Abstract
Background
Persistent infection with a subset of human papillomavirus (HPV) genotypes can cause abnormal cytology and invasive cervical cancer. This study examines the circulating HPV genotype strains in a local population of the province of Alberta (a largely unvaccinated population) to establish baseline frequency of vaccine and non-vaccine genotypes causing abnormal cervical cytology.
Method
Remnant liquid-based cytology specimens from the Alberta Cervical Cancer Screening Program (March 2014–January 2016) were examined. Only specimens from women who had a cytology grading of atypical squamous cells of undetermined significance or higher were included. HPV genotype was determined for all samples, and results were stratified by demographics and cytology results.
Results
Forty-four unique HPV genotypes were identified from 4,794 samples. Of the 4,241 samples with a genotype identified, the most common genotypes were HPV 16, 18, 31, and 51, with 1,599 (37.7%), 441 (12.2%), 329 (7.8%), and 354 (8.4%), respectively. HPV9 vaccine genotypes made up 73.2% of these genotyped samples. Compared with specimens in which HPV9 vaccine genotypes were not detected, those with a genotype covered by the HPV9 vaccine were from younger women (33 [interquartile range {IQR] 28 to 42] y versus 40 [IQR 32 to 51] y; p < 0.00001).
Conclusions
The baseline distribution of HPV genotypes in this largely unvaccinated population indicates that the HPV9 vaccine provides good protection from high-risk HPV infections. Determining the frequency of genotypes causing abnormal cytology in this population post–vaccine implementation will be important to assess efficacy of vaccination and monitor for any potential genotype replacement.
Key words: genotype distribution, HPV genotypes, human papillomavirus (HPV)
Abstract
Historique
L’infection persistante par un sous-groupe de génotypes du virus du papillome humain (VPH) peut être responsable d’une cytologie anormale et d’un cancer invasif du col de l’utérus. La présente étude porte sur les souches du génotype du VPH en circulation dans une population locale de la province de l’Alberta (largement non vaccinée) pour établir la fréquence de base des génotypes vaccinaux et non vaccinaux responsables de cytologies du col de l’utérus anormales.
Méthodologie
Les chercheurs ont examiné les résidus de prélèvement de cytologie en milieu liquide du programme albertain de dépistage du cancer du col de l’utérus (de mars 2014 à janvier 2016). Seuls les prélèvements de femmes dont la cytologie était classée d’après la présence de cellules squameuses (ou malpighiennes) atypiques à caractère non déterminé ou plus graves ont été retenus. Les chercheurs ont déterminé le génotype du VPH de tous les prélèvements et stratifié les résultats selon les résultats démographiques et cytologiques.
Résultats
Les chercheurs ont retenu 44 génotypes uniques du VPH à partir de 4 794 prélèvements. Les VPH 16, 18, 31 et 51 étaient les principaux génotypes observés, correspondant à 1 599 (37,7 %), 441 (12,2 %), 329 (7,8 %) et 354 (8,4 %) cas, respectivement. Les génotypes du vaccin contre le VPH9 représentaient 73,2 % des prélèvements. Par rapport aux prélèvements dans lesquels les génotypes du vaccin contre le VPH9 n’avaient pas été décelés, ceux dont le génotype était couvert par le vaccin contre le VPH9 provenaient de femmes plus jeunes (33 ans [plage interquartile {PIQ} de 28 à 42] par rapport à 40 ans chez les autres [PIQ de 32 à 51]; p < 0,00001).
Conclusion
Selon la distribution de référence des génotypes du VPH dans cette population largement non vaccinée, le vaccin contre le VPH9 assure une bonne protection contre les infections à VPH à haut risque. Il sera important de déterminer la fréquence des génotypes responsables d’une cytologie anormale dans cette population après l’adoption du vaccin pour en évaluer l’efficacité et surveiller le remplacement potentiel des génotypes.
Mots-clés : distribution du génotype, génotypes du VPH, virus du papillome humain (VPH)
Introduction
Human papillomavirus (HPV) is the most common sexually transmitted infection worldwide (1). Estimates of new infections range from 1 to 5.5 million annually in the United States (2). An estimated 75% of sexually active Canadians will contract HPV at least once in their lifetime, leading to 1,550 diagnoses of cervical cancer and approximately 400 HPV-related deaths, based on 2017 population data (3).
Overall, three HPV vaccines have been by approved by Health Canada for use in Canada: HPV2 (Cervarix; HPV 16, 18) (4), HPV4 (Gardasil; HPV 6, 11, 16, 18) (4), and HPV9 (Gardasil; HPV 6, 11, 16, 18, 31, 33, 45, 52, 58) (5). Currently, the HPV2 and HPV9 vaccines are commercially available in Canada. In 2006, the quadrivalent vaccine was licensed in Canada for females aged 9–26 years and then expanded in 2011 to include females aged 9–45 years and males aged 9–26 years. In this same year, the bivalent vaccine was also approved for use in females aged 10–26 years. (6). The HPV9 vaccine has been approved for use in Canada since February 2015 and is currently the sole vaccine used in the province of Alberta for HPV vaccination. Currently, male and female grade 6 students in Alberta are offered the HPV9 vaccine, with a catch-up program in grade 9 (7), using a two-dose schedule (8), as are any other interested adults aged younger than 27 years. Literature from Alberta examining the time period from 2008 to 2014 has reported that 31.3% of females aged 9–26 years had received three doses of vaccine (6).
In Alberta, cervical cancer screening uses primary smear-based cytology graded using the Bethesda system (9). Low-grade abnormal cytology results (eg, atypical squamous cells of undetermined significance [ASCUS] and low-grade squamous intraepithelial lesion [LSIL]) are reflexed to a high-risk HPV DNA (hrHPVDNA) screening assay, and higher-grade cytology results (eg, atypical squamous cells in which HSIL cannot be excluded [ASC-H] and high-grade squamous intraepithelial lesion [HSIL]) are referred directly to colposcopy (10,11). Women with negative hrHPVDNA are referred for Pap test rescreening in 1 year’s time (12).
The purpose of this research was to determine the distribution of HPV genotypes among women with abnormal cervical screening results circulating in the Canadian province of Alberta shortly after provincial HPV vaccination programs were implemented. Remnant liquid-based cytology (LBC) samples of women with cervical screening results of ASCUS or greater were genotyped, and the frequency of HPV genotypes was determined for our cohort. Identification of baseline HPV distribution in our pre-vaccinated population will assist in appropriate vaccine selection for our population and allow for comparison with future post-vaccination studies.
Methods
HPV genotype frequencies were investigated through a retrospective cohort study in the Canadian province of Alberta. During the study period, the female population of Alberta was approximately 2 million. Between March 12, 2014, and January 27, 2016, remnant LBC samples of ASCUS or greater from women aged 17 to 91 years undergoing routine cytology screening through the Alberta Cervical Cancer Screening program were collected and shipped to the Alberta Public Health Laboratory (ProvLab).
Samples from women with ASCUS and the majority of women with LSIL were tested by means of a nucleic acid test (hrHPVDNA; Cobas 4800, Roche Molecular Systems, Pleasanton, California) as per routine provincial screening algorithms for detection of HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68. Women with ASC-H or HSIL and a small subset of women with LSIL were referred directly to colposcopy.
Extraction and genotyping
Viral nucleic acid was extracted using the MagMAX™-96 Viral RNA Isolation Kit (Ambion Life Technologies, Carlsbad, California). Nucleic acid extracts were genotyped at the National Microbiology Laboratory (NML; Winnipeg, Manitoba) using a previously described Luminex microsphere genotyping assay capable of detecting 46 HPV genotypes and up to 30 genotypes in a single sample (13). The previously published study used an EasyMag extraction system to isolate DNA before genotyping, which was substituted for with MagMAX extraction in the current study. Results from this study were not used for clinical decision making.
Database creation and statistical analysis
Demographic data including date of birth, unique personal identifier, postal code, and region of testing were accessed through the ProvLab’s laboratory information system (LIS). Smear-based cytology results were accessed through the Calgary Laboratory Services’ LIS and the DynaLIFE Medical Lab’s LIS and linked to the database using a unique provincial personal identifier. Cytology results were analyzed as four different groups: (1) ASCUS, (2) ASC-H, (3) LSIL, or (4) HSIL+ (HSIL or squamous cell carcinoma). Income quintile and residence area type (metro, urban, rural, or rural remote) were determined by postal code using average income and census data for 2011 and algorithms established by Alberta Health (14). HPV genotyping results were merged into the database using a unique identifier. High-risk genotypes were defined as HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, 73, and 82 (15). Low-risk genotypes were defined as HPV 6, 11, 40, 42, 43, 44, 54, 61, 70, 72, and 81 (15).
Data from the Alberta Ministry of Health Immunization and Adverse Reaction to Immunization repository (for publicly funded vaccines) and from the population-based Pharmaceutical Information Network (for privately purchased vaccines) were used to identify women who had received HPV vaccination. Vaccinations were categorized into pre– and post–abnormal Pap test on the basis of the date of the women’s initial HPV vaccine. Only those women with vaccinations dated before the Pap testing were classified as vaccinated in this analysis. Women with incomplete information on either vaccination or cytology dates were not included in the vaccination analysis.
Data were analyzed using STATA (Stata Corp LLP, College Station, Texas) and Microsoft Excel (Microsoft Corp, Redmond, Washington). All duplicate testing was removed; only the first recorded test (based on the date of specimen collection) for a patient was used in the analysis. Genotype frequencies were stratified by age, residence region, income quintile, and cytology result. Basic descriptive analysis including the Fisher exact test and χ2 tests were used to compare categorical variables, and the Mann–Whitney test was used to compare continuous variables.
Ethics
Ethics approval for this research was received from the University of Alberta Health Research Ethics Board. As per this approval, women were not informed of the additional retrospective genotyping results determined from this research.
Results
In total, 5,157 LBC samples were collected from women aged between 17 and 91 years. Three hundred sixty-three duplicate specimens were removed. Of the remaining 4,794 specimens, median age of the cases was 35 years (interquartile range [IQR] 30 to 45 y), and 66.2% were from one of the two major cities in Alberta (Edmonton and Calgary). The greatest proportion of cytology results were ASCUS (40.2%) and ASC-H (30.9%) (Table 1).
Table 1:
Demographics of overall population and by cytology result for women with abnormal Pap test in Alberta (N = 4,794)
| Variable | No. (%)* | p-value | ||||
|---|---|---|---|---|---|---|
| Overall | ASCUS | ASC-H | LSIL | HSIL+ | ||
| N (%) | 4,794 (100) | 1,929 (40.2) | 1,482 (30.9) | 277 (5.8) | 1,106 (23.1) | |
| Median age, y (IQR) | 35 (30 to 45) | 38 (33 to 47) | 30 (25 to 37) | 54 (52 to 59) | 30 (26 to 39) | <0.0001 |
| Residence type† | ||||||
| Metro | 3,130 (66.2) | 1,291 (67.8) | 894 (60.8) | 208 (77.9) | 737 (67.7) | <0.0001 |
| Urban | 647 (13.7) | 253 (13.3) | 150 (13.8) | 32 (12.0) | 150 (13.8) | |
| Rural | 835 (17.7) | 318 (16.7) | 308 (21.0) | 25 (9.4) | 184 (16.9) | |
| Remote rural | 118 (2.5) | 42 (2.2) | 56 (3.8) | 2 (0.8) | 18 (1.7) | |
| Income quintile† | 0.26 | |||||
| Q1 (lowest) | 929 (19.7) | 389 (26.5) | 267 (18.3) | 58 (21.7) | 215 (19.8) | |
| Q2 | 939 (19.9) | 355 (18.7) | 328 (22.4) | 52 (19.5) | 204 (18.8) | |
| Q3 | 878 (18.6) | 355 (18.7) | 287 (19.6) | 44 (16.5) | 192 (17.7) | |
| Q4 | 969 (20.6) | 373 (19.7) | 284 (19.4) | 69 (25.8) | 243 (22.4) | |
| Q5 (highest) | 995 (21.1) | 425 (22.4) | 296 (20.3) | 44 (16.5) | 230 (21.2) | |
| HPV DNA‡ | <0.0001 | |||||
| Positive | 2,101 (43.8) | 1,834 (95.1) | 0 | 267 (96.4) | 0 | |
| Negative | 77 (1.6) | 77 (4.0) | 0 | 0 | 0 | |
| Not tested | 2,616 (54.6) | 18 (0.9) | 1,482 (100.0) | 10 (3.6) | 1,106 (100.0) | |
| Genotyping results§ | ||||||
| High risk¶ | 3,920 (81.8) | 1,451 (75.2) | 1,221 (82.4) | 227 (82.0) | 1,021 (92.3) | <0.0001 |
| HPV 16 or 18 | 1,893 (39.5) | 1,143 (59.3) | 600 (40.5) | 67 (24.2) | 638 (57.7) | <0.0001 |
| Negative | 553 (11.5) | 286 (14.8) | 194 (13.1) | 11 (4.0) | 62 (5.6) | <0.001 |
| Vaccine history† | 183 (3.6) | 51 (2.6) | 75 (5.1) | 2 (0.7) | 45 (4.1) | <0.0001 |
* Unless otherwise indicated
† ns may not total overall n due to missing data
‡ Routine testing performed as part of the provincial algorithm using Cobas 4800, Roche Molecular Systems; HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68 detected
§ Performed using NML Luminex assay
¶ High-risk genotypes defined as HPV-16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, 73 and 82
ASCUS = Atypical squamous cells of undetermined significance; ASC-H = Atypical squamous cells where HSIL cannot be excluded; LSIL = Low-grade squamous intraepithelial lesion; HSIL = High-grade squamous intraepithelial lesion; IQR = Interquartile range; HPV = Human papillomavirus; NML = National Microbiology Laboratory
History of HPV vaccination was found for 592 (12.4%) women; 173 (29.2%) were vaccinated before their abnormal Pap test, and 367 (62.0%) were vaccinated after their abnormal Pap test. Of these 592 women, 52 (8.7%) had missing data on either date of vaccination or date of cytology testing, which precluded them from any analysis examining vaccination status. Overall, 3.6% of women in our cohort had at least one dose of HPV vaccine before the abnormal cytology results used in this study (Table 1). The median age at initial vaccination for this group was 26 years (IQR 21 to 29), and the median time of the first dose was 3.4 years before the abnormal cytology result. In total, 56% had received three (or more) doses of the vaccine.
On the basis of cytology results, hrHPVDNA testing was performed on 45.4% (2,178) of the total 4,794 specimens as per the Alberta provincial testing algorithm. Of these, 96.5% (2,101) tested positive and 3.5% (77) tested negative. The remaining 54.6% of total specimens were not tested by hrHPVDNA screening and would have been referred directly to colposcopy. The NML Luminex assay detected genotypes in 88.6% (1,861) of the hrHPVDNA-positive samples, 26.0% (20) of the negative specimens, and 90.2% (2,360) of the samples not tested for hrHPVDNA per provincial protocols. High-risk genotypes were detected by the NML Luminex assay in 10% (2) of the samples that were originally tested negative by the hrHPVDNA screening assay (genotypes 31 and 58). Conversely, of the 553 specimens for which the NML Luminex assay identified no genotype, 240 (43.4%) were originally positive and 57 (10.3%) were negative on hrHPVDNA testing, and 256 (46.3%) were never tested by hrHPVDNA. Among the 4,241 specimens with a positive genotype result from the NML Luminex assay, 44 different HPV genotypes were identified (Table 2). HPV 16 was the most frequent HPV genotype (1,599; 37.7%), followed by HPV 31 (441; 10.4%), HPV 18 (354; 8.4%), and HPV 51 (329; 7.8%). Among the 2,360 specimens not undergoing routine hrHPVDNA screening, the most frequent genotypes were HPV 16 (1,196; 50.7%), HPV 31 (287; 12.2%), HPV 33 (188; 8.0%), and HPV 18 (182; 7.7%). In total, high-risk genotypes were found in 92.4% (3,920) of specimens with a genotype result, and low-risk genotypes were found in 13.9% (587). Specifically, genotypes associated with genital warts were found in 2.3% of cases (HPV 6, 83 [2.0%], and HPV 11, 18 [0.4%]). In comparing genotypes between women with and without vaccination, we found no statistically significant differences except for HPV 11, which was higher in the vaccinated group (1.9% versus 0.37%, p = 0.01)
Table 2:
Genotype frequencies and multiple infections among vaccinated and unvaccinated women with abnormal Pap tests in Alberta
| Genotype or no. of infections | No. (%) | ||
|---|---|---|---|
| Total; N = 4,241 | Vaccinated; n = 157 | Unvaccinated; n = 4,034 | |
| Genotype* | |||
| 6 | 83 (2.0) | 3 (1.9) | 77 (1.9) |
| 11† | 18 (0.4) | 3 (1.9) | 15 (0.4) |
| 16 | 1,599 (37.7) | 54 (34.4) | 1,524 (37.8) |
| 18 | 354 (8.4) | 9 (5.7) | 340 (8.4) |
| 26 | 8 (0.2) | 0 | 8 (0.2) |
| 30 | 42 (1.0) | 2 (1.3) | 40 (1.0) |
| 31 | 441 (10.4) | 16 (10.2) | 421 (10.4) |
| 32 | 15 (0.4) | 0 | 14 (0.4) |
| 33 | 267 (6.3) | 7 (4.5) | 253 (6.3) |
| 35 | 264 (6.2) | 15 (9.6) | 249 (6.2) |
| 39 | 199 (4.7) | 12 (7.6) | 187 (4.6) |
| 40 | 21 (0.5) | 0 | 21 (0.5) |
| 42 | 131 (3.1) | 8 (5.1) | 121 (3.0) |
| 43 | 28 (0.7) | 1 (0.6) | 27 (0.7) |
| 44 | 13 (0.3) | 0 | 13 (0.3) |
| 45 | 247 (5.8) | 7 (4.5) | 237 (5.9) |
| 51 | 329 (7.8) | 16 (10.2) | 306 (7.6) |
| 52 | 278 (6.6) | 11 (7.0) | 264 (6.5) |
| 53 | 74 (1.7) | 0 | 72 (1.8) |
| 54 | 93 (2.2) | 3 (1.9) | 88 (2.2) |
| 56 | 272 (6.4) | 13 (8.3) | 255 (6.3) |
| 58 | 267 (6.3) | 15 (9.6) | 250 (6.2) |
| 59 | 208 (4.9) | 12 (7.6) | 194 (4.8) |
| 61 | 2 (0.05) | 0 | 2 (0.05) |
| 62 | 101 (2.4) | 2 (1.3) | 97 (2.4) |
| 66 | 240 (5.7) | 8 (5.1) | 229 (5.7) |
| 67 | 110 (2.6) | 5 (3.2) | 104 (2.6) |
| 68 | 58 (1.4) | 2 (1.3) | 58 (1.4) |
| 69 | 14 (0.3) | 0 | 14 (0.4) |
| 70 | 118 (2.8) | 6 (3.8) | 111 (2.8) |
| 71 | 1 (0.02) | 0 | 1 (0.02) |
| 72 | 35 (0.8) | 1 (0.6) | 33 (0.8) |
| 73 | 85 (2.0) | 5 (3.2) | 77 (1.9) |
| 74 | 18 (0.4) | 0 | 18 (0.5) |
| 81 | 96 (2.3) | 5 (3.2) | 90 (2.2) |
| 82 | 44 (1.0) | 2 (1.3) | 41 (1.0) |
| 83 | 66 (1.6) | 1 (0.6) | 62 (1.5) |
| 84 | 37 (0.9) | 0 | 36 (0.9) |
| 85 | 4 (0.09) | 0 | 4 (0.10) |
| 86 | 2 (0.05) | 0 | 2 (0.05) |
| 87 | 11 (0.3) | 1 (0.6) | 10 (0.3) |
| 89 | 21 (0.5) | 0 | 20 (0.5) |
| 90 | 61 (1.4) | 3 (1.9) | 58 (1.4) |
| 91 | 9 (0.2) | 0 | 8 (0.2) |
| Multiple infections, no. of genotypes in individual specimen | |||
| 1 | 2,773 (65.4) | 94 (59.9) | 2,652 (65.7) |
| 2 | 1,000 (23.6) | 44 (28.0) | 942 (23.4) |
| 3 | 325 (7.7) | 13 (8.3) | 307 (7.6) |
| ≥4 | 143 (3.4) | 6 (3.8) | 133 (3.3) |
Note: Among those specimens with a positive genotyping result; excludes 553 (11.5%) specimens that had a negative genotyping result by NML Luminex genotyping. ns for vaccinated and unvaccinated women will not total table N because 50 women with missing information were not included in this analysis. Boldface indicates high-risk HPV genotype.
* Specimens may have multiple infections; therefore, the total proportion will not add up to 100%.
† Only statistically significant difference between genotypes for vaccinated and unvaccinated women is HPV 11 (p = 0.01 using Fisher exact test)
HPV = Human papillomavirus; NML = National Microbiology Laboratory
Infections with multiple HPV genotypes were observed in 1,468 (34.6%) samples typed by NML Luminex (Table 2). The highest number of multiple HPV genotypes detected from one sample was eight. High-risk genotypes were less likely than lower-risk genotypes to be found in women with multiple infections. In total, of the 3,920 women infected with high-risk genotypes, only 35.1% had multiple infections, whereas 86.7% of the 587 women with a low-risk infection had multiple infections (p <0.0001). Infection with HPV 16 or 18 was found in 33.8% of specimens with multiple infections; however, this was not significantly different from those with a single infection (35.3%; p = 0.32). There was a small trend of higher income quintiles being associated with multiple infections (p = 0.03); however, no other significant association with multiple infections was seen with any other variable examined (data not shown).
High-risk genotypes made up 92.4% of total samples tested by the NML Luminex with a genotype result. These samples were from significantly younger women (34 y [IQR 29 to 44) versus 42 years [IQR 33 to 52]; p <0.0001) and were less likely to be from a remote rural community (2.2% versus 5.0%; p = 0.0005) (Table 3). No differences were seen in income quintile or vaccine history. High-risk genotype specimens were more likely to not have undergone the initial hrHPVDNA screening (57.9% versus 28.4%; p <0.0001).
Table 3:
Correlates of high-risk HPV infection among women with abnormal Pap test results (N = 4,241)
| Variable | HPV genotype, no. (%)* | p-value | |
|---|---|---|---|
| High risk | Not high risk | ||
| Total | 3,920 (92.4) | 321 (7.6) | |
| Median age, y, (IQR) | 34 (29 to 44) | 42 (33 to 52) | <0.0001 |
| Residence type† | 0.01 | ||
| Metro | 2,557 (66.2) | 209 (65.5) | |
| Urban | 543 (14.1) | 39 (12.2) | |
| Rural | 677 (17.5) | 55 (17.2) | |
| Remote rural | 84 (2.2) | 16 (5.0) | |
| Income quintile† | 0.76 | ||
| Q1 (lowest) | 750 (19.5) | 67 (21.1) | |
| Q2 | 758 (19.7) | 57 (17.9) | |
| Q3 | 704 (18.3) | 65 (20.4) | |
| Q4 | 810 (21.1) | 63 (19.8) | |
| Q5 (highest) | 821 (21.3) | 66 (20.8) | |
| HPV DNA‡ | <0.0001 | ||
| Positive | 1,647 (42.0) | 214 (66.7) | |
| Negative | 4 (0.10) | 16 (5.0) | |
| Not tested§ | 2,269 (57.9) | 91 (28.4) | |
| Vaccination history† | 146 (3.8) | 11 (3.5) | 0.79 |
Note: High-risk genotypes are defined as HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, 73, and 82. Table includes values only among those with positive genotyping results by NML Luminex Assay.
* Unless otherwise indicated.
† ns may not total column n because of missing data.
‡ Performed using Cobas 4800, Roche Molecular Systems; high-risk genotypes detected are HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68.
§ All HSIL+ and ASC-H and a subset of LSIL cases were referred directly to colposcopy and were not tested for hrHPVDNA.
HPV = Human papillomavirus; IQR = Interquartile range; HSIL+ = High-grade squamous intraepithelial lesion or squamous cell carcinoma; ASC-H = Atypical squamous cells where HSIL cannot be excluded; LSIL = Low-grade squamous intraepithelial lesion; hrHPBDNA = High-risk HPV DNA; NML = National Microbiology Laboratory
HPV9 vaccine genotypes made up 73.2% of all samples and 88.9% of HSIL samples (Table 4). Compared with specimens without any HPV9 vaccine genotypes, specimens with a genotype covered by the HPV9 vaccine were associated with younger age (33 y [IQR 28 to 42] versus 40 y [IQR 32 to 51]; p <0.0001) and those with higher risk cytology results (eg, HSIL; 29.9% versus 10.2%, p <0.0001) (Table 4).
Table 4:
Correlates of HPV infection with HPV9 vaccine–covered genotypes among women with abnormal Pap test results (N = 4,241)
| Variable | Vaccine genotypes, no. (%) | p-value | |
|---|---|---|---|
| HPV9 | Non-HPV9 | ||
| Total | 3,102 (73.2) | 1,139 (26.8) | |
| Median age (IQR) | 33 (28 to 42) | 40 (32 to 51) | <0.0001 |
| Residence type* | 0.54 | ||
| Metro | 2,006 (65.7) | 760 (67.4) | |
| Urban | 436 (14.3) | 146 (13.0) | |
| Rural | 541 (17.7) | 191 (17.0) | |
| Remote rural | 70 (2.3) | 30 (2.7) | |
| Income quintile* | 0.45 | ||
| Q1 (lowest) | 584 (19.2) | 233 (20.8) | |
| Q2 | 604 (19.9) | 211 (18.8) | |
| Q3 | 556 (18.3) | 213 (19.0) | |
| Q4 | 630 (20.7) | 243 (21.7) | |
| Q5 (highest) | 665 (21.9) | 222 (19.8) | |
| HPV DNA† | |||
| Positive | 1,091 (35.2) | 770 (67.6) | <0.0001 |
| Negative | 3 (0.10) | 17 (1.5) | |
| Not tested‡ | 2,008 (64.8) | 352 (30.9) | |
| Cytology results | <0.001 | ||
| HSIL+ | 928 (29.9) | 116 (10.2) | |
| LSIL | 137 (4.4) | 129 (11.3) | |
| ASC-H | 1,062 (34.2) | 226 (19.8) | |
| ASCUS | 975 (31.4) | 668 (58.7) | |
| Vaccination history* | 108 (3.5) | 49 (4.4) | 0.21 |
Note: The HPV9 vaccine covers genotypes 6, 11, 16, 18, 31, 33, 45, 52, and 58; values include only those with positive genotyping results by NML Luminex Assay.
* ns may not total row n due to missing data.
† Performed using Cobas 4800, Roche Molecular Systems; detected HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68
‡ All HSIL+, ASC-H, and a subset of LSIL cases are referred directly to colposcopy and are not tested for hrHPVDNA.
HPV = Human papillomavirus; IQR = Interquartile range; HSIL = High-grade squamous intraepithelial lesion or squamous cell carcinoma; LSIL = Low-grade squamous intraepithelial lesion ASC-H = Atypical squamous cells where HSIL cannot be excluded; ASCUS = Atypical squamous cells of undetermined significance; hrHPBDNA = High-risk HPV DNA; NML = National Microbiology Laboratory
Discussion
This study describes the frequency of HPV genotypes found among abnormal cytology specimens early after the implementation of provincial vaccination programs and before cohorts of females vaccinated through school-based programs reached the age of routine cervical cancer screening. In our population, the most common genotypes associated with abnormal cervical cytology were HPV 16, 31, 18, and 51, which made up 59.6% of all detectable HPV. HPV genotypes that are part of the HPV9 vaccine made up 73.2% of HPV infection in our cohort. This suggests that use of the HPV9 vaccine has the potential to offer high protection from cervical abnormalities in our cohort, in addition to the other beneficial impacts of this vaccine on HPV-related cancers.
Prevalence of HPV genotypes has been shown to vary geographically and within localized populations. Local circulating genotype data can be useful to policy-makers and governments when deciding which populations to vaccinate, and they may help guide vaccine development in the future. In Canada, northern Indigenous populations have been shown to have higher rates of detectable HPV than non-Indigenous populations (9,16). In the United States, higher rates of HPV-associated cervical cancer are seen among Black and Hispanic women (17). Likewise, globally, the genotype distribution can vary within populations, with HPV 16 and HPV 18 generally being the most frequent genotypes (18,19). In our study population, high-risk genotypes made up the overwhelming majority of samples and were significantly associated with younger women (p < 0.0001) (20). Numerous studies have shown increased HPV infections among younger-aged women, and the association we found specifically between younger age and high-risk genotypes has also been reported elsewhere (21).
HPV 51 was the most common non-vaccine genotype detected in our population, accounting for 329 (7.8%) cases of abnormal cytology. In other parts of the world, HPV 51 has been shown to be highly associated with abnormal cytology and invasive cervical cancers (ICCs). In Italy, HPV 51was the second most prevalent genotype (after 16) in tissue biopsy samples with CIN 1 or greater grading (22), and in Korea HPV 51 accounted for nearly 6% of all abnormal cytologies (23). Bzhalava et al conducted a systematic review of the prevalence of mucosal and cutaneous HPV types; HPV 51 was found in low-grade lesions (9.4% and 8.1% of LSIL and CIN 1, respectively), although it was seen less in invasive cancers (1% of ICCs) (24). In a Mexican evaluation of cervical cytology samples, HPV 51 accounted for 42% of all detectable HPV (25). Given the significance of HPV 51 in many global populations, it may be appropriate to consider this genotype for inclusion in new vaccine compositions.
Routine assessments of circulating HPV genotypes may be helpful in identifying any potential genotype replacement occurring after vaccination, that is, non-vaccine genotypes filling niches made vacant from the lack of vaccine genotypes. For example, serotype replacement was seen after implementation of the pneumococcal conjugate vaccine with increases in the prevalence of non–vaccine-type strains (26,27). Although there is currently no conclusive evidence for the occurrence of HPV-type replacement (28,29), there have been some reports of increased rates of non-vaccine HPV genotypes after HPV vaccine implementation (30,31). Some specific genotypes, including HPV 51, have been highlighted as being at risk for type replacement (30–33), which suggests that these genotypes should be closely monitored over time as the prevalence of vaccinated individuals increases in the population. Given that HPV 51 is already a high-prevalence genotype in our study population, this genotype may be of particular interest in our population for future surveillance.
The proportion of vaccinated women in our cohort was small, and no significant differences were noted between those who had been vaccinated and those who had not. The lack of differences was most likely a result of the older age at vaccination in this group (average age 26 y) and therefore probable pre-vaccination HPV exposure and infection. However, as the provincial cohort of school-based vaccinated women ages (the oldest women in the vaccinated cohort were aged 23 y in 2020), the burden of abnormal cytology should decrease significantly, because 73.1% of the women in our study with ASCUS or greater also had an HPV genotype that is covered by the HPV9 vaccine. Screening algorithms for HPV may need to evolve as the composition of circulating HPV genotypes changes. It will be particularly important to continue monitoring circulating genotypes to inform the best vaccination and screening strategies for the population.
Overall, the concordance between hrHPVDNA (performed as part of the routine clinical care pathway) and the NML Luminex assay (performed as part of this study) was 83.1% when examining only genotypes shared by both assays. The hrHPVDNA assay missed 2 (0.12%) samples identified by the NML Luminex, whereas the NML Luminex either identified no genotype (240) or identified only non-shared genotypes (125) in 365 (17.4%) samples that were initially positive on the hrHPVDNA assay. Both the Roche Cobas 4800 and the NML Luminex assay have previously demonstrated high performance characteristics (13, 34); therefore, the reduction in the ability of the NML Luminex assay to detect genotypes initially identified by the hrHPVDNA assay may be due to sample shipping, storage, or age, because DNA may have degraded.
We were not able to report on genotype information for women with normal screening results, which therefore limits our ability to fully understand the genotype distribution in the screened population. We have, however, been able to describe the distribution of HPV genotypes causing abnormal cervical cytologies on a population level using a methodology that can be replicated to identify changes in genotypes over time as vaccination types, coverage, and policies change provincially.
Here we have provided a baseline of the circulating HPV genotypes causing abnormal cervical cytology in the Alberta population. The majority of high-grade cervical lesions were caused by genotypes included in the HPV9 vaccine; therefore, it will be important to continue with vaccine efforts using this vaccine and to examine genotype composition of both cases and the population at risk to guide further vaccination policy in the future.
Funding Statement
This research was supported in part by Alberta Health (RES0023577) and by a research grant from the Investigator Initiated Studies Program of Merck Canada Inc (SFR1431).
Ethics Approval:
The University of Alberta Health Research Ethics review board approved this study.
Informed Consent:
N/A
Funding:
This research was supported in part by Alberta Health (RES0023577) and by a research grant from the Investigator Initiated Studies Program of Merck Canada Inc (SFR1431). The opinions expressed in this article are those of the authors and do not necessarily represent those of Merck Canada Inc. or its affiliates or related companies.
Disclosures:
CLC received grant funding from both Merck and Alberta Health for this study and outside the submitted work.
Peer Review:
This manuscript has been peer reviewed.
Animal Studies:
N/A
References
- 1.Khan MJ, Castle PE, Lorincz AT, et al. The elevated 10-year risk of cervical precancer and cancer in women with human papillomavirus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. J Natl Cancer Inst. 2005; 97(14):1072–9. 10.1093/jnci/dji187. Medline: [DOI] [PubMed] [Google Scholar]
- 2.Burd EM. Human papillomavirus and cervical cancer. Clin Microbiol Rev. 2003; 16(1):1–17. 10.1128/CMR.16.1.1-17.2003. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Canadian Cancer Society's Advisory Committee on Cancer Statistics. Canadian Cancer Statistics 2017. Toronto: Canadian Cancer Society; 2017. [Google Scholar]
- 4.Brotherton JM. Human papillomavirus vaccination: where are we now? J Paediatr Child Health. 2014;50(12): 959–65. 10.1111/jpc.12627. Medline: [DOI] [PubMed] [Google Scholar]
- 5.Joura EA, Giuliano AR, Iversen OE, et al. A 9-valent HPV vaccine against infection and intraepithelial neoplasia in women. N Engl J Med. 2015;372(8): 711–23. 10.1056/NEJMoa1405044. Medline: [DOI] [PubMed] [Google Scholar]
- 6.Liu XC, Bell CA, Simmonds KA, Russell ML, Svenson LW. HPV vaccine utilization, Alberta 2008/09-2013/14 school year. BMC Infect Dis. 2016;16:15. 10.1186/s12879-016-1340-6. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Colucci R, Hryniuk W, Savage C. HPV vaccination programs in Canada, are we hitting the mark? Cancer Advocacy Coalition of Canada. Report Card on Cancer in Canada. 2008;11:7–10. [Google Scholar]
- 8.Alberta Health Services. Human papillomavirus vaccine (HPV-9). Edmonton: Alberta Health Services; 2018. [Google Scholar]
- 9.Jiang Y, Brassard P, Severini A, et al. The prevalence of human papillomavirus and its impact on cervical dysplasia in Northern Canada. Infect Agent Cancer. 2013;8(1):25. 10.1186/1750-9378-8-25. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wright TC, Stoler MH, Behrens CM, Apple R, Derion T, Wright TL. The ATHENA human papillomavirus study: design, methods, and baseline results. Am J Obstet Gynecol. 2012;206(1):46. e1-e11. 10.1016/j.ajog.2011.07.024. Medline: [DOI] [PubMed] [Google Scholar]
- 11.Cox JT, Castle PE, Behrens CM, et al. Comparison of cervical cancer screening strategies incorporating different combinations of cytology, HPV testing, and genotyping for HPV 16/18: results from the ATHENA HPV study. Am J Obstet Gynecol. 2013;208(3):184. e1-e11. 10.1016/j.ajog.2012.11.020. Medline: [DOI] [PubMed] [Google Scholar]
- 12.Toward Optimized Practice. Cervical cancer screening: clinical practice guidelines. Accelerating Change Transformation Team; 2016. https://actt.albertadoctors.org/CPGs/Lists/CPGDocumentList/Cervical-Cancer-Screening-CPG.pdf (May 5, 2021) [Google Scholar]
- 13.Zubach V, Smart G, Ratnam S, Severini A. Novel microsphere-based method for detection and typing of 46 mucosal human papillomavirus types. J Clin Microbiol. 2012;50(2):460–4. 10.1128/JCM.06090-11. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Passi A, Plitt S, Lai F, Simmonds K, Charlton C. The economic impact of prenatal varicella immunity among pregnant women in Alberta. Vaccine. 2016; 35(4):570–6. 10.1016/j.vaccine.2016.12.014. [DOI] [PubMed] [Google Scholar]
- 15.Muñoz N, Bosch FX, de Sanjosé S, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med. 2003;348(6): 518–27. 10.1056/NEJMoa021641. Medline: [DOI] [PubMed] [Google Scholar]
- 16.Severini A, Jiang Y, Brassard P, et al. Type-specific prevalence of human papillomavirus in women screened for cervical cancer in Labrador, Canada. Int J Circumpolar Health. 2013;72. 10.3402/ijch.v72i0.19743. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Viens LJ, Henley SJ, Watson M, et al. Human papillomavirus-associated cancers—United States, 2008-2012. MMWR Morb Mortal Wkly Rep. 2016; 65(26):661–6. 10.15585/mmwr.mm6526a1. Medline: [DOI] [PubMed] [Google Scholar]
- 18.Andersen B, Christensen BS, Christensen J, et al. HPV-prevalence in elderly women in Denmark. Gynecol Oncol. 2019;154(1):118–23. 10.1016/j.ygyno.2019.04.680. Medline: [DOI] [PubMed] [Google Scholar]
- 19.Moghoofei M, Keshavarz M, Ghorbani S, et al. Association between human papillomavirus infection and prostate cancer: a global systematic review and meta-analysis. Asia Pac J Clin Oncol. 2019;15(5):e59–e67. 10.1111/ajco.13124. Medline: [DOI] [PubMed] [Google Scholar]
- 20.Dempsey AF. Human papillomavirus: the usefulness of risk factors in determining who should get vaccinated. Rev Obstet Gynecol. 2008;1(3):122–8. [PMC free article] [PubMed] [Google Scholar]
- 21.Torres-Poveda K, Ruiz-Fraga I, Madrid-Marina V, Chavez M, Richardson V. High risk HPV infection prevalence and associated cofactors: a population-based study in female ISSSTE beneficiaries attending the HPV screening and early detection of cervical cancer program. BMC Cancer. 2019;19(1):1205. 10.1186/s12885-019-6388-4. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Piana A, Sotgiu G, Cocuzza C, et al. High HPV-51 prevalence in invasive cervical cancers: results of a pre-immunization survey in North Sardinia, Italy. PLoS One. 2013;8(5):e63395. 10.1371/journal.pone.0063395. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.So KA, Kim MJ, Lee KH, et al. The impact of high-risk HPV genotypes other than HPV 16/18 on the natural course of abnormal cervical cytology: a Korean HPV cohort study. Cancer Res Treat. 2016;48(4):1313–20. 10.4143/crt.2016.013. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bzhalava D, Guan P, Franceschi S, Dillner J, Clifford G. A systematic review of the prevalence of mucosal and cutaneous human papillomavirus types. Virology. 2013;445(1-2):224–31. 10.1016/j.virol.2013.07.015. Medline: [DOI] [PubMed] [Google Scholar]
- 25.Gallegos-Bolaños J, Rivera-Domínguez JA, Presno-Bernal JM, Cervantes-Villagrana RD. High prevalence of co-infection between human papillomavirus (HPV) 51 and 52 in Mexican population. BMC Cancer. 2017; 17(1):531. 10.1186/s12885-017-3519-7. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Demczuk WH, Martin I, Griffith A, et al. Serotype distribution of invasive Streptococcus pneumoniae in Canada after the introduction of the 13-valent pneumococcal conjugate vaccine, 2010-2012. Can J Microbiol. 2013;59(12):778–88. 10.1139/cjm-2013-0614. Medline: [DOI] [PubMed] [Google Scholar]
- 27.Weinberger DM, Malley R, Lipsitch M. Serotype replacement in disease after pneumococcal vaccination. Lancet. 2011;378(9807):1962–73. 10.1016/S0140-6736(10)62225-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Covert C, Ding L, Brown D, Franco EL, Bernstein DI, Kahn JA. Evidence for cross-protection but not type-replacement over the 11 years after human papillomavirus vaccine introduction. Hum Vaccin Immunother. 2019;15(7-8):1962–9. 10.1080/21645515.2018.1564438. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mesher D, Soldan K, Lehtinen M, et al. Population-level effects of human papillomavirus vaccination programs on infections with nonvaccine genotypes. Emerg Infect Dis. 2016;22(10):1732–40. 10.3201/eid2210.160675. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Merikukka M, Kaasila M, Namujju PB, et al. Differences in incidence and co-occurrence of vaccine and nonvaccine human papillomavirus types in Finnish population before human papillomavirus mass vaccination suggest competitive advantage for HPV33. Int J Cancer. 2011;128(5):1114–9. 10.1002/ijc.25675. Medline: [DOI] [PubMed] [Google Scholar]
- 31.Gray P, Luostarinen T, Vänskä S, et al. Occurrence of human papillomavirus (HPV) type replacement by sexual risk-taking behaviour group: post-hoc analysis of a community randomized clinical trial up to 9 years after vaccination (IV). Int J Cancer. 2019;145(3):785–96. 10.1002/ijc.32189. Medline: [DOI] [PubMed] [Google Scholar]
- 32.Gray P, Palmroth J, Luostarinen T, et al. Evaluation of HPV type-replacement in unvaccinated and vaccinated adolescent females—post-hoc analysis of a community-randomized clinical trial (II). Int J Cancer. 2018;142(12):2491–500. 10.1002/ijc.31281. Medline: [DOI] [PubMed] [Google Scholar]
- 33.Yang Z, Cuzick J, Hunt W, Wheeler C. Concurrence of multiple human papillomavirus infections in a large US population-based cohort. Am J Epidemiol. 2014;180(11):1066–75. 10.1093/aje/kwu267. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Cui M, Chan N, Liu M, et al. Clinical performance of Roche Cobas 4800 HPV test. J Clin Microbiol. 2014; 52(6): 2210–1. 10.1128/JCM.00883-14. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]