Automatic quantification of tumor-stroma ratio as a prognostic marker for pancreatic cancer.

PURPOSE: This study aims to introduce an innovative multi-step pipeline for automatic tumor-stroma ratio (TSR) quantification as a potential prognostic marker for pancreatic cancer, addressing the limitations of existing staging systems and the lack of commonly used prognostic biomarkers. METHODS: The proposed approach involves a deep-learning-based method for the automatic segmentation of tumor epithelial cells, tumor bulk, and stroma from whole-slide images (WSIs). Models were trained using five-fold cross-validation and evaluated on an independent external test set. TSR was computed based on the segmented components. Additionally, TSR's predictive value for six-month survival on the independent external dataset was assessed. RESULTS: Median Dice (inter-quartile range (IQR)) of 0.751(0.15) and 0.726(0.25) for tumor epithelium segmentation on internal and external test sets, respectively. Median Dice of 0.76(0.11) and 0.863(0.17) for tumor bulk segmentation on internal and external test sets, respectively. TSR was evaluated as an independent prognostic marker, demonstrating a cross-validation AUC of 0.61±0.12 for predicting six-month survival on the external dataset. CONCLUSION: Our pipeline for automatic TSR quantification offers promising potential as a prognostic marker for pancreatic cancer. The results underscore the feasibility of computational biomarker discovery in enhancing patient outcome prediction, thus contributing to personalized patient management.

Lymph node metastases and recurrence in pT1 colorectal cancer: Prediction with the International Budding Consortium Score-A retrospective, multi-centric study.

BACKGROUND: The International Collaboration on Cancer Reporting proposes histological tumour type, lymphovascular invasion, tumour grade, perineural invasion, extent, and dimensions of invasion as risk factors for lymph node metastases and tumour progression in completely endoscopically resected pT1 colorectal cancer (CRC). OBJECTIVE: The aim of the study was to propose a predictive and reliable score to optimise the clinical management of endoscopically resected pT1 CRC patients. METHODS: This multi-centric, retrospective International Budding Consortium (IBC) study included an international pT1 CRC cohort of 565 patients. All cases were reviewed by eight expert gastrointestinal pathologists. All risk factors were reported according to international guidelines. Tumour budding and immune response (CD8+ T-cells) were assessed with automated models using artificial intelligence. We used the information on risk factors and least absolute shrinkage and selection operator logistic regression to develop a prediction model and generate a score to predict the occurrence of lymph node metastasis or cancer recurrence. RESULTS: The IBC prediction score included the following parameters: lymphovascular invasion, tumour buds, infiltration depth and tumour grade. The score has an acceptable discrimination power (area under the curve of 0.68 [95% confidence intervals (CI) 0.61-0.75]; 0.64 [95% CI 0.57-0.71] after internal validation). At a cut-off of 6.8 points to discriminate high-and low-risk patients, the score had a sensitivity and specificity of 0.9 [95% CI 0.8-0.95] and 0.26 [95% 0.22, 0.3], respectively. CONCLUSION: The IBC score is based on well-established risk factors and is a promising tool with clinical utility to support the management of pT1 CRC patients.

The Complexity of Shapes: How the Circularity of Tumor Nodules Affects Prognosis in Colorectal Cancer.

The current stratification of tumor nodules in colorectal cancer (CRC) staging is subjective and leads to high interobserver variability. In this study, the objective assessment of the shape of lymph node metastases (LNMs), extranodal extension (ENE), and tumor deposits (TDs) was correlated with outcomes. A test cohort and a validation cohort were included from 2 different institutions. The test cohort consisted of 190 cases of stage III CRC. Slides with LNMs and TDs were annotated and processed using a segmentation algorithm to determine their shape. The complexity ratio was calculated for every shape and correlated with outcomes. A cohort of 160 stage III CRC cases was used to validate findings. TDs showed significantly more complex shapes than LNMs with ENE, which were more complex than LNMs without ENE (P < .001). In the test cohort, patients with the highest sum of complexity ratios had significantly lower disease-free survival (P < .01). When only the nodule with the highest complexity was considered, this effect was even stronger (P < .001). This maximum complexity ratio per patient was identified as an independent prognostic factor in the multivariate analysis (hazard ratio, 2.47; P < .05). The trends in the validation cohort confirmed the results. More complex nodules in stage III CRC were correlated with significantly worse disease-free survival, even if only based on the most complex nodule. These results suggest that more complex nodules reflect more invasive tumor biology. As most of the more complex nodules were diagnosed as TDs, we suggest providing a more prominent role for TDs in the nodal stage and include an objective complexity measure in their definition.

Deep learning for multi-class semantic segmentation enables colorectal cancer detection and classification in digital pathology images.

In colorectal cancer (CRC), artificial intelligence (AI) can alleviate the laborious task of characterization and reporting on resected biopsies, including polyps, the numbers of which are increasing as a result of CRC population screening programs ongoing in many countries all around the globe. Here, we present an approach to address two major challenges in the automated assessment of CRC histopathology whole-slide images. We present an AI-based method to segment multiple ([Formula: see text]) tissue compartments in the H &E-stained whole-slide image, which provides a different, more perceptible picture of tissue morphology and composition. We test and compare a panel of state-of-the-art loss functions available for segmentation models, and provide indications about their use in histopathology image segmentation, based on the analysis of (a) a multi-centric cohort of CRC cases from five medical centers in the Netherlands and Germany, and (b) two publicly available datasets on segmentation in CRC. We used the best performing AI model as the basis for a computer-aided diagnosis system that classifies colon biopsies into four main categories that are relevant pathologically. We report the performance of this system on an independent cohort of more than 1000 patients. The results show that with a good segmentation network as a base, a tool can be developed which can support pathologists in the risk stratification of colorectal cancer patients, among other possible uses. We have made the segmentation model available for research use on https://grand-challenge.org/algorithms/colon-tissue-segmentation/ .

Fully Automated Tumor Bud Assessment in Hematoxylin and Eosin-Stained Whole Slide Images of Colorectal Cancer.

Tumor budding (TB), the presence of single cells or small clusters of up to 4 tumor cells at the invasive front of colorectal cancer (CRC), is a proven risk factor for adverse outcomes. International definitions are necessary to reduce interobserver variability. According to the current international guidelines, hotspots at the invasive front should be counted in hematoxylin and eosin (H&E)-stained slides. This is time-consuming and prone to interobserver variability; therefore, there is a need for computer-aided diagnosis solutions. In this study, we report an artificial intelligence-based method for detecting TB in H&E-stained whole slide images. We propose a fully automated pipeline to identify the tumor border, detect tumor buds, characterize them based on the number of tumor cells, and produce a TB density map to identify the TB hotspot. The method outputs the TB count in the hotspot as a computational biomarker. We show that the proposed automated TB detection workflow performs on par with a panel of 5 pathologists at detecting tumor buds and that the hotspot-based TB count is an independent prognosticator in both the univariate and the multivariate analysis, validated on a cohort of n = 981 patients with CRC. Computer-aided detection of tumor buds based on deep learning can perform on par with expert pathologists for the detection and quantification of tumor buds in H&E-stained CRC histopathology slides, strongly facilitating the introduction of budding as an independent prognosticator in clinical routine and clinical trials.

Semi-Supervised Learning to Automate Tumor Bud Detection in Cytokeratin-Stained Whole-Slide Images of Colorectal Cancer.

Tumor budding is a histopathological biomarker associated with metastases and adverse survival outcomes in colorectal carcinoma (CRC) patients. It is characterized by the presence of single tumor cells or small clusters of cells within the tumor or at the tumor-invasion front. In order to obtain a tumor budding score for a patient, the region with the highest tumor bud density must first be visually identified by a pathologist, after which buds will be counted in the chosen hotspot field. The automation of this process will expectedly increase efficiency and reproducibility. Here, we present a deep learning convolutional neural network model that automates the above procedure. For model training, we used a semi-supervised learning method, to maximize the detection performance despite the limited amount of labeled training data. The model was tested on an independent dataset in which human- and machine-selected hotspots were mapped in relation to each other and manual and machine detected tumor bud numbers in the manually selected fields were compared. We report the results of the proposed method in comparison with visual assessment by pathologists. We show that the automated tumor bud count achieves a prognostic value comparable with visual estimation, while based on an objective and reproducible quantification. We also explore novel metrics to quantify buds such as density and dispersion and report their prognostic value. We have made the model available for research use on the grand-challenge platform.

Deep learning based tumor-stroma ratio scoring in colon cancer correlates with microscopic assessment.

Background: The amount of stroma within the primary tumor is a prognostic parameter for colon cancer patients. This phenomenon can be assessed using the tumor-stroma ratio (TSR), which classifies tumors in stroma-low (≤50% stroma) and stroma-high (>50% stroma). Although the reproducibility for TSR determination is good, improvement might be expected from automation. The aim of this study was to investigate whether the scoring of the TSR in a semi- and fully automated method using deep learning algorithms is feasible. Methods: A series of 75 colon cancer slides were selected from a trial series of the UNITED study. For the standard determination of the TSR, 3 observers scored the histological slides. Next, the slides were digitized, color normalized, and the stroma percentages were scored using semi- and fully automated deep learning algorithms. Correlations were determined using intraclass correlation coefficients (ICCs) and Spearman rank correlations. Results: 37 (49%) cases were classified as stroma-low and 38 (51%) as stroma-high by visual estimation. A high level of concordance between the 3 observers was reached, with ICCs of 0.91, 0.89, and 0.94 (all P < .001). Between visual and semi-automated assessment the ICC was 0.78 (95% CI 0.23-0.91, P-value 0.005), with a Spearman correlation of 0.88 (P < .001). Spearman correlation coefficients above 0.70 (N=3) were observed for visual estimation versus the fully automated scoring procedures. Conclusion: Good correlations were observed between standard visual TSR determination and semi- and fully automated TSR scores. At this point, visual examination has the highest observer agreement, but semi-automated scoring could be helpful to support pathologists.

Deep learning for fully-automated nuclear pleomorphism scoring in breast cancer.

To guide the choice of treatment, every new breast cancer is assessed for aggressiveness (i.e., graded) by an experienced histopathologist. Typically, this tumor grade consists of three components, one of which is the nuclear pleomorphism score (the extent of abnormalities in the overall appearance of tumor nuclei). The degree of nuclear pleomorphism is subjectively classified from 1 to 3, where a score of 1 most closely resembles epithelial cells of normal breast epithelium and 3 shows the greatest abnormalities. Establishing numerical criteria for grading nuclear pleomorphism is challenging, and inter-observer agreement is poor. Therefore, we studied the use of deep learning to develop fully automated nuclear pleomorphism scoring in breast cancer. The reference standard used for training the algorithm consisted of the collective knowledge of an international panel of 10 pathologists on a curated set of regions of interest covering the entire spectrum of tumor morphology in breast cancer. To fully exploit the information provided by the pathologists, a first-of-its-kind deep regression model was trained to yield a continuous scoring rather than limiting the pleomorphism scoring to the standard three-tiered system. Our approach preserves the continuum of nuclear pleomorphism without necessitating a large data set with explicit annotations of tumor nuclei. Once translated to the traditional system, our approach achieves top pathologist-level performance in multiple experiments on regions of interest and whole-slide images, compared to a panel of 10 and 4 pathologists, respectively.

Unleashing the potential of digital pathology data by training computer-aided diagnosis models without human annotations.

The digitalization of clinical workflows and the increasing performance of deep learning algorithms are paving the way towards new methods for tackling cancer diagnosis. However, the availability of medical specialists to annotate digitized images and free-text diagnostic reports does not scale with the need for large datasets required to train robust computer-aided diagnosis methods that can target the high variability of clinical cases and data produced. This work proposes and evaluates an approach to eliminate the need for manual annotations to train computer-aided diagnosis tools in digital pathology. The approach includes two components, to automatically extract semantically meaningful concepts from diagnostic reports and use them as weak labels to train convolutional neural networks (CNNs) for histopathology diagnosis. The approach is trained (through 10-fold cross-validation) on 3'769 clinical images and reports, provided by two hospitals and tested on over 11'000 images from private and publicly available datasets. The CNN, trained with automatically generated labels, is compared with the same architecture trained with manual labels. Results show that combining text analysis and end-to-end deep neural networks allows building computer-aided diagnosis tools that reach solid performance (micro-accuracy = 0.908 at image-level) based only on existing clinical data without the need for manual annotations.

Shrinkage versus fragmentation response in neoadjuvantly treated oesophageal adenocarcinoma: significant prognostic relevance.

AIMS: No consensus exists on the clinical value of tumour regression grading (TRG) systems for therapy effects of neoadjuvant chemoradiotherapy (nCRT) in oesophageal adenocarcinoma. Existing TRG systems lack standardization and reproducibility, and do not consider the morphological heterogeneity of tumour response. Therefore, we aim to identify morphological tumour regression patterns of oesophageal adenocarcinoma after nCRT and their association with survival. METHODS AND RESULTS: Patients with oesophageal adenocarcinoma, who underwent nCRT followed by surgery and achieved a partial response to nCRT, were identified from two Dutch upper-gastrointestinal (GI) centres (2005-18; test cohort). Resection specimens were scored for regression patterns by two independent observers according to a pre-defined three-step flowchart. The results were validated in an external cohort (2001-17). In total, 110 patients were included in the test cohort and 115 in the validation cohort. In the test cohort, two major regression patterns were identified: fragmentation (60%) and shrinkage (40%), with an excellent interobserver agreement (κ = 0.87). Here, patients with a fragmented pattern had a significantly higher pathological stage (stages III/IV: 52 versus 16%; P < 0.001), less downstaging (48 versus 91%; P < 0.001), a higher risk of recurrence [risk ratio (RR) = 2.9, 95% confidence interval (CI) = 1.5-5.6] and poorer 5-year overall survival (30 versus 80% respectively, P = 0.001). CONCLUSIONS: The validation cohort confirmed these findings, although had more advanced cases (case-stages = III/IV 91 versus 73%, P = 0.005) and a higher prevalence of fragmented-pattern cases (80 versus 60%, P = 0.002). When combining the cohorts in multivariate analysis, the pattern of response was an independent prognostic factor [hazard ratio (HR) = 1.76, 95% CI = 1.0-3.0]. In conclusion, we established an externally validated, reproducible and clinically relevant classification of tumour response.

Improving tumor budding reporting in colorectal cancer: a Delphi consensus study.

Tumor budding is a long-established independent adverse prognostic marker in colorectal cancer, yet methods for its assessment have varied widely. In an effort to standardize its reporting, a group of experts met in Bern, Switzerland, in 2016 to reach consensus on a single, international, evidence-based method for tumor budding assessment and reporting (International Tumor Budding Consensus Conference [ITBCC]). Tumor budding assessment using the ITBCC criteria has been validated in large cohorts of cancer patients and incorporated into several international colorectal cancer pathology and clinical guidelines. With the wider reporting of tumor budding, new issues have emerged that require further clarification. To better inform researchers and health-care professionals on these issues, an international group of experts in gastrointestinal pathology participated in a modified Delphi process to generate consensus and highlight areas requiring further research. This effort serves to re-affirm the importance of tumor budding in colorectal cancer and support its continued use in routine clinical practice.

Taking tumour budding to the next frontier - a post International Tumour Budding Consensus Conference (ITBCC) 2016 review.

Tumour budding in colorectal cancer, defined as single tumour cells or small clusters containing four or fewer tumour cells, is a robust and independent biomarker of aggressive tumour biology. On the basis of published data in the literature, the evidence is certainly in favour of reporting tumour budding in routine practice. One important aspect of implementing tumour budding has been to establish a standardised and evidence-based scoring method, as was recommended by the International Tumour Budding Consensus Conference (ITBCC) in 2016. Further developments have aimed at establishing methods for automated tumour budding assessment. A digital approach to scoring tumour buds has great potential to assist in performing an objective budding count but, like the manual consensus method, must be validated and standardised. The aim of the present review is to present general considerations behind the ITBCC scoring method, and a broad overview of the current situation and challenges regarding automated tumour budding detection methods.

Are tumour grade and tumour budding equivalent in colorectal cancer? A retrospective analysis of 771 patients.

BACKGROUND: Tumour grade is traditionally considered in the management of patients with colorectal cancer. However, a large body of literature suggests that a related feature, namely tumour budding, may have a more important clinical impact. The aim of our study is to determine the correlation between tumour grade and tumour budding and their impact on patient outcome. METHODS: A retrospective collective of 771 patients with colorectal cancer were included in the study. Clinicopathological information included tumour grade (World Health Organisation 2010; G1, G2 and G3) and tumour budding evaluated as BD1, BD2 and BD3 and representing 0-4 buds, 5-9 buds and 10 or more buds per 0.785 mm2, respectively. RESULTS: Tumour grade and tumour budding were correlated (p < 0.0001, percent concordance: 33.8%). Of the BD1 cases, 18.1% were of G3. Only two BD3 cases were G1. Both high tumour grade and tumour budding were associated with higher pT, lymph node metastasis, distant metastasis and lymphatic and venous vessel invasion (p < 0.01, all), but only tumour grade was additionally associated with right-sided tumour location and mucinous histology. Higher tumour budding led to worse overall (p = 0.0286) and disease-free survival (p = 0.001), but tumour grade did not. Budding was independent of tumour grade in multivariate analysis. DISCUSSION: Tumour grade and tumour budding are distinct features, as recognised by their different clinicopathological associations, reflecting different underlying biological processes. Nonetheless, tumour budding seems to outperform tumour grade in terms of predicting disease-free survival.

Mathematical Models for Tumor Growth and the Reduction of Overtreatment.

Background To improve our understanding of the natural course of head and neck paragangliomas (HNPGL) and ultimately differentiate between cases that benefit from early treatment and those that are best left untreated, we studied the growth dynamics of 77 HNPGL managed with primary observation. Methods Using digitally available magnetic resonance images, tumor volume was estimated at three time points. Subsequently, nonlinear least squares regression was used to fit seven mathematical models to the observed growth data. Goodness of fit was assessed with the coefficient of determination ( R 2 ) and root-mean-squared error. The models were compared with Kruskal-Wallis one-way analysis of variance and subsequent post-hoc tests. In addition, the credibility of predictions (age at onset of neoplastic growth and estimated volume at age 90) was evaluated. Results Equations generating sigmoidal-shaped growth curves (Gompertz, logistic, Spratt and Bertalanffy) provided a good fit (median R 2 : 0.996-1.00) and better described the observed data compared with the linear, exponential, and Mendelsohn equations ( p < 0.001). Although there was no statistically significant difference between the sigmoidal-shaped growth curves regarding the goodness of fit, a realistic age at onset and estimated volume at age 90 were most often predicted by the Bertalanffy model. Conclusions Growth of HNPGL is best described by decelerating tumor growth laws, with a preference for the Bertalanffy model. To the best of our knowledge, this is the first time that this often-neglected model has been successfully fitted to clinically obtained growth data.