Off-target effect of the BMI1 inhibitor PTC596 drives epithelial-mesenchymal transition in glioblastoma multiforme

Dual-target EZH2 inhibitor: latest advances in medicinal chemistry

Enhancer of zeste homolog 2 (EZH2), a histone methyltransferase, plays a crucial role in tumor progression by regulating gene expression. EZH2 inhibitors have emerged as promising anti-tumor agents due to their potential in cancer treatment strategies. However, single-target inhibitors often face limitations such as drug resistance and side effects. Dual-target inhibitors, exemplified by EZH1/2 inhibitor HH-2853(28), offer enhanced efficacy and reduced adverse effects. This review highlights recent advancements in dual inhibitors targeting EZH2 and other proteins like BRD4, PARP1, and EHMT2, emphasizing rational design, structure-activity relationships, and safety profiles, suggesting their potential in clinical applications.

Diagnostic accuracy of ESR1 mutation detection by cell-free DNA in breast cancer: a systematic review and meta-analysis of diagnostic test accuracy

BACKGROUND

Estrogen receptors express in nearly 70% of breast cancers (ER-positive). Estrogen receptor alpha plays a fundamental role as a significant factor in breast cancer progression for the early selection of therapeutic approaches. Accordingly, there has been a surge of attention to non-invasive techniques, including circulating Cell-free DNA (ccfDNA) or Cell-Free DNA (cfDNA), to detect and track ESR1 genotype. Therefore, this study aimed to examine the diagnosis accuracy of ESR1 mutation detection by cell-free DNA in breast cancer patientsthrough a systematic review and comprehensive meta-analysis.

METHODS

PubMed, Embase, and Web of Science databases were searched up to 6 April 2022. Diagnostic studies on ESR1 measurement by cfDNA, which was confirmed using the tumour tissue biopsy, have been included in the study. The sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR) and negative likelihood ratio (NLR) were considered to analyse the data.

RESULTS

Out of 649 papers, 13 papers with 15 cohorts, including 389 participants, entered the meta-analyses. The comprehensive meta-analysis indicated a high sensitivity (75.52, 95% CI 60.19-90.85), specificity (88.20, 95% CI 80.99-95.40), and high accuracy of 88.96 (95% CI 83.23-94.69) for plasma ESR1. We also found a moderate PPV of 56.94 (95% CI 41.70-72.18) but a high NPV of 88.53 (95% CI 82.61-94.44). We also found an NLR of 0.443 (95% CI 0.09-0.79) and PLR of 1.60 (95% CI 1.20-1.99).

CONCLUSION

This systematic review and comprehensive meta-analysis reveal that plasma cfDNA testing exhibits high sensitivity and specificity in detecting ESR1 mutations in breast cancer patients. This suggests that the test could be a valuable diagnostic tool. It may serve as a dependable and non-invasive technique for identifying ESR1 mutations in breast cancer patients. However, more extensive research is needed to confirm its prognostic value.

Prediction of DNA methylation-based tumor types from histopathology in central nervous system tumors with deep learning

Precision in the diagnosis of diverse central nervous system (CNS) tumor types is crucial for optimal treatment. DNA methylation profiles, which capture the methylation status of thousands of individual CpG sites, are state-of-the-art data-driven means to enhance diagnostic accuracy but are also time consuming and not widely available. Here, to address these limitations, we developed Deep lEarning from histoPathoLOgy and methYlation (DEPLOY), a deep learning model that classifies CNS tumors to ten major categories from histopathology. DEPLOY integrates three distinct components: the first classifies CNS tumors directly from slide images ('direct model'), the second initially generates predictions for DNA methylation beta values, which are subsequently used for tumor classification ('indirect model'), and the third classifies tumor types directly from routinely available patient demographics. First, we find that DEPLOY accurately predicts beta values from histopathology images. Second, using a ten-class model trained on an internal dataset of 1,796 patients, we predict the tumor categories in three independent external test datasets including 2,156 patients, achieving an overall accuracy of 95% and balanced accuracy of 91% on samples that are predicted with high confidence. These results showcase the potential future use of DEPLOY to assist pathologists in diagnosing CNS tumors within a clinically relevant short time frame.

Fractionated radiation therapy alters energy metabolism and induces cellular quiescence exit in patient-derived orthotopic xenograft models of high-grade glioma

Radiation is one of the standard therapies for pediatric high-grade glioma (pHGG), of which the prognosis remains poor. To gain an in-depth understanding of biological consequences beyond the classic DNA damage, we treated 9 patient-derived orthotopic xenograft (PDOX) models, including one with DNA mismatch repair (MMR) deficiency, with fractionated radiations (2 Gy/day x 5 days). Extension of survival time was noted in 5 PDOX models (P < 0.05) accompanied by γH2AX positivity in >95 % tumor cells in tumor core and >85 % in the invasive foci as well as ∼30 % apoptotic and mitotic catastrophic cell death. The model with DNA MMR (IC-1406HGG) was the most responsive to radiation with a reduction of Ki-67(+) cells. Altered metabolism, including mitochondria number elevation, COX IV activation and reactive oxygen species accumulation, were detected together with the enrichment of CD133+ tumor cells. The latter was caused by the entry of quiescent G0 cells into cell cycle and the activation of self-renewal (SOX2 and BMI1) and epithelial mesenchymal transition (fibronectin) genes. These novel insights about the cellular and molecular mechanisms of fractionated radiation in vivo should support the development of new radio-sensitizing therapies.

3D stem-like spheroids-on-a-chip for personalized combinatorial drug testing in oral cancer

BACKGROUND

Functional drug testing (FDT) with patient-derived tumor cells in microfluidic devices is gaining popularity. However, the majority of previously reported microfluidic devices for FDT were limited by at least one of these factors: lengthy fabrication procedures, absence of tumor progenitor cells, lack of clinical correlation, and mono-drug therapy testing. Furthermore, personalized microfluidic models based on spheroids derived from oral cancer patients remain to be thoroughly validated. Overcoming the limitations, we develop 3D printed mold-based, dynamic, and personalized oral stem-like spheroids-on-a-chip, featuring unique serpentine loops and flat-bottom microwells arrangement.

RESULTS

This unique arrangement enables the screening of seven combinations of three drugs on chemoresistive cancer stem-like cells. Oral cancer patients-derived stem-like spheroids (CD 44+) remains highly viable (> 90%) for 5 days. Treatment with a well-known oral cancer chemotherapy regimen (paclitaxel, 5 fluorouracil, and cisplatin) at clinically relevant dosages results in heterogeneous drug responses in spheroids. These spheroids are derived from three oral cancer patients, each diagnosed with either well-differentiated or moderately-differentiated squamous cell carcinoma. Oral spheroids exhibit dissimilar morphology, size, and oral tumor-relevant oxygen levels (< 5% O2). These features correlate with the drug responses and clinical diagnosis from each patient's histopathological report.

CONCLUSIONS

Overall, we demonstrate the influence of tumor differentiation status on treatment responses, which has been rarely carried out in the previous reports. To the best of our knowledge, this is the first report demonstrating extensive work on development of microfluidic based oral cancer spheroid model for personalized combinatorial drug screening. Furthermore, the obtained clinical correlation of drug screening data represents a significant advancement over previously reported personalized spheroid-based microfluidic devices. Finally, the maintenance of patient-derived spheroids with high viability under oral cancer relevant oxygen levels of less than 5% O2 is a more realistic representation of solid tumor microenvironment in our developed device.

Artificial intelligence in digital pathology: a systematic review and meta-analysis of diagnostic test accuracy

Ensuring diagnostic performance of artificial intelligence (AI) before introduction into clinical practice is essential. Growing numbers of studies using AI for digital pathology have been reported over recent years. The aim of this work is to examine the diagnostic accuracy of AI in digital pathology images for any disease. This systematic review and meta-analysis included diagnostic accuracy studies using any type of AI applied to whole slide images (WSIs) for any disease. The reference standard was diagnosis by histopathological assessment and/or immunohistochemistry. Searches were conducted in PubMed, EMBASE and CENTRAL in June 2022. Risk of bias and concerns of applicability were assessed using the QUADAS-2 tool. Data extraction was conducted by two investigators and meta-analysis was performed using a bivariate random effects model, with additional subgroup analyses also performed. Of 2976 identified studies, 100 were included in the review and 48 in the meta-analysis. Studies were from a range of countries, including over 152,000 whole slide images (WSIs), representing many diseases. These studies reported a mean sensitivity of 96.3% (CI 94.1-97.7) and mean specificity of 93.3% (CI 90.5-95.4). There was heterogeneity in study design and 99% of studies identified for inclusion had at least one area at high or unclear risk of bias or applicability concerns. Details on selection of cases, division of model development and validation data and raw performance data were frequently ambiguous or missing. AI is reported as having high diagnostic accuracy in the reported areas but requires more rigorous evaluation of its performance.

3D magnetic seed localization for augmented reality in surgery

PURPOSE

For tumor resection, surgeons need to localize the tumor. For this purpose, a magnetic seed can be inserted into the tumor by a radiologist and, during surgery, a magnetic detection probe informs the distance to the seed for localization. In this case, the surgeon still needs to mentally reconstruct the position of the tumor from the probe's information. The purpose of this study is to develop and assess a method for 3D localization and visualization of the seed, facilitating the localization of the tumor.

METHODS

We propose a method for 3D localization of the magnetic seed by extending the magnetic detection probe with a tracking-based localization. We attach a position sensor (QR-code or optical marker) to the probe in order to track its 3D pose (respectively, using a head-mounted display with a camera or optical tracker). Following an acquisition protocol, the 3D probe tip and seed position are subsequently obtained by solving a system of equations based on the distances and the 3D probe poses.

RESULTS

The method was evaluated with an optical tracking system. An experimental setup using QR-code tracking (resp. using an optical marker) achieves an average of 1.6 mm (resp. 0.8 mm) 3D distance between the localized seed and the ground truth. Using a breast phantom setup, the average 3D distance is 4.7 mm with a QR-code and 2.1 mm with an optical marker.

CONCLUSION

Tracking the magnetic detection probe allows 3D localization of a magnetic seed, which opens doors for augmented reality target visualization during surgery. Such an approach should enhance the perception of the localized region of interest during the intervention, especially for breast tumor resection where magnetic seeds can already be used in the protocol.

The progressive trend of modeling and drug screening systems of breast cancer bone metastasis

Bone metastasis is considered as a considerable challenge for breast cancer patients. Various in vitro and in vivo models have been developed to examine this occurrence. In vitro models are employed to simulate the intricate tumor microenvironment, investigate the interplay between cells and their adjacent microenvironment, and evaluate the effectiveness of therapeutic interventions for tumors. The endeavor to replicate the latency period of bone metastasis in animal models has presented a challenge, primarily due to the necessity of primary tumor removal and the presence of multiple potential metastatic sites.The utilization of novel bone metastasis models, including three-dimensional (3D) models, has been proposed as a promising approach to overcome the constraints associated with conventional 2D and animal models. However, existing 3D models are limited by various factors, such as irregular cellular proliferation, autofluorescence, and changes in genetic and epigenetic expression. The imperative for the advancement of future applications of 3D models lies in their standardization and automation. The utilization of artificial intelligence exhibits the capability to predict cellular behavior through the examination of substrate materials' chemical composition, geometry, and mechanical performance. The implementation of these algorithms possesses the capability to predict the progression and proliferation of cancer. This paper reviewed the mechanisms of bone metastasis following primary breast cancer. Current models of breast cancer bone metastasis, along with their challenges, as well as the future perspectives of using these models for translational drug development, were discussed.

Transcriptional signals of transformation in human cancer

BACKGROUND

As normal cells transform into cancers, their cell state changes, which may drive cancer cells into a stem-like or more primordial, foetal, or embryonic cell state. The transcriptomic profile of this final state may encode information about cancer's origin and how cancers relate to their normal cell counterparts.

METHODS

Here, we used single-cell atlases to study cancer transformation in transcriptional terms. We utilised bulk transcriptomes across a wide spectrum of adult and childhood cancers, using a previously established method to interrogate their relationship to normal cell states. We extend and validate these findings using single-cell cancer transcriptomes and organ-specific atlases of colorectal and liver cancer.

RESULTS

Our bulk transcriptomic data reveals that adult cancers rarely return to an embryonic state, but that a foetal state is a near-universal feature of childhood cancers. This finding was confirmed with single-cell cancer transcriptomes.

CONCLUSIONS

Our findings provide a nuanced picture of transformation in human cancer, indicating cancer-specific rather than universal patterns of transformation pervade adult epithelial cancers.

Identification of hub genes to determine drug-disease correlation in breast carcinomas

The exact molecular mechanism underlying the heterogeneous drug response against breast carcinoma remains to be fully understood. It is urgently required to identify key genes that are intricately associated with varied clinical response of standard anti-cancer drugs, clinically used to treat breast cancer patients. In the present study, the utility of transcriptomic data of breast cancer patients in discerning the clinical drug response using machine learning-based approaches were evaluated. Here, a computational framework has been developed which can be used to identify key genes that can be linked with clinical drug response and progression of cancer, offering an immense opportunity to predict potential prognostic biomarkers and therapeutic targets. The framework concerned utilizes DeSeq2, Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Cytoscape, and machine learning techniques to find these crucial genes. Total RNA extraction and qRT-PCR were performed to quantify relative expression of few hub genes selected from the networks. In our study, we have experimentally checked the expression of few key hub genes like APOA2, DLX5, APOC3, CAMK2B, and PAK6 that were predicted to play an immense role in breast cancer tumorigenesis and progression in response to anti-cancer drug Paclitaxel. However, further experimental validations will be required to get mechanistic insights of these genes in regulating the drug response and cancer progression which will likely to play pivotal role in cancer treatment and precision oncology.

A fully 3D-printed versatile tumor-on-a-chip allows multi-drug screening and correlation with clinical outcomes for personalized medicine

Optimal clinical outcomes in cancer treatments could be achieved through the development of reliable, precise ex vivo tumor models that function as drug screening platforms for patient-targeted therapies. Microfluidic tumor-on-chip technology is emerging as a preferred tool since it enables the complex set-ups and recapitulation of the physiologically relevant physical microenvironment of tumors. In order to overcome the common hindrances encountered while using this technology, a fully 3D-printed device was developed that sustains patient-derived multicellular spheroids long enough to conduct multiple drug screening tests. This tool is both cost effective and possesses four necessary characteristics of effective microfluidic devices: transparency, biocompatibility, versatility, and sample accessibility. Compelling correlations which demonstrate a clinical proof of concept were found after testing and comparing different chemotherapies on tumor spheroids, derived from ten patients, to their clinical outcomes. This platform offers a potential solution for personalized medicine by functioning as a predictive drug-performance tool.

Development of microRNA-based therapeutics for central nervous system diseases

MicroRNA (miRNA)-mediated gene silencing is a method of RNA interference in which a miRNA binds to messenger RNA sequences and regulates target gene expression. MiRNA-based therapeutics have shown promise in treating a variety of central nervous system diseases, as verified by results from diverse preclinical model organisms. Over the last decade, several miRNA-based therapeutics have entered clinical trials for various kinds of diseases, such as tumors, infections, and inherited diseases. However, such clinical trials for central nervous system diseases are scarce, and many central nervous system diseases, including hemorrhagic stroke, ischemic stroke, traumatic brain injury, intractable epilepsy, and Alzheimer's disease, lack effective treatment. Considering its effectiveness for central nervous system diseases in preclinical experiments, microRNA-based intervention may serve as a promising treatment for these kinds of diseases. This paper reviews basic principles and recent progress of miRNA-based therapeutics and summarizes general procedures to develop such therapeutics for treating central nervous system diseases. Then, the current obstacles in drug development are discussed. This review also provides a new perspective on possible solutions to these obstacles in the future.

Identification of cancer types from gene expressions using learning techniques

Tumor is the major cause of death all around the world in recent days. Early detection and prediction of a cancer type are important for a patient's well-being. Functional genomic data has recently been used in the effective and early detection of cancer. According to previous research, the use of microarray data in cancer prediction has evidenced two main problems as high dimensionality and limited sample size. Several researchers have used numerous statistical and machine learning-based methods to classify cancer types but still, limitations are there which makes cancer classification a difficult job. Deep Learning (DL) and Convolutional Neural Networks (CNN) have been proven with effective analyses of unstructured data including gene expression data. In the proposed method gene expression data for five types of cancer is collected from The Cancer Genome Atlas (TCGA). Prominent features are selected using a hybrid Particle Swarm Optimization (PSO) and Random Forest (RF) algorithm followed by the use of Principal Component Analysis (PCA) for dimensionality reduction. Finally, for classification blend of Convolutional Neural Network (CNN) and Bi-directional Long Short Term Memory (Bi-LSTM) is used to predict the target type of cancer. Experimental results demonstrate that accuracy of the proposed method is 96.89%. As compared to existing work, our method outperformed with better results.

Prediction of cancer treatment response from histopathology images through imputed transcriptomics

Advances in artificial intelligence have paved the way for leveraging hematoxylin and eosin (H&E)-stained tumor slides for precision oncology. We present ENLIGHT-DeepPT, an approach for predicting response to multiple targeted and immunotherapies from H&E-slides. In difference from existing approaches that aim to predict treatment response directly from the slides, ENLIGHT-DeepPT is an indirect two-step approach consisting of (1) DeepPT, a new deep-learning framework that predicts genome-wide tumor mRNA expression from slides, and (2) ENLIGHT, which predicts response based on the DeepPT inferred expression values. DeepPT successfully predicts transcriptomics in all 16 TCGA cohorts tested and generalizes well to two independent datasets. Our key contribution is showing that ENLIGHT-DeepPT successfully predicts true responders in five independent patients' cohorts involving four different treatments spanning six cancer types with an overall odds ratio of 2.44, increasing the baseline response rate by 43.47% among predicted responders, without the need for any treatment data for training. Furthermore, its prediction accuracy on these datasets is comparable to a supervised approach predicting the response directly from the images, which needs to be trained and tested on the same cohort. ENLIGHT-DeepPT future application could provide clinicians with rapid treatment recommendations to an array of different therapies and importantly, may contribute to advancing precision oncology in developing countries.

MSDRP: a deep learning model based on multisource data for predicting drug response

MOTIVATION

Cancer heterogeneity drastically affects cancer therapeutic outcomes. Predicting drug response in vitro is expected to help formulate personalized therapy regimens. In recent years, several computational models based on machine learning and deep learning have been proposed to predict drug response in vitro. However, most of these methods capture drug features based on a single drug description (e.g. drug structure), without considering the relationships between drugs and biological entities (e.g. target, diseases, and side effects). Moreover, most of these methods collect features separately for drugs and cell lines but fail to consider the pairwise interactions between drugs and cell lines.

RESULTS

In this paper, we propose a deep learning framework, named MSDRP for drug response prediction. MSDRP uses an interaction module to capture interactions between drugs and cell lines, and integrates multiple associations/interactions between drugs and biological entities through similarity network fusion algorithms, outperforming some state-of-the-art models in all performance measures for all experiments. The experimental results of de novo test and independent test demonstrate the excellent performance of our model for new drugs. Furthermore, several case studies illustrate the rationality for using feature vectors derived from drug similarity matrices from multisource data to represent drugs and the interpretability of our model.

AVAILABILITY AND IMPLEMENTATION

The codes of MSDRP are available at https://github.com/xyzhang-10/MSDRP.

Whole Slide Imaging-Based Prediction of TP53 Mutations Identifies an Aggressive Disease Phenotype in Prostate Cancer

In prostate cancer, there is an urgent need for objective prognostic biomarkers that identify the metastatic potential of a tumor at an early stage. While recent analyses indicated TP53 mutations as candidate biomarkers, molecular profiling in a clinical setting is complicated by tumor heterogeneity. Deep learning models that predict the spatial presence of TP53 mutations in whole slide images (WSI) offer the potential to mitigate this issue. To assess the potential of WSIs as proxies for spatially resolved profiling and as biomarkers for aggressive disease, we developed TiDo, a deep learning model that achieves state-of-the-art performance in predicting TP53 mutations from WSIs of primary prostate tumors. In an independent multifocal cohort, the model showed successful generalization at both the patient and lesion level. Analysis of model predictions revealed that false positive (FP) predictions could at least partially be explained by TP53 deletions, suggesting that some FP carry an alteration that leads to the same histological phenotype as TP53 mutations. Comparative expression and histologic cell type analyses identified a TP53-like cellular phenotype triggered by expression of pathways affecting stromal composition. Together, these findings indicate that WSI-based models might not be able to perfectly predict the spatial presence of individual TP53 mutations but they have the potential to elucidate the prognosis of a tumor by depicting a downstream phenotype associated with aggressive disease biomarkers.

SIGNIFICANCE

Deep learning models predicting TP53 mutations from whole slide images of prostate cancer capture histologic phenotypes associated with stromal composition, lymph node metastasis, and biochemical recurrence, indicating their potential as in silico prognostic biomarkers. See related commentary by Bordeleau, p. 2809.

Cobind: quantitative analysis of the genomic overlaps

Motivation

Analyzing the overlap between two sets of genomic intervals is a frequent task in the field of bioinformatics. Typically, this is accomplished by counting the number (or proportion) of overlapped regions, which applies an arbitrary threshold to determine if two genomic intervals are overlapped. By making binary calls but disregarding the magnitude of the overlap, such an approach often leads to biased, non-reproducible, and incomparable results.

Results

We developed the cobind package, which incorporates six statistical measures: the Jaccard coefficient, Sørensen-Dice coefficient, Szymkiewicz-Simpson coefficient, collocation coefficient, pointwise mutual information (PMI), and normalized PMI. These measures allow for a quantitative assessment of the collocation strength between two sets of genomic intervals. To demonstrate the effectiveness of these methods, we applied them to analyze CTCF's binding sites identified from ChIP-seq, cancer-specific open-chromatin regions (OCRs) identified from ATAC-seq of 17 cancer types, and oligodendrocytes-specific OCRs identified from scATAC-seq. Our results indicated that these new approaches effectively re-discover CTCF's cofactors, as well as cancer-specific and oligodendrocytes-specific master regulators implicated in disease and cell type development.

Availability and implementation

The cobind package is implemented in Python and freely available at https://cobind.readthedocs.io/en/latest/.

Breast cancer brain metastasis: from etiology to state-of-the-art modeling

Currently, breast carcinoma is the most common form of malignancy and the main cause of cancer mortality in women worldwide. The metastasis of cancer cells from the primary tumor site to other organs in the body, notably the lungs, bones, brain, and liver, is what causes breast cancer to ultimately be fatal. Brain metastases occur in as many as 30% of patients with advanced breast cancer, and the 1-year survival rate of these patients is around 20%. Many researchers have focused on brain metastasis, but due to its complexities, many aspects of this process are still relatively unclear. To develop and test novel therapies for this fatal condition, pre-clinical models are required that can mimic the biological processes involved in breast cancer brain metastasis (BCBM). The application of many breakthroughs in the area of tissue engineering has resulted in the development of scaffold or matrix-based culture methods that more accurately imitate the original extracellular matrix (ECM) of metastatic tumors. Furthermore, specific cell lines are now being used to create three-dimensional (3D) cultures that can be used to model metastasis. These 3D cultures satisfy the requirement for in vitro methodologies that allow for a more accurate investigation of the molecular pathways as well as a more in-depth examination of the effects of the medication being tested. In this review, we talk about the latest advances in modeling BCBM using cell lines, animals, and tissue engineering methods.

Image-guided cancer surgery: a narrative review on imaging modalities and emerging nanotechnology strategies

Surgical resection is the cornerstone of solid tumour treatment. Current techniques for evaluating margin statuses, such as frozen section, imprint cytology, and intraoperative ultrasound, are helpful. However, an intraoperative assessment of tumour margins that is accurate and safe is clinically necessary. Positive surgical margins (PSM) have a well-documented negative effect on treatment outcomes and survival. As a result, surgical tumour imaging methods are now a practical method for reducing PSM rates and improving the efficiency of debulking surgery. Because of their unique characteristics, nanoparticles can function as contrast agents in image-guided surgery. While most image-guided surgical applications utilizing nanotechnology are now in the preclinical stage, some are beginning to reach the clinical phase. Here, we list the various imaging techniques used in image-guided surgery, such as optical imaging, ultrasound, computed tomography, magnetic resonance imaging, nuclear medicine imaging, and the most current developments in the potential of nanotechnology to detect surgical malignancies. In the coming years, we will see the evolution of nanoparticles tailored to specific tumour types and the introduction of surgical equipment to improve resection accuracy. Although the promise of nanotechnology for producing exogenous molecular contrast agents has been clearly demonstrated, much work remains to be done to put it into practice.

Dedifferentiated cells obtained from glioblastoma cell lines are an easy and robust model for mesenchymal glioblastoma stem cells studies

Glioblastoma is an aggressive brain tumor with a poor prognosis. Glioblastoma Stem Cells (GSC) are involved in glioblastoma resistance and relapse. Effective glioblastoma treatment must include GSC targeting strategy. Robust and well defined in vitroGSC models are required for new therapies evaluation. In this study, we extensively characterized 4 GSC models obtained by dedifferentiation of commercially available glioblastoma cell lines and compared them to 2 established patient derived GSC lines (Brain Tumor Initiating Cells). Dedifferentiated cells formed gliospheres, typical for GSC, with self-renewal ability. Gene expression and protein analysis revealed an increased expression of several stemness associated markers such as A2B5, integrin α6, Nestin, SOX2 and NANOG. Cells were oriented toward a mesenchymal GSC phenotype as shown by elevated levels of mesenchymal and EMT related markers (CD44, FN1, integrin α5). Dedifferentiated GSC were similar to BTIC in terms of size and heterogeneity. The characterization study also revealed that CXCR4 pathway was activated by dedifferentiation, emphasizing its role as a potential therapeutic target. The expression of resistance-associated markers and the phenotypic diversity of the 4 GSC models obtained by dedifferentiation make them relevant to challenge future GSC targeting therapies.

Post-acute sequelae of COVID-19 is characterized by diminished peripheral CD8+β7 integrin+ T cells and anti-SARS-CoV-2 IgA response

Several millions of individuals are estimated to develop post-acute sequelae SARS-CoV-2 condition (PASC) that persists for months after infection. Here we evaluate the immune response in convalescent individuals with PASC compared to convalescent asymptomatic and uninfected participants, six months following their COVID-19 diagnosis. Both convalescent asymptomatic and PASC cases are characterised by higher CD8⁺ T cell percentages, however, the proportion of blood CD8⁺ T cells expressing the mucosal homing receptor β7 is low in PASC patients. CD8 T cells show increased expression of PD-1, perforin and granzyme B in PASC, and the plasma levels of type I and type III (mucosal) interferons are elevated. The humoral response is characterized by higher levels of IgA against the N and S viral proteins, particularly in those individuals who had severe acute disease. Our results also show that consistently elevated levels of IL-6, IL-8/CXCL8 and IP-10/CXCL10 during acute disease increase the risk to develop PASC. In summary, our study indicates that PASC is defined by persisting immunological dysfunction as late as six months following SARS-CoV-2 infection, including alterations in mucosal immune parameters, redistribution of mucosal CD8⁺β7Integrin⁺ T cells and IgA, indicative of potential viral persistence and mucosal involvement in the etiopathology of PASC.

Prediction model for drug response of acute myeloid leukemia patients

Despite some encouraging successes, predicting the therapy response of acute myeloid leukemia (AML) patients remains highly challenging due to tumor heterogeneity. Here we aim to develop and validate MDREAM, a robust ensemble-based prediction model for drug response in AML based on an integration of omics data, including mutations and gene expression, and large-scale drug testing. Briefly, MDREAM is first trained in the BeatAML cohort (n = 278), and then validated in the BeatAML (n = 183) and two external cohorts, including a Swedish AML cohort (n = 45) and a relapsed/refractory acute leukemia cohort (n = 12). The final prediction is based on 122 ensemble models, each corresponding to a drug. A confidence score metric is used to convey the uncertainty of predictions; among predictions with a confidence score >0.75, the validated proportion of good responders is 77%. The Spearman correlations between the predicted and the observed drug response are 0.68 (95% CI: [0.64, 0.68]) in the BeatAML validation set, -0.49 (95% CI: [-0.53, -0.44]) in the Swedish cohort and 0.59 (95% CI: [0.51, 0.67]) in the relapsed/refractory cohort. A web-based implementation of MDREAM is publicly available at https://www.meb.ki.se/shiny/truvu/MDREAM/ .

A Physiologically Responsive Nanocomposite Hydrogel for Treatment of Head and Neck Squamous Cell Carcinoma via Proteolysis-Targeting Chimeras Enhanced Immunotherapy

Although immunotherapy has revolutionized oncotherapy, only ≈15% of head and neck squamous cell carcinoma (HNSCC) patients benefit from the current therapies. An immunosuppressive tumor microenvironment (TME) and dysregulation of the polycomb ring finger oncogene BMI1 are potential reasons for the failure. Herein, to promote immunotherapeutic efficacy against HNSCC, an injectable nanocomposite hydrogel is developed with a polymer framework (PLGA-PEG-PLGA) that is loaded with both imiquimod encapsulated CaCO₃ nanoparticles (RC) and cancer cell membrane (CCM)-coated mesoporous silica nanoparticles containing a peptide-based proteolysis-targeting chimeras (PROTAC) for BMI1 and paclitaxel (PepM@PacC). Upon injection, this nanocomposite hydrogel undergoes in situ gelation, after which it degrades in the TME over time, releasing RC and PepM@PacC nanoparticles to respectively perform immunotherapy and chemotherapy. Specifically, the RC particles selectively manipulate tumor-associated macrophages and dendritic cells to activate a T-cell immune response, while CCM-mediated homologous targeting and endocytosis delivers the PepM@PacC particles into cancer cells, where endogenous glutathione promotes disulfide bond cleavage to release the PROTAC peptide for BMI1 degradation and frees the paclitaxel from the particle pores to elicit apoptosis meanwhile enhance immunotherapy. Thus, the nanocomposite hydrogel, which is designed to exploit multiple known vulnerabilities of HNSCC, succeeds in suppressing both growth and metastasis of HNSCC.

Identification of therapeutic sensitivities in a spheroid drug combination screen of Neurofibromatosis Type I associated High Grade Gliomas

Neurofibromatosis Type 1 (NF1) patients develop an array of benign and malignant tumors, of which Malignant Peripheral Nerve Sheath Tumors (MPNST) and High Grade Gliomas (HGG) have a dismal prognosis. About 15-20% of individuals with NF1 develop brain tumors and one third of these occur outside of the optic pathway. These non-optic pathway gliomas are more likely to progress to malignancy, especially in adults. Despite their low frequency, high grade gliomas have a disproportional effect on the morbidity of NF1 patients. In vitro drug combination screens have not been performed on NF1-associated HGG, hindering our ability to develop informed clinical trials. Here we present the first in vitro drug combination screen (21 compounds alone or in combination with MEK or PI3K inhibitors) on the only human NF1 patient derived HGG cell line available and on three mouse glioma cell lines derived from the NF1-P53 genetically engineered mouse model, which sporadically develop HGG. These mouse glioma cell lines were never exposed to serum, grow as spheres and express markers that are consistent with an Oligodendrocyte Precursor Cell (OPC) lineage origin. Importantly, even though the true cell of origin for HGG remains elusive, they are thought to arise from the OPC lineage. We evaluated drug sensitivities of the three murine glioma cell lines in a 3D spheroid growth assay, which more accurately reflects drug sensitivities in vivo. Excitingly, we identified six compounds targeting HDACs, BRD4, CHEK1, BMI-1, CDK1/2/5/9, and the proteasome that potently induced cell death in our NF1-associated HGG. Moreover, several of these inhibitors work synergistically with either MEK or PI3K inhibitors. This study forms the basis for further pre-clinical evaluation of promising targets, with an eventual hope to translate these to the clinic.

The Role of Epigenetics in Brain and Spinal Cord Tumors

Identification of distinct genetic and epigenetic profiles in various neuroepithelial tumors has improved the classification and uncovered novel diagnostic, prognostic, and predictive molecular biomarkers for improved prediction of treatment response and outcome. Especially, in pediatric high-grade brain tumors, such as diffuse midline glioma, H3K27M-altered and posterior fossa group A-ependymoma, epigenetic changes predominate, along with changes in expression of known oncogenes and tumor suppressor genes induced by histone modifications and DNA methylation. The precise role of epigenetic abnormalities is important for understanding tumorigenesis and the establishment of brain tumor treatment strategies. Using powerful epigenetic-based therapies for cancer cells, the aberrantly regulated epigenome can be restored to a more normal state through epigenetic reprogramming. Combinations of agents targeting DNA methylation and/or other epigenetic modifications may be a promising cancer treatment. Therefore, the integration of multi-omics data including epigenomics is now important for classifying primary brain tumors and predicting their biological behavior. Recent advances in molecular genetics and epigenetic integrated diagnostics of brain tumors influence new strategies for targeted therapy.

Biofabrication of 3D breast cancer models for dissecting the cytotoxic response of human T cells expressing engineered MAIT cell receptors

Immunotherapy has revolutionized cancer treatment with the advent of advanced cell engineering techniques aimed at targeted therapy with reduced systemic toxicity. However, understanding the underlying immune-cancer interactions require development of advanced three-dimensional (3D) models of human tissues. In this study, we fabricated 3D tumor models with increasing complexity to study the cytotoxic responses of CD8+T cells, genetically engineered to express mucosal-associated invariant T (MAIT) cell receptors, towards MDA-MB-231 breast cancer cells. Homotypic MDA-MB-231 and heterotypic MDA-MB-231/human dermal fibroblast tumor spheroids were primed with precursor MAIT cell ligand 5-amino-6-D-ribitylaminouracil (5-ARU). Engineered T cells effectively eliminated tumors after a 3 d culture period, demonstrating that the engineered T cell receptor recognized major histocompatibility complex class I-related (MR1) protein expressing tumor cells in the presence of 5-ARU. Tumor cell killing efficiency of engineered T cells were also assessed by encapsulating these cells in fibrin, mimicking a tumor extracellular matrix microenvironment. Expression of proinflammatory cytokines such as interferon gamma, interleukin-13, CCL-3 indicated immune cell activation in all tumor models, post immunotherapy. Further, in corroborating the cytotoxic activity, we found that granzymes A and B were also upregulated, in homotypic as well as heterotypic tumors. Finally, a 3D bioprinted tumor model was employed to study the effect of localization of T cells with respect to tumors. T cells bioprinted proximal to the tumor had reduced invasion index and increased cytokine secretion, which indicated a paracrine mode of immune-cancer interaction. Development of 3D tumor-T cell platforms may enable studying the complex immune-cancer interactions and engineering MAIT cells for cell-based cancer immunotherapies.

Deep learning-based tumor microenvironment segmentation is predictive of tumor mutations and patient survival in non-small-cell lung cancer

BACKGROUND

Despite the fact that tumor microenvironment (TME) and gene mutations are the main determinants of progression of the deadliest cancer in the world - lung cancer, their interrelations are not well understood. Digital pathology data provides a unique insight into the spatial composition of the TME. Various spatial metrics and machine learning approaches were proposed for prediction of either patient survival or gene mutations from this data. Still, these approaches are limited in the scope of analyzed features and in their explainability, and as such fail to transfer to clinical practice.

METHODS

Here, we generated 23,199 image patches from 26 hematoxylin-and-eosin (H&E)-stained lung cancer tissue sections and annotated them into 9 different tissue classes. Using this dataset, we trained a deep neural network ARA-CNN. Next, we applied the trained network to segment 467 lung cancer H&E images from The Cancer Genome Atlas (TCGA) database. We used the segmented images to compute human-interpretable features reflecting the heterogeneous composition of the TME, and successfully utilized them to predict patient survival and cancer gene mutations.

RESULTS

We achieved per-class AUC ranging from 0.72 to 0.99 for classifying tissue types in lung cancer with ARA-CNN. Machine learning models trained on the proposed human-interpretable features achieved a c-index of 0.723 in the task of survival prediction and AUC up to 73.5% for PDGFRB in the task of mutation classification.

CONCLUSIONS

We presented a framework that accurately predicted survival and gene mutations in lung adenocarcinoma patients based on human-interpretable features extracted from H&E slides. Our approach can provide important insights for designing novel cancer treatments, by linking the spatial structure of the TME in lung adenocarcinoma to gene mutations and patient survival. It can also expand our understanding of the effects that the TME has on tumor evolutionary processes. Our approach can be generalized to different cancer types to inform precision medicine strategies.

Minimally invasive detection of cancer using metabolic changes in tumor-associated natural killer cells with Oncoimmune probes

Natural Killer (NK) cells, a subset of innate immune cells, undergo cancer-specific changes during tumor progression. Therefore, tracking NK cell activity in circulation has potential for cancer diagnosis. Identification of tumor associated NK cells remains a challenge as most of the cancer antigens are unknown. Here, we introduce tumor-associated circulating NK cell profiling (CNKP) as a stand-alone cancer diagnostic modality with a liquid biopsy. Metabolic profiles of NK cell activation as a result of tumor interaction are detected with a SERS functionalized OncoImmune probe platform. We show that the cancer stem cell-associated NK cell is of value in cancer diagnosis. Through machine learning, the features of NK cell activity in patient blood could identify cancer from non-cancer using 5uL of peripheral blood with 100% accuracy and localization of cancer with 93% accuracy. These results show the feasibility of minimally invasive cancer diagnostics using circulating NK cells.

NK cells can be affected by tumour cells and this difference could be utilised as a cancer diagnostic. Here the authors use a nickel based plasmonic spectroscopy system to measure metabolic differences in NK cells that have been exposed to cancer cells as a method of cancer detection.

A 3D Bioprinted in vitro Model of Neuroblastoma Recapitulates Dynamic Tumor-Endothelial Cell Interactions Contributing to Solid Tumor Aggressive Behavior

Neuroblastoma (NB) is the most common extracranial tumor in children resulting in substantial morbidity and mortality. A deeper understanding of the NB tumor microenvironment (TME) remains an area of active research but there is a lack of reliable and biomimetic experimental models. This study utilizes a 3D bioprinting approach, in combination with NB spheroids, to create an in vitro vascular model of NB for exploring the tumor function within an endothelialized microenvironment. A gelatin methacryloyl (gelMA) bioink is used to create multi-channel cubic tumor analogues with high printing fidelity and mechanical tunability. Human-derived NB spheroids and human umbilical vein endothelial cells (HUVECs) are incorporated into the biomanufactured gelMA and cocultured under static versus dynamic conditions, demonstrating high levels of survival and growth. Quantification of NB-EC integration and tumor cell migration suggested an increased aggressive behavior of NB when cultured in bioprinted endothelialized models, when cocultured with HUVECs, and also as a result of dynamic culture. This model also allowed for the assessment of metabolic, cytokine, and gene expression profiles of NB spheroids under varying TME conditions. These results establish a high throughput research enabling platform to study the TME-mediated cellular-molecular mechanisms of tumor growth, aggression, and response to therapy.

The Crucial Roles of Bmi-1 in Cancer: Implications in Pathogenesis, Metastasis, Drug Resistance, and Targeted Therapies

B-cell-specific Moloney murine leukemia virus integration region 1 (Bmi-1, also known as RNF51 or PCGF4) is one of the important members of the PcG gene family, and is involved in regulating cell proliferation, differentiation and senescence, and maintaining the self-renewal of stem cells. Many studies in recent years have emphasized the role of Bmi-1 in the occurrence and development of tumors. In fact, Bmi-1 has multiple functions in cancer biology and is closely related to many classical molecules, including Akt, c-MYC, Pten, etc. This review summarizes the regulatory mechanisms of Bmi-1 in multiple pathways, and the interaction of Bmi-1 with noncoding RNAs. In particular, we focus on the pathological processes of Bmi-1 in cancer, and explore the clinical relevance of Bmi-1 in cancer biomarkers and prognosis, as well as its implications for chemoresistance and radioresistance. In conclusion, we summarize the role of Bmi-1 in tumor progression, reveal the pathophysiological process and molecular mechanism of Bmi-1 in tumors, and provide useful information for tumor diagnosis, treatment, and prognosis.

STARD5 as a potential clinical target of hepatocellular carcinoma

To reveal whether STARD5 is a potential biomarker for diagnosis and prognosis of HCC. Using gene expression omnibus and the cancer genome atlas (TCGA) to screen differentially expressed genes in HCC and STARD5 was selected by LASSO algorithm. Then, we analyzed the association between STARD5 and clinical characteristics of HCC patients in TCGA and International Cancer Genome Consortium. Meanwhile, the mRNA and protein level of STARD5 was also verified by collecting 87 cases of HCC patients' liver tissues using qRT-PCR and WB. Next, we applied gene set enrichment analysis (GSEA) for pathways analysis of STARD5. Finally, TIMER1.0 and TISIDB were used to explore the correlation of STARD5 with immune cell infiltration. The expression of STARD5 was lower in HCC and negatively correlated with tumor grade (p < 0.05), while high expression of STARD5 suggested a better prognosis for HCC patients (p < 0.01) and it could be an independent prognostic predictor (p < 0.001). Meanwhile, STARD5 also had strong diagnostic accuracy for HCC patients. GSEA revealed that STARD5-related genes were mainly enriched in E2F targets, G2M checkpoint and KRAS signaling. The TIMER1.0 and TISIDB databases found a negative correlation between STARD5 and tumor immune infiltrating cells. STARD5 could be used as a potential target for HCC diagnosis and prognosis.

Comparative analysis of high- and low-level deep learning approaches in microsatellite instability prediction

Deep learning-based approaches in histopathology can be largely divided into two categories: a high-level approach using an end-to-end model and a low-level approach using feature extractors. Although the advantages and disadvantages of both approaches are empirically well known, there exists no scientific basis for choosing a specific approach in research, and direct comparative analysis of the two approaches has rarely been performed. Using the Cancer Genomic Atlas (TCGA)-based dataset, we compared these two different approaches in microsatellite instability (MSI) prediction and analyzed morphological image features associated with MSI. Our high-level approach was based solely on EfficientNet, while our low-level approach relied on LightGBM and multiple deep learning models trained on publicly available multiclass tissue, nuclei, and gland datasets. We compared their performance and important image features. Our high-level approach showed superior performance compared to our low-level approach. In both approaches, debris, lymphocytes, and necrotic cells were revealed as important features of MSI, which is consistent with clinical knowledge. Then, during qualitative analysis, we discovered the weaknesses of our low-level approach and demonstrated that its performance can be improved by using different image features in a complementary way. We performed our study using open-access data, and we believe this study can serve as a useful basis for discovering imaging biomarkers for clinical application.

Traction of 3D and 4D Printing in the Healthcare Industry: From Drug Delivery and Analysis to Regenerative Medicine

Three-dimensional (3D) printing and 3D bioprinting are promising technologies for a broad range of healthcare applications from frontier regenerative medicine and tissue engineering therapies to pharmaceutical advancements yet must overcome the challenges of biocompatibility and resolution. Through comparison of traditional biofabrication methods with 3D (bio)printing, this review highlights the promise of 3D printing for the production of on-demand, personalized, and complex products that enhance the accessibility, effectiveness, and safety of drug therapies and delivery systems. In addition, this review describes the capacity of 3D bioprinting to fabricate patient-specific tissues and living cell systems (e.g., vascular networks, organs, muscles, and skeletal systems) as well as its applications in the delivery of cells and genes, microfluidics, and organ-on-chip constructs. This review summarizes how tailoring selected parameters (i.e., accurately selecting the appropriate printing method, materials, and printing parameters based on the desired application and behavior) can better facilitate the development of optimized 3D-printed products and how dynamic 4D-printed strategies (printing materials designed to change with time or stimulus) may be deployed to overcome many of the inherent limitations of conventional 3D-printed technologies. Comprehensive insights into a critical perspective of the future of 4D bioprinting, crucial requirements for 4D printing including the programmability of a material, multimaterial printing methods, and precise designs for meticulous transformations or even clinical applications are also given.

Overall Survival and Biomarker Analysis of Neoadjuvant Nivolumab Plus Chemotherapy in Operable Stage IIIA Non-Small-Cell Lung Cancer (NADIM phase II trial)

Neoadjuvant chemotherapy plus nivolumab has been shown to be effective in resectable non–small-cell lung cancer (NSCLC) in the NADIM trial (ClinicalTrials.gov identifier: NCT03081689). The 3-year overall survival (OS) and circulating tumor DNA (ctDNA) analysis have not been reported.

Development of a microfluidic platform to maintain viability of micro-dissected tumor slices in culture

One of the issues limiting the development of personalized medicine is the absence of realistic models that reflect the nature and complexity of tumor tissues. We described a new tissue culture approach that combines a microfluidic chip with the microdissected breast cancer tumor. "Tumor-on-a-chip" devices are suitable for precision medicine since the viability of tissue samples is maintained during the culture period by continuously feeding fresh media and eliminating metabolic wastes from the tissue. However, the mass transport of oxygen, which arguably is the most critical nutrient, is rarely assessed. According to our results, transportation of oxygen provides satisfactory in vivo oxygenation within the system. A high level of dissolved oxygen, around 98%-100% for every 24 h, was measurable in the outlet medium. The microfluidic chip system developed within the scope of this study allows living and testing tumor tissues under laboratory conditions. In this study, tumors were generated in CD-1 mice using MDA-MB-231 and SKBR-3 cell lines. Microdissected tumor tissues were cultured both in the newly developed microfluidic chip system and in conventional 24-well culture plates. Two systems were compared for two different types of tumors. The confocal microscopy analyses, lactate dehydrogenase release, and glucose consumption values showed that the tissues in the microfluidic system remained more viable with respect to the conventional well plate culturing method, up to 96 h. The new culturing technique described here may be superior to conventional culturing techniques for developing new treatment strategies, such as testing chemotherapeutics on tumor samples from individual patients.

3D bioprinted, vascularized neuroblastoma tumor environment in fluidic chip devices for precision medicine drug testing

Neuroblastoma is an extracranial solid tumor which develops in early childhood and still has a poor prognosis. One strategy to increase cure rates is the identification of patient-specific drug responses in tissue models that mimic the interaction between patient cancer cells and tumor environment. We therefore developed a perfused and micro-vascularized tumor-environment model that is directly bioprinted into custom-manufactured fluidic chips. A gelatin-methacrylate/fibrin-based matrix containing multiple cell types mimics the tumor-microenvironment that promotes spontaneous micro-vessel formation by embedded endothelial cells. We demonstrate that both, adipocyte- and iPSC-derived mesenchymal stem cells can guide this process. Bioprinted channels are coated with endothelial cells post printing to form a dense vessel - tissue barrier. The tissue model thereby mimics structure and function of human soft tissue with endothelial cell-coated larger vessels for perfusion and micro-vessel networks within the hydrogel-matrix. Patient-derived neuroblastoma spheroids are added to the matrix during the printing process and grown for more than two weeks. We demonstrate that micro-vessels are attracted by and grow into tumor spheroids and that neuroblastoma cells invade the tumor-environment as soon as the spheroids disrupt. In summary, we describe the first bioprinted, micro-vascularized neuroblastoma - tumor-environment model directly printed into fluidic chips and a novel medium-throughput biofabrication platform suitable for studying tumor angiogenesis and metastasis in precision medicine approaches in future.

Microbiota medicine: towards clinical revolution

The human gastrointestinal tract is inhabited by the largest microbial community within the human body consisting of trillions of microbes called gut microbiota. The normal flora is the site of many physiological functions such as enhancing the host immunity, participating in the nutrient absorption and protecting the body against pathogenic microorganisms. Numerous investigations showed a bidirectional interplay between gut microbiota and many organs within the human body such as the intestines, the lungs, the brain, and the skin. Large body of evidence demonstrated, more than a decade ago, that the gut microbial alteration is a key factor in the pathogenesis of many local and systemic disorders. In this regard, a deep understanding of the mechanisms involved in the gut microbial symbiosis/dysbiosis is crucial for the clinical and health field. We review the most recent studies on the involvement of gut microbiota in the pathogenesis of many diseases. We also elaborate the different strategies used to manipulate the gut microbiota in the prevention and treatment of disorders. The future of medicine is strongly related to the quality of our microbiota. Targeting microbiota dysbiosis will be a huge challenge.

Circulating inflammatory cytokines and risk of five cancers: a Mendelian randomization analysis

BACKGROUND

Epidemiological and experimental evidence has linked chronic inflammation to cancer aetiology. It is unclear whether associations for specific inflammatory biomarkers are causal or due to bias. In order to examine whether altered genetically predicted concentration of circulating cytokines are associated with cancer development, we performed a two-sample Mendelian randomisation (MR) analysis.

METHODS

Up to 31,112 individuals of European descent were included in genome-wide association study (GWAS) meta-analyses of 47 circulating cytokines. Single nucleotide polymorphisms (SNPs) robustly associated with the cytokines, located in or close to their coding gene (cis), were used as instrumental variables. Inverse-variance weighted MR was used as the primary analysis, and the MR assumptions were evaluated in sensitivity and colocalization analyses and a false discovery rate (FDR) correction for multiple comparisons was applied. Corresponding germline GWAS summary data for five cancer outcomes (breast, endometrial, lung, ovarian, and prostate), and their subtypes were selected from the largest cancer-specific GWASs available (cases ranging from 12,906 for endometrial to 133,384 for breast cancer).

RESULTS

There was evidence of inverse associations of macrophage migration inhibitory factor with breast cancer (OR per SD = 0.88, 95% CI 0.83 to 0.94), interleukin-1 receptor antagonist with endometrial cancer (0.86, 0.80 to 0.93), interleukin-18 with lung cancer (0.87, 0.81 to 0.93), and beta-chemokine-RANTES with ovarian cancer (0.70, 0.57 to 0.85) and positive associations of monokine induced by gamma interferon with endometrial cancer (3.73, 1.86 to 7.47) and cutaneous T-cell attracting chemokine with lung cancer (1.51, 1.22 to 1.87). These associations were similar in sensitivity analyses and supported in colocalization analyses.

CONCLUSIONS

Our study adds to current knowledge on the role of specific inflammatory biomarker pathways in cancer aetiology. Further validation is needed to assess the potential of these cytokines as pharmacological or lifestyle targets for cancer prevention.

Spatial transcriptomics using combinatorial fluorescence spectral and lifetime encoding, imaging and analysis

Multiplexed mRNA profiling in the spatial context provides new information enabling basic research and clinical applications. Unfortunately, existing spatial transcriptomics methods are limited due to either low multiplexing or complexity. Here, we introduce a spatialomics technology, termed Multi Omic Single-scan Assay with Integrated Combinatorial Analysis (MOSAICA), that integrates in situ labeling of mRNA and protein markers in cells or tissues with combinatorial fluorescence spectral and lifetime encoded probes, spectral and time-resolved fluorescence imaging, and machine learning-based decoding. We demonstrate MOSAICA's multiplexing scalability in detecting 10-plex targets in fixed colorectal cancer cells using combinatorial labeling of five fluorophores with facile error-detection and removal of autofluorescence. MOSAICA's analysis is strongly correlated with sequencing data (Pearson's r = 0.96) and was further benchmarked using RNAscopeTM and LGC StellarisTM. We further apply MOSAICA for multiplexed analysis of clinical melanoma Formalin-Fixed Paraffin-Embedded (FFPE) tissues. We finally demonstrate simultaneous co-detection of protein and mRNA in cancer cells.

Using a convolutional neural network for classification of squamous and non-squamous non-small cell lung cancer based on diagnostic histopathology HES images

Histological stratification in metastatic non-small cell lung cancer (NSCLC) is essential to properly guide therapy. Morphological evaluation remains the basis for subtyping and is completed by additional immunohistochemistry labelling to confirm the diagnosis, which delays molecular analysis and utilises precious sample. Therefore, we tested the capacity of convolutional neural networks (CNNs) to classify NSCLC based on pathologic HES diagnostic biopsies. The model was estimated with a learning cohort of 132 NSCLC patients and validated on an external validation cohort of 65 NSCLC patients. Based on image patches, a CNN using InceptionV3 architecture was trained and optimized to classify NSCLC between squamous and non-squamous subtypes. Accuracies of 0.99, 0.87, 0.85, 0.85 was reached in the training, validation and test sets and in the external validation cohort. At the patient level, the CNN model showed a capacity to predict the tumour histology with accuracy of 0.73 and 0.78 in the learning and external validation cohorts respectively. Selecting tumour area using virtual tissue micro-array improved prediction, with accuracy of 0.82 in the external validation cohort. This study underlines the capacity of CNN to predict NSCLC subtype with good accuracy and to be applied to small pathologic samples without annotation.

Computational approaches leveraging integrated connections of multi-omic data toward clinical applications

In line with the advances in high-throughput technologies, multiple omic datasets have accumulated to study biological systems and diseases coherently. No single omics data type is capable of fully representing cellular activity. The complexity of the biological processes arises from the interactions between omic entities such as genes, proteins, and metabolites. Therefore, multi-omic data integration is crucial but challenging. The impact of the molecular alterations in multi-omic data is not local in the neighborhood of the altered gene or protein; rather, the impact diffuses in the network and changes the functionality of multiple signaling pathways and regulation of the gene expression. Additionally, multi-omic data is high-dimensional and has background noise. Several integrative approaches have been developed to accurately interpret the multi-omic datasets, including machine learning, network-based methods, and their combination. In this review, we overview the most recent integrative approaches and tools with a focus on network-based methods. We then discuss these approaches according to their specific applications, from disease-network and biomarker identification to patient stratification, drug discovery, and repurposing.

Converging 2D Nanomaterials and 3D Bioprinting Technology: State-of-the-Art, Challenges, and Potential Outlook in Biomedical Applications

The development of next-generation of bioinks aims to fabricate anatomical size 3D scaffold with high printability and biocompatibility. Along with the progress in 3D bioprinting, 2D nanomaterials (2D NMs) prove to be emerging frontiers in the development of advanced materials owing to their extraordinary properties. Harnessing the properties of 2D NMs in 3D bioprinting technologies can revolutionize the development of bioinks by endowing new functionalities to the current bioinks. First the main contributions of 2D NMS in 3D bioprinting technologies are categorized here into six main classes: 1) reinforcement effect, 2) delivery of bioactive molecules, 3) improved electrical conductivity, 4) enhanced tissue formation, 5) photothermal effect, 6) and stronger antibacterial properties. Next, the recent advances in the use of each certain 2D NMs (1) graphene, 2) nanosilicate, 3) black phosphorus, 4) MXene, 5) transition metal dichalcogenides, 6) hexagonal boron nitride, and 7) metal-organic frameworks) in 3D bioprinting technology are critically summarized and evaluated thoroughly. Third, the role of physicochemical properties of 2D NMSs on their cytotoxicity is uncovered, with several representative examples of each studied 2D NMs. Finally, current challenges, opportunities, and outlook for the development of nanocomposite bioinks are discussed thoroughly.

Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures

The utility of pathway signatures lies in their capability to determine whether a specific pathway or biological process is dysregulated in a given patient. These signatures have been widely used in machine learning (ML) methods for a variety of applications including precision medicine, drug repurposing, and drug discovery. In this work, we leverage highly predictive ML models for drug response simulation in individual patients by calibrating the pathway activity scores of disease samples. Using these ML models and an intuitive scoring algorithm to modify the signatures of patients, we evaluate whether a given sample that was formerly classified as diseased, could be predicted as normal following drug treatment simulation. We then use this technique as a proxy for the identification of potential drug candidates. Furthermore, we demonstrate the ability of our methodology to successfully identify approved and clinically investigated drugs for four different cancers, outperforming six comparable state-of-the-art methods. We also show how this approach can deconvolute a drugs' mechanism of action and propose combination therapies. Taken together, our methodology could be promising to support clinical decision-making in personalized medicine by simulating a drugs' effect on a given patient.

Genetic mutation and biological pathway prediction based on whole slide images in breast carcinoma using deep learning

Breast carcinoma is the most common cancer among women worldwide that consists of a heterogeneous group of subtype diseases. The whole-slide images (WSIs) can capture the cell-level heterogeneity, and are routinely used for cancer diagnosis by pathologists. However, key driver genetic mutations related to targeted therapies are identified by genomic analysis like high-throughput molecular profiling. In this study, we develop a deep-learning model to predict the genetic mutations and biological pathway activities directly from WSIs. Our study offers unique insights into WSI visual interactions between mutation and its related pathway, enabling a head-to-head comparison to reinforce our major findings. Using the histopathology images from the Genomic Data Commons Database, our model can predict the point mutations of six important genes (AUC 0.68-0.85) and copy number alteration of another six genes (AUC 0.69-0.79). Additionally, the trained models can predict the activities of three out of ten canonical pathways (AUC 0.65-0.79). Next, we visualized the weight maps of tumor tiles in WSI to understand the decision-making process of deep-learning models via a self-attention mechanism. We further validated our models on liver and lung cancers that are related to metastatic breast cancer. Our results provide insights into the association between pathological image features, molecular outcomes, and targeted therapies for breast cancer patients.