Prognostic impact and causality of age on oncological outcomes in women with endometrial cancer: a multimethod analysis of the randomised PORTEC-1, PORTEC-2, and PORTEC-3 trials.

BACKGROUND: Numerous studies have shown that older women with endometrial cancer have a higher risk of recurrence and cancer-related death. However, it remains unclear whether older age is a causal prognostic factor, or whether other risk factors become increasingly common with age. We aimed to address this question with a unique multimethod study design using state-of-the-art statistical and causal inference techniques on datasets of three large, randomised trials. METHODS: In this multimethod analysis, data from 1801 women participating in the randomised PORTEC-1, PORTEC-2, and PORTEC-3 trials were used for statistical analyses and causal inference. The cohort included 714 patients with intermediate-risk endometrial cancer, 427 patients with high-intermediate risk endometrial cancer, and 660 patients with high-risk endometrial cancer. Associations of age with clinicopathological and molecular features were analysed using non-parametric tests. Multivariable competing risk analyses were performed to determine the independent prognostic value of age. To analyse age as a causal prognostic variable, a deep learning causal inference model called AutoCI was used. FINDINGS: Median follow-up as estimated using the reversed Kaplan-Meier method was 12·3 years (95% CI 11·9-12·6) for PORTEC-1, 10·5 years (10·2-10·7) for PORTEC-2, and 6·1 years (5·9-6·3) for PORTEC-3. Both overall recurrence and endometrial cancer-specific death significantly increased with age. Moreover, older women had a higher frequency of deep myometrial invasion, serous tumour histology, and p53-abnormal tumours. Age was an independent risk factor for both overall recurrence (hazard ratio [HR] 1·02 per year, 95% CI 1·01-1·04; p=0·0012) and endometrial cancer-specific death (HR 1·03 per year, 1·01-1·05; p=0·0012) and was identified as a significant causal variable. INTERPRETATION: This study showed that advanced age was associated with more aggressive tumour features in women with endometrial cancer, and was independently and causally related to worse oncological outcomes. Therefore, our findings suggest that older women with endometrial cancer should not be excluded from diagnostic assessments, molecular testing, and adjuvant therapy based on their age alone. FUNDING: None.

Prediction of recurrence risk in endometrial cancer with multimodal deep learning.

Predicting distant recurrence of endometrial cancer (EC) is crucial for personalized adjuvant treatment. The current gold standard of combined pathological and molecular profiling is costly, hampering implementation. Here we developed HECTOR (histopathology-based endometrial cancer tailored outcome risk), a multimodal deep learning prognostic model using hematoxylin and eosin-stained, whole-slide images and tumor stage as input, on 2,072 patients from eight EC cohorts including the PORTEC-1/-2/-3 randomized trials. HECTOR demonstrated C-indices in internal (n = 353) and two external (n = 160 and n = 151) test sets of 0.789, 0.828 and 0.815, respectively, outperforming the current gold standard, and identified patients with markedly different outcomes (10-year distant recurrence-free probabilities of 97.0%, 77.7% and 58.1% for HECTOR low-, intermediate- and high-risk groups, respectively, by Kaplan-Meier analysis). HECTOR also predicted adjuvant chemotherapy benefit better than current methods. Morphological and genomic feature extraction identified correlates of HECTOR risk groups, some with therapeutic potential. HECTOR improves on the current gold standard and may help delivery of personalized treatment in EC.

Molecular and Clinicopathologic Characterization of Mismatch Repair-Deficient Endometrial Carcinoma Not Related to MLH1 Promoter Hypermethylation.

Universal tumor screening in endometrial carcinoma (EC) is increasingly adopted to identify individuals at risk of Lynch syndrome (LS). These cases involve mismatch repair-deficient (MMRd) EC without MLH1 promoter hypermethylation (PHM). LS is confirmed through the identification of germline MMR pathogenic variants (PV). In cases where these are not detected, emerging evidence highlights the significance of double-somatic MMR gene alterations as a sporadic cause of MMRd, alongside POLE/POLD1 exonuclease domain (EDM) PV leading to secondary MMR PV. Our understanding of the incidence of different MMRd EC origins not related to MLH1-PHM, their associations with clinicopathologic characteristics, and the prognostic implications remains limited. In a combined analysis of the PORTEC-1, -2, and -3 trials (n = 1254), 84 MMRd EC not related to MLH1-PHM were identified that successfully underwent paired tumor-normal tissue next-generation sequencing of the MMR and POLE/POLD1 genes. Among these, 37% were LS associated (LS-MMRd EC), 38% were due to double-somatic hits (DS-MMRd EC), and 25% remained unexplained. LS-MMRd EC exhibited higher rates of MSH6 (52% vs 19%) or PMS2 loss (29% vs 3%) than DS-MMRd EC, and exclusively showed MMR-deficient gland foci. DS-MMRd EC had higher rates of combined MSH2/MSH6 loss (47% vs 16%), loss of >2 MMR proteins (16% vs 3%), and somatic POLE-EDM PV (25% vs 3%) than LS-MMRd EC. Clinicopathologic characteristics, including age at tumor onset and prognosis, did not differ among the various groups. Our study validates the use of paired tumor-normal next-generation sequencing to identify definitive sporadic causes in MMRd EC unrelated to MLH1-PHM. MMR immunohistochemistry and POLE-EDM mutation status can aid in the differentiation between LS-MMRd EC and DS-MMRd EC. These findings emphasize the need for integrating tumor sequencing into LS diagnostics, along with clear interpretation guidelines, to improve clinical management. Although not impacting prognosis, confirmation of DS-MMRd EC may release patients and relatives from burdensome LS surveillance.

MLH1 Promotor Hypermethylation in Colorectal and Endometrial Carcinomas from Patients with Lynch Syndrome.

Screening for Lynch syndrome (LS) in colorectal cancer (CRC) and endometrial cancer patients generally involves immunohistochemical staining of the mismatch repair (MMR) proteins. In case of MLH1 protein loss, MLH1 promotor hypermethylation (MLH1-PM) testing is performed to indirectly distinguish the constitutional MLH1 variants from somatic epimutations. Recently, multiple studies have reported that MLH1-PM and pathogenic constitutional MMR variants are not mutually exclusive. This study describes 6 new and 86 previously reported MLH1-PM CRCs or endometrial cancers in LS patients. Of these, methylation of the MLH1 gene promotor C region was reported in 30 MLH1, 6 MSH2, 6 MSH6, and 3 PMS2 variant carriers at a median age at diagnosis of 48.5 years [interquartile range (IQR), 39-56.75 years], 39 years (IQR, 29-51 years), 58 years (IQR, 53.5-67 years), and 68 years (IQR, 65.6-68.5 years), respectively. For 31 MLH1-PM CRCs in LS patients from the literature, only the B region of the MLH1 gene promotor was tested, whereas for 13 cases in the literature the tested region was not specified. Collectively, these data indicate that a diagnosis of LS should not be excluded when MLH1-PM is detected. Clinicians should carefully consider whether follow-up genetic MMR gene testing should be offered, with age <60 to 70 years and/or a positive family history among other factors being suggestive for a potential constitutional MMR gene defect.

Molecular Classification Predicts Response to Radiotherapy in the Randomized PORTEC-1 and PORTEC-2 Trials for Early-Stage Endometrioid Endometrial Cancer.

PURPOSE: The molecular classification of endometrial cancer (EC) has proven to have prognostic value and is predictive of response to adjuvant chemotherapy. Here, we investigate its predictive value for response to external beam radiotherapy (EBRT) and vaginal brachytherapy (VBT) in early-stage endometrioid EC (EEC). METHODS: Data of the randomized PORTEC-1 trial (n = 714) comparing pelvic EBRT with no adjuvant therapy in early-stage intermediate-risk EC and the PORTEC-2 trial (n = 427) comparing VBT with EBRT in early-stage high-intermediate-risk EC were used. Locoregional (including vaginal and pelvic) recurrence-free survival was compared between treatment groups across the four molecular classes using Kaplan-Meier's methodology and log-rank tests. RESULTS: A total of 880 molecularly classified ECs, 484 from PORTEC-1 and 396 from PORTEC-2, were included. The majority were FIGO-2009 stage I EEC (97.2%). The median follow-up was 11.3 years. No locoregional recurrences were observed in EC with a pathogenic mutation of DNA polymerase-ε (POLEmut EC). In mismatch repair-deficient (MMRd) EC, locoregional recurrence-free survival was similar after EBRT (94.2%), VBT (94.2%), and no adjuvant therapy (90.3%; P = .74). In EC with a p53 abnormality (p53abn EC), EBRT (96.9%) had a substantial benefit over VBT (64.3%) and no adjuvant therapy (72.2%; P = .048). In EC with no specific molecular profile (NSMP EC), both EBRT (98.3%) and VBT (96.2%) yielded better locoregional control than no adjuvant therapy (87.7%; P < .0001). CONCLUSION: The molecular classification of EC predicts response to radiotherapy in stage I EEC and may guide adjuvant treatment decisions. Omitting radiotherapy seems to be safe in POLEmut EC. The benefit of radiotherapy seems to be limited in MMRd EC. EBRT yields a significantly better locoregional recurrence-free survival than VBT or no adjuvant therapy in p53abn EC. VBT is the treatment of choice for NSMP EC as it is as effective as EBRT and significantly better than no adjuvant therapy for locoregional tumor control.

QPOLE: A Quick, Simple, and Cheap Alternative for POLE Sequencing in Endometrial Cancer by Multiplex Genotyping Quantitative Polymerase Chain Reaction.

PURPOSE: Detection of 11 pathogenic variants in the POLE gene in endometrial cancer (EC) is critically important to identify women with a good prognosis and reduce overtreatment. Currently, POLE status is determined by DNA sequencing, which can be expensive, relatively time-consuming, and unavailable in hospitals without specialized equipment and personnel. This may hamper the implementation of POLE-testing in clinical practice. To overcome this, we developed and validated a rapid, low-cost POLE hotspot test by a quantitative polymerase chain reaction (qPCR) assay, QPOLE. MATERIALS AND METHODS: Primer and fluorescence-labeled 5'-nuclease probe sequences of the 11 established pathogenic POLE mutations were designed. Three assays, QPOLE-frequent for the most common mutations and QPOLE-rare-1 and QPOLE-rare-2 for the rare variants, were developed and optimized using DNA extracted from formalin-fixed paraffin-embedded tumor tissues. The simplicity of the design enables POLE status assessment within 4-6 hours after DNA isolation. An interlaboratory external validation study was performed to determine the practical feasibility of this assay. RESULTS: Cutoffs for POLE wild-type, POLE-mutant, equivocal, and failed results were predefined on the basis of a subset of POLE mutants and POLE wild-types for the internal and external validation. For equivocal cases, additional DNA sequencing is recommended. Performance in 282 EC cases, of which 99 were POLE-mutated, demonstrated an overall accuracy of 98.6% (95% CI, 97.2 to 99.9), a sensitivity of 95.2% (95% CI, 90.7 to 99.8), and a specificity of 100%. After DNA sequencing of 8.8% equivocal cases, the final sensitivity and specificity were 96.0% (95% CI, 92.1 to 99.8) and 100%. External validation confirmed feasibility and accuracy. CONCLUSION: QPOLE is a qPCR assay that is a quick, simple, and reliable alternative for DNA sequencing. QPOLE detects all pathogenic variants in the exonuclease domain of the POLE gene. QPOLE will make low-cost POLE-testing available for all women with EC around the globe.

Prognostic refinement of NSMP high-risk endometrial cancers using oestrogen receptor immunohistochemistry.

BACKGROUND: Risk-assessment of endometrial cancer (EC) is based on clinicopathological factors and molecular subgroup. It is unclear whether adding hormone receptor expression, L1CAM expression or CTNNB1 status yields prognostic refinement. METHODS: Paraffin-embedded tumour samples of women with high-risk EC (HR-EC) from the PORTEC-3 trial (n = 424), and a Dutch prospective clinical cohort called MST (n = 256), were used. All cases were molecularly classified. Expression of L1CAM, ER and PR were analysed by whole-slide immunohistochemistry and CTNNB1 mutations were assessed with a next-generation sequencing. Kaplan-Meier method, log-rank tests and Cox's proportional hazard models were used for survival analysis. RESULTS: In total, 648 HR-EC were included. No independent prognostic value of ER, PR, L1CAM, and CTNNB1 was found, while age, stage, and adjuvant chemotherapy had an independent impact on risk of recurrence. Subgroup-analysis showed that only in NSMP HR-EC, ER-positivity was independently associated with a reduced risk of recurrence (HR 0.33, 95%CI 0.15-0.75). CONCLUSIONS: We confirmed the prognostic impact of the molecular classification, age, stage, and adjuvant CTRT in a large cohort of high-risk EC. ER-positivity is a strong favourable prognostic factor in NSMP HR-EC and identifies a homogeneous subgroup of NSMP tumours. Assessment of ER status in high-risk NSMP EC is feasible in clinical practice and could improve risk stratification and treatment.

Interpretable deep learning model to predict the molecular classification of endometrial cancer from haematoxylin and eosin-stained whole-slide images: a combined analysis of the PORTEC randomised trials and clinical cohorts.

BACKGROUND: Endometrial cancer can be molecularly classified into POLEmut, mismatch repair deficient (MMRd), p53 abnormal (p53abn), and no specific molecular profile (NSMP) subgroups. We aimed to develop an interpretable deep learning pipeline for whole-slide-image-based prediction of the four molecular classes in endometrial cancer (im4MEC), to identify morpho-molecular correlates, and to refine prognostication. METHODS: This combined analysis included diagnostic haematoxylin and eosin-stained slides and molecular and clinicopathological data from 2028 patients with intermediate-to-high-risk endometrial cancer from the PORTEC-1 (n=466), PORTEC-2 (n=375), and PORTEC-3 (n=393) randomised trials and the TransPORTEC pilot study (n=110), the Medisch Spectrum Twente cohort (n=242), a case series of patients with POLEmut endometrial cancer in the Leiden Endometrial Cancer Repository (n=47), and The Cancer Genome Atlas-Uterine Corpus Endometrial Carcinoma cohort (n=395). PORTEC-3 was held out as an independent test set and a four-fold cross validation was performed. Performance was measured with the macro and class-wise area under the receiver operating characteristic curve (AUROC). Whole-slide images were segmented into tiles of 360 µm resized to 224 × 224 pixels. im4MEC was trained to learn tile-level morphological features with self-supervised learning and to molecularly classify whole-slide images with an attention mechanism. The top 20 tiles with the highest attention scores were reviewed to identify morpho-molecular correlates. Predictions of a nuclear classification deep learning model serve to derive interpretable morphological features. We analysed 5-year recurrence-free survival and explored prognostic refinement by molecular class using the Kaplan-Meier method. FINDINGS: im4MEC attained macro-average AUROCs of 0·874 (95% CI 0·856-0·893) on four-fold cross-validation and 0·876 on the independent test set. The class-wise AUROCs were 0·849 for POLEmut (n=51), 0·844 for MMRd (n=134), 0·883 for NSMP (n=120), and 0·928 for p53abn (n=88). POLEmut and MMRd tiles had a high density of lymphocytes, p53abn tiles had strong nuclear atypia, and the morphology of POLEmut and MMRd endometrial cancer overlapped. im4MEC highlighted a low tumour-to-stroma ratio as a potentially novel characteristic feature of the NSMP class. 5-year recurrence-free survival was significantly different between im4MEC predicted molecular classes in PORTEC-3 (log-rank p<0·0001). The ten patients with aggressive p53abn endometrial cancer that was predicted as MMRd showed inflammatory morphology and appeared to have a better prognosis than patients with correctly predicted p53abn endometrial cancer (p=0·30). The four patients with NSMP endometrial cancer that was predicted as p53abn showed higher nuclear atypia and appeared to have a worse prognosis than patients with correctly predicted NSMP (p=0·13). Patients with MMRd endometrial cancer predicted as POLEmut had an excellent prognosis, as do those with true POLEmut endometrial cancer. INTERPRETATION: We present the first interpretable deep learning model, im4MEC, for haematoxylin and eosin-based prediction of molecular endometrial cancer classification. im4MEC robustly identified morpho-molecular correlates and could enable further prognostic refinement of patients with endometrial cancer. FUNDING: The Hanarth Foundation, the Promedica Foundation, and the Swiss Federal Institutes of Technology.

p53 immunohistochemistry in endometrial cancer: clinical and molecular correlates in the PORTEC-3 trial.

Standard molecular classification of endometrial cancers (EC) is now endorsed by the WHO and identifies p53-abnormal (p53abn) EC as the subgroup with the poorest prognosis and the most likely to benefit from adjuvant chemo(radio)therapy. P53abn EC are POLE wildtype, mismatch repair proficient and show abnormal immunohistochemical (IHC) staining for p53. Correct interpretation of routinely performed p53 IHC has therefore become of paramount importance. We aimed to comprehensively investigate abnormal p53 IHC patterns and their relation to clinicopathological and molecular features. Tumor material of 411 molecularly classified high-risk EC from consenting patients from the PORTEC-3 clinical trial were collected. p53 IHC was successful in 408 EC and was considered abnormal when the tumor showed a mutant expression pattern (including subclonal): overexpression, null or cytoplasmic. The presence of pathogenic mutations was determined by next generation sequencing (NGS). Abnormal p53 expression was observed in 131/408 (32%) tumors. The most common abnormal p53 IHC pattern was overexpression (n = 89, 68%), followed by null (n = 12, 9%) and cytoplasmic (n = 3, 2%). Subclonal abnormal p53 staining was observed in 27 cases (21%), which was frequently but not exclusively, associated with POLE mutations and/or MMRd (n = 22/27; p < 0.001). Agreement between p53 IHC and TP53 NGS was observed in 90.7%, resulting in a sensitivity and specificity of 83.6% and 94.3%, respectively. Excluding POLEmut and MMRd EC, as per the WHO-endorsed algorithm, increased the accuracy to 94.5% with sensitivity and specificity of 95.0% and 94.1%, respectively. Our data shows that awareness of the abnormal p53 IHC patterns are prerequisites for correct EC molecular classification. Subclonal abnormal p53 expression is a strong indicator for POLEmut and/or MMRd EC. No significant differences in clinical outcomes were observed among the abnormal p53 IHC patterns. Our data support use of the WHO-endorsed algorithm and combining the different abnormal p53 IHC patterns into one diagnostic entity (p53abn EC).

The impact of a pathologist's personality on the interobserver variability and diagnostic accuracy of predictive PD-L1 immunohistochemistry in lung cancer.

OBJECTIVES: Programmed death-ligand 1 (PD-L1) is the only approved predictive biomarker for immunotherapy in non-small cell lung cancer (NSCLC). However, predictive PD-L1 immunohistochemistry is subject to interobserver variability. We hypothesized that a pathologist's personality influences the interobserver variability and diagnostic accuracy of PD-L1 immunoscoring. MATERIALS AND METHODS: Seventeen pathologists performed PD-L1 immunoscoring on 50 resected NSCLC tumors in three categories (<1%;1-49%;≥50%). Also, the pathologists completed a certified personality test (NEO-PI-r), assessing five personality traits: neuroticism, extraversion, openness, altruism and conscientiousness. RESULTS: The overall agreement among pathologists for a series of 47 tumors was substantial (kappa = 0.63). Of these, 23/47 (49%) tumors were entirely negative or largely positive, resulting in a kappa value of 0.93. The remaining 24/47 (51%) tumors had a PD-L1 score around the cutoff value, generating a kappa value of 0.32. Pathologists with high scores for conscientiousness (careful, diligent) had the least interobserver variability (r = 0.6, p = 0.009). Also, they showed a trend towards higher sensitivity (74% vs. 68%, p = 0.4), specificity (86% vs. 82%, p = 0.3) and percent agreement (83% vs. 79%, p = 0.3), although not significant. In contrast, pathologists with high scores for neuroticism (sensitive, anxious) had significantly lower specificity (80% vs. 87%, p = 0.03) and percent agreement (78% vs. 85%, p = 0.03). Also, a trend towards high interobserver variability (r = -0.3, p = 0.2) and lower sensitivity (68% vs. 74%, p = 0.3) was observed, although not significant. Pathologists with relatively high scores for conscientiousness scored fewer tumors PD-L1 positive at the ≥ 1% cut-off (r = -0.5, p = 0.03). In contrast, pathologists with relatively high scores for neuroticism score more tumors PD-L1 positive at ≥ 1% (r = 0.6, p = 0.017) and ≥ 50% cut-offs (r = 0.6, p = 0.009). CONCLUSIONS: This study is the first to demonstrate the impact of a pathologist's personality on the interobserver variability and diagnostic accuracy of immunostaining, in the context of PD-L1 in NSCLC. Larger studies are needed for validation of these findings.

Tertiary lymphoid structures critical for prognosis in endometrial cancer patients.

B-cells play a key role in cancer suppression, particularly when aggregated in tertiary lymphoid structures (TLS). Here, we investigate the role of B-cells and TLS in endometrial cancer (EC). Single cell RNA-sequencing of B-cells shows presence of naïve B-cells, cycling/germinal center B-cells and antibody-secreting cells. Differential gene expression analysis shows association of TLS with L1CAM overexpression. Immunohistochemistry and co-immunofluorescence show L1CAM expression in mature TLS, independent of L1CAM expression in the tumor. Using L1CAM as a marker, 378 of the 411 molecularly classified ECs from the PORTEC-3 biobank are evaluated, TLS are found in 19%. L1CAM expressing TLS are most common in mismatch-repair deficient (29/127, 23%) and polymerase-epsilon mutant EC (24/47, 51%). Multivariable Cox regression analysis shows strong favorable prognostic impact of TLS, independent of clinicopathological and molecular factors. Our data suggests a pivotal role of TLS in outcome of EC patients, and establishes L1CAM as a simple biomarker.

Prognostic relevance of the molecular classification in high-grade endometrial cancer for patients staged by lymphadenectomy and without adjuvant treatment.

INTRODUCTION: The clinical role of the molecular endometrial cancer (EC) classification has not been fully explored in patients staged with lymphadenectomy or without adjuvant treatment, conditions that could potentially moderate the prognostic value of the classification. We aimed to evaluate the clinical outcome of the molecular subgroups in patients with high-grade EC staged by lymphadenectomy and those without adjuvant treatment. METHODS: DNA-sequencing for the detection of pathogenic POLE-exonuclease domain mutations and immunohistochemistry for mismatch repair (MMR) proteins and p53 expression were performed on 412 high-grade EC from the Danish Gynaecological Cancer Database (2005-2012) to classify them as POLE-ultramutated (POLEmut), MMR-deficient (MMRd), p53-mutant (p53abn), or no specific molecular profile (NSMP). Patients with stage IV or residual disease after surgery were excluded. Kaplan-Meier method, log-rank test and Cox proportional hazard models were used for analysis. RESULTS: Molecular analysis was successful in 367 EC; 251 patients had undergone lymphadenectomy. Five-year recurrence rates in this subgroup of patients was 36.7% for women with p53abn EC, 0.0% for POLEmut EC, 13.4% for MMRd EC and 42.9% for NSMP EC (p < 0.001). Similar results were observed among stage IA-IB patients. Among patients without adjuvant treatment (n = 264), none with POLEmut EC (n = 26) had a recurrence. CONCLUSION: The molecular EC classification has strong prognostic value, independent of clinicopathological factors, also among high-grade EC patients staged by lymphadenectomy and those without adjuvant treatment. The unfavourable prognosis of early-stage p53abn EC is not due to undetected lymph node metastasis, and the indolent behaviour of POLEmut EC is independent of adjuvant treatment.

Substantial Lymphovascular Space Invasion Is an Adverse Prognostic Factor in High-Risk Endometrial Cancer.

Approximately 15% of patients with endometrial cancer present with high-risk disease (HREC). Moreover, assessing the extent of lymphovascular space invasion (LVSI) may provide prognostic insight among patients with HREC. The aim of this study was to determine whether the extent of LVSI can serve as a prognostic factor in HREC. All cases of ESMO-ESGO-ESTRO 2016 classified HREC in the Danish Gynecological Cancer Database (DGCD) diagnosed from 2005 to 2012 were reviewed for the presence and extent of LVSI (categorized using a 3-tiered definition). We used the Kaplan-Meier analysis to calculate actuarial survival rates, both adjusted and unadjusted Cox regression analyses were used to calculate the proportional hazard ratio (HR). A total of 376 patients were included in our analysis. Among 305 patients with stage I/II HREC, 8.2% and 6.2% had focal or substantial LVSI, respectively, compared with 12.7% and 38.0% of 71 patients with stage III/IV HREC, respectively. Moreover, the estimated 5-yr recurrence-free survival rate was significantly lower among patients with substantial LVSI compared with patients with no LVSI for both stage I/II (HR: 2.8; P=0.011) and stage III/IV (HR: 2.9; P=0.003) patients. Similarly, overall survival was significantly lower among patients with substantial LVSI for both stage I/II (HR: 3.1; P<0.001) and stage III/IV (HR: 3.2; P=0.020) patients. In patients with HREC, substantial LVSI is an independent adverse prognostic factor for lymph node and distant metastases, leading to reduced survival. Thus, the extent of LVSI should be incorporated into routine pathology reports in order to guide the appropriate choice of adjuvant treatment.

Defining Substantial Lymphovascular Space Invasion in Endometrial Cancer.

Lymphovascular space invasion (LVSI) occurs in a minority of endometrial cancer (EC) cases, and the extent of LVSI is an important risk factor for recurrence and/or metastases. Our aim was to improve the reproducibility of measuring clinically meaningful LVSI by performing a quantitative analysis of the correlation between LVSI and the risk of pelvic lymph node recurrence in EC. EC samples from PORTEC-1 and PORTEC-2 trials were retrieved and used to collect quantitative data, including the number of LVSI-positive vessels per H&E-stained slide. Using a predefined threshold for clinical relevance, the risk of pelvic lymph node recurrence risk was calculated (Kaplan-Meier method, with Cox regression) using a stepwise adjustment for the number of LVSI-positive vessels. This analysis was then repeated in the Danish Gynecological Cancer Database (DGCD) cohort. Among patients in PORTEC-1 and PORTEC-2 trials who did not receive external beam radiotherapy, the 5-yr pelvic lymph node recurrence risk was 3.3%, 6.7% (P=0.51), and 26.3% (P<0.001), respectively when 0, 1 to 3, or ≥4 vessels had LVSI involvement; similar results were obtained for the DGCD cohort. Furthermore, both the average number of tumor cells in the largest embolus and the number of LVSI-positive H&E slides differed significantly between focal LVSI and substantial LVSI. On the basis of these results, we propose a numeric threshold (≥4 LVSI-involved vessels in at least one H&E slide) for defining clinically relevant LVSI in EC, thereby adding supportive data to the semiquantitative approach. This will help guide gynecologic pathologists to differentiate between focal and substantial LVSI, especially in borderline cases.

Development and validation of a supervised deep learning algorithm for automated whole-slide programmed death-ligand 1 tumour proportion score assessment in non-small cell lung cancer.

AIMS: Immunohistochemical programmed death-ligand 1 (PD-L1) staining to predict responsiveness to immunotherapy in patients with advanced non-small cell lung cancer (NSCLC) has several drawbacks: a robust gold standard is lacking, and there is substantial interobserver and intraobserver variance, with up to 20% discordance around cutoff points. The aim of this study was to develop a new deep learning-based PD-L1 tumour proportion score (TPS) algorithm, trained and validated on a routine diagnostic dataset of digitised PD-L1 (22C3, laboratory-developed test)-stained samples. METHODS AND RESULTS: We designed a fully supervised deep learning algorithm for whole-slide PD-L1 assessment, consisting of four sequential convolutional neural networks (CNNs), using aiforia create software. We included 199 whole slide images (WSIs) of 'routine diagnostic' histology samples from stage IV NSCLC patients, and trained the algorithm by using a training set of 60 representative cases. We validated the algorithm by comparing the algorithm TPS with the reference score in a held-out validation set. The algorithm had similar concordance with the reference score (79%) as the pathologists had with one another (75%). The intraclass coefficient was 0.96 and Cohen's κ coefficient was 0.69 for the algorithm. Around the 1% and 50% cutoff points, concordance was also similar between pathologists and the algorithm. CONCLUSIONS: We designed a new, deep learning-based PD-L1 TPS algorithm that is similarly able to assess PD-L1 expression in daily routine diagnostic cases as pathologists. Successful validation on routine diagnostic WSIs and detailed visual feedback show that this algorithm meets the requirements for functioning as a 'scoring assistant'.

Prognostic impact of histological review of high-grade endometrial carcinomas in a large Danish cohort.

The aim of this study was to investigate the outcome of histological subtype review of high-grade endometrial carcinoma (EC) and its prognostic impact in a large well-documented Danish nationwide cohort. From the Danish Gynecological Cancer Database (DGCD) 2005-2012 cohort, we included 425 patients with an original diagnosis of high-grade EC, independent of histologic subtype. Of these, at least one hematoxylin and eosin (H&E)-stained slide from 396 cases (93.2%) was available for review. The histologic subtype was reviewed by specialized gynecopathologists blinded to the original diagnosis and clinical outcome. Interobserver variability between original and revised histologic subtypes was analyzed using simple Kappa statistics. Hazard ratios (HR), recurrence-free survival (RFS), and overall survival were calculated for original and revised subtypes, respectively. Overall histologic subtype agreement was moderate (kappa = 0.42) with the highest agreement for endometrioid-type EC (EEC; 75.5%) and serous-type EC (SEC; 63.8%). For clear cell carcinoma and un-/dedifferentiated EC, agreement was significantly lower: 30.1% and 33.3% respectively. Of the 396 reviewed cases, only two (0.5%) were re-classified as low-grade EEC upon revision. Interestingly, GR3 EEC had better RFS than SEC with stronger significance after revision (HR 2.36 (95% CI 1.43-3.89), p = 0.001), compared to original diagnosis (HR 1.74 (95% CI 1.07-2.81), p = 0.024). In conclusion, this study confirmed that pathology review results in substantial shift in histological subtype in high-grade EC. After review, a stronger prognostic benefit for GR3 EEC as compared to other histological subtypes was observed. This work supports maintaining a low threshold for pathology revision of high-grade EC in clinical practice.

Prevalence and Prognosis of Lynch Syndrome and Sporadic Mismatch Repair Deficiency in Endometrial Cancer.

BACKGROUND: Standard screening of endometrial cancer (EC) for Lynch syndrome (LS) is gaining traction; however, the prognostic impact of an underlying hereditary etiology is unknown. We established the prevalence, prognosis, and subsequent primary cancer incidence of patients with LS-associated EC in relation to sporadic mismatch repair deficient (MMRd)-EC in the large combined Post Operative Radiation Therapy in Endometrial Carcinoma-1, -2, and -3 trial cohort. METHODS: After MMR-immunohistochemistry, MLH1-promoter methylation testing, and next-generation sequencing, tumors were classified into 3 groups according to the molecular cause of their MMRd-EC. Kaplan-Meier method, log-rank test, and Cox model were used for survival analysis. Competing risk analysis was used to estimate the subsequent cancer probability. All statistical tests were 2-sided. RESULTS: Among the 1336 ECs, 410 (30.7%) were MMRd. A total of 380 (92.7%) were fully triaged: 275 (72.4%) were MLH1-hypermethylated MMRd-ECs; 36 (9.5%) LS MMRd-ECs, and 69 (18.2%) MMRd-ECs due to other causes. Limiting screening of EC patients to 60 years or younger or to 70 years or younger would have resulted in missing 18 (50.0%) and 6 (16.7%) LS diagnoses, respectively. Five-year recurrence-free survival was 91.7% (95% confidence interval [CI] = 83.1% to 100%; hazard ratio = 0.45, 95% CI = 0.16 to 1.24, P = .12) for LS, 95.5% (95% CI = 90.7% to 100%; hazard ratio = 0.17, 95% CI = 0.05 to 0.55, P = .003) for "other" vs 78.6% (95% CI = 73.8% to 83.7%) for MLH1-hypermethylated MMRd-EC. The probability of subsequent LS-associated cancer at 10 years was 11.6% (95% CI = 0.0% to 24.7%), 1.5% (95% CI = 0.0% to 4.3%), and 7.0% (95% CI = 3.0% to 10.9%) within the LS, "other," and MLH1-hypermethylated MMRd-EC groups, respectively. CONCLUSIONS: The LS prevalence in the Post Operative Radiation Therapy in Endometrial Carcinoma trial population was 2.8% and among MMRd-ECs was 9.5%. Patients with LS-associated ECs showed a trend towards better recurrence-free survival and higher risk for second cancers compared with patients with MLH1-hypermethylated MMRd-EC.

Endometrial Cancer Risk in Women With Germline BRCA1 or BRCA2 Mutations: Multicenter Cohort Study.

BACKGROUND: Endometrial cancer (EC) risk in BReast CAncer gene 1/2 (BRCA1/2) mutation carriers is uncertain; therefore, we assessed this in a large Dutch nationwide cohort study. METHODS: We selected 5980 BRCA1/2 (3788 BRCA1, 2151 gBRCA2, 41 both BRCA1/BRCA2) and 8451 non-BRCA1/2 mutation carriers from the Hereditary Breast and Ovarian cancer study, the Netherlands cohort. Follow-up started at the date of the nationwide Dutch Pathology Registry coverage (January 1, 1989) or at the age of 25 years (whichever came last) and ended at date of EC diagnosis, last follow-up, or death (whichever came first). EC risk in BRCA1/2 mutation carriers was compared with 1) the general population, estimating standardized incidence ratios (SIRs) based on Dutch population-based incidence rates; and 2) non-BRCA1/2 mutation carriers, using Cox-regression analyses, expressed as hazard ratio (HR). Statistical tests were 2-sided. RESULTS: Fifty-eight BRCA1/2 and 33 non-BRCA1/2 mutation carriers developed EC over 119 296 and 160 841 person-years, respectively (SIR = 2.83, 95% confidence interval [CI] = 2.18 to 3.65; and HR = 2.37, 95% CI = 1.53 to 3.69, respectively). gBRCA1 mutation carriers showed increased risks for EC overall (SIR = 3.51, 95% CI = 2.61 to 4.72; HR = 2.91, 95% CI = 1.83 to 4.66), serous-like EC (SIR = 12.64, 95% CI = 7.62 to 20.96; HR = 10.48, 95% CI = 2.95 to 37.20), endometrioid EC (SIR = 2.63, 95% CI = 1.80 to 3.83; HR = 2.01, 95% CI = 1.18 to 3.45), and TP53-mutated EC (HR = 15.71, 95% CI = 4.62 to 53.40). For BRCA2 mutation carriers, overall (SIR = 1.70, 95% CI = 1.01 to 2.87) and serous-like EC risks (SIR = 5.11, 95% CI = 1.92 to 13.63) were increased compared with the general population. Absolute risks by 75 years remained low (overall EC = 3.0%; serous-like EC = 1.1%). CONCLUSIONS: BRCA1/2 mutation carriers have a two- to threefold increased risk for EC, with highest risk observed for the rare subgroups of serous-like and p53-abnormal EC in BRCA1 mutation carriers.

Long-Term Toxicity and Health-Related Quality of Life After Adjuvant Chemoradiation Therapy or Radiation Therapy Alone for High-Risk Endometrial Cancer in the Randomized PORTEC-3 Trial.

PURPOSE: The survival results of the PORTEC-3 trial showed a significant improvement in both overall and failure-free survival with chemoradiation therapy versus pelvic radiation therapy alone. The present analysis was performed to compare long-term adverse events (AE) and health-related quality of life (HRQOL). METHODS AND MATERIALS: In the study, 660 women with high-risk endometrial cancer were randomly assigned to receive chemoradiation therapy (2 concurrent cycles of cisplatin followed by 4 cycles of carboplatin/paclitaxel) or radiation therapy alone. Toxicity was graded using Common Terminology Criteria for Adverse Events, version 3.0. HRQOL was measured using EORTC QLQ-C30 and CX24/OV28 subscales and compared with normative data. An as-treated analysis was performed. RESULTS: Median follow-up was 74.6 months; 574 (87%) patients were evaluable for HRQOL. At 5 years, grade ≥2 AE were scored for 78 (38%) patients who had received chemoradiation therapy versus 46 (24%) who had received radiation therapy alone (P = .008). Grade 3 AE did not differ significantly between the groups (8% vs 5%, P = .18) at 5 years, and only one new late grade 4 toxicity had been reported. At 3 and 5 years, sensory neuropathy toxicity grade ≥2 persisted after chemoradiation therapy in 6% (vs 0% after radiation therapy, P < .001) and more patients reported significant tingling or numbness at HRQOL (27% vs 8%, P < .001 at 3 years; 24% vs 9%, P = .002 at 5 years). Up to 3 years, more patients who had chemoradiation therapy reported limb weakness (21% vs 5%, P < .001) and lower physical (79 vs 87, P < .001) and role functioning (78 vs 88, P < .001) scores. Both treatment groups reported similar long-term global health/quality of life scores, which were better than those of the normative population. CONCLUSIONS: This study shows a long-lasting, clinically relevant, negative impact of chemoradiation therapy on toxicity and HRQOL, most importantly persistent peripheral sensory neuropathy. Physical and role functioning impairments were seen until 3 years. These long-term data are essential for patient information and shared decision-making regarding adjuvant chemotherapy for high-risk endometrial cancer.

HER2 Status in High-Risk Endometrial Cancers (PORTEC-3): Relationship with Histotype, Molecular Classification, and Clinical Outcomes.

HER2 status has not been investigated in the context of the molecular endometrial cancer (EC) classification. Here, we aimed to determine the clinicopathological features and prognostic significance of the HER2 status in the molecularly classified PORTEC-3 trial population of patients with high-risk EC (HREC). HER2 testing was performed on tumor tissues of 407 molecularly classified HREC. HER2 status was determined by HER2 immunohistochemistry (IHC; all cases) and subsequent HER2 dual in situ hybridization for cases with any (in) complete moderate to strong membranous HER2 IHC expression. The Χ2 test and Spearman's Rho correlation coefficient were used to compare clinicopathological and molecular features. The Kaplan-Meier method, log-rank test, and Cox proportional hazards models were used for survival analysis. We identified 24 (5.9%) HER2-positive EC of various histological subtypes including serous (n = 9, 37.5%), endometrioid (n = 6, 25.0%), and clear cell (n = 5, 20.8%). HER2 positivity was highly associated with the p53-abnormal subgroup (p53abn, 23/24 cases; p < 0.0001). The correlation between p53abn and the HER2 status (ρ = 0.438; p < 0.0001) was significantly stronger (p < 0.0001) than between serous histology and the HER2 status (ρ = 0.154; p = 0.002). HER2 status did not have independent prognostic value for survival after correction for the molecular classification. Our study strongly suggests that molecular subclass-directed HER2 testing is superior to histotype-directed testing. This insight will be relevant for future trials targeting HER2.