Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge

Recognition of mitotic figures in histologic tumor specimens is highly relevant to patient outcome assessment. This task is challenging for algorithms and human experts alike, with deterioration of algorithmic performance under shifts in image representations. Considerable covariate shifts occur when assessment is performed on different tumor types, images are acquired using different digitization devices, or specimens are produced in different laboratories. This observation motivated the inception of the 2022 challenge on MItosis Domain Generalization (MIDOG 2022). The challenge provided annotated histologic tumor images from six different domains and evaluated the algorithmic approaches for mitotic figure detection provided by nine challenge participants on ten independent domains. Ground truth for mitotic figure detection was established in two ways: a three-expert majority vote and an independent, immunohistochemistry-assisted set of labels. This work represents an overview of the challenge tasks, the algorithmic strategies employed by the participants, and potential factors contributing to their success. With an F1 score of 0.764 for the top-performing team, we summarize that domain generalization across various tumor domains is possible with today's deep learning-based recognition pipelines. However, we also found that domain characteristics not present in the training set (feline as new species, spindle cell shape as new morphology and a new scanner) led to small but significant decreases in performance. When assessed against the immunohistochemistry-assisted reference standard, all methods resulted in reduced recall scores, with only minor changes in the order of participants in the ranking.

Image-based consensus molecular subtyping in rectal cancer biopsies and response to neoadjuvant chemoradiotherapy

The development of deep learning (DL) models to predict the consensus molecular subtypes (CMS) from histopathology images (imCMS) is a promising and cost-effective strategy to support patient stratification. Here, we investigate whether imCMS calls generated from whole slide histopathology images (WSIs) of rectal cancer (RC) pre-treatment biopsies are associated with pathological complete response (pCR) to neoadjuvant long course chemoradiotherapy (LCRT) with single agent fluoropyrimidine. DL models were trained to classify WSIs of colorectal cancers stained with hematoxylin and eosin into one of the four CMS classes using a multi-centric dataset of resection and biopsy specimens (n = 1057 WSIs) with paired transcriptional data. Classifiers were tested on a held out RC biopsy cohort (ARISTOTLE) and correlated with pCR to LCRT in an independent dataset merging two RC cohorts (ARISTOTLE, n = 114 and SALZBURG, n = 55 patients). DL models predicted CMS with high classification performance in multiple comparative analyses. In the independent cohorts (ARISTOTLE, SALZBURG), cases with WSIs classified as imCMS1 had a significantly higher likelihood of achieving pCR (OR = 2.69, 95% CI 1.01-7.17, p = 0.048). Conversely, imCMS4 was associated with lack of pCR (OR = 0.25, 95% CI 0.07-0.88, p = 0.031). Classification maps demonstrated pathologist-interpretable associations with high stromal content in imCMS4 cases, associated with poor outcome. No significant association was found in imCMS2 or imCMS3. imCMS classification of pre-treatment biopsies is a fast and inexpensive solution to identify patient groups that could benefit from neoadjuvant LCRT. The significant associations between imCMS1/imCMS4 with pCR suggest the existence of predictive morphological features that could enhance standard pathological assessment.

SST-editing: in silico spatial transcriptomic editing at single-cell resolution

MOTIVATION

Generative Adversarial Nets (GAN) achieve impressive performance for text-guided editing of natural images. However, a comparable utility of GAN remains understudied for spatial transcriptomics (ST) technologies with matched gene expression and biomedical image data.

RESULTS

We propose In Silico Spatial Transcriptomic editing that enables gene expression-guided editing of immunofluorescence images. Using cell-level spatial transcriptomics data extracted from normal and tumor tissue slides, we train the approach under the framework of GAN (Inversion). To simulate cellular state transitions, we then feed edited gene expression levels to trained models. Compared to normal cellular images (ground truth), we successfully model the transition from tumor to normal tissue samples, as measured with quantifiable and interpretable cellular features.

AVAILABILITY AND IMPLEMENTATION

https://github.com/CTPLab/SST-editing.

Pathway level subtyping identifies a slow-cycling biological phenotype associated with poor clinical outcomes in colorectal cancer

Molecular stratification using gene-level transcriptional data has identified subtypes with distinctive genotypic and phenotypic traits, as exemplified by the consensus molecular subtypes (CMS) in colorectal cancer (CRC). Here, rather than gene-level data, we make use of gene ontology and biological activation state information for initial molecular class discovery. In doing so, we defined three pathway-derived subtypes (PDS) in CRC: PDS1 tumors, which are canonical/LGR5+ stem-rich, highly proliferative and display good prognosis; PDS2 tumors, which are regenerative/ANXA1+ stem-rich, with elevated stromal and immune tumor microenvironmental lineages; and PDS3 tumors, which represent a previously overlooked slow-cycling subset of tumors within CMS2 with reduced stem populations and increased differentiated lineages, particularly enterocytes and enteroendocrine cells, yet display the worst prognosis in locally advanced disease. These PDS3 phenotypic traits are evident across numerous bulk and single-cell datasets, and demark a series of subtle biological states that are currently under-represented in pre-clinical models and are not identified using existing subtyping classifiers.

Poor Diagnostic Reproducibility in the Identification of Nonconventional Dysplasia in Colitis Impacts the Application of Histologic Stratification Tools

Due to their increased cancer risk, patients with longstanding inflammatory bowel disease are offered endoscopic surveillance with concomitant histopathologic assessments, aimed at identifying dysplasia as a precursor lesion of colitis-associated colorectal cancer. However, this strategy is beset with difficulties and limitations. Recently, a novel classification criterion for colitis-associated low-grade dysplasia has been proposed, and an association between nonconventional dysplasia and progression was reported, suggesting the possibility of histology-based stratification of patients with colitis-associated lesions. Here, a cohort of colitis-associated lesions was assessed by a panel of 6 experienced pathologists to test the applicability of the published classification criteria and try and validate the association between nonconventional dysplasia and progression. While confirming the presence of different morphologic patterns of colitis-associated dysplasia, the study demonstrated difficulties concerning diagnostic reproducibility between pathologists and was unable to validate the association of nonconventional dysplasia with cancer progression. Our study highlights the overall difficulty of using histologic assessment of precursor lesions for cancer risk prediction in inflammatory bowel disease patients and suggests the need for a different diagnostic strategy that can objectively identify high-risk phenotypes.

Multiplex analysis of intratumoural immune infiltrate and prognosis in patients with stage II-III colorectal cancer from the SCOT and QUASAR 2 trials: a retrospective analysis

BACKGROUND

Tumour-infiltrating CD8+ cytotoxic T cells confer favourable prognosis in colorectal cancer. The added prognostic value of other infiltrating immune cells is unclear and so we sought to investigate their prognostic value in two large clinical trial cohorts.

METHODS

We used multiplex immunofluorescent staining of tissue microarrays to assess the densities of CD8+, CD20+, FoxP3+, and CD68+ cells in the intraepithelial and intrastromal compartments from tumour samples of patients with stage II-III colorectal cancer from the SCOT trial (ISRCTN59757862), which examined 3 months versus 6 months of adjuvant oxaliplatin-based chemotherapy, and from the QUASAR 2 trial (ISRCTN45133151), which compared adjuvant capecitabine with or without bevacizumab. Both trials included patients aged 18 years or older with an Eastern Cooperative Oncology Group performance status of 0-1. Immune marker predictors were analysed by multiple regression, and the prognostic and predictive values of markers for colorectal cancer recurrence-free interval by Cox regression were assessed using the SCOT cohort for discovery and QUASAR 2 cohort for validation.

FINDINGS

After exclusion of cases without tissue microarrays and with technical failures, and following quality control, we included 2340 cases from the SCOT trial and 1069 from the QUASAR 2 trial in our analysis. Univariable analysis of associations with recurrence-free interval in cases from the SCOT trial showed a strong prognostic value of intraepithelial CD8 (CD8IE) as a continuous variable (hazard ratio [HR] for 75th vs 25th percentile [75vs25] 0·73 [95% CI 0·68-0·79], p=2·5 × 10-16), and of intrastromal FoxP3 (FoxP3IS; 0·71 [0·64-0·78], p=1·5 × 10-13) but not as strongly in the epithelium (FoxP3IE; 0·89 [0·84-0·96], p=1·5 × 10-4). Associations of other markers with recurrence-free interval were moderate. CD8IE and FoxP3IS retained independent prognostic value in bivariable and multivariable analysis, and, compared with either marker alone, a composite marker including both markers (CD8IE-FoxP3IS) was superior when assessed as a continuous variable (adjusted [a]HR75 vs 25 0·70 [95% CI 0·63-0·78], p=5·1 × 10-11) and when categorised into low, intermediate, and high density groups using previously published cutpoints (aHR for intermediate vs high 1·68 [95% CI 1·29-2·20], p=1·3 × 10-4; low vs high 2·58 [1·91-3·49], p=7·9 × 10-10), with performance similar to the gold-standard Immunoscore. The prognostic value of CD8IE-FoxP3IS was confirmed in cases from the QUASAR 2 trial, both as a continuous variable (aHR75 vs 25 0·84 [95% CI 0·73-0·96], p=0·012) and as a categorical variable for low versus high density (aHR 1·80 [95% CI 1·17-2·75], p=0·0071) but not for intermediate versus high (1·30 [0·89-1·88], p=0·17).

INTERPRETATION

Combined evaluation of CD8IE and FoxP3IS could help to refine risk stratification in colorectal cancer. Investigation of FoxP3IS cells as an immunotherapy target in colorectal cancer might be merited.

FUNDING

Medical Research Council, National Institute for Health Research, Cancer Research UK, Swedish Cancer Society, Roche, and Promedica Foundation.

Swiss digital pathology recommendations: results from a Delphi process conducted by the Swiss Digital Pathology Consortium of the Swiss Society of Pathology

Integration of digital pathology (DP) into clinical diagnostic workflows is increasingly receiving attention as new hardware and software become available. To facilitate the adoption of DP, the Swiss Digital Pathology Consortium (SDiPath) organized a Delphi process to produce a series of recommendations for DP integration within Swiss clinical environments. This process saw the creation of 4 working groups, focusing on the various components of a DP system (1) scanners, quality assurance and validation of scans, (2) integration of Whole Slide Image (WSI)-scanners and DP systems into the Pathology Laboratory Information System, (3) digital workflow-compliance with general quality guidelines, and (4) image analysis (IA)/artificial intelligence (AI), with topic experts for each recruited for discussion and statement generation. The work product of the Delphi process is 83 consensus statements presented here, forming the basis for "SDiPath Recommendations for Digital Pathology". They represent an up-to-date resource for national and international hospitals, researchers, device manufacturers, algorithm developers, and all supporting fields, with the intent of providing expectations and best practices to help ensure safe and efficient DP usage.

Digital image analysis and artificial intelligence in pathology diagnostics-the Swiss view

Digital pathology (DP) is increasingly entering routine clinical pathology diagnostics. As digitization of the routine caseload advances, implementation of digital image analysis algorithms and artificial intelligence tools becomes not only attainable, but also desirable in daily sign out. The Swiss Digital Pathology Consortium (SDiPath) has initiated a Delphi process to generate best-practice recommendations for various phases of the process of digitization in pathology for the local Swiss environment, encompassing the following four topics: i) scanners, quality assurance, and validation of scans; ii) integration of scanners and systems into the pathology laboratory information system; iii) the digital workflow; and iv) digital image analysis (DIA)/artificial intelligence (AI). The current article focuses on the DIA-/AI-related recommendations generated and agreed upon by the working group and further verified by the Delphi process among the members of SDiPath. Importantly, they include the view and the currently perceived needs of practicing pathologists from multiple academic and cantonal hospitals as well as private practices.

National digital pathology projects in Switzerland: A 2023 update

The Swiss Digital Pathology Consortium (SDiPath) was founded in 2018 as a working group of the Swiss Society for Pathology with the aim of networking, training, and promoting digital pathology (DP) at a national level. Since then, two national surveys have been carried out on the level of knowledge, dissemination, use, and needs in DP, which have resulted in clear fields of action. In addition to organizing symposia and workshops, national guidelines were drawn up and an initiative for a national DP platform actively codesigned. With the growing use of digital image processing and artificial intelligence tools, continuous monitoring, evaluation, and exchange of experiences will be pursued, along with best practices.

Accounting for intensity variation in image analysis of large-scale multiplexed clinical trial datasets

Multiplex immunofluorescence (mIF) imaging can provide comprehensive quantitative and spatial information for multiple immune markers for tumour immunoprofiling. However, application at scale to clinical trial samples sourced from multiple institutions is challenging due to pre-analytical heterogeneity. This study reports an analytical approach to the largest multi-parameter immunoprofiling study of clinical trial samples to date. We analysed 12,592 tissue microarray (TMA) spots from 3,545 colorectal cancers sourced from more than 240 institutions in two clinical trials (QUASAR 2 and SCOT) stained for CD4, CD8, CD20, CD68, FoxP3, pan-cytokeratin, and DAPI by mIF. TMA slides were multi-spectrally imaged and analysed by cell-based and pixel-based marker analysis. We developed an adaptive thresholding method to account for inter- and intra-slide intensity variation in TMA analysis. Applying this method effectively ameliorated inter- and intra-slide intensity variation improving the image analysis results compared with methods using a single global threshold. Correlation of CD8 data derived by our mIF analysis approach with single-plex chromogenic immunohistochemistry CD8 data derived from subsequent sections indicates the validity of our method (Spearman's rank correlation coefficients ρ between 0.63 and 0.66, p ≪ 0.01) as compared with the current gold standard analysis approach. Evaluation of correlation between cell-based and pixel-based analysis results confirms equivalency (ρ > 0.8, p ≪ 0.01, except for CD20 in the epithelial region) of both analytical approaches. These data suggest that our adaptive thresholding approach can enable analysis of mIF-stained clinical trial TMA datasets by digital pathology at scale for precision immunoprofiling.

Bone marrow haematopoiesis in patients with COVID-19

AIMS

Severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) infection broadly affects organ homeostasis, including the haematopoietic system. Autopsy studies are a crucial tool for investigation of organ-specific pathologies. Here we perform an in-depth analysis of the impact of severe coronavirus disease 2019 (COVID-19) on bone marrow haematopoiesis in correlation with clinical and laboratory parameters.

METHODS AND RESULTS

Twenty-eight autopsy cases and five controls from two academic centres were included in the study. We performed a comprehensive analysis of bone marrow pathology and microenvironment features with clinical and laboratory parameters and assessed SARS-CoV-2 infection of the bone marrow by quantitative polymerase chain reaction (qPCR) analysis. In COVID-19 patients, bone marrow specimens showed a left-shifted myelopoiesis (19 of 28, 64%), increased myeloid-erythroid ratio (eight of 28, 28%), increased megakaryopoiesis (six of 28, 21%) and lymphocytosis (four of 28, 14%). Strikingly, a high proportion of COVID-19 specimens showed erythrophagocytosis (15 of 28, 54%) and the presence of siderophages (11 of 15, 73%) compared to control cases (none of five, 0%). Clinically, erythrophagocytosis correlated with lower haemoglobin levels and was more frequently observed in patients from the second wave. Analysis of the immune environment showed a strong increase in CD68+ macrophages (16 of 28, 57%) and a borderline lymphocytosis (five of 28, 18%). The stromal microenvironment showed oedema (two of 28, 7%) and severe capillary congestion (one of 28, 4%) in isolated cases. No stromal fibrosis or microvascular thrombosis was found. While all cases had confirmed positive testing of SARS-CoV-2 in the respiratory system, SARS-CoV-2 was not detected in the bone marrow by high-sensitivity PCR, suggesting that SARS-CoV-2 does not commonly replicate in the haematopoietic microenvironment.

CONCLUSIONS

SARS-CoV-2 infection indirectly impacts the haematological compartment and the bone marrow immune environment. Erythrophagocytosis is frequent and associated with lower haemoglobin levels in patients with severe COVID-19.

Transformer-based biomarker prediction from colorectal cancer histology: A large-scale multicentric study

Deep learning (DL) can accelerate the prediction of prognostic biomarkers from routine pathology slides in colorectal cancer (CRC). However, current approaches rely on convolutional neural networks (CNNs) and have mostly been validated on small patient cohorts. Here, we develop a new transformer-based pipeline for end-to-end biomarker prediction from pathology slides by combining a pre-trained transformer encoder with a transformer network for patch aggregation. Our transformer-based approach substantially improves the performance, generalizability, data efficiency, and interpretability as compared with current state-of-the-art algorithms. After training and evaluating on a large multicenter cohort of over 13,000 patients from 16 colorectal cancer cohorts, we achieve a sensitivity of 0.99 with a negative predictive value of over 0.99 for prediction of microsatellite instability (MSI) on surgical resection specimens. We demonstrate that resection specimen-only training reaches clinical-grade performance on endoscopic biopsy tissue, solving a long-standing diagnostic problem.

Cystatin C is glucocorticoid responsive, directs recruitment of Trem2+ macrophages, and predicts failure of cancer immunotherapy

Cystatin C (CyC), a secreted cysteine protease inhibitor, has unclear biological functions. Many patients exhibit elevated plasma CyC levels, particularly during glucocorticoid (GC) treatment. This study links GCs with CyC's systemic regulation by utilizing genome-wide association and structural equation modeling to determine CyC production genetics in the UK Biobank. Both CyC production and a polygenic score (PGS) capturing predisposition to CyC production were associated with increased all-cause and cancer-specific mortality. We found that the GC receptor directly targets CyC, leading to GC-responsive CyC secretion in macrophages and cancer cells. CyC-knockout tumors displayed significantly reduced growth and diminished recruitment of TREM2+ macrophages, which have been connected to cancer immunotherapy failure. Furthermore, the CyC-production PGS predicted checkpoint immunotherapy failure in 685 patients with metastatic cancer from combined clinical trial cohorts. In conclusion, CyC may act as a GC effector pathway via TREM2+ macrophage recruitment and may be a potential target for combination cancer immunotherapy.

Combinational expression of tumor testis antigens NY-ESO-1, MAGE-A3, and MAGE-A4 predicts response to immunotherapy in mucosal melanoma patients

PURPOSE

Immunotherapy using immune checkpoint inhibitors (ICI) has revolutionized cancer treatment in recent years, particularly in melanoma. While response to immunotherapy is associated with high tumor mutational burden (TMB), PD-L1 expression, and microsatellite instability in several cancers, tumors lacking these biomarkers can still respond to this treatment. Especially, mucosal melanoma, commonly exhibiting low TMB compared to cutaneous melanoma, may respond to immunotherapy with immune checkpoint inhibitors. Therefore, the aim of our study was to investigate novel biomarkers in mucosal melanoma that predict response to combined ipilimumab and nivolumab.

METHODS

We investigated 10 tumor samples from 10 patients (three responders, seven non-responders) before treatment and six tumor samples from five patients after progression using a targeted Next Generation Sequencing (NGS) gene expression panel. The findings were corroborated with an independent method (i.e., immunohistochemical staining) on the same 10 tumor samples before treatment and, to increase the cohort, in addition on three tumor samples before treatment of more recent patients (one responder, two non-responders).

RESULTS

With the targeted gene expression panel, we found the three tumor testis antigens CTAG1B (NY-ESO-1), MAGE-A3, and MAGE-A4 to be predominantly expressed in responding tumors. This marker panel was either not or not completely expressed in non-responders (p < 0.01). Using immunohistochemistry for all three markers, we could confirm the elevated expression in tumors responding to the ipilimumab/nivolumab combination therapy.

CONCLUSION

In conclusion, these three biomarkers await validation in a larger patient cohort and could be easily used in future routine diagnostics to predict the outcome of ipilimumab/nivolumab combination therapy in mucosal melanoma patients.

Nicotinamide N-methyltransferase sustains a core epigenetic program that promotes metastatic colonization in breast cancer

Metastatic colonization of distant organs accounts for over 90% of deaths related to solid cancers, yet the molecular determinants of metastasis remain poorly understood. Here, we unveil a mechanism of colonization in the aggressive basal-like subtype of breast cancer that is driven by the NAD⁺ metabolic enzyme nicotinamide N-methyltransferase (NNMT). We demonstrate that NNMT imprints a basal genetic program into cancer cells, enhancing their plasticity. In line, NNMT expression is associated with poor clinical outcomes in patients with breast cancer. Accordingly, ablation of NNMT dramatically suppresses metastasis formation in pre-clinical mouse models. Mechanistically, NNMT depletion results in a methyl overflow that increases histone H3K9 trimethylation (H3K9me3) and DNA methylation at the promoters of PR/SET Domain-5 (PRDM5) and extracellular matrix-related genes. PRDM5 emerged in this study as a pro-metastatic gene acting via induction of cancer-cell intrinsic transcription of collagens. Depletion of PRDM5 in tumor cells decreases COL1A1 deposition and impairs metastatic colonization of the lungs. These findings reveal a critical activity of the NNMT-PRDM5-COL1A1 axis for cancer cell plasticity and metastasis in basal-like breast cancer.

Abstract 5695: Deep learning risk prediction model of distant recurrence from H&E endometrial cancer slides

Accurate risk prediction of distant recurrence (DR) is crucial for personalized adjuvant systemic therapy of endometrial cancer (EC) stage I-III patients, as DR is associated with a 5-year overall survival of 10-20%. Risk stratification and treatment recommendation are currently based on histopathological and molecular markers, which is challenging due to high inter-observer variability and testing costs respectively. Deep Learning (DL) models can predict prognosis by identifying relevant visual features from H&E whole slide images (WSIs) at different resolutions without prior assumptions. Here, we developed and tested the first interpretable state-of-the-art DL model for WSI-based risk prediction of DR of stage I-III EC (DeREC) from the randomized PORTEC-1/-2/-3 trials and three clinical cohorts with long-term follow-up data. We used one representative H&E WSI each from 1761 EC patients, excluding those who received adjuvant chemotherapy as it lowers the risk of DR. We randomly sampled 20% as a held-out internal test set (N=353 with 62 events; 8.45 year median follow-up) and performed a 5-fold cross-validation on the training set (N=1408). WSIs were partitioned into 360 micron patches at 40x magnification. DeREC combined self-supervised representation learning of patches using a multi-resolution vision transformer and a WSI-level graph attention-based time-to-event prediction model. The model performance of correctly ranking patients by predicted risk scores and true time to DR was measured with the concordance-index and compared with a Cox’ Proportional Hazards (CPH) model fitted on histopathological variables (histotype, grade, lymphovascular space invasion, stage). Discriminative quality of the predicted risk groups was investigated with Kaplan-Meier analysis and the log-rank test. Most predictive patches by predicted risk groups were reviewed by an expert gynecopathologist for identification of prognostic morphological features. DeREC achieved a concordance-index of 0.764 [95%CI 0.754-0.773] on 5-fold cross validation and 0.757 on the test set, as compared to 0.704 [95%CI 0.662-0.746] with CPH. Predicted risk groups around quartiles 1 and 3 accurately stratified patients between low (N=89), intermediate (N=175), high (N=89) risk of DR (p&lt;0.0001). Among the predicted low-risk group only 3 (3.37%) patients relapsed whereas intermediate and high-risk groups counted 27 (15.43%) and 32 (35.96%) events respectively. DeREC is the first DL model accurately distinguishing EC patients at high risk of DR from those at low risk using one H&E WSI, which would aid decisions on adjuvant treatment. DeREC outperformed standard statistical prediction methods using histopathological variables, indicating that it identified prognostic visual features which can be further investigated. Future development includes the integration of clinicopathological and molecular information.

Citation Format: Sarah Fremond, Sonali Andani, Jurriaan Barkey Wolf, Gitte Ørtoft, Estrid Høgdall, Jouke Dijkstra, Jan J. Jobsen, Ina M. Jürgenliemk-Schulz, Ludy CHW Lutgens, Melanie E. Powell, Naveena Singh, Linda R. Mileshkin, Helen J. Mackay, Alexandra Leary, Dionyssios Katsaros, Hans W. Nijman, Stephanie M. de Boer, Remi A. Nout, Vincent T.H.B.M Smit, Carien L. Creutzberg, Nanda Horeweg, Viktor H. Koelzer, Tjalling Bosse. Deep learning risk prediction model of distant recurrence from H&E endometrial cancer slides. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5695.

ROS Induction Targets Persister Cancer Cells with Low Metabolic Activity in NRAS-Mutated Melanoma

Clinical management of melanomas with NRAS mutations is challenging. Targeting MAPK signaling is only beneficial to a small subset of patients due to resistance that arises through genetic, transcriptional, and metabolic adaptation. Identification of targetable vulnerabilities in NRAS-mutated melanoma could help improve patient treatment. Here, we used multiomics analyses to reveal that NRAS-mutated melanoma cells adopt a mesenchymal phenotype with a quiescent metabolic program to resist cellular stress induced by MEK inhibition. The metabolic alterations elevated baseline reactive oxygen species (ROS) levels, leading these cells to become highly sensitive to ROS induction. In vivo xenograft experiments and single-cell RNA sequencing demonstrated that intratumor heterogeneity necessitates the combination of a ROS inducer and a MEK inhibitor to inhibit both tumor growth and metastasis. Ex vivo pharmacoscopy of 62 human metastatic melanomas confirmed that MEK inhibitor-resistant tumors significantly benefited from the combination therapy. Finally, oxidative stress response and translational suppression corresponded with ROS-inducer sensitivity in 486 cancer cell lines, independent of cancer type. These findings link transcriptional plasticity to a metabolic phenotype that can be inhibited by ROS inducers in melanoma and other cancers.

SIGNIFICANCE

Metabolic reprogramming in drug-resistant NRAS-mutated melanoma cells confers sensitivity to ROS induction, which suppresses tumor growth and metastasis in combination with MAPK pathway inhibitors.

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 POLE^mut, 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 POLE^mut 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 POLE^mut (n=51), 0·844 for MMRd (n=134), 0·883 for NSMP (n=120), and 0·928 for p53abn (n=88). POLE^mut and MMRd tiles had a high density of lymphocytes, p53abn tiles had strong nuclear atypia, and the morphology of POLE^mut 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 POLE^mut had an excellent prognosis, as do those with true POLE^mut 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.

Mitosis domain generalization in histopathology images - The MIDOG challenge

The density of mitotic figures (MF) within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. Recognition of MF by pathologists is subject to a strong inter-rater bias, limiting its prognostic value. State-of-the-art deep learning methods can support experts but have been observed to strongly deteriorate when applied in a different clinical environment. The variability caused by using different whole slide scanners has been identified as one decisive component in the underlying domain shift. The goal of the MICCAI MIDOG 2021 challenge was the creation of scanner-agnostic MF detection algorithms. The challenge used a training set of 200 cases, split across four scanning systems. As test set, an additional 100 cases split across four scanning systems, including two previously unseen scanners, were provided. In this paper, we evaluate and compare the approaches that were submitted to the challenge and identify methodological factors contributing to better performance. The winning algorithm yielded an F₁ score of 0.748 (CI95: 0.704-0.781), exceeding the performance of six experts on the same task.

PD-1T TILs as a Predictive Biomarker for Clinical Benefit to PD-1 Blockade in Patients with Advanced NSCLC

PURPOSE

Durable clinical benefit to PD-1 blockade in non-small cell lung cancer (NSCLC) is currently limited to a small fraction of patients, underlining the need for predictive biomarkers. We recently identified a tumor-reactive tumor-infiltrating T lymphocyte (TIL) pool, termed PD-1T TILs, with predictive potential in NSCLC. Here, we examined PD-1T TILs as biomarker in NSCLC.

EXPERIMENTAL DESIGN

PD-1T TILs were digitally quantified in 120 baseline samples from advanced NSCLC patients treated with PD-1 blockade. Primary outcome was disease control (DC) at 6 months. Secondary outcomes were DC at 12 months and survival. Exploratory analyses addressed the impact of lesion-specific responses, tissue sample properties, and combination with other biomarkers on the predictive value of PD-1T TILs.

RESULTS

PD-1T TILs as a biomarker reached 77% sensitivity and 67% specificity at 6 months, and 93% and 65% at 12 months, respectively. Particularly, a patient group without clinical benefit was reliably identified, indicated by a high negative predictive value (NPV) (88% at 6 months, 98% at 12 months). High PD-1T TILs related to significantly longer progression-free (HR 0.39, 95% CI, 0.24-0.63, P < 0.0001) and overall survival (HR 0.46, 95% CI, 0.28-0.76, P < 0.01). Predictive performance was increased when lesion-specific responses and samples obtained immediately before treatment were assessed. Notably, the predictive performance of PD-1T TILs was superior to PD-L1 and tertiary lymphoid structures in the same cohort.

CONCLUSIONS

This study established PD-1T TILs as predictive biomarker for clinical benefit to PD-1 blockade in patients with advanced NSCLC. Most importantly, the high NPV demonstrates an accurate identification of a patient group without benefit. See related commentary by Anagnostou and Luke, p. 4835.

Biological Misinterpretation of Transcriptional Signatures in Tumor Samples Can Unknowingly Undermine Mechanistic Understanding and Faithful Alignment with Preclinical Data

PURPOSE

Precise mechanism-based gene expression signatures (GES) have been developed in appropriate in vitro and in vivo model systems, to identify important cancer-related signaling processes. However, some GESs originally developed to represent specific disease processes, primarily with an epithelial cell focus, are being applied to heterogeneous tumor samples where the expression of the genes in the signature may no longer be epithelial-specific. Therefore, unknowingly, even small changes in tumor stroma percentage can directly influence GESs, undermining the intended mechanistic signaling.

EXPERIMENTAL DESIGN

Using colorectal cancer as an exemplar, we deployed numerous orthogonal profiling methodologies, including laser capture microdissection, flow cytometry, bulk and multiregional biopsy clinical samples, single-cell RNA sequencing and finally spatial transcriptomics, to perform a comprehensive assessment of the potential for the most widely used GESs to be influenced, or confounded, by stromal content in tumor tissue. To complement this work, we generated a freely-available resource, ConfoundR; https://confoundr.qub.ac.uk/, that enables users to test the extent of stromal influence on an unlimited number of the genes/signatures simultaneously across colorectal, breast, pancreatic, ovarian and prostate cancer datasets.

RESULTS

Findings presented here demonstrate the clear potential for misinterpretation of the meaning of GESs, due to widespread stromal influences, which in-turn can undermine faithful alignment between clinical samples and preclinical data/models, particularly cell lines and organoids, or tumor models not fully recapitulating the stromal and immune microenvironment.

CONCLUSIONS

Efforts to faithfully align preclinical models of disease using phenotypically-designed GESs must ensure that the signatures themselves remain representative of the same biology when applied to clinical samples.

Case report: Surgical repair of a large tracheo-esophageal fistula in a patient with post-transplant esophageal lymphoproliferative disorder

Introduction and importance

The management of large malignant tracheo-esophageal fistulas (TEF) is not standardized. Herein, we report a case with a malignant TEF associated with esophageal post-transplant lymphoproliferative disorder (PTLD) for whom we successfully performed a surgical repair. This contributes to the knowledge on how to treat large acquired malignant TEFs.

Case presentation

A 69-year old male presented with a one-week history of fever, productive cough and bilateral coarse crackles. In addition, he described a weight loss of 10 kg during the past three months. The patient's history included a kidney transplantation twenty years ago. Esophagogastroduodenoscopy with a biopsy of the esophagus was performed nine days before. Histopathology showed a PTLD of diffuse large B-cell lymphoma subtype. Subsequent diagnostics revealed a progressive TEF (approx. 2.0 × 1.5 cm) 3.0 cm above the carina. PET-CT scan showed an esophagus with slight tracer uptake in the middle third (approx. 11.5 cm length, SUV max 7.4). After decision against stenting, transthoracic subtotal esophagectomy with closure of the tracheal mouth of the fistula by a pedicled flap was performed. PTLD was treated with prednisone and rituximab. Tumor progression (brain metastasis) led to death 95 days after surgery.

Clinical discussion

The treatment of a malignant TEF is complex and personalized while both the consequences of the esophago-tracheal connection and those of the underlying responsible diagnosis have to be considered concurrently. In this case, we considered surgery as the best treatment option due to a relatively good prognosis of the underlying diagnosis (PTLD) and a large fistula. Esophageal or dual stenting, the treatment of choice for small malignant TEF, would have been associated with a high risk of failure due to the wide trachea, extensively dilated esophagus, proximal location and large diameter of the fistula.

Conclusion

Surgery can be considered for patients with a large acquired malignant TEF and positive long-term prognosis of the underlying diagnosis. Due to the complexity of TEF management, immediate pre-operative multidisciplinary discussion is advised.

•

Surgical repair of a large malignant tracheo-esophageal fistula

•

Immediate pre-operative multidisciplinary discussion is needed for large malignant trachea-esophageal fistulas.

•

Esophageal post-transplant lymphoproliferative disorder (PTLD) of diffuse large B-cell lymphoma (DLBCL) subtype

•

Esophagectomy and closure of the trachea by a pedicled deepithelized muscle flap

Dynamic and adaptive cancer stem cell population admixture in colorectal neoplasia

Intestinal homeostasis is underpinned by LGR5+ve crypt-base columnar stem cells (CBCs), but following injury, dedifferentiation results in the emergence of LGR5-ve regenerative stem cell populations (RSCs), characterized by fetal transcriptional profiles. Neoplasia hijacks regenerative signaling, so we assessed the distribution of CBCs and RSCs in mouse and human intestinal tumors. Using combined molecular-morphological analysis, we demonstrate variable expression of stem cell markers across a range of lesions. The degree of CBC-RSC admixture was associated with both epithelial mutation and microenvironmental signaling disruption and could be mapped across disease molecular subtypes. The CBC-RSC equilibrium was adaptive, with a dynamic response to acute selective pressure, and adaptability was associated with chemoresistance. We propose a fitness landscape model where individual tumors have equilibrated stem cell population distributions along a CBC-RSC phenotypic axis. Cellular plasticity is represented by position shift along this axis and is influenced by cell-intrinsic, extrinsic, and therapeutic selective pressures.

miR-579-3p Controls Hepatocellular Carcinoma Formation by Regulating the Phosphoinositide 3-Kinase-Protein Kinase B Pathway in Chronically Inflamed Liver

Chronic liver inflammation causes continuous liver damage with progressive liver fibrosis and cirrhosis, which may eventually lead to hepatocellular carcinoma (HCC). Whereas the 10-year incidence for HCC in patients with cirrhosis is approximately 20%, many of these patients remain tumor free for their entire lives. Clarifying the mechanisms that define the various outcomes of chronic liver inflammation is a key aspect in HCC research. In addition to a wide variety of contributing factors, microRNAs (miRNAs) have also been shown to be engaged in promoting liver cancer. Therefore, we wanted to characterize miRNAs that are involved in the development of HCC, and we designed a longitudinal study with formalin-fixed and paraffin-embedded liver biopsy samples from several pathology institutes from Switzerland. We examined the miRNA expression by nCounterNanostring technology in matched nontumoral liver tissue from patients developing HCC (n = 23) before and after HCC formation in the same patient. Patients with cirrhosis (n = 26) remaining tumor free within a similar time frame served as a control cohort. Comparison of the two cohorts revealed that liver tissue from patients developing HCC displayed a down-regulation of miR-579-3p as an early step in HCC development, which was further confirmed in a validation cohort. Correlation with messenger RNA expression profiles further revealed that miR-579-3p directly attenuated phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) expression and consequently protein kinase B (AKT) and phosphorylated AKT. In vitro experiments and the use of clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 technology confirmed that miR-579-3p controlled cell proliferation and cell migration of liver cancer cell lines. Conclusion: Liver tissues from patients developing HCC revealed changes in miRNA expression. miR-579-3p was identified as a novel tumor suppressor regulating phosphoinositide 3-kinase-AKT signaling at the early stages of HCC development.

Germline MBD4 deficiency causes a multi-tumor predisposition syndrome

We report an autosomal recessive, multi-organ tumor predisposition syndrome, caused by bi-allelic loss-of-function germline variants in the base excision repair (BER) gene MBD4. We identified five individuals with bi-allelic MBD4 variants within four families and these individuals had a personal and/or family history of adenomatous colorectal polyposis, acute myeloid leukemia, and uveal melanoma. MBD4 encodes a glycosylase involved in repair of G:T mismatches resulting from deamination of 5'-methylcytosine. The colorectal adenomas from MBD4-deficient individuals showed a mutator phenotype attributable to mutational signature SBS1, consistent with the function of MBD4. MBD4-deficient polyps harbored somatic mutations in similar driver genes to sporadic colorectal tumors, although AMER1 mutations were more common and KRAS mutations less frequent. Our findings expand the role of BER deficiencies in tumor predisposition. Inclusion of MBD4 in genetic testing for polyposis and multi-tumor phenotypes is warranted to improve disease management.

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.

Integrated Analysis Of Immunotherapy Treated Clear Cell Renal Cell Carcinomas: An Exploratory Study

Molecular or immunological differences between responders and nonresponders to immune checkpoint inhibitors (ICIs) of clear cell renal cell carcinomas (ccRCCs) remain incompletely understood. To address this question, we performed next-generation sequencing, methylation analysis, genome wide copy number analysis, targeted RNA sequencing and T-cell receptor sequencing, and we studied frequencies of tumor-infiltrating CD8+ T cells, presence of tertiary lymphoid structures (TLS) and PD-L1 expression in 8 treatment-naive ccRCC patients subsequently treated with ICI (3 responders, 5 nonresponders). Unexpectedly, we identified decreased frequencies of CD8+ tumor-infiltrating T cells and TLS, and a decreased expression of PD-L1 in ICI responders when compared with nonresponders. However, neither tumor-specific genetic alterations nor gene expression profiles correlated with response to ICI or the observed immune features. Our results underline the challenge to stratify ccRCC patients for immunotherapy based on routinely available pathologic primary tumor material, even with advanced technologies. Our findings emphasize the analysis of pretreated metastatic tissue in line with recent observations describing treatment effects on the tumor microenvironment. In addition, our data call for further investigation of additional parameters in a larger ccRCC cohort to understand the mechanistic implications of the observed differences in tumor-infiltrating CD8+ T cells, TLS, and PD-L1 expression.

Bone Morphogenetic Protein Pathway Antagonism by Grem1 Regulates Epithelial Cell Fate in Intestinal Regeneration

BACKGROUND & AIMS

In homeostasis, intestinal cell fate is controlled by balanced gradients of morphogen signaling. The bone morphogenetic protein (BMP) pathway has a physiological, prodifferentiation role, predominantly inferred through previous experimental pathway inactivation. Intestinal regeneration is underpinned by dedifferentiation and cell plasticity, but the signaling pathways that regulate this adaptive reprogramming are not well understood. We assessed the BMP signaling landscape and investigated the impact and therapeutic potential of pathway manipulation in homeostasis and regeneration.

METHODS

A novel mouse model was generated to assess the effect of the autocrine Bmp4 ligand on individual secretory cell fate. We spatiotemporally mapped BMP signaling in mouse and human regenerating intestine. Transgenic models were used to explore the functional impact of pathway manipulation on stem cell fate and intestinal regeneration.

RESULTS

In homeostasis, ligand exposure reduced proliferation, expedited terminal differentiation, abrogated secretory cell survival, and prevented dedifferentiation. After ulceration, physiological attenuation of BMP signaling arose through upregulation of the secreted antagonist Grem1 from topographically distinct populations of fibroblasts. Concomitant expression supported functional compensation after Grem1 deletion from tissue-resident cells. BMP pathway manipulation showed that antagonist-mediated BMP attenuation was obligatory but functionally submaximal, because regeneration was impaired or enhanced by epithelial overexpression of Bmp4 or Grem1, respectively. Mechanistically, Bmp4 abrogated regenerative stem cell reprogramming despite a convergent impact of YAP/TAZ on cell fate in remodeled wounds.

CONCLUSIONS

BMP signaling prevents epithelial dedifferentiation, and pathway attenuation through stromal Grem1 upregulation was required for adaptive reprogramming in intestinal regeneration. This intercompartmental antagonism was functionally submaximal, raising the possibility of therapeutic pathway manipulation in inflammatory bowel disease.

Stromal composition predicts recurrence of early rectal cancer after local excision

AIMS

After local excision of early rectal cancer, definitive lymph node status is not available. An alternative means for accurate assessment of recurrence risk is required to determine the most appropriate subsequent management. Currently used measures are suboptimal. We assess three measures of tumour stromal content to determine their predictive value after local excision in a well-characterised cohort of rectal cancer patients without prior radiotherapy.

METHODS AND RESULTS

A total of 143 patients were included. Haematoxylin and eosin (H&E) sections were scanned for (i) deep neural network (DNN, a machine-learning algorithm) tumour segmentation into compartments including desmoplastic stroma and inflamed stroma; and (ii) digital assessment of tumour stromal fraction (TSR) and optical DNA ploidy analysis. 3' mRNA sequencing was performed to obtain gene expression data from which stromal and immune scores were calculated using the ESTIMATE method. Full results were available for 139 samples and compared with disease-free survival. All three methods were prognostic. Most strongly predictive was a DNN-determined ratio of desmoplastic to inflamed stroma >5.41 (P < 0.0001). A ratio of ESTIMATE stromal to immune score <1.19 was also predictive of disease-free survival (P = 0.00051), as was stromal fraction >36.5% (P = 0.037).

CONCLUSIONS

The DNN-determined ratio of desmoplastic to inflamed ratio is a novel and powerful predictor of disease recurrence in locally excised early rectal cancer. It can be assessed on a single H&E section, so could be applied in routine clinical practice to improve the prognostic information available to patients and clinicians to inform the decision concerning further management.

The Tumor Profiler Study: integrated, multi-omic, functional tumor profiling for clinical decision support

The application and integration of molecular profiling technologies create novel opportunities for personalized medicine. Here, we introduce the Tumor Profiler Study, an observational trial combining a prospective diagnostic approach to assess the relevance of in-depth tumor profiling to support clinical decision-making with an exploratory approach to improve the biological understanding of the disease.

Image-based consensus molecular subtype (imCMS) classification of colorectal cancer using deep learning

OBJECTIVE

Complex phenotypes captured on histological slides represent the biological processes at play in individual cancers, but the link to underlying molecular classification has not been clarified or systematised. In colorectal cancer (CRC), histological grading is a poor predictor of disease progression, and consensus molecular subtypes (CMSs) cannot be distinguished without gene expression profiling. We hypothesise that image analysis is a cost-effective tool to associate complex features of tissue organisation with molecular and outcome data and to resolve unclassifiable or heterogeneous cases. In this study, we present an image-based approach to predict CRC CMS from standard H&E sections using deep learning.

DESIGN

Training and evaluation of a neural network were performed using a total of n=1206 tissue sections with comprehensive multi-omic data from three independent datasets (training on FOCUS trial, n=278 patients; test on rectal cancer biopsies, GRAMPIAN cohort, n=144 patients; and The Cancer Genome Atlas (TCGA), n=430 patients). Ground truth CMS calls were ascertained by matching random forest and single sample predictions from CMS classifier.

RESULTS

Image-based CMS (imCMS) accurately classified slides in unseen datasets from TCGA (n=431 slides, AUC)=0.84) and rectal cancer biopsies (n=265 slides, AUC=0.85). imCMS spatially resolved intratumoural heterogeneity and provided secondary calls correlating with bioinformatic prediction from molecular data. imCMS classified samples previously unclassifiable by RNA expression profiling, reproduced the expected correlations with genomic and epigenetic alterations and showed similar prognostic associations as transcriptomic CMS.

CONCLUSION

This study shows that a prediction of RNA expression classifiers can be made from H&E images, opening the door to simple, cheap and reliable biological stratification within routine workflows.

In-depth Clinical and Biological Exploration of DNA Damage Immune Response as a Biomarker for Oxaliplatin Use in Colorectal Cancer

PURPOSE

The DNA damage immune response (DDIR) assay was developed in breast cancer based on biology associated with deficiencies in homologous recombination and Fanconi anemia pathways. A positive DDIR call identifies patients likely to respond to platinum-based chemotherapies in breast and esophageal cancers. In colorectal cancer, there is currently no biomarker to predict response to oxaliplatin. We tested the ability of the DDIR assay to predict response to oxaliplatin-based chemotherapy in colorectal cancer and characterized the biology in DDIR-positive colorectal cancer.

EXPERIMENTAL DESIGN

Samples and clinical data were assessed according to DDIR status from patients who received either 5-fluorouracil (5-FU) or 5FUFA (bolus and infusion 5-FU with folinic acid) plus oxaliplatin (FOLFOX) within the FOCUS trial (n = 361, stage IV), or neoadjuvant FOLFOX in the FOxTROT trial (n = 97, stage II/III). Whole transcriptome, mutation, and IHC data of these samples were used to interrogate the biology of DDIR in colorectal cancer.

RESULTS

Contrary to our hypothesis, DDIR-negative patients displayed a trend toward improved outcome for oxaliplatin-based chemotherapy compared with DDIR-positive patients. DDIR positivity was associated with microsatellite instability (MSI) and colorectal molecular subtype 1. Refinement of the DDIR signature, based on overlapping IFN-related chemokine signaling associated with DDIR positivity across colorectal cancer and breast cancer cohorts, further confirmed that the DDIR assay did not have predictive value for oxaliplatin-based chemotherapy in colorectal cancer.

CONCLUSIONS

DDIR positivity does not predict improved response following oxaliplatin treatment in colorectal cancer. However, data presented here suggest the potential of the DDIR assay in identifying immune-rich tumors that may benefit from immune checkpoint blockade, beyond current use of MSI status.

Prognostic Integrated Image-Based Immune and Molecular Profiling in Early-Stage Endometrial Cancer

Optimum risk stratification in early-stage endometrial cancer combines clinicopathologic factors and the molecular endometrial cancer classification defined by The Cancer Genome Atlas (TCGA). It is unclear whether analysis of intratumoral immune infiltrate improves this. We developed a machine-learning, image-based algorithm to quantify density of CD8+ and CD103+ immune cells in tumor epithelium and stroma in 695 stage I endometrioid endometrial cancers from the PORTEC-1 and -2 trials. The relationship between immune cell density and clinicopathologic/molecular factors was analyzed by hierarchical clustering and multiple regression. The prognostic value of immune infiltrate by cell type and location was analyzed by univariable and multivariable Cox regression, incorporating the molecular endometrial cancer classification. Tumor-infiltrating immune cell density varied substantially between cases, and more modestly by immune cell type and location. Clustering revealed three groups with high, intermediate, and low densities, with highly significant variation in the proportion of molecular endometrial cancer subgroups between them. Univariable analysis revealed intraepithelial CD8+ cell density as the strongest predictor of endometrial cancer recurrence; multivariable analysis confirmed this was independent of pathologic factors and molecular subgroup. Exploratory analysis suggested this association was not uniform across molecular subgroups, but greatest in tumors with mutant p53 and absent in DNA mismatch repair-deficient cancers. Thus, this work identified that quantification of intraepithelial CD8+ cells improved upon the prognostic utility of the molecular endometrial cancer classification in early-stage endometrial cancer.

Two distinct immunopathological profiles in autopsy lungs of COVID-19

Coronavirus Disease 19 (COVID-19) is a respiratory disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which has grown to a worldwide pandemic with substantial mortality. Immune mediated damage has been proposed as a pathogenic factor, but immune responses in lungs of COVID-19 patients remain poorly characterized. Here we show transcriptomic, histologic and cellular profiles of post mortem COVID-19 (n = 34 tissues from 16 patients) and normal lung tissues (n = 9 tissues from 6 patients). Two distinct immunopathological reaction patterns of lethal COVID-19 are identified. One pattern shows high local expression of interferon stimulated genes (ISGhigh) and cytokines, high viral loads and limited pulmonary damage, the other pattern shows severely damaged lungs, low ISGs (ISGlow), low viral loads and abundant infiltrating activated CD8+ T cells and macrophages. ISGhigh patients die significantly earlier after hospitalization than ISGlow patients. Our study may point to distinct stages of progression of COVID-19 lung disease and highlights the need for peripheral blood biomarkers that inform about patient lung status and guide treatment.

Exploiting differential Wnt target gene expression to generate a molecular biomarker for colorectal cancer stratification

OBJECTIVE

Pathological Wnt pathway activation is a conserved hallmark of colorectal cancer. Wnt-activating mutations can be divided into: i) ligand-independent (LI) alterations in intracellular signal transduction proteins (Adenomatous polyposis coli, β-catenin), causing constitutive pathway activation and ii) ligand-dependent (LD) mutations affecting the synergistic R-Spondin axis (RNF43, RSPO-fusions) acting through amplification of endogenous Wnt signal transmembrane transduction. Our aim was to exploit differential Wnt target gene expression to generate a mutation-agnostic biomarker for LD tumours.

DESIGN

We undertook harmonised multi-omic analysis of discovery (n=684) and validation cohorts (n=578) of colorectal tumours collated from publicly available data and the Stratification in Colorectal Cancer Consortium. We used mutation data to establish molecular ground truth and subdivide lesions into LI/LD tumour subsets. We contrasted transcriptional, methylation, morphological and clinical characteristics between groups.

RESULTS

Wnt disrupting mutations were mutually exclusive. Desmoplastic stromal upregulation of RSPO may compensate for absence of epithelial mutation in a subset of stromal-rich tumours. Key Wnt negative regulator genes were differentially expressed between LD/LI tumours, with targeted hypermethylation of some genes (AXIN2, NKD1) occurring even in CIMP-negative LD cancers. AXIN2 mRNA expression was used as a discriminatory molecular biomarker to distinguish LD/LI tumours (area under the curve >0.93).

CONCLUSIONS

Epigenetic suppression of appropriate Wnt negative feedback loops is selectively advantageous in LD tumours and differential AXIN2 expression in LD/LI lesions can be exploited as a molecular biomarker. Distinguishing between LD/LI tumour types is important; patients with LD tumours retain sensitivity to Wnt ligand inhibition and may be stratified at diagnosis to clinical trials of Porcupine inhibitors.

Abstract 4446: Assessment of tissue composition with digital pathology in colorectal cancer

Background: The tumor microenvironment is a key feature to understand cancer biology and may be used clinically. Quantification of tissue composition is usually based either on visual pathological review (VPR) or deconvolution of whole genome molecular data. Although the former is a direct measurement it has modest reproducibility while the latter is an indirect measurement of unclear accuracy, expensive and not always available. Here we test digital pathology coupled with machine learning as a new tool to assess tissue composition.

Methods: As part of the Stratification in COloRecTal cancer (S:CORT) programme, a set of over 500 colorectal cancer (CRC) archival paraffin blocks from resections and biopsies were sequentially sectioned for Hematoxylin and Eosin staining (H&E), RNA extraction, a second H&E and DNA extraction. RNA expression microarrays, targeted DNA sequencing and DNA methylation arrays were applied. Tissue composition from the H&Es was obtained by VPR of expert pathologists and by a deep neural net (DNN) algorithm after supervised training on &gt;1,500 tissue areas from S:CORT, TCGA, TEM and CORGI CRC cohorts. Tumor purity estimates were obtained from RNA and methylation arrays.

Results: DNN estimates including area and cell counts were obtained for tumor, desmoplastic stroma, inflamed stroma, mucin/hypocellular stroma, muscle, necrosis and white space. An average of 6.8x105 total cells (range: 1.2x104-2.8x106) and 1.2x105 (range: 7.2x104-1.8x106) were classified for resections and biopsies respectively. Analyses performed twice on the same H&Es obtained matching results (r=1.0). Comparison of the paired first and second H&E showed very high correlations (r~0.9) and total cell counts correlated with DNA and RNA extraction yields (r~0.6). Tumor purity estimates by VPR mildly correlated with DNN (r~0.5) but they were underestimated and very variable. As a result, copy number adjusted by VPR purity tended to be overestimated compared to adjustment with DNN estimates. The improved performance of DNN is reflected in an accurate capture of non-linear association between area and cell counts in invasive cancer. In contrast, tumor purity estimates derived from RNA or DNA methylation arrays showed better correlations compared with DNN (r~0.6) but both overestimated purity in cases with low cell counts by up to a three-fold difference.

Conclusions: Tissue composition analysis with DNN allows analytical robustness, automatization and standardization and provides very high reproducibility at single cell resolution. DNN-based estimation of tumor purity is more accurate than VPR or extrapolation from molecular data derived from genome-wide omic platforms which tend to under and overestimate tumor purity respectively. DNN could be used to better plan and asses downstream molecular analyses and to investigate tissue-based metrics as potential clinical biomarkers in clinical trials.

Citation Format: Enric Domingo, Aikaterini Chatzipli, Susan Richman, Andrew Blake, Claire Hardy, Celina Whalley, Keara Redmon, Ian Tomlinson, Philip Dunne, Steven Walker, Andrew Beggs, Ultan McDermott, Graeme I. Murray, Leslie M. Samuel, Matthew Seymour, Philip Quirke, Tim Maughan, Viktor H. Koelzer. Assessment of tissue composition with digital pathology in colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4446.

Tracing Clonal Dynamics Reveals that Two- and Three-dimensional Patient-derived Cell Models Capture Tumor Heterogeneity of Clear Cell Renal Cell Carcinoma

BACKGROUND

Extensive DNA sequencing has led to an unprecedented view of the diversity of individual genomes and their evolution among patients with clear cell renal cell carcinoma (ccRCC).

OBJECTIVE

To understand subclonal architecture and dynamics of patient-derived two-dimensional (2D) and three-dimensional (3D) ccRCC models in vitro, in order to determine whether they mirror ccRCC inter- and intratumor heterogeneity.

DESIGN, SETTING, AND PARTICIPANTS

We have established a comprehensive platform of living renal cancer cell models from ccRCC surgical specimens.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

We confirmed the concordance of 2D and 3D patient-derived cell (PDC) models with the original tumor tissue in terms of histology, biomarker expression, cancer driver mutations, and copy number alterations. We addressed inter- and intrapatient heterogeneity by analyzing clonal dynamics during serial passaging.

RESULTS AND LIMITATIONS

In-depth genetic characterization verified the presence of heterogeneous cell populations, and revealed a high degree of similarity between subclonal compositions of monolayer and organoid cell cultures and the corresponding parental ccRCCs. Clonal dynamics were evident during serial passaging of cells in vitro, suggesting that PDC cultures can offer insights into evolutionary potential and treatment susceptibility of ccRCC subclones in vivo. Proof-of-concept drug profiling using selected ccRCC-targeted therapy agents highlighted patient-specific vulnerabilities in PDC models that could not be anticipated by interrogating commercially available cell lines.

CONCLUSIONS

We demonstrate that PDC models mirror inter- and intratumor heterogeneity of ccRCC in vitro. Based on our findings, we envision that the use of these models will advance our understanding of the trajectories that cause genetic diversity and their consequences for treatment on an individual level.

PATIENT SUMMARY

In this study, we developed two- and three-dimensional patient-derived models from clear cell renal cell carcinoma (ccRCC) as "mini-tumors in a dish." We show that these cell models retain important features of the human ccRCCs such as the profound tumor heterogeneity, thus highlighting their importance for cancer research and precision medicine.

[The pathology of adverse events with immune checkpoint inhibitors]

BACKGROUND

Immunotherapy has gained importance with the development of new effective cancer treatments. Immune checkpoint inhibitors (ICI) are monoclonal antibodies that promote T‑cell mediated tumor immune rejection. Checkpoint blockade also carries the risk of inducing autoimmune reactions ("immune related adverse events", irAEs). The diagnosis and classification of irAEs constitute a new and important field in pathology.

AIM

Practice-oriented review of the diagnosis and classification of irAEs.

MATERIALS AND METHODS

Structured, selective literature review based on PubMed und UpToDate ® online.

RESULTS

The most common irAEs affect the skin, the gastrointestinal tract, the liver, and the respiratory system. The correct diagnosis and classification of irAEs by an interdisciplinary care team is essential for appropriate therapy and the prevention of long-term sequelae. Other important irAEs affect the endocrine organs, the heart, the joints, the kidneys and the nervous system. Because of their rarity and/or limited options for bioptic diagnosis, only limited data on the morphology and pathophysiology of these irAEs are currently available. Autopsies carried out after ICI therapy constitute an important element of quality control and allow better documentation of the incidence and pathogenesis of irAEs.

DISCUSSION

Pathology plays a central role in the diagnosis and treatment of irAEs. Future studies may contribute to a better mechanistic understanding of irAEs for individualized knowledge-based risk assessment.

Abstract B050: Identification of PD-1T TILs and CXCL13 as determinants for response to anti-PD-1 treatment using human tumor explants

Reinvigoration of tumor-specific T-cells by cancer immunotherapies, in particular PD-1/PD-L1 blocking agents, has greatly improved clinical outcome in multiple cancer types. Nevertheless, durable clinical benefit is currently limited to a small number of patients. To achieve a better understanding of the immunologic determinants of response to anti-PD-1 therapy, we assessed transcriptional and functional profiles of tumor-infiltrating lymphocyte (TIL) subsets from non-small cell lung cancer specimens. Thereby, we identified a transcriptionally distinct CD8 TIL pool with enriched capacity for tumor recognition. This TIL pool, termed PD-1T TILs, is characterized by bright PD-1 expression and constitutive CXCL13 secretion, which can mediate immune cell recruitment to tertiary lymphoid structures. Notably, the presence of PD-1T TILs correlates with response and survival in a small cohort of lung cancer patients treated with PD-1 blockade. To assess the role of PD-1T TILs and CXCL13 for response to PD-1 blockade in other cancer types, we developed a platform using human tumor explants to visualize immunologic responses to anti-PD-1 on a patient-specific level. Analysis of the cellular and soluble tumor microenvironment composition as well as of treatment-induced changes in tumor-infiltrating immune cells revealed immunologic responses to anti-PD-1 in 5 different cancer types. Of note, responding tumors in different tumor types were characterized by a clear enrichment in both PD-1T TILs and CXCL13 production. Collectively, our data reveal a distinct state of PD-1 bright lymphocytes that are enriched for tumor-reactivity in human cancer, making them an attractive proxy for the antitumor potential of the intratumoral T-cell pool. Furthermore, we established technology using human tumor explants to measure the immunologic response to T-cell checkpoint inhibition on a personalized basis. Finally, with this approach we identified PD-1T TILs and CXCL13 as determinants for response to anti-PD-1 in multiple cancer types, opening potential new avenues for therapeutic intervention and improved patient selection.

Citation Format: Daniela S. Thommen, Viktor H. Koelzer, Petra Herzig, Marjolein de Bruijn, Paula Voabil, Marlous van den Braber, Karlijn Hummelink, Kim Monkhorst, Kirsten D. Mertz, Alfred Zippelius, John B.A.G. Haanen, Ton N.M. Schumacher. Identification of PD-1T TILs and CXCL13 as determinants for response to anti-PD-1 treatment using human tumor explants [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr B050.

The evolutionary landscape of colorectal tumorigenesis

The evolutionary events that cause colorectal adenomas (benign) to progress to carcinomas (malignant) remain largely undetermined. Using multi-region genome and exome sequencing of 24 benign and malignant colorectal tumours, we investigate the evolutionary fitness landscape occupied by these neoplasms. Unlike carcinomas, advanced adenomas frequently harbour sub-clonal driver mutations-considered to be functionally important in the carcinogenic process-that have not swept to fixation, and have relatively high genetic heterogeneity. Carcinomas are distinguished from adenomas by widespread aneusomies that are usually clonal and often accrue in a 'punctuated' fashion. We conclude that adenomas evolve across an undulating fitness landscape, whereas carcinomas occupy a sharper fitness peak, probably owing to stabilizing selection.

Stromal PD-1/PD-L1 Expression Predicts Outcome in Colon Cancer Patients

INTRODUCTION

The programmed cell death 1 (PD-1)/programmed cell death ligand 1 (PD-L1) axis plays an important role in controlling immune suppression by down-regulating T effector cell activities, enabling tumor cells to escape from the host's antitumor immunsurveillance. While only a small part of colon cancer cells express PD-L1, we sought to evaluate the differential impact of stromal and epithelial PD-L1 expression of primary tumors and liver metastasis on overall survival (OS) in colon cancer patients.

PATIENTS AND METHODS

Using a next-generation tissue microarray approach, we assessed both epithelial and stromal PD-L1 expression levels in primary tumors (n = 279) and corresponding liver metastases (n = 14) of colon cancer patients. PD-L1 positivity was graded according to the percentage (0.1%-1%, > 1%, > 5%, > 50%) of tumor cells with membranous PD-L1 expression or as the percentage of positive stroma cells and associated inflammatory infiltrates. We also assessed the interplay between stromal PD-1/PD-L1 and both intratumoral and stromal CD8 count and their impact on outcome. The primary end point was OS.

RESULTS

Stromal PD-L1 and PD-1 expression were both associated with less aggressive tumor behavior in colon cancer patients, which translated into better OS and disease-free survival, respectively. Conversely, PD-L1 staining in the tumor cells was less frequent than stromal staining and was associated with features of aggressive tumor biology, although without impact on outcome. Interestingly, the PD-L1 staining pattern remained similar between primary tumors and corresponding liver metastases. Stromal PD-1 expression correlated significantly with stromal PD-L1 staining and both intratumoral and stromal CD8 expression.

CONCLUSION

Stromal PD-1/PD-L1 expression might serve as a prognostic marker in colon cancer patients.

Digital image analysis improves precision of PD-L1 scoring in cutaneous melanoma

AIMS

Immune checkpoint inhibitors have become a successful treatment in metastatic melanoma. The high response rates in a subset of patients suggest that a sensitive companion diagnostic test is required. The predictive value of programmed death ligand 1 (PD-L1) staining in melanoma has been questioned due to inconsistent correlation with clinical outcome. Whether this is due to predictive irrelevance of PD-L1 expression or inaccurate assessment techniques remains unclear. The aim of this study was to develop a standardised digital protocol for the assessment of PD-L1 staining in melanoma and to compare the output data and reproducibility to conventional assessment by expert pathologists.

METHODS AND RESULTS

In two cohorts with a total of 69 cutaneous melanomas, a highly significant correlation was found between pathologist-based consensus reading and automated PD-L1 analysis (r = 0.97, P < 0.0001). Digital scoring captured the full diagnostic spectrum of PD-L1 expression at single cell resolution. An average of 150 472 melanoma cells (median 38 668 cells; range = 733-1 078 965) were scored per lesion. Machine learning was used to control for heterogeneity introduced by PD-L1-positive inflammatory cells in the tumour microenvironment. The PD-L1 image analysis protocol showed excellent reproducibility (r = 1.0, P < 0.0001) when carried out on independent workstations and reduced variability in PD-L1 scoring of human observers. When melanomas were grouped by PD-L1 expression status, we found a clear correlation of PD-L1 positivity with CD8-positive T cell infiltration, but not with tumour stage, metastasis or driver mutation status.

CONCLUSION

Digital evaluation of PD-L1 reduces scoring variability and may facilitate patient stratification in clinical practice.

Association of genetic variations within the PD-L2 immune checkpoint gene with outcome in stage II and III colon cancer

626

Background: Immune checkpoints can either inhibit or stimulate T-cell responses and therefore play a key role in maintaining an equilibrium between self tolerance and autoimmunity. Targeting immune checkpoint molecules can result in increased antitumor immunity by stimulation of the immune system. We hypothesize, that variations in genes encoding for both inhibitory or stimulatory immune checkpoint proteins may predict outcome in stage II and III colon cancer patients. Methods: The impact of 6 functional single nucleotide polymorphisms (SNPs) within the PD1, PD-L1, PD-L2, TIM3, OX40 and CD27 genes on time to recurrence was evaluated in 201 patients with stage II and III colon cancer. Genomic DNA was extracted from formalin-fixed paraffin embedded tissue and the SNPs were analyzed by PCR-based direct sequencing. Results: Baseline characteristics were as follows: female/male ratio = 42.8% / 57.2%; median age = 70y (19-91); median follow-up = 43months. The PD-L2 rs16923189 SNP showed significant association with recurrence rate. Patients with a G/G genotype had a higher 3-years recurrence rate compared to those harboring any A allele (41% vs 19%) in both univariate (HR 2.68, 95% Confidence interval (CI) 1.28-5.61, p = 0.006) and multivariate analyses (HR 3.13, 95% CI 1.42-6.28, p = 0.003). The high recurrence rate was most evident among patients with stage III and left-sided tumors carrying the G/G genotype (53% vs 24% and 65% vs 18%) in both univariate (HR 2.87, 95% CI 1.24-6.66, p = 0.009 and HR 5.28, 95% CI 2.24-12.44, p < 0.0001) and multivariate analyses (HR 4.32, 95% CI 1.76-9.91, p = 0.001 and HR 5.20, 95% CI 2.02-12.75, p = 0.001). Conclusions: Our results provide the first evidence, that polymorphisms within the PD-L2 gene might serve as prognostic markers in patients with stage II and III colon cancer. Interestingly, the prognostic effect is most significant among patients with stage III and left-sided colon cancers. Targeting PD-L2 might be a promising approach to further optimize treatment options and to improve outcome of colon cancer patients in the adjuvant setting.

Critical research gaps and recommendations to inform research prioritisation for more effective prevention and improved outcomes in colorectal cancer

OBJECTIVE

Colorectal cancer (CRC) leads to significant morbidity/mortality worldwide. Defining critical research gaps (RG), their prioritisation and resolution, could improve patient outcomes.

DESIGN

RG analysis was conducted by a multidisciplinary panel of patients, clinicians and researchers (n=71). Eight working groups (WG) were constituted: discovery science; risk; prevention; early diagnosis and screening; pathology; curative treatment; stage IV disease; and living with and beyond CRC. A series of discussions led to development of draft papers by each WG, which were evaluated by a 20-strong patient panel. A final list of RGs and research recommendations (RR) was endorsed by all participants.

RESULTS

Fifteen critical RGs are summarised below: RG1: Lack of realistic models that recapitulate tumour/tumour micro/macroenvironment; RG2: Insufficient evidence on precise contributions of genetic/environmental/lifestyle factors to CRC risk; RG3: Pressing need for prevention trials; RG4: Lack of integration of different prevention approaches; RG5: Lack of optimal strategies for CRC screening; RG6: Lack of effective triage systems for invasive investigations; RG7: Imprecise pathological assessment of CRC; RG8: Lack of qualified personnel in genomics, data sciences and digital pathology; RG9: Inadequate assessment/communication of risk, benefit and uncertainty of treatment choices; RG10: Need for novel technologies/interventions to improve curative outcomes; RG11: Lack of approaches that recognise molecular interplay between metastasising tumours and their microenvironment; RG12: Lack of reliable biomarkers to guide stage IV treatment; RG13: Need to increase understanding of health related quality of life (HRQOL) and promote residual symptom resolution; RG14: Lack of coordination of CRC research/funding; RG15: Lack of effective communication between relevant stakeholders.

CONCLUSION

Prioritising research activity and funding could have a significant impact on reducing CRC disease burden over the next 5 years.

The T cell repertoire in tumors overlaps with pulmonary inflammatory lesions in patients treated with checkpoint inhibitors

Cancer immunotherapy with antibodies targeting immune checkpoints such as the PD-1/PD-L1 pathway have emerged as breakthrough treatment for multiple solid tumors with high response rates and durable remissions. Despite the benefit for patients and encouraging safety profile, severe inflammatory reactions are observed in some patients. Such immune-related adverse events (irAEs) frequently lead to temporary or permanent cessation of the treatment and require systemic immunosuppression yet underlying mechanisms of irAEs are not known in detail. Here, we describe the T cell-mediated immune reaction in irAE lesions of four patients that developed pneumonitis during therapy with a PD-1 blocking antibody. Immunohistochemical analysis was performed to map the environment of the inflammatory lesions. Tumor infiltrating T cell clones were identified by sequencing the T cell receptor, and comparison with clones from peripheral blood or secondary lymphoid organs. A significant overlap of clones infiltrating irAE lesions and tumors was found. The most prevalent clones were also expanded in peripheral blood, but only a minor fraction of clonal overlap was found. Our findings suggest that irAE lesions in patients under PD-1 blockade are infiltrated by T cells with similar specificity as tumor-infiltrating T cells. These results raise the possibility that the immune response is elicited in these patients against antigens shared by the tumor and distant organs affected by irAEs.

Tumor infiltrating lymphocytes in lymph node metastases of stage III melanoma correspond to response and survival in nine patients treated with ipilimumab at the time of stage IV disease

Prognosis of metastatic melanoma improved with the development of checkpoint inhibitors. The role of tumor infiltrating lymphocytes (TILs) in lymph node metastases of stage III melanoma remains unclear. We retrospectively characterized TILs in primary melanomas and matched lymph node metastases (stage III melanoma) of patients treated with the checkpoint inhibitor ipilimumab. Tumor infiltrating lymphocytes were characterized for CD3, CD4, and CD8 expressions by immunohistochemistry. 4/9 patients (44%) responded to treatment with ipilimumab (1 complete and 2 partial remissions, 1 stable disease). All responders exhibited CD4 and CD8 T-cell infiltration in their lymph node metastases, whereas all non-responders did not show an infiltration of the lymph node metastasis with TILs. The correlation between the presence and absence of TILs in responders vs. non-responders was statistically significant (p = 0.008). Median distant metastases free survival, i.e., progression from stage III to stage IV melanoma, was similar in responders and non-responders (22.1 vs. 19.3 months; p = 0.462). Median progression free and overall survival show a trend in favor of the patients having TIL rich lymph node metastases (6.8 vs. 3.3 months, p = 0.09; and all alive at last follow-up vs. 8.2 months, respectively, p = 0.08). Our data suggest a correlation between the T-cell infiltration of the lymph node metastases in stage III melanoma and the response to ipilimumab once these patients progress to stage IV disease.

Cytokeratin-based assessment of tumour budding in colorectal cancer: analysis in stage II patients and prospective diagnostic experience

Tumour budding in colorectal cancer is an important prognostic factor. A recent consensus conference elaborated recommendations and key issues for future studies, among those the use of pan‐cytokeratin stains, especially in stage II patients. We report the first prospective diagnostic experience using pan‐cytokeratin for tumour budding assessment. Moreover, we evaluate tumour budding using pan‐cytokeratin stains and disease‐free survival (DFS) in stage II patients. To this end, tumour budding on pan‐cytokeratin‐stained sections was evaluated by counting the number of tumour buds in 10 high‐power fields (0.238 mm²), then categorizing counts as low/high‐grade at a cut‐off of 10 buds, in two cohorts. Cohort 1: prospective setting with 236 unselected primary resected colorectal cancers analysed by 17 pathologists during diagnostic routine. Cohort 2: retrospective cohort of 150 stage II patients with information on DFS. In prospective analysis of cohort 1, tumour budding counts correlated with advanced pT, lymph node metastasis, lymphovascular invasion, perineural invasion (all p < 0.0001), and distant metastasis (p = 0.0128). In cohort 2, tumour budding was an independent predictor of worse DFS using counts [p = 0.037, HR (95% CI): 1.007 (1.0–1.014)] and the low‐grade/high‐grade scoring approach [p = 0.02; HR (95% CI): 3.04 (1.2–7.77), 90.7 versus 73%, respectively]. In conclusion, tumour budding assessed on pan‐cytokeratin slides is feasible in a large pathology institute and leads to expected associations with clinicopathological features. Additionally, it is an independent predictor of poor prognosis in stage II patients and should be considered for risk stratification in future clinical studies.

Tumor-associated macrophages and response to 5-fluorouracil adjuvant therapy in stage III colorectal cancer

Tumor-associated macrophages (TAMs) play a role in tumor development and progression. We hypothesized that abundance of TAMs might modify efficacy of 5-fluorouracil chemotherapy in colorectal cancer. We measured the density of CD68+ TAMs at the invasive front of primary tumor of colorectal carcinoma (PT-TAMs; n = 208), at available matched metastatic lymph node (LN-TAMs; n = 149), and in an independent set of primary colorectal cancers (PT-TAMs, n = 111). The hazard ratios for disease-free survival were computed by Cox proportional-hazards model. In exploratory analysis, the interaction between TAMs and 5-fluorouracil adjuvant therapy was significant (PT-TAMs, p = 0.02; LN-TAMs, p = 0.005). High TAMs were independently associated with better disease-free survival only in 5-fluorouracil-treated patients (PT-TAMs, HR 0.23; 95%CI, 0.08–0.65; p = 0.005; LN-TAMs, HR 0.13; 95%CI, 0.04–0.43; p = 0.001). The independent predictive value of PT-TAMs was replicated in the external set (HR, 0.14; 95%CI 0.02–1.00; p = 0.05). In an in vitro experiment, 5-fluorouracil and macrophages showed a synergistic effect and increased colorectal cancer cell death. High densities of TAMs, particularly in metastatic lymph-nodes, identify stage III colorectal cancer patients benefitting from 5-fluorouracil adjuvant therapy.