Image-Based Deep Learning Detection of High-Grade B-Cell Lymphomas Directly from Hematoxylin and Eosin Images.

Deep learning applications are emerging as promising new tools that can support the diagnosis and classification of different cancer types. While such solutions hold great potential for hematological malignancies, there have been limited studies describing the use of such applications in this field. The rapid diagnosis of double/triple-hit lymphomas (DHLs/THLs) involving MYC, BCL2 and/or BCL6 rearrangements is obligatory for optimal patient care. Here, we present a novel deep learning tool for diagnosing DHLs/THLs directly from scanned images of biopsy slides. A total of 57 biopsies, including 32 in a training set (including five DH lymphoma cases) and 25 in a validation set (including 10 DH/TH cases), were included. The DHL-classifier demonstrated a sensitivity of 100%, a specificity of 87% and an AUC of 0.95, with only two false positive cases, compared to FISH. The DHL-classifier showed a 92% predictive value as a screening tool for performing conventional FISH analysis, over-performing currently used criteria. The work presented here provides the proof of concept for the potential use of an AI tool for the identification of DH/TH events. However, more extensive follow-up studies are required to assess the robustness of this tool and achieve high performances in a diverse population.

Direct identification of ALK and ROS1 fusions in non-small cell lung cancer from hematoxylin and eosin-stained slides using deep learning algorithms.

Anaplastic lymphoma kinase (ALK) and ROS oncogene 1 (ROS1) gene fusions are well-established key players in non-small cell lung cancer (NSCLC). Although their frequency is relatively low, their detection is important for patient care and guides therapeutic decisions. The accepted methods used for their detection are immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) assay, as well as DNA and RNA-based sequencing methodologies. These assays are expensive, time-consuming, and require technical expertise and specialized equipment as well as biological specimens that are not always available. Here we present an alternative detection method using a computer vision deep learning approach. An advanced convolutional neural network (CNN) was used to generate classifier models to detect ALK and ROS1-fusions directly from scanned hematoxylin and eosin (H&E) whole slide images prepared from NSCLC tumors of patients. A two-step training approach was applied, with an initial unsupervised training step performed on a pan-cancer sample cohort followed by a semi-supervised fine-tuning step, which supported the development of a classifier with performances equal to those accepted for diagnostic tests. Validation of the ALK/ROS1 classifier on a cohort of 72 lung cancer cases who underwent ALK and ROS1-fusion testing at the pathology department at Sheba Medical Center displayed sensitivities of 100% for both genes (six ALK-positive and two ROS1-positive cases) and specificities of 100% and 98.6% respectively for ALK and ROS1, with only one false-positive result for ROS1-alteration. These results demonstrate the potential advantages that machine learning solutions may have in the molecular pathology domain, by allowing fast, standardized, accurate, and robust biomarker detection overcoming many limitations encountered when using current techniques. The integration of such novel solutions into the routine pathology workflow can support and improve the current clinical pipeline.

Elevated RNA Editing Activity Is a Major Contributor to Transcriptomic Diversity in Tumors.

Genomic mutations in key genes are known to drive tumorigenesis and have been the focus of much attention in recent years. However, genetic content also may change farther downstream. RNA editing alters the mRNA sequence from its genomic blueprint in a dynamic and flexible way. A few isolated cases of editing alterations in cancer have been reported previously. Here, we provide a transcriptome-wide characterization of RNA editing across hundreds of cancer samples from multiple cancer tissues, and we show that A-to-I editing and the enzymes mediating this modification are significantly altered, usually elevated, in most cancer types. Increased editing activity is found to be associated with patient survival. As is the case with somatic mutations in DNA, most of these newly introduced RNA mutations are likely passengers, but a few may serve as drivers that may be novel candidates for therapeutic and diagnostic purposes.

Induction of polyploidy by nuclear fusion mechanism upon decreased expression of the nuclear envelope protein LAP2ß in the human osteosarcoma cell line U2OS.

BACKGROUND: Polyploidy has been recognized for many years as an important hallmark of cancer cells. Polyploid cells can arise through cell fusion, endoreplication and abortive cell cycle. The inner nuclear membrane protein LAP2ß plays key roles in nuclear envelope breakdown and reassembly during mitosis, initiation of replication and transcriptional repression. Here we studied the function of LAP2ß in the maintenance of cell ploidy state, a role which has not yet been assigned to this protein. RESULTS: By knocking down the expression of LAP2ß, using both viral and non-viral RNAi approaches in osteosarcoma derived U2OS cells, we detected enlarged nuclear size, nearly doubling of DNA content and chromosomal duplications, as analyzed by fluorescent in situ hybridization and spectral karyotyping methodologies. Spectral karyotyping analyses revealed that near-hexaploid karyotypes of LAP2ß knocked down cells consisted of not only seven duplicated chromosomal markers, as could be anticipated by genome duplication mechanism, but also of four single chromosomal markers. Furthermore, spectral karyotyping analysis revealed that both of two near-triploid U2OS sub-clones contained the seven markers that were duplicated in LAP2ß knocked down cells, whereas the four single chromosomal markers were detected only in one of them. Gene expression profiling of LAP2ß knocked down cells revealed that up to a third of the genes exhibiting significant changes in their expression are involved in cancer progression. CONCLUSIONS: Our results suggest that nuclear fusion mechanism underlies the polyploidization induction upon LAP2ß reduced expression. Our study implies on a novel role of LAP2ß in the maintenance of cell ploidy status. LAP2ß depleted U2OS cells can serve as a model to investigate polyploidy and aneuploidy formation by nuclear fusion mechanism and its involvement in cancerogenesis.

Consistent levels of A-to-I RNA editing across individuals in coding sequences and non-conserved Alu repeats.

BACKGROUND: Adenosine to inosine (A-to-I) RNA-editing is an essential post-transcriptional mechanism that occurs in numerous sites in the human transcriptome, mainly within Alu repeats. It has been shown to have consistent levels of editing across individuals in a few targets in the human brain and altered in several human pathologies. However, the variability across human individuals of editing levels in other tissues has not been studied so far. RESULTS: Here, we analyzed 32 skin samples, looking at A-to-I editing level in three genes within coding sequences and in the Alu repeats of six different genes. We observed highly consistent editing levels across different individuals as well as across tissues, not only in coding targets but, surprisingly, also in the non evolutionary conserved Alu repeats. CONCLUSIONS: Our findings suggest that A-to-I RNA-editing of Alu elements is a tightly regulated process and, as such, might have been recruited in the course of primate evolution for post-transcriptional regulatory mechanisms.

Alu sequences in undifferentiated human embryonic stem cells display high levels of A-to-I RNA editing.

Adenosine to Inosine (A-to-I) RNA editing is a site-specific modification of RNA transcripts, catalyzed by members of the ADAR (Adenosine Deaminase Acting on RNA) protein family. RNA editing occurs in human RNA in thousands of different sites. Some of the sites are located in protein-coding regions but the majority is found in non-coding regions, such as 3'UTRs, 5'UTRs and introns - mainly in Alu elements. While editing is found in all tissues, the highest levels of editing are found in the brain. It was shown that editing levels within protein-coding regions are increased during embryogenesis and after birth and that RNA editing is crucial for organism viability as well as for normal development. In this study we characterized the A-to-I RNA editing phenomenon during neuronal and spontaneous differentiation of human embryonic stem cells (hESCs). We identified high editing levels of Alu repetitive elements in hESCs and demonstrated a global decrease in editing levels of non-coding Alu sites when hESCs are differentiating, particularly into the neural lineage. Using RNA interference, we showed that the elevated editing levels of Alu elements in undifferentiated hESCs are highly dependent on ADAR1. DNA microarray analysis showed that ADAR1 knockdown has a global effect on gene expression in hESCs and leads to a significant increase in RNA expression levels of genes involved in differentiation and development processes, including neurogenesis. Taken together, we speculate that A-to-I editing of Alu sequences plays a role in the regulation of hESC early differentiation decisions.

Adenosine-to-inosine RNA editing shapes transcriptome diversity in primates.

Human and chimpanzee genomes are almost identical, yet humans express higher brain capabilities. Deciphering the basis for this superiority is a long sought-after challenge. Adenosine-to-inosine (A-to-I) RNA editing is a widespread modification of the transcriptome. The editing level in humans is significantly higher compared with nonprimates, due to exceptional editing within the primate-specific Alu sequences, but the global editing level of nonhuman primates has not been studied so far. Here we report the sequencing of transcribed Alu sequences in humans, chimpanzees, and rhesus monkeys. We found that, on average, the editing level in the transcripts analyzed is higher in human brain compared with nonhuman primates, even where the genomic Alu structure is unmodified. Correlated editing is observed for pairs and triplets of specific adenosines along the Alu sequences. Moreover, new editable species-specific Alu insertions, subsequent to the human-chimpanzee split, are significantly enriched in genes related to neuronal functions and neurological diseases. The enhanced editing level in the human brain and the association with neuronal functions both hint at the possible contribution of A-to-I editing to the development of higher brain function. We show here that combinatorial editing is the most significant contributor to the transcriptome repertoire and suggest that Alu editing adapted by natural selection may therefore serve as an alternate information mechanism based on the binary A/I code.