A dedicated structured data set for reporting of invasive carcinoma of the breast in the setting of neoadjuvant therapy: recommendations from the International Collaboration on Cancer Reporting (ICCR).

AIMS: The International Collaboration on Cancer Reporting (ICCR), a global alliance of major (inter-)national pathology and cancer organisations, is an initiative aimed at providing a unified international approach to reporting cancer. ICCR recently published new data sets for the reporting of invasive breast carcinoma, surgically removed lymph nodes for breast tumours and ductal carcinoma in situ, variants of lobular carcinoma in situ and low-grade lesions. The data set in this paper addresses the neoadjuvant setting. The aim is to promote high-quality, standardised reporting of tumour response and residual disease after neoadjuvant treatment that can be used for subsequent management decisions for each patient. METHODS: The ICCR convened expert panels of breast pathologists with a representative surgeon and oncologist to critically review and discuss current evidence. Feedback from the international public consultation was critical in the development of this data set. RESULTS: The expert panel concluded that a dedicated data set was required for reporting of breast specimens post-neoadjuvant therapy with inclusion of data elements specific to the neoadjuvant setting as core or non-core elements. This data set proposes a practical approach for handling and reporting breast resection specimens following neoadjuvant therapy. The comments for each data element clarify terminology, discuss available evidence and highlight areas with limited evidence that need further study. This data set overlaps with, and should be used in conjunction with, the data sets for the reporting of invasive breast carcinoma and surgically removed lymph nodes from patients with breast tumours, as appropriate. Key issues specific to the neoadjuvant setting are included in this paper. The entire data set is freely available on the ICCR website. CONCLUSIONS: High-quality, standardised reporting of tumour response and residual disease after neoadjuvant treatment are critical for subsequent management decisions for each patient.

Single-cell integrative analysis reveals consensus cancer cell states and clinical relevance in breast cancer.

High heterogeneity and complex interactions of malignant cells in breast cancer has been recognized as a driver of cancer progression and therapeutic failure. However, complete understanding of common cancer cell states and their underlying driver factors remain scarce and challenging. Here, we revealed seven consensus cancer cell states recurring cross patients by integrative analysis of single-cell RNA sequencing data of breast cancer. The distinct biological functions, the subtype-specific distribution, the potential cells of origin and the interrelation of consensus cancer cell states were systematically elucidated and validated in multiple independent datasets. We further uncovered the internal regulons and external cell components in tumor microenvironments, which contribute to the consensus cancer cell states. Using the state-specific signature, we also inferred the abundance of cells with each consensus cancer cell state by deconvolution of large breast cancer RNA-seq cohorts, revealing the association of immune-related state with better survival. Our study provides new insights for the cancer cell state composition and potential therapeutic strategies of breast cancer.

A comprehensive database of exosome molecular biomarkers and disease-gene associations.

Exosomes play a crucial role in intercellular communication and can be used as biomarkers for diagnostic and therapeutic clinical applications. However, systematic studies in cancer-associated exosomal nucleic acids remain a big challenge. Here, we developed ExMdb, a comprehensive database of exosomal nucleic acid biomarkers and disease-gene associations curated from published literature and high-throughput datasets. We performed a comprehensive curation of exosome properties including 4,586 experimentally supported gene-disease associations, 13,768 diagnostic and therapeutic biomarkers, and 312,049 nucleic acid subcellular locations. To characterize expression variation of exosomal molecules and identify causal factors of complex diseases, we have also collected 164 high-throughput datasets, including bulk and single-cell RNA sequencing (scRNA-seq) data. Based on these datasets, we performed various bioinformatics and statistical analyses to support our conclusions and advance our knowledge of exosome biology. Collectively, our dataset will serve as an essential resource for investigating the regulatory mechanisms of complex diseases and improving the development of diagnostic and therapeutic biomarkers.

Brain tumor segmentation using synthetic MR images - A comparison of GANs and diffusion models.

Large annotated datasets are required for training deep learning models, but in medical imaging data sharing is often complicated due to ethics, anonymization and data protection legislation. Generative AI models, such as generative adversarial networks (GANs) and diffusion models, can today produce very realistic synthetic images, and can potentially facilitate data sharing. However, in order to share synthetic medical images it must first be demonstrated that they can be used for training different networks with acceptable performance. Here, we therefore comprehensively evaluate four GANs (progressive GAN, StyleGAN 1-3) and a diffusion model for the task of brain tumor segmentation (using two segmentation networks, U-Net and a Swin transformer). Our results show that segmentation networks trained on synthetic images reach Dice scores that are 80%-90% of Dice scores when training with real images, but that memorization of the training images can be a problem for diffusion models if the original dataset is too small. Our conclusion is that sharing synthetic medical images is a viable option to sharing real images, but that further work is required. The trained generative models and the generated synthetic images are shared on AIDA data hub.

Leucine Motifs Stabilize Residual Helical Structure in Disordered Proteins.

Many examples are known of regions of intrinsically disordered proteins that fold into α-helices upon binding to their targets. These helical binding motifs (HBMs) can be partially helical also in the unbound state, and this so-called residual structure can affect binding affinity and kinetics. To investigate the underlying mechanisms governing the formation of residual helical structure, we assembled a dataset of experimental helix contents of 65 peptides containing HBM that fold-upon-binding. The average residual helicity is 17% and increases to 60% upon target binding. The helix contents of residual and target-bound structures do not correlate, however the relative location of helix elements in both states shows a strong overlap. Compared to the general disordered regions, HBMs are enriched in amino acids with high helix preference and these residues are typically involved in target binding, explaining the overlap in helix positions. In particular, we find that leucine residues and leucine motifs in HBMs are the major contributors to helix stabilization and target-binding. For the two model peptides, we show that substitution of leucine motifs to other hydrophobic residues (valine or isoleucine) leads to reduction of residual helicity, supporting the role of leucine as helix stabilizer. From the three hydrophobic residues only leucine can efficiently stabilize residual helical structure. We suggest that the high occurrence of leucine motifs and a general preference for leucine at binding interfaces in HBMs can be explained by its unique ability to stabilize helical elements.

Disparities in the Demographic Composition of The Cancer Imaging Archive.

Purpose To characterize the demographic distribution of The Cancer Imaging Archive (TCIA) studies and compare them with those of the U.S. cancer population. Materials and Methods In this retrospective study, data from TCIA studies were examined for the inclusion of demographic information. Of 189 studies in TCIA up until April 2023, a total of 83 human cancer studies were found to contain supporting demographic data. The median patient age and the sex, race, and ethnicity proportions of each study were calculated and compared with those of the U.S. cancer population, provided by the Surveillance, Epidemiology, and End Results Program and the Centers for Disease Control and Prevention U.S. Cancer Statistics Data Visualizations Tool. Results The median age of TCIA patients was found to be 6.84 years lower than that of the U.S. cancer population (P = .047) and contained more female than male patients (53% vs 47%). American Indian and Alaska Native, Black or African American, and Hispanic patients were underrepresented in TCIA studies by 47.7%, 35.8%, and 14.7%, respectively, compared with the U.S. cancer population. Conclusion The results demonstrate that the patient demographics of TCIA data sets do not reflect those of the U.S. cancer population, which may decrease the generalizability of artificial intelligence radiology tools developed using these imaging data sets. Keywords: Ethics, Meta-Analysis, Health Disparities, Cancer Health Disparities, Machine Learning, Artificial Intelligence, Race, Ethnicity, Sex, Age, Bias Published under a CC BY 4.0 license.

A Comparative Study of Deep Learning Methods for Multi-Class Semantic Segmentation of 2D Kidney Ultrasound Images.

Ultrasound (US) imaging is a widely used medical imaging modality for the diagnosis, monitoring, and surgical planning for kidney conditions. Thus, accurate segmentation of the kidney and internal structures in US images is essential for the assessment of kidney function and the detection of pathological conditions, such as cysts, tumors, and kidney stones. Therefore, there is a need for automated methods that can accurately segment the kidney and internal structures in US images. Over the years, automatic strategies were proposed for such purpose, with deep learning methods achieving the current state-of-the-art results. However, these strategies typically ignore the segmentation of the internal structures of the kidney. Moreover, they were evaluated in different private datasets, hampering the direct comparison of results, and making it difficult to determination the optimal strategy for this task. In this study, we perform a comparative analysis of 7 deep learning networks for the segmentation of the kidney and internal structures (Capsule, Central Echogenic Complex (CEC), Cortex and Medulla) in 2D US images in an open access multi-class kidney US dataset. The dataset includes 514 images, acquired in multiple clinical centers using different US machines and protocols. The dataset contains the annotation of two experts, but 321 images with complete segmentation of all 4 classes were used. Overall, the results demonstrate that the DeepLabV3+ network outperformed the inter-rater variability with a Dice score of 78.0% compared to 75.6% for inter-rater variability. Specifically, DeepLabV3Plus achieved mean Dice scores of 94.2% for the Capsule, 85.8% for the CEC, 62.4% for the Cortex, and 69.6% for the Medulla. These findings suggest the potential of deep learning-based methods in improving the accuracy of kidney segmentation in US images.Clinical Relevance- This study shows the potential of DL for improving accuracy of kidney segmentation in US, leading to increased diagnostic efficiency, and enabling new applications such as computer-aided diagnosis and treatment, ultimately resulting in improved patient outcomes and reduced healthcare costs.1.

A hybrid machine learning feature selection model-HMLFSM to enhance gene classification applied to multiple colon cancers dataset.

Colon cancer is a significant global health problem, and early detection is critical for improving survival rates. Traditional detection methods, such as colonoscopies, can be invasive and uncomfortable for patients. Machine Learning (ML) algorithms have emerged as a promising approach for non-invasive colon cancer classification using genetic data or patient demographics and medical history. One approach is to use ML to analyse genetic data, or patient demographics and medical history, to predict the likelihood of colon cancer. However, due to the challenges imposed by variable gene expression and the high dimensionality of cancer-related datasets, traditional transductive ML applications have limited accuracy and risk overfitting. In this paper, we propose a new hybrid feature selection model called HMLFSM-Hybrid Machine Learning Feature Selection Model to improve colon cancer gene classification. We developed a multifilter hybrid model including a two-phase feature selection approach, combining Information Gain (IG) and Genetic Algorithms (GA), and minimum Redundancy Maximum Relevance (mRMR) coupling with Particle Swarm Optimization (PSO). We critically tested our model on three colon cancer genetic datasets and found that the new framework outperformed other models with significant accuracy improvements (95%, ~97%, and ~94% accuracies for datasets 1, 2, and 3 respectively). The results show that our approach improves the classification accuracy of colon cancer detection by highlighting important and relevant genes, eliminating irrelevant ones, and revealing the genes that have a direct influence on the classification process. For colon cancer gene analysis, and along with our experiments and literature review, we found that selective input feature extraction prior to feature selection is essential for improving predictive performance.

scAN1.0: A reproducible and standardized pipeline for processing 10X single cell RNAseq data.

Single cell transcriptomics has recently seen a surge in popularity, leading to the need for data analysis pipelines that are reproducible, modular, and interoperable across different systems and institutions.To meet this demand, we introduce scAN1.0, a processing pipeline for analyzing 10X single cell RNA sequencing data. scAN1.0 is built using the Nextflow DSL2 and can be run on most computational systems. The modular design of Nextflow pipelines enables easy integration and evaluation of different blocks for specific analysis steps.We demonstrate the usefulness of scAN1.0 by showing its ability to examine the impact of the mapping step during the analysis of two datasets: (i) a 10X scRNAseq of a human pituitary gonadotroph tumor dataset and (ii) a murine 10X scRNAseq acquired on CD8 T cells during an immune response.

Segmentation of glioblastomas in early post-operative multi-modal MRI with deep neural networks.

Extent of resection after surgery is one of the main prognostic factors for patients diagnosed with glioblastoma. To achieve this, accurate segmentation and classification of residual tumor from post-operative MR images is essential. The current standard method for estimating it is subject to high inter- and intra-rater variability, and an automated method for segmentation of residual tumor in early post-operative MRI could lead to a more accurate estimation of extent of resection. In this study, two state-of-the-art neural network architectures for pre-operative segmentation were trained for the task. The models were extensively validated on a multicenter dataset with nearly 1000 patients, from 12 hospitals in Europe and the United States. The best performance achieved was a 61% Dice score, and the best classification performance was about 80% balanced accuracy, with a demonstrated ability to generalize across hospitals. In addition, the segmentation performance of the best models was on par with human expert raters. The predicted segmentations can be used to accurately classify the patients into those with residual tumor, and those with gross total resection.

A review of the machine learning datasets in mammography, their adherence to the FAIR principles and the outlook for the future.

The increasing rates of breast cancer, particularly in emerging economies, have led to interest in scalable deep learning-based solutions that improve the accuracy and cost-effectiveness of mammographic screening. However, such tools require large volumes of high-quality training data, which can be challenging to obtain. This paper combines the experience of an AI startup with an analysis of the FAIR principles of the eight available datasets. It demonstrates that the datasets vary considerably, particularly in their interoperability, as each dataset is skewed towards a particular clinical use-case. Additionally, the mix of digital captures and scanned film compounds the problem of variability, along with differences in licensing terms, ease of access, labelling reliability, and file formats. Improving interoperability through adherence to standards such as the BIRADS criteria for labelling and annotation, and a consistent file format, could markedly improve access and use of larger amounts of standardized data. This, in turn, could be increased further by GAN-based synthetic data generation, paving the way towards better health outcomes for breast cancer.

WebSTR: A Population-wide Database of Short Tandem Repeat Variation in Humans.

Short tandem repeats (STRs) are consecutive repetitions of one to six nucleotide motifs. They are hypervariable due to the high prevalence of repeat unit insertions or deletions primarily caused by polymerase slippage during replication. Genetic variation at STRs has been shown to influence a range of traits in humans, including gene expression, cancer risk, and autism. Until recently STRs have been poorly studied since they pose significant challenges to bioinformatics analyses. Moreover, genome-wide analysis of STR variation in population-scale cohorts requires large amounts of data and computational resources. However, the recent advent of genome-wide analysis tools has resulted in multiple large genome-wide datasets of STR variation spanning nearly two million genomic loci in thousands of individuals from diverse populations. Here we present WebSTR, a database of genetic variation and other characteristics of genome-wide STRs across human populations. WebSTR is based on reference panels of more than 1.7 million human STRs created with state of the art repeat annotation methods and can easily be extended to include additional cohorts or species. It currently contains data based on STR genotypes for individuals from the 1000 Genomes Project, H3Africa, the Genotype-Tissue Expression (GTEx) Project and colorectal cancer patients from the TCGA dataset. WebSTR is implemented as a relational database with programmatic access available through an API and a web portal for browsing data. The web portal is publicly available at https://webstr.ucsd.edu.

Advancing prostate cancer detection: a comparative analysis of PCLDA-SVM and PCLDA-KNN classifiers for enhanced diagnostic accuracy.

This investigation aimed to assess the effectiveness of different classification models in diagnosing prostate cancer using a screening dataset obtained from the National Cancer Institute's Cancer Data Access System. The dataset was first reduced using the PCLDA method, which combines Principal Component Analysis and Linear Discriminant Analysis. Two classifiers, Support Vector Machine (SVM) and k-Nearest Neighbour (KNN), were then applied to compare their performance. The results showed that the PCLDA-SVM model achieved an impressive accuracy rate of 97.99%, with a precision of 0.92, sensitivity of 92.83%, specificity of 97.65%, and F1 score of 0.93. Additionally, it demonstrated a low error rate of 0.016 and a Matthews Correlation Coefficient (MCC) and Kappa coefficient of 0.946. On the other hand, the PCLDA-KNN model also performed well, achieving an accuracy of 97.8%, precision of 0.93, sensitivity of 93.39%, specificity of 97.86%, an F1 score of 0.92, a high MCC and Kappa coefficient of 0.98, and an error rate of 0.006. In conclusion, the PCLDA-SVM method exhibited improved efficacy in diagnosing prostate cancer compared to the PCLDA-KNN model. Both models, however, showed promising results, suggesting the potential of these classifiers in prostate cancer diagnosis.

Feature-aware unsupervised lesion segmentation for brain tumor images using fast data density functional transform.

We demonstrate that isomorphically mapping gray-level medical image matrices onto energy spaces underlying the framework of fast data density functional transform (fDDFT) can achieve the unsupervised recognition of lesion morphology. By introducing the architecture of geometric deep learning and metrics of graph neural networks, gridized density functionals of the fDDFT establish an unsupervised feature-aware mechanism with global convolutional kernels to extract the most likely lesion boundaries and produce lesion segmentation. An AutoEncoder-assisted module reduces the computational complexity from [Formula: see text] to [Formula: see text], thus efficiently speeding up global convolutional operations. We validate their performance utilizing various open-access datasets and discuss limitations. The inference time of each object in large three-dimensional datasets is 1.76 s on average. The proposed gridized density functionals have activation capability synergized with gradient ascent operations, hence can be modularized and embedded in pipelines of modern deep neural networks. Algorithms of geometric stability and similarity convergence also raise the accuracy of unsupervised recognition and segmentation of lesion images. Their performance achieves the standard requirement for conventional deep neural networks; the median dice score is higher than 0.75. The experiment shows that the synergy of fDDFT and a naïve neural network improves the training and inference time by 58% and 51%, respectively, and the dice score raises to 0.9415. This advantage facilitates fast computational modeling in interdisciplinary applications and clinical investigation.

Authigenic mineral phases as a driver of the upper-ocean iron cycle.

Iron is important in regulating the ocean carbon cycle1. Although several dissolved and particulate species participate in oceanic iron cycling, current understanding emphasizes the importance of complexation by organic ligands in stabilizing oceanic dissolved iron concentrations2-6. However, it is difficult to reconcile this view of ligands as a primary control on dissolved iron cycling with the observed size partitioning of dissolved iron species, inefficient dissolved iron regeneration at depth or the potential importance of authigenic iron phases in particulate iron observational datasets7-12. Here we present a new dissolved iron, ligand and particulate iron seasonal dataset from the Bermuda Atlantic Time-series Study (BATS) region. We find that upper-ocean dissolved iron dynamics were decoupled from those of ligands, which necessitates a process by which dissolved iron escapes ligand stabilization to generate a reservoir of authigenic iron particles that settle to depth. When this 'colloidal shunt' mechanism was implemented in a global-scale biogeochemical model, it reproduced both seasonal iron-cycle dynamics observations and independent global datasets when previous models failed13-15. Overall, we argue that the turnover of authigenic particulate iron phases must be considered alongside biological activity and ligands in controlling ocean-dissolved iron distributions and the coupling between dissolved and particulate iron pools.

Phylogenetic inference from single-cell RNA-seq data.

Tumors are comprised of subpopulations of cancer cells that harbor distinct genetic profiles and phenotypes that evolve over time and during treatment. By reconstructing the course of cancer evolution, we can understand the acquisition of the malignant properties that drive tumor progression. Unfortunately, recovering the evolutionary relationships of individual cancer cells linked to their phenotypes remains a difficult challenge. To address this need, we have developed PhylinSic, a method that reconstructs the phylogenetic relationships among cells linked to their gene expression profiles from single cell RNA-sequencing (scRNA-Seq) data. This method calls nucleotide bases using a probabilistic smoothing approach and then estimates a phylogenetic tree using a Bayesian modeling algorithm. We showed that PhylinSic identified evolutionary relationships underpinning drug selection and metastasis and was sensitive enough to identify subclones from genetic drift. We found that breast cancer tumors resistant to chemotherapies harbored multiple genetic lineages that independently acquired high K-Ras and ß-catenin, suggesting that therapeutic strategies may need to control multiple lineages to be durable. These results demonstrated that PhylinSic can reconstruct evolution and link the genotypes and phenotypes of cells across monophyletic tumors using scRNA-Seq.

Deep learning model fusion improves lung tumor segmentation accuracy across variable training-to-test dataset ratios.

This study aimed to investigate the robustness of a deep learning (DL) fusion model for low training-to-test ratio (TTR) datasets in the segmentation of gross tumor volumes (GTVs) in three-dimensional planning computed tomography (CT) images for lung cancer stereotactic body radiotherapy (SBRT). A total of 192 patients with lung cancer (solid tumor, 118; part-solid tumor, 53; ground-glass opacity, 21) who underwent SBRT were included in this study. Regions of interest in the GTVs were cropped based on GTV centroids from planning CT images. Three DL models, 3D U-Net, V-Net, and dense V-Net, were trained to segment the GTV regions. Nine fusion models were constructed with logical AND, logical OR, and voting of the two or three outputs of the three DL models. TTR was defined as the ratio of the number of cases in a training dataset to that in a test dataset. The Dice similarity coefficients (DSCs) and Hausdorff distance (HD) of the 12 models were assessed with TTRs of 1.00 (training data: validation data: test data = 40:20:40), 0.791 (35:20:45), 0.531 (31:10:59), 0.291 (20:10:70), and 0.116 (10:5:85). The voting fusion model achieved the highest DSCs of 0.829 to 0.798 for all TTRs among the 12 models, whereas the other models showed DSCs of 0.818 to 0.804 for a TTR of 1.00 and 0.788 to 0.742 for a TTR of 0.116, and an HD of 5.40 ± 3.00 to 6.07 ± 3.26 mm better than any single DL models. The findings suggest that the proposed voting fusion model is a robust approach for low TTR datasets in segmenting GTVs in planning CT images of lung cancer SBRT.

A deep learning approach to estimate x-ray scatter in digital breast tomosynthesis: From phantom models to clinical applications.

BACKGROUND: Digital breast tomosynthesis (DBT) has gained popularity as breast imaging modality due to its pseudo-3D reconstruction and improved accuracy compared to digital mammography. However, DBT faces challenges in image quality and quantitative accuracy due to scatter radiation. Recent advancements in deep learning (DL) have shown promise in using fast convolutional neural networks for scatter correction, achieving comparable results to Monte Carlo (MC) simulations. PURPOSE: To predict the scatter radiation signal in DBT projections within clinically-acceptable times and using only clinically-available data, such as compressed breast thickness and acquisition angle. METHODS: MC simulations to obtain scatter estimates were generated from two types of digital breast phantoms. One set consisted of 600 realistically-shaped homogeneous breast phantoms for initial DL training. The other set was composed of 80 anthropomorphic phantoms, containing realistic internal tissue texture, aimed at fine tuning the DL model for clinical applications. The MC simulations generated scatter and primary maps per projection angle for a wide-angle DBT system. Both datasets were used to train (using 7680 projections from homogeneous phantoms), validate (using 960 and 192 projections from the homogeneous and anthropomorphic phantoms, respectively), and test (using 960 and 48 projections from the homogeneous and anthropomorphic phantoms, respectively) the DL model. The DL output was compared to the corresponding MC ground truth using both quantitative and qualitative metrics, such as mean relative and mean absolute relative differences (MRD and MARD), and to previously-published scatter-to-primary (SPR) ratios for similar breast phantoms. The scatter corrected DBT reconstructions were evaluated by analyzing the obtained linear attenuation values and by visual assessment of corrected projections in a clinical dataset. The time required for training and prediction per projection, as well as the time it takes to produce scatter-corrected projection images, were also tracked. RESULTS: The quantitative comparison between DL scatter predictions and MC simulations showed a median MRD of 0.05% (interquartile range (IQR), -0.04% to 0.13%) and a median MARD of 1.32% (IQR, 0.98% to 1.85%) for homogeneous phantom projections and a median MRD of -0.21% (IQR, -0.35% to -0.07%) and a median MARD of 1.43% (IQR, 1.32% to 1.66%) for the anthropomorphic phantoms. The SPRs for different breast thicknesses and at different projection angles were within ± 15% of the previously-published ranges. The visual assessment showed good prediction capabilities of the DL model with a close match between MC and DL scatter estimates, as well as between DL-based scatter corrected and anti-scatter grid corrected cases. The scatter correction improved the accuracy of the reconstructed linear attenuation of adipose tissue, reducing the error from -16% and -11% to -2.3% and 4.4% for an anthropomorphic digital phantom and clinical case with similar breast thickness, respectively. The DL model training took 40 min and prediction of a single projection took less than 0.01 s. Generating scatter corrected images took 0.03 s per projection for clinical exams and 0.16 s for one entire projection set. CONCLUSIONS: This DL-based method for estimating the scatter signal in DBT projections is fast and accurate, paving the way for future quantitative applications.

A harmonized resource of integrated prostate cancer clinical, -omic, and signature features.

Genomic and transcriptomic data have been generated across a wide range of prostate cancer (PCa) study cohorts. These data can be used to better characterize the molecular features associated with clinical outcomes and to test hypotheses across multiple, independent patient cohorts. In addition, derived features, such as estimates of cell composition, risk scores, and androgen receptor (AR) scores, can be used to develop novel hypotheses leveraging existing multi-omic datasets. The full potential of such data is yet to be realized as independent datasets exist in different repositories, have been processed using different pipelines, and derived and clinical features are often not provided or  not standardized. Here, we present the curatedPCaData R package, a harmonized data resource representing >2900 primary tumor, >200 normal tissue, and >500 metastatic PCa samples across 19 datasets processed using standardized pipelines with updated gene annotations. We show that meta-analysis across harmonized studies has great potential for robust and clinically meaningful insights. curatedPCaData is an open and accessible community resource with code made available for reproducibility.

Cellular dynamics in tumour microenvironment along with lung cancer progression underscore spatial and evolutionary heterogeneity of neutrophil.

BACKGROUND: The cellular dynamics in the tumour microenvironment (TME) along with non-small cell lung cancer (NSCLC) progression remain unclear. METHODS: Multiplex immunofluorescence test detecting 10 immune-related markers on 553 primary tumour (PT) samples of NSCLC was conducted and spatial information in TME was assessed by the StarDist depth learning model. The single-cell transcriptomic atlas of PT (n = 4) and paired tumour-draining lymph nodes (TDLNs) (n = 5 for tumour-invaded, n = 3 for tumour-free) microenvironment was profiled. Various bioinformatics analyses based on Gene Expression Omnibus, TCGA and Array-Express databases were also used to validate the discoveries. RESULTS: Spatial distances of CD4+ T cells-CD38+ T cells, CD4+ T cells-neutrophils and CD38+ T cells-neutrophils prolonged and they were replaced by CD163+ macrophages in PT along with tumour progression. Neutrophils showed unique stage and location-dependent prognostic effects. A high abundance of stromal neutrophils improved disease-free survival in the early-stage, whereas high intratumoural neutrophil infiltrates predicted poor prognosis in the mid-to-late-stage. Significant molecular and functional reprogramming in PT and TDLN microenvironments was observed. Diverse interaction networks mediated by neutrophils were found between positive and negative TDLNs. Five phenotypically and functionally heterogeneous subtypes of tumour-associated neutrophil (TAN) were further identified by pseudotime analysis, including TAN-0 with antigen-presenting function, TAN-1 with strong expression of interferon (IFN)-stimulated genes, the pro-tumour TAN-2 subcluster, the classical subset (TAN-3) and the pro-inflammatory subtype (TAN-4). Loss of IFN-stimulated signature and growing angiogenesis activity were discovered along the transitional trajectory. Eventually, a robust six neutrophil differentiation relevant genes-based model was established, showing that low-risk patients had longer overall survival time and may respond better to immunotherapy. CONCLUSIONS: The cellular composition, spatial location, molecular and functional changes in PT and TDLN microenvironments along with NSCLC progression were deciphered, highlighting the immunoregulatory roles and evolutionary heterogeneity of TANs.