Human papillomavirus: genetic diversity, vaginal microbiota, and local immunity in cervical intraepithelial neoplasia

Full Text

Introduction

Cervical cancer (CC) is a significant global public health problem and is one of the four most common types of cancer affecting women worldwide, including breast cancer, colorectal cancer, lung cancer and CC. Virtually every case of CC is associated with infection caused by the human papillomavirus (HPV). Prevention of this disease is based on identifying risk groups and vaccinating patients, but there are differences in CC incidence between developed and developing countries. Numerous factors contribute to the development of cervical cancer, including sexually transmitted diseases, reproductive and hormonal disorders, and genetic factors [1].

The World Health Organization (WHO) estimates the prevalence of HPV infection to be between 9% and 13% among the world's population, and in the United States alone, more than 6.2 million people are diagnosed with HPV each year [2]. According to the results of a systematic review and meta-analysis conducted in 2025 by German scientists, the highest prevalence of any HPV was found in Africa (83.7%, 95% CI 65.9–95.8%) and South and Central America (66.7%, 95% CI 48.8–82.5%), and the lowest in North America (42.1%, 95% CI 7.3–82.5%) (Q = 21.3; p = 0.0007). The most frequently identified genotypes worldwide were 16, 52, and 53, with a combined prevalence of 12.0% (95% CI 8.0–17.7%), 8.4% (95% CI 4.4–15.4%), and 5.8% (95% CI 4.1–8.0%), respectively [3]. There are more than 100 known types of HPV, of which only two genotypes (16 and 18) are associated with 70% of cases of cervical cancer and precancerous lesions. The lack of therapy and insufficient knowledge of molecular mechanisms support critical interest in HPV-related pathology and are crucial for the development of new treatments and more effective control of the consequences of infection, especially given that HPV often persists for a long time and eventually leads to the development of malignant neoplasms [2]. In the latest WHO classification of female genital tumors (2020, 5th ed.), cervical carcinomas are classified based on their association with HPV infection. According to the classification, cervical tumors are divided into HPV-associated and HPV-independent¹.

Cervical intraepithelial neoplasia (CIN) of varying severity always precedes cervical cancer. Cytological and histological methods are used to detect and determine the extent of cervical cancer. The results of cytological examination are described according to the Bethesda classification, and histological results — according to R.M. Richart et al. [4]. According to Bethesda, the results are interpreted as Negative for Intraepithelial Lesion or Malignancy (NILM), atypical squamous cells of undetermined significance (ASC-US), low-grade squamous intraepithelial lesions (LSIL, corresponding to CIN1); high-grade squamous intraepithelial lesions (HSIL, corresponding to CIN2 or CIN3, depending on the severity of the changes) [5].

WHO approaches to reducing mortality from cervical cancer — screening and treatment, in which the decision to treat is based solely on a positive primary screening test without confirmatory tests (i.e., without a second screening test and without a histopathological diagnosis) [6]. The WHO recommends HPV DNA detection as the primary screening test, rather than visual inspection or cytology². This approach does not take into account HPV-independent cervical cancer, which remains undiagnosed at this stage of the examination.

Most previous studies of CC have focused on identifying HPV in CC cases due to its prevalence, resulting in significantly less attention being paid to HPV-independent CC. Accordingly, HPV-independent CC has been insufficiently studied compared to HPV-associated cases, leading to limited understanding of its pathogenesis, clinical characteristics, improving diagnostic accuracy, and developing effective treatments, including specific biomarkers and targeted therapies for this less common but potentially aggressive form of the disease [7]. HPV-independent CC is often diagnosed at later stages, with a more unfavorable prognosis compared to HPV-associated cases. Understanding the pathogenic mechanisms of this disease may help identify potential causes and triggers, which is important for the development of preventive measures and early diagnosis of HPV-independent CC [8].

The development of cancer is a complex process and is usually the result of the combined effects of multiple factors. The clearer the understanding of the relationship between the tumor and its local immune microenvironment, the more personalized treatment plans can be developed for patients, improving prognosis and outcomes [8].

One important factor is the vaginal microbiome, which plays a key role in maintaining vaginal pH and preventing various urogenital diseases. Evidence suggests that vaginal lactobacilli provide a certain level of protection against bacterial vaginosis, sexually transmitted diseases, and urinary tract infections [9]. Recent studies have increasingly focused on the role of the vaginal microbiota in CIN and CC [10]. Normally, lactobacilli predominate in the vaginal microbiota (95–98%). When characterizing the vaginal microbiocenosis, two states can be distinguished: eubiosis, indicating a normal content of lactic acid bacteria and microbial balance, and, in contrast, dysbiosis, indicating microbial imbalance. Eubiosis is promoted by the predominance of Lactobacillus spp. species, which colonize the vaginal mucosa, thereby preventing the colonization of the vagina by pathogenic microorganisms or the excessive reproduction of conditionally pathogenic microorganisms that are part of the normal microbiocenosis and their spread beyond their ecological niches. This protection is achieved through two mechanisms: the first is specific adhesion to epithelial cells, and the second is the production of compounds with antimicrobial properties [11]. The resident microbiota helps maintain the acidic pH of vaginal contents, which normally ranges from 3.8 to 4.5. Normal vaginal microbiota provides what is known as colonisation resistance of the genital tract, which refers to a combination of local and humoral factors that ensure the consistency of the quantitative and species composition of normal microflora. Among the protective factors, the following components of innate and adaptive immunity can be distinguished: IgA, IgM, IgG, lysozyme, α-lysin, complement, secretory immunoglobulin A (sIgA), etc., bactericidal compounds (hydrogen peroxide (H₂O₂), lactic acid), immune system cells (leukocytes, macrophages, T and B lymphocytes) [12].

There are assumptions that not all types of lactobacilli inhabiting the vagina are equal in their protective role, and that L. crispatus is the most active of all species [13]. Numerous studies have shown that women with vaginal microbiota consisting predominantly of L. crispatus in the cervicovaginal secretion, characteristic of a healthy microbiocenosis, have reduced levels of pro-inflammatory cytokines — interleukins (IL)-1α and -1β, as well as chemokines such as IL-8 [14].

Despite the proven role of HPV infection as the main cause of CC, a number of patients test negative for HPV [15, 16]. Despite the presence of false-negative results in modern tests, improving diagnostic accuracy is crucial for studying the pathogenesis of HPV-independent CC and improving the prognosis in these patients. Existing studies show that HPV-independent CC has a different pathogenesis than HPV-associated CC, although the exact mechanism of development is still unclear. It is currently believed to be associated with changes in microflora, immune response, and a number of other factors [8, 16].

The aim of this study is to assess the contribution of HPV and to examine the state of vaginal microecology in women with HPV-induced CIN (using the Gomel region as an example).

Materials and methods

In 2018–2023, a screening study was conducted on 11,382 women aged 18–79 living in the Gomel region. The study material consisted of scrapings from the cervical canal, cervicovaginal secretions, vaginal discharge, and smears from the posterior vaginal fornix. At the initial stage, all scrapings were examined using a cytological method (liquid cytology with Papanicolaou staining), and molecular genetic testing of these samples for the presence of high-risk HPV DNA was performed in parallel using Abbott Real Time hR HPV (Abbott) and AmpliSens HPV Genotype-FL reagent kits (Central Research Institute of Epidemiology, Rospotrebnadzor). Genotyping was performed for 659 samples to determine the specific HPV genotype.

A study was conducted on 102 samples of cervicovaginal secretions and smears from the posterior vaginal fornix: 24 samples in which high-risk HPV DNA was detected and 78 samples in which negative results for high-risk HPV were obtained from women of reproductive age. Women with HPV-negative PCR results were divided into two groups: a comparison group (n = 39) without cervical pathology (normocytogram) and a group with varying degrees of intraepithelial dysplasia (n = 39). The average age of the women (n = 78) was 35 ± 7 years. Cervicovaginal secretions (n = 102) were tested for sIgA, IL-1β, -2, -6, -8, -10, and tumor necrosis factor-α (TNF-α).

Swabs from the posterior vaginal vault (n = 102) were examined using a microbiological method with semi-quantitative culture on a set of differential diagnostic media (Endo medium, mannitol-egg yolk-salt agar, enterococcus agar, Sabouraud agar with chloramphenicol, 5% blood agar, MRS agar. The microorganisms that grew were identified using the Vitek 2 Compact automatic microbiological analyzer (BioMerieux) and the miniApi semi-automatic analyzer (BioMerieux). The state of the vaginal microbiocenosis was assessed³. A variant of the microbiocenosis in which the number of lactobacilli was more than 10⁷ CFU/mL, Candida spp. less than 10⁴ CFU/mL, opportunistic microorganisms less than 10⁵ CFU/mL, was considered an absolute normocenosis, with Candida spp. > 10⁴ CFU/mL, lactobacilli 10⁵–10⁶ CFU/mL — as moderate dysbiosis, and with a decrease in lactobacilli < 10⁵ CFU/mL or their complete absence, conditionally pathogenic microorganisms > 10⁵ CFU/mL — as pronounced dysbiosis. Next, the ability of lactobacilli to produce H₂O₂ (n = 66), antagonistic activity (n = 66), and the ability to form a biofilm (n = 66) were evaluated. To determine H₂O₂ production by lactobacilli, a method developed in the laboratory was used, involving the addition of sulfuric acid solution and potassium starch iodide solution to the daily culture of lactobacilli⁴. To determine antagonistic activity, the agar block method [18] was used in relation to the test strains Escherichia coli ATCC (American Type Culture Collection) 25922, Staphylococcus aureus ATCC 25923, Enterococcus faecalis ATCC 29212, Pseudomonas aeruginosa ATCC 27853. The phenotypic ability to form a biofilm was evaluated using a previously developed method, using the patent for invention of the Republic of Belarus No. 20326 [19] as a prototype and modified for lactobacilli. Congo red was used to stain the main substance of the biofilm, and gentian violet was used to stain the biomass. The pH level of vaginal secretions was determined using indicator test strips (Macherey-Nagel). To control the quality of lactobacilli identification and determine their functions, the reference strain Lactobacillus acidophilus ATCC 4356 was used.

Study design

In the first stage, molecular genetic screening and cytological studies were performed on 11,382 women in the Gomel region, and in the second and third stages, comparative studies (microbiological and immunological) were performed using a case-control design (Fig. 1). Microbiological and immunological studies were performed on women of reproductive age (n = 102), who were divided into two groups: the first group was high-risk HPV-positive (n = 24), and the second group was high-risk HPV-negative (n = 78). A comparison of the studied indicators in the groups was performed, followed by an additional study in the group with HPV-independent CIN.

 

Fig. 1. Study design outline.

 

Statistical methods

Qualitative indicators are presented as frequencies and percentages in the group. The calculation of 95% confidence intervals (CI) was performed using the Klopper–Pearson method. When studying contingency tables, the χ² criterion was used; in case of violation of the assumptions underlying the χ² criterion, Fisher's exact criterion was used. Quantitative indicators of the study are presented as the mean and standard deviation in the form of M ± SD. Quantitative indicators in the groups were compared using the t-test.

Multivariate analysis of laboratory indicators associated with CIN was performed using logistic regression, according to which the odds ratio (OR) was calculated as an exponential transformation of the corresponding regression coefficients. The CI for OR was also calculated as an exponential transformation of the corresponding CI of the regression coefficients.

All calculations were performed using the statistical package R v. 4.3⁵. When testing statistical hypotheses, the probability of a type I error α was assumed to be equal to 0.05.

Results

Prevalence of HPV in the genital tract and characteristics of its genotypes

According to the results of a molecular genetic screening study, the prevalence of high-risk HPV infection among women in the Gomel region was 9.0% (1,022/11,382), and among women of reproductive age, it was 9.7% (952/9,831). The highest prevalence of high-risk HPV was found in the 18–24 age group, at 19% (95% CI 17.0–21.1). There was a tendency for high-risk HPV detection rates to decrease with age: in the 25–29 age group, high-risk HPV was detected in 12.5% of cases (95% CI 10.7–14.5), and in women aged 30–34, it was detected in 9.5% of cases (95% CI 8.1–11.0) (Table 1).

 

Table 1. Prevalence of high-risk HPV infection among women in different age groups, n (%)

Age, years	All women n = 11 382	Women with identified HPV n = 1022	95% CI
18–24	1454 (12.8)	276 (19.0)	17.0–21.1
25–29	1244 (10.9)	156 (12.5)	10.7–14.5
30–34	1620 (14.2)	154 (9.5)	8.1–11.0
35–39	1726 (15.2)	135 (7.8)	6.6–9.2
40–44	1882 (16.5)	120 (6.4)	5.3–7.6
45–49	1902 (16.7)	111 (5.8)	4.8–7.0
≥ 50	1554 (13.6)	70 (4.5)	3.5–5.7

 

Based on the results of genotyping, the contribution of individual genotypes to the HPV genotypic landscape in women in the Gomel region was determined. It was found that during the study period, HPV genotypes 16, 18, 51, 56 and 31 were the most common (Fig. 2). Then, in descending order, the detection rates of genotypes 52, 33, 45, 58, 66, 39, 68, 59 and 35 were distributed (Fig. 2).

 

Fig. 2. Detection rates of oncogenic HPV genotypes in the Gomel region (n = 659)

 

Human papillomavirus in cervical dysplasia

The evaluation of the results of cytological screening revealed pathological changes in the cervix in 536 women. Among them, HSIL was found in 120 (1.1%) samples, LSIL in 205 (1.8%), and ASC-US in 211 (1.84%). The vast majority of samples (95.3%, n = 10,846) met the criteria for a normocytogram.

High-risk HPV is recognized as the main cause of CIN. Our study found that HPV was more frequently detected in clinical samples with HSIL (65.8% (n = 79/120)). A significant decrease in the frequency of high-risk HPV was noted with a decrease in the severity of precancerous changes: HSIL > LSIL > ASC-US > NILM (p < 0.001). In this study, women with HSIL had almost twice as many positive high-risk HPV results as those with ASC-US (p < 0.001). The lowest percentage was in the NILM group — 7.1% (n = 775/10,071) (Table 2).

 

Table 2. Detection rates of high-risk HPV in women depending on the severity of cervical dysplasia, n (%)

Bethesda cytology result	HSIL	LSIL	ASC-US	NILM	p
n	120	205	211	10,846
HPV not detected	41 (34.2)	108 (52.7)	139 (65.9)	10071 (92.9)	< 0.001
HPV detected	79 (65.8)	97 (47.3)	72 (34.1)	775 (7.1)

 

A detailed analysis of the HPV genotypic landscape in women with varying degrees of cervical lesions revealed differences in their spectrum. Thus, genotype 16 was significantly more frequently identified in HSIL (73.8%), genotypes 45 (14.9%) and 58 (11.9%) in LSIL, and genotype 33 (14.8%) in ASC-US. In women without pathological changes, high-risk HPV genotype 18 (18%) was the most common. According to the phylogenetic classification, representatives of the HPV phylogenetic groups α7 and α9 significantly more often prevailed in cases of varying degrees of cervical lesions and with a normocytogram (Table 3).

 

Table 3. Distribution of high-risk HPV genotypes in patients with different degrees of CIN and in normal subjects (n = 659)

HPV type	Phylogenetic group based on whole-genome sequencing	Bethesda cytology results				p
		HSIL (n = 61)	LSIL (n = 67)	ASC-US, (n = 54)	NILM, (n = 477)
16	α9	45 (73.8)*	29 (43.3)	23 (42.6)	224 (47.0)	0.001
18	α7	4 (6.6)	4 (6.0)	1 (1.9)	86 (18.0)*	< 0.001
31	α9	5 (8.2)	8 (11.9)	8 (14.8)	38 (8.0)	0.305
33	α9	3 (4.9)	8 (11.9)	8 (14.8)*	27 (5.7)	0.025
35	α9	1 (1.6)	2 (3.0)	1 (1.9)	12 (2.5)	0.953
39	α7	0	4 (6.0)	1 (1.9)	19 (4.0)	0.262
44	α10	2 (3.3)	2 (3.0)	0	4 (0.8)	0.161
45	α7	1 (1.6)	10 (14.9)*	6 (11.1)	24 (5.0)	0.003
51	α5	3 (4.9)	8 (11.9)	6 (11.1)	57 (11.9)	0.437
52	α9	2 (3.3)	6 (9.0)	2 (3.7)	46 (9.6)	0.207
53	α6	0	1 (1.5)	0	5 (1.0)	0.704
56	α6	5 (8.2)	12 (17.9)	5 (9.3)	38 (8.0)	0.069
58	α9	2 (3.3)	8 (11.9)*	3 (5.6)	18 (3.8)	0.028
59	α7	0	1 (1.5)	4 (7.4)	15 (3.1)	0.113
66	α6	2 (3.3)	2 (3.0)	2 (3.7)	23 (4.8)	0.858
68	α7	1 (1.6)	6 (9.0)*	0	15 (3.1)	0.029
73	α11	0	2 (3.0)	0	4 (0.8)	0.233
82	α5	1 (1.6)	2 (3.0)	0	4 (0.8)	0.340
6	α10	0	0	0	2 (0.4)	0.858

Note. *Significant differences in the groups studied.

 

Multiple high-risk HPV genotypes were most frequently detected in LSIL, with four genotypes simultaneously present in HSIL (3.3%). At the same time, one HPV genotype was widely detected in HSIL (86.9%), but the differences between the groups were insignificant (p = 0.326). Among women with NILM, both 1 (77.1%) and 2 (14.3%) high-risk HPV genotypes were identified, which may increase the risk of developing precancerous pathology in these women over time (Table 4). This is the basis for including them in risk groups and dynamic monitoring for the purpose of diagnosing early stages of CIN.

 

Table 4. Distribution of detection of single and multiple high-risk HPV genotypes in the study groups

Number of identified HPV genotypes	HSIL (n = 61)	LSIL (n = 67)	ASC-US, (n = 54)	NILM, (n = 477)	Total
1	53 (86.9)	41 (61.2)	40 (74.1)	368 (77.1)	502 (76.2)
2	4 (6.6)	13 (19.4)	12 (22.2)	68 (14.3)	97 (14.7)
3	1 (1.6)	9 (13.4)	2 (3.7)	23 (4.8)	35 (5.3)
4	2 (3.3)	2 (3.0)	0	8 (1.7)	12 (1.8)
5 and more	1 (1.6)	2 (3.0)	0	10 (2.1)	13 (2.0)

 

The state of the microbiocenosis and indicators of local vaginal immunity in HPV-positive and HPV-negative women

Of the 536 samples in which CIN was detected, 248 (46.3%) were HPV-positive and 288 (53.7%) were HPV-negative. When comparing indicators characterizing the state of the vaginal microbiocenosis and local immunity, no significant differences were found in the study groups (Table 5). The functions of lactobacilli in these groups did not differ. Therefore, it is not possible to say with certainty that HPV affects the composition of the resident vaginal microbiota and the state of local vaginal immunity.

 

Table 5. Results of the assessment of the microbiocenosis and local immunity indicators in the study groups

Indicator	High-risk HPV+		High-risk HPV–		p
	n	value	n	value
IL-6, Me [Q25; Q75]	24	25 [7.5; 39]	78	19 [9.7; 34]	0.694
IL-10, Me [Q25; Q75]	24	6.7 [5.4; 9.2]	78	6.6 [4; 11]	0.287
IL-8, Me [Q25; Q75]	24	85 [48; 167]	78	81 [45; 136]	0.549
IL-1β, Me [Q25; Q75]	24	41 [13; 128]	78	28 [14; 67]	0.399
IL-2, Me [Q25; Q75]	15	18 [7.8; 129]	45	6.3 [3.5; 37]	0.396
TNF-α, Me [Q25; Q75]	24	14 [10; 16]	78	14 [9.4; 17]	0.67
sIgA, Me [Q25; Q75]	24	3.4 [1.2; 9.9]	78	3.1 [1.1; 5.8]	0.3
pH, Me [Q25; Q75]	24	4.5 [4.0. 5.2]	78	4.0 [4.0. 5.0]	0.289
Normocenosis, n (%):	24	10 (41.7)	78	32 (41.0)	0.333
Moderate dysbiosis, n (%)	24	3 (12.5)	78	20 (25.6)
Severe dysbiosis, n (%)	24	11 (45.8)	78	26 (33.3)
Antagonistic activity, n (%):
E. coli < 12 mm	14	2 (14.3)	52	7 (13.5)	> 0.99
S. aureus < 12 mm	14	2 (14.3)	52	8 (15.4)	> 0.99
E. faecalis < 13 mm	14	6 (42.9)	52	15 (28.8)	0.499
P. aeruginosa < 14 mm	14	3 (21.4)	52	11 (21.2)	> 0.99
Production of H2O2, n (%):
none	14	7 (50.0)	52	12 (23.1)	0.101
yes	14	7 (50.0)	52	40 (76.9)
Ability to form the main substance, n (%):
low	14	7 (50.0)	52	11 (21.2)	0.096
medium	14	2 (14.3)	52	14 (26.9)
high	14	5 (35.7)	52	27 (51.9)
Ability to form biomass, n (%):
none	14	3 (21.4)	52	4 (7.7)	0.264
low	14	2 (14.3)	52	18 (34.6)
medium	14	7 (50.0)	52	20 (38.5)
high	14	2 (14.3)	52	10 (19.2)

Note. HPV “+” — DNA of genotypes 16, 18, and other genotypes detected; HPV “–” — high-risk HPV DNA not detected

 

Despite the proven role of high-risk HPV in the development of cervical dysplasia and cervical cancer, the etiology of HPV-independent cervical dysplasia is increasingly discussed in the literature [17, 20, 21]. According to our results, 34.2% of samples with HSIL were found to be high-risk HPV-negative. Therefore, in the next stage, a detailed study of the state of the microbiota and local immunity of the vagina was conducted in the group with HPV-negative CIN. To date, no similar studies have been conducted in the Republic of Belarus, therefore our results are new.

Characteristics of factors associated with HPV-independent dysplasia

According to the results of microbiological research, lactobacilli were isolated significantly less frequently in the group of women with CIN compared to the group of women with a normocytogram: in 13 (33.3%) clinical samples with CIN and in all samples (n = 39) with normocytogram (Table 5). The resident microbiota was represented by various species, but no significant differences in the prevalence of any lactobacillus species depending on the condition of the cervix were obtained (p = 0.264). Nevertheless, L. crispatus, L. gasseri, and L. plantarum were more frequently isolated in the normocytogram group, while L. acidophilus, L. gasseri, and L. plantarum were more frequently isolated in the CIN group (Table 6).

 

Table 6. Species composition of lactobacilli in the study groups, n (%)

Type of lactobacilli	CIN, n = 13	Normocytogram, n = 39	p
L. acidophilus	3 (23.1)	3 (7.7)	0.264
L. crispatus	1 (7.7)	14 (35.9)
L. delbruckii	1 (7.7)	0
L. fermentum	1 (7.7)	1 (2.6)
L. gasseri	3 (23.1)	10 (25.6)
L. paracasei	1 (7.7)	3 (7.7)
L. plantarum	3 (23.1)	7 (17.9)
L. ramnosus	0	1 (2.6)

 

In women with a normocytogram, the vaginal microbiocenosis more often met the criteria for normocenosis (69.2%). In most patients with CIN, dysbiosis was observed (87.2%), with pronounced changes in the composition of the microbiota in this group being recorded with high frequency (66.7%), while no such changes were detected in healthy women (Table 7).

 

Table 7. Results of the assessment of the state of the vaginal microbiocenosis in the study groups, n (%)

Condition of the vaginal microbiome	CIN, n = 39	Normocytogram, n = 39	p
Normocenosis	5 (12.8)	27 (69.2)	< 0.001
Moderate dysbiosis	8 (20.5)	12 (30.8)
Severe dysbiosis	26 (66.7)	0

 

A significant imbalance in the microbiota in the group with CIN is also evidenced by an increase in the pH level of vaginal secretions compared to healthy women. Thus, in the group with CIN, the pH level was 5.0 [4.0; 6.0], and in the group with a normocytogram, it was 4.0 [4.0; 4.0] (p = 0.001). This is explained by a decrease in the number of lactobacilli in the group with CIN, which produce lactic acid and thus maintain the acidity of the vagina at the proper level.

Noting the decrease in the number of lactic acid bacteria and the low percentage (33.3%) of samples in which their growth was observed in the group with HPV-independent CIN, it was interesting to evaluate the functional characteristics of lactobacilli in the study groups.

The antagonistic activity of lactobacilli in the group with CIN was 2 times lower in relation to all test strains compared to the group with a normocytogram. At the same time, the differences between the groups were statistically significant (p < 0.001). When comparing the growth inhibition zones in determining the antagonistic activity in the group with CIN with those for the control strain L. acidophilus ATCC 4356, it can be seen that clinical isolates with CIN did not have antagonistic activity against all tested strains (Table 8).

 

Table 8. Diameters of growth inhibition zones (mm) for test strains in the study of the antagonistic activity of lactobacilli in the test groups, M ± SD

Antagonistic activity	Growth inhibition zones for L. acidophilus ATCC 4356, mm	CIN n = 13	Normocytogram n = 39	p
E. coli	12	9.6 ± 6.3	19.0 ± 4.2	< 0.001
S. aureus	12	8.8 ± 5.9	15.2 ± 1.9	< 0.001
E. faecalis	13	8.3 ± 4.8	15.1 ± 2.6	< 0.001
P. aeruginosa	14	11.2 ± 5.8	18.2 ± 2.4	< 0.001

 

The capability of certain types of lactobacilli to produce H₂O₂ is one of the significant characteristics of the isolate. In this study, lactic acid bacteria from the group with CIN significantly (p < 0.001) more often lacked this ability compared to lactobacilli from the group without cervical pathology. In the group with a normocytogram, all strains produced H₂O₂ (n = 39), while in the group with CIN, this ability was absent in 12 (92.3%) cases.

Lactobacilli also differed significantly in their ability to form a biofilm in the study groups (Table 9). Thus, in the group with CIN, a pronounced ability to form the main substance by isolates was noted in 7.7% of cases, moderate in 23.1%, and low in 69.2%. A pronounced ability to accumulate biomass was characteristic of only 1 (7.7%) strain of lactobacilli, a low ability was characteristic of 6 (46.2%) strains, and it was completely absent in 3 (23.1%) strains in the group with CIN. In the group with a normocytogram, on the contrary, lactobacilli were more likely to have the ability to synthesize the main substance (9.5 times) and accumulate biomass (6.5 times).

 

Table 9. Intensity of biofilm formation by lactobacilli in the study groups

Capability		CIN n = 13	Normocytogram n = 39	p
To synthesize the main substance:	low	9 (69.2)	2 (5.1)	< 0,001
	medium	3 (23.1)	11 (28.2)
	high	1 (7.7)	26 (66.7)
To accumulate biomass:	none	3 (23.1)	1 (2.6)	0,041
	low	6 (46.2)	12 (30.8)
	medium	3 (23.1)	17 (43.6)
	high	1 (7.7)	9 (23.1)

 

Another group of indicators is cytokines, which play an important role in pro-inflammatory and anti-inflammatory processes and carcinogenesis. A significant increase in IL-6 was found — 25 [12; 56] pg/mL (p = 0.004), IL-8 — 107 [65; 181] (p < 0.001), IL-2 — 43 [26; 127] (p = 0.001) in the group with CIN. In the case of a normocytogram, they were 15 [7.8; 25], 65 [31; 88], and 4.4 [3.2; 5.5] pg/mL, respectively. The levels of IL-1β, TNF-α, and IL-10 did not differ between the study groups (p = 0.065, 0.059, and 0.384, respectively). In the group with CIN, the IL-10 level was 6.6 [3.7; 11] pg/mL, IL-1β — 37 [24; 97], TNF-α — 13 [10; 16], and in the normocytogram group — 6.7 [4; 12], 20 [12; 43] and 16 [9; 20] pg/mL, respectively.

The amount of sIgA was also significantly reduced in the group with CIN compared to the group without cervical pathology: 2.3 [1.1; 4.6] and 3.6 [1.2; 6.5] pg/mL, respectively (p = 0.041).

To identify potential non-viral factors associated with CIN, a multivariate analysis was performed with the determination of OR for the studied indicators. The analysis revealed risk factors associated with CIN (Table 10). According to our data, unfavorable factors in HPV-independent CIN may include vaginal dysbiosis (OR = 18.7; 95% CI 4.1–129.0; p < 0.001), increased pH of vaginal discharge (OR = 2.4; 95% CI 1.4–5.2), increased IL-6 levels (OR = 1.03; 95% CI 1.00–1.08), and decreased sIgA content (OR = 0.85; 95% CI 0.69–0.99). For the other indicators, OR was not significant in HPV-independent CIN according to the results of multivariate analysis. Therefore, it is not possible to attribute changes in IL-8, IL-1β, and TNF-α levels to potential factors in the development of CIN.

 

Table 10. Multivariate analysis of laboratory parameters reflecting the state of the vaginal microbiota, the level of inflammation, and the local immune response associated with CIN

Laboratory indicator	OR (95% CI)	p
Vaginal pH per 1 unit of measurement	2.5 (1.4–5.5)	< 0.001
Vaginal dysbiosis	18.7 (4.1–129.0)	< 0.001
IL-6 per 1 pg/mL	1.03 (1.00–1.08)	< 0.001
IL-8 per 1 pg/mL	1.0 (0.99–1.02)	0.228
IL-10 per 1 pg/mL	1.0 (0.89–1.14)	0.524
IL-1β per 1 pg/mL	1.0 (0.99–1.02)	0.457
TNF-α per 1 pg/mL	0.91 (0.76–1.06)	0.316
sIgA per 1 pg/mL	0.85 (0.69–0.99)	0.037

 

Discussion

This study found that the prevalence of high-risk HPV in women of late reproductive age was lower than in women of early reproductive age. A similar trend was observed in a study by M. Yamaguchi et al. [22]. Japanese scientists found that the prevalence of HPV was highest in the 20–30 age group, at 15.2%. This figure was significantly higher than in the 35–36 age group (6.9%) and the 40–41 age group (5.8%), p for the trend < 0.01. Moreover, we found that the higher the degree of cervical epithelial damage, the more frequently high-risk HPV was found in the samples studied. Similar results were published by J. Mishra et al., based on data from a survey of 1,788 women [23]. The authors showed that high-risk HPV was found in 45% of samples in the ASC-US group, in 73.07% in the LSIL group, in 86.67% in the HSIL group, and in 83.34% in the CC group. There is an increase in positive cases of high-risk HPV with an increase in the severity of cytological changes in the cervix.

R. Correa et al. demonstrated the leading role of HPV genotype 16 in all degrees of malignancy, with slight fluctuations between ≤ CIN1 (14.5%) and CIN2 (19.8%), but its proportion increased in CIN3 (51.5%) and CC (65.1%) [24], which is consistent with our data. The results of the analysis showed that among the samples studied, genotype 16 was detected in 73.8% of cases with HSIL.

High-risk HPV was most frequently detected in women with HSIL, with the number of cases decreasing as the severity of precancerous cervical changes decreased. In women without pathological cervical changes, the percentage of high-risk HPV detection was the lowest. No increase in the frequency of multiple HPV genotypes was observed in women with HSIL. On the contrary, in 86.9% of cases, one of the HPV genotypes was found to be dominant in HSIL. Similar results were shown in a study by X. Tang et al., who noted that the rates of positive high-risk HPV results in squamous cell lesions of the cervix progressively increased with the severity of the disease [25]. At the same time, the percentage of concomitant multiple HPV infections among cases with a positive high-risk HPV result decreased significantly with increasing severity of squamous cell abnormalities. There was no increase in the frequency of CIN3+ (CIN3 and squamous cell carcinoma) detection in cases with concomitant infections caused by 2 or 3 high-risk HPV genotypes compared to patients with HPV infection caused by 1 genotype. Therefore, HPV is associated with CIN in the vast majority of cases.

However, in almost a third of the samples, we did not detect molecular genetic markers of HPV in HSIL. It should be noted that we used reagent kits with high sensitivity and specificity, followed the rules of the pre-analytical stage, and employed highly qualified personnel. Therefore, our further efforts were focused on analyzing the state of the vaginal microbiocenosis and local immunity in a group of HPV-negative women. Recent studies show that HPV-negative CC has a different pathogenesis compared to HPV-positive CC, but the exact mechanism is still unclear. It is believed to be associated with the immune microenvironment and tumor gene mutations, leading to an unfavorable prognosis [8].

One of the important components of the microbiota are lactobacilli, which can significantly influence the likelihood of developing an infectious process and dysbacteriosis. The use of the culture method allowed us to isolate viable, metabolically active lactobacilli, fully study their functional activity, and indirectly, based on correlation analysis, assess their contribution to the development of precancerous pathology of the cervix. The results of microbiological studies of the vaginal microbiocenosis showed that the species composition of lactobacilli in the study groups did not differ significantly, but in the group with a normocytogram, the species L. crispatus predominated, and in HPV-independent CIN, it was detected in only 1 sample. L. crispatus is described in the literature as a species characteristic of a normal healthy vaginal microbiocenosis, which fully possesses all protective functions against bacterial and viral pathogens [26]. It has been established that vaginal dysbiosis and increased pH of vaginal discharge were significantly more common in CIN. Similar results were demonstrated in a study by Y. Ma et al., who found that higher microbiome diversity with a decrease in Lactobacillus spp., especially L. crispatus, is associated with the severity of cervical lesions, most often with HSIL and CC [27]. Vaginal dysbiosis causes changes in immune and metabolic signals, including chronic inflammation, disruption of the epithelial barrier, changes in cell proliferation and apoptosis, genome instability, angiogenesis, and metabolic dysregulation. These pathophysiological changes can lead to gynecological cancer. New data show that genital dysbiosis and/or specific bacteria may play an active role in the development and/or progression and metastasis of gynecological malignancies, such as cervical cancer, endometrial cancer and ovarian cancer, through direct and indirect mechanisms [28]. Consequently, vaginal dysbiosis may be a trigger for chronic inflammation and initiate the mechanism of cervical carcinogenesis.

Furthermore, we found a significant decrease in antagonistic activity, the ability to produce H₂O₂, and the formation of biofilm by lactobacilli in the group with HPV-independent cervical dysplasia. Accordingly, lactic acid bacteria in the group of women with precancerous cervical pathology had reduced functional activity, which led to a decrease in vaginal colonization resistance. After all, the main role of resident autochthonous microflora is to ensure colonization resistance [29]. Consequently, lactobacilli in the group with CIN did not have the ability to suppress the growth of pathogenic and conditionally pathogenic microorganisms on the mucous membranes of the vagina, which contributed to the development of inflammation, and with prolonged inflammation, its transition to a chronic form.

In a study by L. Cheng et al., patients with CIN and CC also had a higher prevalence of pH > 5 and lower H₂O₂ levels compared to healthy women [9]. These results are consistent with the data obtained in our study: in the group with CIN, the pH level was elevated and amounted to 5.0 [4, 0; 6, 0].

The state of local vaginal immunity in HPV-independent CIN also underwent changes: the levels of proinflammatory IL-6, IL-8, and IL-2 were elevated, while the level of sIgA was reduced. A study by J. Zheng et al. also points to reduced levels of sIgA in patients with HPV-negative cervical lesions compared to healthy women [16]. As the severity of CIN increases, significantly lower levels of sIgA are observed. An increase in pro-inflammatory cytokines with low levels of anti-inflammatory cytokines, a decrease in sIgA, and the presence of vaginal dysbiosis indicate the development of inflammation and contribute to its transition to a chronic form, which can subsequently lead to cell damage and HPV-negative tumor transformation of the cervix, on the one hand. On the other hand, vaginal dysbiosis, decreased functional activity of lactobacilli, impaired colonization resistance of the vagina, and decreased local immunity contribute to HPV infection of the cervix and may trigger the development of high-risk HPV-associated tumors.

Conclusion

Both HPV and factors characterizing the microecology of the vagina contribute to the development of CIN. Of course, high-risk HPV infection of women of early reproductive age with high-risk HPV genotypes (16, 18, 45, 33, and 58) plays a leading role in HPV-associated CIN. It is important to note that the unfavorable factors associated with HPV-independent CIN include vaginal dysbiosis, increased pH of vaginal secretions, increased concentration of pro-inflammatory cytokines, decreased sIgA levels, and decreased functional activity of lactobacilli: decreased H₂O₂ production, biofilm formation capacity, and antagonistic activity. All these factors contribute to the onset and progression of HPV-independent CIN.

In practical terms, the results obtained allow us to identify and form risk groups among women in whom the above-mentioned adverse factors are detected. It is these studies that allow us to assess the contribution of each factor studied in CIN, which will subsequently enable the application of personalized, predictive, and preventive approaches for each woman in order to prevent CC.

For patients, this means that treatment strategies can be tailored based on the characteristics of the microbiome, local immune response, and HPV genotype, which will optimize treatment effectiveness and enable a personalized approach.

The presented results are not only practical but also fundamental, as they broaden our understanding of the mechanisms of CC development through the study of the role of vaginal microecology, and the identified biomarkers of microbiota and local immunity in various HPV infection statuses may help in the development of new directions in the diagnosis and prevention of CIN and CC development.

 

¹ WHO Classification of Tumours Editorial Board. Female Genital Tumours. 5th ed. Vol. 4. WHO Classification of Tumours. Geneva; 2020.

² WHO guideline for screening and treatment of cervical pre-cancer lesions for cervical cancer prevention. 2nd ed. Geneva; 2021.

³ Gynecology: National Guidelines / edited by G.M. Savelyeva et al. (eds.) Moscow; 2017.

⁴ Loginova, O.P. Method for determining hydrogen peroxide production by Lactobacilli. State Institution "Republican Scientific and Practical Center for Radiology and Epidemiology of the Russian Federation", No. 1338 of March 20, 2024. URL: https://www.rcrm.by/science/rezultaty-nauchno-issledovatelskoy-deyatelnosti

⁵ R Core Team (2024). R: A language and environment for statistical computing. URL: https://www.R-project.org/

About the authors

Volha P. Lohinava

Republican Scientific and Practical Center for Radiation Medicine and Human Ecology

Author for correspondence.
Email: loginovaolga81@mail.ru
ORCID iD: 0000-0001-7189-3799

doctor of clinical laboratory diagnostics, Laboratory of cellular technologies

Belarus, Gomel

Natalia I. Shevchenko

Republican Scientific and Practical Center for Radiation Medicine and Human Ecology

Email: shevchenkoni@bk.ru
ORCID iD: 0000-0003-0579-6215

Cand. Sci. (Biol.), Associate Professor, Head, Laboratory of Cellular Technologies

Belarus, Gomel

Elena L. Gasich

Republican Center for Hygiene, Epidemiology, and Public Health

Email: elena.gasich@gmail.com
ORCID iD: 0000-0002-3662-3045

Dr. Sci. (Biol.), Associate Professor, Head, Laboratory for the diagnostics of HIV and concomitant infections, Research Institute of Hypertension and Computer Science

Belarus, Minsk

References

Supplementary files