Journal

Histopathology

Papers (110)

Role of the pathologist in assessing response to treatment of ovarian and endometrial cancers

Standardisation of pathological evaluation of tissue responses to therapy permits robust stratification of patient outcomes for management decisions and allows comparison of results across clinical trials. In gynaecological pathology there are two major areas where pathological assessment of treatment response is currently used to determine ongoing therapy. High‐grade serous carcinoma (HGSC) of tubo‐ovarian origin frequently presents as high‐stage disease and may be managed by neoadjuvant chemotherapy with debulking surgery. The chemotherapy response score (CRS) is a reproducible, validated three‐tiered morphological scoring system to assess the response of HGSC to treatment. Interobserver agreement is shown to be substantial following online training, and women with CRS3 have significantly improved progression‐free and overall survival. Low‐grade endometrioid endometrial cancer and atypical hyperplasia/endometrioid intraepithelial neoplasia may be managed by progestogenic therapy in women who wish to preserve fertility or for whom medical co‐morbidities preclude surgical management. The response to treatment is assessed histologically in successive endometrial biopsies. The histological parameters are well described, but the pathological classification of treatment response is still under development. Pathological assessment of the response to treatment is incorporated into clinical guidelines.

Undifferentiated endometrial carcinoma arising in the background of high‐grade endometrial carcinoma – Expanding the definition of dedifferentiated endometrial carcinoma

Dedifferentiated endometrial adenocarcinoma (DEC) is defined by the coexistence of undifferentiated carcinoma with low‐grade (FIGO grade 1 or 2) endometrioid carcinoma. Few cases of DEC arising in the background of high‐grade carcinoma (DEC‐HG) have been reported, however, this phenomenon is poorly characterized. In this study we describe the morphologic, immunohistochemical and clinico‐pathologic characteristics of DEC‐HG. 18 DECs were diagnosed at our institution between 2008‐2019, and in 11 (61%), the undifferentiated component was associated with high‐grade carcinoma (8 endometrioid FIGO grade 3, 2 with ambiguous features, 1 serous). The remaining 7 (39%) represented DEC‐LG (3 FIGO grade 1 and 4 FIGO grade 2). 7/11 (64%) patients with DEC‐HG presented with advanced stage (FIGO stage III/IV), whereas most with DEC‐LG (6/7, 86%) were stage I. On follow up, 2 patients in the DEC‐HG group died of disease and 2 had progressive disease within 2 months of surgery. There was only one recurrence in the DEC‐LG, 6 months post‐surgery. The DEC component in both groups showed similar morphology and immunophenotype, with predominantly focal or complete loss of expression of pan‐keratin, EMA, E‐cadherin, CK8/18, PAX8 and ER. The DEC component in the DEC‐HG group had wild‐type p53 expression in 8/11 (73%) cases, loss of MLH1 and PMS2 in 6/11 (55%) and loss of SMARCA4 in 3/9 (33%). Although numbers are small, we show that DEC‐HG is a previously under‐recognized phenomenon, with morphologic and immunophenotypic similarities to DEC‐LG, which supports expanding the definition of DEC to include DEC‐HG. DEC‐HG may be more aggressive than DEC‐LG.

Serous tubal intra‐epithelial carcinoma: what do we really know at this point?

Serous tubal intra‐epithelial carcinoma (STIC) is the earliest morphologically recognisable step in the development of invasive high‐grade serous carcinoma of the fallopian tube. Lesions occurring prior to STIC within the carcinogenic sequence for the pathogenesis of invasive high‐grade serous carcinoma include the p53 signature and secretory cell outgrowth (SCOUT). Variable histological criteria have been used for diagnosing STIC, but a combination of morphology and immunohistochemistry for p53/Ki‐67 improves interobserver agreement. Half of all carcinomas identified in risk‐reducing salpingo‐oophorectomy specimens are in the form of STIC; however, STIC also may be incidentally found on occasion in specimens from women at low or average risk of ovarian/tubal/peritoneal carcinoma. TP53 mutation is the earliest known DNA sequence alteration in STIC and almost all invasive high‐grade serous carcinomas of the ovary and peritoneum. Data on the clinical behaviour of STIC are limited. While the short‐term follow‐up in the prior literature suggests a low risk of malignant progression, a more recent meta‐analysis indicates a 10‐year risk of 28%. STIC probably should be best regarded as a lesion with uncertain malignant potential at present, and future molecular analysis will help to classify those with higher risk of dissemination. This review provides an update on the current knowledge of STIC and related issues.

A 5‐marker immunohistochemical panel of CK17 , MEP1A , PAX8 , SMAD4 , and CDX2 to distinguish ovarian mucinous carcinoma from metastatic pancreatic ductal adenocarcinoma

Aims Metastatic pancreatic adenocarcinoma (PDAC), albeit uncommon, may involve the ovary, and distinction from primary ovarian mucinous tumours (OMT) poses a diagnostic challenge. Our aim was to develop an ancillary immunohistochemical (IHC) panel to aid in diagnosis and to validate the morphological features of metastatic PDAC. Methods and results Six IHC markers (CDX2, CK17, MEP1A, MUC2, PAX8, SMAD4) selected based on a literature review were stained on tissue microarrays containing 256 PDAC, 102 mucinous ovarian carcinomas (MC) and 58 mucinous borderline ovarian tumours (MBOT). Detailed morphological features were reviewed in 16 ovarian metastases from PDAC, 25 MC, and 9 MBOT. We confirmed that tumours with a size less than 13 cm, bilaterality, ovarian surface involvement, low‐power nodularity, infiltrative invasion, pseudomyxoma ovarii despite cystadenoma or borderline areas, and moderate nuclear atypia should raise suspicion for metastatic PDAC and prompt evaluation with the recommended IHC panel. A 5‐marker panel consisting of CK17, MEP1A, PAX8, SMAD4, and CDX2 had an overall accuracy of 91.8% (95% CI 88.8%–94.3%) using recursive partitioning, with the highest weight resting on CK17. CK17 was expressed in 80.9% of PDAC compared to 18.6% of MC and 1.7% of MBOT, respectively. Conclusions This is the first ancillary IHC panel to distinguish between PDAC and OMT with high accuracy. These results inform further studies on diagnostic workflows tailored to the complexity of metastatic presentations of tumours at the ovary.

Assessment and classification of sex cord‐stromal tumours of the testis: recommendations from the testicular sex cord‐stromal tumour (TESST) group, an Expert Panel of the Genitourinary Pathology Society (GUPS) and International Society of Urological Pathology (ISUP)

AimsTesticular sex cord‐stromal tumours (TSCSTs) are relatively rare, accounting for ~5% of all testicular neoplasms. They were historically classified into Leydig cell tumour, Sertoli cell tumour, granulosa cell tumour, and unclassified sex cord‐stromal tumour. More recently, classification was expanded to incorporate additional histologic types, including some associated with inherited cancer predisposition syndromes. However, the classification of TSCSTs still relies entirely on morphology, with some tumour types being defined based on their resemblance to ovarian counterparts. In recent years, molecular studies have identified drivers and genomic alterations associated with aggressive behaviour and progression; however, these findings have not yet impacted classification and management.Methods and resultsUnder sponsorship of the International Society of Urological Pathology (ISUP) and the Genitourinary Pathology Society (GUPS), a group of genitourinary pathologists was assembled in 2023 with the aim of assessing how to use these new data to improve the classification and management of TSCSTs.ConclusionsThis paper summarizes the recommendations derived from the consensus activities and the first meeting of the testicular sex cord‐stromal tumour (TESST) group (held at Johns Hopkins Hospital, Baltimore, USA, 3/23/2024).

Prognostic values of molecular subtypes and SWI/SNF protein expression in de‐differentiated/undifferentiated endometrial carcinoma

AimsClassification and risk stratification of endometrial carcinoma (EC) has transitioned from histopathological features to molecular classification, e.g. the ProMisE classifier, identifying four prognostic subtypes: POLE mutant (POLEmut) with almost no recurrence or disease‐specific death events, mismatch repair deficient (MMRd) and no specific molecular profile (NSMP), with intermediate outcome and p53 abnormal (p53abn) with poor outcomes. However, the applicability of molecular classification is unclear in rare but aggressive histotypes of EC, e.g. de‐differentiated and undifferentiated endometrial cancers (DD/UDEC). Here, we aim to assembled a cohort of DD/UDEC from a single institution and analysed the prognostic significance of ProMisE molecular subtypes and the expression of SWItch/sucrose non‐fermentable (SWI/SNF) chromatin remodelling complex members, previously implicated in the pathogenesis of DD/UDEC.Methods and resultsWe accrued 88 DD/UDEC cases, assessed POLE status by Sanger sequencing and performed immunohistochemistry for p53, mismatch repair and SWI/SNF proteins on the tissue microarrays assembled. Assignment of molecular subtypes was possible in 80 tumours; POLE sequencing failed in the remaining eight cases. There were 12 (15%) POLEmut, 44 (55%) MMRd, 14 (17.5%) p53abn and 10 (12.5%) NSMP DD/UDEC. POLEmut DD/UDECs had excellent outcomes, but the other three molecular subtypes all had poor outcomes, with no significant differences among them. The loss of one or more SWI/SNF proteins [AT‐rich interactive domain‐containing protein 1A (ARID1A), ARID1B, SWI/SNF‐related, matrix‐associated, actin‐dependent regulator of chromatin, subfamily A, member 4 (SMARCA4), SMARCA2], observed in 66% (55 of 83) cases, was not of prognostic significance.ConclusionsThese results indicate that all molecular subtypes of DD/UDEC except POLEmut behave in an aggressive fashion. Further study is needed to determine whether these molecular alterations can be targeted with adjuvant therapy, in order to improve outcomes of patients with DD/UDEC.

Identifying mesonephric‐like adenocarcinoma of the endometrium by combining SOX17 and PAX8 immunohistochemistry

AimsMesonephric‐like adenocarcinoma (MLA) of the endometrium or ovary is a rare but distinct endometrial carcinoma which has a combination of characteristic morphological, immunohistochemical (IHC) and molecular features. SOX17 has been recently identified as a highly sensitive and specific marker for endometrial and ovarian carcinomas. In this study, we aimed to investigate SOX17 expression in MLA together with other IHCs to differentiate MLAs from other endometrial carcinomas.MethodsSeventeen previously diagnosed endometrial/ovarian MLAs were collected, and multiple IHCs were performed. Additionally, we performed SOX17, PAX8 and ER on tissue microarrays (TMAs) composed of 652 endometrial carcinomas from 2012 to 2015 when MLA diagnostic criteria were not established.ResultsAll 17 MLAs showed diffuse strong positive PAX8, negative ER and variable TTF1/GATA3 staining. Notably, all MLAs showed negative (n = 10) or focal weak/moderate (n = 7) staining for SOX17, which is more diffuse and stronger than PAX8 in other endometrial carcinoma subtypes. This finding prompted us to screen TMAs with 652 endometrial carcinomas diagnosed before MLA by an approach of combined SOX17 and PAX8 IHCs, and 14 cases with positive PAX8 but negative/focal weak SOX17 were identified. We further studied the 14 cases by examining morphology and performing additional IHCs (TTF1, GATA3, ER and CD10) and would classify seven (50%) of them as MLAs based on morphological features and positive CD10, TTF1 and/or GATA3 staining.ConclusionOur results suggest that a combination of SOX17 and PAX8 IHCs would aid in diagnosing MLA if the results show strong positive PAX8, but negative SOX17.

Intratumoral budding is associated with poor clinical outcome in early‐stage clear cell carcinoma of ovary*

AimsClear cell carcinoma of ovary (CCC) is considered a high‐grade malignancy by default and the role of histological grading for assessing clinical outcome is not established. We aimed to evaluate histopathological features associated with clinical outcome in CCC patients.Methods and resultsSeventy‐six cases of CCC with available clinical follow‐up information were studied. Histopathological features, including tumour size, architectural patterns, nuclear atypia, mitotic activity, intratumoral and peritumoral inflammation, presence of endometriosis, peritumoral and intratumoral budding, were evaluated. Multivariate analysis was performed with logistic regression and Kaplan–Meier survival curves with the log‐rank test were used for survival analysis. Forty cases (53%) presented at stage I. Complete response to treatment was achieved in 65%, while 35% of patients had tumour recurrence or progression of disease despite treatment. At last follow‐up, 13% had died of disease, 20% were alive with disease and 67% had no evidence of disease. Higher stage (P = 0.0016) and presence of intratumoral budding (P = 0.0454) were independently associated with recurrence/disease progression. Advanced stage (P = 0.0011), presence of lymph node involvement (P = 0.0003), intratumoral budding (P = 0.0023) and peritumoral budding (P = 0.0334) were significantly associated with shorter survival. Intratumoral budding was significantly associated with recurrent/progressive disease (P = 0.0195) and also shorter survival (P = 0.0277) within the cohort of low‐stage (I/II) patients as well.ConclusionWe have shown that besides the classic prognostic factors of stage and lymph node status, the presence of tumour budding is associated with poorer outcome in patients with CCC. Specifically, evaluation of intratumoral budding may help to more clearly predict prognosis in patients with early‐stage disease.

The spectrum of oestrogen receptor expression in endometrial carcinomas of no specific molecular profile

AimsDecreased oestrogen receptor (ER) expression is a marker of poor prognosis in endometrial carcinomas (EC) of no specific molecular profile (NSMP), but the optimal cut‐off to separate high‐risk ‘low ER’ versus low‐risk ‘high ER’ expression has not been defined. Here we characterised the distribution of ER staining in a cohort of ECs.Methods and resultsBiopsy specimens from 120 cases of NSMP EC were stained for ER and assigned an Allred score. In 66 additional cases ER staining of matched biopsy and hysterectomy were compared. Twelve of 120 tumours had an Allred score of 0–3, including three endometrioid carcinomas (EEA) (one G1, two G3), four clear cell carcinomas (CCC), two mesonephric‐like adenocarcinoma (MLA) and one each of: gastric‐type adenocarcinoma, carcinosarcoma and endometrial carcinoma NOS. Three had Allred scores of 4–5: two MLA and one high‐grade carcinoma with yolk sac differentiation. Five had Allred scores of 6: four EEA (one G1, one G2, two G3) and one mixed clear cell and endometrioid carcinoma. The remaining 100 tumours with Allred scores ≥ 7 were all EEA (66 G1, 28 G2, five G3 and one grade unknown). Comparing the biopsy versus hysterectomy ER staining (n = 66), the results were within a single Allred score point, except two cases with strong diffuse expression in the biopsy (Allred 8) and moderate expression in the hysterectomy (Allred 5).ConclusionsMost NSMP ECs (> 80%) show high ER expression (Allred score ≥ 7). All non‐endometrioid carcinomas and a few endometrioid carcinomas had lower ER expression (Allred score ≤ 6) or were completely negative.

Papillary and ductal patterns of mesonephric‐like adenocarcinomas are often overlooked: a retrospective revaluation of over 1000 endometrial carcinomas

AimsMesonephric‐like adenocarcinoma (MLA) of the endometrium is often a diagnostic challenge, due to its morphological resemblance to other more common Müllerian neoplasms. This study aimed to retrospectively identify overlooked MLA in a large endometrial carcinoma cohort, using a combination of immunohistochemistry (IHC), morphology and KRAS sequencing.Methods and resultsIHC was conducted on 1094 endometrial carcinomas, identifying 16 potential MLA cases based on GATA3+ and/or TTF1+ and ER− staining patterns, which subsequently underwent detailed histological review, KRAS sequencing and ProMisE molecular classification. Of the IHC screen‐positive cases, one was positive for both GATA3 and TTF1, nine were positive for GATA3 only and six were positive for TTF1 only. All IHC screen‐positive cases were POLE wild‐type. All five tumours in the NSMP category showed morphological features of MLA, while the three MMRd and eight p53abn tumours did not show MLA morphology. The five cases diagnosed as MLA on review were all originally diagnosed as low‐grade endometrioid adenocarcinoma probably because of rare morphological patterns, being predominantly papillary or ductal. Four of the five cases harboured a KRAS mutation.ConclusionThis study highlights the importance of a comprehensive diagnostic approach for accurately identifying endometrial MLA and for pathologists to be aware of papillary and ductal patterns in endometrial carcinoma assessment. Further exploration into the molecular landscape of MLA is essential for refining diagnostic criteria and developing targeted therapies.

Practical guidance for assessing and reporting lymphovascular space invasion (LVSI) in endometrial carcinoma

Lymphovascular space invasion (LVSI) is an important prognostic parameter in endometrial carcinoma (EC) and has gained increasing interest in recent years due to an expanding body of evidence of its independent prognostic value, especially when the presence of LVSI is quantified. A key strength of LVSI as a prognostic factor is that it can be detected on routine microscopic examination, without ancillary tests, and thus can be used in low‐resource settings. A weakness, however, is the lack of uniformly applied criteria for assessment and quantification of LVSI, resulting in interobserver variation in diagnosis. This is confounded by artefacts and other morphological features that may mimic LVSI (commonly referred to as pseudo‐LVSI). Despite these issues, multiple studies have shown that LVSI is strongly associated with lymph node (LN) metastasis and is an independent risk factor for LN recurrence and distant metastasis. Consequently, the presence of substantial/extensive LVSI has become an important consideration in formulating adjuvant treatment recommendations in patients with EC, and this has been incorporated in the recent International Federation of Gynecology and Obstetrics (FIGO) 2023 staging system. Herein, we review the current literature on LVSI in EC and discuss its role as a prognostic marker, the reproducibility of LVSI assessment and distinction between LVSI and its mimics. We provide illustrations of key diagnostic features and discuss the two‐tiered (none/focal versus substantial) system of LVSI classification. This work is intended to provide guidance to practising pathologists and unify the approach towards LVSI assessment in EC.

Subclonal p53 immunostaining in the diagnosis of endometrial carcinoma molecular subtype

AimsThe significance of subclonal expression of p53 (abrupt transition from wild‐type to mutant‐pattern staining) is not well understood, and the arbitrary diagnostic cut‐off of 10% between NSMP and p53abn molecular subtypes of endometrial carcinoma (EC) has not been critically assessed. Our aim was to characterise subclonal p53 and discrepant p53 expression/TP53 sequencing results in EC and assess their clinical significance.Methods and resultsSubclonal p53 immuostaining on whole sections from 957 ECs was recorded. Agreement between TP53 mutational assessment and p53 immunostaining was evaluated. Subclonal p53 IHC staining was seen in 4.0% (38 of 957) of cases, with 23 of 957 (2.4%) showing mutant‐pattern p53 staining in ≥10% of tumour cells. It was most commonly seen in POLEmut (nine of 65, 14%) and MMRd (13 of 274, 4.7%) EC (‘multiple classifier’ ECs), where subclonal p53 staining does not impact the molecular subtype diagnosis. Excluding POLEmut and MMRd EC, 11 of 957 (1.1%) showed ≥10% subclonal p53 from which four patients died of disease, while there were no deaths due to disease in the five patients with <10% mutant‐pattern p53 staining. Agreement between p53 immunostaining and TP53 sequencing was 92.6%; most of the discrepant results were in the ultramutated POLEmut or hypermutated MMRd ECs. In NSMP and p53abn EC the agreement between IHC and sequencing was 95.8%.ConclusionsSubclonal p53 staining ≥10% is present in only 1.1% of EC after excluding ‘multiple classifier’ ECs. The cut‐off of ≥10% subclonal p53 staining identified patients at increased risk of dying from EC, supporting its use to diagnose p53abn molecular subtype.

Mesonephric‐like adenocarcinoma of the female genital tract: novel observations and detailed molecular characterisation of mixed tumours and mesonephric‐like carcinosarcomas

AimsTo report novel observations in five mesonephric‐like adenocarcinomas (MLAs) of the female genital tract.Methods and resultsWe report two endometrial MLAs in association with endometrioid carcinoma and atypical hyperplasia and three (one endometrial, two ovarian) cases with a sarcomatoid component (mesonephric‐like carcinosarcoma). Pathogenic KRAS mutations, which are characteristic of MLA, were identified in all cases although interestingly, in one of the mixed carcinomas, this was confined to the endometrioid component. The concurrent MLA, endometrioid carcinoma and atypical hyperplasia components in one case harboured identical EGFR, PTEN and CCNE1 mutations, suggesting that the atypical hyperplasia gave rise to a Müllerian carcinoma with both endometrioid and mesonephric‐like components. The carcinosarcomas all contained a component of MLA and a sarcomatous component with chondroid elements. In the ovarian carcinosarcomas, the coexisting epithelial and sarcomatous components shared some mutations including KRAS and CREBBP, suggesting that they are clonally related. Furthermore, in one case CREBBP and KRAS mutations detected in the MLA and sarcomatous components were also detected in an associated undifferentiated carcinoma component, suggesting that it was clonally related to the MLA and sarcomatous components.ConclusionsOur observations provide additional evidence that MLAs have a Müllerian origin and characterise mesonephric‐like carcinosarcomas in which chondroid elements appear to be characteristic. In reporting these findings, we provide recommendations for distinction between a mesonephric‐like carcinosarcoma and a MLA with a spindle cell component.

Utility of p63 and PTEN staining in distinguishing cervical microglandular hyperplasia from endometrial endometrioid carcinoma with microglandular/mucinous features

AimsDistinction between well‐differentiated endometrial carcinoma (EMCA) with microglandular/mucinous features and benign endocervical microglandular hyperplasia (MGH) can be a diagnostic challenge, especially when tissue is limited. The immunostains used to distinguish endocervical and endometrial carcinoma are less useful when the differential diagnosis is MGH. Here, we investigate the utility of p63 and phosphatase and tensin homologue (PTEN) to aid accurate classification.Methods and resultsCases obtained from our pathology archives included 25 EMCA with mucinous/microglandular features, 26 MGH and nine atypical microglandular proliferations. Cases were assessed for glandular architecture, presence of mucinous and/or eosinophilic luminal secretions, subnuclear vacuoles, foamy histiocytes, inflammation, squamous metaplasia, cytological atypia and mitotic activity. The presence and pattern of immunohistochemical staining for p63 and PTEN was recorded. Microglandular proliferations with cytological atypia, mitotic activity, foamy histiocytes and complex glandular architecture were more commonly seen in EMCA, while small glands, bland nuclei and subnuclear vacuoles were enriched in MGH. All MGH cases displayed p63‐positive subcolumnar reserve cells and retained PTEN expression. Four EMCA cases showed non‐specific focal p63 staining either at the surface of the tumour or in areas of squamous differentiation. p63 and PTEN immunostains accurately predicted the final diagnosis for 3 atypical microglandular proliferation cases with follow‐up.ConclusionsWhile there are morphological characteristics that differentiate EMCA and MGH, there is frequent overlap between these entities. Nonetheless, the pattern and extent of p63 and PTEN can aid accurate classification. Consistent p63‐positive subcolumnar reserve cells were seen only in MGH.

Performance of a HER2 testing algorithm specific for p53‐abnormal endometrial cancer

AimsHuman epidermal growth factor receptor 2 (HER2) amplification in endometrial cancer (EC) is almost completely confined to the p53‐abnormal (p53abn) molecular subtype and independent of histological subtype. HER2 testing should therefore be molecular subtype‐directed. However, the most optimal approach for HER2 testing in EC has not been fully established. Therefore, we developed an EC‐specific HER2 immunohistochemistry (IHC) scoring method and evaluated its reproducibility and performance to establish an optimal diagnostic HER2 testing algorithm for p53abn EC.Methods and resultsHER2 IHC slides of 78 p53abn EC were scored by six gynaecopathologists according to predefined EC‐specific IHC scoring criteria. Interobserver agreement was calculated using Fleiss’ kappa and the first‐order agreement coefficient (AC1). The consensus IHC score was compared with HER2 dual in‐situ hybridisation (DISH) results. Sensitivity and specificity were calculated. A substantial interobserver agreement was found using three‐ or two‐tiered scoring [κ = 0.675, 95% confidence interval (CI) = 0.633–0.717; AC1 = 0.723, 95% CI = 0.643–0.804 and κ = 0.771, 95% CI = 0.714–0.828; AC1 = 0.774, 95% CI = 0.684–0.865, respectively]. Sensitivity and specificity for the identification of HER2‐positive EC was 100 and 97%, respectively, using a HER2 testing algorithm that recommends DISH in all cases with moderate membranous staining in >10% of the tumour (IHC+). Performing DISH on all IHC‐2+ and ‐3+ cases yields a sensitivity and specificity of 100%.ConclusionsOur EC‐specific HER2 IHC scoring method is reproducible. A screening strategy based on IHC scoring on all cases with subsequent DISH testing on IHC‐2+/‐3+ cases has perfect test accuracy for identifying HER2‐positive EC.

Incorporation of molecular characteristics into endometrial cancer management

Histopathological evaluation including subtyping and grading is the current cornerstone for endometrial cancer (EC) classification. This provides clinicians with prognostic information and input for further treatment recommendations. Nonetheless, patients with histologically similar ECs may have very different outcomes, notably in patients with high‐grade endometrial carcinomas. For endometrial cancer, four molecular subgroups have undergone extensive studies in recent years: POLE ultramutated (POLEmut), mismatch repair‐deficient (MMRd), p53 mutant (p53abn) and those EC lacking any of these alterations, referred to as NSMP (non‐specific molecular profile). Several large studies confirm the prognostic relevance of these molecular subgroups. However, this ‘histomolecular’ approach has so far not been implemented in clinical routine. The ongoing PORTEC4a trial is the first clinical setting in which the added value of integrating molecular parameters in adjuvant treatment decisions will be determined. For diagnostics, the incorporation of the molecular parameters in EC classification will add a level of objectivity which will yield biologically more homogeneous subclasses. Here we illustrate how the management of individual EC patients may be impacted when applying the molecular EC classification. We describe our current approach to the integrated diagnoses of EC with a focus on scenarios with conflicting morphological and molecular findings. We also address several pitfalls accompanying the diagnostic implementation of molecular EC classification and give practical suggestions for diagnostic scenarios.

Consensus based recommendations for the diagnosis of serous tubal intraepithelial carcinoma: an international Delphi study

AimReliably diagnosing or safely excluding serous tubal intraepithelial carcinoma (STIC), a precursor lesion of tubo‐ovarian high‐grade serous carcinoma (HGSC), is crucial for individual patient care, for better understanding the oncogenesis of HGSC, and for safely investigating novel strategies to prevent tubo‐ovarian carcinoma. To optimize STIC diagnosis and increase its reproducibility, we set up a three‐round Delphi study.Methods and resultsIn round 1, an international expert panel of 34 gynecologic pathologists, from 11 countries, was assembled to provide input regarding STIC diagnosis, which was used to develop a set of statements. In round 2, the panel rated their level of agreement with those statements on a 9‐point Likert scale. In round 3, statements without previous consensus were rated again by the panel while anonymously disclosing the responses of the other panel members. Finally, each expert was asked to approve or disapprove the complete set of consensus statements. The panel indicated their level of agreement with 64 statements. A total of 27 statements (42%) reached consensus after three rounds. These statements reflect the entire diagnostic work‐up for pathologists, regarding processing and macroscopy (three statements); microscopy (eight statements); immunohistochemistry (nine statements); interpretation and reporting (four statements); and miscellaneous (three statements). The final set of consensus statements was approved by 85%.ConclusionThis study provides an overview of current clinical practice regarding STIC diagnosis amongst expert gynecopathologists. The experts' consensus statements form the basis for a set of recommendations, which may help towards more consistent STIC diagnosis.

Endometrial carcinoma and immune escape: prognostic relevance of HLA class I loss in NSMP subtype

Aims This study aims to define and characterize human leukocyte antigen class I (HLA‐I) expression in a consecutive series of molecularly classified endometrial carcinomas (ECs), and to evaluate its association with clinicopathologic features, spatial cancer–immune phenotypes and patient prognosis, with a focus on the NSMP (no specific molecular profile) subtype. Methods and results HLA‐I expression was assessed by immunohistochemistry on whole tissue sections from 208 ECs, classified into POLE ‐mutated, MMR‐deficient (MMRd), p53‐abnormal (p53abn) and NSMP subtypes. Loss of HLA‐I was identified in 31% of cases and was associated with adverse features including high‐grade, aggressive histotypes, deep myometrial invasion, substantial lymphovascular space invasion (LVSI), extensive tumour necrosis and an ‘excluded’ immune phenotype. While HLA‐I loss showed no significant prognostic impact in POLE , MMRd or p53abn tumours, it significantly correlated with worse disease‐free survival in NSMP tumours ( P  < 0.001). Multivariate analysis confirmed HLA‐I loss as an independent prognostic factor in early‐stage NSMP ECs, in addition to substantial LVSI, presence of lymph node metastases and spatial cancer–immune phenotypes. Integration of HLA‐I status improved the performance of predictive models over time. Conclusions HLA‐I loss defines a biologically aggressive subgroup within NSMP ECs and is associated with adverse clinicopathologic and immune features. Assessment of HLA‐I expression could refine risk stratification in NSMP ECs, a group traditionally lacking robust prognostic markers and may help identify patients who could benefit from intensified clinical surveillance and future immunomodulatory treatment strategies.

Pan‐TRK immunohistochemistry in gynaecological mesenchymal tumours: diagnostic implications and pitfalls

AimsNTRK‐rearranged sarcomas of the female genital tract mainly occur in the uterus (more commonly cervix than corpus) and are characterized by a “fibrosarcoma‐like” morphology and NTRK gene rearrangements. These neoplasms may exhibit histological overlap with other entities and can present diagnostic difficulties without molecular confirmation. Pan‐TRK immunohistochemistry was developed to identify tumours harbouring NTRK rearrangements. The aim of this study was to characterize pan‐TRK immunohistochemical expression in a large cohort of gynaecological mesenchymal neoplasms and investigate the utility of pan‐TRK immunohistochemistry to distinguish NTRK‐rearranged sarcoma from its mimics.Methods and resultsA total of 473 gynaecological mesenchymal tumours (461 without known NTRK fusions and 12 NTRK‐rearranged sarcomas) were selected. Pan‐TRK immunohistochemistry (EPR17341, Abcam) was performed on whole tissue sections and tissue microarrays. Molecular interrogation of pan‐TRK positive tumours was performed by RNA sequencing or fluorescence in situ hybridization (FISH). Of the 12 NTRK‐rearranged sarcomas, 11 (92%) exhibited diffuse (≥70%) cytoplasmic pan‐TRK staining with moderate/marked intensity, while the other was negative. Eleven (2.4%) additional tumours also exhibited pan‐TRK immunohistochemical expression: three low‐grade endometrial stromal sarcomas, seven high‐grade endometrial stromal sarcomas, and an undifferentiated uterine sarcoma. Molecular confirmation of the absence of NTRK rearrangements was possible in nine of these tumours. Of these nine neoplasms, seven exhibited focal/multifocal (<70%) pan‐TRK cytoplasmic staining with weak/moderate intensity.ConclusionEven though pan‐TRK immunohistochemical expression is not entirely sensitive or specific for NTRK‐rearranged sarcomas, these neoplasms tend to exhibit diffuse staining of moderate/strong intensity, unlike its mimics. Pan‐TRK should be performed in monomorphic uterine (corpus and cervix) spindle cell neoplasms that are negative for smooth muscle markers and hormone receptors and positive for CD34 and/ or S100. Ultimately, the diagnosis requires molecular confirmation.

Extrauterine epithelioid trophoblastic tumour and its somatic carcinoma mimics: short tandem repeat genotyping meets the diagnostic challenges

AimsWhile epithelioid trophoblastic tumour (ETT) primarily arises from the uterus, cases have been increasingly documented at extrauterine sites, originating from an ectopic gestation or presenting as a metastatic tumour, leading to the major differential diagnosis of somatic carcinoma with trophoblastic differentiation. The precise separation of a gestational trophoblastic tumour from its somatic carcinoma mimics is highly relevant and crucial for patient management and prognosis.Methods and resultsWe summarise the clinicopathological and molecular features of four challenging epithelioid malignancies presenting at extrauterine sites, with ETT as the main differential diagnosis. All four tumours demonstrated histological and immunohistochemical features overlapping between a somatic carcinoma and an ETT, combined with inconclusive clinical and imaging findings. Serum beta‐hCG elevation was documented in two cases. Short tandem repeat (STR) genotyping was performed and was informative in all cases. The presence of a unique paternal allelic pattern in the tumour tissue confirmed the diagnosis of ETT in two cases with an initial consideration of either somatic carcinoma or suspicion of a gestational trophoblastic tumour. The presence of matching genetic profile with the patient's paired normal tissue was seen in two other cases (both initially considered as ETT), confirming their somatic origin, including one metastatic triple‐negative breast carcinoma and one primary lung carcinoma.ConclusionsDiagnostic separation of ETT at an extrauterine site from its somatic carcinoma mimics can be difficult at the histological and immunohistochemical levels. STR genotyping offers a robust ancillary tool that precisely separates ETT from somatic carcinomas with trophoblastic differentiation.

Analytical validation of human epidermal growth factor receptor 2 immunohistochemistry by the use of the A0485 antibody versus the 4B5 antibody and breast versus gastric scoring guidelines in ovarian clear cell carcinoma

AimsThe aim of this study was to evaluate human epidermal growth factor receptor 2 (HER2) immunohistochemistry (IHC) in ovarian clear cell carcinoma (OCCC) by using two antibodies and two scoring guidelines in correlation with HER2 amplification and clinicopathological features.Methods and resultsA tissue microarray was constructed by use of a total of 71 OCCC cases for IHC (the A0485 antibody and the 4B5 antibody) and dual‐colour silver in‐situ hybridisation (DISH). Two pathologists independently scored the IHC according to the 2018 American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) breast cancer guidelines (breast guidelines) and the 2016 ASCO/CAP gastro‐oesophageal adenocarcinoma guidelines (gastric guidelines). IHC concordances between A0485 and 4B5 were 87.3–93.0%. Three to 16 (4.2–22.5%) cases had an IHC score of 2+/3+ with frequent basolateral/lateral membranous staining. The 4B5 antibody yielded fewer IHC 2+ cases than the A0485 antibody (n = 2–6 versus n = 5–12). Five (7.0%) cases had HER2 amplification as determined with DISH. Cases with papillary‐predominant growth patterns were significantly more likely to have HER2 amplification (P = 0.0051). In predicting DISH results, IHC scored according to the gastric guidelines yielded 100%/100% sensitivity and 83.3–95.5%/98.2–100% specificity, and IHC scored according to the breast guidelines yielded 60–80%/33.3–66.7% sensitivity and 95.5–100%/100% specificity (including/excluding IHC 2+ cases). One case had intratumoral heterogeneity, with discordant results between primary and metastatic tumour specimens.ConclusionWe demonstrated HER2 amplification in 7% of OCCC cases, and the molecular change is significantly associated with papillary‐predominant growth patterns. In predicting HER2 amplification, a combination of 4B5 IHC and gastric guidelines provides the best sensitivity and fewer equivocal (IHC 2+) cases. Given the intratumoral heterogeneity, assessment of HER2 status on whole tissue sections and on both primary and metastatic tumour specimens is recommended.

NTRK fusion cervical sarcoma: a report of three cases, emphasising morphological and immunohistochemical distinction from other uterine sarcomas, including adenosarcoma

AimsA unique fibrosarcoma‐like tumour of the uterine cervix harbouring a rearrangement of a neurotrophic tyrosine kinase receptor (NTRK) gene (NTRK1 or NTRK3) has recently been described in 11 young women, some with recurrence and/or metastasis. The aims of this study were to expand the morphological spectrum of this tumour by reporting three additional cases that showed adenosarcoma‐like features not previously described, one of which is the first reported to respond to targeted therapy, and to evaluate 19 conventional uterine adenosarcomas for evidence of NTRK rearrangement.Methods and resultsThree patients presented with a polyp or mass confined to the cervix. The constellation of polypoid growth, spindle cell morphology, entrapped endocervical glands and intraglandular stromal projections raised diagnostic consideration for adenosarcoma with stromal overgrowth. Deep cervical wall invasion was present in two cases at hysterectomy, and the third was removed by polypectomy. All three stained for S100 and pan‐Trk, but were negative for a spectrum of other diagnostic markers. All three harboured NTRK rearrangements (TPM3–NTRK1, TPR–NTRK1, and SPECC1L–NTRK3). One patient developed pleural metastases at 16 months, received the NTRK inhibitor larotrectinib, and is free of disease 15 months later. Two others are alive without disease. None of the uterine adenosarcomas showed any S100 or pan‐Trk staining, or rearrangement of NTRK1, NTRK2 or NTRK3 on next‐generation sequencing.ConclusionsUnusual adenosarcoma‐like spindle cell neoplasms of the cervix may represent an NTRK fusion sarcoma, which can be detected by S100 and pan‐Trk staining and confirmed by NTRK molecular testing. Conventional uterine adenosarcomas do not harbour NTRK rearrangements.

Gastric‐type mucinous carcinoma of the cervix and its precursors – historical overview

The emerging concept of gastric‐type mucinous carcinoma (GAS) of the uterine cervix has been accepted worldwide because of its aggressive clinical behaviour and the absence of high‐risk human papillomavirus (HPV). GAS is included as a variant of mucinous carcinoma in the 2014 World Health Organization classification, and its recognition has provoked a discussion on endocervical adenocarcinoma as a single entity such that endocervical adenocarcinoma is now divided into HPV‐associated and HPV‐independent groups. This article reviews historical and conceptual aspects of GAS and its precursors, starting with minimal deviation adenocarcinoma (MDA), through the ensuing confusion, up to the recent paradigm shift in cervical adenocarcinoma subclassification. The gastric immunophenotype of MDA was demonstrated by a Japanese group in 1998 using the HIK1083 antibody, which recognises gastric pyloric gland mucin, and this elucidated the pathogenesis of this particular tumour. However, this information resulted in overdiagnosis of lobular endocervical glandular hyperplasia (LEGH), first described in 1999 and which represents pyloric gland metaplasia (PGM), as malignant. In the early 2000s the relationship between MDA and LEGH/PGM became a matter of controversy. In 2007 HIK1083 immunohistochemistry extended the morphological spectrum of endocervical adenocarcinoma showing gastric differentiation beyond MDA, which resulted in the proposal of GAS as a distinct entity including MDA as its very well‐differentiated subtype. GAS is now considered to be an aggressive and chemoresistant neoplasm that is not related to high‐risk HPV. The LEGH/PGM–GAS sequence is currently regarded as an HPV‐independent pathway of carcinogenesis. Understanding the underlying molecular events in this process is key to the development of biomarkers for early detection and molecular targeted therapy.

A clinicopathological and molecular analysis of cervical carcinomas with basaloid features

AimsTo investigate the relationship between adenoid basal carcinoma (ABC), adenoid cystic carcinoma (ACC) and squamous cell carcinoma (SCC) in the uterine cervix.Methods and resultsWe analysed the clinicopathological and molecular features in two pure ABCs, 15 SCCs with ABC‐/ACC‐like features and seven basaloid SCCs (BSCCs) by chart review, immunohistochemistry, human papillomavirus (HPV) RNA in‐situ hybridisation and fluorescence in‐situ hybridisation. All patients were alive with no evidence of disease, except for one patient with ACC‐like features who died of disease at 18 months post diagnosis. The mixed carcinomas comprised variable SCCs and ABC‐/ACC‐like components displaying vague transitional zones. All components consistently showed diffuse p16, p63 and SOX2, variable cytokeratin (CK)7 and CK17 and rare Ber‐EP4 and MYB expression; there was a substantially lower Ki67 index in pure ABCs and the ABC‐like components. The ACC‐like components showed no myoepithelial differentiation (SMA, calponin and S100) and MYB gene fusions. CK7, CK17 and Ber‐EP4 were characteristically stronger in BSCCs than in the mixed carcinomas (P < 0.01). High‐risk HPV (HR‐HPV) E6/E7 mRNA was detected in 12 mixed carcinomas and seven BSCCs, but not in pure ABCs. The HR‐HPV mRNA expression was higher in the SCC components and BSCCs than in the ABC‐like components of mixed carcinomas (P < 0.05).ConclusionsThe ACC‐like components in mixed carcinomas probably represent the morphological mimics of salivary ACCs. ABC‐like components may be the potential precursor of the ACC‐like and SCC components. HR‐HPV oncogenes may play a role in the pathogenesis of SCCs with ABC‐/ACC‐like features.

PD‐L1 expression in vulvar cancer: a systematic review and meta‐analysis

Programmed cell death ligand‐1 (PD‐L1) expression in cancer may predict clinical response to immunotherapeutic treatment with PD‐1/PD‐L1 inhibitors. Within the vulvar cancer field, PD‐L1 expression has only been assessed by a few studies. We conducted a meta‐analysis to examine the prevalence of PD‐L1 positivity in vulvar cancer. PubMed, Embase, and Cochrane were searched for articles reporting on PD‐L1 expression in vulvar cancer. Study selection and data extraction were performed independently by two authors. We extracted data on PD‐L1 prevalence in vulvar cancer according to combined positive score (CPS) and tumour proportion score (TPS). Cutoff values for positivity were ≥1 or ≥10 for CPS and ≥1% and ≥5% for TPS. Random‐effects models were used to estimate pooled PD‐L1 prevalence, with 95% confidence intervals (CIs). Tests of between‐study heterogeneity were evaluated by the I2 statistics. Sources of heterogeneity were explored by subgroup analyses and meta‐regression. In total, 19 studies were included. Pooled PD‐L1 prevalence in vulvar cancer was 83.4% (95% CI: 70.8–91.3; I2 = 80.0) and 53.9% (95% CI: 37.4–69.6; I2 = 93.0) according to CPS and TPS, respectively. Based on TPS, human papillomavirus (HPV)‐associated vulvar squamous cell carcinomas (SCC) showed a lower PD‐L1 prevalence (39.9%; 95% CI: 13.3–74.2) compared with HPV‐independent SCC (62.6%; 95% CI: 33.7–84.6), but meta‐regression showed no significant variation in PD‐L1 prevalence by HPV status. PD‐L1 prevalence was similar in advanced (44.9%; 95% CI: 29.8–61.1) and localized vulvar cancer (56.7%; 95% CI: 18.9–76.7). In conclusion, PD‐L1 expression in vulvar cancer is frequent but between‐study heterogeneity was high. Based on a subgroup of heterogenous studies, we found no strong variation in PD‐L1 prevalence according to HPV status and stage.

Loss of ARID1B and SMARCB1 expression are specific for the diagnosis of dedifferentiated/undifferentiated carcinoma in tumours of the upper gynaecological tract and cervix

Aims Genomic inactivation of ARID1B in ARID1A‐inactivated tumour and genomic inactivation of SMARCB1 represent two recurrent mechanisms, core SWItch/sucrose non‐fermentable (SWI/SNF) complex inactivation, that are associated with de‐differentiation in endometrial carcinoma. Approximately one‐third of dedifferentiated/undifferentiated endometrial carcinomas (DDEC/UEC) show loss of ARID1B expression with a minor subset showing loss of SMARCB1 expression, but little is known regarding the specificity of ARID1B or SMARCB1 loss in gynaecological tract tumours in general. The aim of this study was to examine the frequency of ARID1B and SMARCB1 loss by immunohistochemistry in a series of gynaecological tract epithelial/mesenchymal neoplasms. Methods and results We evaluated 1849 tumours that included 748 endometrial carcinomas, 101 uterine carcinosarcomas/adenosarcomas, 64 uterine sarcomas, 221 cervical carcinomas and 715 ovarian carcinomas/borderline tumours by tissue microarrays (TMA). We observed ARID1B loss in 35 of 86 (41%) and SMARCB1 loss in three of 86 (3%) DDEC/UEC, but not in any other uterine tumour types examined. ARID1B‐deficient DDEC/UEC also showed concurrent loss of ARID1A expression. All SMARCB1‐deficient tumours showed loss of MLH1 and PMS2, while 29 of 35 ARID1B‐deficient tumours showed loss of MLH1 and PMS2 or loss of MSH6. All ovarian carcinomas/borderline tumours and cervical carcinomas showed intact expression of ARID1B and SMARCB1. Conclusion Our findings indicate that the loss of expression of ARID1B or SMARCB1 by immunohistochemistry is highly specific for undifferentiated carcinoma among tumours of the upper gynaecological tract and cervix, and therefore can be used to identify these highly aggressive malignant tumours.

A Subset of SMARCB1 (INI‐1)‐deficient vulvar neoplasms express germ cell markers

AimsSMARCB1 (INI‐1)‐deficient vulvar neoplasms comprise a group of rare tumours that include epithelioid sarcoma (ES), myoepithelial carcinoma (MEC), the recently described myoepithelioma‐like tumour of the vulvar region (MELTVR), and sarcomas that are difficult to classify. It has been suggested that so‐called vulvar yolk sac tumours (YST) may represent morphologic variants of SMARCB1‐deficient tumours; thus, we investigated the immunoreactivity of germ cell markers in SMARCB1‐deficient vulvar neoplasms.Methods and resultsTen SMARCB1‐deficient vulvar neoplasms were stained with germ cell tumour markers (SALL4, glypican‐3, OCT3/4, and AFP) and re‐reviewed for morphologic features. The tumours occurred in adult females (median age 41 years) and included ES (n = 7), MELTVR (n = 2), and MEC (n = 1). All cases showed loss of SMARCB1 expression. Four cases (40%) were focally positive for SALL4 in areas with morphology of typical‐appearing ES. One of these cases also showed focal staining for OCT3/4. One ES showed a transition from typical‐appearing ES to YST‐like morphology, with diffuse expression of SALL4 and glypican‐3, and focal expression of AFP, in these latter areas. All other tested cases were negative for AFP.ConclusionOur study reveals that SALL4, glypican‐3, and OCT3/4 are positive in a subset of SMARCB1‐deficient vulvar neoplasms, which may pose a diagnostic challenge and result in consideration of a germ cell tumour. We also highlight a case with transition from ES to YST‐like morphology, lending further support that YSTs of the vulva are somatically derived SMARCB1‐deficient neoplasms and do not represent true germ‐cell neoplasia.

DEVIL, VAAD and vLSC constitute a spectrum of HPV‐independent, p53‐independent intra‐epithelial neoplasia of the vulva

AimsWe aimed to characterise a large cohort of non‐invasive, human papillomavirus (HPV) and p53‐independent verruciform lesions, such as ‘vulvar acanthosis with altered differentiation’ (VAAD), ‘differentiated exophytic vulvar intra‐epithelial lesion’ (DEVIL) and ‘verruciform lichen simplex chronicus’ (vLSC).Methods and resultsFrom January 2008 to December 2020 we retrospectively identified 36 eligible patients with verruciform non‐invasive lesions (n = 36) and collected clinical, histological and follow‐up parameters. Verruciform non‐invasive lesions occurred at a median age of 71 years, with a median follow‐up of 33.5 months. Clinically, pruritus was only reported in patients with VAAD (n = 3, 21%). Lesion colour was significantly different across categories (P = 0.028). Apart from the histopathological criteria already known to distinguish these entities (hypogranulosis, epithelial pallor and low‐magnification architecture), no other significant criteria were discovered and significant overlap was observed, particularly between VAAD and DEVIL. Patients with vLSC trended towards longer survival without recurrence compared to VAAD and DEVIL (P = 0.082), but showed comparable invasion‐free survival interval (P = 0.782). Squamous cell carcinomas (SCC) associated with either VAAD, DEVIL or vLSC displayed similar clinical, histopathological and biological parameters. In non‐invasive precursor lesions, stromal oedema was associated with invasion (P = 0.015) and remained so upon Cox regression analysis (P = 0.009).ConclusionOur study of HPV and p53 independent non‐invasive verruciform lesions of the vulva highlights significant clinical, histopathological and biological overlap between VAAD, DEVIL and vLSC, suggesting that these pre‐invasive lesions should be viewed as a spectrum. We also show that stromal features such as oedema might play an import role in progression to invasion.

Comparison of p53 immunohistochemical staining in differentiated vulvar intraepithelial neoplasia (dVIN) with that in inflammatory dermatoses and benign squamous lesions in the vulva

AimsDifferentiated vulvar intraepithelial neoplasia (dVIN), the precursor lesion to human papillomavirus‐independent vulvar squamous cell carcinoma (VSCC), can be difficult to distinguish from vulvar inflammatory dermatoses. Our goal was to determine if p53 could be a useful biomarker for dVIN, by characterizing p53 percentage, intensity and patterns of staining in dVIN and its histological mimics.Methods and resultsWe studied p53 immunohistochemical staining patterns in 16 dVIN cases and 46 vulvar non‐neoplastic squamous lesions [12 lichen sclerosus (LS); seven lichen simplex chronicus; three lichen planus (LP); six psoriasis; 13 spongiotic dermatitis (SPO); and five candidiasis]. dVIN cases were adjacent to a p16‐negative invasive VSCC in resection specimens. All dVIN cases showed null‐type or moderate to strong uniform p53 staining in >70% of basal cells, with moderate to strong continuous parabasal staining extending to two‐thirds of the epidermis. This was in contrast to weak or weak to moderate patchy p53 staining in the majority of other lesions. Moderate to strong and increased basal p53 staining (≥70%) was also observed in a subset of LS cases (5/12, 42%), LP cases (1/3, 33%), and SPO cases (36%, 4/11); however, in all categories, this was limited to the basal layer, and any staining in the parabasal layers was patchy.ConclusionStrong and uniform p53 staining of basal cells, extending into the parabasal layers, and a complete absence of staining (null type) is useful in distinguishing dVIN from other mimics in the vulva. p53 staining of lesser intensity or quantity, particularly basal overexpression only, overlaps with that in vulvar inflammatory lesions.

Programmed death‐ligand 1 expression in human papillomavirus‐independent cervical adenocarcinoma and its prognostic significance

AimsIn the 2020 World Health Organization classification of female genital tumours, endocervical adenocarcinomas (ECAs) are subclassified into human papillomavirus (HPV)‐associated (HPVA) and HPV‐independent (HPVI) groups on the basis of their distinct aetiologies and clinical behaviours. The aim of this study was to investigate programmed death‐ligand 1 (PD‐L1) expression and its prognostic value in HPVI ECA and HPVA ECA, and compare these between the two entities.Methods and resultsA total of 93 ECAs accessioned between 2013 and 2020 were selected for further analysis, including 48 usual‐type HPVA ECAs and 45 HPVI ECAs. Then, we evaluated PD‐L1 expression in whole tissue sections of these cases by using the tumour proportion score (TPS) and the combined positive score (CPS). Heterogeneous PD‐L1 expression was observed in both HPVI ECAs and usual‐type HPVA ECAs. However, no significant difference in PD‐L1 expression was seen among different histological types of ECA when either the CPS or the TPS was used. Gastric‐type ECA (GEA) was associated with higher clinical stage (P = 0.001), worse progression‐free survival (PFS) (P = 0.008) and worse overall survival (OS) (P = 0.02) than usual‐type HPVA ECA and non‐GEA HPVI ECA. When the TPS was used, PD‐L1‐positive GEA was associated with significantly worse PFS (P = 0.03) and OS (P = 0.015) than PD‐L1‐negative GEA.ConclusionsOur data show frequent PD‐L1 expression in HPVI ECAs, supporting the potential role of the programmed cell death protein 1/PD‐L1 pathway as a therapeutic target for these tumours. Our data also support PD‐L1 as a negative prognostic marker associated with a potentially unfavourable outcome for GEAs.

Surface prostatic metaplasia, transitional cell metaplasia and superficial clusters of small basophilic cells in the uterine cervix: prevalence in gender‐affirming hysterectomies and comparison with benign hysterectomies from cisgender women

AimsAs gender‐affirming surgery is becoming more common, it is important for pathologists to recognize potential benign findings to avoid misinterpretation. Cervical transitional cell metaplasia and superficial clusters of small basophilic cells have been described in the context of gender‐affirming testosterone therapy; these findings may be misdiagnosed as high‐grade squamous intraepithelial lesions or endometrial cells on Pap smears. Prostatic metaplasia has been reported in the surface squamous epithelium of the vagina and the uterine cervix in individuals undergoing gender‐affirming androgen therapy; this finding is often associated with NKX3.1‐positive basal keratinocytes. The aim of this study was to assess the morphological and immunohistochemical features of the uterine cervix in gender‐affirming hysterectomies in comparison with benign hysterectomies from cisgender women.Methods and resultsWe assessed the morphological and immunohistochemical features of the uterine cervix in 49 gender‐affirming hysterectomies as compared with 57 hysterectomies from cisgender patients to establish the relative prevalences of surface prostatic metaplasia, NKX3.1‐positive basal keratinocytes, transitional cell metaplasia, and small basophilic cells in the cervical squamous epithelium. The cervical tissue from the gender‐affirming therapy cohort showed significantly higher prevalences of NKX3.1‐positive basal keratinocytes (86% versus 1.8%), transitional cell metaplasia (80% versus 3.5%), superficial clusters of small basophilic cells (67% versus 7%), and surface prostatic metaplasia (43% versus 3.5%).ConclusionNKX3.1‐positive basal keratinocytes, transitional cell metaplasia, small basophilic cells and surface prostatic metaplasia are all more prevalent in the cervices of individuals receiving gender‐affirming testosterone therapy; awareness of this fact allows pathologists to avoid the overdiagnosis of dysplasia or the recommendation of unnecessary follow‐up procedures.

Isolated morular squamous metaplasia in endometrial biopsies and curettings: is there a role for repeated sampling?

Aims Endometrial atypical hyperplasia and low‐grade endometrioid carcinoma are often associated with morular squamous metaplasia. Limited evidence suggests that the finding of isolated morular squamous metaplasia without concomitant glandular neoplasia in biopsies is associated with a 6.5% risk of endometrial cancer in subsequent samples and warrants close follow‐up. However, its prognostic value has not been clearly determined. Methods and results The Massachusetts General Hospital pathology database was queried to identify endometrial samples with a diagnosis of ‘adenoacanthosis’, ‘morular metaplasia’, ‘squamous metaplasia’ or ‘morular squamous metaplasia’ between 2012 and 2024. Cases associated with endometrioid carcinoma, non‐atypical or atypical hyperplasia, atypical polypoid adenomyoma and gland crowding insufficient for diagnosis of atypical hyperplasia were excluded. Clinicopathologic data were collected. Outcomes were categorized as regression, persistence and progression to carcinoma. Of 32,800 endometrial samples reported during the study period, isolated morular squamous metaplasia was diagnosed in 57 (0.17%), including 42 (73.7%) biopsies and 15 (26.3%) curettings. The median patient age was 45 (21–70) years. Histologic follow‐up (at least one follow‐up sample) was available in 22 patients (median 30 months, 1–120) and included endometrial biopsy, curettage or hysterectomy. Of these 22 patients, a single follow‐up biopsy was performed in 9 (40.9%), a single curettage in 1 (4.5%), hysterectomy in 6 (27.3%), a single biopsy followed by hysterectomy in 2 (9.1%), multiple biopsies in 3 (13.6%) and a curettage followed by multiple biopsies and hysterectomy in 1 (4.5%). The median number of follow‐up samples was 2 (2–9) per patient. Histologically, the follow‐up samples were unremarkable (regression) in most patients (19/22, 86.4%), 2 (9.1%, aged 49 and 57 years) were diagnosed with grade 1 endometrioid carcinoma in subsequent hysterectomies, and 1 (4.5%, age 62) had persistent squamous morular metaplasia (follow‐up 69 months). In 15 patients with clinical follow‐up but no further pathology sampling, none had clinical symptoms at their last visit (100% clinical regression). Thus, the overall rate of endometrioid carcinoma was 5.4% (2/37). Conclusions Isolated squamous morular metaplasia without associated glandular neoplasia is a rare finding, reported only in 0.17% of endometrial samples. The risk of subsequent endometrioid neoplasia appears to be low (5.4%), although the possibility of undersampled atypical hyperplasia/endometrioid carcinoma cannot be completely ruled out without additional sampling. Persistent squamous morular metaplasia is relatively uncommon (4.5%) and may not lead to the subsequent diagnosis of endometrial cancer, questioning the utility of numerous repeat samplings in patients without progression after one repeat sample.

Mucinous cystic neoplasm in men: a comparative study

AimsMucinous cystic neoplasms (MCN) are defined by the presence of an ovarian‐type stroma (OTS). Once ovarian stroma has become a requirement for the diagnosis of MCNs, studies using this criterion have disclosed that MCNs are seen almost exclusively (97%) in women. The occurrence of MCNs (with ovarian stroma) in men is exceedingly rare and raises questions about the origin of OTS and the role of hormones in tumourigenesis. This study aims to investigate the clinical and histopathological features of pancreatic or hepatic MCNs in men.Methods and resultsWe examined the MCN cases in men and compared them with age‐matched women. We further investigated MCN in premenopausal and postmenopausal women to explore the impact of hormonal status.The stromal cellularity tended to be lower in men as compared to women; however, limited numbers prevented statistical significance. Clinical presentation, tumour location, body mass index and cyst complexity were similar between men and age‐matched women, similarly in pre‐ and postmenopausal woman groups. All MCN cases, regardless of gender or hormonal status, showed OTS with positivity for oestrogen receptor (ER), androgen receptor (AR) and SF‐1 immunostains. Additionally, the presence of AR and ER‐beta in lining epithelial cells, in addition to the stromal cells, raised the probability for a potential role for local hormonal signalling in the pathogenesis of these tumours.ConclusionsMCNs in men had overlapping histopathologic and immunohistochemical profiles with those in women, regardless of menopausal status.

Molecular and immunohistochemical characterization of ERBB2 activating mutations in low‐grade serous ovarian carcinoma

AimsLow‐grade serous carcinoma (LGSC) of the ovary presents unique therapeutic challenges due to its resistance to platinum‐based chemotherapies and a tendency to present at an advanced stage. Approximately 50% of LGSC possess activating mutations in KRAS, NRAS, and BRAF, a finding associated with better overall survival. However, many tumours lack obvious driver alterations against which to direct targeted treatment strategies, necessitating further investigation into molecular drivers of LGSC and their impact on clinical outcomes.Methods and ResultsWe conducted a retrospective analysis of 84 LGSC patients who underwent tumour‐only targeted next‐generation sequencing at our institution. Molecular data were correlated with clinical outcomes, HER2 immunohistochemistry, and supplemented with additional tumour sequencing data from the AACR GENIE cohort v15.1 (n = 295). Approximately 5% of LGSC cases across the combined cohort harboured activating alterations in ERBB2 (n = 17/369), which encodes the HER2 receptor tyrosine kinase. These alterations were mutually exclusive of other MAP kinase pathway mutations and included exon 20 insertions (n = 6), extracellular domain/transmembrane domain missense alterations (n = 4), and exon 16 skipping mutations (n = 7). ERBB2 exon 16 emerged as a mutational hotspot in LGSC when compared to other tumour types. Immunohistochemistry revealed variable HER2 expression patterns that were independent of ERBB2 mutational status. In our institutional cohort, patients with RAS/RAF mutant tumours (n = 38) showed better overall survival compared to RAS/RAF wildtype tumours (n = 35). No tumours in our internal cohort (n = 84) harboured ERBB2 amplifications.ConclusionAs the landscape of HER2‐directed therapies continues to evolve, these findings suggest that ERBB2 alterations and HER2 expression may represent a potential therapeutic target in LGSC.

Pattern A endocervical adenocarcinomas with ovarian metastasis are indolent and molecularly distinct from destructively invasive adenocarcinomas

AimsThe invasive pattern in HPV‐associated endocervical adenocarcinoma (HPVA) has prognostic value. Non‐destructive (pattern A) HPVA has excellent prognosis mirroring adenocarcinoma in‐situ (AIS). However, the rare occurrence of ovarian spread in these tumours suggests aggressiveness in a subset of patients with these otherwise indolent lesions. We hypothesise that AIS/pattern A HPVA with ovarian metastases are biologically different than metastatic destructively invasive HPVA.Methods and resultsSamples from patients with HPVA and synchronous or metachronous metastases were retrieved and reviewed to confirm diagnosis and determine the Silva pattern in the primary lesion. For each case, normal tissue, cervical tumour and at least one metastasis underwent comprehensive sequencing using a 447‐gene panel. Pathogenic single‐nucleotide variants and segmental copy‐number alterations (CNA), tumour mutational burden and molecular signatures were evaluated and compared between primary and metastases and among invasive pattern categories. We identified 13 patients: four had AIS/pattern A primaries, while nine had pattern B/C tumours. All AIS/pattern A lesions had metastasis only to ovary; 50% of patients with ovarian involvement, regardless of invasive pattern, also had involvement of the endometrium and/or fallopian tube mucosa by HPVA. In the ovary, AIS/pattern A HPVA showed deceptive well‐differentiated glands, often with adenofibroma‐like appearance. Conversely, pattern C HPVAs consistently showed overt infiltrative features in the ovary. Sequencing confirmed the genetic relationship between primary and metastatic tumours in each case. PIK3CA alterations were identified in three of four AIS/pattern A HPVAs and three of eight pattern B/C tumours with sequenced metastases. Pattern C tumours showed a notably higher number of CNA in primary tumours compared to pattern A/B tumours. Only one metastatic AIS/pattern A HPVA had a novel pathogenic variant compared to the primary. Conversely, five of eight pattern B/C tumours with sequenced metastases developed novel pathogenic variants in the metastasis not seen in the primary. All four AIS/pattern A patients were alive and free of disease at 31, 47, 58 and 212 months after initial diagnosis. Conversely, cancer‐related death was documented in five of nine pattern B/C patients with follow‐up at 7, 20, 20, 43 and 87 months.ConclusionMorphologically and genomically, AIS/pattern A HPVA with secondary ovarian involvement appears distinct from destructively invasive tumours. In at least a subset of these cases, ovarian spread appears to occur via trans‐Mullerian superficial extension, different from the stromal and lymphatic vascular spread typical of more aggressive tumours (pattern C). These differences may explain the indolent outcome observed in the rare subset of patients with AIS/pattern A HPVA and ovarian metastasis. Our data underscore the potential for conservative surgical management approaches to pattern A HPVA.

Molecular classification of metastatic and recurrent endometrial endometrioid carcinoma: prognostic relevance among low‐ and high‐stage tumours

AimsMolecular classification according to The Cancer Genome Atlas (TCGA) improves endometrial endometrioid carcinoma (EEC) prognostication and has specific treatment implications; however, original data were skewed towards low‐grade and low‐stage tumours. Herein, we molecularly classify EECs metastatic at the time of diagnosis or with subsequently documented recurrent/metastatic disease to examine correlation with clinical outcomes.MethodsTCGA categories include POLE‐mutated, microsatellite instability (MSI), p53 abnormal (p53 abnl) and no specific molecular profile (NSMP). POLE targeted sequencing at exons 9, 11, 13 and 14 and immunohistochemistry (IHC) for PMS2, MSH6 and p53 were performed to establish molecular classification.ResultsThe distribution in our cohort of 141 EECs was similar to that generally reported in EEC, with nine POLE‐mutated (6%), 45 MSI (32%), 16 p53 abnl (11%) and 71 NSMP (50%), with similar distributions between low‐ and high‐stage cohorts. We demonstrate that when stratified by molecular subtype, disease‐specific survival from the time of high‐stage (stages III–IV) presentation or time of recurrence in low‐stage (stages I–II) disease among metastatic and/or recurrent EEC is strongly associated with TCGA classification (high‐stage P = 0.02, low‐stage P = 0.017). Discordant molecular classification between primary and metastatic/recurrent tumours occurred in four of 105 (3.8%) patients, two related to PMS2/MSH6 IHC and two related to p53 IHC.ConclusionsWe demonstrate that molecular classification is prognostically relevant not only at the time of diagnosis, but also at the time of recurrence and in the metastatic setting. Rare subclonal alterations occur and suggest a role for confirming TCGA classification in recurrent/metastatic tumours.

Correlation between P53 immunohistochemical staining and TP53 molecular testing in endometrial carcinomas: a detailed assessment of discrepant cases with implications for patient management

Aims The 2013 Cancer Genome Atlas (TCGA) study identified four molecular types of endometrial carcinoma (EC) that are prognostic and predictive of therapy response. The p53 abnormal (p53abn) group of tumours is associated with aggressive clinical behaviour, chemoresponsiveness and generally high‐grade histology. p53abn tumours may be identified by p53 immunohistochemical staining (a surrogate marker) or molecular testing. In this study, we evaluated the concordance between p53 immunohistochemistry and TP53 molecular testing in a consecutive cohort of ECs from a population‐based setting. Our aim was to investigate the rate of concordance and reasons for discordance between the immunohistochemistry and molecular testing and to provide recommendations for pathologists and clinicians dealing with these discordant cases. Methods and results A total of 386 ECs were included where all biopsy specimens underwent molecular testing using a next‐generation sequencing (NGS) panel (including POLE and TP53 genes and MSI testing) and immunohistochemistry for oestrogen receptor (ER), p53 and mismatch repair (MMR) proteins. Concordance between p53 immunohistochemistry and TP53 NGS was initially 88.6% (discordance of 11.4%) following review of the pathology and molecular reports; most of the discordant cases comprised carcinomas with wild‐type p53 immunohistochemistry but TP53 mutations identified on NGS. The discordance reduced to 6.5% after review of the p53 stained slides, which revealed subclonal mutation‐type staining in some tumours, and to 5% after excluding POLE mutated and mismatch repair deficient carcinomas. However, there remained a small cohort of 19 POLE wild‐type/MMR proficient carcinomas (8 low‐grade endometrioid, 9 high‐grade endometrioid, 2 carcinosarcomas), with wild‐type p53 staining but with TP53 mutations on NGS. Altogether, there were 12 POLE wild‐type/MMR proficient low‐grade endometrioid carcinomas with TP53 mutations on NGS; all were stage I (11 IA, 1 IB). Conclusions Our study demonstrated a good overall concordance between p53 immunohistochemical staining and TP53 molecular results. The concordance can be increased by reviewing the p53 stained slides in discrepant cases but there remains a small cohort of cases, mostly low‐grade endometrioid carcinomas ( POLE wild‐type/MMR proficient), where TP53 mutations are present on NGS but p53 immunohistochemistry is wild‐type. Such cases present a dilemma for the pathologist (which TCGA group should they be placed into) and the clinician (should adjuvant therapy be instigated based on the presence of a TP53 mutation alone with no other adverse features). For now, we advise classifying such cases as p53abn but not to administer adjuvant therapy based on the presence of a TP53 mutation alone without other adverse pathological factors. The significance of TP53 mutations in such cases should be determined by larger studies with long‐term follow‐up.

Challenges of a tailored immunohistochemistry algorithm for uterine leiomyosarcoma: an integrated analysis of leiomyomas with bizarre nuclei and fumarate hydratase (FH) deficiency

AimsLeiomyomas (LM) are the most common uterine mesenchymal neoplasms and encompass a variety of histological subtypes. Bizarre nuclei are described in both leiomyomas with bizarre nuclei (LM‐BN) and fumarate hydratase‐deficient leiomyomas (FH‐LM), which raise diagnostic concerns regarding leiomyosarcoma (LMS). Recently, an immunohistochemical algorithm to support the diagnosis of LMS based on the genomic landscape of these neoplasms was proposed. This study aimed to evaluate the algorithm's accuracy in distinguishing LM‐BN and FH‐LM from LMS.Methods and ResultsWe collected 68 LM (29 LM‐BN, 30 FH‐LM, and 9 LM) and 9 LMS, along with clinicopathological and molecular data. An immunohistochemical panel comprising p53, Rb, PTEN, ATRX, DAXX, and MDM2 was applied. Nine cases were non‐interpretable due to fixation issues. The algorithm demonstrated 100% accuracy for LM without bizarre nuclei (9/9) and for nonmyxoid LMS (5/5). Notably, 28.6% (14/49) of LM‐BN and FH‐LM exhibited at least two abnormalities, leading to potential misclassification as LMS. However, their clinical course, morphology, and genomic profile supported a benign diagnosis. Frequent alterations included Rb (20/49; 40.8%) and p53 (19/49; 38.8%), particularly in bizarre cells, while no abnormal staining was observed for ATRX, DAXX, or MDM2.ConclusionThe proposed algorithm has limitations in differentiating LMS from LM‐BN and FH‐LM, misclassifying 28.6% of the latter. Accurate interpretation requires proper internal controls, particularly for markers whose loss of expression favours malignancy. Morphology remains central for diagnosis, although integration of molecular data may provide additional insights for a definitive classification in challenging cases.

Folate receptor alpha ( FRα / FOLR1 ) and HER2 immunohistochemical staining in high‐grade endometrial carcinoma with aberrant p53 expression

Aims Mirvetuximab soravtansine is an anti‐FOLR1 (folate receptor 1/alpha) antibody–drug conjugate with a companion diagnostic immunohistochemical (IHC) biomarker for platinum‐resistant ovarian, fallopian tube and primary peritoneal carcinoma. Its effectiveness has sparked interest in FOLR1 expression in endometrial carcinoma. We evaluate the relationship of FOLR1 and HER2‐IHC expression in high‐grade p53‐aberrantly expressed endometrial carcinomas, as overexpression of both is associated with high‐grade histologic features and aggressiveness. Methods and results Carcinomas were scored for HER2‐IHC by both gastric and endometrial serous criteria and FOLR1‐IHC by Ventana package insert on tissue microarray cores. Intratumoral heterogeneity was quantified for HER2 and FOLR1 expression across multiple cores from the same tumour. A total of 291 cores (226 endometrial serous, 47 high‐grade endometrioid and 18 high‐grade Müllerian carcinoma, nos) were collected from 66 cases (discrete accession dates) from 62 patients. When stratified by two‐category HER2‐IHC status (0/1+ vs. 2+/3+, either criteria), significantly more HER2 2+/3+ cores (16/133, 12%) were FOLR1‐positive by a ≥75% cut‐off than HER2 0/1+ specimens (8/158, 5%, P  = 0.031). Results were similar by a ≥25% cut‐off (49/133, 37% vs. 38/158, 24%; P  = 0.018). More cases with high HER2‐IHC heterogeneity (2–3‐pt difference) also demonstrated FOLR1‐IHC heterogeneity (≥50% score difference between cores from the same case) than cases with no or low (1‐pt) HER2‐IHC heterogeneity, though most cases did not show high HER2 or FOLR1 intratumoral heterogeneity (16.7%–18.2% 2–3‐pt difference and 12.1% ≥50% score difference). Conclusions A positive correlation between HER2 and FOLR1 overexpression may be seen in high‐grade endometrial carcinomas with aberrant p53 expression, which may extend to intratumoral heterogeneity of biomarker expression.

Initiation of molecular testing of endometrial carcinomas in a population‐based setting: practical considerations and pitfalls

AimsSince the publication of The Cancer Genome Atlas (TCGA) molecular Classification of endometrial carcinomas in 2013, multiple studies have demonstrated the prognostic and therapeutic importance of this. However, there is great variability on whether and how this is undertaken in different institutions, and this is often dependent on resources and availability of molecular testing. Points of controversy include whether molecular classification is needed on all endometrial carcinomas and whether pure molecular testing is undertaken or a surrogate such as the ProMisE (Proactive Molecular Risk Classifier for Endometrial Cancer) Classifier. Herein we report our experience instigating molecular classification of endometrial carcinomas in Northern Ireland.Methods and ResultsFrom 1st March 2023, all endometrial carcinomas diagnosed on biopsy in the four pathology laboratories in Northern Ireland were referred to the central molecular pathology laboratory for genomic analysis using a custom next‐generation sequencing (NGS) panel; the NGS panel included the entire coding regions of polymerase epsilon (POLE) and TP53 genes, as well as microsatellite instability (MSI) analysis. All cases also underwent immunohistochemical staining with oestrogen receptor (ER), p53, and the mismatch repair (MMR) proteins MLH1, PMS2, MSH2, and MSH6. The molecular results were available by the time of surgery (if a hysterectomy was performed) allowing integration into the final pathology report where a TCGA molecular type was assigned. Two hundred and sixty‐seven endometrial carcinomas underwent molecular testing; in five cases, there was insufficient material for testing, leaving 262 cases. The TCGA groups were POLEmut (19; 7.3%), MMRd (63; 24%), p53abn (62; 23.7%), and no specific molecular profile (NSMP) 118 (45%). Seventeen tumours (6.5%) were “multiple‐classifiers”: five POLEmut‐p53abn, two POLEmut‐MMRd, one POLE‐MMRd‐p53abn (all included in the POLEmut TCGA group), and nine MMRd‐p53abn (included in the MMRd group).ConclusionThis represents one of the first population‐based studies investigating the prevalence of the different TCGA molecular groups of endometrial carcinomas in an unselected population. Performing molecular testing on biopsies enables management to be tailored to the molecular group and allows integration of the TCGA group into the report of the final resection specimen. We hope our experience will facilitate other laboratories in undertaking TCGA molecular classification.

SOX17 expression in mesonephric‐like adenocarcinomas and mesonephric remnants/hyperplasia of the female genital tract: Expanding its utility as a Müllerian biomarker

AimsRecently, SOX17 has emerged as a promising biomarker for non‐mucinous Müllerian (ovarian and endometrial) carcinomas, demonstrating increased specificity in comparison to PAX8 while maintaining similar sensitivity. However, expression of SOX17 in mesonephric‐like adenocarcinoma (MLA), a carcinoma of the female genital tract with uncertain, but probably Müllerian histogenesis, remains unexplored. This study aims to address this gap.Methods and resultsSOX17 immunohistochemistry was performed on whole tissue sections from 68 MLAs originating from the endometrium or ovary and seven cervical mesonephric carcinomas, as well as six mesonephric remnants/hyperplasias. Using a four‐tiered scoring system based on distribution and intensity of staining, 68% of MLA displayed a negative/low (< 10%) SOX17 expression pattern, which contrasts with the high expression observed in most Müllerian carcinomas. However, 22% of MLA demonstrated high SOX17 expression, similar to other endometrial and ovarian carcinomas. Similarly, five of seven (72%) mesonephric carcinomas of the cervix were SOX17‐negative, but two cases (28%) were positive. All mesonephric remnants/hyperplasias were SOX17 negative.ConclusionsThe majority of MLA are negative or exhibit low SOX17 expression, in contrast to the diffuse and strong expression commonly seen in other types of Müllerian carcinoma. However, a subset of MLAs demonstrate high SOX17 expression. Therefore, absence of SOX17 staining is supportive for MLA when the differential includes another non‐mucinous Müllerian carcinoma. SOX17 may also be useful for differentiating mesonephric remnants/hyperplasias from Müllerian malignancies and benign Müllerian glandular lesions.

Prognostic performance of FIGO 2023 endometrial carcinoma staging: a comparison to FIGO 2009 staging in the setting of known and unknown molecular classification

AimsThe 2023 FIGO staging criteria for endometrial cancer (EC) introduced marked changes from the 2009 version. The full implication of these changes for patient diagnosis and treatment is unknown. We evaluate the differences in staging and prognostication between the two systems, with and without inclusion of molecular classification.Methods and resultsWe assigned (1) FIGO 2009, (2) 2023 molecular‐agnostic and (3) 2023 molecular‐informed stages to 404 fully staged and molecularly classified patients with EC. Disease‐specific and progression/relapse‐free survival were analysed via the Kaplan–Meier method and compared with log‐rank testing; 118 of 252 (47%) FIGO 2009 stage I patients were upstaged based on histopathological findings alone. Stage I/II subgroup survival distribution analysis showed a worse prognosis in FIGO 2023 IIB and IIC patients. In the molecular‐informed FIGO 2023 system, three of 15 (20%) POLE‐mutated stage I/II cases were downstaged from FIGO 2009 and eight (53%) were downstaged from molecular‐agnostic FIGO 2023. Fifty‐one of 60 (85%) p53‐abnormal tumours were upstaged from the FIGO 2009, whereas 13 of 60 (22%) were upstaged from the 2023 molecular‐agnostic stage. Molecular classification improved prognostic stratification for both 2009 and 2023 FIGO systems.ConclusionsDownstaging based on POLE mutation more accurately represents patient outcomes. However, in the absence of known POLE status, applying molecular‐agnostic FIGO 2023 criteria for stage I/II disease should be conducted with caution. For aggressive histotypes, additionally reporting FIGO 2009 stage should be considered. Upstaging based on substantial lymphovascular space invasion, aggressive histotype with any myometrial invasion and abnormal p53 improves prognostic discernment. Further subdivisions within stage I/II provide minimal additional prognostic information.

Genomic profiling of dedifferentiated endometrial carcinomas arising in the background of high‐grade carcinoma: a targeted next‐generation sequencing study

AimsOur understanding of dedifferentiated endometrial carcinoma (DEC), a rare and aggressive malignancy, mainly reflects undifferentiated carcinomas (UC) arising in the setting of low‐grade endometrial cancer (DEC‐LG). However, cases of UC arising in the setting of high‐grade EC (DEC‐HG) have been noted in the literature. Our knowledge of the genomics of DEC‐HG is limited. To characterise the molecular landscape of DEC‐HC, targeted genomic sequencing and immunohistochemical analysis was carried out on seven DEC‐HG and four DEC‐LG.Methods and resultsDEC‐HG and DEC‐LG, including undifferentiated and differentiated components, both showed a similar frequency and spectrum of mutations. ARID1A mutations were identified in 6/7 (86%) DEC‐HG and 4/4 (100%) DEC‐LG, while SMARCA4 mutations were present in 4/7 (57%) DEC‐HG and in 1/4 (25%) DEC‐LG. Concurrent SMARCA4/BRG1 protein loss by immunohistochemistry was observed in 3/4 and 1/1 SMARCA4 mutated DEC‐HG and DEC‐LG, respectively. Neither genomic alterations nor protein loss in SMARCB1/INI1 were observed in any of our cases. TP53 mutations were detected in 4/7 (57%) DEC‐HG and in 2/4 (50%) DEC‐LG, while mutation‐pattern p53 immunohistochemistry expression was observed in 2/7 (29%) DEC‐HG and none of the DEC‐LG. MLH1 mutations were observed in 1/7 (14%) DEC‐HG and 1/4 (25%) DEC‐LG. MSH2 and MSH6 mutations were each detected in 1/7 (14%) DEC‐HG, but neither was associated with corresponding loss of protein expression.ConclusionThe findings support expanding the definition of DEC to include DEC‐HG, a previously under‐recognised phenomenon with genomic similarities to DEC‐LG.

ZEB1 expression is frequently detected in undifferentiated and de‐differentiated carcinomas, but is not specific among endometrial carcinomas

AimsThe pathological diagnosis of undifferentiated and de‐differentiated endometrial carcinomas (UC/DCs) is prognostically important. However, undifferentiated components may be confused with other subtypes, particularly grade 3 endometrioid carcinomas (G3ECs). Zinc finger E‐box binding homeobox 1 (ZEB1) has recently been identified as a promising marker because it is frequently expressed in the undifferentiated components of UC/DCs, but not in other carcinomas. Therefore, we herein evaluated the diagnostic utility of ZEB1 with an emphasis on distinguishing between UC/DCs and G3ECs using an expanded cohort of endometrial carcinomas and carcinosarcomas.Methods and resultsImmunostaining for ZEB1 was performed on whole‐tissue sections of 19 UC/DCs, 194 non‐UC/DC endometrial carcinomas and 29 carcinosarcomas. Staining was defined as negative (< 5%), focal (5–50%) and diffuse expression (> 50%). ZEB1 was expressed in 84% of the undifferentiated components of UC/DCs (diffuse in 14, focal in two). Focal expression was observed in eight non‐UC/DC endometrial carcinomas and diffuse expression in seven, with the latter comprising G3ECs (four of 76), serous carcinoma (one of 37), clear cell carcinoma (one of 21) and neuroendocrine carcinoma (one of three). Epithelial differentiation was morphologically and immunohistochemically less evident in G3ECs and neuroendocrine carcinoma with diffuse ZEB1 expression. All carcinosarcomas showed diffuse ZEB1 expression in their sarcomatous components.ConclusionImmunostaining for ZEB1 was sufficiently sensitive to detect undifferentiated components. Diffuse ZEB1 expression showed high specificity for distinguishing between undifferentiated components and G3ECs; however, ZEB1 expression was not entirely specific to UC/DCs. The integration of ZEB1 into the diagnosis of UC/DCs requires careful examination to exclude other tumours, such as less differentiated G3ECs, neuroendocrine carcinomas and carcinosarcomas.

Patterns of SATB2 and p16 reactivity aid in the distinction of atypical polypoid adenomyoma from myoinvasive endometrioid carcinoma and benign adenomyomatous polyp on endometrial sampling

AimsAtypical polypoid adenomyoma (APAM) is an uncommon uterine lesion composed of complex endometrioid glands with frequent squamous morular metaplasia and fibromuscular stroma. On endometrial curettage, biopsy or polypectomy specimens, the admixture of endometrioid glands and smooth muscle raises the differential diagnosis of myoinvasive endometrioid carcinoma. Reproductive‐age APAM patients may opt for fertility preservation, whereas myoinvasive carcinoma is treated surgically. One previous study reported an incidental finding that the stroma of APAM, in contrast to that of other polypoid lesions, was SATB2‐positive. APAM has also been reported to show increased stromal p16 staining. We aimed to assess whether SATB2 and p16 are useful stains for the distinction of APAM from myoinvasive carcinoma and benign adenomyomatous polyps.Methods and resultsCases of ‘atypical polypoid adenomyoma’ (n = 32), ‘adenomyomatous polyp’ (n = 39) and ‘myoinvasive endometrioid carcinoma’ (n = 30) were identified. Morphological features were assessed, along with the intensity and extent of SATB2 and p16 staining in the stromal component of each lesion. SATB2 expression was seen in the stromal components of 30 of 32 (94%) APAMs, versus none of 39 (0%) benign adenomyomatous polyps and five of 30 (17%) myoinvasive endometrioid carcinomas. Stromal p16 expression was seen in 31 of 31 (100%) APAMs, versus 20 of 39 (51%) benign adenomyomatous polyps and 12 of 30 (40%) myoinvasive endometrioid carcinomas.ConclusionsPatchy to diffuse SATB2 and block‐type p16 staining of fibromuscular stroma separating atypical endometrioid glands is more consistent with APAM than with myoinvasive endometrioid carcinoma. These stains are potentially useful adjuncts to careful morphological evaluation of endometrial biopsies/curettings.

Endometrial neuroendocrine carcinoma and undifferentiated carcinoma are distinct entities with overlap in neuroendocrine marker expression

AimsDedifferentiated endometrial carcinomas (DDECs)/undifferentiated endometrial carcinomas (UDECs) frequently harbour genomic activation of switch/sucrose non‐fermentable (SWI/SNF)‐complex proteins, and can show histological overlap with neuroendocrine carcinoma (NEC). The aim of this study was to compare the extent of the expression of neuroendocrine markers, SWI/SNF proteins and mismatch repair (MMR) proteins in DDEC/UDEC and NEC.Methods and resultsThe extent of expression of synaptophysin, chromogranin, CD56, ARID1A, ARID1B, SMARCA4, SMARCB1 and MMR proteins was evaluated by immunohistochemistry on 44 SWI/SNF‐deficient DDECs/UDECs and 15 NECs. Thirty‐three of 44 (75%) DDECs/UDECs showed expression of at least one neuroendocrine marker, with 18 of 44 (41%) expressing two or more neuroendocrine markers, whereas all 15 NECs showed expression of at least one neuroendocrine marker, with 14 of 15 (93%) expressing two or more neuroendocrine markers. Neuroendocrine marker expression in DDECs/UDECs was typically focal when present, with average extents of 17%, 4% and 8% for synaptophysin, chromogranin and CD56 in the positive cases, respectively, in contrast to 73%, 40% and 62% in the positive NEC cases, respectively. All 15 NECs showed intact expression of SWI/SNF‐complex proteins, except for one that showed isolated loss of ARID1A. Thirty‐eight of 44 DDECs/UDECs were MMR‐abnormal (34 with loss of MLH1 and PMS2, and four with loss of PMS2 alone), whereas all NECs retained MMR protein expression.ConclusionsOur study demonstrates frequent but typically focal neuroendocrine marker expression in SWI/SNF‐deficient DDECs/UDECs, whereas NECs typically express two or more neuroendocrine markers, with diffuse expression of at least one marker. ARID1B, SMARCA4 and SMARCB1 immunohistochemistry can be used to aid in the differentiation between DDEC/UDEC and NEC.

Endometrial endometrioid carcinoma, grade 1, is more aggressive in the elderly than in the young

AimsThe aim of this study was to characterise grade 1 (G1) endometrioid carcinoma in the elderly, by using clinicopathological features and immunohistochemical features of surrogate markers of molecular subtypes.Methods and resultsWe retrospectively analysed tumour samples from 268 patients with G1 endometrioid carcinoma (<40 years, n = 24; 40–59 years, n = 169; ≥60 years, n = 75) for whom long‐term clinical follow‐up data were available. G1 endometrioid carcinoma in the elderly (≥60 years) was characterised by frequent deep myometrial invasion, less frequent endometrioid intraepithelial neoplasia (EIN), lack of benign hyperplasia (BH), less frequent squamous differentiation, and occasional aberrant p53 expression. In contrast, this condition in the young (<40 years) was characterised by frequent EIN, BH, and squamous differentiation. Univariate analysis revealed that elderly status (≥60 years), International Federation of Obstetrics and Gynecology (FIGO) 2009 stage and aberrant p53 expression were significantly associated with shorter progression‐free survival, and multivariate analysis revealed that elderly status and FIGO 2009 stage were independently associated with a poor prognosis.ConclusionsG1 endometrioid carcinoma in the elderly is more aggressive than that in the young, and elderly status is an independent predictor of shorter progression‐free survival in this condition. We propose that type 1 tumours can be subdivided into type 1a (young age at onset and indolent) and type 1b (old age at onset and relatively aggressive).

Malignant tumours of the uterus and ovaries with Mullerian and germ cell or trophoblastic components have a somatic origin and are characterised by genomic instability

AimsTumours of the female genital tract with a combination of malignant Mullerian and germ cell or trophoblastic tumour (MMGC/T) components are usually diagnosed in postmenopausal women, and pursue an aggressive clinical course characterised by poor response to therapy and early relapses. These clinical features suggest that MMGC/T are somatic in origin, but objective molecular data to support this interpretation are lacking. This study evaluates the molecular features of nine MMGC/T, including seven tumours containing yolk sac tumour (YST), one tumour containing choriocarcinoma and one tumour containing epithelioid trophoblastic tumour. The objectives were to: (i) investigate whether MMGC/T show a distinct genetic profile and (ii) explore the relationship between the different histological components.Methods and resultsNext‐generation sequencing of paired samples demonstrated that the mutational profile of the Mullerian and non‐Mullerian components of the tumour were almost identical in all cases. Moreover, the driver mutations identified were those expected in the specific subtype of Mullerian component present in each case. In contrast, variants expected in postpubertal germ cell tumours and gestational trophoblastic tumours were not identified, and FISH for i(12p) was negative in all cases tested. In this study, mismatch repair‐proficient MMGC/T (eight of nine) were characterised by a complex copy‐number variant profile, including numerous focal, regional, arm‐level and chromosome‐level events.ConclusionsComparison of paired samples supports that the YST and trophoblastic tumour components of MMGC/T have a somatic origin and often show numerous copy‐number variants, suggestive of underlying genomic instability.

Metastases to the ovary arising from endometrial, cervical and fallopian tube cancer: recent advances

The introduction of genomic studies has enabled assessment of the clonality of synchronous tumours involving the ovary and other sites in the female genital tract in a definitive way. This has led to the abandonment of conventional approaches to primary site assignment, and the recognition that most such synchronous neoplasms are clonally related single tumours with metastatic spread, rather than independent primary tumours. These discoveries have implications for diagnostic practice, analogous to the gradual change over the last few decades in our approach to mucinous neoplasms of the ovary metastatic from the gastrointestinal tract. In this review, we first examine the routes of metastasis to the ovary, and then discuss the diagnostic and clinical implications of concurrent ovarian carcinomas arising in combination with endometrial, endocervical and tubal carcinomas. It is proposed that cases of primary low‐grade endometrioid endometrial carcinoma with a secondary unilateral ovarian tumour, both with indolent characteristics, may be classified as ‘FIGO stage IIIA‐simulating independent primary tumours’, with a comment that conservative management would be appropriate. It should be recognised that human papillomavirus‐associated endocervical adenocarcinomas may result in synchronous or metachronous ovarian metastases that appear to be unrelated to the primary tumour, and that these may be managed conservatively in the absence of other sites of disease. In cases of tubo‐ovarian high‐grade serous carcinoma, tubal intraepithelial or contralateral adnexal involvement should count as a pelvic disease site for staging purposes.

Intratumour heterogeneity in endometrial serous carcinoma assessed by targeted sequencing and multiplex ligation‐dependent probe amplification: a descriptive study

AimsEndometrial serous carcinoma (ESC) represents the most aggressive subtype of endometrial carcinoma (EC). According to The Cancer Genome Atlas (TCGA), ESC exhibits a genomic profile characterised by frequent TP53 mutations and somatic copy‐number alterations (SCNA). Several studies have suggested the role of intratumour heterogeneity (ITH) in tumour progression and therapy resistance, highlighting ITH as a challenge for personalised medicine. ITH is described as the co‐existence of clonal and subclonal cellular populations within a single tumour. To date, the extent and prevalence of ITH in ESC have not been fully evaluated. The aim of this study was to address ITH analysis in ESC. We performed a descriptive integrated molecular approach using targeted sequencing and multiplex ligation‐dependent probe amplification (MLPA) to identify mutations and SCNA patterns, respectively.Methods and resultsEight ESC were examined, selecting three tumour regions per case and their corresponding normal tissue. For targeted sequencing a gene panel of 40 genes based on TCGA and other survey data was performed. For MLPA different probe mixes were used to detect SCNA in 106 genes. Analysis of mutations and SCNA were performed in each sample and comparative analysis of the three tumour regions was also conducted. Targeted sequencing showed that mutations in TP53, PIK3CA and PPP2R1A were ubiquitous in all tumour regions. Moreover, MLPA results demonstrated a high frequency of SCNA, according to the already known presence of genomic instability in ESC. Unlike the homogeneous distribution of somatic mutations, SCNA exhibited ITH affecting targetable genes such as ERBB2.ConclusionsOur study suggests that somatic gene copy‐number alterations are the main source of ITH in ESC.

Clinicopathological and molecular analyses of linearly expanded epithelial cells with β‐catenin alterations, “β‐catenin signature”, in the normal fallopian tube

AimsRecent advances in next‐generation sequencing have made it clear that clonal expansion of cells harbouring driver gene mutations occurs in physiologically normal epithelium. Molecular analysis of tubal epithelium has been almost exclusively confined to the TP53 pathway, which is involved in serous carcinogenesis. Other oncogenic events have not been explored in detail. Here, we report the linear expansion of fallopian tubal epithelial cells exhibiting an altered β‐catenin profile (β‐catenin signature). Through molecular analyses, we determined the incidence and clinicopathological significance of β‐catenin signatures.Methods and resultsWe evaluated 64 specimens of surgically removed bilateral fallopian tubes. Thirty‐three β‐catenin signatures were identified in 13 cases (20.3%); these patients were significantly younger than those without β‐catenin signatures (median ages of 44 and 57 years, respectively, P = 0.0317). No correlation between β‐catenin signature and any clinical factor was observed. CTNNB1 mutations were detected in three of eight β‐catenin signatures when tissues were microdissected and subjected to Sanger sequencing in two representative cases.ConclusionsThis is the first report of the CTNNB1 mutation in clusters of morphologically bland tubal epithelial cells. The results of this study indicate that β‐catenin signatures are common, and they may be a part of diverse molecular alterations occurring in normal tubal epithelium.

Tumour budding in pretreatment cervical biopsies: a prognosticator for personalised therapy in the era of precision oncology

Aims Tumour budding (TB) is a noteworthy morphologic indicator for tumour microenvironment (TME) especially because it is detectable with routine haematoxylin and eosin (H&E) staining. Its prognostic relevance has been demonstrated across various cancers, but its significance in pretreatment biopsy specimens of cervical cancer is unknown. This is the first study to investigate the prognostic value of TB in pretreatment cervical biopsy. Additional TME features identifiable with H&E such as cell nest size (CNS) were evaluated. Methods and results A retrospective review was conducted on the 2018 International Federation of Gynaecology and Obstetrics (FIGO) stage IIVA cervical cancer patients ( N  = 182) who had completed standard treatment. In multivariate analysis, TB (hazard ratio [HR], 2.06) and CNS (HR, 2.16) independently predicted overall survival. While TB (AUC, 0.7065) slightly outperformed CNS (AUC, 0.6975) in discriminating overall survival, the combination of TB and CNS demonstrated the highest performance (AUC, 0.7192) in time‐dependent receiver operating characteristic analysis. Conclusions This study is the first to suggest TB in pretreatment biopsy specimens as a reliable morphologic prognosticator in cervical cancer. TME features may enhance precision oncology by offering insights into the individual tumour biology. The fact that these morphologic features are available from routine H&E slides, reserving immunohistochemistry or molecular analysis for indeterminate cases, is of particular value in low‐resource settings where the burden of cervical cancer is most significant.

Claudin‐18 expression in gastric type adenocarcinoma and HPV ‐associated adenocarcinoma of the uterine cervix

Aims Claudin‐18 (CLDN18) is both a marker for the gastric phenotype and a therapeutic target. However, little is known about its immunoexpression in endocervical adenocarcinomas (ECAs), particularly as detected using the clone 43‐14A antibody, or about the gene expression of its isoforms in ECAs. Methods and results We examined CLDN18, HIK1083, p16 and Rb expression by immunohistochemistry and high‐risk human papillomavirus (HR‐HPV) mRNA by in situ hybridization (ISH) in 121 ECAs, including 35 HPV‐independent adenocarcinomas (gastric type [GAS], n  = 24; non‐GAS, n  = 11) and 86 HPV‐associated ECAs. We also analysed mRNA expression of the CLDN18.1 (lung type) and CLDN18.2 (gastric type) isoforms by quantitative polymerase chain reaction (qPCR) in selected cases. CLDN18 positivity was detected in 8/24 (33%) GASs, 0/11 (0%) non‐GASs and 2/86 (2%) HPV‐associated ECAs, with positivity defined as staining in ≥75% of tumour cells, as in gastric cancer. When a 5% cut‐off was used, CLDN18 positivity was detected in 22/24 (92%) GASs, 0/11 (0%) non‐GASs and 6/86 (7%) HPV‐associated ECAs; CLDN18 expression was thus significantly associated with GAS histology ( P  < 0.0001). Among the 6 cases of HPV‐associated ECAs with CLDN18 expression (ranging from 5% to 80%), the histological patterns included a mix of usual and mucinous features in 4 cases, pure usual type in 1 and villoglandular variant in 1. Otherwise features such as p16 overexpression and the Rb partial loss pattern were consistent with those of HPV‐associated ECAs. Six of 22 (27%) CLDN18‐positive GASs were also positive for p16, but their other features—such as CLDN18 expression and the Rb preserved pattern—were the same as in p16 negative GASs. Expression of CLDN18.2 mRNA but not CLDN18.1 mRNA was confirmed in both GASs and HPV‐associated ECAs. Conclusions CLDN18 (43‐14A) emerged as a potential diagnostic and therapeutic marker for GAS. A minor subset of HPV‐associated ECAs also can be immunoreactive for CLDN18 and express CLDN18.2 mRNA, suggesting divergent gastric phenotypic differentiation. The caution is that GAS and HPV‐associated ECAs can share overlapping histological features and similar expression of CLDN18 and p16.

Further confirmation of a highly prognostic grading scheme based upon tumour budding and cell cluster size in cervical squamous cell carcinoma

AimsOur study aimed to further confirm the clinical significance of the tumour budding activity and cell nest size‐based (TBNS) grading scheme in cervical squamous cell carcinomas (SCC).Methods and resultsWe applied the TBNS system to assess the prognostic value in an institutional cohort of well‐annotated cervical SCC consisting of 312 consecutive cases with surgical resection, no neoadjuvant chemotherapy and higher than stage pT1a. We found that high budding activity, single cell and TBNS grade 3 were more frequently associated with a decreased overall survival (OS) time and disease‐free survival (DFS) time (P < 0.001) and several other clinicopathological factors, including lymphovascular space invasion, lymph node metastasis, advanced Federation of Gynecology and Obstetrics (FIGO) stage and deep invasion of the cervical wall (> 2/3) (P < 0.05). On multivariate analysis, TBNS grade 3 was an adverse indicator for OS and DFS independently of age, invasion of the cervical wall and FIGO stage (P < 0.05). By comparison, the conventional three‐tiered grading system was not associated with OS and DFS in cervical SCC (P > 0.05).ConclusionsOur study further confirms that the TBNS grading scheme is robust in prognostic assessment in cervical SCC that outperforms the conventional three‐tiered grading system. It is applicable to add TBNS grade into routine diagnostic practice.

STK11 (LKB1) immunohistochemistry is a sensitive and specific marker for STK11 adnexal tumours

AimsSTK11 adnexal tumour is a rare, recently described malignant neoplasm that is associated with Peutz–Jeghers syndrome. [Correction added on 3 October 2024, after first online publication: ‘ST11’ in preceding sentence has been corrected to ‘STK11’ in this version.] It predominantly originates from the para‐adnexal soft tissues and often shows secondary involvement of the fallopian tube and ovary. STK11 adnexal tumours have a broad differential diagnosis due to their variable morphology and non‐specific immunoprofile, and diagnostic confirmation currently requires sequencing to identify an STK11 mutation. We investigate the diagnostic utility of STK11 (LKB1) immunohistochemistry (IHC) in a cohort of STK11 adnexal tumours and morphological mimics.Methods and resultsIHC for STK11 was performed on 122 tumours, including 17 STK11 adnexal tumours and 105 morphological mimics (10 female adnexal tumours of Wolffian origin, 22 adult granulosa cell tumours, 10 juvenile granulosa cell tumours, four Sertoli–Leydig cell tumours, two Leydig cell tumours, one Sertoli cell tumour, one steroid cell tumour, four extra‐ovarian sex cord‐stromal tumours, 16 ovarian endometrioid carcinomas, eight tubo‐ovarian high‐grade serous carcinomas, five ovarian mesonephric‐like adenocarcinomas, 14 ovarian carcinosarcomas, five peritoneal malignant mesotheliomas, two pelvic plexiform leiomyomata and one ovarian solid pseudopapillary tumour). All STK11 adnexal tumours showed complete loss of cytoplasmic staining for STK11. All other tumour types showed retained cytoplasmic staining, except for one endometrioid carcinoma with mucinous differentiation which showed complete loss of STK11 expression and a high‐grade serous carcinoma with subclonal loss.ConclusionsSTK11 is a highly sensitive and specific immunohistochemical marker for distinguishing STK11 adnexal tumour from its histological mimics, and can obviate the need for confirmatory molecular studies in the appropriate morphological context.

Diandric triploid partial mole versus digynic nonmolar triploidy: is morphological assessment sufficient for the diagnostic distinction?

AimsDiagnostic separation of diandric triploid gestation, i.e. partial mole from digynic triploid gestation, is clinically relevant, as the former may progress to postmolar gestational trophoblastic neoplasia. The aim of the study was to investigate if the combination of abnormal histology combined with ploidy analysis‐based triploidy is sufficient to accurately diagnose partial mole.Methods and ResultsA genotype–phenotype correlation study was undertaken to reappraise histological parameters among 20 diandric triploid gestations and 22 digynic triploid gestations of comparable patient age, gestational weeks, and clinical presentations. Two villous populations, irregular villous contours, pseudoinclusions, and syncytiotrophoblast knuckles, were common in both groups. Villous size ≥2.5 mm, cistern formation, trophoblastic hyperplasia, and syncytiotrophoblast lacunae were significantly more common in the partial hydatidiform mole. Cistern formation had the highest positive predictive value (PPV) (93%) and highest specificity (96%) for diandric triploid gestation, although the sensitivity was 70%. Cistern formation combined with villous size ≥2.5 mm or trophoblast hyperplasia or syncytiotrophoblast lacunae had 100% specificity and PPV, but a marginal sensitivity of 60%–65%. A moderate interobserver agreement (Kappa = 0.57, Gwet's AC1 = 0.59) was achieved among four observers who assigned diagnosis of diandric triploid gestation or digynic triploidy solely based on histology.ConclusionsNone of histological parameters are unique to either diandric triploid gestation or digynic triploid gestation. Cistern formation is the most powerful discriminator, with 93% PPV and 70% sensitivity for diandric triploid gestation. While cistern formation combined with either trophoblastic hyperplasia or villous size ≥2.5 mm or syncytiotrophoblast lacunae has 100% PPV and specificity for diandric triploid gestation, the sensitivity is only 60% to 65%. Therefore, in the presence of triploidy, histological assessment is unable to precisely classify 35% to 40% of diandric triploid gestations or partial moles.

Gastric‐type glandular lesions of the female genital tract excluding the cervix: emerging pathological entities

In the last two decades or so, a spectrum of benign, premalignant and malignant cervical glandular lesions exhibiting gastric differentiation has been described, with gastric‐type adenocarcinoma representing the most common human papillomavirus (HPV)‐independent cervical adenocarcinoma. More recently, limited literature has reported a variety of gastric‐type glandular lesions at other sites within the female genital tract and, as in the cervix (the most common site for these lesions), a spectrum of benign, premalignant and malignant lesions has been proposed. We provide an update and review of the emerging spectrum of gastric‐type glandular lesions at female genital tract sites other than the cervix. In the endometrium, putative gastric‐type glandular lesions include mucinous metaplasia of gastric‐type, atypical mucinous proliferation of gastric‐type and gastric‐type adenocarcinoma. Similarly in the vagina, gastric‐type adenosis, atypical adenosis and adenocarcinoma have been described. There have also been occasional reports of gastric‐type lesions involving the ovary and fallopian tube. We provide guidance on how to recognise gastric‐type lesions morphologically and immunophenotypically and stress that sometimes these lesions occur at more than one site within the female genital tract (synchronous/multifocal gastric‐type lesions of the female genital tract), sometimes in association with Peutz–Jeghers syndrome.

BRCA1 and BRCA2 carriers with breast, ovarian and prostate cancer demonstrate a different pattern of metastatic disease compared with non‐carriers: results from a rapid autopsy programme

AimTo catalogue and compare the pattern of metastatic disease in germline BRCA1/2 pathogenic mutation carriers and non‐carriers with breast, ovarian and prostate cancer from a rapid autopsy programme.Methods and resultsThe number of metastases in the major body systems and the proportion of participants with metastases were documented in 50 participants (19 germline mutation carriers). Analysis was conducted on the participants’ pattern of disease for the different cancers and mutation subgroups. The four commonly affected organ systems were the digestive (liver only) (82%), respiratory (76%), gastrointestinal (65%) and reticuloendothelial (42%). There were significant differences in the pattern of metastatic breast cancer in BRCA1/2 germline carriers compared with non‐carriers. Breast cancer carriers had significantly fewer organ systems involved (median n = 3, range = 1–3) compared with non‐carriers (median n = 9, range = 1–7) (P = 0.03). BRCA1/2 carriers with ovarian carcinomas had significantly more organ systems with metastatic carcinoma (median n = 10, range = 3–8) than non‐carriers (median n = 5, range = 3–5) (P < 0.001). There were no significant differences in the number of involved systems in BRCA2 carriers compared with non‐carriers with prostate cancer (P = 1.0). There was an absence of locoregional disease (6.5%) compared with distant disease (93.5%) among the three cancer subtypes (P < 0.001). The majority of metastatic deposits (97%) collected during the autopsy were identified by recent diagnostic imaging.ConclusionEven though a major limitation of this study is that our numbers are small, especially in the breast cancer carrier group, the metastatic patterns of breast and ovarian cancers may be impacted by BRCA1/2 carrier status, suggesting that tumours derived from patients with these mutations use different mechanisms of dissemination. The findings may focus clinical diagnostic imaging for monitoring metastases where whole‐body imaging resources are scant.

Use of p53 immunohistochemistry can improve diagnostic agreement for differentiated vulvar intraepithelial neoplasia ( dVIN ): an international reproducibility study

Aims Differentiated or HPV‐independent vulvar intraepithelial neoplasia (dVIN) can progress rapidly to invasive cancer and accurate pathological diagnosis is essential to facilitate appropriate interventions. Histological similarities of dVIN with non‐neoplastic lesions, however, often make the diagnosis less reproducible. We investigated among a diverse group of pathologists whether the diagnostic agreement improves with the use of p53 immunohistochemistry (IHC) interpreted using the pattern‐based schema. Methods and results Fifty haematoxylin–eosin (HE) stained archival slides (30 dVIN and 20 non‐dysplastic vulvar lesions) were selected and p53‐IHC was performed. Twenty‐four board‐certified pathologists from eight countries first assessed the HE slides alone, and after a washout period, re‐evaluated them alongside the p53‐IHC slides. During both rounds, slides were diagnosed as dVIN, favour dVIN, favour no‐VIN or no‐VIN. p53‐IHC was scored as wild‐type or mutant (diffuse, basal, cytoplasmic or null). Kappa ( κ ) statistics and McNemar's test were used for statistical analyses. Overall diagnostic agreement for dVIN saw a significant increase in the Kappa value ( κ  = 0.6 vs. κ  = 0.4, P  = 0.002) when HE and p53‐IHC slides were assessed together compared with histology assessment alone, although the level of agreement remained moderate. For p53‐IHC assessment, overall agreement was substantial ( κ  = 0.7). Diagnoses changing from no‐VIN/favour no‐VIN to dVIN correlated significantly with the identification of a p53‐mutant pattern ( P  < 0.001). Conclusions Our findings indicate that p53‐IHC is a robust ancillary tool that can be reproducibly interpreted by pathologists with varying experience levels and supports the routine use of p53‐IHC in cases where dVIN is considered in the differential diagnosis.

Infiltrative pattern of invasion is independently associated with shorter survival and desmoplastic stroma markers FAP and THBS2 in mucinous ovarian carcinoma

AimsMucinous ovarian carcinoma (MOC) is a rare ovarian cancer histotype with generally good prognosis when diagnosed at an early stage. However, MOC with the infiltrative pattern of invasion has a worse prognosis, although to date studies have not been large enough to control for covariables. Data on reproducibility of classifying the invasion pattern are limited, as are molecular correlates for infiltrative invasion. We hypothesized that the invasion pattern would be associated with an aberrant tumour microenvironment.Methods and resultsFour subspecialty pathologists assessed interobserver reproducibility of the pattern of invasion in 134 MOC. Immunohistochemistry on fibroblast activation protein (FAP) and THBS2 was performed on 98 cases. Association with survival was tested using Cox regression. The average interobserver agreement for the infiltrative pattern was moderate (kappa 0.60, agreement 86.3%). After reproducibility review, 24/134 MOC (18%) were determined to have the infiltrative pattern and this was associated with a higher risk of death, independent of FIGO stage, grade, and patient age in a time‐dependent manner (hazard ratio [HR] = 10.2, 95% confidence interval [CI] 3.0–34.5). High stromal expression of FAP and THBS2 was more common in infiltrative MOC (FAP: 60%, THBS2: 58%, both P < 0.001) and associated with survival (multivariate HR for FAP: 1.5 [95% CI 1.1–2.1] and THBS2: 1.91 [95% CI 1.1–3.2]).ConclusionsThe pattern of invasion should be included in reporting for MOC due to the strong prognostic implications. We highlight the histological features that should be considered to improve reproducibility. FAP and THBS2 are associated with infiltrative invasion in MOC.

Role of gene sequencing in classifying struma ovarii: BRAF p.G469A mutation and TERT promoter alterations favour malignant struma ovarii

AimsStruma ovarii (SO) are rare, accounting for 0.3–1% of ovarian tumours, and include benign and malignant lesions. In most cases, histology is not predictive of clinical outcome and prognosis. The prognosis of histologically malignant thyroid‐type carcinomas can indeed be excellent, while SO, composed of normal thyroid tissue, can recur and are designated highly differentiated follicular carcinoma of the ovary. Clearer diagnostic criteria are therefore required.Methods and resultsWe retrospectively studied 31 SO using DNA and RNA sequencing with pan‐cancer gene panels, including eight biologically malignant SO (BMSO) defined based on ovarian serosal or extra‐ovarian dissemination at presentation or during follow‐up, 10 stage IA histologically malignant SO (HMSO) with thyroid‐type carcinoma morphology and 13 biologically and histologically benign SO (BSO), with none of the above‐mentioned characteristics. Molecular alterations were observed in 87.5% of BMSO, 70% of HMSO and 7.7% of BSO (P < 0.001). All patients with a peritoneal dissemination at presentation or during follow‐up had at least one gene alteration. BRAF mutations (44.5%) were only observed in malignant forms (HMSO and BMSO) and TERT promoter alterations (25%) only in cases of BMSO. The BRAF p.G469A mutation, which is extremely rare in thyroid carcinomas, was the molecular alteration most frequently associated with malignant SO (28.5%).ConclusionOur results highlight the clinical utility of molecular sequencing in SO, based on this limited number of cases. However, as malignant SO evolve slowly, more extensive molecular studies in SO with more than 10 years’ follow‐up are required to draw any conclusions on the prognostic value of the associated gene alterations.

Immunohistochemical p16 overexpression and Rb loss correlate with high‐risk human papillomavirus infection in endocervical adenocarcinomas

Aimsp16 is a sensitive surrogate marker for transcriptionally active high‐risk human papillomavirus (HR‐HPV) infection in endocervical adenocarcinoma (ECA); however, its specificity is not perfect.Methods and resultsWe examined p16 and Rb expressions by immunohistochemistry (IHC) and the transcriptionally active HR‐HPV infection by mRNA in‐situ hybridisation (ISH) with histological review in 108 ECA cases. Thirteen adenocarcinomas of endometrial or equivocal origin (six endometrioid and seven serous carcinomas) were compared as the control group. HR‐HPV was detected in 83 of 108 ECA cases (77%), including five HPV‐associated adenocarcinomas in situ and 78 invasive HPV‐associated adenocarcinomas. All 83 HPV‐positive cases showed consistent morphology, p16 positivity and partial loss pattern of Rb. Among the 25 cases of HPV‐independent adenocarcinoma, four (16%) were positive for p16, and of these four cases, three of 14 (21%) were gastric type adenocarcinomas and one of 10 (10%) was a clear cell type adenocarcinoma. All 25 HPV‐independent adenocarcinomas showed preserved expression of Rb irrespective of the p16 status. Similarly, all 13 cases of the control group were negative for HR‐HPV with preserved expression of Rb, even though six of 13 (46%) cases were positive for p16. Compared with p16 alone, the combination of p16 overexpression and Rb partial loss pattern showed equally excellent sensitivity (each 100%) and improved specificity (100 versus 73.6%) and positive predictive values (100 versus 89.2%) in the ECA and control groups. Furthermore, HR‐HPV infection correlated with better prognosis among invasive ECAs.ConclusionsThe results suggest that the combined use of p16 and Rb IHC could be a reliable method to predict HR‐HPV infection in primary ECAs and mimics. This finding may contribute to prognostic prediction and therapeutic strategy.

FoxA2 is a reliable marker for the diagnosis of yolk sac tumour postpubertal‐type

AimsYolk sac tumour postpubertal‐type (YSTpt) shows a wide range of histological patterns and is challenging to diagnose. Recently, forkhead box transcription factor A2 (FoxA2) emerged as a driver of YSTpt formation and a promising marker for diagnosing YSTpt. However, FoxA2 has not been tested in the different patterns of YSTpt. This study aimed to assess the staining pattern of FoxA2 in te different patterns of YSTpt and other germ cell tumours of the testis (GCTT), comparing it with glypican‐3 (GPC3) and α‐fetoprotein (AFP).Methods and resultsFOXA2, GPC3 and AFP immunohistochemistry was performed on 24 YSTpt (24 microcystic/reticular, 10 myxoid, two macrocystic, five glandular/alveolar, two endodermal sinus/perivascular, four solid, two polyembryoma/embryoid body and two polyvesicular vitelline) and 81 other GCTT. The percentage of positive cells (0, 1+, 2+, 3+) and the intensity (0, 1, 2, 3) were evaluated regardless of and within each YSTpt pattern. FoxA2 was positive in all YSTpt (24 of 24) and all but one (23 of 24) exhibited 2+/3+ stain, with higher intensity [median value (mv): 2.6] than AFP (1.8) and GPC3 (2.5). Both FoxA2 and GPC3 were positive in all microcystic/reticular (24 of 24), myxoid (10 of 10), macrocystic (two of two), endodermal sinus/perivascular (four of four) and polyembryoma/embryoid body (two of two) patterns. Nevertheless, only FoxA2 was positive in all glandular/alveolar (five of five), solid (four of four) and polyvesicular vitelline (two of two) patterns. The intensity of FoxA2 was higher than AFP and GPC3 in almost all YST patterns. In the other GCTT, FoxA2 was positive only in teratoma postpubertal‐type (Tpt) [13 of 20 (65%)], with staining almost exclusively confined to the mature gastrointestinal/respiratory tract epithelium.ConclusionsFoxA2 is a highly sensitive and specific biomarker that supports the diagnosis of YSTpt. FoxA2 is superior to GPC3 and AFP, especially in rare and difficult‐to‐diagnose histological patterns of YSTpt, but mature glands of Tpt could represent a potential diagnostic pitfall.

Retained PAX2 expression associated with DNA mismatch repair deficiency in endometrial endometrioid adenocarcinoma

AimsLoss of expression of tumour suppressor PAX2 and MMR deficiency (dMMR) has been frequently seen in endometrial endometrioid adenocarcinoma (EEC). However, the relationship between PAX2 expression and MMR status is unknown.Methods and ResultsWe studied the PAX2 expression and examined its association with MMR status at the protein and genetic levels in 180 cases of EEC. Overall, total loss of PAX2 expression was found in about 70%, while retained PAX2 expression was seen in 30% of EEC. Among 125 cases with loss of PAX2, 68.8% were found in EECs with pMMR, while 31.2% were seen in those with dMMR. Among 55 cases of EECs with retained PAX2 expression, 92.7% were EECs with dMMR and 7.3% were those with pMMR (P < 0.001). While dMMR cases with MLH1 hypermethylation show almost equal retained or loss of PAX2 expression (52% versus 48%), dMMR with genetic alterations had significantly more retained PAX2 expression than loss of PAX2 (92.3% versus 7.7%), regardless of somatic or germline mutations. Loss of PAX2 was observed in 97.3% of dMMR with MLH1 hypermethylation compared to 2.7% of dMMR with genetic alterations (P < 0.001). Aggressive features such as higher tumour grades (FIGO 2–3) and advanced clinical stage (T2–T4) were significantly more frequently seen in dMMR with retained PAX2 expression, compared those to pMMR with loss of PAX2 expression.ConclusionOur study demonstrates a close correlation between retained PAX2 expression and dMMR in EEC. The molecular mechanism and clinical significance linking these two pathways in EEC remains to be unravelled.

Mismatch repair deficiency: how reliable is the two‐antibody approach? A national real‐life study

AimsTraditionally, mismatch repair (MMR) status is determined by a panel of four antibodies (MLH1, PMS2, MSH2, MSH6). If all proteins are retained, cases are MMR proficient (pMMR), while loss of one or more proteins is indicative of MMR deficiency (dMMR). This approach has been challenged in favour of a two‐antibody approach, using PMS2 and MSH6 as a first screening. Their retainment is deemed sufficient to declare cases pMMR. In this study we aim to verify the validity of the two‐antibody approach.Methods and ResultsWe performed a nationwide study in colorectal cancer (CRC) and endometrial cancer (EC) diagnosed between 2016 and 2023, including 47,657 patients to evaluate the two‐antibody approach. In 0.17% and 0.4% of cases of CRC and EC, respectively, dMMR cases would be missed with the two‐antibody approach. Subgroup analyses pointed towards slightly increased miss rates in younger patients (under the age of 50 years) in both groups and identified special subtypes (signet ring cell carcinoma, medullary carcinoma, and mucinous carcinoma in CRC and clear cell carcinoma in EC) with increased miss rates. For these specific subgroups, a low threshold should be used for further testing. In case of ambiguous or heterogeneous staining patterns, four antibodies should be used.ConclusionIn general, the application of a two‐antibody MMR testing strategy does not lead to considerable failure of dMMR identification and saves costs.

Diagnosis of verruciform acanthotic vulvar intra‐epithelial neoplasia (vaVIN) using CK17, SOX2 and GATA3 immunohistochemistry

AimsVerruciform acanthotic vulvar intra‐epithelial neoplasia (vaVIN) is an HPV‐independent, p53 wild‐type lesion with distinct morphology and documented risk of recurrence and cancer progression. vaVIN is rare, and prospective distinction from non‐neoplastic hyperplastic lesions can be difficult. CK17, SOX2 and GATA3 immunohistochemistry has emerging value in the diagnosis of HPV‐independent lesions, particularly differentiated VIN. We aimed to test the combined value of these markers in the diagnosis of vaVIN versus its non‐neoplastic differentials in the vulva.Methods and resultsCK17, SOX2 and GATA3 immunohistochemistry was evaluated on 16 vaVINs and 34 mimickers (verruciform xanthoma, lichen simplex chronicus, lichen sclerosus, psoriasis, pseudo‐epitheliomatous hyperplasia). CK17 was scored as 3+ = full‐thickness, 2+ = partial‐thickness, 1+ = patchy, 0 = absent; SOX2 as 3+ = strong staining ≥ 10% cells, 2+ = moderate, 1 + =weak, 0 = staining in < 10% cells; and GATA3 as pattern 0 = loss in < 25% basal cells, 1 = loss in 25–75% basal cells, 2 = loss in > 75% basal cells. For analysis, results were recorded as positive (CK17 = 3+, SOX2 = 3+, GATA3 = patterns 1/2) or negative (CK17 = 2+/1+/0, SOX2 = 2+/1+/0, GATA3 = pattern 0). CK17, SOX2 and GATA3 positivity was documented in 81, 75 and 58% vaVINs, respectively, versus 32, 17 and 22% of non‐neoplastic mimickers, respectively; ≥ 2 marker positivity conferred 83 sensitivity, 88 specificity and 86% accuracy in vaVIN diagnosis. Compared to vaVIN, SOX2 and GATA3 were differentially expressed in lichen sclerosus, lichen simplex chronicus and pseudo‐epitheliomatous hyperplasia, whereas CK17 was differentially expressed in verruciform xanthoma and adjacent normal mucosa.ConclusionsCK17, SOX2 and GATA3 can be useful in the diagnosis of vaVIN and its distinction from hyperplastic non‐neoplastic vulvar lesions. Although CK17 has higher sensitivity, SOX2 and GATA3 are more specific, and the combination of all markers shows optimal diagnostic accuracy.

Lesions sub‐diagnostic of endometrioid intra‐epithelial neoplasia/atypical hyperplasia: value of morphology and immunohistochemistry in predicting neoplastic outcome

AimsAreas of gland crowding that do not fulfil diagnostic criteria of endometrioid intra‐epithelial neoplasia (EIN) are often encountered in endometrial biopsies. In this study, we document the prevalence of neoplastic outcome in patients with these subdiagnostic lesions (SL) and assess the utility of morphological features and a three‐marker immunohistochemistry panel (PAX2, PTEN, beta‐catenin) to predict outcome.Methods and resultsOf 430 women with SL on endometrial sampling at Brigham and Women's Hospital between 2001 and 2021 with available follow‐up biopsy, 72 (17%) had a neoplastic outcome (EIN or endometrioid carcinoma). Multilayered epithelium and mitoses in SL were statistically associated with a neoplastic outcome. Abnormal three‐marker staining was observed in 93% (53 of 57) of SL with neoplastic outcome and 60% (37 of 62) of a control group with benign outcome. Among the 72 patients with neoplastic outcome, EIN/carcinoma tissue was available in 33; of these, 30 (91%) showed abnormal staining for one or more markers. Remarkably, in 84% of these cases the EIN/carcinoma had the aberrant expression seen in the preceding SL. Based on a prevalence of 17%, the positive and negative predictive values of abnormal staining in one or more markers were 24 and 97%, respectively.ConclusionsThe presence of SL warrants clinical surveillance and repeat sampling because it is followed by endometrioid neoplasia in a significant subset of patients. Normal three‐marker staining identifies women with a very low risk of neoplastic outcome. Conversely, abnormal staining is frequent in SL with benign outcome leading to poor specificity and positive predictive value.

Teratoma‐associated and so‐called pure Wilms tumour of the ovary represent two separate tumour types with distinct molecular features

AimsOvarian Wilms tumour (WT)/nephroblastoma is an extremely rare neoplasm that has been reported to occur in pure form or as a component of a teratomatous neoplasm. We hypothesized that teratoma‐associated and pure ovarian WT may represent different tumour types with diverging molecular backgrounds. To test this hypothesis, we comprehensively characterized a series of five tumours originally diagnosed as ovarian WT.Methods and ResultsThe five cases comprised three teratoma‐associated (two mature and one immature) and two pure WTs. Two of the teratoma‐associated WTs consisted of small nodular arrangements of “glandular”/epithelial structures, while the third consisted of both an epithelial and a diffuse spindle cell/blastemal component. The pure WTs consisted of “glandular” structures, which were positive for sex cord markers (including inhibin and SF1) together with a rhabdomyosarcomatous component. The two pure WTs harboured DICER1 pathogenic variants (PVs), while the three associated with teratomas were DICER1 wildtype. Panel‐based DNA sequencing of four of the cases did not identify PVs in the other genes investigated. Analysis of the HA19/IGF2 imprinting region showed retention of imprinting in the pure WTs but loss of heterozygosity with hypomethylation of the ICR1 region in two of three teratoma‐associated WTs. Furthermore, copy number variation and clustering‐based whole‐genome DNA methylation analyses identified divergent molecular profiles for pure and teratoma‐associated WTs.ConclusionBased on the morphological features, immunophenotype, and molecular findings (DICER1 PVs, copy number, and DNA methylation profiles), we suggest that the two cases diagnosed as pure primary ovarian WT represent moderately to poorly differentiated Sertoli Leydig cell tumours (SLCTs), while the tumours arising in teratomas represent true WTs. It is possible that at least some prior cases reported as pure primary ovarian WT represent SLCTs.

Cervical adenocarcinoma: integration of HPV status, pattern of invasion, morphology and molecular markers into classification

Cervical adenocarcinoma is a heterogenous group of tumours with various aetiologies, molecular drivers, morphologies, response to treatment and prognosis. It has become evident that human papillomavirus (HPV) infection does not drive all adenocarcinomas, and appropriate classification is critical for patient management, especially in the era of the HPV vaccine and HPV‐only screening. Identified as one of the most important developments in gynaecological pathology during the past 50 years, the separation of cervical adenocarcinomas into HPV‐associated (HPVA) and HPV‐independent has resulted in a transformation of the classification system for cervical adenocarcinomas. HPVA has been traditionally subclassified by morphology, such as usual type (UEA), mucinous and villoglandular, etc. However, it has become evident that cell type‐based histomorphological classification is not clinically meaningful, and the newly proposed International Endocervical Adenocarcinoma Criteria and Classification (IECC) is a necessary and relevant break from this prior system. Non‐HPV‐associated adenocarcinomas can be divided by their distinct morphology and molecular genomics with very different responses to standard therapies and potential for future targeted therapies. These include gastric‐type, clear‐cell, mesonephric and endometrioid adenocarcinomas. So‐called ‘serous’ carcinomas of the cervix probably represent morphological variants of UEA or drop metastases from uterine or adnexal serous carcinomas, and the existence of true cervical serous carcinomas is in question. This review will discuss the advances since WHO 2014, and how HPV status, pattern of invasion as described by Silva and colleagues, histological features and molecular markers can be used to refine diagnosis and prognostication for patients with cervical adenocarcinoma.

PD‐L1 expression and immune stromal features in HPV‐independent cervical adenocarcinoma

AimsHuman papilloma virus (HPV)‐independent cervical adenocarcinoma (CA) is usually diagnosed at an advanced stage, while the therapeutic options are limited. Therefore, effective treatment options are required. The programmed cell death 1 (PD‐1) inhibitor pembrolizumab has been approved for the treatment of patients with recurrent or metastatic cervical squamous cell carcinoma expressing PD‐ligand 1 (PD‐L1). However, no data regarding PD‐L1 expression in HPV‐independent CA are available. Thus, we evaluated the association between PD‐L1 expression and the clinicopathological characteristics and survival of patients with HPV‐independent CA.MethodsWe evaluated PD‐L1, mismatch repair (MMR) protein expression and the immune stromal features of 44 patients with HPV‐independent CA. PD‐L1 expression was defined as a combined positive score (CPS) ≥1 and a tumour proportion score (TPS) ≥1%.ResultsPD‐L1 expression was observed in 14 cases (31.8%) with CPS ≥1 and 12 cases (27.3%) with TPS ≥1%. PD‐L1 expression, based on either the CPS or the TPS, was associated with a high tumour‐infiltrating lymphocyte percentage (CPS = P < 0.001; TPS = P < 0.001). Patients with a PD‐L1 CPS ≥1 showed worse progression‐free survival and overall survival than PD‐L1‐negative patients (P = 0.004 and P = 0.023, respectively). Forty‐two cases demonstrated intact MMR expression and two cases demonstrated loss of MSH2/MSH6.ConclusionsOur data demonstrated that PD‐L1 was expressed in HPV‐independent CA, especially in clear cell carcinoma, and that PD‐L1 expression is a negative prognostic marker. Our data support the role of PD‐L1 in HPV‐independent CA and its potential as an immunotherapeutic target.

Histological clues to the diagnosis of metastasis to the breast from extramammary malignancies

AbstractAimsThe aims of this study were to review the histological features useful for the identification of metastases to the breast and to investigate the impression that this diagnosis has become more common.Methods and resultsThe histological features of metastases to the breast from 2008 to 2018 were reviewed. Seventy‐four biopsies from 66 patients were identified: 1% compared with primary carcinoma of the breast. Non‐haematological metastases comprised 0.75% compared with 0.3% in a series from 1996 to 2005. The most common tumour types were pulmonary carcinoma (22), lymphoma (15), melanoma (13), gastrointestinal carcinoma (eight) and serous papillary carcinoma (four). In 73% there were histological features that were not typical of primary mammary carcinoma. Some metastases were histologically similar to breast cancer and the history was essential to making the correct diagnosis. Useful histological clues included small‐cell morphology for pulmonary carcinoma, glands containing necrosis for gastrointestinal carcinoma, intranuclear inclusions, marked pleomorphism and spindle cells for melanoma, clear cells for renal carcinoma, papillary architecture for serous papillary carcinoma and sheets of centroblasts or nodules of centroblasts and centrocytes for lymphoma. Useful immunohistochemical markers included TTF‐1 for pulmonary carcinoma, S100, melan‐A and HMB45 for melanoma, CK20 and CDX2 for colorectal carcinoma, PAX8 and WT1 for serous papillary carcinoma and lymphoid markers for lymphomas, in addition to the absence of expression of mammary markers ER, GATA3 and GCDFP‐15.ConclusionThe majority of metastases to the breast have histological clues to the diagnosis. Immunohistochemistry is helpful. This diagnosis is being made more frequently.

SWI/SNF protein and claudin‐4 expression in anaplastic carcinomas arising in mucinous tumours of the ovary and retroperitoneum

AimsAnaplastic carcinoma arising in a mucinous tumour of the ovary and rarely in the retroperitoneum is an uncommon neoplasm with three morphological patterns; rhabdoid, sarcomatoid and pleomorphic. We investigated expression of switch/sucrose non‐fermentable (SWI/SNF) chromatin remodelling complex components and claudin‐4 expression.Methods and resultsTwenty‐two ovarian and three retroperitoneal mucinous tumours were investigated using antibodies against SMARCB1, SMARCA4, SMARCA2, ARID1A and claudin‐4. Loss of nuclear staining for any SWI/SNF protein was observed in the anaplastic component of nine of 25 (36%), with retained expression within the mucinous component of all tumours. Five (56%) showed loss of more than one protein, with dual loss of SMARCA4 and SMARCA2 in two, loss of SMARCA2 and ARID1A in two and loss of SMARCB1 and SMARCA2 in one. Retained expression of claudin‐4 was seen in 39% of the anaplastic carcinomas and within the mucinous component of all tumours. Rhabdoid morphology was associated with poor prognosis [stages III or IV disease (six of six, 100% versus four of 14, 29%; P = 0.0108] and death from disease (three of four, 75% versus one of 13, 8%; P = 0.0223). Although loss of a SWI/SNF protein was not significantly associated with death from disease (three of five, 60% versus one of 12, 8%; P = 0.0525), it showed a trend in correlation with poor prognosis and was often noted in tumours with rhabdoid morphology within this small cohort.ConclusionsOur report adds to the growing list of female genital tract malignancies with loss of SWI/SNF proteins, underlining their broad differential diagnosis and the importance of careful, context‐dependent interpretation of SWI/SNF protein loss.

Sclerosing stromal tumour: a clinicopathological study of 100 cases of a distinctive benign ovarian stromal tumour typically occurring in the young

AimsSince the sclerosing stromal tumour (SST) of the ovary was first described in 1973, few studies have expanded upon its histological features or overlap with other tumours. We thus investigate these aspects based on our experience with 100 cases.Methods and resultsThe patients, 14 of whom were pregnant, ranged from 12 to 63 years (median = 26 years). Ten patients had hormonal manifestations (seven oestrogenic, three androgenic). Bilateral ovarian involvement was present in two cases. Size ranged from 1 to 23 cm (mean = 8.4 cm). Most tumours were solid and white with focal yellow areas; oedema with cystic degeneration (seen in 25 cases) resulted in eight being predominantly cystic. On microscopic examination, alternating cellular and paucicellular areas (pseudolobulation) were prominent in 94 cases but seen to a limited degree in the remaining neoplasms. Admixed spindled and luteinised cells were present in all tumours, but 13 demonstrated mainly spindled cells and 19 demonstrated mainly lutein cells; 14 of the latter were from pregnant patients. The stroma was typically oedematous or collagenous, but in 14 cases was prominently hyalinised and, in four, myxoid. Prominent vascularity was present in most cases. The mitotic rate ranged from 0 to 8/10 high‐power fields (HPF), but most demonstrated <1/10 HPF.ConclusionsThe differential diagnosis of SST is broad, including fibromas, thecomas, solitary fibrous tumours, pregnancy luteomas, myxomas, other ovarian sex cord‐stromal tumours with sclerosis and, rarely, Krukenberg tumours. Strict adherence to the requirement of pseudolobulation, prominent (usually ectatic) vessels, and lutein cells and fibroblasts admixed in a jumbled manner, will distinguish the neoplasm from others in the differential.

Histological grading of ovarian mucinous carcinoma – an outcome‐based analysis of traditional and novel systems

AimsGrading of primary ovarian mucinous carcinoma (OMC) is inconsistent among practices. The International Collaboration on Cancer Reporting recommends grading OMC using the International Federation of Gynecology and Obstetrics (FIGO) system for endometrial endometrioid carcinoma, when needed. The growth pattern (expansile versus infiltrative), a known prognostic variable in OMC, is not considered in any grading system. We herein analysed the prognostic value of various grading methods in a well‐annotated cohort of OMC.Methods and resultsInstitutional OMCs underwent review and grading by the Silverberg and FIGO schemes and a novel system, growth‐based grading (GBG), defined as G1 (expansile growth or infiltrative invasion in ≤10%) and G2 (infiltrative growth >10% of tumour). Of 46 OMCs included, 80% were FIGO stage I, 11% stage II and 9% stage III. On follow‐up (mean = 52 months, range = 1–190), five patients (11%) had adverse events (three recurrences and four deaths). On univariate analysis, stage (P = 0.01, Cox proportional analysis), Silverberg grade (P = 0.01), GBG grade (P = 0.001) and percentage of infiltrative growth (P < 0.001), but not FIGO grade, correlated with disease‐free survival. Log‐rank analysis showed increased survival in patients with Silverberg grade 1 versus 2 (P < 0.001) and those with GBG G1 versus G2 (P < 0.001). None of the parameters evaluated was significant on multivariate analysis (restricted due to the low number of adverse events).ConclusionsSilverberg and the new GBG system appear to be prognostically significant in OMC. Pattern‐based grading allows for a binary stratification into low‐ and high‐grade categories, which may be more appropriate for patient risk stratification. Despite current practices and recommendations to utilise FIGO grading in OMC, our study shows no prognostic significance of this system and we advise against its use.

Six‐pattern p53 interpretation in 1293 vulvar squamous cell carcinomas: inter‐pathologist variation and pattern distribution according to p16 status

Aims Vulvar squamous cell carcinoma (VSCC) is classified into human papillomavirus (HPV)‐associated and HPV‐independent types, primarily using p16 immunohistochemistry, with p53 staining playing a complementary role since a subset of HPV‐independent VSCC is driven by TP53 mutations. We aimed to assess the robustness of the six‐pattern p53 classification by evaluating interobserver agreement and mapping pattern distribution in relation to p16 status. Methods We performed p53 immunohistochemistry on 1293 VSCC cases, comprising 832 p16‐negative and 461 p16‐positive cases. Eight pathologists independently evaluated p53, with each case assessed by two pathologists. Expression was classified as wild‐type (scattered or mid‐epithelial) or mutated (basal overexpression, parabasal/diffuse overexpression, absent or cytoplasmic). Interobserver agreement was measured using kappa statistics. Results Overall concordance across the six p53 patterns was 66.7%, increasing to 86.9% when dichotomized as wild‐type versus mutated. In the p16‐negative cases, concordance was 68.8% across all six patterns and 82.6% when dichotomized. Corresponding rates in the p16‐positive cases were 62.9% and 94.6%. Kappa values for pairwise assessments ranged from 0.44 to 0.73 (six‐pattern) and from 0.60 to 0.88 (dichotomized). After resolving discordant cases, 79.9% of p16‐negative cases showed a mutated pattern, and 20.1% were wild type (scattered). Among the p16‐positive cases, 93.1% exhibited a wild‐type pattern. Conclusions Findings support the clinical robustness of the six‐pattern p53 framework, as interobserver agreement was high and most discrepancies were unlikely to impact tumour classification. While p16 proved helpful in p53 interpretation, certain cases remained challenging due to p53 heterogeneity or ambiguous p16/p53 combinations indicating a need for additional molecular testing in such instances.

High‐grade vulvar intraepithelial neoplasia: comprehensive characterization and long‐term vulvar carcinoma risk

AimsAdequate diagnosis of human papillomavirus (HPV)‐associated high‐grade squamous intraepithelial lesion (HSIL) and HPV‐independent vulvar intraepithelial neoplasia (VIN) is essential but can be challenging. We comprehensively characterized a large population‐based series of vulvar lesions, originally reported as high‐grade VIN, and assessed the cancer risk.Methods and resultsBaseline high‐grade VIN of 751 patients were categorized by histopathological reassessment, integrating the results of immunohistochemistry (p16INK4a, p53, Ki‐67) and HPV DNA testing. Integrated analyses resulted in 88.4% HPV‐associated lesions (77.0% HSIL, 10.9% low‐grade SIL [LSIL], and 0.4% vulvar squamous cell carcinoma [VSCC]), 10.9% HPV‐independent lesions (6.1% HPV‐independent VIN, 4.7% nondysplastic lesions, and 0.1% VSCC) and 1.1% inconclusive lesions. HSIL demonstrated p16INK4a block‐positivity in 99.0%, increased Ki‐67 in ≥2/3rd of the epithelium in 93.6%, and HPV positivity in 99.6%. In HSIL, a p53 wildtype mid‐epithelial staining pattern was common (51.6%) while this was not observed in HPV‐independent lesions. HPV‐independent VIN harboured mutant p53 patterns in 65.2% and showed a wide morphological spectrum, ranging from differentiated to nondifferentiated (‘HPV‐associated‐like’, in 41.3%). Kaplan–Meier analyses showed a 10‐year cancer risk of 8.0% in HPV‐associated HSIL, 67.4% in HPV‐independent VIN/p53mutant, and 27.8% in HPV‐independent VIN/p53wildtype. Strikingly, the 10‐year cancer risk was 73.3% in HPV‐independent VIN with nondifferentiated (‘HPV‐associated‐like’) morphology.ConclusionImmunohistochemistry by p16INK4a and p53 is highly recommended for optimal categorization into HPV‐associated and HPV‐independent VIN, which is of utmost importance given the different cancer risk. The high cancer risk of HPV‐independent VIN underscores the need for surgical treatment and close follow‐up, especially in case of a p53 mutant pattern and/or nondifferentiated morphology.

Human papillomavirus and p53 status define three types of vulvar squamous cell carcinomas with distinct clinical, pathological, and prognostic features

IntroductionBased on their etiological relationship with human papillomavirus (HPV), the 2020 WHO classification has divided vulvar squamous cell carcinomas (VSCC) into two distinct types, HPV‐associated and HPV‐independent, and HPV‐independent tumours have recently been divided according to p53 status. Nevertheless, the clinical and prognostic significance of this classification has not been clearly established. We analysed the differential clinical, pathological, and behavioural characteristics of these three types of VSCC in a large series of patients.Methods and resultsVSCC samples from patients who underwent primary surgery at the Hospital Clinic of Barcelona, Spain, during a 47‐year period (January 1975 to January 2022) were analysed (n = 190). HPV detection, p16, and p53 immunohistochemical staining were evaluated. We also analysed recurrence‐free survival (RFS) and disease‐specific survival (DSS). Thirty‐three tumours (17.4%) were HPV‐associated and 157 (82.6%) HPV‐independent. Of these, 20 showed normal and 137 abnormal p53 expression. The two types of HPV‐independent tumours showed worse RFS in the multivariate analysis (hazard ratio [HR] = 3.63; P = 0.023 for the HPV‐independent p53 normal VSCC and HR = 2.78; P = 0.028 for the HPV‐independent p53 abnormal VSCC). Although the differences were not significant, HPV‐independent VSCC had worse DSS than HPV‐associated VSCC. Although patients with HPV‐independent p53 normal tumours had worse RFS than patients with HPV‐independent p53 abnormal tumours, the DSS was better for the former group. Only advanced FIGO stage was associated with worse DSS in multivariate analysis (HR = 2.83; P = 0.010).ConclusionThe association of HPV and p53 status have prognostic implications, reinforcing a three‐tier molecular classification of VSCC (HPV‐associated VSCC, HPV‐independent VSCC with normal p53, HPV‐independent VSCC with abnormal p53).

Vulvar squamous cell carcinoma arising on human papillomavirus‐independent precursors mimicking high‐grade squamous intra‐epithelial lesion: a distinct and highly recurrent subtype of vulvar cancer

AimsEach category of vulvar squamous cell carcinoma (VSCC), human papillomavirus (HPV)‐associated and HPV‐independent, arises on a specific intra‐epithelial precursor: high‐grade squamous intra‐epithelial lesions (HSIL) and differentiated vulvar intra‐epithelial neoplasia (dVIN), respectively. However, a subset of HPV‐independent VSCC arises on an intra‐epithelial precursor closely mimicking HSIL. We aimed to explore the clinicopathological features of the HPV‐independent tumours with HSIL‐like lesions and compare them with HPV‐independent VSCC with dVIN and HPV‐associated tumours with HSIL.Methods and resultsWe retrospectively identified 105 cases of surgically treated VSCC with adjacent intra‐epithelial precursors. The cases were classified into three groups based on the HPV status and the adjacent precursor identified: (i) HPV‐associated VSCC with HSIL (n = 26), (ii) HPV‐independent VSCC with dVIN lesions (n = 54) and (iii) HPV‐independent VSCC with HSIL‐like lesions (n = 25). We analysed the histological and clinical features including the recurrence‐free survival and disease‐specific survival in the three groups. Patients with HPV‐independent VSCC with HSIL‐like lesions and with dVIN were older than patients with HPV‐associated VSCC (76 and 77 versus 66 years, respectively, P < 0.001). HPV‐independent VSCC with HSIL‐like lesions recurred more frequently [hazard ratio (HR) = 3.87; P < 0.001] than HPV‐independent VSCC with dVIN (HR = 2.27; P = 0.1) and HPV‐associated VSCC (HR = 1). In the multivariate analysis, HPV‐independent VSCC with HSIL‐like lesions remained significant for recurrence. No differences in disease‐specific survival were observed between the three groups.ConclusionsEven though VSCC with HSIL‐like lesions are not associated with higher mortality, they are more likely to recur and might benefit from more intensive treatment strategies and closer surveillance after treatment.

Performance of the pattern‐based interpretation of p53 immunohistochemistry as a surrogate for TP53 mutations in vulvar squamous cell carcinoma

AimsThe most commonly mutated gene in vulvar squamous cell carcinoma (VSCC) is TP53 and its prognostic value, particularly in HPV‐independent VSCC, is uncertain. In other tumours, p53 immunohistochemistry (IHC) is an excellent surrogate marker for TP53 mutations. In order to study this in VSCC, we assigned six p53 IHC patterns into two final classes: ‘wild‐type’ or ‘mutant’. We determined the performance and interobserver variability of this pattern‐based p53 IHC approach.Methods and resultsTwo experienced gynaecological pathologists scored the predefined p53 IHC patterns of 59 VSCC, independently and blinded for molecular data. Agreement was calculated by Cohen's kappa. All disagreements regarding p53 IHC patterns were resolved by a consensus meeting. After DNA isolation, the presence of pathogenic TP53 variants was determined by next‐generation sequencing (NGS). Sensitivity, specificity and accuracy of p53 IHC as a surrogate marker for TP53 mutation status were calculated. Initial p53 IHC pattern interpretation showed substantial agreement between both observers (k = 0.71, P < 0.001). After consensus, 18 cases (30.5%) were assigned a final p53 IHC class as TP53 wild‐type and 41 cases (69.5%) as mutant. The accuracy between the p53 IHC class and TP53 mutation status, after the consensus meeting, was 96.6%. Moreover, the sensitivity and specificity were high 95.3% [95% confidence interval (CI) = 82.9–99.1% and 100% (95% CI = 75.9–100%)].ConclusionsPattern‐based p53 IHC classification is highly reproducible among experienced gynaecological pathologists and accurately reflects TP53 mutations in VSCC. This approach to p53 IHC interpretation offers guidance and provides necessary clarity for resolving the proposed prognostic relevance of final p53 IHC class within HPV‐independent VSCC.

GREB1 ‐rearranged uterine tumour shares a common DNA methylation signature with ESR1 ‐rearranged UTROSCT

Background and objectives GREB1 ‐rearranged uterine tumours encompass a group of uterine mesenchymal tumours with varied histologic appearances. The fusion partners to GREB1 include NCOA1‐3, SS18 and NR4A3 . Given that some GREB1 ‐rearranged uterine tumours exhibit histologic features of uterine tumours resembling ovarian sex cord tumour (UTROSCT), there is a general belief that GREB1 ‐rearranged uterine mesenchymal tumours are part of the UTROSCT family. Methods In this study, we applied global DNA methylation and copy number analyses to a series of 10 GREB1‐ rearranged uterine tumours and 21 classic UTROSCTs (7 of which were molecularly confirmed to harbour ESR1::NCOA2/3 fusions). Results We found that GREB1 ‐rearranged uterine tumors show an overlap in their global methylation profiles with UTROSCT, including ESR1::NCOA2/3 positive cases. Together, these tumours form a DNA methylation cluster separate from uterine smooth muscle tumours (leiomyomas and leiomyosarcomas), endometrial stromal sarcomas (low‐grade and high‐grade), embryonal rhabdomyosarcoma and SMARCA4‐deficient uterine sarcomas. However, despite their epigenetic similarity, there were two notable differences. First, GREB1 ‐rearranged uterine tumours as a group displayed a greater degree of genomic complexity with more extensive copy number alterations than conventional UTROSCTs, including those harbouring ESR1::NCOA2/3 . Second, GREB1 ‐rearranged uterine tumours frequently lacked overt sex cord morphology: while all 7 ESR1::NCOA2/3 UTROSCTs demonstrated corded, nested, trabecular and/or tubular/sertoliform patterns, only 1 GREB1 ‐rearranged uterine tumour displayed a prominent trabecular pattern, with the remaining cases showing exclusively or predominantly diffuse/solid growth. Conclusions Overall, our findings confirm that GREB1 ‐rearranged uterine tumours are part of the UTROSCT spectrum, though they frequently exhibit a more diffuse growth pattern and a higher degree of genomic instability.

JAZF1, YWHAE and BCOR gene translocation in primary extrauterine low‐grade and high‐grade endometrial stromal sarcomas

AimsJAZF1 translocation is the most common genetic change in low‐grade (LG) endometrial stromal sarcoma (ESS), and YWHAE and BCOR translocations are common in high‐grade (HG) ESS. Primary extrauterine ESS is rare, and there are limited data on molecular alterations in these tumours.Methods and resultsCases of primary extrauterine ESS, comprising eight LG‐ESS cases and five HG‐ESS cases were collected. Haematoxylin and eosin and immunohistochemical staining were used to observe the histomorphology and analyse related protein expression. JAZF1, YWHAE and BCOR rearrangements were explored with fluorescence in‐situ hybridisation (FISH). In LG‐ESS, the tumour cells resembled normal proliferative‐phase endometrial stromal cells; CD10, oestrogen receptor and progesterone receptor were expressed in all eight cases. In HG‐ESS, the tumour cells had uniform HG round and/or spindle morphology, sometimes with an LG component; CD10 was fully expressed in one case and focally expressed in four cases; BCOR was expressed in all five cases, and cyclin D1 in four of five cases. FISH analysis showed JAZF1 translocation in one of eight LG‐ESS cases (12.5%). YWHAE translocation occurred in four of five HG‐ESS cases, with a positivity rate of 80%. BCOR translocation was absent in all five cases.ConclusionsIn extrauterine LG‐ESS, the rate of JAZF1 rearrangement was significantly lower than in uterine LG‐ESS. This result limited the value of JAZF1 translocation for diagnosis. YWHAE rearrangement is a common genetic change in extrauterine HG‐ESS. Further studies are required to confirm these findings, especially in LG‐ESS.

Publisher

Wiley

ISSN

0309-0167