidly execute their cytotoxic functions and secrete high levels of effector cytokines such as IFN-γ and TNF-α [8]. Defined memory T-cell subpopulations are characterized by the expression of distinct surface molecules (e.g., costimulatory molecules and inhibitory receptors) that function as immunological biomarkers, facilitating the prediction of their specialized antitumour capacity and the stratification of patients on the basis of potential response to immune intervention therapies.

The expression of the costimulatory receptors CD27 and CD28 further defines the functional spectrum of memory T cells. Expression of both markers is progressively lost during differentiation, leading to significant heterogeneity, especially within the T_EM compartment, where different CD27 and CD28 profiles correlate with varying functional stages, providing mechanistic insight into T-cell responsiveness and antitumour potential [9, 10]. This immunological heterogeneity is clinically relevant across numerous cancers, and previous studies have examined the expression of CD27 and/or CD28 on T cells in the context of treatment response in various malignancies [1, 3, 11–15]. However, data concerning single expression or coexpression of CD27 and CD28 on T-cell subsets across multiple biological compartments—namely, peripheral blood, ascites, and tumour tissue—in ovarian cancer are limited, particularly in relation to clinical parameters.

This study aims to address this gap by specifically investigating the presence and clinical significance of CD27 and CD28 expression on T-cell subsets in peripheral blood, ascites fluid and tumour tissue from high-grade ovarian cancer patients. Here, we report that the sustained presence of a high frequency of circulating CD27⁺CD28⁺ T_EM cells functions as an independent, favourable prognostic biomarker for enhanced survival in ovarian carcinoma patients, irrespective of patient age or therapeutic modality.

Materials and methods

Patients and clinical data

Peripheral blood (n = 30), ascites (n = 14), and tumour tissue samples (n = 31) were obtained from patients with ovarian cancer (n = 31: nonmalignant, n = 7; malignant, n = 24; median age, 56; min age, 29; max age, 80) who had received either neoadjuvant therapy (carboplatin/paclitaxel, n = 8; carboplatin/paclitaxel/bevacizumab, n = 4) or no therapy prior to oncological surgery (n = 11). All patients were at clinical grade FIGO-III-C. On the basis of histopathological evaluation, patients with borderline or benign pathology were collectively categorized as having nonmalignant tumours, while all other patients were classified as having malignant tumours. Age-matched peripheral blood samples were also obtained from heathy donors (n = 16; median age, 57; min age, 30; max age, 74) who did not present with comorbidities or inflammatory diseases. Postoperative therapies included paclitaxel alone (n = 1), carboplatin/paclitaxel (n = 3), carboplatin/paclitaxel/bevacizumab (n = 15; two patients also received olaparib) and gemcitabine/docetaxel (n = 1) in accordance with the National Comprehensive Cancer Network (NCCN) guidelines (Suppl. Tables 1–2). All experiments were conducted in accordance with the Declaration of Helsinki after approval from the local ethics committees (University of Health Sciences Ethics Committee, approval no. 2025 − 414; Hacettepe University Ethics Committee, approval no. 2025/21–35), and informed consent was obtained from the participants.

Cell isolation

Peripheral blood samples from patients and healthy volunteers were collected into EDTA vacutainer tubes (BD Biosciences). Blood samples and ascites were diluted 1:1 with phosphate-buffered saline (PBS) and gently layered over Histopaque-1077 (Sigma Aldrich). Histopaque-1077 is crucial for isolating peripheral blood mononuclear cells (PBMCs) because of its precise density. With this technique, when diluted blood samples and ascites are carefully layered and centrifuged, components are separated on the basis of density. Denser components, including red blood cells (RBCs) and granulocytes, sediment at the bottom of the tube, whereas PBMCs form a distinct buffy coat layer at the plasma–Histopaque interface, thereby eliminating the need for additional RBC lysis steps. The buffy coat is then collected for flow cytometric analysis.

To obtain cell suspensions from ovarian tumours, surgical specimens were finely chopped and incubated in culture medium supplemented with 0.075% collagenase type II and 0.01% DNase I at 37 °C. The resulting cell suspension was passed through a 40-µm mesh filter to eliminate cell aggregates and debris. Since leukocyte size ranges from 10 to 30 μm, filtration with a 40-µm mesh did not exclude any leukocyte population. This step also minimized the presence of necrotic material. Cell viability and the exclusion of residual debris were subsequently ensured during flow cytometric analysis through forwards- and side-scatter–based gating strategies.

Flow cytometry

Immunophenotyping was performed using fluorochrome-conjugated monoclonal antibodies targeting human CD45RA (HI100), CD45RO (UCHL1), CCR7 (150503) (BD Biosciences), CD3 (UCHT1), CD56 (MEM-188), CD4 (RPA-T4), CD8 (SK1), CD27 (O323), and CD28 (CD28.2) (BioLegend). The samples were run and analysed on a FACSCanto II (BD Biosciences) flow cytometer, and the percentage of positive cells was determined according to isotype-matched control antibody staining or autofluorescence. The specific gating strategy employed for subset identification is detailed in the Supplementary Information (Suppl. Figure 1). The median fluorescence intensity (MFI) values were also recorded.

Statistical analysis

Student’s t test (paired or unpaired) or one-way/two-way ANOVA was used for statistical analyses, as appropriate. A p value of < 0.05 was considered to indicate statistical significance. Unless otherwise noted, the data are shown as the median ± SEM.

Results

Helper and cytotoxic T cells with an effector memory phenotype populate blood, ascites and tumours in ovarian cancer patients

Immunophenotyping of peripheral blood samples collected from age-matched healthy volunteers, along with peripheral blood, ascites, and corresponding tumour tissue obtained from ovarian carcinoma patients, consistently revealed T_EM cells as the predominant T lymphocyte subset among both the CD4⁺ and CD8⁺ populations (Suppl Fig. 2). Quantitative analysis revealed a significant increase in the percentage of CD4⁺ T_EM cells in healthy controls (44.2 ± 2.7% in blood), whereas a significant increase was detected in peripheral blood of patients (57.7 ± 2.9%), followed by marked enrichment in ascites (69.1 ± 3.4%) and the highest level of CD4⁺ T_EM cells within solid tumour tissue (96.5 ± 3.1%). CD8⁺ T_EM frequencies demonstrated less variation across the fluid compartments—65.3 ± 2.2% in healthy blood, 65.6 ± 3.5% in patient blood, and 64.9 ± 3.6% in ascites—but also reached high dissemination within the tumour tissue (94.1 ± 4%) (Fig. 1A).

Fig. 1.

CD4⁺ and CD8⁺ T_EM cells populate blood, ascites and tumour tissue in ovarian cancer patients independent of age, disease stage and neoadjuvant therapy. Percentages of CD4⁺ and CD8⁺ naïve and memory T-cell subsets (a) and correlations of CD4⁺ and CD8⁺ T_EM cells with age (b) in the peripheral blood of healthy controls and in the peripheral blood, ascites, and tumour tissue of ovarian cancer patients were analysed by simple linear regression. The regression line represents the estimated linear relationship between T_EM cells and age. A p value of < 0.05 was considered to indicate statistical significance. Percentages of CD4⁺ and CD8⁺ naïve and memory T-cell subsets (c) in peripheral blood and tumour tissue of ovarian cancer patients stratified by histopathological and neoadjuvant therapy status (healthy control, blood n = 16; ovarian cancer patient, blood n = 30, ascites fluid n = 14, tumour tissue n = 31; *p < 0.05, **p < 0.01; two-way ANOVA)

Following the confirmation of T_EM cell predominance, the influence of age on T_EM distribution was investigated. In healthy volunteers, the frequencies of both CD4⁺ and CD8⁺ T_EM cells were significantly positively correlated with age. In sharp contrast, T_EM cells consistently dominated all analysed samples from ovarian carcinoma patients, suggesting that their elevated presence is disease-associated rather than age-dependent (Fig. 1B). Finally, the distribution of CD4⁺ and CD8⁺ T-cell subsets was assessed on the basis of clinical covariates, including histopathological diagnosis (malignant versus nonmalignant) and the administration of neoadjuvant chemotherapy prior to surgical resection. The data indicated that T_EM cells remained the overwhelmingly dominant T-cell subset across all analysed patient groups, suggesting that their pronounced distribution is not significantly influenced by the specific disease histopathology or prior therapeutic intervention (Fig. 1C).

Collectively, these findings establish T_EM cells as the paramount T-cell subset not only in the systemic circulation but also profoundly within the local TME cells of ovarian carcinoma, underscoring the potential significance of T_EM cells in mediating local antitumour or protumour immune responses.

CD27⁺CD28⁺ T_EM cells dominate the tumour microenvironment and circulation in ovarian cancer

Having established the predominance of T_EM cells in samples from both healthy controls and patients with ovarian carcinoma, we proceeded to further classify the CD4⁺ and CD8⁺ T_EM populations on the basis of the expression profiles of the costimulatory molecules CD27 and CD28. This phenotyping yielded four distinct T_EM subgroups. Among the CD4⁺ T_EM cells, CD27⁺CD28⁺ cells were the most abundant, accounting for 47.3 ± 1.3% of the cells in the peripheral blood of healthy controls, 63.6 ± 3.9% of the cells in the peripheral blood of patients with ovarian cancer, and, remarkably, 89.9 ± 1.7% of the cells within the tumour tissue. With respect to the CD8⁺ T_EM cells, the CD27⁺CD28⁺ subset also represented the major fraction, observed at 53.9 ± 1.3% in the peripheral blood of healthy controls, 62.6 ± 3.9% in the peripheral blood of patients with ovarian cancer, and 58.9 ± 2.1% within the tumour tissue (Fig. 2A and B).

Fig. 2.

CD27⁺CD28⁺ T_EM cells represent the predominant subset of CD4⁺ and CD8⁺ T_EM cells in ovarian cancer. The percentage distribution of CD4⁺ and CD8⁺ T_EM cell subsets according to CD27 and CD28 expression was plotted in blood from healthy controls, blood from patients with ovarian carcinoma, and tumour tissue (a). Representative flow cytometry dot plots illustrate CD27 and CD28 expression patterns on T_EM cells determined in distinct samples (b)

In summary, these findings strongly demonstrate that the subpopulation coexpressing the costimulatory molecules CD27 and CD28 is the unequivocally dominant T_EM fraction observed in the circulation of patients and is prominently enriched within the malignant ovarian tumour microenvironment.

High levels of circulating CD27⁺CD28⁺ T_EM cells are associated with good prognosis and survival in patients with ovarian cancer

Following the identification that CD27⁺CD28⁺ subsets of both helper and cytotoxic T_EM cells predominate in the peripheral blood and tumour microenvironment of ovarian cancer patients, the data were stratified by histopathological subtype and neoadjuvant chemotherapy status. The CD27⁺CD28⁺ phenotype remained the dominant population, irrespective of histopathological features or treatment history. Among circulating CD4⁺ T_EM cells, the frequency of the CD27⁺CD28⁺ subset was significantly greater in patients with nonmalignant tumours (83.2 ± 8.7%) than in those with malignant tumours (57.7 ± 4%). Conversely, no statistically significant differences were observed between the malignant and nonmalignant groups when the percentages of CD27⁺CD28⁺ cells among peripheral CD8⁺ T_EM cells or tumour-infiltrating CD4⁺ and CD8⁺ T_EM populations were compared (Fig. 3A).

Fig. 3.

Tumour-infiltrating CD27⁺CD28⁺ T_EM cells predict favourable prognosis in patients with ovarian cancer. The percentage of circulating and tumour-infiltrating helper and cytotoxic T_EM cells was distributed according to CD27 and CD28 expression (a) in patients with malignant and nonmalignant (benign and borderline) tumours and (b) according to the neoadjuvant therapy received. c Kaplan‒Meier curves for progression-free survival (PFS) of ovarian cancer patients categorized according to the percentage of CD27⁺CD28⁺ T_EM cells among CD4⁺ and CD8⁺ T_EM cells (*p < 0.05, **p < 0.01; two-way ANOVA). The median percentage of CD27⁺CD28⁺ T cells (63%) was taken as the cut-off value (n = 18; *p < 0.05; statistical Gehan–Breslow–Wilcoxon test applied)

With respect to treatment status, the peripheral blood CD4⁺ T_EM CD27⁺CD28⁺ frequency was 61.9 ± 6% in treatment-naïve patients and 52.8 ± 5.7% in patients receiving neoadjuvant therapy. Notably, within tumour tissue, the percentage of the CD27⁺CD28⁺ subset decreased from 85.6 ± 2.3% to 64.9 ± 4.7% following neoadjuvant therapy. In contrast, CD27⁺CD28⁺ CD8⁺ T_EM cell abundance remained comparable in both the blood and tumour compartments regardless of therapeutic administration (Fig. 3B).

Survival analysis utilized a median cut-off of 63%. Patients harbouring CD27⁺CD28⁺ helper or cytotoxic T_EM cell frequencies above this threshold exhibited significantly prolonged progression-free survival (PFS). Specifically, high helper T_EM cell levels correlated with a PFS of 15 ± 2.5 months (vs. 10.5 ± 1.9 months in the low-frequency group), whereas high cytotoxic T_EM cell levels correlated with a PFS of 16 ± 2.1 months (vs. 9 ± 1.9 months in the low-frequency group) (Fig. 3C). Lymph node status, CA125 kinetics, and median survival were analysed according to the frequency of CD27⁺CD28⁺ T_EM cell subsets in peripheral blood (Table 1). Higher proportions of both CD4⁺ and CD8⁺ CD27⁺CD28⁺ T_EM cells (≥ 63%) were associated with increased reactive lymph node rates, reduced metastatic involvement, lower CA125 values, and longer median survival than lower proportions of these cell subtypes were.

Table 1.

Association of CD27⁺CD28⁺ T_EM cell subsets with lymph node status and CA125 dynamics (% of patients), and survival in ovarian cancer patients

	Reactive LN (%)	Metastatic LN (%)	CA125 (%)	Median survival (months)
CD4+ CD27+CD28+ TEM ≥ 63%	91.38	6.36	Stable (20) Decreased (70) Increased (10)	15
CD4+ CD27+CD28+ TEM < 63%	80.61	13.24	Stable (12.5) Decreased (37.5) Increased (50)	10.5
CD8+ CD27+CD28+ TEM ≥ 63%	93.23	8.42	Stable (12.5) Decreased (62.5) Increased (25)	16
CD8+ CD27+CD28+ TEM < 63%	85.66	14.82	Stable (10) Decreased (50) Increased (40)	9

CA125 kinetics were evaluated during follow-up and categorized as stable, decreased, or increased

Values are presented as percentages or median values as indicated. LN, lymph node; T_EM, effector memory T cells

Collectively, these findings suggest a significant correlation between peripheral CD27⁺CD28⁺ T_EM cell levels and the clinical prognosis of ovarian cancer patients.

Discussion

Memory T-cell subsets, including T_CM, T_EM, and terminally differentiated effector memory cells re-expressing CD45RA (T_EMRA) cells, circulating among different compartments are fundamental to the durability and efficacy of the antitumour immune response. Their importance in cancer immunology is defined by a clear division of labour and distinct anatomical compartmentalization. The relative distributions of these subsets across the blood-tumour axis may provide a valuable picture of the immune system state in cancer patients. By investigating the clinical relevance of memory T-cell types and, especially, T_EM cell subsets defined by CD27 and CD28 expression patterns in peripheral blood, ascites, and tumour tissue of ovarian cancer patients, a significant correlation between the distribution of specific circulating subsets and clinical outcomes was observed. Patients harbouring a higher frequency of CD27⁺CD28⁺ T_EM cells exhibited prolonged PFS, a higher number of reactive lymph nodes, and a more pronounced decrease in CA125 levels posttreatment, indicating a favourable immune profile. Conversely, enrichment of the CD27⁻CD28⁻ T_EM cell subset was associated with decreased PFS, an increased burden of metastatic lymph nodes, and persistently elevated CA125 levels. These data suggest that the CD27/CD28 expression signature of circulating T_EM cells may serve as a minimally invasive and clinically accessible biomarker for predicting disease progression and therapeutic response.

The shift towards a CD27⁻CD28⁻ phenotype in advanced malignancy likely reflects terminal T-cell differentiation driven by chronic antigen exposure [16]. The progressive loss of CD28 and CD27 is a hallmark of replicative senescence and impaired proliferative potential. This CD27⁻CD28⁻ T cells population resembles effector memory T cells that transition towards a terminally differentiated state that is often characterized by homeostatic dysregulation as T_EMRA cells [10]. In the context of the chronic inflammation typical of ovarian cancer, the accumulation of these cells may indicate an exhausted or functionally impaired immune response.

Our observations align with findings in other malignancies; for instance, in diffuse large B-cell lymphoma (DLBCL), a low frequency of CD27⁻CD28⁻ cells has been determined to be a favourable biomarker for long-term CAR-T-cell therapy response [17]. Similarly, in lung cancer, systemic accumulation of CD27⁻CD28⁻ T_EM cells is correlated with reduced proliferative capacity and poor responses to immune checkpoint inhibitors [18]. Compared with their CD27⁺CD28⁺ counterparts, these cells exhibit reduced cytokine production [17, 18]. Thus, the absence of CD27 and CD28 may indicate a transition from high to low proliferative potential, compromising effective antitumour immunity. In contrast, the T_EM cell subset, which retains the critical costimulatory molecules CD27 and CD28, represents a less differentiated, highly proliferative phenotype. The correlation of this subset with active immune surveillance, as evidenced by the number of reactive lymph nodes and favourable CA125 kinetics, is consistent with the benefits of CD27 signalling. Indeed, CD27 agonism has been shown to enhance anti-PD-1 efficacy [3], and high CD27 expression has been determined to be an independent protective factor in serous ovarian cancer [1]. Alternatively, the loss of peripheral CD8⁺CD28⁺ T cells is significantly associated with lymph node metastasis in patients with breast cancer [11].

A primary strength of this study is the use of specific immunophenotyping across three biological compartments affected by ovarian cancer. The specificity and potency of antitumour immune responses are inherently dictated by the evolution of local immune reactions within the tumour microenvironment. While these reactions have a limited impact on surrounding tissues, they typically manifest in the peripheral blood only in advanced stages of the disease. In malignancies that are characterized by the accumulation of ascites fluid, such as ovarian cancer, a unique compartment that may serve as a more representative proxy for the tumour’s immune microenvironment and the resulting immunomodulation emerges. In accordance with previous literature [8], our study suggests that, in several aspects, ascites fluid provides a closer reflection of the TME than the systemic circulation does. Moreover, comparing the distribution of memory T-cell subsets in the peripheral blood of ovarian cancer patients with that in the peripheral blood of healthy individuals allows for characterization of the systemic impact of malignancy. Notably, the expansion of memory T-cell populations, which is typically associated with chronological ageing, occurs in these patients regardless of age [9]. This shift may indicate systemic immunomodulation and premature contraction of the T-cell immune reservoir, both of which are detectable in the peripheral blood.

However, this study is limited by its cohort size, particularly after stratification by histopathological subtype and neoadjuvant treatment status, necessitating validation in larger prospective trials. Further mechanistic studies are needed to elucidate the reduction in CD27⁺CD28⁺ T_EM cells in patients receiving neoadjuvant therapy and to determine whether this decline is a direct consequence of chemotherapy-induced lymphotoxicity. Additionally, while this phenotypic profile is suggestive of senescence, direct functional assays (e.g., cytotoxicity or cytokine secretion) were not performed. This study also did not account for tissue-resident memory T cells (T_RM cells), which play a pivotal role in local tumour control. As CD103⁺ T_RM cells represent a specialized, nonrecirculating lineage that facilitates rapid, localized anticancer activity within the tumour epithelium, they may serve as critical determinants of the robustness and durability of the local immune response [19]. Future research should focus on functional characterization of these subsets (T_EM and T_RM cells) and their ability to predict the relative contributions of T_RM cells to resident versus circulating T-cell populations and responses to next-generation therapies, such as CD27 agonists. In conclusion, CD27 and CD28 coexpression on T_EM cells may provide a useful, quantifiable peripheral marker for assessing disease aggressiveness and prognosis in ovarian cancer.

Conclusion

In conclusion, this study identifies the CD27⁺CD28⁺ T_EM cell subset as a prominent T-cell population across the peripheral blood, ascites, and tumour tissues of ovarian cancer patients. Maintained expression of these costimulatory molecules appears to correlate with favourable clinical indicators, including prolonged progression-free survival and stable CA125 kinetics. In contrast, the transition towards a terminal CD27⁻CD28⁻ phenotype may indicate replicative senescence and reduced antitumour capacity. These findings suggest that the CD27/CD28 expression profile could serve as a supplementary tool for patient stratification and may provide a rationale for exploring therapeutic targets, such as CD27 agonists, to support immune persistence in ovarian cancer.

Abbreviations

T_EM

Effector memory T

T_N

Naïve cells

T_CM

Central memory T cells

T_EMRA

Terminal effector memory cells

T_RM

Tissue-resident memory T cells

Ag

Antigen

PBMCs

Peripheral blood mononuclear cells

PFS

Progression-free survival

SOC

Serous ovarian cancer

ICI

Immune checkpoint inhibitor

CA125

Cancer antigen 125

PB

Peripheral blood

PBS

Phosphate-buffered saline

Authors’ contributions

G.E. conceived and supervised the project. S.U. and G.E. designed the protocol. S.U. and I.A.A. performed the experiments. S.U. analysed the data. S.U., G.E., and K.B.Y. wrote, reviewed and edited the manuscript. K.B.Y., I.A.A., M.M.H., B.K., V.A., V.S., S.G., I.B.B, O.O., I.E., and M.A.G. validated the data. All the authors read and approved the final manuscript.

Funding

This work was supported by The Scientific and Technological Research Council of Turkey (TÜBİTAK) under the 2211 National PhD Scholarship Programme (Application No. 1649B032205227) awarded to S.U.

Data availability

The data can be obtained through email under reasonable request: gunese@hacettepe.edu.tr.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Hacettepe University (approval No. 2025/21–35) and the University of Health Sciences (approval No. 2025 − 414). The use of patient samples conformed to the Declaration of Helsinki. Written informed consent was obtained from all participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.