A cross-sectional study to estimate high-risk human papillomavirus prevalence and type distribution in Italian women aged 18–26 years

This cross-sectional study was carried out in the period 2007–2009, coordinated by the Istituto Superiore di Sanità (ISS, Italy’s National Institute of Public Health) and partially sponsored by the Ministry of Health. It was part of a national project (PreGio), which also included a KAP (knowledge-attitudes-practices) survey on HPV infection and CC prevention [13] and a study to estimate the acceptance rate of HPV vaccination [14] .

A sample of 2,289 women was randomly selected from the resident population lists of ten Local Health Units (LHU) located in six Italian Regions scattered across the country: Abruzzo and Campania in southern Italy (1,026), Lazio and Tuscany in central Italy (388) and Emilia-Romagna and Piedmont in northern Italy (875). Both rural and urban LHUs were involved; all 10 LHUs have population-based CC screening programmes.

The inclusion criteria for the study were: women aged between 18 and 26 years and living in the selected LHUs. Pregnant women and women who did not speak Italian were excluded. The women were stratified into two age groups (18–24 and 25–26 years), partially overlapping (25–26 years) with the target age of population-based CC screening. Procedures for enrolling participants as well as study methodology have already been described [13, 14].

All women were offered Pap-test and HPV test; a written informed consent was obtained by each participant. Selected midwives, especially trained for this study, were in charge of contacting sampled women, collecting socio-demographic data and information regarding sexual and reproductive history, getting informed consent and performing cervical smears. The study was approved by the national ethics committee of the ISS.

Cervical smears were taken according to the consolidate CC-screening procedures at the LHU. In most Regions (Tuscany, Piedmont, Emilia-Romagna, Lazio, Campania) two cervical samples were taken: one for Pap-test and one for the HPV test in Specimen Transport Medium (STM, DNAPAP cervical sampler, Qiagen, Gaithersburg, USA); for STM sample, cytobrush was used to obtain samples. In Abruzzo Region the cervical cell specimens were put in PreservCyt solution (Hologic, Inc., Marlborough, MA) and used both for HPV testing and for cytological examination; for Preservcyt sample, Ayre spatula and cytobrush were used. The cytology was interpreted locally according to Bethesda 2001 system [15]. Information on reproductive, sexual and cervical screening history were collected (Table 1).

Women aged 25–26 years followed the routine protocol of treatment or follow-up according to cytological results. A specific protocol for 18–24 year-old women was adopted: colposcopy was offered to women with HSIL (high squamous intraepithelial lesions) and ASCH (atypical squamous cells that cannot exclude high-grade lesions), while women with LSIL (low squamous intraepithelial lesions) or less severe lesions were invited to repeat pap-test two years later and colposcopy was offered only if, after two years, the diagnosis was still LSIL or it progressed to a more severe lesion.

The HPV test was performed using Hybrid Capture 2 (HC2 High-Risk HPV DNA, Qiagen; Hilden, Germany) according to the manufacturer’s instructions. It is a hybridization assay which detects the presence of HPV-DNA using cocktails of RNA probes and an amplified, chemiluminescent signal. The high-risk group of probes B, designed to detect HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68, was used. The assay is calibrated on a positive cut off (pc) of 1 pg/ml of HPV-DNA. Samples were considered positive when the ratio between the Relative Light Units (RLU) of specimen and the pc attained or exceeded the value of 1.0.

A regional laboratory was identified in Tuscany, Piedmont, Abruzzo and Emilia-Romagna Regions for performing HPV testing; the Analytical and Biomolecular Cytology Unit of the Cancer Prevention and Research Institute (ISPO) in Tuscany also analysed samples from Campania and Lazio Regions.

Genotyping was performed on HR-HC2 positive samples. DNA was extracted using the QIAamp DNAMini kit (Qiagen) according to the manufacturer’s instructions. For PreservCyt™ samples the elution volume varied from 80 to 100 μl depending on the size of the pellet, while for STM samples 100 μl of elution buffer (Buffer AE) were used. To facilitate a higher recovery of DNA a double elution was done for all the samples. HR-HC2 positive samples were typed using the “Consensus High Risk HPV genotyping” kit assay (Digene Corporation, Gaithersburg, USA). The test is based on the reverse hybridization principle. Denaturated biotynilated amplicons, resulting from the amplification of a part of the L1 region with GP5+/GP6+ primers, were hybridized with specific oligonucleotide probes, which were immobilized as parallel lines on membrane strips (Reverse Line Blot Hybridization). The hybrids were detected with alkaline phosphatase–streptavidin conjugate and substrate (5-bromo-4-chloro-3-indolylphosphate and nitroblue tetrazolium), resulting in a purple precipitate at positive probe lines. The kit allows the identification of 18 HPV types (16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, 82). One HPV-positive control and two negative PCR controls (a purified DNA sample negative for HPV and a DNA free sample) were included in each PCR run and subsequent Reverse Line Blot (RLB). GP5+/GP6+ PCR-negative and RLB-negative samples were amplified for the β-globin gene sequence using GH20-PC04 primers (268-bp amplicon length) to assess DNA integrity (Bauer).

The molecular Laboratory of ISPO performed genotyping of samples collected in all Italian regions involved in the study except samples collected in Piedmont Region, which were genotyped at the Center for Experimental Research and Medical Science of the Turin University. ISPO also coordinated the molecular activities of the whole project, standardizing the procedures for storing and analysing the samples according to a protocol defined to guarantee the quality of molecular methods [16].

Confounding was assessed by a multivariate logistic regression approach; all variables with a p-value ≤0.15 in the univariate analysis were included in the multivariate model and retained in the final model according to a log-likelihood-ratio test for goodness-of-fit. Statistical analysis was performed using STATA 11.2 (Stata Corporation, College Station, Texas, USA).

Recruitment was performed in the period ranging from February 2008 to April 2009. The initial sample included 2,289 women; 239 did not meet the inclusion criteria (124 did not live in the selected LHU, 57 were pregnant, 45 were not included in the 18–26 years range of age and 13 did not speak Italian) and were replaced, and 290 could not be contacted. Thus 1,999 women were offered Pap test and HPV test. Among them, 897 (44.9%) declined participation for several issues (non interested in the study, practical constraints, health disorders, HPV-positivity, recent previous Pap-test, embarrassment, parental advice, safety concerns, no sexual activity) (Figure 1). Women who did not participate were slightly younger than participants (mean age 23.2 ± 0.09 vs. 23.9 ± 0.07 years; p < 0.001); the proportion of women with a lower education level (24.6% vs 19.9%; p = 0.017), living with parents (87.9 vs 74.6%; p = 0 < 001), unmarried (93.5 vs 89.5%; p = 0.003) and living in southern Italy (53.3% vs 38.0%; p < 0.001) was higher in the group of women who refused to participate.

In conclusion a total of 1,102 out of 2,289 women (48.1%) were enrolled in the study. In order to calculate the participation rate to the prevalence study, we considered more appropriate to exclude from the denominator (2,289) the 154 women (among the 897 women who declined participation) who spontaneously declared not to be sexually active: therefore the participation rate was 51.6% (1,102/2,135), with a range among LHUs of 23.9-80.7.

The mean age of participants was 23.8 ± 0.07 years; 40.7% lived in northern Italy, 37.8% in southern Italy, and 21.5% in central Italy. Most women were Italian (93.7%), unmarried (89.3%) and had a high educational level, defined as > 8 years of education (80.2%). Detailed socio-behavioural characteristics are shown in Table 1.

The Pap-test was performed on 1,099 cervical smears because three samples were missed (they were included in the CC screening programme, instead of being included in the PreGio study) and the HR-HPV testing was performed on 1,094 out of 1,999 samples (the material of five samples was not sufficient for HR-HC2 test). The proportion of HR-HC2 positive samples was 18.7% (205/1,094) and did not differ by geographic area and age group (Table 2).

The genotyping was performed on 202/205 HR-HPV positive samples (three samples were not adequate because material was not sufficient for genotyping) and 194 (96%) showed positivity at least for one of the 18 types identifiable by the kit assay used for genotyping. In contrast, eight samples (4%) did not result positive at any of these 18 types: this proportion is in line with data reported in literature [17, 18] and is mainly due to the cross-reactivity of the probe B with low-risk HPV types.

Among Hybrid Capture positive women HPV16 was the most prevalent type (30.9%), followed by 31 (19.6%), 66 (12.9%), 51 (11.3%), 18 (8.8%), 56 (8.8%) and 58 (8.3%); all the other types were detected at frequencies lower than 5.2% (Table 3); the cumulative rate of 16 and 18 was 38.7%.

The prevalence of HPV 26, 53, 66, 73, 82 is underestimated because these HPV types are detected by the “Consensus High Risk HPV genotyping” assay, but they are not targeted by HR-HC2 test; the detection of HPV 26, 53, 66, 73, 82 is due to an occasional cross-hybridization of HR-HC2 method or co-infection with HPV-targeted types, therefore they could more likely appear in multiple than single infections. For this reason we limited the analysis of multiple infections to the 13 HR-HPV types targeted by HR-HC2 test.

Out of 181 women infected by HR-HC2 targeted types, 37 (20.4%) had a multiple infection (36 infected by two types and one by three types). No difference in socio-demographic characteristics and sexual and reproductive history was found between women presenting a multiple and single infection. HPV16 was the most common genotype detected in multiple infections, followed by HPV31 and 18. Table 4 shows the distribution of HPV types in single and multiple infections; the proportion of single and co-infections by genotype is also reported: the proportion of infections occurring in conjunction with other types did not significantly differ by HPV type (Fisher’s exact test: p = 0.264).

The univariate analysis of socio-behavioural characteristics and HR-HPV infection is reported in Table 2. The risk of HR-HPV infection resulted significantly higher in women who declared a greater number of partners in lifetime (OR = 3.78, 95%CI: 2.40-5.94 if 2–4 partners and OR = 9.29, 95%CI: 5.54-15.57 if ≥5 partners) and in the last six months (OR = 2.96, 95%CI 1.73-5.07 if ≥2 partners), and in women who have used any contraceptive method in the last six months (OR = 1.70, 95%CI 1.14-2.53). In contrast, the risk of HR-HPV infection resulted significantly lower in women living with their partner compared to women living with parents/friends/alone (OR = 0.53, 95%CI: 0.33-0.86), in married women (OR = 0.30, 95%CI: 0.14-0.63) and women with children (OR = 0.43, 95%CI: 0.22-0.84). Also a higher age at the first intercourse showed a protective effect (OR = 0.92, 95%CI: 0.85-0.99).

The following variables were included in the multivariate model: marital status, “living with”, age at first sexual intercourse, age at first delivery, parity, number of lifetime partners, number of partners in the last six months, use of contraceptives in the last six months.

In the multivariate logistic regression model (Table 2), the effect of the number of lifetime sexual partner remained statistically significant (OR_adj = 4.15, 95%CI: 2.56-6.72 if 2–4 partners and OR_adj = 10.63, 95%CI: 6.16-18.36 if ≥5 partners). Also, living with the sexual partner and having used any contraceptive in the last six months remained significantly associated with the detection of HR-HPV infection (OR_adj = 0.56, 95%CI: 0.34-0.92 and OR_adj = 1.67, 95%CI: 1.09-2.54 respectively).

The pap-test result was available for 1,093/1,094 cervical smears tested for HR-HC2; 5.7% (62/1,093) of samples were inadequate, cytology was normal in 83.0% samples (907/1,093) and abnormal in 11.3% (124/1,093) samples: 62 ASCUS/AGUS (atypical squamous/glandular cervical cells of undetermined significance), 52 LSIL, 6 ASCH and 4 HSIL. The proportion of HR-HPV positivity was significantly higher in women with abnormal cytology (65/124, 52.4%) than in women with normal cytology (132/907, 14.6%) (p < 0,001). Among the 197 HPV-infected women with adequate cytology, 65 (33.0%) had an abnormal finding.

HPV16 was the most frequent virus strain in normal cytology, ASCUS/AGUS, LSIL and ASCH, detected in 29.3, 47.8, 24.2 and 66.7% of the samples respectively. All the four HSIL were HR-HPV positive and four different types were detected respectively (16, 18, 31, 33).