Challenges of the 5th Edition of the WHO Classification of Thyroid Tumours Open Access

The latest WHO Classification aimed (and succeeded) to extend beyond histopathologic parameters in categorizing thyroid tumours in order to understand the etiopathogenesis of tumours and to advance the adequacy of patients treatment, taking in consideration the new information regarding molecular data and therapy developments based on thorough followed-up cases.

The 5th edition incorporated a newly defined category of “low-risk neoplasms” in an attempt to avoid “overdefinition” of thyroid carcinoma, thus protecting patients from being overtreated. The aforementioned point concerns mainly the substitution of encapsulated noninvasive follicular variant of papillary thyroid carcinoma (PTC) by noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP), but it also encompasses “uncertain malignant potential (UMP) cases” and hyalinizing trabecular tumour (HTT).

Besides the importance of classic staging for prognosis and therapy selection (T, N, M), the 5th edition stresses the need to evaluate precisely the presence and degree of vascular invasion in this context. Additionally, it is also very important to separate lymphatic invasion from venous invasion – ruling out the utilization of lympho-vascular invasion in thyroid pathology.

Two new aspects of the 5th WHO Classification regards, in differentiated carcinomas, the criteria needed to separate papillary carcinomas from follicular carcinomas, turning less important the nuclear features per se, and the creation of the category high-grade differentiated thyroid carcinoma (HGDTC) based upon the presence of necrosis and/or high number of mitoses, or high Ki-67 index. Finally, the 5th edition incorporates genetic (germline and, more frequently, somatic) markers, acknowledging their value for diagnosing rare forms of thyroid cancer and for providing crucial information as predictive and/or prognostic and/or therapeutic biomarkers.

The 5th edition includes recommendations for classifying thyroid tumours into categories (follicular cell-derived and C-cell derived neoplasms), family (class), type and tumour subtype, reserving the term “variant” in reference to genetic alterations in tumours [1]. Other generic recommendations include reporting tumour size in mm, as well as assessing mitotic activity by area in mm² instead of by high-power fields; this last change is facilitated by the incorporation of digital systems and will avoid biases due to the different high-power field sizes of the different microscopes [1].

Within the category of follicular cell-derived neoplasms, tumours have been classified into three families (class): benign tumours, low-risk neoplasms, and malignant neoplasms [1, 2]. The family of benign tumours includes thyroid follicular nodular disease (FND), follicular thyroid adenoma (FA), FA with papillary architecture, and oncocytic adenoma of the thyroid [1, 2]. FND designates multiple nodular proliferations both hyperplastic and neoplastic (clonal) previously referred clinically as multinodular goitre, multinodular hyperplasia, adenomatous goitre, and colloid nodules, but which may also be subclinical. FND presented in childhood or adolescence may be associated with germline mutation of DICER1 (see below): the thyroid gland typically shows multiple follicular cell nodules with variable morphology, FAs with intrafollicular centripetal papillary growth, some nuclear features of PTC (intermediate-type nuclei), and involutional changes in the parenchyma [1‒3]. FA with papillary architecture is a new tumour type characterized by its papillary architecture, lack of nuclear features of PTC and, sometimes, autonomous hyperfunction [1‒3].

The family of low-risk neoplasms encompasses, NIFTP, thyroid tumours of uncertain malignant potential, and HTT [1, 2]. Oncocytic NIFTP, composed of ≥75% oncocytic cells, and subcentimeter NIFTP, when measuring <10 mm in diameter, are the two new NIFTP subtypes incorporated in this edition [1, 2]. In our opinion, because determining infiltrative margins in tiny NIFTPs can be problematic, the alternative diagnosis of sub-centimetric PTC or papillary thyroid microtumour should be considered [4, 5].

HTT is a neoplasm composed of large trabeculae with marked inter-trabecular hyalinization along with nuclear features characterized by prominent grooves, vacuoles, and membrane irregularities like PTC. The diagnosis of HTT can be confirmed by membrane staining with Ki-67 (MIB-1 clone) and determination of GLIS rearrangements (or GLIS protein expression) [1, 2].

In this 2020 WHO edition, the family of malignant follicular cell neoplasms covers follicular thyroid carcinoma (FTC), invasive encapsulated follicular variant of papillary thyroid carcinoma (IEFVPTC), PTC, oncocytic carcinoma of the thyroid (OTC), high-grade follicular cell-derived non-anaplastic thyroid carcinoma (ATC), and ATC [1, 2]. IEFVPTC appears as a new tumour type independent of PTC due to its greater morphological (encapsulation and follicular growth pattern) and molecular (RAS-like) similarity with FTC, with the idea of merging both entities into a single diagnostic category (differentiated thyroid carcinoma with a follicular pattern), in the next edition [1], a reasonable and practical proposal.

PTC is considered a BRAF-like neoplasm that includes some more aggressive histological subtypes such as tall cell-PTC, hobnail-PTC, and columnar cell-PTC. The diagnosis of papillary microcarcinoma has disappeared from the classification and sub-centimetric PTCs must be subtyped in the same way as larger PTCs. Because of the significant percentage of sub-centimetric PTCs that have extremely low malignant potential, particularly if they are incidentally discovered, it remains a challenge to obtain biomarkers that identify clinically aggressive small PTCs. The new diagnostic criteria for the tall cell subtype of PTC require ≥30% of tumour cells to be at least 3 times as tall as they are wide [1, 2]. Although they are not diagnostic of tall cell-PTC subtype, it is possible that with these strict criteria, even a lower percentage of tall cells may be associated with more aggressive behaviour [6, 7]. Outside of the more conventional approaches, the use of deep learning-based algorithm and the detection of somatic mitochondrial DNA (mtDNA) mutations can help in the characterization of tall cell-PTC [8, 9].

The diagnosis of oncocytic adenoma and carcinoma is used to designate follicular cell tumours composed of at least 75% oncocytic cells in which the nuclear features of PTC are absent [1, 2]. The term Hürthle [10] cells should be avoided when referring to oncocytic tumours because what Hürthle [10] described were C cells. FTC, IEFVPTC, and OTC are classified into the following subtypes: (1) minimally invasive (capsular invasion only); (2) encapsulated angioinvasive (with or without capsular invasion); and (3) widely invasive. Tumours with extensive vascular invasion (4 or more foci of venous invasion counted separately) have a worse prognosis than those with limited invasion of vessels [11‒13].

High-grade follicular cell-derived thyroid carcinoma occupies a biological and molecular intermediate position between differentiated carcinomas and ATC [1, 2]. These tumours are diagnosed based on mitotic activity and tumour necrosis. Non-anaplastic high-grade follicular cell-derived carcinomas are subtyped into (1) HGDTC when the distinctive architectural and/or cytologic features of well-differentiated tumour types are preserved; and (2) poorly differentiated thyroid carcinoma (PDTC) when the Turin consensus criteria are met [11]. HGDTC, i.e., high-grade PTC, high-grade FTC, and high-grade OTC, may show poorly differentiated areas but lack anaplastic morphology. Unlike ATC, high-grade follicular cell-derived carcinomas are always immunohistochemically positive for thyroglobulin, TTF1, and PAX8, although some PDTCs may show lower expression of thyroglobulin with a characteristic dot-like pattern [1, 2].

Because ATC and primary thyroid squamous cell carcinoma (TSC) have a similar prognosis, share the BRAF p.V600E mutation [14], and ATC frequently shows areas with squamous differentiation, TSC is now considered a histologic pattern of ATC [1, 2]. Other studies, on the contrary, indicate that TSC has distinctive clinical, pathological and molecular profiles [15].

In the 5th edition, the division into high and low grade is also applied, with similar criteria, to the family of thyroid tumours derived from C cells [1, 2]. The diagnosis of high-grade medullary thyroid carcinoma (MTC) based on the presence of tumour necrosis and/or proliferative activity (≥5 mitoses per 2 mm² and/or Ki-67 index ≥5%), is associated with adverse outcome [16].

The family of salivary gland-type carcinomas of the thyroid includes (a) mucoepidermoid carcinoma, defined according to the criteria of the previous edition; and (b) secretory carcinoma [1, 2]. Primary thyroid mucinous carcinoma is now considered a subtype of mucoepidermoid carcinoma [1, 2]. Secretory carcinoma, formerly known as mammary analogue secretory carcinoma, is a rare tumour negative for thyroglobulin and TTF1 associated with ETV6 translocations (ETV6::NTRK3) [17].

The family of thyroid tumours of uncertain histogenesis covers: (a) sclerosing mucoepidermoid carcinoma with eosinophilia based on morphological criteria without distinctive molecular features; and (b) cribriform-morular thyroid carcinoma (CMTC) [1, 2]. CMTC is a thyroid carcinoma in patients with familial adenomatous polyposis (FAP), although sporadic cases also exist [18, 19]. CMTC is associated with alterations in the Wnt/β-catenin pathway, showing immunohistochemical negativity for thyroglobulin and calcitonin and characteristic nuclear and cytoplasmic positivity for β-catenin [18].

The family of thymic tumours arising within the thyroid includes (a) thymoma; (b) spindle epithelial tumour with thymus-like elements (SETTLE); and (c) intrathyroid thymic carcinoma, whose morphology is not different from that of its mediastinal counterpart [1, 2].

The family of embryonal thyroid neoplasms is represented by thyroblastoma, a high-grade neoplasm reported for the first time in the WHO Classification [1, 2]. Thyroblastoma consists of primitive follicular epithelium positive for thyroglobulin, small cell blastemal component, and mesenchymal stroma, which should be differentiated primarily from teratoma and carcinosarcoma [20, 21].

The latest WHO Classification aims to extend beyond histologic parameters in categorizing thyroid tumours. The 5th edition incorporates genetic markers, acknowledging their significant value in diagnosing rare forms of thyroid cancer and serving as predictive and/or prognostic and/or therapeutics biomarkers (Table 1) [1, 2].

Starting with benign lesions, the previous terms “colloid nodules,” “multinodular goitre,” “adenomatous goitre,” and “multinodular hyperplasia” have been consolidated under the new designation FND. This change is related to the recognition that many nodular lesions are genetically monoclonal, thereby qualifying as neoplastic lesions (corresponding to adenomas). In some situations, particularly in paediatric cases as referred above, these lesions can represent manifestations of germline mutations, such as DICER1 or PTEN.

The condition previously known as hyperfunctioning, toxic adenoma, has been renamed as follicular adenoma with papillary architecture. This updated terminology is a morphologic descriptor, indicating that while these lesions exhibit a papillary architecture, they lack the nuclear features characteristic of PTC. Genetically these lesions differ from traditional follicular adenomas which frequently harbour RAS mutations. Instead, hyperfunctioning follicular adenomas with papillary architecture are identified by mutations in the thyroid-stimulating hormone receptor (TSHR), GNAS, or enhancer of zeste 1 polycomb repressive complex 2 subunit (EZH1) genes, which can occur associated with Carney complex, McCune Albright syndrome, and DICER1 syndrome. Of note, a subtype of non-hyperfunctioning, follicular adenoma with papillary architecture, lacking the characteristic PTC-like nuclear features, is also considered in the 5th WHO Classification [1, 2].

A newly defined category of “Low-risk neoplasms” has been incorporated into the WHO Classification, which mainly includes RAS-like neoplasms such as NIFTP, FT-UMP, and WD-UMP. As a novelty, the genetic criteria outlined in the 5th WHO Classification specify that the presence of BRAF p.V600E or telomerase reverse transcriptase (TERT) promoter mutations are used as exclusion criteria for NIFTP [1, 2]. An exception within this group to the typical RAS-like genetic background, is the rare neoplasm HTT, distinguished by frequent PAX8::GLIS3 and PAX8::GLIS1 fusions, which are diagnostic markers for these tumours [22].

Regarding differentiated thyroid carcinomas (PTC, FTC, OTC, and MTC), the major alterations that has been introduced was the inclusion of a group of HGDTCs. However, this classification is not based on genetic biomarkers; instead, the number of mitoses and, most importantly, the presence of necrosis serve as the main criteria, regardless of genetic background [1, 2]. It remains to be determined whether the so-called “high-grade mutations,” such as TERT or TP53, are more prevalent in HGDTC.

Several changes in the 5th WHO Classification aim to clarify the distinct, or previously unclear, etiopathogenesis of certain tumours. For instance, the category of salivary gland-type neoplasms includes rare thyroid tumours with specific genetic alterations, such as secretory carcinoma harbouring ETV6::NTRK3 fusions [17]. Additionally, the tumour previously known as the “cribriform-morular variant of PTC” is no longer classified as a PTC variant. Its derivation from follicular cells is not clearly established since it often lacks PAX8 and thyroglobulin expression, and TTF1 expression is limited to the cribriform elements. Consequently, this tumour has been reclassified as one of uncertain histogenesis and renamed cribriform-morular thyroid carcinoma (CMTC), characterized by mutations in the Wnt/beta-catenin pathway [1, 2]. Although the histogenesis is controversial, it has been advanced that the absence of follicular differentiation markers is secondary to the permanent activation of the Wnt/β-catenin pathway (APC, CTNNB1 and AXIN1 genes) [18, 19].

Malignant teratoma of the thyroid has been recognized as a distinct entity, now called thyroblastoma. This extremely rare high-grade thyroid neoplasm is composed of primitive follicular cells and is frequently associated with DICER1 mutations, which may help in the in the differential diagnosis [21].

In the context of thyroid cancer molecular pathology is increasingly significant, particularly in the realm of personalized oncology. For example, BRAF genotyping in ATC is crucial as patients benefit from early determination of BRAF status to facilitate targeted therapy. Similarly, in advanced MTC, RET-targeted therapies have shown promising effects on patient survival. Additionally, differentiated thyroid carcinomas harbouring NTRK or ALK rearrangements, although rare, often respond favourably to specific inhibitors.

The new WHO edition also includes a comprehensive review of syndromic familial non-MTC, as well as non-syndromic familial non-medullary thyroid carcinoma (NSFNMTC) [1, 3]. While in daily diagnostic practice immunohistochemical determination of β-catenin can help confirm the diagnosis of FAP-associated thyroid carcinoma (CMTC) [18, 19], loss of PTEN protein expression facilitates the screening of follicular cell tumours associated with PTEN hamartoma tumour syndrome [23, 24]. The definition of clinical criteria for NSFNMTC may help characterize this complex and heterogeneous group in which many susceptibility genes have been implicated [1, 25, 26].

The 5th edition of the WHO Classification clarified some problems regarding the diagnosis of thyroid tumours, treatment, and follow-up providing some hints about the most adequate surgical treatment in each patient [1]. Low-risk neoplasms are correctly treated by hemithyroidectomy (lobectomy and isthmusectomy) [1]. The biologic potential of UMP tumours is not certain, so they require close follow-up [2].

Tumours below 10 mm should be subtyped, and no longer referred as microcarcinomas. Occasionally, there are <10 mm thyroid tumours with aggressive pathologic features and guarded prognosis that develop recurrence and distant metastasis after initial treatment [2, 27]. When treating the aforementioned <10 mm tumours, histological subtype must be considered. Tall cell, columnar cell, and hobnail-PTC subtypes have more aggressive clinicopathologic features comparing to classic PTC and have intermediate risk of structural recurrence [2].

Separation of OTC from FTC is very important because OTC patients have more often RAI-refractory disease than FTC patients [28]. Patients with OTC may benefit from molecular evaluation since RAI therapy may be ineffective. Unfortunately, most OTC lack molecular alterations besides mutations in mitochondrial DNA, turning difficult to find oncogenic targets.

FTC, OTC, and IEFVPTC have been subtyped as minimally invasive, encapsulated angioinvasive and widely invasive. Invasiveness extension enhanced distinction between low-, intermediate-, and high-risk patients, especially for disease-specific survival and incidence of RAI-refractory disease [28]. Minimally invasive subtype (FTC, FVPTC) can be treated just by hemithyroidectomy since they are considered to be low risk [2]. Encapsulated angioinvasive and widely invasive subtypes (FTC, FVPTC) are more aggressive than minimally invasive, and their treatment requires total thyroidectomy and RAI administration in an attempt to prevent cervical and/or distant metastasis [2]. Patients with encapsulated angioinvasive and widely invasive subtypes are at higher risk of RAI-refractory disease and disease-specific death [28]. In encapsulated angioinvasive OTC (EAIOTC) patients, the incidence of RAI-refractory disease is in-between minimal invasive OTC and widely invasive OTC patients [28].

Concerning vascular invasion, the 5th edition of the WHO Classification requires separation of lymphatic invasion from venous invasion, and to evaluate, as precisely as possible, the degree of vascular invasion [2]. In FTC and OTC patients, presence and extent of venous invasion were strongly associated with survival [28]. Angioinvasion is associated with prognosis, and tumours with extensive vascular invasion carry worse prognosis [29]. Two or more foci of angioinvasion may be useful for predicting the prognosis among FTCs [30]. Treatment options for patients with encapsulated angioinvasive FTC should consider the number of independent vessels invaded in cases with vascular invasion [31]. Patients having tumours with angioinvasion, particularly >2 [32], should be submitted to total thyroidectomy and RAI treatment. Despite the aforementioned recommendations it remains unclarified the way of evaluating the degree of vascular invasion, in order to provide recommendations about the need to do total thyroidectomy and RAI treatment. For the moment we think this recommendation is followed whenever there are unequivocal signs of venous invasion, regardless of the number of vessels involved [29‒31].

DHGTC present more advanced histopathologic stages and lower radioactive iodine avidity than differentiated thyroid carcinomas [31]. Patients with DHGTC have similar overall prognosis to patients with PDTC [31]. These patients should be submitted to total thyroidectomy and RAI treatment. Molecular evaluation should be considered to guide further treatment.

High-grade MTCs occur in approximately 25% of cases and have worse prognosis than low-grade MTC [31]. Patients with DHGTC and high-grade MTC, have higher risk of recurrence, require more intense treatments and should have a close follow-up [31]. Clinicopathological evaluation is critical when considering the group of patients with high-grade neoplasms; besides the pathological characterization of the tumours, including size and invasiveness, treatment and follow-up also have to take in consideration other parameters such as patient’s age and presence of local and/or distant metastasis.

The 5th edition of the WHO Classification also raises challenges in the medical treatment and follow-up of patients with follicular cell-derived thyroid carcinomas. The group of low-risk neoplasms is helpful to emphasize the very good prognosis of patients with these lesions, avoiding overtreatment, namely, completion of thyroidectomy and radioiodine therapy. Furthermore, the terminology also helps reduce the psychosocial burden of a cancer diagnosis. However, it is recognized that the risk of recurrence is low, but not zero, and a different follow-up from patients with a purely benign diagnosis is under discussion [33]. At the present time, there is no consensus on the follow-up strategy for patients with low-risk neoplasms, with some authors suggesting a follow-up similar to patients with low-risk thyroid carcinoma, including thyroglobulin + thyroglobulin antibodies measurements and neck ultrasound for about 5–8 years [34]. On the other hand, other authors suggest that this approach is not cost-effective, advocating that patients with low-risk neoplasms should have a follow-up similar to patients with strictly benign diseases [35]. Considering that the risk is extremely low but not zero, we think that these patients should not be followed as if they had a clear cut diagnosis of benignity.

Another challenge is the communication with patients with a previous diagnosis of cancer, namely, of the former noninvasive encapsulated follicular variant of PTC, that now fulfil the criteria for the diagnosis of NIFTP after revision. On one hand, the change is very reassuring for both clinicians and patients; on the other hand, it may raise issues when patients realize they have been treated for a condition that now is no longer considered malignant.

From the clinical standpoint, the accurate definition of vascular invasion may also be challenging. In fact, in the case of FTC, the presence of vascular invasion immediately assigns a tumour to the high-risk category of the American Thyroid Association [36], with strong implications on the decision to submit patients to radioiodine treatment. It is well established that purely lymphatic invasion and venous invasion (the latter, with different degrees) have different prognostic implications, but there is lack of guidance on how they should be considered for treatment decisions.

The new high-risk category (high grade) is also useful to pinpoint tumours with aggressive behaviour, leading clinicians to more comprehensive strategies of treatment and follow-up. PDTC has been recognized has an aggressive cancer for a long time, but the simple presence of necrosis or mitosis may have not been considered strong indicators of aggressive disease for less experienced physicians.

As previously mentioned, the 5th edition of the WHO Classification also incorporates genetic markers, acknowledging their significant value in serving as predictive and/or prognostic and/or therapeutics biomarkers. Regarding prognosis and from the clinical standpoint, two poor prognosis markers emerge as relevant: TERTp and p53 mutations [37]. But the most relevant contribution of genetic markers in clinical practice is their use as therapeutic targets in advanced thyroid cancer. BRAF, NTRK, and RET genetic alterations (the latter in both follicular cell-derived and medullary thyroid cancer) are now potential drug targets, with specific inhibitors showing benefit in patients with advanced thyroid cancer [38‒40]. In comparison with multikinase inhibitors, specific inhibitors showed not only efficacy but also better adverse events profile. The paradigm of this shift is ATC harbouring BRAF mutations, in which the combination of dabrafenib and trametinib was approved by the FDA, with a substantial improvement in overall survival [41]. Targeted therapy has also been used with success for the redifferentiation of radioiodine-refractory FCDTC [42]. For the aforementioned reasons, an early approach to molecular studies may also be of value in patients with high-grade tumours because it can avoid further delays for treatment initiation once the disease becomes advanced, progressive and radioiodine-refractory.

The four previous sections about the most impressive aspects of the 5th edition of WHO Classification demonstrate a number of major advances in terms of (a) differential diagnosis (e.g., separating RAS-like from BRAF-like tumours and supporting some very difficult diagnosis using gene-rearrangements); (b) different types of surgical approach (e.g., avoiding lymph node dissection in cases of follicular carcinomas, regardless of being or not HGDTC, encapsulated follicular variant PTC and oncocytic [Hürthle] carcinomas); (c) strategy of therapy selection based upon the identification of molecular targets.

Besides the aforementioned examples, there are several clinicopathological advances regarding precise diagnosis, and more accurate prognostic evaluation and therapy selection, separating lymphatic from venous invasion, as well as the evaluation of the degree of vascular invasion and, in selected cases, the utilization of molecular biomarkers.

The last point – importance of genetic markers – has to be considered in cost/benefit terms since there is an increasing tendency to use molecular data everywhere, whereas cytopathology has not gained significant advantages from their utilization, namely, for identifying PTC and PDTC, thus leading us back to the importance of morphological evidence (hail to histopathology, together with immunohistochemistry, as the basis of diagnosis and treatment).

Despite realizing that most authors usually stress “positive” findings as the basis for “take home lessons,” we decided to finish this article identifying the most challenging issues, at present, after publishing the 5th edition of WHO Classification.

• It is appropriate to consider that micro PTCs (microMPT) do not represent a subtype of PTC, thus removing it from the histopathology list, but it remains its clinical importance of (very) small PTC. Like the group of Cameselle-Teijeiro [5], we think it would be appropriate to designate “so-called” micro-PTC as micro papillary thyroid tumours.

• Taking in consideration the huge data about the diagnosis and prognosis of NIFTP accumulated to date, it is very interesting to figure out when we will jump to tumours diagnosed as follicular adenomas instead of NIFTP (we think NIFTP will sooner or later disappear from the list of thyroid tumours).

• Following the molecular data (and follow-up) obtained to date on encapsulated follicular (RAS-like) patterned tumours, it remains to be clarified whether or not we will reclassify encapsulated, invasive, follicular variant PTC as follicular carcinoma.

• The importance of the evaluation of the degree of vascular invasion for prognosis and therapy selection remains unclarified [13].

• The creation of HGDTC in the spectrum of PDTCs was a very important step. The problem, now, regards how to designate, and how to treat, “encapsulated noninvasive tumours with foci of necrosis and/or numerous mitoses.” We think these tumours should not be diagnosed as carcinoma but it is important to make a close follow-up of these cases. Incidentally, numerous mitoses are observed in follicular adenomas of young patients (so-called mitotically active follicular adenomas) without clinical meaning.

• Targeted therapy of thyroid carcinomas, namely, anaplastic carcinomas, as well as some PDTC and HGDTC, witnessed major advances, and the 5th edition of WHO Classification provides a solid basis for treating patients. The most challenging issue in this field concerns the treatment of patients who do not respond to I131 iodine therapy.

Dr. Paula Soares, Dr. Miguel Melo, and Prof. Manuel Sobrinho-Simões were members of the journal’s Editorial Board at the time of submission. The authors have no other conflicts of interest to declare.

J.M.C.-T. is supported by grant PI23/00722 from Instituto de Salud Carlos III (Spain), co-funded by the European Union (EU). This study is part of the project “Institute for Research and Innovation in Health Sciences” (UID/BIM/04293/2019) and the project “The Porto Comprehensive Cancer Centre” ref. NORTE-01-0145-FEDER-072678 – Consórcio PORTO.CCC – Porto. Comprehensive Cancer Centre Raquel Seruca and obtained further funding through Fundação para a Ciência e a Tecnologia (FCT). The funders had no role in the design, data collection, data analysis, and reporting of this study.

Paula Soares drafted the first version of the manuscript and contributed to the “Challenges in genetics.” José Manuel Cameselle-Teijeiro contributed to the “Challenges in Pathology.” Antónia Póvoa contributed to the “Challenges in Surgical Approach.” Miguel Melo contributed to the “Challenges in Medical Treatment and Follow-Up.” Manuel Sobrinho-Simões contributed to the “Perspective” and “Take Home Lessons.” All the authors contributed to the final review of the text.