G Model
ARTICLE IN PRESS
JVAC-15415; No. of Pages 7
Vaccine xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions Emma T. Callegari a,c,1 , Sepehr N. Tabrizi b,c,d,1 , Jan Pyman e,1 , Marion Saville f,1 , Alyssa M. Cornall b,c,1 , Julia M.L. Brotherton f,g,1 , Suzanne M. Garland b,c,d,∗,1 a
The University of Melbourne, Grattan Street, Parkville, Melbourne 3010, VIC, Australia Regional HPV Reference Laboratory, Department of Microbiology and Infectious Diseases, The Royal Women’s Hospital, Parkville, VIC, Australia c Murdoch Childrens Research Institute, Parkville, VIC, Australia d Department of Obstetrics and Gynaecology, University of Melbourne, Melbourne, VIC, Australia e Department of Anatomical Pathology, Royal Women’s Hospital, 5th Floor, 20 Flemington Road, Parkville, Melbourne 3052, VIC, Australia f VCS Incorporated, 265 Faraday St, Carlton, Melbourne 3053, VIC, Australia g School of Population and Global Health, University of Melbourne, Melbourne, VIC, Australia b
a r t i c l e
i n f o
Article history: Received 12 February 2014 Received in revised form 2 May 2014 Accepted 7 May 2014 Available online xxx Keywords: Human papillomavirus Cervical cancer CIN3 AIS Laser capture microdissection Multiple infections
a b s t r a c t Objectives: This study aimed to determine human papillomavirus (HPV) genotypes present in biopsy sections from young women of vaccine eligible age living in Victoria, Australia, with confirmed cervical intraepithelial neoplasia grade 3 (CIN3) or adenocarcinoma in situ (AIS) using laser capture microdissection (LCM). Methods: Histologically confirmed CIN3 or AIS positive biopsies from vaccine eligible women (born after 30th June 1981, n = 169), between May 2011 and March 2013, were identified. CIN3 or AIS lesions were isolated from biopsy material using LCM, and the HPV genotypes present in whole tissue sections (WTS) as well as LCM-isolated lesion tissue were determined by a sensitive reverse hybridisation assay; RHA kit HPV SPF10-LiPA25, version 1 (Labo Bio-medical Products, Rijswijk, The Netherlands). Results: One hundred and sixty-eight cases were shown to be HPV positive (99%), of which 20 (12%) had more than one HPV genotype detected using WTS-PCR. Evaluation by LCM of individual biopsies with mixed infections showed 18 cases (90%) had only one HPV genotype associated with each CIN3 lesion. HPV 16 was the most common HPV type, found in 95/168 cases (57%). Conclusion: LCM-PCR allowed us to confirm the presence of a single HPV genotype associated with each biologically separate CIN3 lesion, supporting the theory that only one virus type causes each independent CIN lesion. LCM will provide an important tool in assessing vaccine effectiveness in HPV vaccine programs. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Abbreviations: AIS, adenocarcinoma in situ; CIN, cervical intraepithelial neoplasia; DNA, deoxyribonucleic acid; FFPE, formalin-fixed, paraffin-embedded; H&E, haematoxylin and eosin; HPV, human papillomavirus; HR, high-risk; LCM, laser capture microdissection; LEEP, loop electrical excision procedure; PEN, polyethylene naphthalate; PCR, polymerase chain reaction; VACCINE, vaccine against cervical cancer impact and effectiveness study; WTS, whole tissue section. ∗ Corresponding authors at: Department of Microbiology and Infectious Diseases, Murdoch Childrens Research Institute, Level 1, Building 404, Bio 21 Institute, 30 Flemington Road, Parkville, VIC 3052, Australia. Tel.: +61 418170334; fax: +61 383453671/+61 3 9347 8235. E-mail addresses: [email protected] (E.T. Callegari), [email protected] (S.N. Tabrizi), [email protected] (J. Pyman), [email protected] (M. Saville), [email protected] (A.M. Cornall), [email protected] (J.M.L. Brotherton), [email protected] (S.M. Garland). 1 On behalf of the VACCINE Study Group.
Worldwide, cervical cancer is responsible for approximately 275,000 deaths annually [1] and is the third most common cancer of women [1,2]. It is well established that human papillomavirus (HPV) is a necessary cause of all cervical cancers, with high-risk (HR) types, particularly HPV16 and 18, being the most oncogenic, with relative risks of several hundred-fold [3]. Genital HPV infection is the most common viral sexually transmitted infection amongst males and females [4,5]. Transmissibility is rapid with over 50% of young women acquiring HPV infection within the first two years of becoming sexually active [6]. Around 70–80% of women will have an HPV infection during their lifetime [7,8]. However, the majority of HPV infections are transient and do not lead to cancer, which takes years to develop and is preceded by persistent infection with HR-HPV genotypes, followed by a
http://dx.doi.org/10.1016/j.vaccine.2014.05.041 0264-410X/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
2
combination of multiple aberrant cellular events. Thus cancer is a rare outcome of a very common infection [9,10]. Cervical lesions and cancers are believed to be the result of a monoclonal event related to a single HPV genotype [11]. However, co-detection of DNA from multiple HPV genotypes in cervical swabs and biopsy material is not infrequent, particularly amongst younger women [12–14]. In more than 20–30% of women with cervical HPV infection, DNA from multiple genotypes are found regardless of diagnosis [11]. In a study of 3444 liquid based cytological cervical samples, researchers found multiple HR-HPV genotypes in 40.4% of women with high-grade dyskaryosis (n = 38) [13]. A Melbourne study examining cervical biopsies from patients being treated for moderate (n = 122) or severe (n = 180) cervical intraepithelial neoplasia (CIN II/III) or cervical cancer (n = 191) detected multiple HPV genotypes in 14.4% of samples [15]. Therefore, defining the aetiological genotype in a biopsy from cancer or CIN lesions can prove difficult, when utilising methods such as whole tissue section PCR (WTS-PCR). Detection of multiple types may be due to two abutting lesions caused by different genotypes, or one lesion with surrounding cervical epithelium harbouring an infection, without cytological change, with a different HPV genotype. Recent deposition from intercourse, shedding of virus from an infection in another region of the cervix or potential contamination during the biopsy procedure are other causes [16,17]. Until recently, molecular pathological mapping of HPV DNA specifically to cervical lesions in the presence of multiple genotypes had not been reported [16]. Laser capture microdissection (LCM) is a technique for precisely excising cells from a specific lesion. Unlike whole tissue sectioning, it can be used to discriminate between adjacent lesions, or between infection and recent deposition of HPV [16,18,19]. Accurately determining the cause of precursor lesions as surrogates of cancer is important in measuring vaccine effectiveness following public health campaigns of HPV vaccination. The Vaccine against cervical cancer impact and effectiveness (VACCINE) study aims to measure the effectiveness of the HPV vaccine in an Australian setting [20]. The VACCINE study has 2 sub-studies recruiting young women living in Victoria, Australia. The first (sub-study A) is investigating the prevalence of vaccine-related genotype carriage, whilst the aim of the present study (sub-study B) is to examine the distribution of HPV genotypes detected in high-grade CIN and AIS lesions (in women who were age-eligible for the National HPV Vaccination Program i.e. born after 30th June 1981). To accomplish this, a sample size of 500 CIN3/AIS cases is being evaluated [20]. We have specifically chosen to characterise HPV genotypes associated with CIN3, as the true precursor to cancer, rather than CIN2/3 [20]. 2. Materials and methods In this report we evaluate the use of LCM technology to attribute a specific HPV type to a lesion in cervical sections where mixed infections are present, within sub study B. 2.1. Ethics The study protocol was approved by the Royal Women’s Hospital Human Research Ethics Committees, and is being carried out according to the National Statement on Ethical Conduct in Research Involving Humans (June 1999) produced by the National Health and Medical Research Council of Australia. 2.2. Cases investigated All consecutive patient biopsy blocks submitted for pathological diagnosis between May 2011 and August 2012 containing CIN3 or AIS lesions were retrieved from the Royal Women’s Hospital
Fig. 1. Sandwich cutting procedure. Adapted from [16] Abbreviations: H&E, haematoxylin and eosin; LCM, laser capture microdissection; WTS, whole tissue section.
Dysplasia Clinic (Parkville, Victoria, Australia) and VCS Pathology (Carlton, Victoria, Australia). All colposcopically-guided cervical biopsies and cervical excision specimens were fixed in formalin and embedded in paraffin (FFPE). All cases were reviewed by one specialist gynaecological anatomical pathologist at each respective institution for inclusion in the study. Any cases which did not meet the inclusion criteria (patient age, diagnosis of CIN3 or AIS) as well as any cases without remaining tissue were excluded from consideration for this study. 2.3. Serial sectioning Paraffin embedded blocks were cut using a sandwich cutting procedure as previously described [21] and as shown in Fig. 1. Briefly, all biopsy blocks were serially sectioned with an initial 3 M section for histopathological review (haematoxylin and eosin stained or H&E pre), followed by one 9 M section mounted onto an Arcturus PEN (polyethylene naphthalate) membrane glass slide (Applied Biosystems, Foster City, CA) for laser capture microdissection PCR (LCM-PCR), two unmounted 9 M sections for whole tissue section-PCR (WTS-PCR) and lastly another 3 M section for histopathological review following haematoxylin and eosin staining (H&E post). To avoid contamination, sections were cut from the basal side towards the superficial (epithelial) side. The microtome stage and forceps were cleaned with Para-Kleaner (United Biosciences, Carindale, Australia) followed by ethanol, and a fresh blade and a new container for water for floating sections was used for each paraffin block, to minimise cross-contamination. 2.4. Review diagnosis and grading The H&E stained sections were scanned using the Aperio ScanScope (Aperio Technologies, Inc., Vista, CA, USA) slide scanner. H&E slides and scanned images were reviewed and annotated by one of two anatomical pathologists (JP and MS) to identify regions of CIN3 or AIS to be excised by LCM [22]. Cases which did not contain areas of CIN3 or AIS after sectioning were excluded from further study. Stromal areas were also annotated and used as negative controls. Of 169 patient biopsy blocks or cases which were eligible upon review, 20 biopsy samples contained DNA from multiple HPV genotypes. 2.5. DNA extraction of whole tissue section Methods as described previously were utilised for deparaffinisation and HPV DNA isolation [15]. Briefly, WTS was deparaffinised with histolene, and incubated at 55 ◦ C for 1 h with 160 l Tissue Lysis Buffer (Roche Molecular Systems, Alameda, CA) and 40 l Proteinase K (Roche Molecular Systems) followed by overnight incubation at 37 ◦ C. The entire sample was then extracted on the MagNA Pure LC (Roche Diagnostics GmbH, Penzberg, Germany)
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
3
using the DNA-I kit (blood cells high-performance protocol) according to the manufacturer’s instructions. The final volume of isolated DNA in elution buffer was 100 l. 2.6. Laser capture microdissection LCM was performed on all samples regardless of whether a single or multiple HPV genotypes were found by whole tissue section-PCR. LCM was used to obtain CIN3- or AIS-specific tissue from a section placed onto an unstained PEN membrane slide. Slides were deparaffinised using two separate washes of 100% xylene (5 min each) followed by two separate washes in 100% ethanol (5 min each). Slides were air-dried for at least 5 min prior to dissection. Tissue dissection was performed on the Veritas 704 Laser Capture Microdissecting System (Arcturus Bioscience, Mountain View CA, USA) as previously described [23]. The Arcturus® PicoPure® DNA extraction kit (Applied Biosystems) was used to extract DNA from cells captured using CapSure Macro LCM Caps (Applied Biosystems) as per protocol F of the PicoPure DNA Extraction Kit User guide. Fifty microliters of Proteinase K Extraction Solution (Applied Biosystems) was added to each CapSure Macro LCM Cap and incubated overnight (>16 h) at 65 ◦ C, followed by 95 ◦ C for 10 min to inactivate the Proteinase K. Subsequently the digested cells were either subjected to PCR analysis or were stored at −20 ◦ C until later testing.
Fig. 2. HPV genotype distribution of 168 CIN3 or AIS cases positive for HPV-DNA. Note: ‘Mixed’ refers to cases where more than 1 HPV type was found in a lesion.
2.7. HPV DNA detection and genotyping DNA extracted from both LCM and WTS was PCR-amplified in a 50 l reaction volume comprising 10 l of purified DNA sample or positive control, and 40 l of the PCR master mix (DNA ELISA kit HPV SPF10, Labo Biomedical Products, Rijswijk, The Netherlands). The DNA ELISA kit HPV SPF10, version 1 kit is able to amplify and detect 69 HPV genotypes. A 65 bp region of the L1 region was amplified utilizing SPF10 primers to test DNA as previously described [16,24]. Samples positive for HPV DNA by DNA ELISA were genotyped by the highly sensitive reverse hybridisation assay with the RHA kit HPV SPF10-LiPA25, version 1 (Labo Bio-medical Products, Rijswijk, The Netherlands) [25]. The kit contains probes for the detection and identification of 25 different HPV genotypes (HPV 6, 11, 16, 18, 31, 33, 34,35, 39, 40, 42, 43, 44, 45, 51, 52, 53, 54, 56, 58, 59,66, 68/73, 70, and 74). Samples positive for HPV DNA using the ELISA, but which could not be typed using SPF10-LiPA25 were considered untypable and labelled HPV-X. 2.8. Beta globin analysis To assess satisfactory quality and quantity of total DNA, all samples (169) were assayed for a 110 bp section of the human beta-globin gene by quantitative real-time PCR using 5 l of DNA sample or standard as previously described [21]. Any samples negative for both HPV DNA detection and beta-globin qPCR detection were considered unassessable and excluded from further analysis. 3. Results The patients ages ranged from 18 to 31 years (mean 26). Of the 169 cases examined, 168 cases were HPV positive (99%) whilst 1 case was negative for both HPV and beta-globin DNA and therefore unassessable (1%). There were 3 AIS cases (2%). Of the HPV-positive cases, 144 (86%) were positive for a single HPV genotype [4 (2%) were HPV-X], whilst 20 (12%) were positive for multiple HPV genotypes on WTS-PCR. All 168 HPV positive CIN3 cases had adequate internal controls and therefore were analysed by LCM-PCR. Fig. 2 illustrates the distribution of HPV genotypes found for all positive cases. HPV 16 was the most common genotype, being
found in 58% (95 cases), followed by HPV 31 in 14% (23 cases) and HPV 18 in 5% (8 cases). Table 1 shows HPV genotypes for CIN3 cases positive for more than one HPV genotype by WTS-PCR, as well as by LCM. In 18 cases LCM-PCR found a single HPV type within the CIN3 lesion. HPV 16 was the most common single causal HPV genotype found in cases where only one HPV type was found within the lesion (14/18 cases, 78%). Fig. 3 shows two representative cases, patient 4 (target biopsy) and patient 8 (loop electrical excision procedure or LEEP). Genotypes HPV 31 and HPV 35 were detected by WTS-PCR for patient 4; however after dissection of the annotated CIN3 epithelium, LCM-PCR revealed HPV 31 as the associated type. Similarly
Table 1 Final results for 20 cases of CIN3 cases with multiple HPV genotypes detected by WTS PCR. Patient number
WTS HPV genotype
LCM HPV genotype
1 2 3 4 5 6 7 8 9 10 11 12 (lesion 1) 12 (lesion 2) 13 14 15 16 17 18 19 20
16, 39 16, 39, 66 16, 18 31, 35 16, 52 11, 16 16, 56, 58 16, 51 16, 52 16, 18, 31 16, 51 33, 51, 52 33, 51, 52 16, 31 16, 18 16, 66 16, 33 16, 31 52, 56 16, 51 16, 52
16 16 18 31 16 16 16 16 52 16 16 331 521 16 16 16 332 16 52 16 16
(1) Primary result: 33, 52. (2) Primary result: 16, 33.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7 4
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
Fig. 3. Representative cases of CIN3; case 4 was target biopsy and case 8 a LEEP. Abbreviations: H&E, haematoxylin and eosin; LCM, laser capture microdissection.
both HPV 16 and HPV 51 were detected by WTS-PCR in the biopsy of patient 8. After dissection of the annotated CIN3 epithelium, LCM-PCR revealed HPV 16 as the associated type. A stromal area and a CIN1 lesion were also annotated and dissected for case 8, on separate LCM caps (image not shown). The CIN1 epithelium tested positive for HPV 16, whilst the stromal control was negative for HPV-DNA. There were two cases where more than one HPV type was found in the microdissected lesion (cases 12 and 16) and these were further investigated. The biopsy blocks of these particular cases were
re-sectioned; the lesions recut using LCM and tested for the presence of HPV. Case 12 has been used as an example. The paraffin embedded block contained 2 target biopsies. DNA from HPV 33, 51, 52 was detected in the WTS, while HPV 33 and 52 were found in the original dissected CIN3 lesion. Fig. 4a illustrates the HPV genotypes found in different areas of the section as separately cut out by LCM. We hypothesised that due to the direction of cutting (as indicated by the arrow) HPV 52 DNA was carried across the section to contaminate the second lesion. To confirm this, the biopsy block was recut in the opposite direction (see Fig. 4b). In this instance HPV 33
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
5
Fig. 4. (a) Case 12—Histological map of HPV genotypes present in lesions in the original section. The arrow represents the direction of movement of the microtome blade during sectioning across two embedded lesions within the one paraffin block. Based on these results, it is proposed that HPV 52 DNA from one CIN3 lesion (red) was carried across to the second CIN3 lesion on the same section (blue). HPV 52 DNA was also detected in another area of the cervical section that was not annotated as an area of CIN (green), supporting the theory of contamination by the microtome blade in the cutting procedure. (b) Case 12—Histological map of HPV genotypes present in the same section when recut and retested. The arrow represents the direction of movement of the microtome blade during sectioning. Based on these results, it is proposed that HPV 33 DNA from one CIN3 lesion (blue) was carried across to the other CIN3 lesion present on the same section (red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
DNA was carried across the section to contaminate the lesion that was originally positive for HPV 52 only. Case 16, originally HPV 16 and 33 in the microdissected lesion, was found to be positive for HPV 33 alone when only basal epithelium was microdissected, indicating the likely presence of surface contamination by HPV 16 DNA (image not shown). 4. Discussion The technique of LCM has been a major advance in allowing scientists to isolate a pure population of cells for genetic analysis [19]. The methodology has been newly applied to the investigation of HPV genotypes associated with specific CIN3, AIS and/or cervical cancers, particularly when seemingly mixed infection is detected or there are multiple contiguous lesions. Moreover, LCM has the advantage of detecting HPV at a low copy number, with high sensitivity [26]. The current widely-used method of HPV detection in cervical epithelium, whole tissue section PCR, is limited in that it cannot differentiate between HPV associated with individual lesions, HPV deposition or surrounding infection. A further advantage of LCM is that it can be used on FFPE tissue sections, allowing archival tissue to be dissected with ease. Its use however has been limited to date, as it is expensive, low-throughput and requires anatomical pathologists to map lesions. Care needs to be taken when cutting an individual section for LCM to avoid contamination. To ensure this the section should always be cut from basal towards the superficial side of the epithelium. In addition, it is essential to microdissect cells within a lesion from the basal and parabasal layers rather than the superficial layer of epithelium. If multiple HPV genotypes are found in a particular cervical section it may be necessary to take tissue from multiple areas of it to ascertain HPV association with specific lesions if more than one is present [16]. Techniques used may not totally eliminate the deposition or carryover of HPV DNA, but greatly reduces it. In the present study evaluating LCM in the effectiveness of a large public vaccination program, HPV genotyping of isolated cervical tissue using LCM demonstrates that only one HPV genotype was present in each specific CIN3 lesion. This confirms that a single CIN lesion is the result of a single clonal HPV infection and hence
each CIN lesion is biologically independent [16]. In our study LCM was conducted on all of our 168 cases, whether they were single or mixed infections. For single HPV infections the same HPV type was found both in the whole tissue section and in the lesion dissections. Accordingly, we now only perform LCM for cases where multiple HPV genotypes are present in the WTS. Our findings on deciphering causality using LCM for mixed infections are important as a single cervical biopsy sample may have separate cervical lesions of differing severity that are caused by different genotypes of HPV; yet this distinction is not able to be determined with genotyping WTS-DNA alone. In a diagnostic setting, where more than one biopsy specimen may be embedded per paraffin block, HPV DNA may be carried across from one lesion to another in cutting. The use of LCM allows the designation of a HPV genotype to a specific lesion in such a scenario, as depicted in Fig. 4a and b. The prevalence of HPV 16 (58%) is consistent with similar studies, adding to and confirming existing evidence that HPV 16 is aetiologically dominant in the pathogenesis of cervical cancer [27]. The ability to isolate the associated HPV genotype using LCM-PCR allows us to better understand the natural history of specific HPV genotypes, as well as their associated lesions. This will be highly beneficial in the fields of vaccine development, both preventative and therapeutic, as well as accurately determining the outcome of vaccine programs. LCM is a more specific methodology to assign causality and therefore a better assessment of HPV vaccine effectiveness, as well as efficacy. In the phase 3 clinical trials for the bivalent vaccine as well as the quadrivalent vaccine, mixed infections – whether they be determined by swabs or by biopsies – were not uncommon [28]. Accordingly, several different methods were used to attribute causality. In the phase 3 clinical trial of the bivalent vaccine, causality was assigned based on the presence of oncogenic HPV genotypes in cytological samples preceding the development of CIN; the HPV type assignment algorithm [29]. Alternatively, in a study examining HPV genotype-specific infection in relation to vulval intraepithelial neoplasia, causality was assigned based on statistical adjustments according to genotype prevalence data (proportional adjustment approach) [30]. Either methodology could potentially lead to misidentification of HPV genotypes associated with specific CIN lesions. Specifically, within this report there are
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
6
two cases where the associated HPV type would be misidentified using the proportional adjustment approach. Case 3 would be assigned HPV 16, instead of HPV 18. Similarly, case 9 would have been assigned HPV 16, instead of HPV 52. Whilst the HPV type assignment algorithm and proportional adjustment methods have been used in the past, these methods are not definitive in case assignment. The use of LCM to measure outcomes of vaccination would be highly beneficial as misclassification of associated HPV genotypes could be avoided in cases where multiple HPV types are present. In conclusion, LCM-PCR is the preferred methodology to attribute HPV genotype to specific cervical lesions where multiple HPV genotypes are detected. It provides the ability to distinguish between lesion-associated HPV infection and incidental HPV DNA. The use of LCM in the future will allow us to better understand the natural history of specific HPV genotypes and the interactions of multiple HPV infections within the cervix. LCM-PCR will also allow investigation of the relationship between HPV infection and disease progression in other forms of anogenital disease such as anal cancer. Such knowledge would greatly improve diagnosis and subsequent treatment of patients with persistent HPV infection and resultant neoplasia, as well as improve assessment of vaccine effectiveness. VACCINE Study Group Callegari ET, Tabrizi SN, Pyman J, Saville M, Cornall AM, Brotherton JML, Pitts M (La Trobe University, Australian Research Centre for Sex, Health & Society), Gertig DM (VCS Incorporated; School of Population and Global Health, University of Melbourne), Wark JD (Department of Medicine, Bone and Mineral Service, The University of Melbourne, The Royal Melbourne Hospital), Jayasinghe Y (Department of Obstetrics and Gynaecology, The University of Melbourne; Department of Gynaecology, Royal Children’s Hospital), Tan J (Department of Oncology & Dysplasia, Royal Women’s Hospital), Wrede CD (Department of Oncology & Dysplasia, Royal Women’s Hospital), and Garland SM. Conflict of interest statement SMG has received advisory board fees and grant support from bioCSL and GlaxoSmithKline Australia, and lecture fees from Merck, GSK and Sanofi Pasteur; in addition, SMG has received funding to conduct HPV vaccine studies for MSD and GSK. SMG is a member of the Merck Global Advisory Board as well as the Merck Scientific Advisory Committee for HPV. JMLB, DMG, MS and SMG were partner investigators on an Australian Research Council Linkage Grant 2008–2011 on which bioCSL Biotherapies was a partner organisation. Author contributions ETC was involved in data collection, and is the primary author of this manuscript. SMG, chief investigator of the study, conceived the study and was involved in study design, study coordination and helped to draft this manuscript. SNT has overseen all molecular HPV laboratory aspects of the study, including study design and data collection and helped to draft this manuscript. AMC was involved in overseeing molecular HPV laboratory aspects and data collection and helped to draft this manuscript. MS, JP were involved in study design, planning and co-ordination of histology processing and pathologic review of all study cases for sub study B. JT was involved in the study design for sub study B. JMLB, MP, DG, JDW, YJ were involved in study design. DCW is involved in the coordination of sub study B. All authors read and approved this manuscript.
Acknowledgements We grateful acknowledge funding from the Victorian Cancer Agency (VCA). We thank the Anatomical Pathology Department at Royal Melbourne Hospital for providing slide-imaging services and Elizabeth McKinnon (Department of Anatomical Pathology, Royal Children’s Hospital, Melbourne) and Eileen Tan (VCS) who sectioned the tissue blocks and prepared the slides. We thank Tania Tabone for assistance in sub study B set up, Elisa Young for project co-ordination and Houda Abdo, Angela Hurley, Fiona Tan and for sample preparation and genotype analysis. We acknowledge the initial assistance and advice of LCM methods from Wim H. Quint.
References [1] Ferlay J, Shin H, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127:2893–917. [2] Parkin DM, Bray F. Chapter 2: The burden of HPV-related cancers. Vaccine 2006;24:S11–25. [3] Garland SM, Tabrizi SN. H.P.V. D.N.A. testing: potential clinical applications. Int J Health Promot Educ 2006;44:107–12. [4] Markowitz L, Dunne EF, Saraiya M, Lawson HW, Chesson H, Unger ER. Quadrivalent human papillomavirus vaccine. Morb Mortal Wkly Rep 2007;56:1–23. [5] Burchell AN, Winer RL, de Sanjose S, Franco EL. Chapter 6: Epidemiology and transmission dynamics of genital HPV infection. Vaccine 2006;24S3::52–61. [6] Winer RL, Lee S-K, Hughes JP, Adam DE, Kiviat NB, Joutsky LA. Genital human papillomavirus infection: incidence and risk factors in a cohort of female university students. Am J Epidemiol 2003;157:218–26. [7] Moscicki AB. HPV infections in adolescents. Dis Markers 2007;23:229–34. [8] Baseman JG, Koutsky LA. The epidemiology of human papillomavirus infections. J Clin Virol 2005;32:16. [9] Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, Snijders PJF, Peto J, Meijer CJL, Munoz N. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:12–9. [10] Galani E, Christodoulou C. Human papilloma viruses and cancer in the postvaccine era. Clin Microbiol Infect 2009;15:977–81. [11] Moscicki AB, Schiffman M, Kjaer S, Villa LL. Updating the natural history of HPV and anogenital cancer. Vaccine 2006;24:S42–51. [12] Thomas KK, Hughes JP, Kuypers JM, Kiviat NB, Lee S-K, Adam DE, Koutsky LA. Concurrent and sequential acquisition of different genital human papillomavirus types. J Infect Dis 2000;182:1097. [13] Cuschieri KS, Cubie HA, Whitley MW, Seagar AL, Arends MJ, Moore C, et al. Multiple high risk HPV infections are common in cervical neoplasia and young women in a cervical screening population. J Clin Pathol 2004;57:68–72. [14] Garland SM, Brotherton JML, Condon JR, McIntyre PB, Stevens MP, Smith DW, Tabrizi SN. Human papillomavirus prevalence among indigenous and nonindigenous Australian women prior to a national HPV vaccination program. BMC Med 2011;9:104. [15] Stevens MP, Tabrizi SN, Quinn MA, Garland SM. Human papillomavirus genotype prevalence in cervical biopsies from women diagnosed with cervical intraepithelial neoplasia or cervical cancer in Melbourne, Australia. Int J Gynecol Cancer 2006;16:1017–24. [16] Quint W, Jenkins D, Molijn A, Struijk L, van de Sandt M, Doorbar J, Mols J, Van Hoof C, Hardt K, Struyf F, Colau B. One virus, one lesion—individual components of CIN lesions contain a specific HPV type. J Pathol 2011;227:62–71. [17] Kjaer SK, De Villiers EM, Ca˘glayan H, Svare E, Haugaard BJ, Engholm G, Christensen RB, Møller KA, Poll P, Jensen H. Human papillomavirus, herpes simplex virus and other potential risk factors for cervical cancer in a high-risk area (Greenland) and a low-risk area (Denmark)—a second look. Br J Cancer 1993;67:830. [18] Curran S, McKay JA, McLeod HL, Murray GI. Laser capture microscopy. Mol Pathol 2000;53:64–8. [19] Emmert Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA. Laser capture microdissection. Science 1996;274:998–1001. [20] Young EJ, Tabrizi SN, Brotherton JML, Wark JD, Payman J, Saville M, Wrede DC, Jayasinghe Y, Tan J, Gertig DM, Pitts M, Garland SM. Measuring effectiveness of the cervical cancer vaccine in an Australian setting (the VACCINE study). BMC Cancer 2013;13:296. [21] Cornall AM, Roberts JM, Garland SM, Hillman RJ, Grulich AE, Tabrizi SN. Anal and perianal squamous carcinomas and high-grade intraepithelial lesions exclusively associated with “low-risk” HPV genotypes 6 and 11. Int J Cancer 2013. [22] Jenkins D. Histopathology and cytopathology of cervical cancer. Dis Markers 2007;23:199. [23] Espina V, Wulfkuhle JD, Calvert VS, VanMeter A, Zhou W, Coukos G, Geho DH, Petricoin EF, Liotta LA. Laser-capture microdissection. Nat Protoc 2006;1:586–603. [24] Quint WGV, Scholte G, van Doorn LJ, Kleter B, Smits PHM, Lindeman J. Comparative analysis of human papillomavirus infections in cervical scrapes and biopsy specimens by general SPF10 PCR and HPV genotyping. J Pathol 2001;194:51.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
[25] Kleter B, Van Doorn L-J, Schrauwen L, Molijn A, Sastrowijoto S, ter Schegget J, et al. Development and clinical evaluation of a highly sensitive PCR-reverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. J Clin Microbiol 1999;37:2508–17. [26] Kalantari M, Garcia-Carranca A, Morales-Vazquez CD, Zuna R, Montiel DP, Calleja-Macias IE, Johansson B, Andersson S, Bernard HU. Laser capture microdissection of cervical human papillomavirus infections: copy number of the virus in cancerous and normal tissue and heterogeneous DNA methylation. Virology 2009;390:261–7. [27] van der Marel J, Quint WGV, Schiffman M, van de Sandt MM, Zuna RE, Terence Dunn S, Smith K, Mathews CA, Gold MA, Walker J, Wentzensen N. Molecular mapping of high-grade cervical intraepithelial neoplasia shows etiological dominance of HPV16. Int J Cancer 2012;131:E946–53.
7
[28] The FUTURE.II Study Group. Quadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions. N Engl J Med 2007;356:1915–27. [29] Paavonen J, Jenkins D, Bosch FX, Naud P, Salmerón J, Wheeler CM, Chow S-N, Apter DL, Kitchener HC, Castellsague X, de Carvalho NS, Skinner SR, Harper DM, Hedrick JA, Jaisamrarn U, Limson GAM, Dionne M, Quint W, Spiessens B, Peeters P, Struyf F, Wieting SL, Lehtinen MO, Dubin G. Efficacy of a prophylactic adjuvanted bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a phase III double-blind, randomised controlled trial. Lancet 2007;369:2161–70. [30] Garland SM, Insinga RP, Sings HL, Haupt RM, Joura EA. Human papillomavirus infections and vulvar disease development. Cancer Epidemiol Biomarkers Prev 2009;18:1777–84.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
ARTICLE IN PRESS
JVAC-15415; No. of Pages 7
Vaccine xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions Emma T. Callegari a,c,1 , Sepehr N. Tabrizi b,c,d,1 , Jan Pyman e,1 , Marion Saville f,1 , Alyssa M. Cornall b,c,1 , Julia M.L. Brotherton f,g,1 , Suzanne M. Garland b,c,d,∗,1 a
The University of Melbourne, Grattan Street, Parkville, Melbourne 3010, VIC, Australia Regional HPV Reference Laboratory, Department of Microbiology and Infectious Diseases, The Royal Women’s Hospital, Parkville, VIC, Australia c Murdoch Childrens Research Institute, Parkville, VIC, Australia d Department of Obstetrics and Gynaecology, University of Melbourne, Melbourne, VIC, Australia e Department of Anatomical Pathology, Royal Women’s Hospital, 5th Floor, 20 Flemington Road, Parkville, Melbourne 3052, VIC, Australia f VCS Incorporated, 265 Faraday St, Carlton, Melbourne 3053, VIC, Australia g School of Population and Global Health, University of Melbourne, Melbourne, VIC, Australia b
a r t i c l e
i n f o
Article history: Received 12 February 2014 Received in revised form 2 May 2014 Accepted 7 May 2014 Available online xxx Keywords: Human papillomavirus Cervical cancer CIN3 AIS Laser capture microdissection Multiple infections
a b s t r a c t Objectives: This study aimed to determine human papillomavirus (HPV) genotypes present in biopsy sections from young women of vaccine eligible age living in Victoria, Australia, with confirmed cervical intraepithelial neoplasia grade 3 (CIN3) or adenocarcinoma in situ (AIS) using laser capture microdissection (LCM). Methods: Histologically confirmed CIN3 or AIS positive biopsies from vaccine eligible women (born after 30th June 1981, n = 169), between May 2011 and March 2013, were identified. CIN3 or AIS lesions were isolated from biopsy material using LCM, and the HPV genotypes present in whole tissue sections (WTS) as well as LCM-isolated lesion tissue were determined by a sensitive reverse hybridisation assay; RHA kit HPV SPF10-LiPA25, version 1 (Labo Bio-medical Products, Rijswijk, The Netherlands). Results: One hundred and sixty-eight cases were shown to be HPV positive (99%), of which 20 (12%) had more than one HPV genotype detected using WTS-PCR. Evaluation by LCM of individual biopsies with mixed infections showed 18 cases (90%) had only one HPV genotype associated with each CIN3 lesion. HPV 16 was the most common HPV type, found in 95/168 cases (57%). Conclusion: LCM-PCR allowed us to confirm the presence of a single HPV genotype associated with each biologically separate CIN3 lesion, supporting the theory that only one virus type causes each independent CIN lesion. LCM will provide an important tool in assessing vaccine effectiveness in HPV vaccine programs. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Abbreviations: AIS, adenocarcinoma in situ; CIN, cervical intraepithelial neoplasia; DNA, deoxyribonucleic acid; FFPE, formalin-fixed, paraffin-embedded; H&E, haematoxylin and eosin; HPV, human papillomavirus; HR, high-risk; LCM, laser capture microdissection; LEEP, loop electrical excision procedure; PEN, polyethylene naphthalate; PCR, polymerase chain reaction; VACCINE, vaccine against cervical cancer impact and effectiveness study; WTS, whole tissue section. ∗ Corresponding authors at: Department of Microbiology and Infectious Diseases, Murdoch Childrens Research Institute, Level 1, Building 404, Bio 21 Institute, 30 Flemington Road, Parkville, VIC 3052, Australia. Tel.: +61 418170334; fax: +61 383453671/+61 3 9347 8235. E-mail addresses: [email protected] (E.T. Callegari), [email protected] (S.N. Tabrizi), [email protected] (J. Pyman), [email protected] (M. Saville), [email protected] (A.M. Cornall), [email protected] (J.M.L. Brotherton), [email protected] (S.M. Garland). 1 On behalf of the VACCINE Study Group.
Worldwide, cervical cancer is responsible for approximately 275,000 deaths annually [1] and is the third most common cancer of women [1,2]. It is well established that human papillomavirus (HPV) is a necessary cause of all cervical cancers, with high-risk (HR) types, particularly HPV16 and 18, being the most oncogenic, with relative risks of several hundred-fold [3]. Genital HPV infection is the most common viral sexually transmitted infection amongst males and females [4,5]. Transmissibility is rapid with over 50% of young women acquiring HPV infection within the first two years of becoming sexually active [6]. Around 70–80% of women will have an HPV infection during their lifetime [7,8]. However, the majority of HPV infections are transient and do not lead to cancer, which takes years to develop and is preceded by persistent infection with HR-HPV genotypes, followed by a
http://dx.doi.org/10.1016/j.vaccine.2014.05.041 0264-410X/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
2
combination of multiple aberrant cellular events. Thus cancer is a rare outcome of a very common infection [9,10]. Cervical lesions and cancers are believed to be the result of a monoclonal event related to a single HPV genotype [11]. However, co-detection of DNA from multiple HPV genotypes in cervical swabs and biopsy material is not infrequent, particularly amongst younger women [12–14]. In more than 20–30% of women with cervical HPV infection, DNA from multiple genotypes are found regardless of diagnosis [11]. In a study of 3444 liquid based cytological cervical samples, researchers found multiple HR-HPV genotypes in 40.4% of women with high-grade dyskaryosis (n = 38) [13]. A Melbourne study examining cervical biopsies from patients being treated for moderate (n = 122) or severe (n = 180) cervical intraepithelial neoplasia (CIN II/III) or cervical cancer (n = 191) detected multiple HPV genotypes in 14.4% of samples [15]. Therefore, defining the aetiological genotype in a biopsy from cancer or CIN lesions can prove difficult, when utilising methods such as whole tissue section PCR (WTS-PCR). Detection of multiple types may be due to two abutting lesions caused by different genotypes, or one lesion with surrounding cervical epithelium harbouring an infection, without cytological change, with a different HPV genotype. Recent deposition from intercourse, shedding of virus from an infection in another region of the cervix or potential contamination during the biopsy procedure are other causes [16,17]. Until recently, molecular pathological mapping of HPV DNA specifically to cervical lesions in the presence of multiple genotypes had not been reported [16]. Laser capture microdissection (LCM) is a technique for precisely excising cells from a specific lesion. Unlike whole tissue sectioning, it can be used to discriminate between adjacent lesions, or between infection and recent deposition of HPV [16,18,19]. Accurately determining the cause of precursor lesions as surrogates of cancer is important in measuring vaccine effectiveness following public health campaigns of HPV vaccination. The Vaccine against cervical cancer impact and effectiveness (VACCINE) study aims to measure the effectiveness of the HPV vaccine in an Australian setting [20]. The VACCINE study has 2 sub-studies recruiting young women living in Victoria, Australia. The first (sub-study A) is investigating the prevalence of vaccine-related genotype carriage, whilst the aim of the present study (sub-study B) is to examine the distribution of HPV genotypes detected in high-grade CIN and AIS lesions (in women who were age-eligible for the National HPV Vaccination Program i.e. born after 30th June 1981). To accomplish this, a sample size of 500 CIN3/AIS cases is being evaluated [20]. We have specifically chosen to characterise HPV genotypes associated with CIN3, as the true precursor to cancer, rather than CIN2/3 [20]. 2. Materials and methods In this report we evaluate the use of LCM technology to attribute a specific HPV type to a lesion in cervical sections where mixed infections are present, within sub study B. 2.1. Ethics The study protocol was approved by the Royal Women’s Hospital Human Research Ethics Committees, and is being carried out according to the National Statement on Ethical Conduct in Research Involving Humans (June 1999) produced by the National Health and Medical Research Council of Australia. 2.2. Cases investigated All consecutive patient biopsy blocks submitted for pathological diagnosis between May 2011 and August 2012 containing CIN3 or AIS lesions were retrieved from the Royal Women’s Hospital
Fig. 1. Sandwich cutting procedure. Adapted from [16] Abbreviations: H&E, haematoxylin and eosin; LCM, laser capture microdissection; WTS, whole tissue section.
Dysplasia Clinic (Parkville, Victoria, Australia) and VCS Pathology (Carlton, Victoria, Australia). All colposcopically-guided cervical biopsies and cervical excision specimens were fixed in formalin and embedded in paraffin (FFPE). All cases were reviewed by one specialist gynaecological anatomical pathologist at each respective institution for inclusion in the study. Any cases which did not meet the inclusion criteria (patient age, diagnosis of CIN3 or AIS) as well as any cases without remaining tissue were excluded from consideration for this study. 2.3. Serial sectioning Paraffin embedded blocks were cut using a sandwich cutting procedure as previously described [21] and as shown in Fig. 1. Briefly, all biopsy blocks were serially sectioned with an initial 3 M section for histopathological review (haematoxylin and eosin stained or H&E pre), followed by one 9 M section mounted onto an Arcturus PEN (polyethylene naphthalate) membrane glass slide (Applied Biosystems, Foster City, CA) for laser capture microdissection PCR (LCM-PCR), two unmounted 9 M sections for whole tissue section-PCR (WTS-PCR) and lastly another 3 M section for histopathological review following haematoxylin and eosin staining (H&E post). To avoid contamination, sections were cut from the basal side towards the superficial (epithelial) side. The microtome stage and forceps were cleaned with Para-Kleaner (United Biosciences, Carindale, Australia) followed by ethanol, and a fresh blade and a new container for water for floating sections was used for each paraffin block, to minimise cross-contamination. 2.4. Review diagnosis and grading The H&E stained sections were scanned using the Aperio ScanScope (Aperio Technologies, Inc., Vista, CA, USA) slide scanner. H&E slides and scanned images were reviewed and annotated by one of two anatomical pathologists (JP and MS) to identify regions of CIN3 or AIS to be excised by LCM [22]. Cases which did not contain areas of CIN3 or AIS after sectioning were excluded from further study. Stromal areas were also annotated and used as negative controls. Of 169 patient biopsy blocks or cases which were eligible upon review, 20 biopsy samples contained DNA from multiple HPV genotypes. 2.5. DNA extraction of whole tissue section Methods as described previously were utilised for deparaffinisation and HPV DNA isolation [15]. Briefly, WTS was deparaffinised with histolene, and incubated at 55 ◦ C for 1 h with 160 l Tissue Lysis Buffer (Roche Molecular Systems, Alameda, CA) and 40 l Proteinase K (Roche Molecular Systems) followed by overnight incubation at 37 ◦ C. The entire sample was then extracted on the MagNA Pure LC (Roche Diagnostics GmbH, Penzberg, Germany)
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
3
using the DNA-I kit (blood cells high-performance protocol) according to the manufacturer’s instructions. The final volume of isolated DNA in elution buffer was 100 l. 2.6. Laser capture microdissection LCM was performed on all samples regardless of whether a single or multiple HPV genotypes were found by whole tissue section-PCR. LCM was used to obtain CIN3- or AIS-specific tissue from a section placed onto an unstained PEN membrane slide. Slides were deparaffinised using two separate washes of 100% xylene (5 min each) followed by two separate washes in 100% ethanol (5 min each). Slides were air-dried for at least 5 min prior to dissection. Tissue dissection was performed on the Veritas 704 Laser Capture Microdissecting System (Arcturus Bioscience, Mountain View CA, USA) as previously described [23]. The Arcturus® PicoPure® DNA extraction kit (Applied Biosystems) was used to extract DNA from cells captured using CapSure Macro LCM Caps (Applied Biosystems) as per protocol F of the PicoPure DNA Extraction Kit User guide. Fifty microliters of Proteinase K Extraction Solution (Applied Biosystems) was added to each CapSure Macro LCM Cap and incubated overnight (>16 h) at 65 ◦ C, followed by 95 ◦ C for 10 min to inactivate the Proteinase K. Subsequently the digested cells were either subjected to PCR analysis or were stored at −20 ◦ C until later testing.
Fig. 2. HPV genotype distribution of 168 CIN3 or AIS cases positive for HPV-DNA. Note: ‘Mixed’ refers to cases where more than 1 HPV type was found in a lesion.
2.7. HPV DNA detection and genotyping DNA extracted from both LCM and WTS was PCR-amplified in a 50 l reaction volume comprising 10 l of purified DNA sample or positive control, and 40 l of the PCR master mix (DNA ELISA kit HPV SPF10, Labo Biomedical Products, Rijswijk, The Netherlands). The DNA ELISA kit HPV SPF10, version 1 kit is able to amplify and detect 69 HPV genotypes. A 65 bp region of the L1 region was amplified utilizing SPF10 primers to test DNA as previously described [16,24]. Samples positive for HPV DNA by DNA ELISA were genotyped by the highly sensitive reverse hybridisation assay with the RHA kit HPV SPF10-LiPA25, version 1 (Labo Bio-medical Products, Rijswijk, The Netherlands) [25]. The kit contains probes for the detection and identification of 25 different HPV genotypes (HPV 6, 11, 16, 18, 31, 33, 34,35, 39, 40, 42, 43, 44, 45, 51, 52, 53, 54, 56, 58, 59,66, 68/73, 70, and 74). Samples positive for HPV DNA using the ELISA, but which could not be typed using SPF10-LiPA25 were considered untypable and labelled HPV-X. 2.8. Beta globin analysis To assess satisfactory quality and quantity of total DNA, all samples (169) were assayed for a 110 bp section of the human beta-globin gene by quantitative real-time PCR using 5 l of DNA sample or standard as previously described [21]. Any samples negative for both HPV DNA detection and beta-globin qPCR detection were considered unassessable and excluded from further analysis. 3. Results The patients ages ranged from 18 to 31 years (mean 26). Of the 169 cases examined, 168 cases were HPV positive (99%) whilst 1 case was negative for both HPV and beta-globin DNA and therefore unassessable (1%). There were 3 AIS cases (2%). Of the HPV-positive cases, 144 (86%) were positive for a single HPV genotype [4 (2%) were HPV-X], whilst 20 (12%) were positive for multiple HPV genotypes on WTS-PCR. All 168 HPV positive CIN3 cases had adequate internal controls and therefore were analysed by LCM-PCR. Fig. 2 illustrates the distribution of HPV genotypes found for all positive cases. HPV 16 was the most common genotype, being
found in 58% (95 cases), followed by HPV 31 in 14% (23 cases) and HPV 18 in 5% (8 cases). Table 1 shows HPV genotypes for CIN3 cases positive for more than one HPV genotype by WTS-PCR, as well as by LCM. In 18 cases LCM-PCR found a single HPV type within the CIN3 lesion. HPV 16 was the most common single causal HPV genotype found in cases where only one HPV type was found within the lesion (14/18 cases, 78%). Fig. 3 shows two representative cases, patient 4 (target biopsy) and patient 8 (loop electrical excision procedure or LEEP). Genotypes HPV 31 and HPV 35 were detected by WTS-PCR for patient 4; however after dissection of the annotated CIN3 epithelium, LCM-PCR revealed HPV 31 as the associated type. Similarly
Table 1 Final results for 20 cases of CIN3 cases with multiple HPV genotypes detected by WTS PCR. Patient number
WTS HPV genotype
LCM HPV genotype
1 2 3 4 5 6 7 8 9 10 11 12 (lesion 1) 12 (lesion 2) 13 14 15 16 17 18 19 20
16, 39 16, 39, 66 16, 18 31, 35 16, 52 11, 16 16, 56, 58 16, 51 16, 52 16, 18, 31 16, 51 33, 51, 52 33, 51, 52 16, 31 16, 18 16, 66 16, 33 16, 31 52, 56 16, 51 16, 52
16 16 18 31 16 16 16 16 52 16 16 331 521 16 16 16 332 16 52 16 16
(1) Primary result: 33, 52. (2) Primary result: 16, 33.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7 4
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
Fig. 3. Representative cases of CIN3; case 4 was target biopsy and case 8 a LEEP. Abbreviations: H&E, haematoxylin and eosin; LCM, laser capture microdissection.
both HPV 16 and HPV 51 were detected by WTS-PCR in the biopsy of patient 8. After dissection of the annotated CIN3 epithelium, LCM-PCR revealed HPV 16 as the associated type. A stromal area and a CIN1 lesion were also annotated and dissected for case 8, on separate LCM caps (image not shown). The CIN1 epithelium tested positive for HPV 16, whilst the stromal control was negative for HPV-DNA. There were two cases where more than one HPV type was found in the microdissected lesion (cases 12 and 16) and these were further investigated. The biopsy blocks of these particular cases were
re-sectioned; the lesions recut using LCM and tested for the presence of HPV. Case 12 has been used as an example. The paraffin embedded block contained 2 target biopsies. DNA from HPV 33, 51, 52 was detected in the WTS, while HPV 33 and 52 were found in the original dissected CIN3 lesion. Fig. 4a illustrates the HPV genotypes found in different areas of the section as separately cut out by LCM. We hypothesised that due to the direction of cutting (as indicated by the arrow) HPV 52 DNA was carried across the section to contaminate the second lesion. To confirm this, the biopsy block was recut in the opposite direction (see Fig. 4b). In this instance HPV 33
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
5
Fig. 4. (a) Case 12—Histological map of HPV genotypes present in lesions in the original section. The arrow represents the direction of movement of the microtome blade during sectioning across two embedded lesions within the one paraffin block. Based on these results, it is proposed that HPV 52 DNA from one CIN3 lesion (red) was carried across to the second CIN3 lesion on the same section (blue). HPV 52 DNA was also detected in another area of the cervical section that was not annotated as an area of CIN (green), supporting the theory of contamination by the microtome blade in the cutting procedure. (b) Case 12—Histological map of HPV genotypes present in the same section when recut and retested. The arrow represents the direction of movement of the microtome blade during sectioning. Based on these results, it is proposed that HPV 33 DNA from one CIN3 lesion (blue) was carried across to the other CIN3 lesion present on the same section (red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
DNA was carried across the section to contaminate the lesion that was originally positive for HPV 52 only. Case 16, originally HPV 16 and 33 in the microdissected lesion, was found to be positive for HPV 33 alone when only basal epithelium was microdissected, indicating the likely presence of surface contamination by HPV 16 DNA (image not shown). 4. Discussion The technique of LCM has been a major advance in allowing scientists to isolate a pure population of cells for genetic analysis [19]. The methodology has been newly applied to the investigation of HPV genotypes associated with specific CIN3, AIS and/or cervical cancers, particularly when seemingly mixed infection is detected or there are multiple contiguous lesions. Moreover, LCM has the advantage of detecting HPV at a low copy number, with high sensitivity [26]. The current widely-used method of HPV detection in cervical epithelium, whole tissue section PCR, is limited in that it cannot differentiate between HPV associated with individual lesions, HPV deposition or surrounding infection. A further advantage of LCM is that it can be used on FFPE tissue sections, allowing archival tissue to be dissected with ease. Its use however has been limited to date, as it is expensive, low-throughput and requires anatomical pathologists to map lesions. Care needs to be taken when cutting an individual section for LCM to avoid contamination. To ensure this the section should always be cut from basal towards the superficial side of the epithelium. In addition, it is essential to microdissect cells within a lesion from the basal and parabasal layers rather than the superficial layer of epithelium. If multiple HPV genotypes are found in a particular cervical section it may be necessary to take tissue from multiple areas of it to ascertain HPV association with specific lesions if more than one is present [16]. Techniques used may not totally eliminate the deposition or carryover of HPV DNA, but greatly reduces it. In the present study evaluating LCM in the effectiveness of a large public vaccination program, HPV genotyping of isolated cervical tissue using LCM demonstrates that only one HPV genotype was present in each specific CIN3 lesion. This confirms that a single CIN lesion is the result of a single clonal HPV infection and hence
each CIN lesion is biologically independent [16]. In our study LCM was conducted on all of our 168 cases, whether they were single or mixed infections. For single HPV infections the same HPV type was found both in the whole tissue section and in the lesion dissections. Accordingly, we now only perform LCM for cases where multiple HPV genotypes are present in the WTS. Our findings on deciphering causality using LCM for mixed infections are important as a single cervical biopsy sample may have separate cervical lesions of differing severity that are caused by different genotypes of HPV; yet this distinction is not able to be determined with genotyping WTS-DNA alone. In a diagnostic setting, where more than one biopsy specimen may be embedded per paraffin block, HPV DNA may be carried across from one lesion to another in cutting. The use of LCM allows the designation of a HPV genotype to a specific lesion in such a scenario, as depicted in Fig. 4a and b. The prevalence of HPV 16 (58%) is consistent with similar studies, adding to and confirming existing evidence that HPV 16 is aetiologically dominant in the pathogenesis of cervical cancer [27]. The ability to isolate the associated HPV genotype using LCM-PCR allows us to better understand the natural history of specific HPV genotypes, as well as their associated lesions. This will be highly beneficial in the fields of vaccine development, both preventative and therapeutic, as well as accurately determining the outcome of vaccine programs. LCM is a more specific methodology to assign causality and therefore a better assessment of HPV vaccine effectiveness, as well as efficacy. In the phase 3 clinical trials for the bivalent vaccine as well as the quadrivalent vaccine, mixed infections – whether they be determined by swabs or by biopsies – were not uncommon [28]. Accordingly, several different methods were used to attribute causality. In the phase 3 clinical trial of the bivalent vaccine, causality was assigned based on the presence of oncogenic HPV genotypes in cytological samples preceding the development of CIN; the HPV type assignment algorithm [29]. Alternatively, in a study examining HPV genotype-specific infection in relation to vulval intraepithelial neoplasia, causality was assigned based on statistical adjustments according to genotype prevalence data (proportional adjustment approach) [30]. Either methodology could potentially lead to misidentification of HPV genotypes associated with specific CIN lesions. Specifically, within this report there are
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
6
two cases where the associated HPV type would be misidentified using the proportional adjustment approach. Case 3 would be assigned HPV 16, instead of HPV 18. Similarly, case 9 would have been assigned HPV 16, instead of HPV 52. Whilst the HPV type assignment algorithm and proportional adjustment methods have been used in the past, these methods are not definitive in case assignment. The use of LCM to measure outcomes of vaccination would be highly beneficial as misclassification of associated HPV genotypes could be avoided in cases where multiple HPV types are present. In conclusion, LCM-PCR is the preferred methodology to attribute HPV genotype to specific cervical lesions where multiple HPV genotypes are detected. It provides the ability to distinguish between lesion-associated HPV infection and incidental HPV DNA. The use of LCM in the future will allow us to better understand the natural history of specific HPV genotypes and the interactions of multiple HPV infections within the cervix. LCM-PCR will also allow investigation of the relationship between HPV infection and disease progression in other forms of anogenital disease such as anal cancer. Such knowledge would greatly improve diagnosis and subsequent treatment of patients with persistent HPV infection and resultant neoplasia, as well as improve assessment of vaccine effectiveness. VACCINE Study Group Callegari ET, Tabrizi SN, Pyman J, Saville M, Cornall AM, Brotherton JML, Pitts M (La Trobe University, Australian Research Centre for Sex, Health & Society), Gertig DM (VCS Incorporated; School of Population and Global Health, University of Melbourne), Wark JD (Department of Medicine, Bone and Mineral Service, The University of Melbourne, The Royal Melbourne Hospital), Jayasinghe Y (Department of Obstetrics and Gynaecology, The University of Melbourne; Department of Gynaecology, Royal Children’s Hospital), Tan J (Department of Oncology & Dysplasia, Royal Women’s Hospital), Wrede CD (Department of Oncology & Dysplasia, Royal Women’s Hospital), and Garland SM. Conflict of interest statement SMG has received advisory board fees and grant support from bioCSL and GlaxoSmithKline Australia, and lecture fees from Merck, GSK and Sanofi Pasteur; in addition, SMG has received funding to conduct HPV vaccine studies for MSD and GSK. SMG is a member of the Merck Global Advisory Board as well as the Merck Scientific Advisory Committee for HPV. JMLB, DMG, MS and SMG were partner investigators on an Australian Research Council Linkage Grant 2008–2011 on which bioCSL Biotherapies was a partner organisation. Author contributions ETC was involved in data collection, and is the primary author of this manuscript. SMG, chief investigator of the study, conceived the study and was involved in study design, study coordination and helped to draft this manuscript. SNT has overseen all molecular HPV laboratory aspects of the study, including study design and data collection and helped to draft this manuscript. AMC was involved in overseeing molecular HPV laboratory aspects and data collection and helped to draft this manuscript. MS, JP were involved in study design, planning and co-ordination of histology processing and pathologic review of all study cases for sub study B. JT was involved in the study design for sub study B. JMLB, MP, DG, JDW, YJ were involved in study design. DCW is involved in the coordination of sub study B. All authors read and approved this manuscript.
Acknowledgements We grateful acknowledge funding from the Victorian Cancer Agency (VCA). We thank the Anatomical Pathology Department at Royal Melbourne Hospital for providing slide-imaging services and Elizabeth McKinnon (Department of Anatomical Pathology, Royal Children’s Hospital, Melbourne) and Eileen Tan (VCS) who sectioned the tissue blocks and prepared the slides. We thank Tania Tabone for assistance in sub study B set up, Elisa Young for project co-ordination and Houda Abdo, Angela Hurley, Fiona Tan and for sample preparation and genotype analysis. We acknowledge the initial assistance and advice of LCM methods from Wim H. Quint.
References [1] Ferlay J, Shin H, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127:2893–917. [2] Parkin DM, Bray F. Chapter 2: The burden of HPV-related cancers. Vaccine 2006;24:S11–25. [3] Garland SM, Tabrizi SN. H.P.V. D.N.A. testing: potential clinical applications. Int J Health Promot Educ 2006;44:107–12. [4] Markowitz L, Dunne EF, Saraiya M, Lawson HW, Chesson H, Unger ER. Quadrivalent human papillomavirus vaccine. Morb Mortal Wkly Rep 2007;56:1–23. [5] Burchell AN, Winer RL, de Sanjose S, Franco EL. Chapter 6: Epidemiology and transmission dynamics of genital HPV infection. Vaccine 2006;24S3::52–61. [6] Winer RL, Lee S-K, Hughes JP, Adam DE, Kiviat NB, Joutsky LA. Genital human papillomavirus infection: incidence and risk factors in a cohort of female university students. Am J Epidemiol 2003;157:218–26. [7] Moscicki AB. HPV infections in adolescents. Dis Markers 2007;23:229–34. [8] Baseman JG, Koutsky LA. The epidemiology of human papillomavirus infections. J Clin Virol 2005;32:16. [9] Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, Snijders PJF, Peto J, Meijer CJL, Munoz N. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:12–9. [10] Galani E, Christodoulou C. Human papilloma viruses and cancer in the postvaccine era. Clin Microbiol Infect 2009;15:977–81. [11] Moscicki AB, Schiffman M, Kjaer S, Villa LL. Updating the natural history of HPV and anogenital cancer. Vaccine 2006;24:S42–51. [12] Thomas KK, Hughes JP, Kuypers JM, Kiviat NB, Lee S-K, Adam DE, Koutsky LA. Concurrent and sequential acquisition of different genital human papillomavirus types. J Infect Dis 2000;182:1097. [13] Cuschieri KS, Cubie HA, Whitley MW, Seagar AL, Arends MJ, Moore C, et al. Multiple high risk HPV infections are common in cervical neoplasia and young women in a cervical screening population. J Clin Pathol 2004;57:68–72. [14] Garland SM, Brotherton JML, Condon JR, McIntyre PB, Stevens MP, Smith DW, Tabrizi SN. Human papillomavirus prevalence among indigenous and nonindigenous Australian women prior to a national HPV vaccination program. BMC Med 2011;9:104. [15] Stevens MP, Tabrizi SN, Quinn MA, Garland SM. Human papillomavirus genotype prevalence in cervical biopsies from women diagnosed with cervical intraepithelial neoplasia or cervical cancer in Melbourne, Australia. Int J Gynecol Cancer 2006;16:1017–24. [16] Quint W, Jenkins D, Molijn A, Struijk L, van de Sandt M, Doorbar J, Mols J, Van Hoof C, Hardt K, Struyf F, Colau B. One virus, one lesion—individual components of CIN lesions contain a specific HPV type. J Pathol 2011;227:62–71. [17] Kjaer SK, De Villiers EM, Ca˘glayan H, Svare E, Haugaard BJ, Engholm G, Christensen RB, Møller KA, Poll P, Jensen H. Human papillomavirus, herpes simplex virus and other potential risk factors for cervical cancer in a high-risk area (Greenland) and a low-risk area (Denmark)—a second look. Br J Cancer 1993;67:830. [18] Curran S, McKay JA, McLeod HL, Murray GI. Laser capture microscopy. Mol Pathol 2000;53:64–8. [19] Emmert Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA. Laser capture microdissection. Science 1996;274:998–1001. [20] Young EJ, Tabrizi SN, Brotherton JML, Wark JD, Payman J, Saville M, Wrede DC, Jayasinghe Y, Tan J, Gertig DM, Pitts M, Garland SM. Measuring effectiveness of the cervical cancer vaccine in an Australian setting (the VACCINE study). BMC Cancer 2013;13:296. [21] Cornall AM, Roberts JM, Garland SM, Hillman RJ, Grulich AE, Tabrizi SN. Anal and perianal squamous carcinomas and high-grade intraepithelial lesions exclusively associated with “low-risk” HPV genotypes 6 and 11. Int J Cancer 2013. [22] Jenkins D. Histopathology and cytopathology of cervical cancer. Dis Markers 2007;23:199. [23] Espina V, Wulfkuhle JD, Calvert VS, VanMeter A, Zhou W, Coukos G, Geho DH, Petricoin EF, Liotta LA. Laser-capture microdissection. Nat Protoc 2006;1:586–603. [24] Quint WGV, Scholte G, van Doorn LJ, Kleter B, Smits PHM, Lindeman J. Comparative analysis of human papillomavirus infections in cervical scrapes and biopsy specimens by general SPF10 PCR and HPV genotyping. J Pathol 2001;194:51.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
G Model JVAC-15415; No. of Pages 7
ARTICLE IN PRESS E.T. Callegari et al. / Vaccine xxx (2014) xxx–xxx
[25] Kleter B, Van Doorn L-J, Schrauwen L, Molijn A, Sastrowijoto S, ter Schegget J, et al. Development and clinical evaluation of a highly sensitive PCR-reverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. J Clin Microbiol 1999;37:2508–17. [26] Kalantari M, Garcia-Carranca A, Morales-Vazquez CD, Zuna R, Montiel DP, Calleja-Macias IE, Johansson B, Andersson S, Bernard HU. Laser capture microdissection of cervical human papillomavirus infections: copy number of the virus in cancerous and normal tissue and heterogeneous DNA methylation. Virology 2009;390:261–7. [27] van der Marel J, Quint WGV, Schiffman M, van de Sandt MM, Zuna RE, Terence Dunn S, Smith K, Mathews CA, Gold MA, Walker J, Wentzensen N. Molecular mapping of high-grade cervical intraepithelial neoplasia shows etiological dominance of HPV16. Int J Cancer 2012;131:E946–53.
7
[28] The FUTURE.II Study Group. Quadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions. N Engl J Med 2007;356:1915–27. [29] Paavonen J, Jenkins D, Bosch FX, Naud P, Salmerón J, Wheeler CM, Chow S-N, Apter DL, Kitchener HC, Castellsague X, de Carvalho NS, Skinner SR, Harper DM, Hedrick JA, Jaisamrarn U, Limson GAM, Dionne M, Quint W, Spiessens B, Peeters P, Struyf F, Wieting SL, Lehtinen MO, Dubin G. Efficacy of a prophylactic adjuvanted bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a phase III double-blind, randomised controlled trial. Lancet 2007;369:2161–70. [30] Garland SM, Insinga RP, Sings HL, Haupt RM, Joura EA. Human papillomavirus infections and vulvar disease development. Cancer Epidemiol Biomarkers Prev 2009;18:1777–84.
Please cite this article in press as: Callegari ET, et al. How best to interpret mixed human papillomavirus genotypes in high-grade cervical intraepithelial neoplasia lesions. Vaccine (2014), http://dx.doi.org/10.1016/j.vaccine.2014.05.041
