Tacrolimus Modulates Peripheral Blood NK Cell Subsets and IL-17/IL-10 Ratio in Women With Recurrent Implantation Failure.

This study was found to be ethically acceptable by the Ethical Committee of Mazandaran University of Medical Sciences (IR.MAZUMS.REC.1403.088).

This study was supported by Mazandaran University of Medical Sciences (Project number: 21743).

This study was conducted as a mechanistic continuation using samples obtained from our previous registered clinical trial (IRCT20231127060207N1) [ 20 ]. It included a baseline case–control comparison between RIF patients and fertile controls, followed by a longitudinal within‐subject analysis of RIF patients before and after tacrolimus treatment. Importantly, the immunological analyses presented in this manuscript were not predefined primary endpoints of the original clinical trial and were performed as exploratory post hoc assessments to investigate potential mechanistic pathways associated with the clinical findings. The study was conducted at the Infertility Clinic of Imam Khomeini Educational Hospital in Sari, Iran, affiliated to Mazandaran University of Medical Sciences. A total of 60 women aged ≤ 40 years were enrolled between February and July 2025, comprising 30 patients with RIF who received tacrolimus therapy and 30 age‐matched fertile healthy volunteers. The control group consisted of women with normal reproductive outcomes, all of whom had at least one prior uncomplicated pregnancy resulting in a live birth and no history of infertility or recurrent pregnancy loss. This cohort was used to identify dysregulated immune markers in RIF compared to a healthy population at baseline. Patients in the tacrolimus group received oral tacrolimus at a daily dose of 3 mg, starting 2 days before embryo transfer and continuing until the day of serum pregnancy testing, approximately 14 days post‐transfer (total duration: 16 days). RIF was defined as failure to achieve pregnancy after at least three consecutive IVF cycles involving transfer of morphologically high‐quality embryos under favorable endometrial conditions, following exclusion of anatomical, genetic, and endocrine causes. Control participants had no history of immunologic, rheumatologic, or allergic disorders, and all had experienced at least one previous normal pregnancy resulting in a live birth. The study protocol was reviewed and approved by the Ethics Committee of Mazandaran University of Medical Sciences (IR.MAZUMS.REC.1403.088). Written informed consent was obtained from all participants before enrollment. Peripheral venous blood samples were collected from RIF patients at two time points: (i) prior to initiation of tacrolimus therapy and embryo transfer, and (ii) on the day of the serum pregnancy test, approximately 14 days after embryo transfer. Blood was obtained from healthy volunteers at a single time point. At each collection, two 5 mL samples were drawn into EDTA tubes: one for flow cytometric analysis of NK cell‐associated subsets and the other for evaluation of gene expression of IL‐17 and IL‐10. To evaluate NK cell phenotyping, peripheral blood samples were analyzed by flow cytometry. Briefly, peripheral blood samples were diluted with flow cytometry staining buffer (PBS containing 0.5% BSA) and individually incubated with fluorescence‐conjugated monoclonal antibodies, including anti‐human CD16‐FITC and anti‐human CD56‐PE (BioLegend, San Diego, CA, USA), for 45 min at 4°C. Following antibody staining, red blood cells (RBCs) were lysed using an RBC lysis buffer (Farhan Azma, Iran). The samples were then analyzed using a Beckman Coulter flow cytometer (Beckman Coulter, Brea, CA, USA). Isotype‐matched control antibodies were used to subtract nonspecific background staining. Importantly, CD3 was not included in the flow cytometry panel; therefore, NK cell subsets were defined based on CD16/CD56 expression, and a minor contribution of CD3 + CD56 + NKT cells cannot be fully excluded. Total RNA was extracted from peripheral blood samples using the Denazist RNA isolation kit (Mashhad, Iran) according to the manufacturer's instructions. The quality and integrity of the RNA were evaluated using nanospectrophotometry (WPA, England) and electrophoresis (Bio‐Rad, UK). Subsequently, 1 μg of total RNA was reverse‐transcribed into complementary DNA (cDNA) using the Yekta‐Tajhiz cDNA synthesis kit (Tehran, Iran) in a 20 μL reaction volume. The reaction mixture contained 1 μL random hexamer primer, 2 μL of 10 mM deoxyribonucleotide triphosphate mix, 4 μL of 5× reaction buffer, 1 μL RNase inhibitor, 200 units of RevertAid MMuLV reverse transcriptase, and RNase‐/DNase‐free water. cDNA synthesis was carried out at 42°C for 1 h, followed by 70°C for 5 min to inactivate the enzyme. The expression levels of IL‐10 and IL‐17 genes were evaluated using Real‐time PCR. Reactions were performed with the SYBR Green PCR Master Mix (Amplicon, Copenhagen, Denmark) on an ABI Step‐One Real‐time PCR system (Applied Biosystems, Foster City, CA, USA). Primers were designed using AlleleID software and synthesized by Metabion International AG (Planegg, Germany) (Table  1 ). Primer efficiency was validated in preliminary experiments, and standard curves were generated. β‐actin served as the reference (housekeeping) gene for normalization. Relative gene expression was calculated using the Pfaffl method, which accounts for primer amplification efficiencies. All reactions were performed in triplicate to ensure reproducibility of the results. Primer sequences used for real‐time PCR. F: TCTCCGAGATGCCTTCAG R: CCCAGGTAACCCTTAAAGTC F: GAATCTCCACCGCAATGA R: GGACAGAGTTCATGTGGTAG F: CCTTCCTGGGCATGGAGTCCT R: TGGGTGCCAGGGCAGTGAT Abbreviations: bp, base pair; F, Forward; R, Reverse. Statistical analyses were conducted using GraphPad Prism software (version 9). Data are expressed as mean ± standard deviation (SD). The normality of data distribution was assessed using the Kolmogorov–Smirnov test. Based on the results of normality testing, parametric analyses were applied. Differences between two independent groups (RIF and control) were analyzed using an unpaired t ‐test, whereas paired t ‐tests were applied for within‐group comparisons before and after treatment. To strengthen the analysis, the potential risk of type I error due to multiple comparisons was acknowledged given the number of immune endpoints evaluated, and no formal adjustment for multiple testing was applied. Analyses of subgroup comparisons based on clinical pregnancy outcome were considered exploratory due to limited sample size and should be interpreted with caution. In addition, effect sizes (Cohen's d) and 95% confidence intervals (95% CI) were additionally calculated for key comparisons to estimate the magnitude and precision of observed differences. A p ‐value of less than 0.05 was considered statistically significant.

Patients included in this study were part of a previously conducted clinical trial assessing the effect of tacrolimus on pregnancy outcomes [ 20 ]. A total of 60 women were included in the study, comprising 30 patients with RIF who received tacrolimus and 30 age‐matched fertile healthy women for baseline comparison of immune markers. Baseline characteristics were comparable between groups, with mean ages of 35.07 ± 3.75 years for RIF patients and 35.47 ± 3.82 years for controls ( p  = 0.683). In the RIF group, all patients had primary infertility, with a mean infertility duration of 6.3 ± 3.6 years. Additional demographic and clinical characteristics, including indications for ART, comorbidities, and gynecological findings, are summarized in Table  2 . Baseline demographic and clinical characteristics of the study population. Abbreviations: ART, assisted reproductive technology; BMI, body mass index; IVF, In vitro fertilization; RIF, recurrent implantation failure. Defined as no history of clinically recognized pregnancy loss prior to enrollment. Representative flow cytometry dot plots from one control subject and one RIF patient illustrating the gating strategy used to quantify CD16 + , CD56 + , and CD16 + CD56 + NK cell–associated subsets are shown in Figure  1A . At baseline, the proportion of CD16 + NK cells was significantly higher in RIF patients compared with controls (13.79% ± 5.17% vs. 11.05% ± 3.44%, p  = 0.019; mean difference: 2.74%, 95% CI: 0.47–5.01, Cohen's d = 0.62) (Figure  1B ). A more significant elevation was detected in CD56 + NK cells, as RIF patients exhibited almost a two‐fold higher frequency compared with controls (19.63% ± 5.11% vs. 9.93% ± 2.52%, p  < 0.0001; mean difference: 10.0%, 95% CI: 7.92–12.08, Cohen's d = 2.48) (Figure  1C ). Similarly, the frequency of CD16 + CD56 + NK cells was markedly elevated in the RIF group compared with controls (12.59% ± 4.96% vs. 7.56% ± 2.46%, p  < 0.0001; mean difference: 5.03%, 95% CI: 3.01–7.05, Cohen's d = 1.28) (Figure  1D ). Baseline quantification of CD16 + , CD56 + , and CD16 + CD56 + NK cells in RIF patients and controls. (A) The gating strategy included initial identification of lymphocytes based on forward‐ and side‐scatter characteristics, followed by gating on lymphocyte population. Representative flow cytometry dot plots from a control subject and an RIF patient illustrating CD16 + and CD56 + NK cell distribution. Bar plots comparing the percentage of CD16 + NK cells (B), CD56 + NK cells (C), and CD16 + CD56 + NK cells (D) between controls and RIF patients. Data are presented as mean ± SD (* p  < 0.05; **** p  < 0.0001). At baseline, the expression of IL‐17, a Th17‐associated cytokine, was significantly higher in RIF patients compared with controls (2.13 ± 0.51 vs. 1.10 ± 0.47 fold change, p  < 0.0001; mean difference: 1.03, 95% CI: 0.78–1.28, Cohen's d = 2.10) (Figure  2A ). In contrast, IL‐10, an anti‐inflammatory cytokine associated with regulatory responses, did not show a significant difference between the two groups (0.97 ± 0.40 vs. 1.12 ± 0.50 fold change, p  = 0.235; mean difference: −0.15, 95% CI: −0.38 to 0.08, Cohen's d = −0.33) (Figure  2B ). Importantly, the IL‐17/IL‐10 ratio, representing a surrogate marker of Th17/Treg‐associated cytokine balance, was markedly elevated in RIF patients relative to controls (2.57 ± 1.28 vs. 1.07 ± 0.41, p  < 0.0001; mean difference: 1.50, 95% CI: 1.03–1.97, Cohen's d = 1.64) (Figure  2C ). Baseline expression of IL‐17, IL‐10, and the IL‐17/IL‐10 ratio in RIF patients and controls. (A) Bar graph showing significantly elevated IL‐17 expression in RIF patients compared with controls; (B) Bar graph showing no significant difference in IL‐10 expression between groups; (C) Bar graph illustrating a markedly increased IL‐17/IL‐10 ratio in RIF patients relative to controls. Data are presented as mean ± SD (ns = not significant; **** p  < 0.0001). Following tacrolimus treatment, a significant reduction in NK cell–associated subsets was observed. The percentage of CD16 + NK cells decreased from 13.79% ± 5.17% at baseline to 11.60% ± 4.88% post‐treatment ( p  = 0.011; mean difference: 2.19%, 95% CI: 0.28–4.10, Cohen's d = 0.44) (Figure  3A ). Similarly, CD56 + NK cells declined from 19.63% ± 5.11% to 16.59% ± 4.62% ( p  = 0.0014; mean difference: 3.04%, 95% CI: 0.56–5.52, Cohen's d = 0.62) (Figure  3B ). A comparable decrease was also noted in CD16 + CD56 + NK cells, dropping from 12.59% ± 4.96% to 10.15% ± 4.18% ( p  = 0.0015; mean difference: 2.44%, 95% CI: 0.09–4.79, Cohen's d = 0.53) (Figure  3C ). Changes in CD16 + , CD56 + , and CD16 + CD56 + NK cells before and after tacrolimus treatment in RIF patients. Paired plots comparing the percentage of CD16 + NK cells (A), CD56 + NK cells (B), and CD16 + CD56 + NK cells (C) before and after tacrolimus therapy. Data are presented as mean ± SD (* p  < 0.05; ** p  < 0.01). Tacrolimus treatment was associated with a significant reduction in inflammatory cytokine expression. The level of IL‐17 decreased from 1.03 ± 0.24 at baseline to 0.70 ± 0.23 after treatment ( p  < 0.0001; mean difference: 0.33, 95% CI: 0.21–0.45, Cohen's d = 1.40) (Figure  4A ). In contrast, IL‐10 expression remained unchanged between baseline and post‐treatment (1.07 ± 0.44 vs. 1.16 ± 0.61, p  = 0.406; mean difference: −0.09, 95% CI: −0.36 to 0.18, Cohen's d = −0.17) (Figure  4B ). Notably, the IL‐17/IL‐10 ratio, a surrogate marker of Th17/Treg‐associated cytokine balance, declined significantly following tacrolimus therapy (1.12 ± 0.56 vs. 0.74 ± 0.43, p  = 0.0036; mean difference: 0.38, 95% CI: 0.12–0.64, Cohen's d = 0.76) (Figure  4C ). Changes in IL‐17, IL‐10, and IL‐17/IL‐10 ratio before and after tacrolimus treatment in RIF patients. (A) Paired analysis showing a significant reduction in IL‐17 expression after tacrolimus; (B) No significant difference observed in IL‐10 expression before and after treatment; (C) Paired analysis showing a significant decrease in the IL‐17/IL‐10 ratio following tacrolimus therapy. Data are presented as mean ± SD (*ns = not significant; **** p  < 0.0001). To investigate the association between immunological alterations and clinical pregnancy outcomes, confirmed by ultrasound evidence of a gestational sac, further analyses were performed based on clinical pregnancy outcome. Among the 30 patients with RIF who received tacrolimus treatment, 12 (40%) achieved a positive clinical pregnancy outcome, whereas 18 (60%) did not achieve clinical pregnancy. For this analysis, immunological changes before and after tacrolimus treatment were quantified as percentage decrease (%Decrease), calculated as: %Decrease = [(Value before treatment−Value after treatment)/Value before treatment] × 100. As shown in Figure  5A , the reduction in CD16 + NK cells did not differ significantly between patients with positive and negative clinical pregnancy outcomes. Similarly, patients who achieved a positive clinical pregnancy outcome exhibited a greater reduction in CD56 + NK cells and CD16 + CD56 + NK cell subsets compared with patients without clinical pregnancy; however, these differences did not reach statistical significance (Figure  5B,C ). In contrast, the reduction in the IL‐17/IL‐10 ratio was significantly greater in patients with a positive clinical pregnancy outcome than in those without clinical pregnancy (Figure  5D , p  = 0.048). Collectively, these findings suggest a potential association between tacrolimus‐induced immunomodulation and implantation success. These changes may represent a candidate exploratory biomarker of clinical pregnancy outcome. However, these subgroup analyses were exploratory and underpowered due to the limited sample size and should therefore be interpreted with caution. Changes in NK cell subsets and IL‐17/IL‐10 ratio after tacrolimus treatment according to clinical pregnancy outcome in RIF patients. Percentage decrease of CD16 + NK cells (A), CD56 + NK cells (B), CD16 + CD56 + NK cells (C), and IL‐17/IL‐10 ratio (D) in tacrolimus‐treated RIF patients, stratified according to clinical pregnancy outcome. Among the 30 patients, 12 (40%) achieved a positive clinical pregnancy outcome, whereas 18 (60%) did not achieve clinical pregnancy. Percentage decrease (%Decrease) was calculated as: %Decrease = [(Value before treatment−Value after treatment)/Value before treatment] × 100. Data are presented as mean ± SEM (ns = not significant; * p  < 0.05).

This study provides novel mechanistic insight into how tacrolimus may enhance implantation and pregnancy outcomes in women with RIF by modulating immune cell subsets and cytokine regulation. Specifically, we demonstrate that tacrolimus reduces pro‐inflammatory NK cell–associated subsets and shifts the IL‐17/IL‐10 ratio toward a more tolerogenic profile. These immunological effects complement our previous clinical trial in the same cohort, in which tacrolimus treatment was associated with a clinical pregnancy rate of 42.5% versus 15% in controls and a live birth rate of 27.5% versus 7.5% in the placebo group [ 20 ], providing supportive clinical context rather than definitive evidence of efficacy. Our baseline analyses revealed that women with RIF exhibit immune dysregulation compared with fertile controls, which may reflect infertility‐associated immune alterations. The frequencies of CD16 + , CD56 + , and CD16 + CD56 + NK cell–associated subsets were markedly elevated in RIF patients, consistent with a heightened cytotoxic immune environment. In parallel, IL‐17 expression and the IL‐17/IL‐10 ratio were significantly increased, while IL‐10 levels remained unchanged. Together, these data indicate a pro‐inflammatory bias characterized by expansion of cytotoxic NK cell–associated subsets and Th17‐type cytokine dominance. Following low‐dose tacrolimus treatment, we observed significant reductions in the frequencies of all NK cell subsets examined, as well as a decrease in IL‐17 expression and IL‐17/IL‐10 ratio, without significant alteration of IL‐10 levels. These immunological changes suggest that tacrolimus selectively attenuates pro‐inflammatory pathways while preserving IL‐10–associated anti‐inflammatory signaling rather than implying Treg‐specific activity, a shift that may support a more favorable implantation environment [ 21 ]. Our findings are in line with and extend previous clinical studies reporting beneficial effects of tacrolimus in RIF patients with immune abnormalities. Earlier investigations primarily focused on the Th1/Th2 ratio as a marker of immune imbalance, showing that tacrolimus reduces Th1‐type responses and shifts the Th1/Th2 balance toward a more tolerant profile, which has been associated with improved implantation and pregnancy outcomes [ 19 , 22 , 23 , 24 ]. Moreover, these studies largely concentrated on clinical endpoints, providing limited mechanistic insight into other immunological pathways [ 22 ]. By examining both NK cell phenotypes and the Th17‐associated inflammatory pathway together with IL‐10–mediated regulatory signaling, our study extends this understanding and offers a more integrated view of how tacrolimus rebalances maternal immune responses. The elevated frequencies of CD56 + and CD16 + CD56 + NK cells observed in our RIF cohort are consistent with prior reports implicating aberrant NK cell activation in implantation failure [ 11 , 25 , 26 ]. Uterine CD56 + CD16 − NK cells, as described in decidual tissue studies, normally support trophoblast invasion and vascular remodeling, whereas peripheral‐type NK cells, including CD56 − CD16 + and CD56 + CD16 + , are more cytotoxic and secrete pro‐inflammatory cytokines such as IFN‐γ and TNF‐α. Excess recruitment or activation of these subsets can disrupt endometrial receptivity and lead to pregnancy loss [ 27 , 28 , 29 ]. Importantly, NK cells include a distinct CD56 bright CD16 − subset with key cytokine‐producing and immunoregulatory functions essential for endometrial receptivity and early pregnancy [ 25 ]. In this study, this subset was not separately quantified; therefore, CD56 + cells may represent a heterogeneous NK cell population. As CD3 staining was not included, the analyzed populations should be considered NK‐enriched rather than definitively defined NK cells, and should be interpreted with caution. Tacrolimus acts by inhibiting calcineurin signaling and suppressing T cell activation, but emerging evidence suggests it can also indirectly modulate NK cell activity [ 30 , 31 ]. Our data support this, showing significant reductions in CD16 + , CD56 + , and CD16 + CD56 + NK cell frequencies after treatment. This may reflect decreased activation and cytotoxicity of NK populations, leading to a more permissive systemic immune environment rather than direct uterine immune modulation, potentially supportive of implantation [ 22 , 32 , 33 ]. Importantly, these immune changes likely reflect coordinated modulation of innate and adaptive immunity. In our cohort, tacrolimus was associated with a shift in the Th1/Th2 balance toward a more tolerant profile, and together with reduced cytotoxic NK cells, may contribute to a less inflammatory systemic environment during the peri‐implantation period. These findings should not be interpreted as reduced efficacy of tacrolimus, but rather as evidence of partial immune modulation in the context of RIF. Such integrated immune modulation may indirectly improve endometrial receptivity and support implantation. Similarly, our cytokine findings highlight the role of the IL‐17/IL‐10 balance in implantation. Elevated IL‐17 and IL‐17/IL‐10 ratio in RIF patients at baseline are consistent with a shift toward Th17‐driven inflammation, which has been associated with impaired maternal‐fetal tolerance and implantation failure [ 13 , 14 , 34 ]. Tacrolimus treatment significantly reduced IL‐17 expression and the IL‐17/IL‐10 ratio, but IL‐10 levels remained stable. However, this response may vary among patients, reflecting immune heterogeneity in RIF and the potential presence of non‐responders. This pattern suggests that tacrolimus dampens pro‐inflammatory Th17 responses while maintaining IL‐10–associated anti‐inflammatory signaling, aligning with previous studies showing that tacrolimus suppresses Th17 differentiation and cytokine secretion [ 35 , 36 ]. Further analyses stratified by clinical pregnancy outcome revealed that patients who achieved a positive clinical pregnancy outcome exhibited greater reductions in CD56 + NK cells, CD16 + CD56 + NK cells, and notably in the IL‐17/IL‐10 ratio compared with patients without clinical pregnancy, although differences in NK subsets did not reach statistical significance. These findings suggest a potential association between tacrolimus‐induced immunomodulatory changes and implantation success, and may reflect underlying immune shifts linked to favorable reproductive outcomes in this cohort. In our clinical trial protocol, tacrolimus was administered from 2 days before embryo transfer until 14 days after, indicating that these effects occur within the peri‐implantation window. Despite these insights, several limitations warrant consideration. First, the relatively small sample size, particularly in subgroup analyses, limits statistical power and generalizability, and the findings should therefore be considered exploratory and hypothesis‐generating. Second, the absence of a placebo or untreated immunological control group limits causal attribution, as observed changes may also be influenced by time‐related factors, hormonal environment, embryo transfer, or early implantation‐related immune modulation. In addition, the use of healthy fertile women as controls does not fully account for infertility status, ART exposure, or hormonal treatment, and may therefore reflect broader infertility‐associated immune differences rather than RIF‐specific alterations. Peripheral blood was used instead of endometrial tissue, limiting direct interpretation of local uterine immune mechanisms, and NK cell subsets were defined without CD3 exclusion. Treg activity was also inferred indirectly from IL‐10. Finally, embryonic genetic status was not evaluated, which may represent a potential confounder. Accordingly, peripheral findings should be interpreted as systemic immune modulation rather than direct endometrial immune changes. Future larger studies incorporating uterine immune profiling and comprehensive immune phenotyping are warranted.

In conclusion, our findings provide preliminary mechanistic evidence suggesting that tacrolimus may be associated with modulation of immune pathways relevant to implantation by reducing pro‐inflammatory NK cell–associated subsets and shifting the IL‐17/IL‐10–associated cytokine balance toward a more tolerogenic profile. These immunological changes may contribute to a more favorable immune environment potentially supportive of implantation in women with RIF. This study supports the potential role of tacrolimus as an immunomodulatory agent in selected RIF patients with evidence of immune dysregulation and highlights the importance of integrating immunological assessment into personalized reproductive medicine. However, given the observational nature of the study, the absence of an immunological no‐treatment control group, and the lack of formal predictive analyses or robust clinical endpoints, these findings should be interpreted as exploratory and hypothesis‐generating and require validation in larger, well‐controlled studies incorporating uterine immune profiling and comprehensive clinical outcome assessment.

Despite significant advances in reproductive immunology, infertility remains a major global health challenge. Assisted reproductive technologies (ART), particularly in vitro fertilization (IVF) and embryo transfer (ET), have improved clinical outcomes in infertility treatment [ 1 ]. However, a subset of patients continues to experience recurrent implantation failure (RIF), defined as repeated failure to achieve pregnancy despite the transfer of high‐quality embryos [ 2 ]. The etiology of RIF is multifactorial and often unexplained, with dysregulation of the maternal immune response emerging as a principal contributor [ 2 , 3 ]. This persistent challenge underscores the need for targeted therapeutic strategies to manipulate immune imbalance and improve ART outcomes [ 4 ]. Successful embryo implantation requires a precisely balanced immunological environment. The maternal immune system must simultaneously tolerate the semi‐allogeneic embryo and support its development. Disruption of this balance can compromise implantation and early pregnancy maintenance [ 5 ]. Among immune cell populations, natural killer (NK) cells are critical during the peri‐implantation period and early placentation [ 6 ]. Distinct subsets of NK cells are identified by surface markers. CD56 + CD16 − NK cells have been described as the predominant NK cell subset in the decidua during early pregnancy, where they are involved in immune regulation and tissue remodeling. These cells secrete cytokines and growth factors that contribute to trophoblast invasion, spiral artery remodeling, and the establishment of endometrial receptivity [ 7 , 8 ]. In contrast, CD16 + NK cells are the dominant NK cell subset in peripheral blood and exhibit cytotoxic and pro‐inflammatory functions. Alterations in the balance of NK cell–associated subsets in peripheral blood have been suggested to be associated with adverse reproductive outcomes, including implantation failure [ 9 , 10 ]. In addition, CD56 + CD16 + NK cells represent a highly cytotoxic subset capable of producing pro‐inflammatory cytokines and mediating antibody‐dependent cellular cytotoxicity, and alterations in their frequency have been implicated in impaired implantation and pregnancy loss [ 10 , 11 ]. A complex interplay of pro‐ and anti‐inflammatory cytokines regulates implantation [ 12 ]. Pro‐inflammatory cytokines such as interferon‐γ (IFN‐γ), tumor necrosis factor‐α (TNF‐α), and interleukin‐17 (IL‐17) are essential for initiating the inflammatory responses required for tissue remodeling and placentation. However, dysregulated expression may contribute to pregnancy failure [ 13 , 14 ]. Anti‐inflammatory cytokines such as IL‐10 are equally critical for maintaining maternal tolerance to the embryo [ 15 ]. IL‐17, secreted primarily by T‐helper 17 (Th17) cells, promotes controlled inflammation at the maternal‐fetal interface, while IL‐10, produced by regulatory T cells (Treg) as well as other immune cells including Th2 cells and macrophages, suppresses excessive pro‐inflammatory responses, creating a tolerogenic environment favorable for implantation. Thus, the IL‐17/IL‐10 ratio, reflecting the Th17/Treg balance, has emerged as a key determinant of implantation success and early pregnancy maintenance [ 16 , 17 ]. Taking these considerations into account, immunomodulation has therefore been introduced as a therapeutic approach for RIF patients. Interventions such as intravenous immunoglobulin (IVIG), corticosteroids, and intralipid therapy have been trialed to enhance implantation by modulating immune responses [ 18 ]. Among these, tacrolimus, a calcineurin inhibitor widely used in organ transplantation, has shown promising results. Emerging clinical evidence suggests that low‐dose tacrolimus during IVF cycles can significantly improve implantation and live birth rates in women with RIF suffering immune dysregulation [ 17 , 19 ]. Nevertheless, the mechanistic pathways through which tacrolimus influences NK cell subsets and cytokine balance remain incompletely understood. We hypothesized that tacrolimus would reduce cytotoxic NK cell–associated subsets and shift the IL‐17/IL‐10 cytokine balance toward a more tolerogenic profile, thereby supporting improved implantation outcomes in women with RIF. The present study was therefore designed to evaluate the immunomodulatory effects of tacrolimus in patients with RIF. We examined NK cell phenotypic shifts defined by CD16 and CD56 expression, alongside changes in Th17/Treg‐associated cytokine dynamics (IL‐17 and IL‐10). By addressing these mechanisms, this study aims to provide new insight into how tacrolimus improves implantation outcomes and to support the development of tailored immunotherapeutic strategies in reproductive medicine.

The authors declare no conflicts of interest.