Background

Trichomoniasis is the most common non-viral sexually transmitted infection (STI) in human beings which is responsible for a range of symptoms such as increasing vaginal secretion, pruritus and irritation of perineum in female [1]. According to the survey of the World Health Organization (WHO), at least 370 million people worldwide suffered from trichomoniasis. The global prevalence of trichomoniasis was 5.3% in females and 0.6% in males with a growing trend [2]. Trichomonas vaginalis (T. vaginalis) infection prevalence was 1.8% and 0.5% among females and males in the U.S [3]. In France, the total infection rates of T. vaginalis was 1.7% [4]. In Africa, the prevalence was as higher as 29% in Natal and 7.1% in Tanzania [5, 6]. In China, the prevalence was varied in different regions, for instance, 13.9% in Zhengzhou, 1.6% in Xinxiang [7] and 0.7% in Sichuan [8].

The infection of T. vaginalis can cause female vaginitis, cervicitis and adverse birth outcomes [9, 10]. Most of male infected with T. vaginalis are asymptomatic, but the infection can also cause urethritis or prostatitis and even lead to infertility [11]. Moreover, T. vaginalis usually increases the risk of other pathogens infection [12], for example HIV, which greatly threaten public health [13]. Some studies indicated that T. vaginalis infection might be a risk factor for cervical cancer and prostate cancer [14, 15].

As is known to all, the main pathogenic pathway of T. vaginalis is to cause the inflammatory reaction of parasitic sites, and the repeated inflammation may induce carcinogenesis. It was reported that T. vaginalis infection could lead to cervical precancerous lesions and neoplastic lesions [16]. The risk of cervical cancer increased about twice in the presence of T. vaginalis [17], and T. vaginalis infection could promote prostate cancer by damaging prostate epithelial cells [18]. Moreover, T. vaginalis infection was associated with hrHPV, the causative agent in most cervical cancers [13, 16].

Although many published articles have shown that T. vaginalis infection can increase the risk of cervical cancer or prostate cancer, the correlation between T. vaginalis infection and reproductive system cancer and whether T. vaginalis can cause reproductive system cancer remain unclear. Therefore, in this study, we analyzed the correlation between T. vaginalis infection and reproductive system cancer through meta-analysis of relevant epidemiological data. In addition, we summarized the potential pathogenic mechanism of cancer caused by T. vaginalis infection through consulting the relevant review and research articles. The results of the study provided a direction for exploring the pathogenic mechanism of T. vaginalis leading to cancer.

Methods

Search strategy and classification

The literature retrieval databases mainly included PubMed database, EMBASE, Ovid MEDLINE medical literature library, Web of Science, Science Direct and Google Scholar, and the literature search time was up to December 2021. The keywords searched by MeSH and commonly used for literature retrieval were used alone or in combination: “Trichomonas vaginalis”, “Trichomonas infections”, “Trichomoniasis”, “Trichomonas vaginitis”, “Neoplasms”, “cancer”, “neoplasia”, “carcinoma in situ”, “canceration” and “tumour” with “OR” and / or “AND” operators. The systematic search of literature was performed by two independent researchers. The retrieved articles were manually checked the title, abstract and full-text by two independent researchers, and the irrelevant articles and duplicate were removed. The retained articles were categorized as epidemiological investigations, reviews, and research articles. This systematic review with meta-analysis was registered in the International Prospective Register of Systematic Reviews (PROSPERO, https://www.crd.york.ac.uk/prospero) and a registration ID was assigned (CRD42022340263).

Inclusion criteria and exclusion criteria

Inclusion criteria for epidemiological investigations

1. The article contains statistical data on cancer in people infected with T. vaginalis and without T. vaginalis. 2. The article contains statistical data on T. vaginalis infection in cancer patients and noncancer patients. 3. In the articles, the data of T. vaginalis infection in cancer patients and noncancer patients are compared. 4. All articles have clear data sources. 5. For the literature with repeated relevant data, the latest and most comprehensive article is selected.

Exclusion criteria for epidemiological investigations

1. The effect of T. vaginalis on cancer is not mentioned in the article, and the relevant data is incomplete. 2. The article only contains data on cancer in people infected with T. vaginalis with no corresponding data from people without T. vaginalis infection. 3. The article only contains data on T. vaginalis infection in cancer patients without corresponding data from noncancer patients. 4. In the articles, the sample size is too small.

Inclusion criterion for review articles

1. The article concerns the correlation between T. vaginalis infection and cancer. 2. The article has a clear conclusion.

Exclusion criterion for review articles

1. The article does not concern the correlation between T. vaginalis infection and cancer. 2. The conclusion about the correlation between T. vaginalis infection and cancer is ambiguous.

Inclusion criterion for research articles

There is a clear carcinogenic mechanism caused by T. vaginalis infection in the article.

Exclusion criterion for research articles

The relevant mechanism of cancer caused by T. vaginalis infection was not written in the article.

Quality assessment and data analysis for epidemiological investigations

Newcastle–Ottawa scale (NOS) was used to evaluate the quality of the epidemiological investigations with a total score of 9 points, and the articles with a score ≥ 6 points were included in statistics. NOS considers the following items comprising selection criteria (0–4 points), subject comparability (0–2 points) and exposure (0–3 points). The selected epidemiological investigations were analyzed by Stata 16 software, and the forest plot was drawn. I2 test was used for heterogeneity analysis. If P < 0.05 or I2 ≥ 50%, it was considered that there was heterogeneity among the research, and the origin of heterogeneity should be further analyzed by the random-effect model and Galbraith plot. The funnel chart was used to test the publication bias of the included articles, and the Begg and egger tests were further used to evaluate the publication bias. If there was publication bias, we excluded articles one by one to make a sensitivity test, and the total effect quantity was observed to judge the stability of the analysis.

Results

In this study, a total of 411 articles were retrieved (Fig. 1), and the title, abstract and full-text of these articles were checked. After removing 267 irrelevant articles which did not mention both “T. vaginalis” and “Neoplasms”, the remaining 144 articles were classified into three categories, 68 epidemiological investigations, 30 review articles, and 46 research articles. In the 144 articles, 143 articles were about T. vaginalis and reproductive system cancer (cervical cancer, prostate cancer and vaginal cancer), and the other one was about T. vaginalis and anal canal carcinoma.

Fig. 1
figure 1

Flow diagram of articles retrieval, identification, classification and inclusion

Full size image

Epidemiological investigations

Literature inclusion

The 68 epidemiological selected articles were identified according to the NOS (NOS score ≥ 6), inclusion and exclusion criteria. Finally, 22 relevant articles were obtained. Fourteen articles (Table 1) were concerned the infection of T. vaginalis in reproductive system cancer patients. In these articles, we set the cancer patients as the experimental group and healthy people or noncancer patients as the control group. The other eight articles were concerned the reproductive system cancer incidence in T. vaginalis infected people. The people infected with T. vaginalis were set as the experimental group, and the people without T. vaginalis infection were set as the control group. The samples of these articles were sufficient for meta-analysis, which were deemed as high-quality articles based on the NOS quality assessment score (NOS score ≥ 6).

Table 1 The information of relevant epidemiological articles
Full size table

Meta analysis of the epidemiological investigations

The data of the 14 articles on T. vaginalis infection in cancer patients were analyzed by Stata 16 software. The results showed that 292,456 research subjects were included in the articles, including 57,467 in the experimental group (cancer patients) and 234,989 in the control group (healthy or non-cancer patients) [15, 19,20,21,22,23,24,25,26,27,28,29,30,31]. As shown in the forest plot (Fig. 2a), P = 0.01, I2 = 52% which indicated that there was heterogeneity among the research results (P < 0.1, I2 ≥ 50%). So random-effect model was used for further analysis, and Galbraith plot analysis was performed for the included articles (Fig. 2c). In the plot, the two articles, Panpan (2019) and Costa (2017), had influence on heterogeneity. The reason for the heterogeneity of Panpan (2019) was that the sample size was relatively small, while the reason for the heterogeneity of Costa (2017) was that the sample size of the experimental group and the control group was quite different.

Fig. 2
figure 2

The articles concerned the incidence of T. vaginalis infection among patients with cancer. A Forest plot of the meta-analysis. The heterogeneity test yielded Chi2 = 27.01, p = 0.01, and I2 = 52%. The OR and 95% confidence interval were calculated as 1.87 and 1.29–2.71, respectively, and z and p values of the combined effect size were as 3.29 and less than 0.001. T. vaginalis infection among patients with cervical tumor: The heterogeneity test yielded Chi2 = 17.22, p = 0.05, and I2 = 48%. The OR and 95% confidence interval were calculated as 1.81 and 1.10–2.96, respectively, and z and p values of the combined effect size were as 2.36 and 0.02. B Funnel plot of the meta-analysis. C Galbraith plot. D Begg and Egger plot of the meta-analysis. P = 0.696 (P > 0.1) E Sensitivity analysis of the articles

Full size image

Combining the data of these 14 articles, the pooled OR was 1.87 (95% CI 1.29–2.71). The results of the combined effect were Z = 3.29 and P = 0.001 which meant that there was a significant difference in T. vaginalis infection between the experimental group and the control group (P < 0.05). As shown in the funnel plot (Fig. 2b), the excellent distribution symmetry of the individual sample points indicated that there was no publication bias. Further, the Begg and Egger plot was used to test (Fig. 2d) the bias, and the result was found that P = 0.696 (P > 0.1) and showed that there was no publication bias in these data. The sensitivity analysis (Fig. 2e) found that the impact on the analysis was not significant after excluding each article, and the 95% confidence interval was still within 1.29–2.70, indicating that the analysis was relatively stable.

In addition, 8 articles on cancer among patients infected with T. vaginalis were screened for meta-analysis (Fig. 3a). The results showed that 746,630 subjects were included in the articles, including 8376 cases in the experimental group (with T. vaginalis infection) and 738,254 cases in the control group (without T. vaginalis infection) [32,33,34,35,36,37,38]. The pooled OR was 2.77 (95% Cl: 2.37–3.25). The results of the combined effect were Z = 12.64 and P = 0 (P < 0.05) which meant that there was a significant difference in cancer patients between the experimental group and the control group (P < 0.05). The results of I2 = 31%, P = 0.18 proved that the heterogeneity was small (P > 0.1). The funnel plot (Fig. 3b) were asymmetric, indicating that there might be publication bias. The Begg and Egger plot was used to test (Fig. 3c), P = 0.051 (P < 0.1) proved that there was publication bias. Because few retrieved articles were included, so it was necessary to add more articles for further study. The sensitivity analysis (Fig. 3d) indicated that the analysis was relatively stable (95%Cl: 2.01–2.58).

Fig. 3
figure 3

The articles concerned the incidence of cancer among people with T. vaginalis infection. A Forest plot of the meta-analysis. The heterogeneity test yielded Chi2 = 10.21, p = 0.18, and I2 = 31%. The OR and 95% confidence interval were calculated as 2.77 and 2.37–3.25, respectively, and z and p values of the combined effect size were as 12.64 and 0 (P < 0.05). B Funnel plot of the meta-analysis. C Begg and Egger plot of the meta-analysis. P = 0.051 (P < 0.1). D Sensitivity analysis of the articles

Full size image

Review articles

In this study, 30 reviews were selected and five articles were finally obtained after screening by inclusion and exclusion criteria. As shown in Table 2, we analyzed the conclusions of each article. These five articles suggested that T. vaginalis could lead to cervical cancer and prostate cancer [16, 17, 39,40,41]. The possible reasons were cell carcinogenesis caused by long-term inflammatory response [16, 17, 39, 40] and cancer caused by T. vaginalis increasing the infection of other pathogens (hrHPV, Chlamydia) [41].

Table 2 The information of review articles
Full size table

Research articles

In this study, 48 research articles were selected and 25 articles were obtained according to the inclusion and exclusion criteria of research articles, of which 24 articles were related to T. vaginalis infection and reproductive system cancer, and the other articles involved T. vaginalis infection and anal canal carcinoma [18, 42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67]. As shown in Table 3, we listed the information of the articles and concluded the mechanism of cancer caused by T. vaginalis infection. Most of the articles showed that T. vaginalis infection could promote cancer procession but 3 articles hold the opposite view. As described in these articles, the main pathogenic mechanisms were as follows: (1) T. vaginalis promoted inflammatory reaction through various ways and leaded to cell carcinogenesis; (2) The metabolites secreted by T. vaginalis promoted the occurrence of cancer; (3) The infection of T. vaginalis affected the vivo environment and signal transduction pathway which was associated with cancer; (4) T. vaginalis increasing other pathogenic microbial infection (HPV) promoted the occurrence of cancer.When phagocytosis of Candida spp. by T. vaginalis occurs, Candida spp. are protected by T. vaginalis from the defences of the host and the inhibitory effects of antimycotic drugs used for treatment finally lead to anal canal carcinoma

Table 3 The information of research articles
Full size table

Discussion

In recent years, there are increasing number of studies on T. vaginalis, and many studies indicated that T. vaginalis was closely related to reproductive system cancer. In this study, we searched for 14 articles on incidence of T. vaginalis infection in cancer patients. By meta-analysis, the combined effect size of the forest plot was determined to Z = 3.29, P = 0.001, which was statistically significant. However, the analysis result of forest plot (P = 0.01, I2 = 52%) indicated a moderate degree of heterogeneity in these articles. Galbraith plot indicated that there was heterogeneity in Panpan (2019) and Costa (2017), but sensitivity analysis showed that the removal of the 2 articles has little impact on the results, and the data was stable. These results signified that the infection rate of T. vaginalis in cancer patients was significantly higher than that in healthy group or non cancer group, which indicated there was a correlation between T. vaginalis infection and cancer. By analyzing funnel plot, Begg and Egger plot, we found that there was no publication bias in our analysis. However, the heterogeneity still affected the quality of the analysis. Although the result indicated that T. vaginalis infection was related to cancer, more epidemiological data should be further excavated.

In addition, there were eight epidemiological investigations on incidence of cancer in people with T. vaginalis infection. In the study, the people infected with T. vaginalis were set as the experimental group, and the people without T. vaginalis infection were set as the control group. By comparing the prevalence of cancer between the two groups, the results indicated that the group infected with T. vaginalis had a high proportion of cancer. There was no heterogeneity in the data (I2 = 31%, P = 0.18). However, the Begg and egger analysis of the data (P = 0.051, P < 0.1) showed publication bias in the result. The reason for this might be related to the small number of such articles in this study. Therefore, the epidemiological investigations on the prevalence of cancer in the people infected with T. vaginalis need further research and more relevant articles need to be included to consolidate the results.

Several review articles on T. vaginalis and cancer were selected in this study. By consulting these articles, we found that T. vaginalis plays a positive role in the occurrence and evolution of cancer. T. vaginalis infection could directly or indirectly lead to cancer. It was found that T. vaginalis infection in women mainly caused inflammation of reproductive tract and the release of proinflammatory factors, changed in vaginal environment and promoted pathogenic microbial infections (HPV) [16, 41] which led to cervical cancer. The main factors leading to prostate cancer in men infected with T. vaginalis were the inflammatory response induced by T. vaginalis or its cytotoxic effect. The infection of T. vaginalis could induce the production of a large number of different proinflammatory cytokines [40].

From the research articles, we found several potential mechanisms of cancer induced by T. vaginalis infection.

The infection of T. vaginalis can promote the inflammatory response and lead to cell carcinogenesis in a variety of ways. Chronic T. vaginalis infection leads to the production of macrophage migration factor and macrophage polarization to M2 type, driving inflammation and abnormal cell proliferation to promote the progress of cervical and prostate cancer [60, 62]. The inflammatory response by BPH (benign prostatic hyperplasia) epithelial cells stimulated with T. vaginalis induce the proliferation of prostate stromal cells via crosstalk with mast cells [55, 67]. Chronic T. vaginalis infections result in TvMIF (T. vaginalis macrophage migration inhibitory factor)-driven inflammation and cell proliferation, thus triggering pathways that contribute to progression of prostate cancer [51]. The initiation of inflammation driven tumor-like cell signaling in parasite-infected human prostatic epithelial cells and the prostate tumor cells are more sensitive to T. vaginalis than normal prostatic cells [63]. The level of IL-6 in BPH-1 cells infected with T. vaginalis is increased, and IL-6 is considered to be a factor promoting the development of benign prostatic hyperplasia and prostate cancer [54, 66].

During the metabolism of T. vaginalis, the contents of propionic acid and iso-valeric acid increase [44]. Both of them had a promoter-like activity and/or promoter-enhancing effect in vitro studies on viral oncology, which could promote the occurrence of cervical cancer or vaginal cancer. Moreover, T. vaginalis infect the prostate cells and lead prostate epithelial cells to express P2X1, P2X2 and P2X7 receptors, affecting the purinergic signaling of host, which is related to prostate cancer [18, 47]. T. vaginalis count and growth rates were significantly higher in trophozoites from CN (cervical neoplasia) and cervical neoplasia proliferates, and the trophozoite surface of CN isolate was creased and rough implying that these were virulent forms which could aggravate cervical neoplasia conditions [47]. There is a relationship between seropositivity for ACT-P2 (truncated protein of trichomonad α-actinin) of T. vaginalis and prostate cancer [49]. T. vaginalis may promote cancer by directly damaging or dissolving prostate epithelial cells [52].

T. vaginalis infection associated with other pathogen infection increase the risk of cancer. The presence of Mycoplasma hominis (M. hominis) in T. vaginalis play a key role in inflammation. The synergistic upregulation of the macrophage proinflammatory response also increase the risk of acquiring cervical cancer [50]. When phagocytosis of Candida spp by T. vaginalis occurs, Candida spp are protected by T. vaginalis from the defences of the host and the inhibitory effects of antimycotic drugs used for treatment finally lead to anal canal carcinoma [48]. T. vaginalis infection associated with Gardnerella vaginalis infection might increase the progression of low-grade cervical intraepithelial neoplasia [61]. Morever, some published articles indicated that T. vaginalis significantly increased the infection of hrHPV [41]. However, which mechanism plays a major role in cancer induced by T. vaginalis infection needs to be further studied.

Conclusions

Through meta-analysis of relevant epidemiological data, we found that there was a correlation between T. vaginalis infection and reproductive system cancer. By consulting the relevant review and research articles, we discovered that T. vaginalis infection could lead to cervical cancer or prostate cancer, and the articles showed that the main potential carcinogenic mechanisms involved inflammatory reaction change the environment around the parasitic site and signal transduction pathway, and aggravate the infection of other carcinogenic pathogenic microorganisms. Our study provides a foundation for further investigation to the mechanism of cancer caused by T. vaginalis in the future.