Risk factors for the development of cervical cancer: analysis of the evidence

Introduction Cervical cancer (CC) is the fourth most prevalent female cancer globally. Understanding its epidemiology is crucial for devising practical strategies suited to geographic and social contexts to attain the global eradication of CC. Hence, this study examined the latest evidence of risk factors contributing to CC development. Methods An independent literature search was conducted on PubMed using MESH terms. The primary sources were meta-analyses published from 2010 to 2023, which detail updated evidence on risk factors associated with CC. Additionally, the quality of the evidence was evaluated using the GRADE system and recommendations were made accordingly. Results The main risk factors related to the cause of CC include co-infections with other sexually transmitted infections, genetic markers, cervicovaginal microbiota, nutritional factors, comorbidities that affect the immune response, smoking, and the use of hormonal contraceptives with a quality evidence based on the GRADE scale moderate. Conclusions Since the necessary cause for CC is persistent cervicovaginal HPV, all the risk factors implicated in the causality of CC act as non-independent cofactors that increase the risk of CC. Thus, changes in public policies aimed at addressing these risk factors are highly recommended and can substantially decrease the risk of CC.


Introduction
Cervical cancer (CC) is the fourth most prevalent form of cancer among women worldwide.In 2020, the age-standardized incidence rate was 13.3 cases per 100,000 woman-years, and the mortality rate stood at 7.3 deaths per 100,000 woman-years (1).Sub-Saharan Africa, Latin America, and Asia-regions with countries possessing a low human development index-have the highest incidence and mortality rates (2,3).This discrepancy predominantly stems from the absence of extensive screening programs and insufficient healthcare infrastructure.Conversely, countries with well-established screening initiatives have seen a significant decline in CC cases (4).
In Mexico, 9,439 new CC cases were reported in 2020 (constituting 4.8% of total cases) and 4,335 deaths, with an estimated prevalence of 25,026 cases (5).Approximately 77% of women receive a diagnosis in locally advanced stages, 16% in early stages, and 7% in advanced stages (6).While mortality rates from this cancer have been decreasing since 2001 in the country's central region, the highest mortality rates are found in some of the most marginalized states, such as Chiapas, Tabasco, and Morelos (7).
A persistent infection with high-risk human papillomavirus (HPV) is the leading cause of CC development.There exist certain factors that increase the potential for exposure to and acquisition of an HPV infection at the cervicovaginal level, as well as structural elements that make a woman more susceptible to CC.
As the female cancer is the greatest prevention potential (8,9), understanding the etiology of CC is crucial in response to the Global Initiative to eradicate CC as a public health issue (10).Given this context, this study aimed to analyze and present the available evidence of risk factors associated with the development of CC.

Methods
We conducted a scoping review of meta-analyses published from 2010-2023 in the MEDLINE database via the PubMed database.The search criteria consisted of combined MeSH terms: "risk factors," "smoking," "contraceptives," "genetic markers," "microbiota," "immunity," and "uterine cervical neoplasms."The Boolean operator "AND" was applied to link the search terms and answer the question: "What is the updated evidence on risk factors associated with cervical cancer?"The search was limited to full-text articles published in English.The quality of the evidence presented in these meta-analyses was assessed as high, moderate, low, or very low according to the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system.Recommendations, based on the strength of evidence, were made by members of the Mexican National Consensus on Cervical Cancer Epidemiology.

Results
This review provides an update on the evidence found in the literature regarding the association of risk factors with biological plausibility for CC.Given the essential role that persistent high-risk human papillomavirus (HR-HPV) infection at the cervical level plays in the development of CC, we examined the factors associated with HR-HPV infection (exposure variable), the persistence of HR-HPV infection (intermediate phenotype), and the specific outcome of CC (Figure 1).The analyses of evidence analysis using the GRADE approach on risk factors related to the causality of CC are listed in Table 1.3.1 Risk factors associated with a greater probability of HR-HPV infection at the cervical level

Number of sexual partners
Studies have observed that an increased number of sexual partners correlates with a heightened risk of obtaining an abnormal Pap smear result [odds ratio (OR): 5.5] (11).This risk for HPV infection similarly escalates with increasing numbers of sexual partners (12).In Peru, reports indicated a higher risk of HPV infection among individuals who had more than five sexual partners throughout their lifetime (13).Similarly, researchers in the United States and China found an association between HPV infection and having two or more sexual partners (14).A rise in risk with multiple partners (OR: 1.91) and high-risk genotypes was also reported in Tunisia (15).Moreover, studies in Mexico revealed that having more than five sexual partners heightened the risk for HPV-16 and non-HPV 16/18 infection (16).

Age of onset of sexual life
Another extensively researched risk factor concerning sexual history is engaging in sexual activity at an early age.A study conducted in China found that the risk of HPV infection increased when sexual activity commenced at 19 years old or younger (OR: 1.51) (17).Similarly, researchers in Peru noted an elevated risk of HPV infection (OR: 1.4) when sexual relations began at an age younger than 18 years (13).

Coinfection with Chlamydia
A meta-analysis found an increased risk for CC associated with Chlamydia infection (OR: 1.96) (18), a finding consistent with another meta-analysis (OR: 2.21) (19).The same meta-analysis suggested an even greater increase in CC risk with concurrent Chlamydia and HPV infections (OR: 2.13) (18).Separate studies have also reported Chlamydia to be more prevalent in HPV-positive women compared to those who are HPV-negative (20).Furthermore, in women with Chlamydia, the risk for HPV infection increases (OR: 2.21) (21).In addition, the literature has shown that past Chlamydia infection is a risk factor for contracting HPV (OR: 1.72) (21).

Co-infection with herpes simplex virus
A study in Mexico on herpes simplex virus (HSV-2) documented that the likelihood of having an active HSV-2 infection in HR-HPV-positive cases was nine times higher than in negative cases (p = 0.03).Furthermore, the primary factors related to an active HSV-2 infection were a history of risky sexual behavior and HR-HPV infection (22).
3.2 Risk factors associated with a higher likelihood of persistent HPV infection at the cervical level 3.2.1 Endogenous factors

Genetic factors
Most studies examining genetic factors as risk factors for CC have been association studies focusing on single nucleotide polymorphisms (SNPs) within candidate genes involved in oncogenesis and cellular immune response (23).These studies relate primarily to immune response evasion in patients persistently infected with HPV and CC (24,25).
To date, few comprehensive meta-analyses have investigated the association between genetic polymorphisms unrelated to a specific biological pathway and the risk of CC.These studies primarily focus on the evidence reported in the existing literature (26)(27)(28)(29).A recent meta-analysis of studies that examined the association of SNPs in genes coding for cytokines found SNPs in IL-17A, IL-17F, IL-12A, IL-12B, TNFA, IL-1B, IL-6, and IL-10 (26).Another meta-analysis focusing exclusively on case-control studies reported a negative association between CC and a polymorphism of the p21 gene, a potent cell proliferation and DNA replication inhibitor, as well as two polymorphisms of the BRIP1 gene, a crucial gene in the BRCA-associated DNA repair process (27).In contrast, the meta-analyses of Wang et al. (29) and Zhang et al. (28) reported divergent SNPs significantly associated with the risk of developing CC, potentially due to differences in defined inclusion criteria.

Cervicovaginal microbiota
Emerging evidence suggests that increased diversity of the vaginal microbiota, coupled with a reduced relative abundance of Lactobacillus spp.may play a role in the acquisition and persistence of HPV, as well as the development of precancer and CC (30,31).There are only a few published meta-analyses to date that focus on the results from studies exploring the causal relationship between vaginal microbiota and CC (32,33).The 2019 meta-analysis by Brusselaers et al. reported an association between vaginal dysbiosis and a higher risk of HPV incidence (relative risk [RR]: 1.33), HPV persistence (RR: 1.14), high-grade lesions, and CC (RR: 2.01) (33).In contrast, the 2019 meta-analysis by Wang et al. focused solely on the relationship between cervicovaginal microbiota dominated by Lactobacillus spp.and HR-HPV, cervical intraepithelial neoplasia (CIN), and CC infection using data from cross-sectional studies (32).This analysis reported a protective association related to the detection of HR-HPV infection (OR: 0.64), CIN (OR: 0.53), and CC (OR: 0.12) (32).Given that malnutrition exists in some patients at the time of CC diagnosis, it is conceivable that nutritional deficiencies may contribute to the disease's pathogenesis, given their intimate association with the immune system.Some of the nutrients potentially involved include vitamins A, C, D, and E, calcium, and various antioxidants (34).Nevertheless, one meta-analysis found elevated beta-carotene levels in the blood to be protective against CC development (OR: 0.48) in stark contrast to high vitamin A levels (35).In another study, consumption of over 502.6 mg/dL of calcium (OR: 0.54) and more than 291 IU of vitamin D (OR: 0.51) was identified as a protective factor against the development of invasive CC, except in individuals who smoke or consume alcohol (36).Other meta-analyses reported higher blood levels of vitamin E (OR: 0.52) and selenium (OR: 0.55) associated with protective effects against CC (37, 38).

Other nutritional indicators
Patients recently diagnosed with CC may exhibit lower circulating levels of non-enzymatic antioxidants (e.g., glutathione) and enzymatic antioxidants (e.g., glutathione S transferase, glutathione peroxidase, and superoxide dismutase).They may also have lower levels of vitamins C and E compared to patients who do not have cancer.These discrepancies could be attributed to the more significant elimination of lipid peroxides and the sequestration of glutathione by tumor cells (39).In addition, another meta-analysis revealed that while being overweight did not have a significant association with CC, obesity did have a slight correlation (OR: 1.1) (40).

3.2.1.4
Comorbidities that condition the immune response 3.2.1.4.1 Acquired immunodeficiency virus Acquired immunodeficiency virus (HIV) infection significantly accelerates carcinogenesis in the progression of HPV infection.Studies have found that women with HIV are at a higher risk of contracting and spreading HPV (41, 42), predominantly when their CD4 count decreases (42).This demographic has a heightened incidence of both low-and high-grade squamous intraepithelial lesions and an elevated risk for developing CC, commonly as a result of HPV strains 16 and 18 (42)(43)(44).A recent meta-analysis found that women with HIV have a 4.1 higher risk of developing CC than women without HIV (RR: 4.1) (42).

Systemic lupus erythematosus
Systemic lupus erythematosus (SLE) can lead to a dysregulated immune system, potentially prompting persistent HPV infection and subsequent CC development.Women with SLE have an increased risk of HPV infection (45), developing cervical atypia (46), accruing low-grade squamous intraepithelial lesions, and CC (46,47).The connection between SLE and CC development was substantiated in two independent meta-analyses, both of which indicated an increased risk for CC (46,47).

Other comorbidities
Gestational diabetes and rheumatoid arthritis have not been found to increase the risk of CC (48,49).Furthermore, HPV infection has not been shown to be associated with immunosuppressive therapy or any other treatment (49).

Smoking
Epidemiological studies have suggested a dose-response relationship between cervical neoplasia/CC and smoking, a proposition partially corroborated by experimental studies (50).Malevolti and collaborators performed a meta-analysis that reported a combined RR of preinvasive lesions and CC for current smokers (RR: 2.11) versus never-smokers (RR: 1.70) and for former smokers (RR: 1.29) versus never-smokers (RR: 1.13).The risk increases to over 2 with a habit of approximately 20 cigarettes/ day or 15 pack-years for invasive CC and about nine cigarettes/day or eight pack-years for preinvasive lesions.However, the risk subsides about 15 years after cessation of smoking (51).In terms of passive smoking as a risk factor for CC, the results are inconsistent.A multicenter case-control study did not identify passive smoking as a risk factor for invasive CC (52).
Nonetheless, a separate case-control meta-analysis reported a 73% heightened risk of CC (53).

Extended use of hormonal contraceptives
In a study by Gadducci et al. ( 54), an increased incidence of CC was reported in users of oral contraceptives for periods of 5-9 years (RR: 1.3-1.6)and 10 years (RR 2.2-2.5).These findings corroborate a meta-analysis that reported an increased OR with oral contraceptive use for 2-5 years (OR: 1.36), >5 years (OR: 1.93), and >10 years (OR: 2.24) (55).Another study reported a significant association between oral contraceptive use for 15 years and higher risk of CIN3/cancer in situ (hazard ratio [HR]: 1.6) and invasive CC (HR: 1.8) compared to non-use.Former menopausal hormone therapy use was associated with a reduced risk of invasive CC (HR: 0.5).A restricted analysis of HPV-seropositive cases and controls revealed an inverse association between intrauterine device use and CIN3 (56).A meta-analysis examining the relationship between hormonal contraceptives and the risk of CC in various ethnic groups found no link between Caucasian, African, and mixed populations' oral contraceptive use and CC, although there was a higher risk for CC in Asian women (OR: 1.43) (57).

Structural risk factors related to CC
Socioeconomic factors, while not direct causes of CC, can increase a woman's susceptibility to it.Women with low socioeconomic status (SES), residing in rural areas, or with limited education often delay medical care, thereby elevating their risk levels (58)(59)(60)(61).Evidence has also shown that a lower SES correlates with a higher chance of advanced CC (62,63).One Turkish study discovered a linear relationship between education level, understanding of CC, and coping capacity for certain situations (64).In Ethiopia, lower levels of education also corresponded to limited knowledge of CC (65).Meanwhile, in Uganda, less education was associated with a higher risk of contracting high-risk human papillomavirus (HR-HPV) (66).In Canada, there were more CC cases in rural and poorer areas than in urban or affluent locations (67).A meta-analysis suggested that low SES (OR: 2.68) or a low level of education (OR: 1.97) increases the risk of developing CC (68).
One strength of this study is its synthetic and updated review of risk factors related to the causality of CC, which will be valuable to individuals interested in epidemiology and causality analysis.Nevertheless, the main limitation is that some meta-analyses about the causes of CC are based on primary observational studies with a high risk of bias.n.Co-infection of HPV with other sexually transmitted infections (e.g., Chlamydia and HSV-2) is the only known sexual life history risk factor that has been moderately proven to be causally related to cervical cancer.o.An updated evaluation of the available scientific literature evidence via the GRADE system on CC causality and risk factors provides valuable, synthesized information for those interested in CC causality analysis and epidemiology.

Recommendations
a. From a public health standpoint, all the lifestyle-related risk factors examined, irrespective of the proposed causal mechanism being direct or confounding, are modifiable elements that, when reduced, can significantly decrease the risk of cervical cancer.The quality of the supporting evidence is high (GRADE), and the strength of the recommendation is strong.b.The medical literature consistently underscores a scientifically-backed public health message regarding the importance of preventing and eliminating lifestyle factors tied to the causation of cervical cancer.By addressing these factors, the risk of cervical cancer can be reduced.The quality of evidence supporting this is high (GRADE), and the strength of the recommendation is strong.c.Cervical cancer is the type of cancer in women with the highest potential for prevention.That said, there is an urgent need to enhance health promotion efforts, including providing health-related information on the risks of tobacco use and hormonal contraceptive utilization, implementing age-and culture-appropriate sex education, and promoting condom use.The quality of evidence supporting these measures is high (GRADE), and the strength of the recommendation is strong.d.Fortifying health promotion programs offered in primary care units, such as adolescent and youth-friendly services at the national level, is paramount.The quality of supporting evidence is high (GRADE), and the strength of recommendation is strong.e. Increased coverage of the primary prevention program at the national level is necessary.The quality of the supporting evidence is high (GRADE), and the strength of the recommendation is strong.f.It is necessary to ensure adherence to available regulations and official guidelines for HPV infection prevention and CC control, considering HPV infection is the most common sexually transmitted infection leading to CC.The quality of evidence supporting this is high (GRADE), and the strength of recommendation is strong.

1 A
FIGURE 1 A directed acyclic diagram that represents the causal structure of cervical cancer.This directed acyclic diagram illustrates the exposure variable and the interplay of other risk factors affecting CC causality.The variables are represented by nodes (circles), while the arrows represent the causal direction between variables.Green arrows represent open causal pathways, black arrows symbolize closed non-causal pathways, and pink arrows indicate open non-causal pathways.(A) exhibits risk factors associated with the onset of HPV-HR infection (depicted by green circles).Blue circles with bidirectional arrows denote common causes predisposing to HPV-HR infection persistence and CC development.Blue circles with a single arrow represent factors linked to the suppression of the immune response and CC progression.Furthermore, infection persistence is considered a mediator in the association between HPV-HR infection and CC.Sexual history-related risk factors that can increase the likelihood of HPV-HR infection.Lifestyle-related factors such as smoking and use of hormonal contraceptives for periods (>5 years), along with the host's intrinsic features such as genetic factors, cervicovaginal microbiota, and immune response elements, are linked to a higher likelihood of HPV-HR infection persistence.Structural factors related to social determinants impacting women's health, particularly in countries like Mexico, are associated with a higher incidence of CC. (B) displays the variables associated with exposure and the outcome; however, they are not part of the causal chain (confounders).These confounding variables (revealed as red circles) must be controlled during the study design or adjusted during data analysis to prevent spurious associations when examining the relationship between HPV-HR infection persistence and CC.HR-HPV, high-risk human papillomavirus; CC, cervical cancer; STI, sexually transmitted infection; HIV, human immunodeficiency virus; HC, hormone contraceptives; y, years.

TABLE 1
Analysis of evidence using the GRADE system by risk factors related to the causality of cervical cancer.