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.

Emergence of Tn1999.7, a New Transposon in blaOXA-48-Harboring Plasmids Associated with Increased Plasmid Stability.

OXA-48 is the most common carbapenemase in Enterobacterales in Germany and many other European countries. Depending on the genomic location of blaOXA-48, OXA-48-producing isolates vary in phenotype and intra- and interspecies transferability of blaOXA-48. In most bacterial isolates, blaOXA-48 is located on one of seven variants of Tn1999 (Tn1999.1 to Tn1999.6 and invTn1999.2). Here, a novel Tn1999 variant, Tn1999.7, is described, which was identified in 11 clinical isolates from 2016 to 2020. Tn1999.7 differs from Tn1999.1 by the insertion of the 8,349-bp Tn3 family transposon Tn7442 between the lysR gene and blaOXA-48 open reading frame. Tn7442 carries genes coding for a restriction endonuclease and a DNA methyltransferase as cargo, forming a type III restriction modification system. Tn1999.7 was carried on an ~71-kb IncL plasmid in 9/11 isolates. In one isolate, Tn1999.7 was situated on an ~76-kb plasmid, harboring an additional insertion sequence in the plasmid backbone. In one isolate, the plasmid size is only ~63 kb due to a deletion adjacent to Tn7442 that extends into the plasmid backbone. Mean conjugation rates of the Tn1999.7-harboring plasmids in J53 ranged from 4.47 × 10-5 to 2.03 × 10-2, similar to conjugation rates of other pOXA-48-type IncL plasmids. The stability of plasmids with Tn1999.7 was significantly higher than that of a Tn1999.2-harboring plasmid in vitro. This increase in stability could be related to the insertion of a restriction-modification system, which can promote postsegregational killing. The increased plasmid stability associated with Tn1999.7 could contribute to the further spread of OXA-48.

Pathogenicity of Clinical OXA-48 Isolates and Impact of the OXA-48 IncL Plasmid on Virulence and Bacterial Fitness.

OXA-48 is the most common carbapenemase in Enterobacterales in Germany and one of the most frequent carbapenemases worldwide. Several reports have associated bla OXA - 48 with a virulent host phenotype. To challenge this hypothesis, 35 OXA-48-producing clinical isolates of Escherichia coli (n = 15) and Klebsiella pneumoniae (n = 20) were studied in vitro, in vivo employing the Galleria mellonella infection model and by whole-genome sequencing. Clinical isolates belonged to 7 different sequence types (STs) in E. coli and 12 different STs in K. pneumoniae. In 26/35 isolates bla OXA- 48 was located on a 63 kb IncL plasmid. Horizontal gene transfer (HGT) to E. coli J53 was high in isolates with the 63 kb IncL plasmid (transconjugation frequency: ∼103/donor) but low in isolates with non-IncL plasmids (<10-6/donor). Several clinical isolates were both highly cytotoxic against human cells and virulent in vivo. However, 63 kb IncL transconjugants generated from these highly virulent isolates were not more cytotoxic or virulent when compared to the recipient strain. Additionally, no genes associated with virulence were detected by in silico analysis of OXA-48 plasmids. The 63 kb plasmid was highly stable and did not impair growth or fitness in E. coli J53. In conclusion, OXA-48 clinical isolates in Germany are diverse but typically harbor the same 63 kb IncL plasmid which has been reported worldwide. We demonstrate that this 63 kb IncL plasmid has a low fitness burden, high plasmid stability and can be transferred by highly efficient HGT which is likely the cause of the rapid dissemination of OXA-48 rather than the expansion of a single clone or gain of virulence.

Transgenic tobacco plants accumulating osmolytes show reduced oxidative damage under freezing stress.

We studied the reaction to the oxidative component of freezing in several tobacco lines, transformed with genes coding for enzymes involved in the synthesis of osmoprotectants (proline, fructan or glycine betaine) along with their wild type. The levels of some oxidative stress markers (leakage of electrolytes, hydrogen peroxide and malondialdehyde) as well as the activity of antioxidative enzymes catalase (EC 1.11.1.6.) and guaiacol peroxidase (EC 1.11.1.7.) have been followed at acclimation, 12 and 24 h freezing and at recovery. Freezing for 24 h resulted in severe damages for the wild type. A corresponding increase of electrolyte leakage, hydrogen peroxide and malondialdehyde contents, a rise of peroxidase activity and inhibition of catalase activity occurred in the non-transformants. Similar, but significantly lower trend of the same parameters has been found for the transgenic lines. Moreover, the oxidative markers returned to their normal levels when the transformants were able to recover from freezing. It could be speculated that transfer of genes, coding for accumulation of osmoprotectants, is related to reduced intensity of freezing-induced oxidative processes. Our lines and model system could serve as a good prerequisite for additional studies to gain further insights into the complex role of osmoprotectants in freezing tolerance.