Anovulatory infertility is one of the most common endocrine causes of female reproductive dysfunction. When conservative hormone therapy is ineffective, in vitro fertilization (IVF) with controlled ovarian stimulation (COS) becomes the treatment of choice. According to current data, IVF efficacy varies widely among individual cases, and a substantial proportion of patients fail to achieve clinical pregnancy even when standardized stimulation protocols are used [1, 2]. One promising direction in personalized medicine is the search for genetic markers associated with treatment response [3–5]. Of particular interest are genes of the cytochrome P450 (CYP) family, which encode enzymes involved in the metabolism of exogenous and endogenous substrates, including sex steroids and drugs used in IVF protocols [6–8]. It has been shown that CYP gene polymorphisms may influence the individual pharmacokinetics of gonadotropins as well as the endogenous hormonal milieu, collectively determining the efficacy of ovarian stimulation and the likelihood of pregnancy [3, 9–11].

From the perspective of reproductive medicine, the isoenzymes CYP3A4 and CYP3A5, which participate in the biotransformation of progestins, are of particular importance, as are CYP2D6, CYP2C9, and CYP2C19, which are additionally involved in this process [12–14]. CYP2A6 is a steroid-sensitive gene whose activity is modulated by estrogen and progesterone [15, 16]. At the same time, data on the role of these CYP isoenzyme genes in predicting IVF outcomes remain fragmentary and require further investigation.

To evaluate the association of polymorphic variants of the genes CYP2A6, CYP2C9, CYP2C19, CYP2D6, CYP3A4, and CYP3A5 with the achievement of clinical pregnancy in IVF programs in patients with anovulatory infertility.

This work was conducted as part of a prospective pilot cohort study carried out at the Assisted Reproductive Technology (ART) Unit of the Perinatal Center of the Mordovian Republic Clinical Hospital No. 1 (GBUZ RM "MRTsKB") during the period from May 2023 to January 2025. The study protocol was approved by the Local Ethics Committee of the National Research Ogarev Mordovia State University (Protocol No. 116 dated 12.05.2023). All participants provided written informed consent to participate in the study. Initial screening was conducted among 265 patients with a verified diagnosis of anovulatory infertility (ICD‑10: N97.0); after applying inclusion and exclusion criteria, the final analytical sample comprised 96 patients.

Inclusion criteria: age 25–35 years; documented infertility due to absence of ovulation, verified on the basis of clinical and laboratory data; normal ovarian reserve; absence of endometrial pathology according to ultrasound and hysteroscopy; normal karyotype in both partners; transfer of embryos graded 3–6 with inner cell mass and trophectoderm quality not lower than A/B according to the Gardner scale.

Exclusion criteria: male factor infertility; body mass index >30 kg/m²; extragenital or genital pathology constituting an absolute contraindication to IVF; smoking; voluntary withdrawal of the patient from participation at any stage of the study.

The mean age of participants was 30.5 ± 2.5 years; all patients were of Caucasian ethnicity. All participants underwent a standardized IVF program using their own oocytes. Ovarian stimulation was performed using a short protocol with gonadotropin‑releasing hormone antagonists: ganirelix or cetrorelix at a dose of 0.25 mg/day subcutaneously. Recombinant follicle‑stimulating hormone (FSH) — follitropin alfa — was used as the gonadotropic agent at a starting dose of 100–150 IU/day subcutaneously, with subsequent individual dose adjustment based on folliculogenesis dynamics. The trigger for final oocyte maturation was choriogonadotropin alfa at a dose of 250 μg subcutaneously. Patient management was carried out in accordance with current National Clinical Guidelines for the treatment of female infertility and the use of assisted reproductive technologies.

When forming the study groups, the endpoint was the occurrence of clinical pregnancy, defined as echographic visualization of a gestational sac on day 21 after embryo transfer. According to the IVF cycle outcome, all patients were divided into two groups: Group 1 (main, n = 48) — patients who achieved clinical pregnancy after their first IVF attempt; Group 2 (comparison, n = 48) — patients with three or more failed IVF cycles in their history.

Venous blood sampling (6 mL) was performed on an empty stomach from the cubital vein on the day of patient enrollment into the IVF protocol. Biomaterial collection was carried out into disposable sterile vacuum tubes with EDTA anticoagulant. Samples were stored at −20 °C until extraction; long‑term storage was performed at −70 °C. Genomic DNA extraction was performed using magnetic sorbent reagents (Genotek, Russia) on an Allsheng Auto‑Pure 96 automated platform (China) according to the manufacturer's protocol. Genotyping was performed using whole‑genome typing on Infinium Global Screening Array‑24 v3.0 biochips (Illumina, USA) with an iScan scanning system according to the Infinium HTS Assay protocol [17]. This paper presents the results of the analysis of CYP family gene polymorphisms included in the genotyping panel: CYP2A6 (rs1809810, rs7259706, rs7260629, rs8192733, rs12459249, rs28399433, rs56113850, rs57897628, rs113288603), CYP2C9 (rs1057910, rs1799853), CYP2C19 (rs4244285, rs4986893, rs12248560), CYP2D6 (rs1065852, rs3892097, rs35742686), CYP3A4 (rs35599367), and CYP3A5 (rs776746).

Statistical data analysis was performed using Statistica 12.5 software (StatSoft, USA). The Shapiro–Wilk test was used to assess the conformity of quantitative variables to a normal distribution; data are presented as M ± SD. Intergroup comparisons of quantitative traits were performed using Student's t‑test. Comparison of genotype and allele frequencies between groups was performed using the Pearson χ² test. In cases where the expected number of observations in one of the contingency table cells was less than 5, Fisher's exact test was used. The conformity of observed genotype frequencies to Hardy–Weinberg equilibrium (HWE) was assessed in the comparison group using the χ² method. The strength of association of polymorphic variants with IVF outcome was expressed as odds ratios (OR) with calculation of 95% confidence intervals (95% CI). The threshold for statistical significance was set at p < 0.05. Additionally, given the limited sample size (n = 96), a permutation test (Monte Carlo procedure with 10,000 phenotype‑label permutations under fixed genotype structure) was used for additional verification of the identified associations. This method provides an empirical estimate of significance level (p‑perm) independent of normality assumptions and is considered the "gold standard" for nonparametric hypothesis testing in genetic studies. A result was considered statistically significant at an empirical significance level of p‑perm < 0.05. The permutation test was applied only to genotypes that demonstrated statistically significant associations under standard calculation.

As a result of the analysis, of the 6 studied CYP genes, statistically significant associations with IVF efficacy were identified for the CYP2A6 and CYP2C19 genes (Table 1).

The most pronounced associations were recorded for polymorphisms of the CYP2A6 gene. For the rs8192733 locus, the genotype distribution between groups was characterized by extremely high differences; the G/G genotype was predominantly found in the group with unsuccessful IVF attempts (33 out of 48 vs. 7 out of 48 in the main group), corresponding to OR = 0.105 (95% CI 0.105–0.316; p < 0.001) (Table 1). The heterozygous C/G genotype and the homozygous C/C genotype showed significant predominance in the main group and were associated with an effective IVF procedure (OR = 3.176; 95% CI 1.202–8.395; p = 0.031 and OR = 4.795; 95% CI 1.212–18.963; p < 0.001, respectively). The rs56113850 polymorphism of the CYP2A6 gene also showed statistically significant intergroup differences. According to the analysis, the C/T genotype was associated with a 3.1‑fold increased odds of pregnancy (OR = 3.095; 95% CI 1.284–7.458; p = 0.010), whereas the homozygous T/T variant, conversely, was more frequent in the comparison group and was associated with an unfavorable outcome (OR = 0.286; 95% CI 0.100–0.812; p = 0.015) (Table 1). For the rs57897628 polymorphism of the CYP2A6 gene, carriage of the A/A genotype was associated with a lower probability of pregnancy (OR = 0.256; 95% CI 0.091–0.725; p = 0.005), while the heterozygous A/G and homozygous G/G genotypes did not reach statistical significance (p > 0.05) (Table 1). The remaining studied polymorphisms of this gene (rs1809810, rs7259706, rs7260629, rs12459249, rs28399433, rs113288603) did not demonstrate statistically significant associations with IVF outcome.

Table 1. Distribution of CYP gene polymorphisms in women with anovulation

GenePolymorphismGenotypeGroup 1 (n = 48)Group 2 (n = 48)χ²OR95% CIpCYP2A6rs1809810T/T4848––––¹ rs7259706C/T22190.3231.2980.577–2.0950.536  T/T2629 0.7740.344–1.742  rs7260629G/G28290.0430.9170.406–2.0720.835  G/T2019 1.0900.483–2.463  rs8192733C/C194–4.7951.212–18.963<0.001²,³  C/G22114.6883.1761.202–8.3950.031³  G/G73328.9710.1050.105–0.316<0.001³ rs12459249C/C27280.0430.8270.363–1.8810.837  C/T18170.0451.2400.580–2.8990.833  T/T3301.0–1.0 rs28399433A/A42400.3341.40.446–4.3930.563  A/C68 0.7140.228–2.272  rs56113850C/C19210.1710.8420.374–1.8980.679  C/T23116.5583.0951.284–7.4580.010³  T/T6165.8970.2860.1–0.8120.015³ rs57897628A/A6188.0000.2560.091–0.7250.005³  A/G21180.3891.4580.638–3.3350.534  G/G21123.7402.00.826–4.8440.054 rs113288603C/C36350.0540.7710.289–2.0590.817  C/T1262.4622.1110.716–6.2220.117  T/T07––––¹CYP2C9rs1057910A/A4841––––¹  A/C07 ––  rs1799853C/C38370.0611.1300.429–2.9760.805  C/T1080.2741.3160.470–3.6870.601  T/T03––––¹CYP2C19rs4244285G/G3245–0.1330.036–0.4960.001²,³  G/A163 7.52.016–27.901  rs4986893G/G4848–––– rs12248560C/C282801.00.444–2.2511.0  C/T171701.00.433–2.3081.0  T/T3301.00.192–5.2221.0CYP2D6rs1065852A/A06––––¹  A/G19160.4051.3100.569–3.0160.525  G/G29260.3831.2910.574–2.9050.536 rs3892097C/C29260.3831.2910.574–2.9050.536  C/T1922 0.7740.344–1.742  rs35742686C/C4845––––¹  C/T03 –– CYP3A4rs35599367G/G4245–2.1430.504–9.1200.243²  G/A63 0.4670.110–1.986 CYP3A5rs776746C/C43410.3811.4680.431–4.9970.537  T/T57 0.6810.2–2.318 

Notes: ¹ — not calculated due to absence of cases in one of the groups; ² — p corresponds to Fisher's exact test; ³ — p corresponds to p‑perm when the value is less than 0.05. OR — odds ratio; 95% CI — 95% confidence interval; p — significance level; p‑perm — empirical p‑value after Monte Carlo permutation test.

According to the obtained results, for the CYP2C19 gene, a significant association was identified for the rs4244285 polymorphism. The G/G genotype was significantly more frequent in the comparison group (45 out of 48 vs. 32 out of 48 in the main group) and was associated with an unfavorable procedure (OR = 0.133; 95% CI 0.036–0.496; p = 0.001); the heterozygous G/A genotype, conversely, predominated in the group with achieved pregnancy and was associated with a 7.5‑fold odds of pregnancy (OR = 7.5; 95% CI 2.016–27.901; p = 0.001), while the rs4986893 and rs12248560 polymorphisms of the CYP2C19 gene showed no statistically significant differences between groups (p < 0.05) (Table 1).

It should be noted that for the CYP2C9, CYP2D6, CYP3A4, and CYP3A5 genes, no statistically significant differences in genotype distribution between groups were identified. Also, in several cases (rs1057910, rs1799853, rs35742686, rs700519), individual genotypes occurred exclusively in one of the groups, which did not allow calculation of odds ratios and attainment of statistical significance (Table 1).

In this pilot study, a comprehensive analysis of associations of polymorphisms of six cytochrome P450 family genes (CYP2A6, CYP2C9, CYP2C19, CYP2D6, CYP3A4, CYP3A5) with IVF outcome in patients with anovulatory infertility was conducted for the first time. According to the obtained data, statistically significant associations were identified for polymorphisms of CYP2A6 (rs8192733, rs56113850, rs57897628) and CYP2C19 (rs4244285).

For rs8192733, associations were recorded for the G/G genotype of the CYP2A6 gene with an ineffective IVF cycle (p < 0.001) and for C/C and C/G with an effective procedure (p = 0.031 and p = 0.031, respectively). According to current research, the CYP2A6 gene encodes an enzyme involved in the metabolism of xenobiotics and endogenous substrates, including steroid hormones [18, 19]. The rs8192733 polymorphism, located in the intronic region of the gene, may affect splicing efficiency and enzyme expression levels, which in turn influences the metabolism of endogenous steroids and drugs used in COS protocols [18, 19]. A study investigating the effect of the CYP2A6 rs8192733 polymorphism on letrozole efficacy in infertile women with PCOS did not reveal statistically significant associations, which may be related to population characteristics and differences in study design [20].

In the present study, for the rs4244285 polymorphism of the CYP2C19 gene, an association was identified for the G/A genotype with a 7.5‑fold odds of pregnancy (p = 0.001), while the G/G genotype was associated with an unfavorable outcome (p = 0.001). It has been shown that CYP2C19 polymorphisms may affect progesterone metabolism [20]. Among carriers of the intermediate metabolizer phenotype for CYP2C19, statistically significantly higher AUC and Cmax of progesterone were noted compared to carriers of the normal (extensive) phenotype, while no associations with CYP2C9 polymorphisms were identified, suggesting that CYP2C19 may contribute more to progesterone metabolism than CYP2C9 [22]. According to current research, the relatively well‑studied functional variant associated with the "poor metabolizer" phenotype is the rs4244285 polymorphism of the CYP2C19 gene (G/A) [20–22]. In the present study, the G/G genotype, corresponding to normal metabolism, was significantly more frequent in the group with failed IVF attempts. Based on this fact, it can be hypothesized that patients who are carriers of the CYP2C19 G/G genotype may require higher doses of progestins for luteal phase support, which may lead to suboptimal response when standardized protocols are used.

The absence of significant associations for the CYP2C9, CYP2D6, CYP3A4, and CYP3A5 genes in the present study does not exclude their potential role in predicting IVF outcomes. Further investigation of their role in predicting IVF outcomes requires an increased sample size, which would also allow the detection of less prominent associations. Furthermore, it cannot be excluded that the influence of these genes may manifest not at the stage of pregnancy achievement but at other stages of the reproductive process, such as oocyte or embryo quality. Alongside this, it should be noted that in the present study, the permutation test was applied only to genotypes that showed significant associations under standard calculation. This approach made it possible to minimize the risk of false‑positive results in multiple testing while maintaining statistical power to confirm the identified associations; as a result, all significant associations were confirmed by the permutation test (p‑perm < 0.05), which increases the reliability of the obtained results.

Among the limitations of this study, the small sample size (n = 96), characteristic of pilot studies, and the absence of a validation cohort can be mentioned. In addition, the study was conducted exclusively on patients of Caucasian ethnicity, which limits the possibility of extrapolating the results to other populations, and the combined contribution of several polymorphisms (haplotype analysis) was not evaluated, which may become a subject of future research.

The results of this pilot study demonstrated an association between polymorphic variants of the CYP2A6 (rs8192733, rs56113850, rs57897628) and CYP2C19 (rs4244285) genes and the probability of achieving clinical pregnancy in IVF programs in patients with anovulatory infertility. Overall, the obtained data are consistent with the results of earlier studies and confirm the significant role of CYP gene polymorphisms in interindividual differences in treatment response. The obtained data also open prospects for the development of personalized approaches to the management of patients with anovulatory infertility based on pharmacogenetic testing. Further studies on larger samples with the inclusion of validation cohorts and multifactorial analysis are needed to confirm the obtained results and to more thoroughly evaluate their clinical significance.