Human Papillomavirus Primary Screening and Collaboration in Quality Assurance Work between Laboratories

Cytology-based screening for cervical cancer has been widely used and has been successful in reducing the incidence of invasive cervical cancer. There is now a fast change in screening methodology with human papillomavirus (HPV)-testing replacing cytology as the primary test [1]. This change from primary cytology to primary HPV testing is not only a question of one method replacing the other but also a question of one strategy replacing the other. While screening with cytology depended on identifying defined lesions, e.g., HSIL and treating them to prevent cervical cancer, HPV testing relies on identifying virus with more or less risk of cervical cancer in the future with different follow-up and treatment alternatives for high- and low-risk HPV types.

The organization of the screening program differs from country to country. The introduction of HPV screening has also been dealt with differently. Some countries have kept the existing screening organization and step-wise introduced HPV testing. In Sweden, for example, this means that the decentralized model with over 20 screening laboratories has been kept [2], while, for example, the Netherlands restructured screening and a more centralized organization has been chosen [3]. There will also be differences between countries and within countries whether the HPV laboratory forms a separate screening unit, separate microbiology unit or is part of a department of pathology or cytology.

Quality assurance work in cytology has traditionally been laboratory based and focused on distribution of diagnoses, individual screening profiles, and follow-up of false negative slides. Differences between laboratories have been easy to identify but sometimes difficult to interpret and handle. Detailed knowledge of quality assurance work has been necessary for cytopathologists and pathologists compiling annual follow-ups. The introduction of HPV testing places high demands on validation of the test itself and on follow-up of screening. Responsibility for HPV follow-up will differ depending on the organization of screening. Regardless of the organization, it is important that cytopathologists, pathologists, and cytotechnicians understand how quality assurance with HPV is organized and works [4], how it is performed and how results should be interpreted and communicated, sometimes in conjunction with cytology and histology. This has been emphasized in the ESGO-EFC position article on cervical screening [5]. A recent, extensive, and detailed overview of quality assurance in HPV screening has been given by Cuschieri et al. [6]. Equally important is the knowledge of which demands could be made on HPV analysis and HPV laboratories, e.g., when contracting an external HPV laboratory.

Integrating HPV analysis with morphology is important for cervical screening and cervical pathology and increasingly so for other organs as well. HPV status is becoming more and more important in the classification of squamous cell carcinoma metastases in the neck and in the differential diagnosis of benign squamous cysts [7]. HPV is the also main cause of precancerous lesions and cancer in the anal region [8] and of more than one-third of the carcinomas in the vulva [9]. Also, immunocompromised women and men are at higher risk for persistent HPV infections and cancers [10]. Important concepts for understanding HPV analysis and quality assurance work are extended genotyping, external quality assessment/proficiency panels, screening registries and population-based cancer registries, internal quality control/laboratory audits, and multidisciplinary conferences, HPV platforms, and biobanking.

The risk of cancer between various types of HPV differs markedly [11] and must be known to all cytologists and pathologists taking part in diagnostic work with cytological samples and histological biopsies from the cervix. These different cancer risks are the underlying concept of partial genotyping, where HPV is classified as high risk oncogenic (HPV16/18) and other oncogenic (not HPV16/18), and extended genotyping of HPV, where most of the HPV types with notable cancer risks are typed individually. The highest risks are found for HPV16 and 18, followed by HPV45. HPV types 31, 33, 52, and 58 also cause >2% of the cancer cases and HPV35 about 1.5% of the cancers (“medium risk viruses”), whereas the other formally oncogenic viruses cause <1% of the cancers (low-oncogenicity viruses). Some HPV tests also test for HPV types that are not established to be oncogenic, such as HPV66 and 68. Genotyping, particularly extended genotyping, allows for a better risk stratification of HPV-positive women in HPV-based screening [12‒14] and this knowledge is fundamental for understanding recommendations in national screening guidelines. For example, should women positive for HPV16/18 in screening be triaged with cytology or referred to colposkopy directly due to the higher risk of severe dysplasias and cancer in this group compared to women positive for medium or low-oncogenicity types. Also, in Sweden, for example, women under the age of 30 with a histological diagnosis of HSIL should always be graded as CIN2 or CIN3 [2, 15] to allow for active expectancy instead of treatment of women with CIN2. This could be particularly important for the management of middle- or low-oncogenic types as cancers caused by these types are rare below 30 years of age.

The positive predictive value of HPV-positive samples for HSIL, AIS and cancer on histological biopsies should be evaluated on a yearly basis. In connection to this, an important question for pathology departments would be if immunohistochemistry with p16 or other markers always should be performed when HPV is detected.

Important implications in quality assurance work and follow-up at cytology departments could be how the performance of cytology is affected by the knowledge of HPV status. Better inter-observer agreement and an improved detection rate of severe lesions have been shown to be the benefits of information about HPV status [14, 16‒18]. However, there might be differences in the interpretation of both squamous and glandular changes depending on whether the virus is of high-, medium-, or low-oncogenic type, although this has not been properly studied.

The International Human Papilloma Reference Center (IHRC) [19] was established in Heidelberg in 1985 and, in 2012, transferred to the Karolinska Institutet. Since 2008, the center has conducted global HPV proficiency studies [20‒27] to evaluate the sensitivity, type specificity and proficiency of HPV typing. The proficiency panels (PPs) were initially designed for vaccinology and HPV typing, but since 2022, there is also a panel better meeting the demands of laboratories analyzing HPV-based screening samples [28].

While taking part in proficiency studies should be a prerequisite for HPV-reference laboratories and HPV laboratories analyzing screening samples [6], cytopathologists and other health professionals involved in cervical screening must be aware of the importance of PPs [5]. Results from proficiency panels should always be paid attention to and included or mentioned in quality assurance summaries. The definition of an optimal result must be known and also what measures should be taken in case of suboptimal performance. This is especially important in organizations where the HPV laboratory is not immediately connected to the cytopathology laboratory. Communication of results as good or less good is an important step in creating understanding between different specialties and laboratories.

High-quality registries of vaccination, screening, cancer diagnosis, and mortality are a prerequisite for a successful quality assurance work and research [29]. Screening registries, vaccination registries, and laboratory information systems are powerful tools in the yearly evaluation of screening programs [30, 31] and inter- and intra-laboratory follow-ups. Population-based cancer registries (PBCR) are important tools to estimate cancer burden and identify changes in cancer incidence and mortality. When screening registries are linked to PBCR the short- and long-term impact of screening and vaccination can be monitored [32]. However, the quality of PBCR differs between countries and is lower in some countries with an estimated high incidence of cervical cancer [29]. In all laboratories, pathologists and cytopathologists must have a good knowledge of which data are reported and how data will have impact on the quality of the register, and this knowledge should nowadays also include HPV.

The use of uniform and standardized coding not only for morphology but also for HPV types and findings on a national level is necessary [2]. Cytopathologists with an interest and knowledge in coding have an important role in the interpretation of data when morphological data are linked to HPV findings and to cancer. They often have first-hand knowledge of local deviations from recommended coding and data exporting and direct communication between national registries, HPV laboratories, and departments of pathology and cytology is necessary [30, 31].

Follow-up of negative cytology before histologically confirmed HSIL, AIS, or cancer – laboratory audit – has, for decades, been an integral part of quality assurance work in many cytology and pathology laboratories [33, 34]. In Sweden, for example, it has been one of the cornerstones in the accreditation of cytopathology laboratories [34, 35].

Laboratory audits can be used as an estimation of the clinical sensitivity of detecting high-grade lesions in the screening program. The basic principle in laboratory audits in cytology is the re-review of benign cytology. When findings in rescreened samples are compared with findings in biopsies, they will also give an important opportunity of education, training, and personal feedback.

The same principles used in laboratory audits for cervical cytology should be applied to HPV-based screening to monitor sensitivity [36‒38]. The basic principle in these audits is a stepwise HPV reanalysis of biobanked liquid-based cytology (LBC) and biopsies taken before cancer or HSIL. Whether reanalysis should be performed in-house or in a different laboratory may differ between settings, but the use of a reference laboratory has been shown to improve detection rate of HPV in “negative” samples taken before HSIL or cancer (unpublished data). As with audits in cytology, negative samples before cancer and samples before HSIL and AIS can and should be assessed separately. Primary HPV testing should be positive in >95% of samples before HSIL and in >90% of samples taken before invasive cancer [36‒38]. Systematic and repeated laboratory audits will not only monitor the sensitivity of the programs. They can also serve the purpose of showing which HPV types continue to cause dysplasias and cancer, the emergence of new types, and differences between HPV platforms.

An important task for cytopathologists and pathologists in both types of laboratory audits will be to review the identified HPV-negative HSIL, AIS or cancer samples to scrutinize the inclusion criteria. Even in high-quality screening registries and laboratory information systems, incorrect coding of morphology and HPV occurs and the morphologic difference between, e.g., LSIL and HSIL [39] or between AIS and adenocarcinoma is not always straightforward. This is well known by cytopathologists but not always by HPV laboratories or other health professionals. Additional benefits from laboratory audits could, therefore, be the detection of miscoding or of problematic areas in histopathology [38]. The complete work-up of a laboratory audit should include other possible causes of HPV negativity apart from pathological issues, e.g., sample quality or assay quality, before classifying samples as truly HPV negative [38].

Multidisciplinary team conferences are an important opportunity to discuss difficult or complicated cases, e.g., divergent, or surprising findings in screening on follow-up. In this discussion, expertise not only in cytology, histopathology, and oncology should be included but also expert skill in colposcopy, management of cervical dysplasias, and interpretation of HPV analysis [2, 40, 41]. Good examples how HPV results should be interpreted in conjunction with cytology are given by the differences in management of, e.g., younger women with cytologic glandular atypia positive for HPV18, women with postmenopausal bleeding and glandular atypia negative for HPV, or women with HPV-negative, ASCH-positive test of cure after adenocarcinoma in situ. These examples all show that cytology and histopathology nowadays benefit from being interpreted together with HPV results. Cytologists and pathologists must be aware of that and acquire knowledge of HPV interpretation. The composition of the multidisciplinary group will differ from hospital to hospital depending on the size of the hospital or its level of specialization.

Biobanking of HPV samples is vital to establish and maintain quality assurance work, follow-up and research on HPV-based screening and testing [42, 43]. Ideally biobanking should include both LBC samples and self-samples. Close cooperation between cytology departments, HPV laboratories, reference laboratories, and universities is necessary to meet legal, technical, and economic demands and to ensure long-term quality of stored samples [4, 44]. Once again solutions will differ from country to country depending on how HPV screening is organized and who will be responsible for follow-up and quality assurance work. In Sweden, for example, updated screening guidelines [2] states that biobanking of LBC samples and self-samples is mandatory for university hospitals but not for community hospitals. In this way, about two-thirds of the screening population and all regions will be covered. High-quality laboratory audits are dependent on the use of biobanked samples [37] as are follow-up of individual patients.

The technical validation of the HPV test and HPV platform itself is beyond the scope of most cytopathologists, but basic knowledge of the underlying technology of the HPV test is important to understand performance and results. Is the test DNA-based or mRNA-based and is the target the L1-gene or the E6/E7-genes? Will extended genotyping be possible or only partial typing? All available HPV screening tests perform well on LBC specimens and self-collected specimens are at least as well suited for HPV detection [45]. How the HPV test performs on formalin-fixed paraffin-embedded specimens (FFPE) will also be of importance when it comes to lab-audits and re-evaluation of possible HPV-negative HSIL and cancers on biopsies and cones [46‒49].

The rapid and continuing introduction of HPV testing as a primary screening tool for the prevention of cervical cancer places high demands on all parties engaged in screening and diagnostics. Cytopathologists, cytotechnicians, and pathologists must be aware of how quality assurance works and follow-up of HPV works and which demands can be made. They should also understand how a successful follow-up of HPV is connected to and mutually dependent on quality assurance work in pathology and cytology. While standardized coding of cytological and pathological diagnoses is necessary for register-based follow-ups of screening programs, it is equally important for successful laboratory audits. Audits also depend on biobanking of both cytological and histological samples. Genotyping of HPV gives important information about HPV types and forms the basis of updated screening programs. It will also influence the way cytology and histopathology are interpreted and reported. Discussions in multidisciplinary team conferences and patient management will also benefit from the information given by the genotyping of HPV. Taking part in external quality programs, e.g., proficiency panels, gives opportunities to monitor and follow the performance of the laboratory and its HPV platform and to compare it with other platforms worldwide. No matter how good or bad the results in each QA-program are, a successful QA-work ends with the interpretation of the outcome and communication. Screening, hospital, and laboratory organization will differ both between and within countries, but results of quality assurance work, follow-ups, and possible measures should always be communicated and explained to all parts involved. Openness and communication will probably encourage internal and external feedback which is an important source of knowledge in both HPV and cytology and pathology laboratories.

An ethics statement is not applicable because this review is based on published literature.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Drafting of manuscript was done by Henrik Edvardsson and Joakim Dillner. Both authors contributed to revision of manuscript for important intellectual content and approved of the final version.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.