The endometrium is a complex, highly differentiated tissue lining the uterine cavity, with the unique ability to undergo cyclic regeneration, differentiation, and desquamation throughout a woman's reproductive period [1]. Endometrial function is under strict control of ovarian steroid hormones, among which estrogens play a key role [1, 2]. Disruption of hormonal regulation of the endometrium leads to a wide spectrum of pathological conditions, including hyperplastic processes, endometriosis, and malignant neoplasms, and is also one of the causes of reduced endometrial receptivity in in vitro fertilization (IVF) programs [1].

The effectiveness of estrogen action in the endometrium is determined not only by hormone concentrations in the blood but also by the rate of their biotransformation, governed by enzyme activity [1]. The first phase of oxidative detoxification of estrogens is provided by isoenzymes of the cytochrome P450 (CYP) superfamily—CYP1A1, CYP1A2, and CYP1B1 [3]. Individual differences in the functioning of these enzymes, caused by genetic polymorphism, can lead to substantial variability in endometrial response to stimulation in IVF programs [3–5].

CYP1A1, CYP1A2, and CYP1B1 hydroxylate estradiol and estrone to form metabolites (2-hydroxyestrone, 2-hydroxyestradiol, 4-hydroxyestrone) that differ in estrogenic activity and genotoxic potential [3]. In the liver, CYP1A2 plays the primary role in 2-hydroxylation of estradiol, while CYP1A1 exhibits similar activity in extrahepatic tissues. CYP1B1 is characterized by high expression in estrogen-sensitive tissues and specifically catalyzes the formation of 4-hydroxyestrogens, which possess genotoxic potential [6].

The protocol of controlled ovarian stimulation (COS) used in IVF programs involves the administration of exogenous gonadotropins in doses exceeding physiological levels to trigger the growth of multiple follicles, which is predictably accompanied by excessive estradiol secretion. The resulting steroid overload modifies the endometrial maturation process and promotes a shift in the period of maximum uterine receptivity—the so-called "implantation window" [7, 8].

Understanding how estrogens regulate proliferative processes in the endometrium has direct practical significance, particularly for IVF programs, given the widespread problem of insufficient endometrial growth ("thin" endometrium) when using standard hormonal support regimens.

To determine the association of polymorphic variants of estrogen metabolism enzyme genes—CYP1A1, CYP1A2, and CYP1B1—with endometrial response to pharmacotherapy in IVF programs in women with anovulatory infertility.

Study design and participants. A prospective observational pilot clinical study incorporating pharmacogenetic analysis, approved by the local ethics committee of the National Research Ogarev Mordovia State University (Protocol No. 116 dated 12.05.2023), included 265 patients with female infertility associated with anovulation (ICD-10 code: N97.0), aged 25–35 years (30.5±2.5 years). Enrollment was conducted at the Assisted Reproductive Technologies (ART) Department of the Perinatal Center of the Mordovian Republican Central Clinical Hospital (Saransk) from May 2023 to January 2025. All participants signed informed consent to participate in the study.

Selection criteria. After applying inclusion and exclusion criteria, the final sample comprised 96 patients. Inclusion criteria: age 25–35 years; documented anovulatory infertility (ICD-10 code: N97.0); normal ovarian reserve; absence of endometrial pathology on ultrasound and hysteroscopy; normal karyotype of the couple. Exclusion criteria: male factor infertility; BMI >30 kg/m²; smoking; extragenital or genital pathology contraindicated for IVF program; patient refusal to participate.

IVF protocol and endometrial assessment. All participants received standardized treatment according to a short GnRH antagonist protocol (ganirelix or cetrorelix 0.25 mg/day subcutaneously) and recombinant follicle-stimulating hormone (FSH) (follitropin alfa, starting dose 100–150 IU/day subcutaneously with dose adjustment). The trigger for final oocyte maturation was choriogonadotropin alfa (250 μg subcutaneously). Patient management followed national clinical guidelines for the treatment of female infertility and the use of ART.

The study endpoint was endometrial thickness (mm) on the day of ovulation trigger administration as assessed by ultrasound, serving as a marker of proliferative endometrial response to stimulation. Threshold values: thin—less than 7.0 mm; optimal—7–14 mm; hyperplastic—greater than 14 mm. Eight women were diagnosed with "thin endometrium," and 88 had optimal thickness.

Genotyping. Venous blood (6 mL) was collected in EDTA tubes and stored at –20 °C (short-term) and –70 °C (long-term). DNA was extracted using the magnetic-sorbent method (Genotek, Russia) on an Allsheng Auto-Pure 96 automated station (China). Genotyping was performed using genome-wide analysis on Infinium GSA-24 v3.0 biochips (Illumina, USA) with the iScan system according to the Infinium HTS Assay protocol [9] at the GENOTEK LLC laboratory. Thirteen single nucleotide polymorphisms (SNPs) were analyzed, selected based on a systematic literature review as markers with potential prognostic significance for IVF outcomes: CYP1A1—rs1048943, rs1800031, rs2606345, rs4646903; CYP1A2—rs762551, rs2069514, rs2470890; CYP1B1—rs1056836, rs1056837, rs10012, rs1056827, rs2855658, rs2617266. All stages of genetic analysis, including DNA extraction, genotyping, and biomaterial storage, were conducted exclusively within the territory of the Russian Federation. No cross-border transfer of genetic data occurred, fully complying with the requirements of Federal Law No. 86-FZ.

Statistical analysis. Statistical data processing was performed using Statistica 12.5 software (StatSoft, USA). Normality of quantitative trait distribution was tested using the Shapiro-Wilk test; data are presented as arithmetic mean and standard deviation (M±SD). For intergroup comparisons of quantitative variables, Student's t-test was used. To control for group-wise type I error probability in multiple testing, the Benjamini-Hochberg false discovery rate (FDR) correction was applied. To assess the independent contribution of the studied genetic variants and to mitigate the potential influence of confounding variables, multiple regression analysis was employed. Statistical power calculation was performed using G*Power software (version 3.1). With a sample size of n=96, a significance level of α=0.05, and the observed difference in endometrial thickness between homo- and heterozygotes (on average 0.8–1.2 mm, standardized effect size Cohen's d >0.6), the calculated statistical power (1-β) for key associations exceeded 80%, indicating that the sample size was sufficient to detect clinically significant effects. The critical level of statistical significance for null hypothesis testing was set at p <0.05.

Assessment of allele and genotype frequency distribution for conformity with the Hardy-Weinberg equilibrium was performed for all 13 analyzed polymorphic markers (Table 1). As the calculations showed, in the studied cohort (n=96), the actual genotype distribution across all studied loci of the CYP1A1, CYP1A2, and CYP1B1 genes did not significantly deviate from theoretically expected values (χ² test, p >0.05 in all cases). These data objectively indicate that the examined sample is in a state of population equilibrium. This, in turn, confirms the correctness of the genotyping stage and allows for the exclusion of potential systematic errors that could be associated with population stratification.

Table 1. Hardy-Weinberg equilibrium for the studied polymorphisms

GenePolymorphismGenotypesχ²p-value HWECYP1A1rs4646903A/A (n=81)A/G (n=15)G/G (n=0)0.6890.406 rs2606345C/C (n=6)C/A (n=38)A/A (n=47)0.2070.649 rs1048943T/T (n=91)T/C (n=5)C/C (n=0)0.06860.793 rs1800031A/A (n=96)A/G (n=0)G/G (n=0)——CYP1A2rs762551C/C (n=11)C/A (n=39)A/A (n=46)0.3810.537 rs2470890C/C (n=17)C/T (n=45)T/T (n=34)0.09860.753 rs2069514G/G (n=93)G/A (n=3)A/A (n=0)0.02430.876CYP1B1rs1056836C/C (n=22)C/G (n=48)G/G (n=26)0.000260.987 rs1056837A/A (n=22)A/G (n=48)G/G (n=26)0.000260.987 rs10012G/G (n=47)C/G (n=38)C/C (n=11)0.59550.440 rs1056827C/C (n=47)C/A (n=38)A/A (n=11)0.59550.440 rs2855658T/T (n=22)C/T (n=48)C/C (n=26)0.000260.987 rs2617266G/G (n=47)A/G (n=38)A/A (n=11)0.59550.440

Notes: χ² — Pearson's chi-square; p-value HWE — p-value of Hardy-Weinberg equilibrium.

In the pilot study, an assessment was made of the association of polymorphic variants of the CYP1 family enzyme genes (CYP1A1, CYP1A2, CYP1B1) with endometrial thickness on the day of ovulation trigger administration in patients with anovulatory infertility. Statistically significant differences between carriers of different genotypes were identified for several polymorphic loci (Table 2).

In the CYP1A1 gene, for polymorphism rs2606345, differences were found between the C/A and A/A genotypes (9.1±0.8 mm vs. 8.6±0.6 mm, p=0.003).

In the CYP1A2 gene group, for polymorphism rs762551, genotype-dependent gradation was observed: endometrial thickness sequentially increased from the C/C genotype (8.2±0.3 mm) to C/A (8.4±0.8 mm) and A/A (9.1±0.7 mm), with differences between C/C and A/A, as well as between C/A and A/A, being statistically significant (p <0.001). A similar pattern was observed for rs2470890, where the C/C homozygous genotype (8.1±0.2 mm) was associated with significantly lower endometrial thickness compared to both C/T (8.8±0.9 mm, p=0.007) and T/T (9.0±0.5 mm, p <0.001).

In the CYP1B1 gene, statistically significant differences were recorded for three polymorphisms: rs10012, rs1056827, and rs2617266, with identical data for these loci. Carriers of the homozygous genotype C/C rs10012, A/A rs1056827, and A/A rs2617266 had the greatest endometrial thickness—9.4±0.5 mm—compared to carriers of heterozygous genotypes (8.6±0.4 mm, p <0.001) and carriers of the G/G rs10012, C/C rs1056827, and G/G rs2617266 genotypes (p=0.015).

For the remaining studied polymorphic variants, no statistically significant differences in endometrial thickness between groups of carriers of different genotypes were found (p >0.05).

Among women with "thin" endometrium (n=8), all patients were carriers of the C allele of rs762551 and rs2470890 of the CYP1A2 gene, the G/G genotype rs2069514, the T/T genotype rs1048943, the A/A genotype rs4646903 of the CYP1A1 gene, the G/G genotype rs10012, C/C genotype rs1056827, and G/G genotype rs2617266 of the CYP1B1 gene. However, due to the limited number of observations, these data should be considered preliminary and requiring confirmation in larger samples.

The constructed regression model confirmed that CYP1A2 gene polymorphisms determine changes in endometrial thickness independently of baseline clinical and anthropometric characteristics. The presence of the A/A genotype at locus rs762551 was found to be an independent factor associated with increased endometrial thickness (unstandardized coefficient β=0.82; 95% CI: 0.35–1.29; β=0.41; p=0.002). Conversely, carriage of the C/C homozygous variant rs2470890 was a significant independent predictor of insufficient endometrial growth (β = –0.85; 95% CI: –1.31 to –0.39; β = –0.44; p=0.001).

Table 2. Endometrial thickness in carriers of different genotypes of CYP1A1, CYP1A2, CYP1B1 polymorphisms (M±SD, mm)

GenePolymorphismGenotypeEndometrial thickness, mmp-valueCYP1A1rs2606345C/A (n=38)9.1±0.8   A/A (n=47)8.6±0.60.003CYP1A2rs762551C/C (n=11)8.2±0.3   C/A (n=39)8.4±0.80.072*  A/A (n=46)9.1±0.7<0.001** rs2470890C/C (n=17)8.1±0.2   C/T (n=45)8.8±0.90.007*  T/T (n=34)9.0±0.5<0.001**CYP1B1rs10012G/G (n=47)8.3±0.8   C/G (n=38)8.6±0.40.015  C/C (n=11)9.4±0.5<0.001 rs1056827C/C (n=47)8.3±0.8   C/A (n=38)8.6±0.40.015  A/A (n=11)9.4±0.5<0.001 rs2617266G/G (n=47)8.3±0.8   A/G (n=38)8.6±0.40.015  A/A (n=11)9.4±0.5<0.001

*Notes: *compared to C/C genotype; **compared to C/C and C/A genotypes; p-values calculated using Student's t-test.

Despite the novelty of the obtained data, the study limitations must be considered. The relatively small sample size (n=96) does not allow for confident extrapolation of the results to the entire population, requiring confirmation in larger cohorts. Furthermore, the functional significance of the identified polymorphic variants remains unexplored. For a definitive conclusion on causality, additional functional studies (e.g., assessment of CYP1B1 activity in endometrial cell cultures with different genotypes) and replication of results in independent samples are necessary.

Endometrial thickness on the day of ovulation trigger administration is an important marker for assessing endometrial readiness for embryo transfer in IVF programs and a key element for evaluating pharmacogenetic influences. The found associations of the rs2470890 and rs762551 CYP1A2 gene polymorphisms and rs10012, rs1056827, rs2617266 CYP1B1 gene polymorphisms with endometrial thickness on the day of ovulation trigger administration indicate that genetically determined features of estrogen metabolism may directly affect the proliferative response of the endometrium under hyperestrogenic conditions induced by COS.

The enzyme CYP1A2 is predominantly expressed in the liver and is responsible for the hydroxylation of estradiol and estrone to form 2-hydroxyestrogens, which have significantly lower biological activity than estradiol itself [3]. Two polymorphisms—rs762551 and rs2470890—are located in the intronic regions of the CYP1A2 gene and may modulate the activity of the encoded enzyme, thereby influencing systemic estrogen metabolism. The found associations of these polymorphisms with endometrial thickness indicate that genetically determined hepatic clearance of estradiol may directly determine the proliferative response of uterine tissue under COS conditions.

The rs762551 (-163C>A) polymorphism of the CYP1A2 gene is an important marker of basal enzyme activity. The peculiarity of this locus lies in its high sensitivity to external inducers, such as smoking [10]. In the absence of induction, carriers of the A/A genotype exhibit lower basal enzyme activity ("slow metabolizers") compared to carriers of the C allele [10, 11]. In our study, patients with the A/A genotype had the greatest endometrial thickness. From a pharmacokinetic perspective, this is easily explained: in non-smoking women with the A/A genotype, the hepatic clearance of estradiol synthesized by growing follicles is reduced. The hormone circulates longer in the blood, reaches higher systemic concentrations, and provides a more prolonged effect on uterine receptors. In turn, the C allele (A/C and C/C genotypes) is associated with increased basal transcription of the enzyme [12, 13]. In such patients, systemic clearance of estradiol occurs more rapidly; therefore, the stimulatory effect of estrogens on the endometrium may be weaker at the same dose of recombinant FSH. Thus, slow metabolism in A/A patients provides better estrogenic support for the endometrium in IVF protocols.

The regulatory polymorphism rs2470890 (-1545C>T) deserves special attention. In our sample, the C/C homozygous genotype was associated with significantly lower endometrial thickness compared to heterozygotes and T allele homozygotes. The literature contains isolated data (from a study in a Chinese population) linking the C/C genotype with lower CYP1A2 protein expression [14]. However, the clinical phenotype recorded in our study ("thin" endometrium) indicates that under COS conditions, the C/C genotype manifests as a profile of accelerated metabolism leading to insufficient estrogenic stimulation of the uterine receptor apparatus. This discrepancy with the literature may be explained by population specificity. It is likely that in the European population, the C allele of the rs2470890 polymorphism is in strong linkage disequilibrium with other high-activity promoter variants of the CYP1A2 gene, leading to accelerated estradiol elimination. Thus, carriage of the C/C genotype rs2470890 in the studied cohort serves as a prognostic marker of risk for insufficient endometrial growth due to forced systemic estrogen clearance.

CYP1B1 is the only enzyme that most efficiently performs 4-hydroxylation of estradiol [6]. This metabolic pathway is of fundamental importance, since 4-hydroxyestrogens, unlike 2-hydroxymetabolites, retain high affinity for estrogen receptors and possess pronounced estrogenic activity. Furthermore, during redox cycling, 4-hydroxyestrogens form semiquinones and quinones capable of causing DNA damage; however, for physiological endometrial proliferation, it is precisely their receptor-mediated action that is key [15, 16].

This study is the first to identify an association of the homozygous genotypes C/C rs10012, A/A rs1056827, and A/A rs2617266 of the CYP1B1 gene with the greatest endometrial thickness on the day of ovulation trigger administration in patients with anovulatory infertility in IVF programs. Notably, for three different polymorphic loci, an identical distribution pattern of endometrial thickness was obtained, which may indicate either complete linkage disequilibrium between them or a combined effect on enzyme function.

Given that rs10012 and rs1056827 are missense polymorphisms leading to amino acid substitutions Arg48Gly and Ala119Ser, respectively, and rs2617266 is located in the 5'-untranslated region 12 bp upstream of the first coding exon, it can be assumed that they are in linkage disequilibrium. This is supported by literature data on the formation of haplotypes within the CYP1B1 gene, combining, in particular, rs2617266, rs10012, and rs1056827 [17].

Thus, the combination of genotypes C/C rs10012, A/A rs1056827, and A/A rs2617266 likely forms a haplotype characterized by reduced local CYP1B1 activity in the endometrium, leading to decreased formation of genotoxic 4-hydroxymetabolites and local accumulation of active estradiol, providing prolonged receptor stimulation and enhanced endometrial proliferation. During stimulation, when estradiol levels are very high, this haplotype more actively converts it to a protective, less active metabolite (2-hydroxyestradiol) and does not overload the endometrium with toxic 4-hydroxyestradiol. This leads to a reduction in local "hormonal stress" in the endometrium and creates a more favorable environment for proliferation, manifesting as increased thickness.

The data obtained for CYP1B1 are consistent with the results for CYP1A2, creating a unified picture of estrogen balance. Competitive metabolism of estradiol involving CYP1A2 (predominantly in the liver) and CYP1B1 (locally in the endometrium) may determine the balance of active and inactive metabolites, influencing the proliferative response.

The results of this study showed that endometrial thickness on the day of ovulation trigger administration is associated with polymorphisms rs762551 and rs2470890 of the CYP1A2 gene, and rs10012, rs1056827, and rs2617266 of the CYP1B1 gene, indicating the potential significance of these genetic markers for personalizing endometrial preparation protocols in IVF cycles.

In the examined patients with anovulatory infertility, associations of CYP1A2 and CYP1B1 polymorphisms with endometrial thickness were identified, which may indicate the role of genetically determined estrogen metabolism in shaping the proliferative response. While CYP1A2 primarily determines systemic estrogen levels, CYP1B1 acts locally, directly in the endometrium. It can be hypothesized that optimal endometrial proliferative response requires a fine balance between systemic hormonal stimulation and local metabolic control. Any deviation in the functioning of this complex system—disruption of systemic or local metabolism, or local estrogen conversion—may lead to a reduced proliferative response. The obtained data complement current understanding of the role of genetically determined features of estrogen metabolism in the regulation of endometrial proliferative activity under controlled ovarian stimulation conditions.