Prevalence of autoantibodies against VEGF-A and association with adverse outcomes of assisted reproductive technology programs in patients with infertility

The vascular endothelial growth factor (VEGF) family include VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E, VEGF-F, placental growth factor (PlGF),  and VEGFR-1, VEGFR-2, and VEGFR-3 receptors, that are  key mediators of vasculogenesis and angiogenesisis, and lymphangiogenesis [1, 2]. VEGF ligands and their receptors control endothelial cell proliferation, migration, and survival, playing a central role in dynamic vascular remodeling and angiogenesis [3]. VEGF-dependent mechanisms play important role in the female reproductive system,  since cyclic angiogenesis accompanies folliculogenesis, ovulation, corpus luteum formation, embryo implantation, and placentation [1–4]. Ovarian angiogenesis involves coordinated interaction between the network of signaling pathways and molecular factors, among which VEGF-A and VEGFR-2 are the most significant [4].

Experimental and clinical data indicate that disruption of VEGF-mediated angiogenesis is accompanied by suppression of ovulation, the corpus luteum dysfunction, impairment of steroidogenesis and decreased oocyte quality [2]. VEGF is secreted by granulosa and theca cells into follicular fluid. High VEGF level is associated with higher fertilization rate, good quality embryo, and increased chance of getting pregnant [3]. Inhibition of VEGF-dependent signaling pathways can impair corpus luteum vascularization and early embryonic development [4].

Dysregulation of angiogenesis is considered one of the mechanisms in the development of gynecological conditions associated with infertility. The authors reported increased VEGF expression in the eutopic endometrium and altered expression of VEGF receptors in endometriosis [5, 6]; increased VEGF level and hypervascularization of the ovaries in polycystic ovary syndrome [7, 8]; reduced VEGF level in primary ovarian insufficiency, as  potential components of pathogenesis [9]. VEGF also contributes to the processes of implantation and placentation. VEGF level is considered as the potential predictor of outcomes of assisted reproductive technology (ART) programs. Abnormal VEGF expression  is associated with recurrent implantation failures, recurrent miscarriage, non-viable pregnancy, and placenta-associated complications [3, 10, 11].

In recent years, exploration of the role of immunological mechanisms in gynecological conditions and assisted reproductive technology failures is of the most scientific interest [12, 13]. A wide range of autoantibodies against reproductive system hormones (gonadotropic and steroid hormones), steroidogenic enzymes, ovarian tissue and endometrial antigens have been identified in patients with endometriosis and primary ovarian insufficiency [14, 15]. At the same time, autoimmune mechanisms associated with the formation of autoantibodies against regulatory molecules of angiogenesis remain poorly understood [16, 17]. Since these antibodies are potentially capable of inhibiting the biological activity of VEGF or disrupting its interaction with receptors, the assessment of autoantibodies against VEGF-A, the most significant factor in ovarian angiogenesis, can be important for clarifying the immunopathogenesis of failures in ART programs.

In view of the above, the objective of this study was to assess the prevalence and levels of autoantibodies against VEGF-A in women with infertility and their association with outcomes of ART programs.

Materials and methods

This cross-sectional study included the patients with infertility who underwent treatment in 2021 using ART programs at the Department of Assisted Reproductive Technologies for Infertility Treatment of В.V. Leonov Institute of Reproductive Medicine (Head of Department – Kalinina E.A., MD, Professor), V.I. Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology of the Ministry of Health of Russia.  Based on analysis of clinical and anamnestic data, 2 groups were formed. Group 1 included the patients with external genital endometriosis (EGE) (n=42). Group 2 consisted of the patients without the diagnosis of EGE (n=36). The diagnosis of endometriosis was confirmed based on the results of ultrasound (US) scans and magnetic-resonance imaging, as well as by the results of visual evaluation of pelvic organs during laparoscopic surgery and pathomorphological examination.

Inclusion criteria in the groups were women aged 18–45 years, previously signed written voluntary informed consent to participate in the study, undergoing infertility treatment using ART. Non-inclusion criteria were contraindications for using ART, programs with donor oocytes or surrogacy, morbid obesity – body mass index (BMI) ≥40.0 kg/m², HIV infection, severe somatic, autoimmune and oncological diseases.

All women in the studied groups were examined according to the current clinical recommendations “Female infertility”. Ovarian stimulation was performed according to gonadotropin-releasing hormone (GnRH) antagonist protocol using recombinant follicle-stimulating hormone and/or luteinizing hormone-containing preparations. The gonadotropin dose was individually adjusted taking into account woman’s age, medical history, and ovarian reserve indicators.   Human chorionic gonadotropin or its combination with a GnRH agonist was used as an ovulation trigger. Following transvaginal puncture, the numbers of cumulus-oocyte complexes (COCs) and mature oocytes were assessed. Fertilization occurred using in vitro fertilization or intracytoplasmic sperm injection. Morphological assessment of embryos was performed using the Gardner grading system. One or two embryos were transferred into the uterine cavity using a soft catheter. Gestagens were used to support the post transfer period. The outcomes of ART programs included clinical pregnancy occurrence based on a positive beta human chorionic gonadotropin (β-hCG) blood test (about 14 days after embryo transfer) and ultrasound visualization of the gestational sac, as well as progressive intrauterine pregnancy that ended with live birth.

Blood samples for autoantibody testing were collected from patients on days 2–5 of the menstrual cycle in the follicular phase before starting ART program.  Serum samples were stored until testing at -80°C in biobank at the Laboratory for collection and storage of biomaterials, V.I. Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology of the Ministry of Health of Russia from 2021 to 2025.

Serum autoantibodies of M and G classes against VEGF-A were determined using a modified enzyme-linked immunosorbent assay (ELISA), that was developed at the Laboratory of clinical immunology, V.I. Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology of the Ministry of Health of Russia. Recombinant human VEGF-A expressed in Human Embryonic Kidney (HEK) 293 cell line (Sigma-Aldrich Co. LLC), mouse monoclonal antibodies against human IgM and IgG labeled with horseradish peroxidase, buffer solutions and ELISA reagents (LLC “CHEMA”, Russia) were used to perform ELISA. VEGF-A was adsorbed onto the polystyrene surface of the microplate wells with maximum sorption capacity at a concentration of 2.5 μg/mL in phosphate-buffered saline (PBS) (pH 7.4) at a temperature of 20±2°C for 16 hours. To block free binding centers, the surface of the wells was treated with 1% solution of bovine serum albumin (BSA) in PBS at a temperature of 20±2°C for 1.5 hours. Serum samples diluted 1:100 were tested in duplicate. Then conjugates against human IgM and IgG were added to the wells in working dilutions using PBS with 0.5% BSA and 0.05% Tween-20, and were incubated at a temperature of 20±2°C for 1 hour.  After performance of enzymatic reaction with chromogenic substrate solution and stopping it with stop reagent, the optical density (OD) of the solution in the wells was measured using the Infinite F50 photometer (TECAN, Austria) at a wavelength of 450 nm. The result of analysis was considered to be positive, if the mean of the OD value in the wells with the test sample exceeded the cut-off value (calculated as the mean OD value of the negative control + 0.2) by more than 1.25 times.

 Statistical analysis

Statistical data processing was performed using Microsoft Excel tables and software programs MedCalc v. 12 and Statistica v. 10. The normal distribution of the quantitative data in the groups was assessed using Kolmogorov–Smirnov test and Shapiro–Wilk W test. In normal distribution,the data was represented as the arithmetic mean and standard deviation – M (SD). In non-normal distribution, the data was represented as the median and interquartile range from the 25th percentile to the 75th percentile (Q1; Q3) or from the minimum value to the maximum value – Me (min; max). The differences between the groups were compared using the nonparametric Mann–Whitney U test. The differences between the groups in  the qualitative data, that were represented as absolute numbers (n) and relative values (%), were assess using Pearson’s chi-squared (χ²) test. Association between the risk factor and outcome was determined by calculation of the relative risk (RR) with 95% confidence interval (95% CI). The differences were considered statistically significant at p<0.05.

Results

The patients in the studied groups were comparable in age, BMI, gravidity, medical abortions and miscarriages in history, the prevalence and duration of primary and secondary infertility, anti-Müllerian hormone (AMH), the prevalence of male factor infertility, the number of ART programs in medical history, and the rate of the majority of somatic disorders (Table 1).

Group 1 comprised patients with EGE stages 2-3, including ovarian endometriosis in 21/42 patients (50%) and endometriosis of the pelvic peritoneum in 33/42 (78.6%). In this group of patients, periods lasted longer, the number of cases of ovarian resection was higher and gastrointestinal tract (GIT) diseases were more often compared to the women in group 2 (Table 1). However, in group 2, the prevalence of sexually transmitted infections (STIs) and the number of ectopic pregnancies was higher.  Due to this, in patients in group 2, the cases of tubectomy in history were registered significantly more often compared to group 1. Only 5/42 (11.9%) patients with EGE had a history of childbirth.

The patients in both groups were diagnosed with primary and secondary infertility. However, in the group with EGE, the rate of primary infertility was higher versus secondary infertility (p<0.01). A large number of patients in groups 1 and 2 had a history of ART failures (Table 1). Pregnancies occurred after using ART only in 2/42 (4.8%) patients in group 1 and in 3/36 (8,3%) in group 2.  Ovarian reserve was comparable in patients in the studied groups. Reduced AMH levels (<1.2 ng/ml) were in 7/42 (16.7%) patients in group 1 and in 3/36 (8.3%) in group 2 (p=0.30).

Serum autoantibodies of M and G classes against VEGF-A in patients in the studied groups were determined using a modified ELISA. In group 1, antibodies against VEGF-A (total IgM and IgG) were identified in 13/42 (31,0%) patients with EGE, whereas in group 2 slightly more often– in 15/36 (41.7%) (p=0.33). In women with endometriosis IgM antibodies against VEGF-A were identified more often versus IgG antibodies (p=0.06) (Table 2). In the group without EGE the frequency of IgM-antibodies and IgG-antibodies against VEGF-A did not differ significantly (p=0.17).

The two groups did not differ significanltly in the levels of IgM and IgG antibodies against VEGF-A (Table 3), though in group 2, the mean level of antibodies was slightly higher than in group 1. At the same time, the highest values of IgG-antibodies were obtained in group 2 in patients with a history of ectopic pregnancy and tubectomy.

In all patients included in the study (n=78), 10. 0 (7.0; 13.0) COCs, 8.0 (5.0; 10.3) mature oocytes, 6.0 (4.0; 9.3) zigotes, and 3.0 (1.0; 5.3) blastocysts were obtained on average; 2 of them (1.0; 4.0) were high-quality and good-quality blastocysts. The average fertilization rate was 0.92 (0.78; 1.00), the blastulation rate was 0.55 (0.26; 0.75). The clinical pregnancy rate was 20/78 (25.6%), the live birth rate was 13/78 (16.7%). The obtained average values were used as thresholds in comparison of embryological outcomes in patients in groups 1 and 2.

The number of obtained COCs and mature oocytes was comparable. However, the number of obtained zygotes below the average number for all women included in the study was more often observed in patients of group 1 with EGE compared to women in group 2 without EGE – 52.4% (22/42) of cases versus 30.6% (11/36), (p=0.05). Also, the blastulation rate was slightly lower in women with endometriosis than in the comparison group – 0.41 (0.20; 0.72) and 0.64 (0.33; 0.79), respectively, (p=0.07). The blastulation rate below the average (<0.55) was registered significantly more often – in 61.9% (26/42) of cases versus 38.9% (14/36), (p=0.04). The number of zygotes less than 6 was obtained more often – in 52.4% (22/42) of cases versus 30.6% (11/36), (p=0.05). Moreover, in patients without endometriosis, the total number of obtained blastocysts was significantly higher – 4.5 (2.0; 7.0) and 3.0 (1.0; 5.0), (p<0.05). The number of high-quality blastocysts was 2.0 (1.0; 4.0) and 1.0 (0.0; 2.0), (p=0.01), and the number of good-quality and high-quality blastocysts suitable for transfer was 3.0 (1.0; 4.0) and 1.5 (0.0; 3.0), (p=0.03). However, the differences in the clinical pregnancy rates (19.0% and 33.3%) and the birth rates (11.9% and 22.2%) between groups 1 and 2 were not statistically significant (p>0.05).

However, analysis of the outcomes of ART programs in seropositive and seronegative patients in the studied groups found the differences in clinical pregnancy and live birth rates (Table 4). Thus, low clinical pregnancy rate and low live birth rate was observed in seropositive women without the diagnosis of endometriosis compared to seronegative patients. Moreover, the risk of adverse clinical outcome of ART programs (absence of clinical pregnancy) in seropositive patients – 13/15 (86.7%) was 1.7 times higher compared to seronegative patients – 11/21 (52.4%), (RR = 1.7; 95% CI (1.1; 2.6), (p=0.03).

Comparison of embryological outcomes of ART programs in seropositive and seronegative patients found no significant differences between the studied groups. However, in seropositive patients in both groups, the trends toward less number of obtained good-quality blastocysts (B) in groups 1 and 2 (p=0.087; p=0.057); frequent absence of blastocysts suitable for transfer (a total of A and B) in group 1 (p=0.051); and the blastulation rate below the average in group 2 (p=0.18) were observed.

In total, in seropositive patients in both groups (n=28), the absence of good-quality blastocysts (B) was significantly more often – in 21/28 (75%) cases compared to seronegative patients (n=50) – in (25/50 (50%) cases, (p=0.03). In seropositive patients, the risk of absence of good-quality blastocysts (B) was 1.5 times higher than in seronegative patients (RR=1.5; 95% CI (1.1; 2.1), (p=0.02) (Table 4).

Analysis of clinical and anamnestic data of seropositive patients in group 1 (n=13) showed that besides being diagnosed with to endometriosis, the women were diagnosed with adenomyosis – 4/13 (30.8%), uterine fibroids –  5/13 (38.5%), endometrial polyps – 3/13 (23.1%), STIs – 2/13 (15.4%), endometritis – 2/13 (15,4%), adnexitis – 2/13 (15.4%), endocrine diseases –  2/13 (15.4%). The number of patients with a history of previous pregnancy was 4/13 (38.5%), that ended in spontaneous miscarriage (3/13) or medical abortion (1/13). Myomectomy was performed for 4/13 (38.5%) and tubectomy for 2/13 (15.4%) patients.  Duration of infertility in all women was from 3 to 12 years; 6/13 (46.2%) patients experienced failures in ART programs.

Among seropositive patients in group 2 (n=15), 5/15 (33,3%) women had a history of previous pregnancies, 4/15 (26.7%) had ectopic pregnancy, and 1/15 (6.7%) had spontaneous miscarriage. Polyps were diagnosed in 4/15 (26.7%) women, uterine fibroids in 2/15 (13.3%), endocrine diseases in 4/15 (26.7%); 4/15 (26.7%) women had a history of tubectomy; and 1/15 (6.7%) woman had a history of myomectomy. Duration of infertility in all women was from 2 to 10 years; 6/15 (40%) women experienced failures in ART programs.

Discussion

Our study assessed the prevalence of autoantibodies against VEGF-A and their possible impact on the outcomes of assisted reproductive technology cycles in patients with endometriosis-associated infertility (group 1) and in patients with other types of female infertility (group 2), including tubal factor infertility, uterine factor infertility, and unexplained infertility. The patients in group 2 had no EGE. It is noteworthy that patients in group 1 often had a history of endometrioid ovarian cysts, adenomyosis, and ovarian resections, while in group 2, the rates of STIs, ectopic pregnancies, and tubectomy were higher. High frequency of fibroids were in both groups. The patients were diagnosed both with primary and secondary infertility.  Infertility rates did not differ significantly between the groups, but primary infertility was more common in group 1 versus secondary infertility. It should be noted that in both groups the frequency of previous failures in ART cycles was high. Only 11.9% of patients in group 1 experienced childbirth.

Comparative analysis of embryological outcomes in the observed ART cycles found several trends and significant differences between the groups. The number of COCs and mature oocytes obtained in both groups was comparable. However, in   patients with EGE, the number of obtained zygotes was below the average value. The blastulation rate was lower than the average rate (<0.55) and the number of zygotes was less than 6. It should be noted that fewer blastocysts, high-quality blastocysts (A), and blastocysts suitable for transfer (A+B) were obtained in patients with EGE. However, clinical pregnancy and live birth rates did not differ significantly between the two groups.

Currently, the scientific literature provides conflicting data on the impact of endometriosis on the outcomes of ART programs [18–21]. The results obtained by us are consistent with the data by a number of researchers, who reported that endometriosis in some cases was associated with poor embryological outcomes,  including the quality of oocytes and embryos, and fertilization rate, that can be explained by impaired energy metabolism, oxidative stress, mitochondrial dysfunction and inflammatory remodeling  of the ovarian microenvironment in endometriosis [18]. However, according to the data in systematic reviews and meta-analyses, no significantly reduced clinical pregnancy and live birth rates were observed in women with endometriosis undergoing ART treatment compared to those without endometriosis [18–20]. Further research is needed to elucidate molecular mechanisms underlying these effects and to develop therapeutic strategies aimed at improving the effectiveness of ART in women with endometriosis.

Given the important impact of angiogenesis factors, primarily VEGF, on folliculogenesis, oocyte quality, corpus luteum function, implantation, placentation, and the potential for formation of autoantibodies that can inhibit their biological activity and binding to specific receptors, we developed an ELISA-based method for detecting antibodies against VEGF-A and investigated the prevalence of these autoantibodies in patients with different types of infertility. The patients with EGE were of particular interest, since high VEGF level in endometriosis is one of the mechanisms underlying its pathogenesis.

This study showed high detection rate of autoantibodies of M and G classes in total against VEGF-A in both groups of patients with infertility, both in women with and without EGE. At the same time, no significant difference was found between the groups in the detection rate. In seropositive patients in both groups. obstetric and gynecological history was burdened and comprised high rates of gynecological and endocrine diseases, myomectomy and tubectomy, and failed ART attempts. It should be noted that in patients with endometriosis, IgM antibodies against VEGF-A were detected significantly more often than IgG antibodies, that can reflect the immune response to high VEGF level in endometriosis and its possible transient nature. Also, no significant difference was found between the groups in the levels of autoantibodies of M and G classes against VEGF-A.

The obtained results were of great scientific interest given the important biological role of VEGF in the female reproductive system, since, according to modern concepts, the presence of autoantibodies against VEGF can potentially be associated with impaired local angiogenesis supporting oocyte maturation, early embryonic development, implantation and the course of the early stage of pregnancy [2-4]. Thereby, analysis of embryological and clinical outcomes in the observed ART cycles was conducted in anti-VEGF-A antibody seropositive and seronegative patients. In total, absence of good-quality blastocysts (B) was observed more often in seropositive patients (75%) in the studied groups, compared to seronegative patients (50%). At the same time, the risk of absence of good-quality blastocysts (B) in seropositive patients was 1.5 times higher than in seronegative patients. Moreover, in seropositive patients with endometriosis, the absence of blastocysts suitable for transfer into the uterine cavity (A+B) was observed more often compared to seronegative patients. Given possible pathogenetic mechanisms, the identified association between autoantibodies against VEGF-A and the absence of good-quality blastocysts can be considered as a result of dysregulated angiogenesis and impaired female immune system, affecting folliculogenesis, oocyte quality, and embryo development [2, 3], and autoantibodies against VEGF-A can serve as a potential biomarker of these impairments [16].

It is important to note that in the group of women without endometriosis, the presence of antibodies against VEGF-A was associated with lower clinical pregnancy rate. The risk of adverse outcomes in ART programs was 1.7 times higher in seropositive patients compared to seronegative women. In this case, autoantibodies against VEGF-A can be considered as a potential biomarker of reproductive failures.  At the same time, in the group of women with EGE, the presence of antibodies against VEGF-A was not associated with significant worsening of clinical outcomes, namely, pregnancy and live birth rates. The obtained results can be probably explained by the complex, multifactorial nature of endometriosis-associated infertility and numerous confounding factors. These findings are also important for understanding the immunopathogenesis of endometriosis. Current data indicate that VEGF is a key mediator of endometriosis-associated angiogenesis [22]. Increased VEGF level in peritoneal fluid and ectopic endometrioid tissue is observed in endometriosis. It contributes to the maintenance of angiogenic microenvironment, can lead to a transient increase in the level of IgM antibodies against VEGF. In genetically predisposed individuals, antibodies can circulate in the blood for a long time, have pathogenetic activity and have impact on the outcomes of ART programs.

From a practical point of view, the results of the study open the possibility to stratify the patients with infertility and recurrent ART failures based on the presence of autoantibodies against VEGF-A. Detection of these autoantibodies is of particular importance for patients with unexplained infertility, whose clinical and laboratory parameters do not indicate adverse outcomes of ART programs. The presence of anti-VEGF-A autoantibodies in these patients suggests autoimmune mechanism of recurrent failures in ART cycles.

Limitations. Limitations of this study include a small sample size, the absence of comparison group of fertile women, impossibility to completely exclude the influence of confounding factors, single determination of the level of autoantibodies against VEGF-A without monitoring the changes in its level in dynamics with subsequent determination at least in 12 weeks.

Conclusion

Thus, this study demonstrated high prevalence of autoantibodies against VEGF-A both in patients with endometriosis-associated infertility, tubal factor infertility and uterine factor infertility undergoing ART programs.  Furthermore, the presence of autoantibodies against VEGF-A is associated with the trends toward worsening of some of the indicators of embryological outcomes and decline in pregnancy and birth rates. The obtained preliminary data give reason for further exploration of autoantibodies against regulatory molecules of angiogenesis to identify potential biomarkers and understand the mechanisms leading to adverse reproductive outcomes, as well as to improve the effectiveness of ART programs.