Clinical Datasets Research Articles

Liver fibrosis automatic diagnosis utilizing dense-fusion attention contrastive learning network.

Liver fibrosis poses a significant public health challenge given its elevated incidence and associated mortality rates. Diffusion-Weighted Imaging (DWI) serves as a non-invasive diagnostic tool for supporting the identification of liver fibrosis. Deep learning, as a computer-aided diagnostic technology, can assist in recognizing the stage of liver fibrosis by extracting abstract features from DWI images. However, gathering samples is often challenging, posing a common dilemma in previous research. Moreover, previous studies frequently overlooked the cross-comparison information and latent connections among different DWI parameters. Thus, it is becoming a challenge to identify effective DWI parameters and dig potential features from multiple categories in a dataset with limited samples. A self-defined Multi-view Contrastive Learning Network is developed to automatically classify multi-parameter DWI images and explore synergies between different DWI parameters. A Dense-fusion Attention Contrastive Learning Network (DACLN) is designed and used to recognize DWI images. Concretely, a multi-view contrastive learning framework is constructed to train and extract features from raw multi-parameter DWI. Besides, a Dense-fusion module is designed to integrate feature and output predicted labels. We evaluated the performance of the proposed model on a set of real clinical data and analyzed the interpretability by Grad-CAM and annotation analysis, achieving average scores of 0.8825, 0.8702, 0.8933, 0.8727, and 0.8779 for accuracy, precision, recall, specificity and F-1 score. Of note, the experimental results revealed that IVIM-f, CTRW-β, and MONO-ADC exhibited significant recognition ability and complementarity. Our method achieves competitive accuracy in liver fibrosis diagnosis using the limited multi-parameter DWI dataset and finds three types of DWI parameters with high sensitivity for diagnosing liver fibrosis, which suggests potential directions for future research.

Development of a minimum clinical dataset for preoperative comprehensive geriatric assessment using a modified Delphi technique.

To construct a standardised, consensus-guided minimum clinical dataset (MCDS) for preoperative comprehensive geriatric assessment and optimisation (CGA) in Australia and Aotearoa New Zealand. We conducted a review of the international perioperative literature to identify CGA domains and tools for potential inclusion in the MCDS. We invited members of the Australian and New Zealand Society for Geriatric Medicine to participate in a Delphi study to obtain consensus on MCDS tools. Participants were asked to rate proposed tools using Likert scales (when >2 tools) or make a binary choice between two proposed tools. Consensus was considered to be achieved when there was at least 75% concordance between the two rounds amongst the participants, and at least one variable attaining over 50% of participants' votes. Domains that did not achieve consensus in Round 1 were carried over to Round 2. There were 73 participants in Round 1 of the Delphi study and 47 participants in Round 2. Consensus was achieved on tool/s recommended for every MCDS domain: Clinical Frailty Scale (frailty); sMMSE, RUDAS, MoCA (cognition); 4AT (delirium); timed-up-and-go (physical function); GDS-15 (mood); Barthel Index (functional status); and MUST (malnutrition). We recommend clinicians delivering preoperative CGA consider the use of the MCDS we have constructed when assessing older people contemplating surgery, as part of a multicomponent and multidisciplinary approach to optimising perioperative outcomes.

PAX6 promotes neuroendocrine phenotypes of prostate cancer via enhancing MET/STAT5A-mediated chromatin accessibility

BackgroundNeuroendocrine prostate cancer (NEPC) is a lethal subset of prostate cancer which is characterized by neuroendocrine differentiation and loss of androgen receptor (AR) signaling. Growing evidence reveals that cell lineage plasticity is crucial in the failure of NEPC therapies. Although studies suggest the involvement of the neural transcription factor PAX6 in drug resistance, its specific role in NEPC remains unclear.MethodsThe expression of PAX6 in NEPC was identified via bioinformatics and immunohistochemistry. CCK8 assay, colony formation assay, tumorsphere formation assay and apoptosis assay were used to illustrate the key role of PAX6 in the progression of in vitro. ChIP and Dual-luciferase reporter assays were conducted to confirm the binding sequences of AR in the promoter region of PAX6, as well as the binding sequences of PAX6 in the promoter regions of STAT5A and MET. For in vivo validation, the xenograft model representing NEPC subtype underwent pathological analysis to verify the significant role of PAX6 in disease progression. Complementary diagnoses were established through public clinical datasets and transcriptome sequencing of specific cell lines. ATAC-seq was used to detect the chromatin accessibility of specific cell lines.ResultsPAX6 expression was significantly elevated in NEPC and negatively regulated by AR signaling. Activation of PAX6 in non-NEPC cells led to NE trans-differentiation, while knock-down of PAX6 in NEPC cells inhibited the development and progression of NEPC. Importantly, loss of AR resulted in an enhanced expression of PAX6, which reprogramed the lineage plasticity of prostate cancer cells to develop NE phenotypes through the MET/STAT5A signaling pathway. Through ATAC-seq, we found that a high expression level of PAX6 elicited enhanced chromatin accessibility, mainly through attenuation of H4K20me3, which typically causes chromatin silence in cancer cells.ConclusionThis study reveals a novel neural transcription factor PAX6 could drive NEPC progression and suggest that it might serve as a potential therapeutic target for the management of NEPC.

Differentially expressed genes associated with high metabolic tumor volume served as diagnostic markers and potential therapeutic targets for pancreatic cancer

BackgroundThe lack of distinct biomarkers for pancreatic cancer is a major cause of early-stage detection difficulty. The pancreatic cancer patient group with high metabolic tumor volume (MTV), one of the values measured from positron emission tomography—a confirmatory method and standard care for pancreatic cancer, showed a poorer prognosis than those with low MTV. Therefore, MTV-associated differentially expressed genes (DEGs) may be candidates for distinctive markers for pancreatic cancer. This study aimed to evaluate the possibility of MTV-related DEGs as markers or therapeutic targets for pancreatic cancer.MethodsTumor tissues and their normal counterparts were obtained from patients undergoing preoperative 18F-FDG PET/CT. The tissues were classified into MTV-low and MTV-high groups (7 for each) based on the MTV2.5 value of 4.5 (MTV-low: MTV2.5 < 4.5, MTV-high: MTV2.5 ≥ 4.5). Gene expression fold change was first calculated in cancer tissue compared to its normal counter and then compared between low and high MTV groups to obtain significant DEGs. To assess the suitability of the DEGs for clinical application, the correlation of the DEGs with tumor grades and clinical outcomes was analyzed in TCGA-PAAD, a large dataset without MTV information.ResultsTotal RNA-sequencing (MTV RNA-Seq) revealed that 44 genes were upregulated and 56 were downregulated in the high MTV group. We selected the 29 genes matching MTV RNA-seq patterns in the TCGA-PAAD dataset, a large clinical dataset without MTV information, as MTV-associated genes (MAGs). In the analysis with the TCGA dataset, MAGs were significantly associated with patient survival, treatment outcomes, TCGA-PAAD-suggested markers, and CEACAM family proteins. Some MAGs showed an inverse correlation with miRNAs and were confirmed to be differentially expressed between normal and cancerous pancreatic tissues. Overexpression of KIF11 and RCC1 and underexpression of ADCY1 and SDK1 were detected in ~ 60% of grade 2 pancreatic cancer patients and associated with ~ 60% mortality in stages I and II.ConclusionsMAGs may serve as diagnostic markers and miRNA therapeutic targets for pancreatic cancer. Among the MAGs, KIF11, RCC1, ADCY, and SDK1 may be early diagnostic markers.

A retrospective prognostic evaluation using unsupervised learning in the treatment of COVID-19 patients with hypertension treated with ACEI/ARB drugs.

This study aimed to evaluate the prognosis of patients with COVID-19 and hypertension who were treated with angiotensin-converting enzyme inhibitor (ACEI)/angiotensin receptor B (ARB) drugs and to identify key features affecting patient prognosis using an unsupervised learning method. A large-scale clinical dataset, including patient information, medical history, and laboratory test results, was collected. Two hundred patients with COVID-19 and hypertension were included. After cluster analysis, patients were divided into good and poor prognosis groups. The unsupervised learning method was used to evaluate clinical characteristics and prognosis, and patients were divided into different prognosis groups. The improved wild dog optimization algorithm (IDOA) was used for feature selection and cluster analysis, followed by the IDOA-k-means algorithm. The impact of ACEI/ARB drugs on patient prognosis and key characteristics affecting patient prognosis were also analysed. Key features related to prognosis included baseline information and laboratory test results, while clinical symptoms and imaging results had low predictive power. The top six important features were age, hypertension grade, MuLBSTA, ACEI/ARB, NT-proBNP, and high-sensitivity troponin I. These features were consistent with the results of the unsupervised prediction model. A visualization system was developed based on these key features. Using unsupervised learning and the improved k-means algorithm, this study accurately analysed the prognosis of patients with COVID-19 and hypertension. The use of ACEI/ARB drugs was found to be a protective factor for poor clinical prognosis. Unsupervised learning methods can be used to differentiate patient populations and assess treatment effects. This study identified important features affecting patient prognosis and developed a visualization system with clinical significance for prognosis assessment and treatment decision-making.

Automatic three-dimensional facial symmetry reference plane construction based on facial planar reflective symmetry net

ObjectivesThree-dimensional (3D) facial symmetry analysis is based on the 3D symmetry reference plane (SRP). Artificial intelligence (AI) is widely used in the dental and oral sciences. This study developed a novel deep learning model called the facial planar reflective symmetry net (FPRS-Net) to automatically construct an SRP and established a method for defining a 3D point-cloud region of interest (ROI) and high-dimensional feature computations suitable for this network model. MethodsOverall, 240 patients were enroled. The deep learning model was trained and predicted using 200 samples, and its clinical suitability was evaluated with 40 samples. Four FPRS-Net models were prepared, each using supervised and unsupervised learning approaches based on full facial and ROI data (FPRS-NetS, FPRS-NetSR, FPRS-NetU, and FPRS-NetUR). These models were trained on 160 3D facial datasets, validated on 20 cases, and tested on another 20 cases. The model predictions were evaluated using an additional 40 clinical 3D facial datasets by comparing the mean square error of the SRP between the parameters predicted by the four FPRS-Net models and the truth plane. The clinical suitability of FPRS-Net models was evaluated by measuring the angle error between the predicted and ground-truth planes; experts evaluated the predicted SRP of the four FPRS-Net models using the visual analogue scales (VAS) method. ResultsThe FPRS-NetSR and FPRS-NetU models achieved an average angle error of 0.84° and 0.99° in predicting 3D facial SRP, respectively, with a VAS value of >8. Using the four FPRS-Net models to create an SRP in 40 cases of 3D facial data required <4 s. ConclusionsOur study demonstrated a new solution for automatically constructing oral clinical 3D facial SRPs. Clinical significanceThis study proposes a novel deep learning algorithm (FPRS-Net) to construct a symmetry reference plane that can reduce workload, shorten the time required for digital design, reduce dependence on expert experience, and improve therapeutic efficiency and effectiveness in dental clinics.

Deep learning for automatic segmentation of vestibular schwannoma: a retrospective study from multi-center routine MRI.

Automatic segmentation of vestibular schwannoma (VS) from routine clinical MRI has potential to improve clinical workflow, facilitate treatment decisions, and assist patient management. Previous work demonstrated reliable automatic segmentation performance on datasets of standardized MRI images acquired for stereotactic surgery planning. However, diagnostic clinical datasets are generally more diverse and pose a larger challenge to automatic segmentation algorithms, especially when post-operative images are included. In this work, we show for the first time that automatic segmentation of VS on routine MRI datasets is also possible with high accuracy. We acquired and publicly release a curated multi-center routine clinical (MC-RC) dataset of 160 patients with a single sporadic VS. For each patient up to three longitudinal MRI exams with contrast-enhanced T1-weighted (ceT1w) (n = 124) and T2-weighted (T2w) (n = 363) images were included and the VS manually annotated. Segmentations were produced and verified in an iterative process: (1) initial segmentations by a specialized company; (2) review by one of three trained radiologists; and (3) validation by an expert team. Inter- and intra-observer reliability experiments were performed on a subset of the dataset. A state-of-the-art deep learning framework was used to train segmentation models for VS. Model performance was evaluated on a MC-RC hold-out testing set, another public VS datasets, and a partially public dataset. The generalizability and robustness of the VS deep learning segmentation models increased significantly when trained on the MC-RC dataset. Dice similarity coefficients (DSC) achieved by our model are comparable to those achieved by trained radiologists in the inter-observer experiment. On the MC-RC testing set, median DSCs were 86.2(9.5) for ceT1w, 89.4(7.0) for T2w, and 86.4(8.6) for combined ceT1w+T2w input images. On another public dataset acquired for Gamma Knife stereotactic radiosurgery our model achieved median DSCs of 95.3(2.9), 92.8(3.8), and 95.5(3.3), respectively. In contrast, models trained on the Gamma Knife dataset did not generalize well as illustrated by significant underperformance on the MC-RC routine MRI dataset, highlighting the importance of data variability in the development of robust VS segmentation models. The MC-RC dataset and all trained deep learning models were made available online.

The significance of partial volume effect on the estimation of hypoxic tumour volume with [18F]FMISO PET/CT

BackgroundThe purpose of this study was to evaluate how a retrospective correction of the partial volume effect (PVE) in [18F]fluoromisonidazole (FMISO) PET imaging, affects the hypoxia discoverability within a gross tumour volume (GTV). This method is based on recovery coefficients (RC) and is tailored for low-contrast tracers such as FMISO. The first stage was the generation of the scanner’s RC curves, using spheres with diameters from 10 to 37 mm, and the same homogeneous activity concentration, positioned in lower activity concentration background. Six sphere-to-background contrast ratios were used, from 10.0:1, down to 2.0:1, in order to investigate the dependence of RC on both the volume and the contrast ratio. The second stage was to validate the recovery-coefficient correction method in a more complex environment of non-spherical lesions of different volumes and inhomogeneous activity concentration. Finally, we applied the correction method to a clinical dataset derived from a prospective imaging trial (DRKS00003830): forty nine head and neck squamous cell carcinoma (HNSCC) cases who had undergone FMISO PET/CT scanning for the quantification of tumour hypoxia before (W0), 2 weeks (W2) and 5 weeks (W5) after the beginning of radiotherapy. Here, PVE was found to cause an underestimation of the activity in small volumes with high FMISO signal.ResultsThe application of the proposed correction method resulted in a statistically significant increase of both the hypoxic subvolume (171% at W0, 691% at W2 and 4.60 × 103% at W5 with p < 0.001) and the FMISO standardised uptake value (SUV) (27% at W0, 21% at W2 and by 25% at W5 with p < 0.001) within the primary GTV.ConclusionsThe proposed PVE-correction method resulted in a statistically significant increase of the hypoxic fraction (HF) with p < 0.001 and demonstrated results in better agreement with published HF data for HNSCC. To summarise, the proposed RC-based correction method can be a useful tool for a retrospective compensation against PVE.

Med-ImageTools: An open-source Python package for robust data processing pipelines and curating medical imaging data

Background Machine learning and AI promise to revolutionize the way we leverage medical imaging data for improving care but require large datasets to train computational models that can be implemented in clinical practice. However, processing large and complex medical imaging datasets remains an open challenge. Methods To address this issue, we developed Med-ImageTools, a new Python open-source software package to automate data curation and processing while allowing researchers to share their data processing configurations more easily, lowering the barrier for other researchers to reproduce published works. Use cases We have demonstrated the efficiency of Med-ImageTools across three different datasets, resulting in significantly reduced processing times. Conclusions The AutoPipeline feature will improve the accessibility of raw clinical datasets on public archives, such as the Cancer Imaging Archive (TCIA), the largest public repository of cancer imaging, allowing machine learning researchers to process analysis-ready formats without requiring deep domain knowledge.

Competing risks and multivariate outcomes in epidemiological and clinical trial research.

Data analysis methods for the study of treatments or exposures in relation to a clinical outcome in the presence of competing risks have a long history, often with inference targets that are hypothetical, thereby requiring strong assumptions for identifiability with available data. Here data analysis methods are considered that are based on single and higher dimensional marginal hazard rates, quantities that are identifiable under standard independent censoring assumptions. These lead naturally to joint survival function estimators for outcomes of interest, including competing risk outcomes, and provide the basis for addressing a variety of data analysis questions. These methods will be illustrated using simulations and Women's Health Initiative cohort and clinical trial data sets, and additional research needs will be described.

Modelling lower-limb peripheral arterial disease using clinically available datasets: impact of inflow boundary conditions on hemodynamic indices for restenosis prediction

Background and ObjectivesThe integration of hemodynamic markers as risk factors in restenosis prediction models for lower-limb peripheral arteries is hindered by fragmented clinical datasets. Computed tomography (CT) scans enable vessel geometry reconstruction and can be obtained at different times than the Doppler ultrasound (DUS) images, which provide information on blood flow velocity. Computational fluid dynamics (CFD) simulations allow the computation of near-wall hemodynamic indices, whose accuracy depends on the prescribed inlet boundary condition (BC), derived from the DUS images. This study aims to: (i) investigate the impact of different DUS-derived velocity waveforms on CFD results; (ii) test whether the same vessel areas, subjected to altered hemodynamics, can be detected independently of the applied inlet BC; (iii) suggest suitable DUS images to obtain reliable CFD results. MethodsCFD simulations were conducted on three patients treated with bypass surgery, using patient-specific DUS-derived inlet BCs recorded at either the same or different time points than the CT scan. The impact of the chosen inflow condition on bypass hemodynamics was assessed in terms of wall shear stress (WSS)-derived quantities. Patient-specific critical thresholds for the hemodynamic indices were applied to identify critical luminal areas and compare the results with a reference obtained with a DUS image acquired in close temporal proximity to the CT scan. ResultsThe main findings indicate that: (i) DUS-derived inlet velocity waveforms acquired at different time points than the CT scan led to statistically significantly different CFD results (p<0.001); (ii) the same luminal surface areas, exposed to low time-averaged WSS, could be identified independently of the applied inlet BCs; (iii) similar outcomes were observed for the other hemodynamic indices if the prescribed inlet velocity waveform had the same shape and comparable systolic acceleration time to the one recorded in close temporal proximity to the CT scan. ConclusionsDespite a lack of standardised data collection for diseased lower-limb peripheral arteries, an accurate estimation of luminal areas subjected to altered near-wall hemodynamics is possible independently of the applied inlet BC. This holds if the applied inlet waveform shares some characteristics – derivable from the DUS report – as one matching the acquisition time of the CT scan.

Unsupervised low-dose CT denoising using bidirectional contrastive network

Background and Objective:Low-dose computed tomography (LDCT) scans significantly reduce radiation exposure, but introduce higher levels of noise and artifacts that compromise image quality and diagnostic accuracy. Supervised learning methods have proven effective in denoising LDCT images, but are hampered by the need for large, paired datasets, which pose significant challenges in data acquisition. This study aims to develop a robust unsupervised LDCT denoising method that overcomes the reliance on paired LDCT and normal-dose CT (NDCT) samples, paving the way for more accessible and practical denoising techniques. Methods:We propose a novel unsupervised network model, Bidirectional Contrastive Unsupervised Denoising (BCUD), for LDCT denoising. This model innovatively combines a bidirectional network structure with contrastive learning theory to map the precise mutual correspondence between the noisy LDCT image domain and the clean NDCT image domain. Specifically, we employ dual encoders and discriminators for domain-specific data generation, and use unique projection heads for each domain to adaptively learn customized embedded representations. We then align corresponding features across domains within the learned embedding spaces to achieve effective noise reduction. This approach fundamentally improves the model’s ability to match features in latent space, thereby improving noise reduction while preserving fine image detail. Results:Through extensive experimental validation on the AAPM-Mayo public dataset and real-world clinical datasets, the proposed BCUD method demonstrated superior performance. It achieved a peak signal-to-noise ratio (PSNR) of 31.387 dB, a structural similarity index measure (SSIM) of 0.886, an information fidelity criterion (IFC) of 2.305, and a visual information fidelity (VIF) of 0.373. Notably, subjective evaluation by radiologists resulted in a mean score of 4.23, highlighting its advantages over existing methods in terms of clinical applicability. Conclusions:This paper presents an innovative unsupervised LDCT denoising method using a bidirectional contrastive network, which greatly improves clinical applicability by eliminating the need for perfectly matched image pairs. The method sets a new benchmark in unsupervised LDCT image denoising, excelling in noise reduction and preservation of fine structural details.

Abstract PS06-03: Differences in ctDNA genomic profiles and outcomes of Black and White patients with metastatic breast cancer: results from a large multicenter consortium

Abstract Background: Black breast cancer patients have a well-documented survival disparity when compared to White patients. It is imperative to explore the reasons for this disparity from both a socioeconomic and biological perspective. Prior studies evaluating somatic genetic differences have primarily focused on tumor tissue analysis, which could be limited by inter- and intra-tumor heterogeneity. Circulating tumor DNA (ctDNA) testing allows for non-invasive detection of these heterogenous somatic mutations in the peripheral blood. We hypothesized that there could be differences in cancer-specific genetic profiles across Black and White patients with metastatic breast cancer (MBC), and that these differences may impact treatment response and clinical outcomes. Methods: This retrospective cohort study included a total of 1327 patients with MBC who were treated at Washington University in St. Louis (N=474 patients), Massachusetts General Hospital (N=412), and Northwestern University (N=441). All patients underwent ctDNA profiling using the commercially available Guardant360 assay. Race information was patient-reported, and ancestry data were not available. Descriptive analysis of clinical variables and pathway variants was performed. Univariate and multivariate analyses were done to evaluate single gene mutations and genetic pathways in both the entire cohort and the hormone-receptor positive (HR+), HER2-negative population (HR+/HER2-). The potential prognostic impact of these somatic mutations was assessed through multivariate analysis in both White and Black patient populations. Results: The cohort of 1327 patients included 1057 White patients (79.6%) and 140 Black patients (10.5%). The remaining patients (9.9%) were not included in the analysis due to low numbers. Black patients had significantly higher rates of GATA3 single nucleotide variants (snv) (OR 2.13, 95% CI 1.07-4.25, P=0.03), PTPN11 snv (OR 7.90, CI 1.10-56.56, P=0.04), and CCND2 copy number variants (cnv) (OR 3.78, CI 1.51-9.45, P=0.004). These alterations were also significantly more common in Black patients in the HR+/HER2- population (N = 812) for GATA3 snv (OR 2.28, CI 1.09-4.76, P=0.028), PTPN11 snv (OR 16.60, CI 1.49-184.93, p=0.022), and CCND2 cnv (OR 4.17, CI 1.02-16.97, P=0.046). Multivariate analysis confirmed that GATA3 snv (OR 2.23, CI 1.12-4.45, P=0.023) and CCND2 cnv (OR 3.97, CI 1.59-9.94, P=0.003) were significantly more common in Black patients in both the full cohort and HR+/HER2- subset. Mutations in the PI3K pathway were more prevalent in White patients, but this difference was not statistically significant in univariate or multivariate analysis. Overall survival (OS) from time of ctDNA collection was significantly worse in Black patients compared to White patients when corrected for lines of therapy and sites of metastasis (median 22 vs. 29 months, log-rank test, P=0.03). Among HR+/HER2- patients specifically, worse prognosis in White patients was associated with TP53 snv (HR 1.55, CI 1.19-2.01, P=0.001), NF1 snv (HR 1.99, CI 1.11-3.56, P=0.021), MYC cnv (HR 1.76, CI 1.09-2.87, P=0.02), PIK3CA cnv pathway (HR 1.69, CI 1.07-2.66, P=0.024), and MYC cnv pathway (HR 1.85, CI 1.15-2.97, P=0.011). No significant gene mutation or pathway association was found in Black patients. Prognostic differences in the cohort based on clinical, pathological, and treatment history were also explored and will be presented. Discussion: To our knowledge, this is the largest clinical genomic dataset examining ctDNA differences across Black and White patients. Our findings revealed that Black patients had higher frequencies of GATA3 snv and CCND2 cnv. The shorter OS observed in Black patients in our study aligns with previous studies and is likely multifactorial, especially given the early separation of the survival curves. Future research should focus on both socioeconomic and genetic factors to explain this disparity. Citation Format: Emily Podany, Lorenzo Foffano, Lorenzo Gerratana, Arielle Medford, Katherine Clifton, Whitney Hensing, Renee Morecroft, Marko Velimirovic, Ami Shah, Carolina REDUZZI, Laura Munoz Arcos, Charles S. Dai, Jennifer Keenan, Elyssa Denault, Foluso Ademuyiwa, Fabio Puglisi, Cynthia Ma, Aditya Bardia, Massimo Cristofanilli, Andrew Davis. Differences in ctDNA genomic profiles and outcomes of Black and White patients with metastatic breast cancer: results from a large multicenter consortium [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS06-03.

Abstract PO5-23-08: Adenylosuccinate lyase is essential for proliferation and mitochondrial function of endocrine therapy resistant breast cancer cells

Abstract Eighty percent of breast cancers express estrogen receptor alpha (ERα, ESR1) at the time of diagnosis. Endocrine therapy (ET) is the standard treatment that either block estrogen mediated ER activation (such as tamoxifen, fulvestrant) or suppress estrogen synthesis (such as exemestane, letrozole, or anastrozole). While ET is initially effective, emergence of ET-resistance is common and possess a major clinical challenge. Hence, there is an ongoing need to develop new treatments for ET-resistant-ERα positive breast cancer. In this study we investigated the role of adenylosuccinate lyase (ADSL), an enzyme of the de-novo purine biosynthesis pathway as a potential target of ET-resistant breast cancer. The protein level of ADSL in two independent ET-resistant cell line model LCC9 and T47D-4HT cells were significantly higher in comparison to its parental ET-sensitive cell lines, MCF7 and T47D cells, respectively. Transient knockdown of ADSL using two independent siRNA in both LCC9 and T47D-4HT cells results in a significant decrease of cell growth, colony, and spheroid formation. Notably, reduced ADSL levels in LCC9 cells prevented the cells from progressing from G1 to S phase of cell cycle with concurrent elevation of cyclin D1/D3, CDK2/4, and cyclin E. On the other hand, in T47D-4HT cells, ADSL reduction caused cells to accumulate in S-phase of cell cycle and showed lower levels of total cyclin D1 protein. Mechanistically, in LCC9 cells (but not in T47D-4HT cells), ADSL depletion induced DNA replication stress which activates (phosphorylate) ataxia-telangiectasia-mutated-and-Rad3-related kinase (ATR) and its major downstream effector checkpoint kinase 1 (Chk1) that in turn failed to de-phosphorylate the inhibitory-phospho-groups of cyclin-dependent kinase 2 (CDK2). In addition, ADSL depletion perturbed the mitochondrial function in both LCC9 and T47D-4HT cells. Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) was lower in both the cells along with the mitochondrial membrane potential. Pertinently, lower concentration of AICAR (5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside), a product of ADSL enzyme function, partially rescued the cell growth and restored mitochondrial membrane potential in ADSL-depleted LCC9 cells. To further understand the global effects of ADSL depletion we performed RNA-sequencing of LCC9 and T47D-4HT cells after ADSL depletion and compared with respective control siRNA transfected cells followed by gene set enrichment analysis. Intriguingly, diverse effect was observed in LCC9 and T47D-4HT cells as very few genes were found to be commonly regulated between these two cell lines. Analysis of clinical data sets revealed high level of ADSL transcript was associated with adverse progression free survival and overall survival in breast cancer patients treated with endocrine therapies. Overall, our findings demonstrate that ADSL-mediated de novo purine synthesis is critical for cellular growth, proliferation, and mitochondrial function of endocrine therapy-resistant breast cancer cells. Therefore, targeting ADSL is a novel potential therapeutic approach for ERα positive ET- resistant breast cancer. Citation Format: Anil Yadav, Lu Jin, Robert Clarke, Surojeet Sengupta. Adenylosuccinate lyase is essential for proliferation and mitochondrial function of endocrine therapy resistant breast cancer cells [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-23-08.

Abstract PO2-02-14: Development of a multi-institutional, photograph-rich clinical dataset to test and validate a novel inflammatory breast cancer (IBC) scoring system

Abstract Purpose: Susan G. Komen, The Inflammatory Breast Cancer (IBC) Research Foundation, and the Milburn Foundation convened a collaborative of patient advocates, clinicians, and researchers and proposed a novel quantitative scoring system for IBC diagnosis. The score includes timing of symptoms, extent of skin edema and erythema/other discoloration, breast and nipple asymmetry, lympho-vascular space invasion, nodal stage, and diffuse extent disease in the breast on breast imaging. Total scores were categorized (&amp;lt; 14 Not IBC, 14-24 Weak Possibility of IBC, 25-41 Strong Possibility of IBC, &amp;gt; 41 Definitely IBC). Herein, we developed a multi-institutional clinical dataset to retrospectively test and validate the proposed scoring system. Methods: IBC (N= 988) and non-IBC (N=322) cases were identified at two institutions with dedicated multi-disciplinary IBC programs, UT MD Anderson Cancer Center (MDA) and Dana-Farber Cancer Institute (DFCI). The DFCI cohort was diagnosed from 1986-2021 and the MDA cohort was diagnosed from 2010-2020. The non-IBC cohorts were identified as consecutive T4b and T4c cases in existing retrospective databases. Photographs were reviewed for available cases. All scoring variables were reviewed with data extractors and standard operating procedures (SOPs) developed for all ambiguous findings and language. A subset of scores were reviewed by IBC clinicians (FL, WW) and discrepancies discussed to inform the SOPs. Missing data imputation using three methods was examined. Area under the curve (AUC) - Receiver operator characteristic (ROC) curves and values, and sensitivity and specificity for the proposed score cut offs were examined. Results: Of 1320 patients currently in the analysis, 421 (31.9%) have missing data on &amp;gt; 1 characteristics and 47 (3.6%) are missing data on &amp;gt; 4. The rubric characteristic most commonly missing was “nipple abnormalities” accounting for 21.0% and 19.3% for IBC and non-IBC cohorts respectively. Unavailable bilateral breast photograph was the primary source of missing data. The highest scoring finding was present in the following percentage of IBC and non-IBC cases, respectively: timing of symptom onset &amp;lt; 3 months (74.1% vs 38.6%), any peau d’orange (61.2% vs 30.1%), clinically apparent enlargement or new asymmetry of breast size (87.8% vs 90.1%), complete or near complete involvement of the breast skin by erythema or discoloration (55.6% vs 25.6%), new nipple inversion (24.6% vs 9.0%), dermal lymphovascular emboli present (17.9% vs 1.8%), diffuse involvement of the breast parenchyma by imaging (72% vs 77.7%). True IBC cases were categorized as Definitely IBC, Strong Possibility, Weak Possibility, not IBC and unknown 19.7%, 49.1%, 0.4%, 0.1%, and 30.7%. True non-IBC were 0.6%, 51.8%, 9.9%, 2.1%, and 35.5%. Random imputation, lowest value imputation, and median value imputation were compared to impute missing data. AUC-ROC values for each method were similar, 0.80 - 0.84. Conclusion: We describe the development of the largest multi-institutional IBC clinical database to date. SOPs for the rubric characteristics were developed. The importance of bilateral photography to assess the score retrospectively is highlighted. The score is associated with a good AUC-ROC but still leaves much overlap among cases between 25 and 42. Further optimization of the current scoring system and association with outcomes are in progress. Citation Format: Filipa Lynce, Samuel Niman, Megumi Kai, Sean Ryan, Elizabeth Troll, Li Li, Kathy Miller, Reshma Jagsi, Ginny Mason, Beth Overmoyer, H. T. Carisa Le-Petross, Faina Nakhlis, Savitri Krishnamurthy, Beth Harrison, Susie X. Sun, Eren Yeh, Jennifer Bellon, Laura Warren, Michael Stauder, Meredith Regan, Wendy Woodward. Development of a multi-institutional, photograph-rich clinical dataset to test and validate a novel inflammatory breast cancer (IBC) scoring system [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-02-14.

Myocardial infarction detection method based on the continuous T-wave area feature and multi-lead-fusion deep features

Objective. Myocardial infarction (MI) is one of the most threatening cardiovascular diseases. This paper aims to explore a method for using an algorithm to autonomously classify MI based on the electrocardiogram (ECG). Approach. A detection method of MI that fuses continuous T-wave area (C_TWA) feature and ECG deep features is proposed. This method consists of three main parts: (1) The onset of MI is often accompanied by changes in the shape of the T-wave in the ECG, thus the area of the T-wave displayed on different heartbeats will be quite different. The adaptive sliding window method is used to detect the start and end of the T-wave, and calculate the C_TWA on the same ECG record. Additionally, the coefficient of variation of C_TWA is defined as the C_TWA feature of the ECG. (2) The multi lead fusion convolutional neural network was implemented to extract the deep features of the ECG. (3) The C_TWA feature and deep features of the ECG were fused by soft attention, and then inputted into the multi-layer perceptron to obtain the detection result. Main results. According to the inter-patient paradigm, the proposed method reached a 97.67% accuracy, 96.59% precision, and 98.96% recall on the PTB dataset, as well as reached 93.15% accuracy, 93.20% precision, and 95.14% recall on the clinical dataset. Significance. This method accurately extracts the feature of the C_TWA, and combines the deep features of the signal, thereby improving the detection accuracy and achieving favorable results on clinical datasets.

Deep learning system for screening AIDS-related cytomegalovirus retinitis with ultra-wide-field fundus images

BackgroundOphthalmological screening for cytomegalovirus retinitis (CMVR) for HIV/AIDS patients is important to prevent lifelong blindness. Previous studies have shown good properties of automated CMVR screening using digital fundus images. However, the application of a deep learning (DL) system to CMVR with ultra-wide-field (UWF) fundus images has not been studied, and the feasibility and efficiency of this method are uncertain. MethodsIn this study, we developed, internally validated, externally validated, and prospectively validated a DL system to detect AIDS-related from UWF fundus images from different clinical datasets. We independently used the InceptionResnetV2 network to develop and internally validate a DL system for identifying active CMVR, inactive CMVR, and non-CMVR in 6960 UWF fundus images from 862 AIDS patients and validated the system in a prospective and an external validation data set using the area under the curve (AUC), accuracy, sensitivity, and specificity. A heat map identified the most important area (lesions) used by the DL system for differentiating CMVR. ResultsThe DL system showed AUCs of 0.945 (95 % confidence interval [CI]: 0.929, 0.962), 0.964 (95 % CI: 0.870, 0.999) and 0.968 (95 % CI: 0.860, 1.000) for detecting active CMVR from non-CMVR and 0.923 (95 % CI: 0.908, 0.938), 0.902 (0.857, 0.948) and 0.884 (0.851, 0.917) for detecting active CMVR from non-CMVR in the internal cross-validation, external validation, and prospective validation, respectively. Deep learning performed promisingly in screening CMVR. It also showed the ability to differentiate active CMVR from non-CMVR and inactive CMVR as well as to identify active CMVR and inactive CMVR from non-CMVR (all AUCs in the three independent data sets >0.900). The heat maps successfully highlighted lesion locations. ConclusionsOur UWF fundus image-based DL system showed reliable performance for screening AIDS-related CMVR showing its potential for screening CMVR in HIV/AIDS patients, especially in the absence of ophthalmic resources.

A seven-step guide to spatial, agent-based modelling of tumour evolution.

Spatial agent-based models are frequently used to investigate the evolution of solid tumours subject to localized cell-cell interactions and microenvironmental heterogeneity. As spatial genomic, transcriptomic and proteomic technologies gain traction, spatial computational models are predicted to become ever more necessary for making sense of complex clinical and experimental data sets, for predicting clinical outcomes, and for optimizing treatment strategies. Here we present a non-technical step by step guide to developing such a model from first principles. Stressing the importance of tailoring the model structure to that of the biological system, we describe methods of increasing complexity, from the basic Eden growth model up to off-lattice simulations with diffusible factors. We examine choices that unavoidably arise in model design, such as implementation, parameterization, visualization and reproducibility. Each topic is illustrated with examples drawn from recent research studies and state of the art modelling platforms. We emphasize the benefits of simpler models that aim to match the complexity of the phenomena of interest, rather than that of the entire biological system. Our guide is aimed at both aspiring modellers and other biologists and oncologists who wish to understand the assumptions and limitations of the models on which major cancer studies now so often depend.

POOLED, LONG-TERM ANALYSIS OF THE GLOBAL SYMPLICITY CLINICAL TRIAL PROGRAM OF RADIOFREQUENCY RENAL DENERVATION FOR THE TREATMENT OF UNCONTROLLED HYPERTENSION

Objective: The Global Symplicity Clinical program is the largest investigation of radiofrequency renal denervation (RDN) to treat uncontrolled hypertension (HTN) to date. Collectively, over 4,000 patients enrolled into Spyral First-in-Man, the Global SYMPLICITY Registry (GSR) DEFINE, and multiple randomized controlled and sham controlled trials including HTN-Japan, Symplicity HTN-3, and the SPYRAL HTN-OFF and ON MED studies. We leveraged the entire Symplicity Global Clinical program patient dataset to develop a multivariate, patient-level, mixed model assessing (1) long-term BP reductions after RDN, and (2) whether specific patient characteristics correlate with future BP reductions after the procedure. Design and method: Repeated BP measurements from each patient in the Global Symplicity Clinical program were analyzed using linear mixed models fitted with 24-h ambulatory systolic (S)BP and office SBP as outcome variables. Baseline SBP, baseline number of antihypertensive (AH) medications, and AH medications over time were included as fixed effects. Additional patient characteristics were included as potential significant fixed effects. To account for multiple BP measurements per patient and multiple studies, nested random effects were included using an auto-regressive correlation structure. Results: The mixed model included data from 4156 patients treated with either the first-generation Symplicity Flex or the next-generation Symplicity Spyral catheter. Estimated, office and 24-h ambulatory SBP changes through 36 months, after adjusting for AH medication effects, were biphasic with a steep reduction after RDN through the first 6 months (Figure), followed by a steady reduction in SBP afterward through 36 months. Higher baseline SBP was correlated with greater BP reductions. History of AF (-3.7 mmHg compared to no AF, p=0.0048) and increasing eGFR at baseline (-0.039 mmHg for every 1 ml/min increase in baseline eGFR, p=0.040) significantly correlated with lower 24-h ambulatory SBP. Conclusions: This is the most comprehensive analysis of radiofrequency RDN to date. Higher baseline SBP was strongly correlated with greater BP reductions. The results demonstrate the durability of RDN as an adjunctive therapy to treat HTN patients. Prospective studies need to validate whether certain patient characteristics are associated with improved response to RDN.

TAI-GAN: A Temporally and Anatomically Informed Generative Adversarial Network for early-to-late frame conversion in dynamic cardiac PET inter-frame motion correction

Inter-frame motion in dynamic cardiac positron emission tomography (PET) using rubidium-82 (82Rb) myocardial perfusion imaging impacts myocardial blood flow (MBF) quantification and the diagnosis accuracy of coronary artery diseases. However, the high cross-frame distribution variation due to rapid tracer kinetics poses a considerable challenge for inter-frame motion correction, especially for early frames where intensity-based image registration techniques often fail. To address this issue, we propose a novel method called Temporally and Anatomically Informed Generative Adversarial Network (TAI-GAN) that utilizes an all-to-one mapping to convert early frames into those with tracer distribution similar to the last reference frame. The TAI-GAN consists of a feature-wise linear modulation layer that encodes channel-wise parameters generated from temporal information and rough cardiac segmentation masks with local shifts that serve as anatomical information. Our proposed method was evaluated on a clinical 82Rb PET dataset, and the results show that our TAI-GAN can produce converted early frames with high image quality, comparable to the real reference frames. After TAI-GAN conversion, the motion estimation accuracy and subsequent myocardial blood flow (MBF) quantification with both conventional and deep learning-based motion correction methods were improved compared to using the original frames. The code is available at https://github.com/gxq1998/TAI-GAN.