Laser capture microdissection shows HPV11 as both a causal and a coincidental infection in cervical cancer specimens with multiple HPV types

Disproportionate preponderance of HPV genotypes associated with anogenital warts among HIV-positive MSM

Background

In this study, we characterized the HPV genotype distribution in a population of 489 adults already positive for HPV DNA. The study population was divided into two groups: 244 HIV-positive (HIV+) men who have sex with men (MSM) undergoing routine anal screening for sexually transmitted diseases (STDs) and 245 women undergoing routine cervical cancer screening. Acknowledging the fact that women and MSM represent two independent circles of sexual practices, which are-largely-exclusive of each other, we were interested in determining if particular genotypes of human papillomavirus (HPV) disproportionately predominate in one of these circles compared to the other.

Results

HIV+ MSM are significantly more likely to be infected with multiple genotypes at a time, with an odds ratio (OR) of 9.30 (95% confidence interval [CI]: 3.91-22.1) and a p-value of <0.001. In addition, multivariable-adjusted logistic regression analysis showed that anal swab samples were significantly more likely to harbor lrHPV infections, with an OR of 6.67 (95% CI: 2.42-18.4) and a p-value of <0.001, in particular, HPV 6, with an OR of 8.92 (95% CI: 3.84-20.7) compared to cervical samples of screening women.

Conclusion

Given the significant impact of recurrent anogenital warts (AGWs) on quality of life and the accompanying predisposition to invasive anal cancer, our data underscore the critical need for HPV vaccination. This includes expanding vaccination eligibility to include both boys and adults within high-risk populations.

HPV-Negative Adenocarcinomas of the Uterine Cervix: From Molecular Characterization to Clinical Implications

Cervical cancer is the fourth most common cancer in women. It is the leading cause of female deaths in developing countries. Most of these cervical neoplasms are represented by squamous lesions. Cervical adenocarcinoma causes about a quarter of cervical cancers. In contrast to squamous lesions, cervical glandular disease is HPV-negative in about 15-20% of cases. HPV-negative cervical adenocarcinomas typically present in advanced stages at clinical evaluation, resulting in a poorer prognosis. The overall and disease-free survival of glandular lesions is lower than that of squamous lesions. Treatment options require definitive treatments, as fertility-sparing is not recommended. Moreover, the impact of HPV vaccination and primary HPV screening is likely to affect these lesions less; hence, the interest in this challenging topic for clinical practice. An updated review focusing on clinical and molecular characterization, prognostic factors, and therapeutic options may be helpful for properly managing such cervical lesions.

HPV42-associated Seborrhoeic Keratosis-like Lesion of the Cervix: First Reported Case With High-grade Morphology

Seborrheic keratosis-like lesion (SKLL) is an extremely rare, morphologically distinct lesion occurring in the cervix and vagina that differs histologically from usual squamous intraepithelial lesions in these sites, by bearing close resemblance to cutaneous seborrheic keratosis and lacking koilocytosis. Like many vulvar seborrheic keratoses, which are associated with low-risk human papillomavirus (HPV), an association between SKLL and low-risk HPV is suggested based on the identification of HPV42, regarded as a low-risk genotype, in 4 of 8 reported cases. We report a further HPV42-associated SKLL of the cervix which differs from the previously reported cases by the presence of high-grade morphology and block-type p16 immunoreactivity. This novel finding challenges the classification of HPV42 as a low-risk genotype and expands the reported morphologic spectrum of SKLL, suggesting that they may not always be clinically indolent.

Molecular Pathology of Human Papilloma Virus-Negative Cervical Cancers

Cervical cancer is the fourth most common cancer in women worldwide and is predominantly caused by infection with human papillomavirus (HPV). However, a small subset of cervical cancers tests negative for HPV, including true HPV-independent cancers and false-negative cases. True HPV-negative cancers appear to be more prevalent in certain pathological adenocarcinoma subtypes, such as gastric- and clear-cell-type adenocarcinomas. Moreover, HPV-negative cervical cancers have proven to be a biologically distinct tumor subset that follows a different pathogenetic pathway to HPV-associated cervical cancers. HPV-negative cervical cancers are often diagnosed at an advanced stage with a poor prognosis and are expected to persist in the post-HPV vaccination era; therefore, it is important to understand HPV-negative cancers. In this review, we provide a concise overview of the molecular pathology of HPV-negative cervical cancers, with a focus on their definitions, the potential causes of false-negative HPV tests, and the histology, genetic profiles, and pathogenesis of HPV-negative cancers.

A proposed new generation of evidence-based microsimulation models to inform global control of cervical cancer

Health decision models are the only available tools designed to consider the lifetime natural history of human papillomavirus (HPV) infection and pathogenesis of cervical cancer, and the estimated long-term impact of preventive interventions. Yet health decision modeling results are often considered a lesser form of scientific evidence due to the inherent needs to rely on imperfect data and make numerous assumptions and extrapolations regarding complex processes. We propose a new health decision modeling framework that de-emphasizes cytologic-colposcopic-histologic diagnoses due to their subjectivity and lack of reproducibility, relying instead on HPV type and duration of infection as the major determinants of subsequent transition probabilities. We posit that the new model health states (normal, carcinogenic HPV infection, precancer, cancer) and corollary transitions are universal, but that the probabilities of transitioning between states may vary by population. Evidence for this variability in host response to HPV infections can be inferred from HPV prevalence patterns in different regions across the lifespan, and might be linked to different average population levels of immunologic control of HPV infections. By prioritizing direct estimation of model transition probabilities from longitudinal data (and limiting reliance on model-fitting techniques that may propagate error when applied to multiple transitions), we aim to reduce the number of assumptions for greater transparency and reliability. We propose this new microsimulation model for critique and discussion, hoping to contribute to models that maximally inform efficient strategies towards global cervical cancer elimination.

Human Papillomavirus-Negative Cervical Cancer: A Comprehensive Review

Human papillomavirus (HPV) has been the leading cause of cervical cancer for over 25 years. Approximately 5.5–11% of all cervical cancers are reported to be HPV-negative, which can be attributed to truly negative and false-negative results. The truly HPV-negative cervical cancers are almost all cervical adenocarcinomas with unclear etiology. False HPV negativity can arise from histological misclassification, latent HPV infection, disruption of the targeting fragment, non-high risk HPV infection, and HPV testing methods. HPV-negative cervical cancers are often diagnosed at an advanced FIGO stage and have a poor prognosis; thus, the management of these cases requires greater attention.

Human Papillomavirus and Factors Associated with Recurrence in Sinonasal Inverted Papillomas from Poland and Spain

Sinonasal inverted papilloma (SNIP) is a benign but locally aggressive tumor that has a tendency for recurrence and malignant transformation. The role of human papillomavirus (HPV) in SNIP is controversial. To determine the HPV-DNA prevalence and type distribution in SNIP in two different geographic areas and assess the association between SNIP recurrence and HPV infection, as well as additional potential etiologic factors. Two retrospective cohorts of SNIP patients from Poland and Spain were evaluated. Demographic, tobacco/alcohol use, clinical, and follow-up data were collected. All samples were subject to histopathologic evaluation, DNA quality control, and HPV-DNA detection by PCR. HPV-DNA positive samples and a random sample of HPV-DNA negative cases were further subject to p16^INK4a analysis. Proportional-hazards models were used to evaluate the risk of recurrence by selected variables. Seventy-nine SNIP patients (46 from Spain diagnosed between 1995 and 2014, and 33 from Poland diagnosed between 2012 and 2017) were included in the study. HPV-DNA was detected in four patients (5.1%), two from each region, and all four were positive for the HPV11 subtype. Seventeen patients (21.5%) experienced recurrence, with a median time to recurrence of 14 months. No association was identified between lesional HPV-DNA positivity, toxic habits, Krouse stage, or malignant transformation and a higher risk of recurrence. The low prevalence of HPV-DNA in SNIPs suggests that HPV is not a main etiology for development of these lesions. With a lack of association between the evaluated factors and recurrence, further research with larger number of patients and additional biomarkers is warranted to further understand predisposing risk factors.

Distinct geographic clustering of oncogenic human papillomaviruses multiple infections in cervical cancers: Results from a worldwide cross-sectional study

Coinfections by multiple Human Papillomaviruses (HPVs) are observed in approximately 6-8% of invasive cervical cancer (ICC) cases worldwide. But neither the presence of persistent HPVs coinfections nor their etiological role in the development of ICC is well understood. Cervical HPVs coinfections have been observed randomly, mostly in women with preneoplastic lesions, and only few studies have globally analyzed ICC cases. Here we explored the HPVs multiple infection patterns in a large worldwide sample of cross-sectional ICC cases. Paraffin-embedded ICC biopsy samples were tested using stringent HPV genotyping. Logistic regression models were used to identify the most likely pairwise HPV types in multiple infections. Multivariate analysis was applied to detect significant HPV coinfection patterns beyond pairwise HPVs comparison. Among 8780 HPV DNA-positive ICC cases worldwide, 6.7% (N = 587) contained multiple HPVs. Pairwise analysis revealed that HPV16|74, HPV31|33, HPV31|44, HPV33|44 and HPV45|70 pairs were significantly more frequently found together in multiple infections compared to any other HPV type combination, which supports the occasional role of Alpha-10 LR-HPVs in cervical cancers. In contrast, HPV16|31, HPV16|45, HPV16|51 and HPV18|HPV45 pairs were significantly less frequently found together than with any other HPV pair combination. Multivariate analysis sustained the results and revealed for the first time a distinct coinfection pattern in African ICCs stemming from the clustering of oncogenic HPV51/35/18/52 coinfections in African women. We suggest that the differential geographic HPVs coinfections clustering observed might be compatible with a specific modulation of the natural history/oncogenic potential of particular HPVs multiple infections and warrant monitoring for post-vaccinated.

Laser Microdissection for Human Papillomavirus (HPV) Genotyping Attribution and Methylation Pattern Analyses of Squamous Intraepithelial Lesions

Human papillomavirus (HPV) is a nearly ubiquitous infectious organism. It is estimated that 80% of sexually active adults will be exposed to anogenital HPVs in their lifetime, and detection of multiple genotypes in an anogenital sample is common. Detection and genotyping of HPV is usually performed by DNA testing, and less frequently by mRNA testing. HPV genotype testing and characterization of DNA methylation patterns of HPV-related lesions can provide important biological, epidemiological, and potentially relevant clinical information in individuals and populations. The use of laser capture microdissection to isolate cells within a specific lesion allows for very precise molecular characterization and hence causal attribution. This chapter describes detailed protocols for the capture of lesion-specific tissue from formalin-fixed, paraffin-embedded (FFPE) biopsy tissue, and downstream DNA testing for lesion-specific HPV genotype and their methylation patterns.

Development and validation of a protocol for optimizing the use of paraffin blocks in molecular epidemiological studies: The example from the HPV-AHEAD study

Worldwide use of formalin-fixed paraffin-embedded blocks (FFPE) is extensive in diagnosis and research. Yet, there is a lack of optimized/standardized protocols to process the blocks and verify the quality and presence of the targeted tissue. In the context of an international study on head and neck cancer (HNC)-HPV-AHEAD, a standardized protocol for optimizing the use of FFPEs in molecular epidemiology was developed and validated. First, a protocol for sectioning the FFPE was developed to prevent cross-contamination and distributed between participating centers. Before processing blocks, all sectioning centers underwent a quality control to guarantee a satisfactory training process. The first and last sections of the FFPEs were used for histopathological assessment. A consensus histopathology evaluation form was developed by an international panel of pathologists and evaluated for four indicators in a pilot analysis in order to validate it: 1) presence/type of tumor tissue, 2) identification of other tissue components that could affect the molecular diagnosis and 3) quality of the tissue. No HPV DNA was found in sections from empty FFPE generated in any histology laboratories of HPV-AHEAD consortium and all centers passed quality assurance for processing after quality control. The pilot analysis to validate the histopathology form included 355 HNC cases. The form was filled by six pathologists and each case was randomly assigned to two of them. Most samples (86%) were considered satisfactory. Presence of >50% of invasive carcinoma was observed in all sections of 66% of cases. Substantial necrosis (>50%) was present in <2% of samples. The concordance for the indicators targeted to validate the histopathology form was very high (kappa > 0.85) between first and last sections and fair to high between pathologists (kappa/pabak 0.21-0.72). The protocol allowed to correctly process without signs of contamination all FFPE of the study. The histopathology evaluation of the cases assured the presence of the targeted tissue, identified the presence of other tissues that could disturb the molecular diagnosis and allowed the assessment of tissue quality.

Human Papillomavirus (HPV) Genotypes in Condylomas, Intraepithelial Neoplasia, and Invasive Carcinoma of the Penis Using Laser Capture Microdissection (LCM)-PCR: A Study of 191 Lesions in 43 Patients

Laser capture microdissection-polymerase chain reaction (LCM-PCR) supported by p16 was used for the first time to demonstrate human papillomavirus (HPV) DNA in histologically specific penile lesions, which were as follows: squamous hyperplasia (12 lesions, 10 patients), flat lesions (12 lesions, 5 patients), condylomas (26 lesions, 7 patients), penile intraepithelial neoplasia (PeIN) (115 lesions, 43 patients), and invasive squamous cell carcinomas (26 lesions, 26 patients). HPV was detected by whole-tissue section and LCM-PCR. LCM proved to be more precise than whole-tissue section in assigning individual genotypes to specific lesions. HPV was negative or very infrequent in squamous hyperplasia, differentiated PeIN, and low-grade keratinizing variants of carcinomas. HPV was strongly associated with condylomas, warty/basaloid PeIN, adjacent flat lesions, and warty/basaloid carcinomas. A single HPV genotype was found in each lesion. Some condylomas and flat lesions, especially those with atypia, were preferentially associated with high-risk HPV. Unlike invasive carcinoma, in which few genotypes of HPV were involved, there were 18 HPV genotypes in PeIN, usually HPV 16 in basaloid PeIN but marked HPV heterogeneity in warty PeIN (11 different genotypes). Variable and multiple HPV genotypes were found in multicentric PeIN, whereas unicentric PeIN was usually related to a single genotype. There was a correspondence among HPV genotypes in invasive and associated PeIN. p16 was positive in the majority of HPV-positive lesions except condylomas containing LR-HPV. p16 was usually negative in squamous hyperplasia, differentiated PeIN, and low-grade keratinizing variants of squamous cell carcinomas. In summary, we demonstrated that LCM-PCR was a superior research technique for investigating HPV genotypes in intraepithelial lesions. A significant finding was the heterogeneity of HPV genotypes in PeIN and the differential association of HPV genotypes with subtypes of PeIN. The presence of atypia and high-risk HPV in condylomas and adjacent flat lesions suggests a precursor role, and the correspondence of HPV genotypes in invasive carcinomas and associated PeIN indicates a causal relation. Data presented support the bimodal hypothesis of penile cancer carcinogenesis in HPV-driven and non-HPV-driven carcinomas and justify the current WHO pathologic classification of PeIN in special subtypes.

Looking beyond human papillomavirus (HPV) genotype 16 and 18: Defining HPV genotype distribution in cervical cancers in Australia prior to vaccination

Australia has implemented a high-coverage HPV vaccination program but has not, to date, established the distribution of HPV types that occur in cervical cancers in Australia. This information is important for determining the potential for cervical cancer prevention with both current and broader spectrum HPV vaccines. We analysed 847 cervical cancers diagnosed 2005 to 2015 in tertiary centres in the three most populous Australian states with resolution of specimens containing multiple HPV types using laser-capture microdissection. Archived FFPE tissue was reviewed by specialist pathologists, sandwich sectioned, and initially whole-tissue sections genotyped for HPV. Samples were first genotyped using SPF10-LiPA25 (version 1). Negative samples were screened with DNA ELISA kit HPV SPF10, followed by genotyping with SPF+ LiPA if ELISA positive. If still negative, samples were tested on a qPCR assay targeting the E6 region of HPV16, 18, 45 and 33. Of the 847 cancers (65.1% squamous, 28.7% adenocarcinoma, 4.3% adenosquamous, 2.0% other), 92.9% had HPV detected. Of the HPV-positive cancers, 607 of 787 (77.1%) contained HPV16 or 18, 125 of 787 (15.9%) contained HPV31/33/45/52 or 58, and 55 (7.0%) another HPV type. There was a strong correlation between HPV type and age, with younger women most likely to have HPV16/18 detected and least likely HPV negative. Our findings indicate that cervical cancers diagnosed in Australia more frequently contain HPV16/18 than in international series. This could be due to cervical screening in Australia increasing the proportion of adenocarcinomas, in which types 18 and 16 more strongly predominate, due to prevention of squamous cancers.

Current and Emerging Molecular Tests for Human Papillomavirus-Related Neoplasia in the Genomic Era

Laboratory tests have a key role in preventing human papillomavirus (HPV)-driven carcinomas and in guiding therapeutic interventions. An understanding of the virology, immunology, and carcinogenesis of HPV is essential for choosing appropriate diagnostic test modalities and developing new and even more effective cancer prevention strategies. HPV infects basal epithelial cells on multiple surfaces and induces carcinoma primarily in the cervix and the oropharynx. HPV types are stratified as high risk or low risk based on their carcinogenic potential. During oncogenesis, HPV interferes with cell cycle regulation and incites DNA damage responses that thwart apoptosis and enable mutations to accumulate. Such mutations are an adverse effect of innate and adaptive antiviral immune responses that up-regulate DNA-editing enzymes, with natural selection of cells having a chromosomally integrated viral genome lacking expression of viral proteins targeted by the immune system. Infected cancers share a similar mutation signature, reflecting the effect of apolipoprotein B mRNA-editing catalytic polypeptide enzyme DNA-editing enzymes. It is feasible that genomic tests for characteristic mutations or methylation signatures, along with tests for dysregulated HPV gene expression, add value in predicting behavior of premalignant lesions. Furthermore, these tumor markers in cell-free DNA of plasma or body fluids may one day assist in early detection or monitoring cancer burden during treatment.

The low-risk papillomaviruses

Human Papillomavirus (HPV) research has been dominated by the study of a subset of Alpha papillomaviruses that together cause almost 5% of human cancers worldwide, with the focus being on the two most prominent of these (HPV16 and 18). These viruses are referred to as 'high-risk' (hrHPV), to distinguish them from the over 200 prevalent HPV types that more commonly cause only benign epithelial lesions. The 'low-risk' (lrHPV) term used to describe this group belies their cumulative morbidity. Persistent laryngeal papillomas, which occur rarely in children and adults, require regular surgical de-bulking to allow breathing. Such infections are not curable, and despite being caused by HPV11 (a lrHPV) are associated with 1-3% risk of cancer progression if not resolved. Similarly, the ubiquitous Beta HPV types, which commonly cause asymptomatic infections at cutaneous sites, can sometimes cause debilitating papillomatosis with associated cancer risk. Recalcitrant genital warts, which affect 1 in 200 young adults in the general population, and even the ubiquitous common warts and verrucas that most of us at some time experience, cannot be reliably eradicated, with treatment strategies advancing little over the last 100 years. The review highlights molecular similarities between high and low-risk HPV types, and focuses on the different pathways that the two groups use to ensure persistent infection and adequate virus shedding from the epithelial surface. Understanding the normal patterns of viral gene expression that underlie lesion formation, and which also prevent loss of the infected basal cells in established lesions, are particularly important when considering new treatment options. Finally, the common requirement for deregulated viral gene expression and genome persistence in development of cancers, unites both high and low-risk HPV types, and when considered alongside viral protein functions, provides us with a working understanding of the mechanisms that underlie HPV-associated pathology.

Squamous intraepithelial lesions of the anal squamocolumnar junction: Histopathological classification and HPV genotyping

Background

Human papillomavirus (HPV)-related anal cancer lesions are often found adjacent to the squamocolumnar junction (SCJ). We have assessed the histopathology and associated HPV genotypes in anal SCJ lesions in surgically excised anal warts in HIV-negative and –positive patients.

Methods

Histopathology identified 47 squamous intraepithelial lesions (SILs) adjacent to the SCJ amongst a total of 145 cases of clinically diagnosed anal condylomata. The anal SCJ lesions were further analyzed with p16, CK7 and p63 immunohistochemistry and HPV genotyping.

Results

Sixteen (16/47) of the excised anal wart lesions contained HSIL; Three were HSIL and exclusively associated with oncogenic HPVs. A further thirteen (13/47) were mixed lesions. Of these eight were HSILs with LSIL and six were HSILs with papillary immature metaplasia (PIM); Ten of the mixed lesions were associated with one or more oncogenic HPVs, while three cases were exclusively associated with HPV6.

Conclusions

Clinically diagnosed anal warts cannot be assumed to be limited to low-grade lesions as anal warts of the SCJ often show heterogeneous lesions, with coexistence of LSIL, PIM, and HSIL. Lesions showing PIM, however, may mimic HSIL, because they are hypercellular, but lack the nuclear atypia and conspicuous mitotic activity of HSIL; and are p16 negative.

•

Anal warts of the SCJ are often heterogeneous; with LSIL, including PIM and HSIL.

•

Anal HSILs only with HPV6 may indicate that LR-HPVs have some role in oncogenesis.

•

Recognition of PIM is important given its potential for being confused with HSIL.

•

Clinical diagnosis of anal warts cannot be assumed to be limited to LSIL.

•

Microscopic examination of anal warts is encouraged with appropriate immunostains.

Global Genomic Diversity of Human Papillomavirus 11 Based on 433 Isolates and 78 Complete Genome Sequences

Human papillomavirus 11 (HPV11) is an etiological agent of anogenital warts and laryngeal papillomas and is included in the 4-valent and 9-valent prophylactic HPV vaccines. We established the largest collection of globally circulating HPV11 isolates to date and examined the genomic diversity of 433 isolates and 78 complete genomes (CGs) from six continents. The genomic variation within the 2,800-bp E5a-E5b-L1-upstream regulatory region was initially studied in 181/207 (87.4%) HPV11 isolates collected for this study. Of these, the CGs of 30 HPV11 variants containing unique single nucleotide polymorphisms (SNPs), indels (insertions or deletions), or amino acid changes were fully sequenced. A maximum likelihood tree based on the global alignment of 78 HPV11 CGs (30 CGs from our study and 48 CGs from GenBank) revealed two HPV11 lineages (lineages A and B) and four sublineages (sublineages A1, A2, A3, and A4). HPV11 (sub)lineage-specific SNPs within the CG were identified, as well as the 208-bp representative region for CG-based phylogenetic clustering within the partial E2 open reading frame and noncoding region 2. Globally, sublineage A2 was the most prevalent, followed by sublineages A1, A3, and A4 and lineage B.

IMPORTANCE

This collaborative international study defined the global heterogeneity of HPV11 and established the largest collection of globally circulating HPV11 genomic variants to date. Thirty novel complete HPV11 genomes were determined and submitted to the available sequence repositories. Global phylogenetic analysis revealed two HPV11 variant lineages and four sublineages. The HPV11 (sub)lineage-specific SNPs and the representative region identified within the partial genomic region E2/noncoding region 2 (NCR2) will enable the simpler identification and comparison of HPV11 variants worldwide. This study provides an important knowledge base for HPV11 for future studies in HPV epidemiology, evolution, pathogenicity, prevention, and molecular assay development.

Detection of alpha human papillomaviruses in archival formalin-fixed, paraffin-embedded (FFPE) tissue specimens

Formalin-fixed, paraffin-embedded (FFPE) tissue specimens stored in pathology departments worldwide are an invaluable source for diagnostic purposes when fresh clinical material is unavailable as well as for retrospective molecular and epidemiological studies, especially when dealing with rare clinical conditions for which prospective collection is not feasible. Accurate detection of HPV infection in these specimens is particularly challenging because nucleic acids are often degraded and therefore, not suitable for amplification of larger fragments of the viral genome or viral gene transcripts. This review provides a brief summary of molecular methods for detecting alpha-HPV DNA/RNA in FFPE tissue specimens. We specifically address the key procedural and environmental factors that have the greatest impact on the quality of nucleic acids extracted from FFPE tissue specimens, and describe some solutions that can be used to increase their integrity and/or amplifiability. Moreover, commonly used methods for HPV DNA/RNA detection in FFPE tissue specimens are presented and discussed, focusing on studies using polymerase chain reaction as an HPV detection method and published after 1999. Finally, we briefly summarize our 22 years of experience with HPV detection in FFPE tissue specimens.

HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis

BACKGROUND

We aimed to provide updated information about the global estimates of attributable fraction and type distribution of human papillomavirus (HPV) in head and neck squamous cell carcinomas by doing a systematic review and meta-analysis.

METHODS

We did a literature search on PubMed to identify studies that used PCR for detection of HPV DNA in head and neck squamous cell carcinomas with information about HPV genotype distribution. We included studies that tested 20 or more biopsies per cancer site and were published between July 15, 1990, and Feb 29, 2012. We collected information about sex, risk factors, HPV detection methods, and biomarkers of potentially HPV-induced carcinogenesis (E6/E7 mRNA and p16(INK4a)). If it was not possible to abstract the required information directly from the paper, we contacted the authors. We did a meta-analysis to produce pooled prevalence estimates including a meta-regression to explore sources of heterogeneity.

FINDINGS

148 studies were included, contributing data for 12 163 cases of head and neck squamous cell carcinoma from 44 countries. HPV DNA was detected in 3837 cases. HPV16 accounted for 82·2% (95% CI 77·7-86·4) of all HPV DNA positive cases. By cancer site, pooled HPV DNA prevalence estimates were 45·8% (95% CI 38·9-52·9) for oropharynx, 22·1% (16·4-28·3) for larynx (including hypopharynx), and 24·2% (18·7-30·2) for oral cavity. The percent positivity of p16(INK4a) positive cases in HPV-positive oropharyngeal cancer cases was 86·7% (95% CI 79·2-92·9) and of E6/E7 mRNA positive cases was 86·9% (73·2-96·8). The estimate of HPV attributable fraction in oropharyngeal cancer defined by expression of positive cases of E6/E7 mRNA was 39·8% and of p16(INK4a) was 39·7%. Of subsites, tonsils (53·9%, 95% CI 46·4-61·3) had the highest HPV DNA prevalence. HPV DNA prevalence varied significantly by anatomical site, geographic region, but not by sex or tobacco or alcohol consumption.

INTERPRETATION

The contribution of HPV prevalence in head and neck squamous cell carcinoma and in particular that of HPV16 in the oropharynx shows the potential benefit of prophylactic vaccines.

FUNDING

European Commission.

The distribution of high-risk human papillomaviruses is different in young and old patients with cervical cancer

Despite numerous human papillomavirus (HPV) frequency studies in women with cervical cancer (CC), little is known of HPV frequency trends according to patient age. In this work, we compare the mean age and frequency distribution by age of CC patients positive for different HPVs. This study included 462 CC patients. HPVs were detected by PCR and typed using DNA sequencing. A total of 456 patients (98.7%) were positive for HPV: 418 (90.5%) had single and 38 (8.2%) had double HPV infections. HPV16 (46.5%), HPV18 (10.4%), HPV45 (6.7%), and HPV31 (4.1%) were the most frequent viral types in single-infected patients. The mean ages of single-infected patients with HPV16 (49.2±13.3), HPV18 (47.9±12.2), HPV45 (47.9±11.7), or HPV39 (42.6±8.9) were significantly lower than the mean ages of patients singly (53.9±12.7; p<0.001, t-test) or doubly (55.4±12.7; p<0.05, t-test) infected with the remaining HPVs. Three different trends were identified: one for HPV16, another for HPVs18/45/39, and a third for the rest of HPVs. The frequency trend of HPV16 shows two peaks. The first (63.2%) was found in the youngest women (≤35 years), followed by a decreasing trend until the age of 55-60 years (31.1%). The second peak arose at 61-65 years (52.5%), followed by a decreasing trend. The trend for HPVs18/45/39 declined from the youngest (19.3%) to the oldest (>70 years; 12.8%) women. In contrast, the trend for the remaining HPVs increased from the youngest (15.8%) to the oldest (46.2%) women. Unlike other life-style factors, low-risk sexual behavior was associated with late onset of CC independent of low-oncogenic HPV types (p<0.05, Wald chi-square statistic). The data indicate that most CCs in young women depend on the presence of high-oncogenic HPVs. In contrast, almost half of CCs in older patients had low-oncogenic HPVs, suggesting they could depend on the presence of other factors.

HPV-negative carcinoma of the uterine cervix: a distinct type of cervical cancer with poor prognosis

OBJECTIVE

Using highly sensitive polymerase chain reaction (PCR) techniques, we reanalysed all cervical carcinomas (CCs) found to be human papillomavirus (HPV)-negative by Hybrid Capture 2 (HC2) to determine the prevalence of true HPV-negativity. We also evaluated the characteristics of the patients with tumours with confirmed HPV-negativity.

DESIGN

Observational study.

SETTING

Barcelona, Spain.

POPULATION

A cohort of 136 women with CC (32 adenocarcinomas, 104 squamous cell carcinomas) who had pre-treatment HC2 testing.

METHODS

All negative cases were reanalysed and genotyped for HPV using three PCR assays (SPF10, GP5+/6+ and E7-specific assay).

MAIN OUTCOME MEASURES

Percentage of confirmed HPV-negative and HPV-positive tumours. Clinicopathological features and disease-free survival (DFS) and overall survival (OS) of both groups.

RESULTS

Fourteen of 136 women (10.2%) were negative for HPV by HC2. After reanalysis by PCR-based techniques only 8/136 (5.8%) tumours were confirmed as HPV-negative, whereas in six cases different HPVs were identified [HPV-11, -16 (two tumours), -18, -45, and -68]. Confirmed HPV-negativity was more frequent in adenocarcinomas than in squamous cell carcinomas (5/32, 15.6% versus 3/104, 2.9%, respectively; P = 0.017). Patients with CCs with confirmed HPV-negativity had significantly worse DFS than women with HPV-positive tumours [51.9 months (95% CI 12.2-91.7 months) versus 109.9 months (95% CI 98.2-121.5 months); P = 0.010]. In the multivariate analysis HPV-negativity and International Federation of Gynecology and Obstetrics (FIGO) staging were associated with increased risk of progression and mortality.

CONCLUSIONS

An HC2-negative result is an uncommon finding in women with CC, but in almost half of these cases HPVs are identified by more sensitive techniques. CCs with confirmed HPV-negativity are more frequently adenocarcinomas, and seem to be associated with worse DFS.