Background

HPV infection can cause a variety of genital diseases, and type-specific persistent infection of high-risk HPV (hrHPV) cases is strongly associated with cervical carcinogenesis [1]. Cervical cancer is the third most common type of cancer in women worldwide [2]. Effective implementation of cervical screening programs in developed countries has resulted in a steady reduction in the incidence of cervical cancer [3]. However, in China, the most populous country, cervical cancer remains the second leading cause of cancer deaths among 15- to 44-year-old females [4]. It is estimated that 75,434 women are diagnosed with cervical cancer annually (11.3/100,000) and 33,914 (45.0 %) of those women die because of cervical cancer [5, 6].

To date, more than 200 HPV genotypes have been identified, and ~40 HPV genotypes have been detected in the female genital tract. HPV16 and HPV18 are well known as oncogenic genotypes; additionally, HPV31, HPV33, HPV35, HPV39, HPV45, HPV51, HPV52, HPV56, HPV58, HPV59, HPV68, HPV69, and HPV 82 are also closely associated with cervical cancer. Therefore, all of these genotypes are classified as “high-risk” HPV. Meanwhile, “low-risk” genotypes, including HPV6, HPV11, HPV42, HPV43, and HPV44 are the causative agents for benign or low-grade changes in cervical cells, such as genital warts [7, 8].

The current hrHPV detection is supposed to serve as an additional approach for the early diagnosis of cervical cancer, and was implemented to complement less sensitive and non-objective cytology-based methods [3]. The high negative predictive value of hrHPV testing is applicable for the indication of a low-risk population, in which the cervical cancer screening interval can be safely extended [3]. Based on the results of clinical trials, several European countries will implement hrHPV testing as the primary screening modality [9]. In addition to HPV screening, HPV vaccination has been shown to be an effective strategy against HPV infection and has been recently implemented in most western countries. Although Cervarix (HPV16/18) and Gardasil (HPV6/11/16/18) protect against infection by HPV16 and HPV18, these vaccines provide no effect on some of the hrHPV types found in at least 25 % of cervical cancers [10]. Furthermore, the role of non-vaccine HPV types in the development of lesions remains unknown, and it remains possible that non-vaccine HPV types could replace these vaccine types as the causative agents for cervical precancerous lesions and cancer in vaccinated cohorts without sufficiently broad cross-protection [10].

The prevalence of HPV infection and type-specific distribution vary greatly both between nations [11], and between different regions within countries [12]. Additionally, several other risk factors can influence the prevalence of HPV, including genetic variation, sexual behavior (age at first sexual intercourse and individuals with multiple sex partners), biological predisposition of the immature cervix, and immunodeficiency [13]. Hence, surveillance of the general population is needed to assess the clinical benefits of screening and vaccination strategies.

Obtaining large-scale information about the epidemiological features of HPV in China is critical for global HPV prevention strategies, because of the need to understand the geographic diversity and age distribution of HPV infections. Although a pilot project aimed to measure the hrHPV infection rate and cervical cancer screening was conducted in a few Chinese areas in 1999 [14], and several similar larger scale investigations were carried out that covered more regions that were conducted in 2003, 2008, 2009, and 2012 [15–18], the available data remain insufficient [9] and outdated. Obtaining a more current dataset will also provide a reference for effective screening and vaccination. Herein, we describe a nationwide cross-sectional and large-scale study that had the following aims: 1) to reveal the prevalence of HPV in regions not yet investigated; 2) to follow-up potential changes in HPV infection in regions that have been previously studied; 3) to clarify the genotypic distribution of HPV in different regions and age-grouped populations. These data should expand the data available regarding HPV-related cervical cancer and aid in future research, screening, and vaccination efforts.

Methods

Ethics Statement

This study was approved by the Ethics Committee of Tian** Medical University in accordance with the Ethical Principles for Biomedical Research Involving Human Subjects (Ministry of Health of the People’s Republic of China) and Declaration of Helsinki for Human Research of 1974 (last modified in 2000). Samples were originally obtained from clinical settings for laboratory diagnoses. After diagnostic testing, excess samples were anonymized and kept for this study. Informed consent was obtained from each female participant. For those individuals younger than 18 years old, the consent form was signed by the parents of each participant.

Study population

KingMed Diagnostics is the largest reference laboratory in China and provides diagnostic testing services for over 13,000 hospitals in 18 provinces and 4 municipalities across the nation. From January to December of 2012, a total of 120,772 samples were obtained for population- or employee-based screening from 37 cities, belonging to 18 regions (Guangdong province was regarded as one region as 20 cities in this investigation were located in that province). Fig. 1 shows a map of all of the geographical sites in China and the 18 regions were further grouped into 4 macro-geographical regions (East, West, South, and North).

Fig. 1
figure 1

National map of China showing all the geographical sites included in this study

Full size image

The women enrolled from population- or employee-based cervical cancer screening programs ranged in age from 15 to 60 years old, were sexually active, and had no history of cervical treatment before the screening. Exclusion criteria included current pregnancy, <3 months post-partum, HIV-seropositivity, and a history of either hysterectomy or treatment for cervical cancer.

Among all of the samples, 111,131 were collected from 15 regions (Haikou, Chongqing, **an, Jilin, Shenyang, Tian**, Shanghai, Nanning, Guangdong, Guiyang, Fuzhou, Hangzhou, Chengdu, Changsha, and Nanchang; Fig. 1) and were analyzed using Hybrid Capture II (HCII); 105,069 (94.5 %) of these samples could be grouped by age. The other 9641 samples from 10 regions (Shanghai, Guiyang, ** was performed using the Tellgenplex™ HPV DNA Test (Tellgen Life Science Co., Shanghai, China). The assay can be used to detect and genotype 26 HPV genotypes including 19 hrHPV genotypes (HPV16, HPV18, HPV26, HPV31, HPV33, HPV35, HPV39, HPV45, HPV51, HPV52, HPV53, HPV55, HPV56, HPV58, HPV59, HPV66, HPV68, HPV82, and HPV83) and 7 lrHPV genotypes (HPV6, HPV11, HPV40, HPV42, HPV44, HPV61, and HPV73). The Tellgenplex HPV™ kit was used to simultaneously examine the presence or absence of the most common 26 HPV genotypes in a single test by multiplex PCR combined with Luminex technology [22, 23]. The three steps of DNA extraction, PCR amplification, and hybridization were included in the procedure and a template of 10–20 pg/ml HPV DNA was needed for each assay.

Statistical analyses

Region-specific prevalence of HPV

The HPV infection rate in each region was calculated by dividing the number of HPV-positive samples by the total number of samples that were successfully tested for HPV. A binomial 95 % confidence interval (95 % CI) was estimated for each calculation of the prevalence of HPV. Chi-squared (χ2) tests were used to compare differences among all regions and every two regions.

Age-specific prevalence

The HPV infection rate was estimated within 5 age groups (15–19, 20–29, 30–39, 40–49, and 50–60). A binomial 95 % confidence interval (95 % CI) was estimated, and P-values for age trends of HPV infection were analyzed using the linear-by-linear association test. Differences between each pairing of two age groups were compared by the chi-squared (χ2) test. Multiple comparisons were performed using the Bonferoni step-down procedure to minimize an inflated risk of type 1 error [24]; P < 0.05 was considered statistically significant.

HPV-type-specific prevalence

The frequency of each hrHPV and lrHPV genotype was presented in hrHPV-positive samples and lrHPV-positive samples, respectively.

All statistical analyses were conducted using SPSS20.0 software (SPSS lnc., Chicago, IL, USA).

Results

The total prevalence of hrHPV infection

The total hrHPV infection rate was 21.07 % (95 % Cl 20.83–21.31 %). The prevalence of hrHPV infection differed significant among the various regions (P < 0.001). The regions with the highest hrHPV prevalence were Haikou (31.94 %) and Chongqing (27.29 %). By contrast, Nanchang, Changsha, Hangzhou, Chengdu, Fuzhou, Guangdong, and Guiyang could be grouped into the low prevalence regions (Table 1).

Table 1 Region-specific prevalence of hrHPV infection by HCII
Full size table

As shown in Fig. 2, hrHPV infection was associated with age. The group of 15–19-year-olds showed the highest prevalence (30.55 %), followed by the 50–60-year-olds (23.30 %). The total infection rate of hrHPV was associated with age (P < 0.001, Additional file 1: Table S1). After the highest peak, the prevalence declined, but then the infection rate significantly increased again in the 50–60-year-old group compared with the 40-49-year-old group (P < 0.001, Additional file 1: Table S1). Among the four macro-regions, only the south showed a similar trend with the overall age-specific hrHPV prevalence, whereas the other three macro-geographic regions did not show significant differences among all five age groups (Fig. 2, Additional file 1: Table S1).

Fig. 2
figure 2

Age-specific hrHPV infection by HCII

Full size image

The distribution of variant HPV genotypes

Among the well-recognized 26 HPV genotypes, all genotypes except HPV73 were examined using the Tellgenplex technique. The total infection rate of each genotype among all of the samples and the distributions of each genotype in the HPV-positive individuals were respectively evaluated. In descending order, the infection rate and distribution are shown in Table 2: the most common hrHPV types were HPV16 (18.02 %) and HPV52 (16.9 %) (P = 0.288), moreover, HPV6 (45.80 %), HPV11 (26.15 %), and HPV61 (14.20 %) were common in lrHPV.

Table 2 The prevalence of each HPV genotype by the Tellgenplex™ HPV DNA Test
Full size table

Regarding the region-specific distribution of hrHPV, the top three genotypes were analyzed in each region. HPV16, HPV58, and HPV52 were dominant in six regions—Guiyang, ** countries or a second peak in developed countries [25]. The trend wherein hrHPV infection shows high rates in younger groups and low rates in middle age groups reflects the natural history of HPV infection. Young females are sensitive to HPV soon after beginning of sexual activity because of immature immune protection. Nevertheless, most cases of HPV infection are usually temporary, so in 70 % and 91 % of women infected with HPV, the virus could be cleared within one or two years, respectively [33]. The slight increase in the HPV infection rate in older females might reflect the viral persistence or reactivation of latent HPV, likely because of the physiological and immunological disorders that can result from hormone fluctuations during the menopausal transition [34]. In this study, the general age distribution showed a first peak of hrHPV in the age group of 15-year-old patients (30.55 %), then gradually decreasing in middle age, which is consistent with data from Bruni et al. [25]. However, the hrHPV infection rate was significantly increased in 50–60-year-old individuals compared with women in their 40s, which is a similar trend that has been observed in most developed countries. The prosperous economy in most major cities could influence the culture and sexual behavior of Chinese individuals [15]. Nonetheless, the exact mechanisms for the increase in HPV prevalence still remain unclear.

Data concerning the distribution of HPV genotypes is important for both vaccine development and HPV-based screening design, particularly for selecting the testing spectrum of HPV genotypes [35] and detecting multiple HPV infections. It is a prerequisite for the genoty** assays in cervical cancer screening programs. However, few HPV devices have been cleared by the FDA, including the Roche Cobas 4800, which also provides limited genoty** data [36]. Therefore, the Tellgenplex™ HPV DNA Test, an established genoty** method [23], was performed in this survey. By following CAP (College of American Pathologists) principles, the validation of Tellgenplex™ HPV DNA Test were conducted compare to HCII in the molecular biology lab of KingMed, and the correlations that we detected were found to be (coincident rate = 90.5 %, Kappa = 0.88).

Consistent with the data generated by some Chinese population-specific investigations, HPV16, HPV52, and HPV58 were found to be the dominant hrHPV types [37, 38]. Compared with the data of Wu et al. [28], a population-based investigation from five regions (Bei**g, Shanghai, ** review. Journal of medical economics 2013;16(6):763-76" href="/article/10.1186/s12879-015-0998-5#ref-CR45" id="ref-link-section-d84523792e2349">45]. The infection can become more serious in immune-compromised individuals [44]. In our present survey, the incidence of HPV6 (4.01 %) infection was inconsistent with the results of Bruni et al. [25], who showed that HPV6 was most frequent among Americans (2.9 %) and was less frequent in Asian individuals (0.2 %), followed by HPV11 and HPV61.

Characterization of the prevalence of multiple HPV infections might be important for understanding its effects on cervical carcinogenesis. Herrero et al. [10] reported that women infected with HPV16 alone were at a similar or higher risk for cervical cancer than those infected with both HPV16 and another HPV type. Lee et al. [46] reported an association between infection with multiple HPV types and an increased risk of cervical cancer. In a recent study, Schmitt et al. [47] confirmed that co-infection would increase the duration of infection. Furthermore, patients with multiple high viral loads showed a 4- to 6-fold increased risk of cervical precancerous cytological lesions compared with patients with single high viral loads. In this present study, 434 hrHPV-positive samples (4.5 %) were multiple-infections. The incidence was in the same range (3.5–5.3 %) as was previously reported in China [37], but was higher than the global average (3.2 %) [25]. Furthermore, the rate of infection (14.19 %) was lower than reported for both domestic (25.8 %) [28] and international (20 %) [25] populations. The most common combinations of two types were HPV16 + HPV52 (26 cases) and HPV52 + HPV58 (14 cases), and the genotypes HPV52 and HPV58 were more likely to be involved in co-infections. Regarding region-specific surveillance, some geographical features were observed. For example, in Shanghai, a city with a large internal population, the situation was close to the world average regarding HPV prevalence, genotypic distribution, and multiple infections.

This present study confirmed the high overall incidence of HPV in China and strongly argues for the necessity of develo** national population-based screening programs. However, the appropriate management of this HPV screening program for a large number of women with HPV-positive specimens and the absence of cytological evidence of cervical pre-cancer or cancer remains a major concern [48]. HPV genoty** could be an option to stratify the HPV-positive women. Furthermore, a stainable HPV detection and close follow-up program for hrHPV carrier women should be implemented. Thus, more cost-effective techniques, such as the Cervista™ HR HPV test, COBAS HPV test, and some other genoty** platforms, might be good alternatives for molecular detection of HPV [49].

Conclusions

In conclusion, China has large population and a variety of territories; meanwhile, economic conditions, cultural habits, and population migrations have dramatically affected Chinese lifestyle and health, for instance, cancers related to sexual transmitted diseases. Therefore, attention should be paid to the prevalence of HPV infection in a timely and regional basis because it has been commonly recognized that HPV infection plays a critical role in the occurrence of cervical cancer and its increase incidence. These surveillance findings indicated that a national plan for a cervical screening program is urgently needed, not only because of the increase in hrHPV infection rate in some previously reported regions, but also because of the high infection rate in most newly investigated regions. In brief, the most significant findings of this study are as follows: (1) the prevalence of hrHPV infection has reached a level that cannot be ignored, and the rate of increase is growing with time; (2) the prevalence of hrHPV infection showed population variations in age and region, and also reflected economic, cultural, and lifestyle relevance; (3) the HPV16, HPV52, and HPV58 strains consistently constituted the three dominant genotypes in different Chinese populations, a characteristic pattern that was significantly different from the epidemiological features in most of industrial countries. These findings defined principles for future proposals for cervical screening and vaccination in China.