Human papillomavirus DNA and mRNA prevalence and association with cervical cytological abnormalities in the Irish HIV population

Abstract

The complex interplay between HIV and human papillomavirus and its link to cervical dysplasia is poorly understood. This is the first study to assess the prevalence of oncogenic human papillomavirus mRNA in HIV-positive women, its relationship to HIV and its potential use in the triage of cervical cancer screening in HIV-positive women. In this cross-sectional study, we included 321 HIV-positive women. In all, 28.7% had abnormal cervical cytology, 51.1% were human papillomavirus DNA-positive and 21.8% tested positive for human papillomavirus mRNA. Women with a CD4 count of <200 × 10⁶/L were more likely to test positive for human papillomavirus DNA and mRNA. Virally suppressed women were less likely to be human papillomavirus DNA-positive; however, the same did not hold true for human papillomavirus mRNA. We found the human papillomavirus mRNA screening to be more specific when screening for low-grade squamous intraepithelial lesion and high-grade squamous intraepithelial lesion than human papillomavirus DNA at 84.53% compared to 57.36%. However, the sensitivity was less at 51.59% versus 91.07% for human papillomavirus DNA. It may be possible in the future to use human papillomavirus mRNA/DNA testing within a triage algorithm for the screening and management of cervical cancer in the HIV-positive patient.

Keywords

HIV AIDS sexually transmitted infection human papillomavirus HPV CIN cervical dysplasia women Europe

Background

It is well established that HIV-positive women are at an increased risk of developing cervical intraepithelial neoplasia (CIN) and cervical cancer when compared with the general population, and for this reason current guidelines recommend yearly cervical cancer screening in this population.^1–3 Human papillomavirus (HPV) has been associated with over 90% of cervical carcinoma.⁴ There are over one hundred different subtypes of HPV; each has a different oncogenic potential.⁵ The oncogenic potential of high-risk HPV genotypes depends on the unregulated expression of the E6 and E7 genes.^6,7 This expression, detected as HPV E6, E7 mRNA, is more prevalent in cervical cancer lesions and its precursors.⁸

Worldwide, HPV prevalence rates in HIV-positive women from Africa and South/Central America are higher when compared to Europe and North America.⁹ This meta-analysis of several HPV studies showed that types 16, 58, 18, 52, 31, 33 and 45 were the most common types found in HIV-positive populations.⁹ Several studies suggest that HPV mRNA testing is more specific than HPV DNA testing for the detection of high-grade squamous intraepithelial lesion (HSIL) in the general population.^10–12 No study to date has looked at using HPV DNA and the expression of E6/E7 HPV mRNA in the HIV population and its utility in cervical cancer screening. For this to be a useful adjunct to Papanicolaou smears, we need first to assess the prevalence of and interplay between HPV and HIV.

Here we present the first study to describe the prevalence of HPV mRNA in an HIV-positive population and its relationship to cervical abnormalities, CD4 count, HIV viral load and age.

Methods

Patients and samples recruitment

Between January 2007 and August 2010, all HIV-positive female patients attending routine HIV clinic appointments in St James’s Hospital, Dublin, Ireland, were invited to participate in this study. Inclusion criterion was: all HIV-positive female patients eligible for a cervical smear who were over 18 years. Written consent was obtained, and patients were interviewed by an advanced nurse practitioner, who filled out a questionnaire with the patient. Clinically-relevant data, age, ethnicity, smoking, contraception, parity, viral load and CD4 cell count were collected and a database of information was maintained.

PreservCyt™ smear samples were collected at baseline and at follow-up as per the Irish Cervical Screening Programme guidelines. Smear samples were processed according to The British Society of Colposcopy and Cervical Pathology guidelines.

Following cytological diagnosis (Table 1), the specimens were analysed for HPV DNA and integrated HPV mRNA.

Table 1.

HPV DNA and HPV mRNA positivity across different cytological disease categories.

Cytology result	HPV DNA (%)	HPV mRNA (%)	Total (%)
Normal	88 (38.4)	32 (14.0)	229 (71.3)
Total abnormal	76 (82.6)	38 (41.3)	92 (28.7)
ASCUS	25 (69.4)	9 (25.0)	36 (11.2)
LSIL	37 (88.1)	18 (42.9)	42 (13.1)
HSIL	14 (100.0)	11 (78.6)	14 (4.3)

ASCUS: atypical squamous cells of undetermined significance; LSIL: low-grade squamous intraepithelial lesion; LSIL: high-grade squamous intraepithelial lesion.

HPV DNA testing

HPV DNA testing was carried out using the Hybrid Capture 2 assay™ (Qiagen Ltd.,) as per the manufacturer’s instructions. Four millilitres of sample were used for DNA extraction by alkali denaturation at 65°C ± 5 for 45 min, and subsequently tested with a cocktail of RNA probes for the detection of 13 HR-HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68). A cut-off point of 1.0 relative light units/cut-off value (RLU/CO) was set as per the manufacturer’s instructions. Samples with RLU/CO of 1.0–2.5 were re-tested where possible and those with an RLU of >2.5 were considered positive. Re-tested samples were considered positive if they had an RLU/CO of >1.0. In the event of a negative re-test (RLU/CO < 1.0), the sample was tested a third and final time and considered positive if it had an RLU/CO of >1.0 and negative if it had an RLU/CO <1.0.

HPV mRNA testing

An aliquot of 5 mL PreservCyt was processed for total nucleic acid extraction using the Qiagen M48 BioRobot™ extraction method (Qiagen Ltd., UK). Cell lysis was performed prior to BioRobot™ extraction. Briefly, cells were centrifuged for 12 min @ 1 130 × g, and washed in 1 mL 100% ethanol. Qiagen lysis buffer (Buffer RLT: 400 µL) was added to the cell pellet, and the sample was vortexed for 1 min. Samples were transferred to the M48 BioRobot for nucleic acid extraction using the ‘Custom NorChip’ programme in the Qiasoft™ software and the MagAttract™ RNA Cell Mini M48 kit (Qiagen Ltd., UK). Sample volume was set to 400 µL, and elution volume was set to 50 µL. HPV mRNA testing was carried out using PreTect™ HPV-Proofer assay (Norchip AS) in accordance with the manufacturer’s instructions. This is a qualitative assay for the detection of HPV E6/E7 oncogenic mRNA from HPV types 16, 18, 31, 33 and 45. The assay also detects mRNA from the human U1 small nuclear ribonucleoprotein-specific protein A (U1A) to monitor sample mRNA integrity (control). The kit uses real-time Nucleic Acid Sequence Based Amplification technology for nucleic acid amplification and molecular beacon probes for detection of amplified mRNA.

Statistical analyses

Pearson’s Chi Square test was used to determine statistical significance for the association between HPV DNA/mRNA and clinical/demographic factors. McNemar’s test was carried out to determine concordance of HPV tests and multiple logistic regression was carried out to determine predictors of HPV infection. All statistical analyses were carried out using Statistical Package for the Social Sciences (SPSS) software version 16.0.

Ethical approval

Written informed consent was obtained from all patients and ethics approval was sought and obtained from the SJH/AMNCH Ethics Committee.

Results

There were 321 women included in this study; none were HPV vaccinated. The age range at inclusion was 17 to 71 years with a median of 34.6 years. The majority, 71.3% (229/321), had normal cervical cytology with 28.7% (92/321) having an abnormal cytological diagnosis at baseline. Most abnormal smears were in the low-grade squamous intraepithelial lesion (LSIL) category (Table 1).

The prevalence of HPV DNA in this population was found to be 51.1% (164/321), with an HPV mRNA prevalence of 21.8% (70/321). The mean ages of women infected with HPV DNA and HPV mRNA were 33.9 years and 34.9 years, respectively. HPV DNA prevalence peaked in the <25 years age group at 64.5% (20/31). Subsequent age groups displayed a falling trend, with a second rise occurring at age >45 years, 51.4% (18/35) (Figure 1). There was no significant difference in the HPV DNA prevalence between women <30 years (58.6% [51/87]) and women ≥30 years (48.3% [113/234]), p = 0.104.

Figure 1.

HPV DNA and HPV mRNA prevalence by age group.

HPV mRNA prevalence was high across all age groups, with a peak of 31.4% (11/35) in >45 years age group. There was no significant difference between HPV mRNA prevalence between women <30 years (25.3% [22/87]) and women ≥30 years (20.5% [48/234]), p = 0.365. Overall, 19.6% (63/321) of women were positive for both HPV DNA and mRNA. In all, 31.5% (101/321) tested positive for HPV DNA only with a further 2.2% (7/321) testing positive for HPV mRNA only.

The majority of women with abnormal cervical cytology tested positive for HPV DNA, 82.6% (76/92), with 41.3% (38/92) also testing positive for HPV mRNA. All 14 HSIL cases tested positive for HPV DNA, with 11/14 (78.6%), also testing positive for HPV mRNA (Table 1).

On logistical regression analysis, the only significant predictor of HPV DNA positivity was an abnormal cytological diagnosis. An inverse relationship was found between the years since HIV diagnosis and HPV positivity (Table 2). LSIL and HSIL cytological status were also the only predictors of HPV mRNA positivity (Table 3).

Table 2.

Logistic regression analysis, predictors of HPV DNA positivity.

	Odds ratio (95% CI)
Normal cervical cytology	1.0
ASCUS	5.0 (2.0–12.4)
LSIL	10.7 (3.7–30.8)
HSIL	2.2 (1.6–10.3)
<5 years since HIV-positive diagnosis	1.0
5–9 years since HIV diagnosis	0.5 (0.3–0.8)
10–19 years since HIV diagnosis	0.5 (0.2–1.2)
>20 years since HIV diagnosis	0.1 (0.2–0.5)
Non-smoker	1.0
Ex-smoker	0.3 (0.1–0.9)
Current smoker	1.5 (0.8–2.7)

ASCUS: atypical squamous cells of undetermined significance; LSIL: low-grade squamous intraepithelial lesion; LSIL: high-grade squamous intraepithelial lesion.

Table 3.

Logistic regression analysis, predictors of HPV mRNA positivity.

	Odds ratio (95% CI)
Normal cervical cytology	1.0
ASCUS	2.1 (0.9–4.8)
LSIL	4.6 (2.3–9.5)
HSIL	22.6 (6.0–85.4)

ASCUS: atypical squamous cells of undetermined significance; LSIL: low-grade squamous intraepithelial lesion; LSIL: high-grade squamous intraepithelial lesion.

Data on smoking habits were available for 94.7% (304/321) of patients. In all, 36.2% (110/304) were current smokers, and 63.8% (194/304) were non-smokers. The HPV DNA and mRNA prevalence in the non-smoking cohort was significantly lower than the smoking cohort (χ²= 6.693, p = 0.001) (χ²= 4.247, p < 0.05).

Data on ethnic origin were available for 310 women; 39.7% (123/310) were European, 58.7% (182/310) were African, 1% (3/310) were Asian, 0.3% (1/310) was South American and 0.3% (1/310) was North American. There was no significant difference between ethnic groups and HPV status.

HPV DNA detection was significantly higher in women who were recently diagnosed HIV positive (64.1% [93/145], [<5 years]) than those diagnosed with HIV for more than five years (χ²= 18.579, p < 0.001); the same did not hold true for HPV mRNA positivity.

There was a significant difference between HPV DNA prevalence but not mRNA prevalence in parous, 72% (36/50), and nulliparous, 49.6% (129/260) women (χ²= 8.440, p < 0.05). No significant difference between HPV status and reported contraception use was found.

Data on CD4 counts were available on 319/321 patients and HIV viral load on 317/321 patients. In all, 13.8% (44/319) of women had CD4 counts <200 × 10⁶/L with 51.4% (163/317) having an HIV viral load <50 copies/mL. The HPV DNA prevalence was significantly higher in women with lower CD4 counts (<200 × 10⁶/L) when compared with the prevalence in women with high CD4 counts (>200 × 10⁶/L), 72.7% (32/44) versus 48% (132/275) (χ²= 9.284, p < 0.005), respectively. This trend was also observed for HPV mRNA data with 40.9% (18/44) of women with low CD4 counts positive for HPV mRNA compared to 18.9% (52/275) of women with CD4 > 200 × 10⁶/L (χ²= 10.718 p = 0.001). In all, 52.3% (23/44) of women with CD4 counts <200 × 10⁶/L had abnormal cytology, significantly higher than the abnormality rate found in women with CD4 counts >200 × 10⁶/L (25.1% [69/275]) (χ²= 13.656, p < 0.001) (Table 4).

Table 4.

HPV DNA, mRNA and cytology data in relation to CD4 cell count.

	CD4 < 200 × 10⁶/L		CD4 > 200 × 10⁶/L
	HPV DNA+ (%)	HPV mRNA + (%)	HPV DNA + (%)	HPV mRNA + (%)
Normal cytology	13/21 (61.9)	6/21 (28.6)	75/206 (36.4)	26/206 (12.6)
Abnormal cytology	19/23 (82.6)	12/23 (52.2)	57/69 (82.6)	26/69 (37.7)
Total	32/44 (72.7)*	18/44 (41)**	132/275 (48)*	52/275 (18.9)**

*p value = (p < 0.005, χ²= 9.284).

**p value = (p < 0.001, χ²= 10.718).

HPV DNA prevalence in virally-suppressed women was significantly lower, 45.4% (74/163), compared to women who were not virally suppressed 57.8%, (89/154) (χ²= 4.869, p < 0.05) (Table 5). However, there was no significant difference between mRNA rates and HIV viral load.

Table 5.

HPV DNA, mRNA and cytology data in relation to HIV viral load.

	HIV viral load <50 copies/mL		HIV viral load >50 copies/mL
	HPV DNA+ (%)	HPV mRNA + (%)	HPV DNA + (%)	HPV mRNA + (%)
Normal cytology	39/121 (32.2)	13/121 (10.7)	49/106 (46.2)	19/106 (17.9)
Abnormal cytology	35/42 (83.3)	19/42 (45.2)	40/48 (83.3)	18/48 (37.5)
Total	74/163 (45.4)*	32/163 (19.6)**	89/154 (57.8)*	37/154 (24)**

p value = (p < 0.05, χ²= 4.869).

p value = 0.4141.

Analysis of the distribution of different oncogenic HPV genotypes in this study cohort by PreTect™ HPV-Proofer for detection of oncogenic mRNA from five high-risk HPV genotypes found that HPV 45 (45.7%; 32/70) was the most prevalent HPV genotype detected. This was followed by HPV 33, 24.3% (17/70), HPV 16, 22.9% (16/70), HPV 18, 21.4% (15/70) and HPV 31, 12.9% (9/70). HPV mRNA indicative of the presence of multiple HPV genotypes was found in 24.3% (17/70) of the mRNA-positive samples. We found that the majority of women with HPV 45-positive samples had normal cytology (62.5% [20/32]), the other four genotypes detected were more likely to be associated with an abnormal cytological diagnosis (χ²= 11.246, p = 0.001).

Discussion

In this study, we have described for the first time, the HPV DNA and mRNA prevalence rates in an HIV-positive cohort of women residing in Ireland. Prevalence rates of HPV DNA and mRNA were 51.1% and 21.8%, respectively. Similar HPV DNA prevalence has been found in other HIV-positive population studies.^13–17 The prevalence of HPV DNA was increased in HIV-positive women in comparison to the HIV-negative population of Ireland, 19.8%.¹⁷ Correspondingly, as expected, the rate of cervical cytology abnormalities was higher in this study population compared to the general Irish population, 28.7% versus 11.1%.¹³

In the general population of Norway, the HPV mRNA prevalence was found to be 14.5% in women <30 years and 3% in women ≥30 years.^11,18

We found a direct correlation between immunosuppression, HPV infection and cervical abnormalities. Those who were virally suppressed had significantly less carriage of HPV DNA than those who were not. However, the same did not hold true for HPV mRNA, with no significant difference found; this may be due to the small sample size, with only 70 testing positive for HPV mRNA. Those who had CD4 counts >200 × 10⁶/L had significantly less carriage of HPV DNA, mRNA and less cervical cytological abnormalities compared to those with more profound immunosuppression with CD4 counts <200 × 106/L. Many previous studies have looked at the association of HPV DNA carriage and its relationship to treatment with anti-retroviral therapy (ART). Conflicting results have been found and no previous studies have looked at the relationship between HPV mRNA and ART. Minkoff et al.¹⁹ have shown that effective anti-retroviral treatment is associated with decreased HPV infection and SIL progression of cervical lesions, while others have shown little or no effect.⁵

We found no significant difference in HPV carriage across the age groups. This study does not, however, look at the persistence of HPV infection and therefore it is impossible to comment on this aspect. However, we can speculate that the uniform distribution of HPV across the age groups may be due to the decreased clearance of HPV infection given the interplay between HIV and HPV, leading to decreased immunity, delayed clearance of HPV and possibly reactivation of latent infections acquired earlier in life.

In our study, we determined the HPV mRNA prevalence of five of the seven most common HPV DNA types found in HIV-positive populations. HPV genotypes 16, 58, 18, 52, 31, 33 and 45 are the most prevalent in the HIV-positive populations worldwide.⁹ In our study, the most dominant oncogenic HPV mRNA was HPV 45, followed by HPV 33, 16, 18 and 31. The HPV mRNA type distribution is different to that reported in mRNA studies carried out in the general population in Norway, where HPV 16 was the most prevalent HPV mRNA type found, followed by: HPV 31, 33, 18 and finally HPV 45.^11,18 In the general Irish population, Keegan et al.¹⁷ found the most common types to be HPV 16 (20%) and 18 (12%) followed by HPV 66, 33, 53, 31 and 58. The differences in these cohorts may be due to the greater ethnic diversity in the HIV-positive cohort in Ireland compared to the general population in Ireland.

HPV 45 mRNA carriage was less likely to be associated with abnormal cervical cytology than the other less common HPV subtypes. However, the number of HPV mRNA-positive samples in the study as a whole is not large enough to make definitive statements regarding oncogenicity of individual HPV mRNA types.

In the general population, using HPV DNA as a triage tool in conjunction with routine cervical cytology has been shown to increase the sensitivity in detecting cervical cytological abnormalities; however, there is a loss of specificity due to the large number of transient HPV infections. Mayrand et al. reported the sensitivity of both tests used together as 100%, and the specificity as 92.5%.^20,21

A meta-analysis of HPV mRNA use, as a triage tool in the general population, has shown that the pooled sensitivity and specificity of HPV mRNA to triage ASCUS to detect underlying CIN3 or worse was 96.2% (95% CI: 91.7–98.3%) and 54.9% (95% CI: 43.5–65.9%), respectively.²² However, no large studies have fully validated its use in the general population.

In this HIV-positive population, the use of HPV mRNA increased the specificity of detecting LSIL and HSIL from 57.36% to 84.53% (95% CI: 79.60–88.66%); however, there was a reduction in sensitivity at 51.59% (95% CI: 38.03–65.34%).

Therefore, we can conclude that using HPV mRNA testing as an adjunct to cervical cytology screening in HIV-positive women there would be a higher negative predictive value than if HPV DNA was used. This is not the case in the general population and most likely reflects the higher rates of HPV carriage in this population. Using a more specific test may decrease the number of referrals to colposcopy, increasing efficiency in the service, decreasing patient worry and inconvenience. However, it is not sensitive enough to be used alone as its negative predictive value is 89.24% (95% CI: 84.74–92.79%), and therefore some patients with LSIL/HSIL would be missed.

Further studies to interrogate the potential usage of HPV mRNA as a triage tool in the HIV population are needed. One potential future use may be in the triage of those with atypical squamous cells of undetermined significance (ASCUS) to ascertain if they are at risk of progressing to HSIL; however, longitudinal data are needed to ensure that the long-term negative predictive value of this mRNA assay is similar to those of validated HPV DNA tests.

The utility of HPV mRNA testing for predicting CIN has been well documented in the general population, with data suggesting HPV mRNA is a more specific and less sensitive that HPV DNA testing and cytology.^10,11,23,24 This is the first study to elicit its potential use in the HIV-positive population. As we begin to understand interplay between HPV, HIV and cervical cytology, it may be possible in the future to use HPV mRNA/DNA testing within a triage algorithm for the screening and management of cervical cancer in HIV-positive patients. We have demonstrated that use of this technology in cervical screening needs to be specifically tailored in the HIV-positive population.

Footnotes

Acknowledgements

This study was undertaken as part of the CERVIVA research consortium (). We would like to acknowledge the Coombe Women and Infants University Hospital, St. James’ Hospital GUIDE clinic, and the women who participated in the study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article..

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Health Research Board, Ireland.

References

Frisch

Biggar

Goedert

. Human papillomavirus-associated cancers in patients with human immunodeficiency virus infection and acquired immunodeficiency syndrome. J Natl Cancer Inst 2000; 92: 1500–1510.

Palefsky

. Human papillomavirus-associated anogenital neoplasia and other solid tumors in human immunodeficiency virus-infected individuals. Curr Opin Oncol 1991; 3: 881–885.

De Vuyst

Lillo

Broutet

. HIV, human papillomavirus, and cervical neoplasia and cancer in the era of highly active antiretroviral therapy. Eur J Cancer Prev 2008; 17: 545–554.

Munoz

Bosch

de Sanjose

. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 2003; 348: 518–527.

Faridi

Zahra

Khan

. Oncogenic potential of Human Papillomavirus (HPV) and its relation with cervical cancer. Virol J 2011; 8: 269–269.

Cuschieri

Wentzensen

. Human papillomavirus mRNA and p16 detection as biomarkers for the improved diagnosis of cervical neoplasia. Cancer Epidemiol, Biomarkers Prev 2008; 17: 2536–2545.

Sotlar

Stubner

Diemer

. Detection of high-risk human papillomavirus E6 and E7 oncogene transcripts in cervical scrapes by nested RT-polymerase chain reaction. J Med Virol 2004; 74: 107–116.

Kraus

Molden

Holm

. Presence of E6 and E7 mRNA from human papillomavirus types 16, 18, 31, 33, and 45 in the majority of cervical carcinomas. J Clin Microbiol 2006; 44: 1310–1317.

Clifford

Goncalves

Franceschi

Hpv Group

HIVS

. Human papillomavirus types among women infected with HIV: a meta-analysis. AIDS 2006; 20: 2337–2344.

10.

Keegan

McInerney

Pilkington

. Comparison of HPV detection technologies: hybrid capture 2, PreTect HPV-Proofer and analysis of HPV DNA viral load in HPV16, HPV18 and HPV33 E6/E7 mRNA positive specimens. J Virol Methods 2009; 155: 61–66.

11.

Molden

Kraus

Karlsen

. Comparison of human papillomavirus messenger RNA and DNA detection: a cross-sectional study of 4,136 women >30 years of age with a 2-year follow-up of high-grade squamous intraepithelial lesion. Cancer Epidemiol Biomarkers Prev 2005; 14: 367–372.

12.

Molden

Nygard

Kraus

. Predicting CIN2+ when detecting HPV mRNA and DNA by PreTect HPV-proofer and consensus PCR: a 2-year follow-up of women with ASCUS or LSIL Pap smear. Int J Cancer 2005; 114: 973–976.

13.

Levi

Kleter

Quint

. High prevalence of human papillomavirus (HPV) infections and high frequency of multiple HPV genotypes in human immunodeficiency virus-infected women in Brazil. J Clin Microbiol 2002; 40: 3341–3345.

14.

Tornesello

Duraturo

Giorgi-Rossi

. Human papillomavirus (HPV) genotypes and HPV16 variants in human immunodeficiency virus-positive Italian women. J Gen Virol 2008; 89(Pt 6): 1380–1389.

15.

Luchters

Vanden Broeck

Chersich

. Association of HIV infection with distribution and viral load of HPV types in Kenya: a survey with 820 female sex workers. BMC Infect Dis 2010; 10: 18–18.

16.

Girianelli

Thuler

e Silva

. Prevalence of HPV infection among women covered by the family health program in the Baixada Fluminense, Rio de Janeiro, Brazil. Rev Bras Ginecol Obstet 2010; 32: 39–46.

17.

Keegan

Ryan

Malkin

. Human papillomavirus prevalence and genotypes in an opportunistically screened Irish female population. Br J Biomed Sci 2007; 64: 18–22.

18.

Molden

Kraus

Karlsen

. Human papillomavirus E6/E7 mRNA expression in women younger than 30 years of age. Gynecol Oncol 2006; 100: 95–100.

19.

Minkoff

Zhong

Burk

. Influence of adherent and effective antiretroviral therapy use on human papillomavirus infection and squamous intraepithelial lesions in human immunodeficiency virus-positive women. J Infect Dis 2010; 201: 681–690.

20.

Boulet

Horvath

Berghmans

. Human papillomavirus in cervical cancer screening: important role as biomarker. Cancer Epidemiol, Biomarkers Prev 2008; 17: 810–817.

21.

Mayrand

Duarte-Franco

Rodrigues

. Human papillomavirus DNA versus Papanicolaou screening tests for cervical cancer. N Engl J Med 2007; 357: 1579–1588.

22.

Arbyn

Roelens

Simoens

. Human papillomavirus testing versus repeat cytology for triage of minor cytological cervical lesions. Cochrane Database Syst Rev 2013; 3: CD008054–CD008054.

23.

Trope

Sjoborg

Eskild

. Performance of human papillomavirus DNA and mRNA testing strategies for women with and without cervical neoplasia. J Clin Microbiol 2009; 47: 2458–2464.

24.

Ratnam

Coutlee

Fontaine

. Clinical performance of the PreTect HPV-Proofer E6/E7 mRNA assay in comparison with that of the Hybrid Capture 2 test for identification of women at risk of cervical cancer. J Clin Microbiol 2010; 48: 2779–2785.