Purpose of investigation: To establish the relationship between high-risk human papillomavirus (hrHPV) serotypes and the odds of developing cervical intraepithelial lesions. Material and Methods: Retrospective cohort study using altered cervical cytology results and hrHPV test obtained by PCR. Groups were divided according to low-grade squamous intraepithelial lesions (LSIL), high-grade squamous intraepithelial lesions (HSIL) and squamous cervical cancer (SCC) results, while hrHPV was divided according to polymerase chain reaction (PCR) results in “other HR-HPV, HPV-16, and HPV-18. Logistic regression analysis was performed to estimate the odds of each hrHPV serotype in developing squamous intraepithelial lesions or SCC. Results: A total of 252 samples were analysed, mean age was 42 ± 12.65 years. Prevalence of hrHPV infection was 60% (n=152). “Other HR-HPV” group was the most common infection in our population (39%); nonetheless, there were no increased odds for any HPV group and the risk for developing LSIL. The odds for HSIL were 4.58 and 7 for “other HR-HPV” and HPV-16, respectively. Odds for SCC were 11.29 and 7.35 for HPV-18 and HPV-16. There were no significant odds in HPV-18 for HSIL and “other HR-HPV” SCC. Conclusion: The present authors found a greater prevalence for “other HR-HPV” group in this population. Nonetheless, HPV-16 and HPV-18 contribute the highest probability of developing HSIL and SCC.
Cervical carcinoma is one of the leading causes of cancer among women in developing countries [1]. Worldwide, an estimated 400,000 cases of invasive cervical carcinoma are diagnosed annually [2]. In Mexico, cervical cancer (CC) shows the highest incidence among women in reproductive age and is the second leading cause of death in the same group [3]. Human papillomavirus (HPV) infection is recognised as the most important factor implicated in the pathogenesis of cervical cancer [4], about 99.7% of squamous cervical cancers (SCCs) underlies an HPV infection [5]. Papillomaviruses are double-stranded DNA viruses that constitute the papillomavirus genus of the papillomaviridae family [6]. There are more than 200 types of HPV, and approximately 30 to 40 HPV serotypes infecting the genital tract mucosa; of which 15 (HPV 16, 18, 31, 33, 35, 39,45, 51, 52, 56, 58, 59, 66, 68, and 73) are classified as high-risk HPVs serotypes (hrHPV) [7]. HPV-16 and -18 are responsible for about 70% of cervical cancer in women worldwide [8] and the persistence of HPV infection has demonstrated to play an important role in carcinogenesis [9], displaying the importance of HPV-DNA testing as a screening test. Meanwhile, studies analysing DNA testing for HR-HPV serotypes, found a test sensitivity ranging from 96% to 99%, compared to the low sensitivity of cervical smear itself [10,11]; this shows the importance of cotesting as a screening method for detection of cervical cancer.
The present study aims to determine the relationship between hrHPV serotypes and the risk of developing high-grade squamous lesions (HSIL) and SCC.
This retrospective cohort study was part of an open-population campaign aiming to reduce cervical cancer prevalence designed by Institute of Social Security and Services of State Workers (ISSSTE) in Mexico City. During a three-year period from December 2013 to October 2016, cervical cytology and HPV samples were obtained from all women attending Lopez Mateos hospital’s colposcopy centre. Eligibility criteria were patients with complete record and results for Pap smear, HPV testing. and age. Patients with abnormal results were followed according to The American Society for Colposcopy and Cervical Pathology (ASCCP) [12]. Results from Pap smear, HPV screening, and age were recorded for analysis. The study protocol was approved by the Hospital’s review board.
Cytology samples were obtained according to World’s Health Organization guidelines [13] using standard smearing of the specimen on a glass slide. Cytology report was made based on 2001 Bethesda system for cervical cytology [14]. HPV sampling was obtained using ThinPrep endocervical brush [15], and storage on PreservCyt solution component, that serves as an alternative transport medium for biological specimens [16].
HPV was analysed using Cobas 4800 HPV Test, an FDA-approved qualitative real-time in vitro test for the detection of HPV in patients specimens. The test uses DNA amplification using polymerase chain reaction (PCR) and nucleic acid hybridization. This test detects 14 high-risk (HR) HPV types in a single analysis. The test identifies HPV 16, HPV 18 and 12 more high-risk types (31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68) using a generic label called “other HR-HPV”. Cobas test uses primers to define a sequence of 200 nucleotides within the polymeric L1 region of the HPV genome. HPV primers employed in the master mix are designed to amplify the 14-targeted HPV types; a fluorescent oligonucleotide probe binds to polymorphic regions within the sequence defined by these primers. Real-time PCR amplification is used for the particularly linked regions. [17]
HPV and cytology were considered positive or negative and then were divided into three groups 1) “other HR-HPV”, 2) “HPV-16”, and 3) “HPV-18”. Type-specific cytology results were reported as low-grade squamous intraepithelial lesion (LSIL), HSIL, and SCC. Age was presented by mean, standard deviation and then divided by groups. HPV prevalence was described according to abnormal cytology results and age. Comparison between typespecific HPV and cytology results was made. Differences between groups were calculated by X2 or Fisher’s exact test for categorical variables and t-test for continues variables. Univariate logistic regression was performed for each LSIL, HSIL, and SCC outcomes in order to obtain specific OR for each HPV serotype. P-value < 0.05 was considered significant. Data was analysed using STATA v.14.1.
A total of 3,274 cervical and HPV samples were obtained, and 3,022 samples were excluded because cytology results were other than LSIL, HSIL. and SCC. A total of 252 samples were used for this analysis.
The mean age was 42 ± 12.65 years. hrHPV distribution according to age group is described in Figure 1. HPV infection occurred in 60% (n=152) of the samples, and according to each squamous cervical lesion, HPV was present in 48% (n=91) of the LSIL samples, a 98% (n=44) in HSIL group, and 100% (n=17) in the case of SCC. The most common infection in the present population was “other HR-HPV” group, infecting 46% (n=115) of the samples, followed by HPV-16 (17%, n=43) and HPV0-18 (6%, n=14). Co-infection occurred in 8% of cytology results. Description HPV test results according to squamous epithelial lesion are depicted in Table 1.
Figure 1.— HPV distribution according to age group.
| Characteristic | LSIL n=190 | HSIL n=45 | SCC n=17 | p-valueb |
|---|---|---|---|---|
| No HPV, n (%) | 99 (39) | 1 (0.4) | 0 | <0.001 |
| Other HR-HPVa, n (%) | 67 (27) | 25 (10) | 7 (3) | |
| HPV 16, n (%) | 10 (4) | 12 (5) | 5 (2) | |
| HPV 18, n (%) | 3 (1) | 2 (1) | 1 (0.4) | |
| HPV HR/16, n (%) | 7 (3) | 5 (2) | 0 | |
| HPV HR/18, n (%) | 4 (2) | 0 | 0 | |
| HPV 16/18, n (%) | 0 | 0 | 4 (2) |
a “Other HR-HPV” (31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68).b p-value calculated by Fisher’s exact test
Logistic regression analysis for LSIL did not show higher odds for any HPV group. While in the HSIL group, “other HR-HPV” odds were 4.58 (2-10; p < 0.001) and 7 (3-16.6; p < 0.001) for HPV16, while HPV-18 infection had no statistical significance. In the squamous cervical cancer group, odds were 7.35 (2.39-22.6; p < 0001) for HPV-16 and 11.29 (0.87-44.4; p = 0.001) for HPV-18 “other HR-HPV” infection showed no statistical significance. (Table 2)
| HPV Group | LSIL | HSIL | SCC | |||
|---|---|---|---|---|---|---|
| OR 95% CIb | p-value | OR 95% CIb | p-value | OR 95% CIb | p-value | |
| Other HR-HPVa | 0.22 (0.1-0.48) | < 0.001 | 4.58 (2-10) | < 0.001 | 1.54 (0.49-4.76) | 0.452 |
| HPV 16 | 0.08 (0.35-0.18 | < 0.001 | 7 (3-16.6) | < 0.001 | 7.35 (2.39-22.6) | < 0.001 |
| HPV 18 | 0.23 (0.06-0.83) | 0.025 | 0.79 (0.16-3.9) | 0.778 | 11.29 (2.87-44.41) | 0.001 |
a “Other HR-HPV” (31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68). b Odds ratio with 95% confidence interval.
HPV infection exhibits a contrasted prevalence according to continents, countries, and even among regions. In the present study, HPV infection was more prevalent between 30 and 59 years of age, similarly to age-distribution patterns between South-eastern Asia and Eastern Africa results [18]. In women with HPV infection and abnormal cytology, the present data depicted comparable results to those reported by Mu-Mu-Shwe et al. [19] who tested 1,771 women for HPV infection and cytology, reporting a 41.4% prevalence of HPV in patients with abnormal cytology result. Meanwhile, in Algeria, Hammouda et al. reported a 33.3% prevalence of HPV infection in women with abnormal cytology result [20], showing an even lower prevalence compared to the results.
Type-specific serotype studies have shown a higher prevalence for HPV-16 and HPV-18, being HPV-16 the most common infection among all cervical cytology results [21]. Iwasaki using the same qualitative assay for HPV testing found “other HR-HPV” group serotype to have the highest prevalence (28.08%) vs. HPV-16 (10.77%) and HPV-18 (2.0%). Meanwhile, the present results also found a higher prevalence for “other HR-HPV” group, similarly to other studies in this population [22].
In this investigation, the authors found that odds for HSIL were higher for HPV-16, and had no statistical significance for “other HR-HPV group”; this differs from other populations were “other HR-HPV” have higher odds for HSIL [23]. In case is squamous cervical lesions, the present authors found HPV-18 to present the highest odds for cervical cancer, unlike other studies that show HPV-16 as the most common infection in case of cervical neoplasia [24].
HPV infection shows a very different distribution and prevalence according to regions, continents, and countries. Overall prevalence according to the present data is very similar to global HPV prevalence, but distribution for type-specific group of infection shows important differences from other studies, being “other HR-HPV” serotypes the most common infection among women with normal and abnormal cytology result. Despite this, odds ratio for HSIL and SCC is higher for HPV-16 and HPV-18, reiterating HPV-16 and -18 as the most important factor for developing high-grade squamous cervical lesions and SCC. Heterogeneity of HPV results requires continued epidemiological studies by country and region; therefore, this investigation may contribute to the continued epidemiological mapping of HPV among regions.