The World Health Organization Reporting System for Lung Cytopathology

The International Academy of Cytology (IAC) has joined with the International Agency for Research on Cancer (IARC) in developing and publishing four international systems for reporting lung, pancreaticobiliary, lymph node, and soft tissue cytopathology. The aims of this project are to improve and standardize the reporting of cytopathology, facilitate communication between cytopathologists and clinicians, improve patient care, synthesize existing evidence, and highlight areas requiring further research. The process of selection of authors, writing and review of the literature, were similar to those used for the WHO Classification of Tumours (the “Blue Books”). In fact, one of the goals of this system is to provide cytopathological correlates with the entities described in the WHO Classification of Tumours, thereby presenting an international approach for reporting cytopathology to mirror the WHO Classification of Tumours, with links particularly on the website between the two series. In addition, information obtained by an international survey promoted by the IAC was taken into account in the final review of the system. This article is a summary of the proposed system for lung cytopathology.

The importance of cytopathology in the investigation of respiratory conditions has been recognized since the earliest days of clinical cytopathology, and respiratory tract cytopathology specimens are established as a vital diagnostic procedure in the evaluation of patients with suspected lung inflammatory/infectious or neoplastic diseases. Lung cancer is the second most common malignancy and the leading cause of cancer-related mortality [1]. In patients with suspected lung cancer, a timely and precise diagnosis is essential to ensure adequate treatment. The study of sputum, bronchial washings (BW), bronchial brushings (BB), bronchoalveolar lavage (BAL) specimens, and fine needle aspirate biopsies (FNAB) provides the cytomorphological basis of the diagnosis. As in other cytopathological fields, the lack of a consistent and standardized system of reporting has hampered communication between disciplines managing these patients.

In 2016, the Papanicolaou Society of Cytopathology (PSC) proposed guidelines for pulmonary cytopathology based on a multidisciplinary approach. This is a 6-tiered reporting system incorporating “Non-diagnostic,” “negative for malignancy,” “Atypical,” “neoplastic, benign neoplasm, low-grade carcinoma,” “Suspicious for malignancy” and “Malignant” categories [2, 3]. The most contentious categories of the guideline are “neoplastic, benign neoplasm, low-grade carcinoma,” and “Non-diagnostic” [3]. In 2020, the Japan Lung Cancer Society and Japanese Society of Clinical Cytology proposed a new four-tiered cytopathology reporting system for lung carcinoma with the following categories: “negative for malignancy,” “Atypical cells,” “Suspicious for malignancy,” and “malignancy” [4]. In this system, inadequate cases are not categorized because they are excluded in an initial step and only adequate cases are categorized. Although these two proposals had merit in that they attempted to systematize the nomenclature and link the categories with management, neither is used globally or was developed internationally and adapted for all conditions of cytopathology practice.

The aims of the WHO Reporting System for Lung Cytopathology (WHO System) include establishing for the first time a system that can be used internationally in all medical infrastructure settings and provides options for diagnostic management that recognize the variation in availability of ancillary diagnostic and prognostic testing modalities in low- and middle-income countries. The WHO System emphasizes the importance of the cell preparation techniques required to optimize quality and enhance cytopathological diagnosis. Management of the different cytopathological specimens and use of immunocytochemistry (ICC), in situ hybridization and molecular techniques are discussed, since these are extremely important in the rapidly developing era of targeted therapy in lung cancer [5].

The WHO System has five categories that can be stratified by their risk of malignancy (ROM):

• Insufficient/Inadequate/Non-diagnostic

• Benign

• Atypical

• Suspicious for malignancy

• Malignant

The standardized structured report should state one of these five descriptive terms as a heading. A laboratory and its cytopathologists should select one of the terms, “Insufficient/Inadequate/Non-diagnostic,” and use this term consistently. The structured report headed by a category term can then include a brief cytopathological description noting where possible the presence or absence of key diagnostic features, which will be detailed in the forthcoming book. In cases with a specific diagnosis, such as “lung adenocarcinoma,” some cytopathologists may simply provide the diagnosis with minimal description. This is followed by a conclusion or summary in which the cytopathologist should give as specific a diagnosis of the lesion as possible, such as squamous cell carcinoma, or, if the diagnosis is uncertain, provide the most likely differential diagnoses. A working group consisting of members of the IAC and ICCR is establishing a minimum data set of core and noncore components to be included in the reports of lung malignancies diagnosed by small biopsies and cytopathology specimens.

There are few published papers showing the ROM for each of the WHO System categories and most of them have use previous nomenclature system [2, 3, 6]. Therefore, the ROM provided in this first edition need to be refined by future research by cytopathologists interested in lung cytopathology. The categories are linked with recommendations as to the further workup and diagnostic management, which are dependent of the availability of local practices and medical resources (Table 1).

An international web-based survey was developed by the IAC in consultation with the Lung Expert Editorial Board to establish a snapshot of current lung cytopathology usage from the international community of cytopathology. The survey was based on an earlier survey developed and utilized to assist the authors of the IAC Yokohama System for Reporting Breast FNAB Cytopathology {33369266}. It also included questions related to the proposed lung reporting system. Participants included pathologists and cytotechnologists. The survey demonstrated a diverse practice among various laboratories and countries using a wide spectrum of techniques for specimen processing and handling and various reporting systems. The information obtained assisted the editors and authors to develop the system.

A specimen categorized as “Insufficient/Inadequate/Non-diagnostic” lacks sufficient material in quantity or quality for reliable diagnosis.

In respiratory cytopathology, specimen adequacy plays a pivotal role in the interpretation and rendering of a diagnosis, the application of ancillary studies and in clinical decision-making. Specimen adequacy is defined in relation to specimen type, medical history, clinical symptoms and imaging findings providing a “triple test” approach [2, 7]. Terms such as Insufficient, Inadequate, and Non-diagnostic have been used interchangeably to describe this category but, until recently, sputum was the only specimen type with relatively well-defined adequacy criteria [8, 9]. The PSC guidelines [2] and the recently revised Cytology Reporting System for Lung Cancer from the Japan Lung Cancer Society and Japanese Society of Clinical Cytology [4] published definitions of an inadequate sample. These reflect the different modalities available for cytopathological diagnosis of pulmonary lesions, each of which strictly requires a different definition of what is “satisfactory” or “adequate.” The PSC guidelines include the category “Non-diagnostic” for a “specimen which provides no useful diagnostic information about the pulmonary nodule, cyst, or mass lesion identified by imaging findings.” The Japanese system utilizes a stepwise approach with an initial assessment of “Adequate,” which is defined as “adequate for cytological evaluation” and clarified as the presence of abundant cellular material, or “Inadequate” where there is no abundant cellular material or obscuring artefact is present, but both of these systems can have limitations in interpretation [10].

In the WHO System the “Insufficient/Inadequate/Non-diagnostic” category is used for cases where there is insufficient material due to low cellularity, poor preparation, fixation or staining, and obscuring by blood, inflammatory or other material. The term “Non-diagnostic” has been used by some cytopathologists for this category to include not only these cases of insufficient material due to technical causes, but also to include cases where there is considerable benign material on the slides, but the material appears to not be representative of a mass lesion or lung nodule seen on imaging. In this situation, an alternative approach is to categorize what is seen on the slides as “Benign” and add a caveat to the report, “that the material may not represent the lesion seen on imaging.” Either approach is acceptable, and an institution or cytopathology service should select one term, “Insufficient,” “Inadequate,” or “Non-diagnostic” and apply it routinely. In order to avoid confusion, it is not recommended to use “Insufficient” or “Inadequate” for cases with technically insufficient material, and “Non-diagnostic” for cases with abundant benign material but a mass lesion on imaging.

The reasons for an “Insufficient/Inadequate/Non-diagnostic” specimen should be documented in the report. If there are any Atypical cells however scant, even if the slides are otherwise insufficient, the case is immediately regarded as “Atypical” and not “Insufficient/Inadequate/Non-diagnostic.”

In order to recognize an “Insufficient/Inadequate/Non-diagnostic” cytopathological specimen, the features of an adequate specimen should be established. Specific features to determine adequacy in each of the following specimen types are listed, and it is noted that none require a specified number of cells or cell types. FNAB of lung should include some alveolar, that is, pulmonary, macrophages which generally contain pigment that can be carbon and/or hemosiderin, and may include tissue fragments of collapsed alveolar septa. Transbronchial ultrasound-guided needle aspiration (EBUS-TBNA), when a lymph node has been targeted, should contain moderate to abundant lymphocytes or a large number of anthracotic pigment-laden macrophages. More specifically more than 40 lymphocytes in a high-power field in the area of highest cellularity have been proposed to define an adequate sample, and these should be seen generally in greater numbers than in a CT-guided transthoracic approach [11, 12]. Sputum samples should include at least a few alveolar macrophages and ciliated columnar cells on smears or in a LBC preparation [8, 9]. BB should include abundant bronchial epithelial cells and macrophages may be present. BW and BAL should include readily identified alveolar macrophages and a BAL should have more than 10 alveolar macrophages per high-power field [13]. Bronchial epithelial cells may be present but should not exceed the number of alveolar macrophages.

“Insufficient/Inadequate/Non-diagnostic” rates and the diagnostic yield of lung cytopathology depend on the sampling method and in particular during FNAB on the availability of rapid on-site evaluation (ROSE) [14]. There are considerable variations in the literature concerning “Insufficient/Inadequate/Non-diagnostic” rates. One recent study based on the PSC system categorizing different lung specimen types showed an incidence of 16% [3]. The “Insufficient/Inadequate/Non-diagnostic” rate can be reduced by better initial training of the operator, greater caseload to build the experience of both the operator and the cytopathologist, use of image guidance and immediate feedback on the adequacy and quality of the specimen through ROSE.

Overall, the “Insufficient/Inadequate/Non-diagnostic” category harbours a ROM of approximately 40–60% [2, 3], depending on the mode of sampling and imaging characteristics of a lung mass [15]. Reporting the reasons for inadequacy in a specimen can contribute to improving the diagnostic yield of a repeat examination [15], for example, to perform an adequate “deep cough” to produce sputum after an initial inadequate predominantly buccal sample. In the case of a BB of an endobronchial mass that yields only normal bronchial cells due to its submucosal localization or the fibrotic nature of the tumour, the approach can be changed and an EBUS-TBNA can be performed. After unsuccessful noninvasive or minimally invasive procedures, such as sputum, BW and BB, FNAB, or transthoracic CT-guided FNAB can be utilized. Mediastinoscopy, if available, may be appropriate. In principle, a lung mass more than 30 mm in diameter is considered Malignant [15], and if an “Insufficient/Inadequate/Non-diagnostic” result is obtained after several attempts, then clinical, imaging, and pathological correlation is required to establish further diagnostic management.

A specimen categorized as “Benign” demonstrates unequivocal cytopathological features which may or may not be diagnostic of a specific process or benign neoplasm.

The “Benign” category includes inflammatory processes and benign neoplasms found in all types of samples of respiratory cytopathology. The category includes those cases where the material is diagnostic of a specific process, such as suppurative or granulomatous inflammation (Fig. 1), or a specific tumour such as a pulmonary hamartoma, as well as, those cases where the normal components of lung tissue, including bronchial cells, alveolar macrophages, alveolar septa and Type 2 pneumocytes are found. Correlation with imaging is required wherever available. If the cytopathological findings do not correlate with the imaging, which may be indeterminate or suspicious, then this should be clearly stated as a caveat in the report and particularly in its conclusion, “that the cytopathological material may not represent the lesion seen on imaging.” Recommendations for further diagnostic workup should be given. Tumours that may have an uncertain Malignant potential, but whose cytopathological features are included in the differential diagnosis of benign tumours and whose specific diagnosis can be made with ancillary testing, are included in the benign chapter with a clear discussion of their possible aggressive course, for example, solitary fibrous tumour.

Rates of benign diagnoses vary between different practices and cytopathology specimen types. In one recent series, approximately 50% of the cases were placed in this category [3]. The ROM is reported as in the range from 20 to 40% [2, 3, 6].

Whenever possible, the categorization as “Benign” should be qualified by as specific a diagnosis as possible. The final diagnosis should be established in the context of the “Triple Test” with correlation of the cytopathology findings with the clinical and imaging presentation [16]. This is especially important in the benign neoplasms such as hamartoma where the material may resemble normal lung elements. If a specific diagnosis is not established, further evaluation such as repeat core needle biopsy (CNB) or in some cases, if clinically appropriate, limited resection should be performed.

A specimen categorized as “Atypical” demonstrates features predominantly seen in benign lesions and minimal features that may raise the possibility of a Malignant lesion, but with insufficient features either in number or quality to diagnose a benign or Malignant process or lesion.

The main causes of an “Atypical” categorization include intrinsic characteristics of the targeted lesion, the expertise of the operator, technical issues related to obtaining and preparing the material, and the experience of the pathologist who is interpreting the specimen [17, 18]. All cellular elements of the respiratory tract including squamous cells lining the buccal mucosa and oropharynx, metaplastic squamous cells, respiratory columnar cells, terminal bronchiolar, or alveolar lining cells and pulmonary macrophages can demonstrate architectural and cellular atypia [19, 20]. Squamous carcinoma precursor lesions of the airways including low, moderate, and high-grade dysplasia and carcinoma in situ are recognized in the WHO Thoracic Tumours 5th Edition, and although not specifically diagnosable on cytopathology, contribute to cases categorizable as “Atypical.” Reactive changes, such as metaplasia and hyperplasia, infections particularly viral and post-therapy changes, are commonly associated with an “Atypical” categorization (Fig. 2). It is recommended clinical and imaging findings are reviewed before categorizing a specimen as “Atypical” [5, 18]. Imaging studies can help differentiate between a localized and a diffuse and/or bilateral disease process.

In acute respiratory distress syndrome, a chest radiograph will show bilateral alveolar infiltrates and no evidence of cardiomegaly [21]. Lung malignancy, especially adenocarcinoma, can show a spectrum of changes on CT. These include ground glass that is nonsolid, and part solid nodules, solid pulmonary nodules or masses, cystic lucencies, and multifocal nodules. Lesions may demonstrate calcifications and cavitation [22]. Typically, small cell carcinoma and lymphoma present as extensive mediastinal lymphadenopathy without a dominant pulmonary lesion [22]. Rates of the “Atypical” category vary between different practices and cytopathology specimen types. In the published series, approximately 3–5% of the cases were placed in this category [3, 23]. The ROM is reported as in the range of 50–60%, which is regarded as far from ideal and too high, but there are few published studies [2, 3, 6].

Correlation with imaging and clinical findings is required. Particularly if the imaging features are Atypical or concerning for malignancy, further investigation, is warranted by an additional study usually using a different modality: an “Atypical” sputum could be repeated or bronchoscopy with BW/BB could be conducted; an “Atypical” BW/BB could be followed by an EBUS-TBNA with or without a CNB; and an “Atypical” FNAB could be repeated with a core needle biopsy. However, if the imaging and clinical features account for the “Atypia” described, a confirmatory repeat study may be obtained or the patient may be clinically observed for a period of time.

A specimen categorized as “Suspicious for malignancy”’ demonstrates some cytopathological features suggestive of malignancy but with insufficient features either in number or quality to make an unequivocal diagnosis of malignancy. The category “Suspicious for malignancy” is generally used in respiratory cytopathology to indicate a degree of uncertainty regarding the diagnosis of malignancy and offers a risk stratification for a Malignant diagnosis [2, 4, 18]. This category is often used when there is insufficient quality or quantity of cellular material in an adequate specimen for a definite diagnosis of malignancy. The category remains subjective with a high rate of inter-observer disagreement [24], as the threshold for the diagnosis of malignancy depends on the pathologist’s experience, the type of cytopathological preparation and the degree of cellular atypia (Fig. 3). The category is used particularly to avoid a false-positive diagnosis of malignancy, which can lead to unnecessary intervention. The diagnosis of “Suspicious for malignancy” should not be used as the sole basis for therapy [2, 4, 25].

When a case is categorized as “Suspicious for malignancy,” the report should include a statement as to which malignancy or malignancies in a differential diagnosis are suspected, including nonsmall cell carcinoma, neuroendocrine tumours, small and large cell neuroendocrine carcinomas, lymphoma, sarcoma, and metastatic carcinomas. The category can be used when reporting any respiratory cytopathological specimen. Although there are no defined cytopathological criteria for the category of “Suspicious for malignancy,” significant cytopathological atypia is present including nuclear enlargement, anisonucleosis, nuclear crowding, varying chromatin, variability in cell size and shape, and other features associated with malignancy.

ROSE has been shown to reduce the rate of the “Suspicious for malignancy” category [14]. However, the use of the diagnostic category may persist despite utilization of ROSE due to limiting factors related to the intrinsic characteristics of the tumour, the clinical circumstances, and technical aspects. Necrosis, dense inflammatory background, association with a granulomatous reaction, cytopathological features that overlap with benign lesions and tumours with extensive fibrosis are examples of intrinsic tumour characteristics that may produce highly Atypical samples or scant Atypical cellularity independent of the number of passes on ROSE.

In contrast, well-differentiated adenocarcinoma of lung with a lepidic pattern can result in sheets of cells with low nuclear atypia that may be categorized as “Suspicious for malignancy” due to the lack of overtly Malignant characteristics. A prior history of radiotherapy and or chemotherapy may result in a “Suspicious for malignancy” categorization because their reactive atypia may mimic carcinoma or there are a small number of highly Atypical cells. Technical aspects including the tumour size particularly if less than 20 mm, location of the targeted lesion and the expertise of the interventionist performing the biopsy can result in specimens with low cellularity [26‒29].

The use of ancillary techniques such as ICC may assist in changing a “Suspicious for malignancy” to a “Malignant” categorization, such as when metastatic tumours and neuroendocrine tumours are suspected, but generally, the quality and quantity of the suspicious cells may still prevent definitive classification. For example, the finding of crushed cells on FNAB smears and even in a cell block preparation may lead to a suspicion of a differential diagnosis of a neuroendocrine carcinoma or lymphoma, but insufficient tumour cells may prevent an ICC diagnosis [30, 31]. Similarly, when the features are suspicious for a primary pulmonary “nonsmall cell carcinoma,” the thyroid transcription factor-1 (TTF1), a marker of adenocarcinoma, is also expressed in reactive alveolar epithelial cells or in basal cells of the peripheral bronchial tree and does not separate benign from Malignant.

Rates for the category “Suspicious for malignancy” vary between different practices and cytopathology specimen types. In a recent publication, approximately 5% of the cases were placed in this category [3]. The ROM of the “Suspicious for malignancy” category is approximately 82% with a range of 54.5–90% [2, 3, 6], in comparison to the “Atypical” category that carries an estimated ROM of 50–60% [2, 3, 6], demonstrating that the use of the two categories enables risk stratification.

The categorization of a case as “Suspicious for malignancy” is not equivalent to a categorization or final diagnosis of malignancy. It is imperative to have good communication with the clinicians with review of all relevant clinical and imaging information to determine a management plan, in order to avoid unnecessary additional risks and costs. In a minority of cases, surgical management can proceed without a final diagnosis of malignancy if clinically indicated, and in this situation, intraoperative pathological confirmation such as frozen section of the diagnosis is recommended. In most cases, however, if systemic therapy is to be considered, a repeat of the diagnostic cytopathological procedure or more commonly a change to another diagnostic modality such as FNAB or CNB is recommended. Use of ancillary studies has limited value in further classification of these cases. The categories of “Suspicious for malignancy” and “Malignant” are within a spectrum with a high rate of inter-observer variability, and therefore, consultation with a more experienced cytopathologist may be helpful to reach the threshold of malignancy in an otherwise limited specimen.

A specimen classified as “Malignant” demonstrates unequivocal cytopathological features of malignancy.

The “Malignant” category should only be used when there is a full constellation of cytopathological findings and no discrepant features. Wherever possible the neoplasm should be subclassified based on the key diagnostic cytopathological features and ICC, if needed. Malignant neoplasms involving the lungs include both primary and secondary tumours. It is important to be aware of the limitations of cytopathology samples, which can be due to qualitative and quantitative reasons [2, 3, 32]. Therefore, a “Malignant” categorization and specific diagnosis should be rendered only in specimens with adequate cellularity meeting definitive criteria for malignancy [32]. Based on cytomorphology, good accuracy, that is, greater than 70%, can be achieved in differentiating between nonsmall cell carcinoma, including adenocarcinoma and squamous cell carcinoma (Fig. 4, 5), and small cell neuroendocrine carcinoma. However, with the advent of targeted therapies, exact classification of the cancer as adenocarcinoma or squamous cell carcinoma is essential and the standard of care in the current practice of pulmonary cytopathology [33, 34]. A major limitation of pulmonary cytopathology is the higher rate of NSCC-NOS resulting from an inability to subtype as compared to small biopsies and resections [35]. Misclassifications and inability to classify, that is, a final report of NSCC-NOS, on cytopathology are mainly seen in exfoliative cytopathology [36], touch preparations [37], Giemsa-stained smears mainly in cases of SCC [38], samples with low cellularity [36, 39], necrosis [36, 39, 40], poorly differentiated histopathology [36, 39, 41], and in rare instances of large cell carcinoma, sarcomatoid carcinoma or adenosquamous carcinomas [36, 40, 42].

In the case of primary epithelial tumours, differentiation between the commonly encountered adenocarcinoma and squamous cell carcinoma can be achieved using a limited ICC panel consisting of TTF1 and p40 (Fig. 6a, b) [32‒34, 40, 43]. Dual stains such as Napsin A/p40 are especially helpful in preserving tissue in very scant samples [44]. Therefore, within the interpretation category “Malignant,” it is important that the final diagnosis is as specific as possible. It is also important to be aware of the diagnostic pitfalls associated with ICC stains, especially in cases of secondary malignancies of the lung, for example, TTF1 positivity in metastatic thyroid carcinoma [33, 40, 45].

The Malignant category also includes low-grade neuroendocrine tumours, previously known as carcinoid and Atypical carcinoid, and neuroendocrine carcinomas of predominantly small or large cell types (Fig. 7, 8), which can be diagnosed by cytopathology and confirmed by use of ICC markers such as INSM1, chromogranin, and synaptophysin. The other Malignant neoplasms included in this category are salivary gland-type carcinomas, mesenchymal tumours, and secondary malignancies (Fig. 9). Rates of malignancy diagnosis can vary between different institutions and countries. In a recent series, approximately 20% of the cases were placed in this category [3]. The reported ROM for cytopathology specimens categorized as “Malignant” is greater than 90% and in most cases approaches 100% [2, 3, 6, 46].

A cytopathology categorization as “Malignant” should be correlated with the clinical and imaging findings and if it is concordant, surgical management if appropriate can proceed, while if systemic treatment is planned in the more common situation of advanced disease presentation, definitive therapy can be commenced if material is available for prognostic and predictive biomarkers. Cytopathology of Malignant lesions is often highly diagnostic, and in the majority of cases, a large number of cells are present with a high percentage of tumour cells. This enables not only the diagnosis but also ancillary studies including predictive comprehensive molecular profiling using cell blocks and smears. However, if the material is mainly necrotic, all specimens including smears, cell blocks and CNB need to be reviewed for morphologically viable tumour cells to allow for accurate diagnosis and subtyping. Otherwise, a repeat diagnostic procedure is required.

As with all reporting systems involving categorization of cytopathology specimens, the new WHO System is designed to improve communication between clinicians and cytopathologists. Each specimen type and its category have a specific ROM and this will directly influence clinical diagnostic management algorithms. The authors recognize that the performance indicators for each category of the system are derived from recent reviews of the literature and hope that the WHO System will encourage research into these indicators to test the current system and its management recommendations and provide ever more precise ROM.

The WHO System also defines through the first international consensus the key diagnostic cytopathological criteria for each lesion or tumour, which is essential to improve the quality of diagnostic assessment and reporting of lung cytopathology. It also provides a differential diagnosis based on these cytopathological criteria of the different entities empowering cytopathologists throughout the world to use the System. Further, the WHO System also provides the current best practice application of ancillary testing, including ICC and molecular pathology, and, importantly, provides detailed descriptions of sampling and processing techniques to optimize the handling and preparation of the cytopathology sample. The integration of ancillary techniques for detecting therapeutic targets in lung cancer has further enhanced the utility of lung cytopathology, and the emergence of prognostic markers and targeted therapies has required more specific classification of the carcinoma and a multidisciplinary approach to patient management.

The authors of the WHO System recognize that local medical and pathology resources and infrastructure will vary, particularly in low- and middle-income countries. To make the WHO System applicable worldwide, the system is based on cytomorphology and provides options for further diagnostic management of the patient.

The WHO System will provide a direct and dynamic link to the WHO Classification for Thoracic Tumours 5th Edition and will raise the profile and use of cytopathology by increasing awareness of its current role and its potential role in the era of personalized medicine based on molecular pathology utilizing “small biopsies.” Ultimately, the System will improve patient care and outcomes.

The IAC IARC WHO Expert Editorial Board for the WHO System sincerely acknowledge and thank all the authors who contributed to the text, tables and microphotographs of the web-based and book formats. Their commitment to the required literature searches and writing to meet deadlines over the 2-year period the system and the book have been in development is very gratefully acknowledged and made the joint project possible.

The content of this article represents the personal views of the authors and does not represent the views of the authors’ employers and associated institutions. Where authors are identified as personnel of the IARC/World Health Organization, the authors alone are responsible for the views expressed in this article and they do not necessarily represent the decisions, policy, or views of the IARC/World Health Organization.”

There are no funding sources.

Conceptualization: Fernando C. Schmitt, Lukas Bubendorf, Sule Canberk, Ashish Chandra, Ian A Cree, Marianne Engels, Kenzo Hiroshima, Deepali Jain, Ivana Kholová, Lester Layfield, Ravi Mehrotra, Claire W. Michael, Robert Osamura, Martha B. Pitman, Sinchita Roy-Chowdhuri, Yukitoshi Satoh, Paul VanderLaan, Maureen F. Zakowski, and Andrew S. Field. Writing original draft: Fernando C. Schmitt with collaboration of all authors. Final review: Andrew S. Field. All authors read and approved the final manuscript.

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