Fast-Track Development and Multi-Institutional Clinical Validation of an Artificial Intelligence Algorithm for Detection of Lymph Node Metastasis in Colorectal Cancer.

Lymph node metastasis (LNM) detection can be automated using artificial intelligence (AI)-based diagnostic tools. Only limited studies have addressed this task for colorectal cancer (CRC). This study aimed to develop of a clinical-grade digital pathology tool for LNM detection in CRC using the original fast-track framework. The training cohort included 432 slides from one department. A segmentation algorithm detecting 8 relevant tissue classes was trained. The test cohorts consisted of materials from 5 pathology departments digitized by 4 different scanning systems. A high-quality, large training data set was generated within 7 days and a minimal amount of annotation work using fast-track principles. The AI tool showed very high accuracy for LNM detection in all cohorts, with sensitivity, negative predictive value, and specificity ranges of 0.980 to 1.000, 0.997 to 1.000, and 0.913 to 0.990, correspondingly. Only 5 of 14,460 analyzed test slides with tumor cells over all cohorts were classified as false negative (3/5 representing clusters of tumor cells in lymphatic vessels). A clinical-grade tool was trained in a short time using fast-track development principles and validated using the largest international, multi-institutional, multiscanner cohort of cases to date, showing very high precision for LNM detection in CRC. We are releasing a part of the test data sets to facilitate academic research.

Artificial Intelligence-Based Tool for Tumor Detection and Quantitative Tissue Analysis in Colorectal Specimens.

Digital pathology adoption allows for applying computational algorithms to routine pathology tasks. Our study aimed to develop a clinical-grade artificial intelligence (AI) tool for precise multiclass tissue segmentation in colorectal specimens (resections and biopsies) and clinically validate the tool for tumor detection in biopsy specimens. The training data set included 241 precisely manually annotated whole-slide images (WSIs) from multiple institutions. The algorithm was trained for semantic segmentation of 11 tissue classes with an additional module for biopsy WSI classification. Six case cohorts from 5 pathology departments (4 countries) were used for formal and clinical validation, digitized by 4 different scanning systems. The developed algorithm showed high precision of segmentation of different tissue classes in colorectal specimens with composite multiclass Dice score of up to 0.895 and pixel-wise tumor detection specificity and sensitivity of up to 0.958 and 0.987, respectively. In the clinical validation study on multiple external cohorts, the AI tool reached sensitivity of 1.0 and specificity of up to 0.969 for tumor detection in biopsy WSI. The AI tool analyzes most biopsy cases in less than 1 minute, allowing effective integration into clinical routine. We developed and extensively validated a highly accurate, clinical-grade tool for assistive diagnostic processing of colorectal specimens. This tool allows for quantitative deciphering of colorectal cancer tissue for development of prognostic and predictive biomarkers and personalization of oncologic care. This study is a foundation for a SemiCOL computational challenge. We open-source multiple manually annotated and weakly labeled test data sets, representing a significant contribution to the colorectal cancer computational pathology field.

Quality control stress test for deep learning-based diagnostic model in digital pathology.

Digital pathology provides a possibility for computational analysis of histological slides and automatization of routine pathological tasks. Histological slides are very heterogeneous concerning staining, sections' thickness, and artifacts arising during tissue processing, cutting, staining, and digitization. In this study, we digitally reproduce major types of artifacts. Using six datasets from four different institutions digitized by different scanner systems, we systematically explore artifacts' influence on the accuracy of the pre-trained, validated, deep learning-based model for prostate cancer detection in histological slides. We provide evidence that any histological artifact dependent on severity can lead to a substantial loss in model performance. Strategies for the prevention of diagnostic model accuracy losses in the context of artifacts are warranted. Stress-testing of diagnostic models using synthetically generated artifacts might be an essential step during clinical validation of deep learning-based algorithms.

Glucagon-like peptide-1 and glucagon-like peptide-2 regulation during human liver regeneration.

Accumulating evidence suggests that metabolic demands of the regenerating liver are met via lipid metabolism and critical regulators of this process. As such, glucagon-like peptide-1 (GLP-1) and glucagon-like peptide-2 (GLP-2) critically affect hepatic regeneration in rodent models. The present study aimed to evaluate potential alterations and dynamics of circulating GLP-1 and GLP-2 in patients undergoing liver resections, focusing on post-hepatectomy liver failure (PHLF). GLP-1, GLP-2, Interleukin-6 (IL-6) and parameters of lipid metabolism were determined perioperatively in fasting plasma of 46 patients, who underwent liver resection. GLP-1 and GLP-2 demonstrated a rapid and consistently inverse time course during hepatic regeneration with a significant decrease of GLP-1 and increase of GLP-2 on POD1. Importantly, these postoperative dynamics were significantly more pronounced when PHLF occurred. Of note, the extent of resection or development of complications were not associated with these alterations. IL-6 mirrored the time course of GLP-2. Assessing the main degradation protein dipeptidyl peptidase 4 (DPP4) no significant association with either GLP-1 or -2 could be found. Additionally, in PHLF distinct postoperative declines in plasma lipid parameters were present and correlated with GLP-2 dynamics. Our data suggest dynamic inverse regulation of GLP-1 and GLP-2 during liver regeneration, rather caused by an increase in expression/release than by changes in degradation capacity and might be associated with inflammatory responses. Their close association with circulating markers of lipid metabolism and insufficient hepatic regeneration after liver surgery suggest a critical involvement during these processes in humans.

Artificial intelligence for tumour tissue detection and histological regression grading in oesophageal adenocarcinomas: a retrospective algorithm development and validation study.

BACKGROUND: Oesophageal adenocarcinoma and adenocarcinoma of the oesophagogastric junction are among the most common malignant epithelial tumours. Most patients receive neoadjuvant therapy before complete tumour resection. Histological assessment after resection includes identification of residual tumour tissue and areas of regressive tumour, data which are used to calculate a clinically relevant regression score. We developed an artificial intelligence (AI) algorithm for tumour tissue detection and tumour regression grading in surgical specimens from patients with oesophageal adenocarcinoma or adenocarcinoma of the oesophagogastric junction. METHODS: We used one training cohort and four independent test cohorts to develop, train, and validate a deep learning tool. The material consisted of histological slides from surgically resected specimens from patients with oesophageal adenocarcinoma and adenocarcinoma of the oesophagogastric junction from three pathology institutes (two in Germany, one in Austria) and oesophageal cancer cohort of The Cancer Genome Atlas (TCGA). All slides were from neoadjuvantly treated patients except for those from the TCGA cohort, who were neoadjuvant-therapy naive. Data from training cohort and test cohort cases were extensively manually annotated for 11 tissue classes. A convolutional neural network was trained on the data using a supervised principle. First, the tool was formally validated using manually annotated test datasets. Next, tumour regression grading was assessed in a retrospective cohort of post-neoadjuvant therapy surgical specimens. The grading of the algorithm was compared with that of a group of 12 board-certified pathologists from one department. To further validate the tool, three pathologists processed whole resection cases with and without AI assistance. FINDINGS: Of the four test cohorts, one included 22 manually annotated histological slides (n=20 patients), one included 62 sides (n=15), one included 214 slides (n=69), and the final one included 22 manually annotated histological slides (n=22). In the independent test cohorts the AI tool had high patch-level accuracy for identifying both tumour and regression tissue. When we validated the concordance of the AI tool against analyses by a group of pathologists (n=12), agreement was 63·6% (quadratic kappa 0·749; p<0·0001) at case level. The AI-based regression grading triggered true reclassification of resected tumour slides in seven cases (including six cases who had small tumour regions that were initially missed by pathologists). Use of the AI tool by three pathologists increased interobserver agreement and substantially reduced diagnostic time per case compared with working without AI assistance. INTERPRETATION: Use of our AI tool in the diagnostics of oesophageal adenocarcinoma resection specimens by pathologists increased diagnostic accuracy, interobserver concordance, and significantly reduced assessment time. Prospective validation of the tool is required. FUNDING: North Rhine-Westphalia state, Federal Ministry of Education and Research of Germany, and the Wilhelm Sander Foundation.

An international multi-institutional validation study of the algorithm for prostate cancer detection and Gleason grading.

Pathologic examination of prostate biopsies is time consuming due to the large number of slides per case. In this retrospective study, we validate a deep learning-based classifier for prostate cancer (PCA) detection and Gleason grading (AI tool) in biopsy samples. Five external cohorts of patients with multifocal prostate biopsy were analyzed from high-volume pathology institutes. A total of 5922 H&E sections representing 7473 biopsy cores from 423 patient cases (digitized using three scanners) were assessed concerning tumor detection. Two tumor-bearing datasets (core n = 227 and 159) were graded by an international group of pathologists including expert urologic pathologists (n = 11) to validate the Gleason grading classifier. The sensitivity, specificity, and NPV for the detection of tumor-bearing biopsies was in a range of 0.971-1.000, 0.875-0.976, and 0.988-1.000, respectively, across the different test cohorts. In several biopsy slides tumor tissue was correctly detected by the AI tool that was initially missed by pathologists. Most false positive misclassifications represented lesions suspicious for carcinoma or cancer mimickers. The quadratically weighted kappa levels for Gleason grading agreement for single pathologists was 0.62-0.80 (0.77 for AI tool) and 0.64-0.76 (0.72 for AI tool) for the two grading datasets, respectively. In cases where consensus for grading was reached among pathologists, kappa levels for AI tool were 0.903 and 0.855. The PCA detection classifier showed high accuracy for PCA detection in biopsy cases during external validation, independent of the institute and scanner used. High levels of agreement for Gleason grading were indistinguishable between experienced genitourinary pathologists and the AI tool.

Clinical and economic aspects of KRAS mutational status as predictor for epidermal growth factor receptor inhibitor therapy in metastatic colorectal cancer patients.

Treatment of metastasized colorectal cancer (mCRC) patients with anti-epidermal growth factor receptor (EGFR)-directed monoclonal antibodies is driven by the results of the KRAS mutational status (wild type [WT]/mutated [MUT]). To find out as to what extent the treatment selection based on the KRAS status had impact on overall costs, a retrospective analysis was performed. Of 73 mCRC patients 31.5% were MUT carriers. Costs of EGFR inhibitor treatment for WT patients were significantly higher compared to those for patients with MUT (p = 0.005). Higher treatment costs in WT carriers reflect a significantly higher number of treatment cycles (p = 0.012) in this cohort of patients. Savings of drug costs minus the costs for the determination of KRAS status accounted for EUR 779.42 (SD ±336.28) per patient per cycle. The routine use of KRAS screening is a cost-effective strategy. Costs of unnecessary monoclonal EGFR inhibitor treatment can be saved in MUT patients.

The OncoMasTR Test Predicts Distant Recurrence in Estrogen Receptor-Positive, HER2-Negative Early-Stage Breast Cancer: A Validation Study in ABCSG Trial 8.

PURPOSE: To validate the clinical performance of the OncoMasTR Risk Score in the biomarker cohort of Austrian Breast and Colorectal Cancer Study Group (ABCSG) Trial 8. EXPERIMENTAL DESIGN: We evaluated the OncoMasTR test in 1,200 formalin-fixed, paraffin-embedded (FFPE) surgical specimens from postmenopausal women with estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative primary breast cancer with 0 to 3 involved lymph nodes in the prospective, randomized ABCSG Trial 8. Time to distant recurrence (DR) was analyzed by Cox models. RESULTS: The OncoMasTR Risk Score categorized 850 of 1,087 (78.2%) evaluable patients as "low risk". At 10 years, the DR rate for patients in the low-risk group was 5.8% versus 21.1% for patients in the high-risk group (P < 0.0001, absolute risk reduction 15.3%). The OncoMasTR Risk Score was highly prognostic for prediction of DR in years 0 to 10 in all patients [HR 1.91, 95% confidence interval (CI) 1.62-2.26, P < 0.0001; C-index 0.73], in patients that were node negative (HR 1.79, 95% CI, 1.43-2.24, P < 0.0001; C-index 0.72), and in patients with 1 to 3 involved lymph nodes (HR 1.93, 95% CI, 1.44-2.58, P < 0.0001; C-index 0.71). The OncoMasTR Risk Score provided significant additional prognostic information beyond clinical parameters, Ki67, Nottingham Prognostic Index, and Clinical Treatment Score. CONCLUSIONS: OncoMasTR Risk Score is highly prognostic for DR in postmenopausal women with ER-positive, HER2-negative primary breast cancer with 0 to 3 involved lymph nodes. In combination with prior validation studies, this fully independent validation in ABCSG Trial 8 provides level 1B evidence for the prognostic capability of the OncoMasTR Risk Score.

Predictive Value of Molecular Subtypes in Premenopausal Women with Hormone Receptor-positive Early Breast Cancer: Results from the ABCSG Trial 5.

PURPOSE: To assess the predictive value of molecular breast cancer subtypes in premenopausal patients with hormone receptor-positive early breast cancer who received adjuvant endocrine treatment or chemotherapy. EXPERIMENTAL DESIGN: Molecular breast cancer subtypes were centrally assessed on whole tumor sections by IHC in patients of the Austrian Breast and Colorectal Cancer Study Group Trial 5 who had received either 5 years of tamoxifen/3 years of goserelin or six cycles of cyclophosphamide, methotrexate, and fluorouracil (CMF). Luminal A disease was defined as Ki67 <20% and luminal B as Ki67 ≥20%. The luminal B/HER2-positive subtype displayed 3+ HER2-IHC or amplification by ISH. Recurrence-free survival (RFS) and overall survival (OS) were analyzed using Cox models adjusted for clinical and pathologic factors. RESULTS: 185 (38%), 244 (50%), and 59 (12%) of 488 tumors were classified as luminal A, luminal B/HER2-negative and luminal B/HER2-positive, respectively. Luminal B subtypes were associated with poor outcome. Patients with luminal B tumors had a significantly shorter RFS [adjusted HR for recurrence: 2.22; 95% confidence interval (CI), 1.41-3.49; P = 0.001] and OS (adjusted HR for death: 3.51; 95% CI, 1.80-6.87; P < 0.001). No interaction between molecular subtypes and treatment was observed (test for interaction: P = 0.84 for RFS; P = 0.69 for OS). CONCLUSIONS: Determination of molecular subtypes by IHC is an independent prognostic factor for recurrence and death in premenopausal women with early-stage, hormone receptor-positive breast cancer but is not predictive for outcome of adjuvant treatment with tamoxifen/goserelin or CMF.See related commentary by Hunter et al., p. 5543.

Nbn heterozygosity renders mice susceptible to tumor formation and ionizing radiation-induced tumorigenesis.

Nijmegen Breakage Syndrome (NBS) is a rare autosomal recessive disease characterized by microcephaly, growth retardation, immunodeficiency, chromosomal instability, and predisposition to cancer. Heterozygous NBS patients show increased chromosomal instability and are suspected to be at a high risk for cancer. To study the impact of NBS1 heterozygosity on malignancy susceptibility, we disrupted the murine homologue (Nbn) of NBS1 in mice using gene targeting techniques. While null mutation in the Nbn gene resulted in embryonic lethality at the blastocyst stage because of growth retardation and increased apoptosis, heterozygous knockout (Nbn(+/-)) mice developed a wide array of tumors affecting the liver, mammary gland, prostate, and lung, in addition to lymphomas. Moreover, gamma-irradiation enhanced tumor development in Nbn(+/-) mice, giving rise to a high frequency of epithelial tumors, mostly in the thyroid and lung, as well as lymphomas. These mice also developed numerous tumors in the ovary and testis. Southern and Western blot analyses showed a remaining wild-type allele and nibrin expression in Nbn(+/-) tumors. Sequencing analysis confirmed no mutation in the Nbn cDNA derived from these tumors. Cytogenetic analysis revealed that primary Nbn(+/-) embryonic fibroblasts and tumor cells exhibit increased chromosomal aberrations. These data suggest that haploinsufficiency, not loss of heterozygosity, of Nbn could be the mechanism underlying the tumor development. Taken together, our heterozygous Nbn-knockout mice represent a novel model to study the consequences of NBS1 heterozygosity on tumor development.

[Histology-based algorithm in the molecular diagnosis of mutations of the Epidernal Growth Factor Receptor (EGFR) in non-small cell lung cancer]. / Histologie-basierter Algorithmus der molekularen Diagnostik des epidermalen Wachstumsfaktor-Rezeptorgens (Epidermal Growth Factor Receptor, EGFR) beim nicht-kleinzelligen Lungenkarzinom (non-small cell lung cancer, NSCLC).

Patients with stage IIIB and IV non-small cell lung carcinoma (NSCLC) harboring an activating mutation of the Epidermal Growth Factor Receptor (EGFR) Gene should be treated first-line with Gefitinib, an EGFR tyrosine kinase inhibitor (TKI). EGF receptor mutations are most common in adenocarcinomas, especially non-mucinous type, rare in squamous cell carcinomas and sarcomatoid carcinomas, and do not occur in neuroendocrine carcinomas. Therefore, the Pulmonary Pathology Working Group of the Austrian Society of Pathology, after intense discussions and in consensus with Oncologists and Pulmonologists, recommends a priori EGFR mutation analysis for all cases of adenocarcinoma, and for all other NSCLC upon clinical request. This will markedly reduce waiting time for those patients, which most likely will have the greatest benefit from EGFR TKI therapy.