Multiple Biomarkers Research Articles

Bayesian and deep-learning models applied to the early detection of ovarian cancer using multiple longitudinal biomarkers.

Ovarian cancer is the most lethal of all gynecological cancers. Cancer Antigen 125 (CA125) is the best-performing ovarian cancer biomarker which however is still not effective as a screening test in the general population. Recent literature reports additional biomarkers with the potential to improve on CA125 for early detection when using longitudinal multimarker models. Our data comprised 180 controls and 44 cases with serum samples sourced from the multimodal arm of UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Our models were based on Bayesian change-point detection and recurrent neural networks. We obtained a significantly higher performance for CA125-HE4 model using both methodologies (AUC 0.971, sensitivity 96.7% and AUC 0.987, sensitivity 96.7%) with respect to CA125 (AUC 0.949, sensitivity 90.8% and AUC 0.953, sensitivity 92.1%) for Bayesian change-point model (BCP) and recurrent neural networks (RNN) approaches, respectively. One year before diagnosis, the CA125-HE4 model also ranked as the best, whereas at 2 years before diagnosis no multimarker model outperformed CA125. Our study identified and tested different combination of biomarkers using longitudinal multivariable models that outperformed CA125 alone. We showed the potential of multivariable models and candidate biomarkers to increase the detection rate of ovarian cancer.

Pleiotropic effects of rifampicin as a mixed-mode perpetrator drug assessed simultaneously using multiple endogenous biomarkers in dogs

Pleiotropic effects of rifampicin as a mixed-mode perpetrator drug assessed simultaneously using multiple endogenous biomarkers in dogs

Investigation of Common Pathways and Putative Biomarker Candidates of Colorectal Cancer and Insomnia by Using Integrative In-Silico Approaches.

Colorectal cancer (CRC) is one of the leading causes of cancer-related mortalities across the globe. Accumulating evidence shows that individuals having sleep disorders such as insomnia are at high risk of developing CRC, yet the association of sleep disorders with CRC risk is still unclear. Here, we investigated the potential molecular connections between CRC and insomnia using integrative in silico approaches. This study aims to explore the potential molecular connections between CRC and insomnia utilizing integrative in-silico methodologies. Gene expression microarray datasets for CRC and insomnia samples were retrieved from the NCBI-GEO database and analyzed using R. Functional enrichment analysis of common differentially expressed genes (DEGs) was performed by the g: Profiler tool. Cytoscape software was used to construct a protein-protein interaction network and hub gene identification. Expression profiles of hub genes in TCGA datasets were also determined, and predicted miRNAs targeting hub genes were analyzed by miRNA target prediction tools. Our results revealed a total of 113 shared DEGs between the CRC and insomnia datasets. Six genes (HSP8A, GAPDH, HSP90AA1, EEF1G, RPS6, and RPLP0), which were also differently expressed in TCGA datasets, were prioritized as hub genes and were found to be enriched in pathways related to protein synthesis. hsa-miR-324-3p, hsa-miR-769-3p, and hsa-miR-16-5p were identified as promising miRNA biomarkers for two diseases. Our in-silico analysis provides promising evidence of the molecular link between CRC and insomnia and highlights multiple potential molecular biomarkers and pathways. Validation of the results by wet lab work can be utilized for novel translational and precision medicine applications to alleviate the public health burden of CRC.

Serum neurofilament light chain levels as multiple sclerosis biomarker

Cumulative neuroaxonal damage is the main cause of disability in multiple sclerosis (MS), although it may remain clinically silent for many years. In this article, the authors describe their research study measuring serum levels of the neurofilament light chain in patients with MS to assess how they correlated to disease severity and degree of disability in an Egyptian population. They demonstrate how serum levels of the neurofilament light chain are associated with clinical and MRI‐related measures of disease activity, supporting their use as an easily accessible, low cost and non‐invasive biomarker.

Comprehensive Genomic Testing: Tissue Stewardship and Best Practices

Biomarker-driven targeted therapies have shaped the oncology treatment landscape for patients with advanced solid tumors over the past decade. Comprehensive genomic profiling (CGP) has played a key role in precision medicine as it enables simultaneous identification of multiple biomarkers to guide cancer diagnosis, therapy selection, and prognostication. As a result, tissue stewardship for successful CGP testing is paramount. In addition, widespread adoption of less invasive sampling techniques leaves less diagnostic tissue for additional or future testing as smaller biopsies are acquired. To help oncology care practitioners overcome these challenges, this paper provides an overview of current genomic testing methodologies and offers guidelines on best practices for tissue stewardship and preanalytic practices for successful CGP testing and efficient turnaround times for laboratory tests.

Quantitative, high-sensitivity measurement of liquid analytes using a smartphone compass

Smartphone ubiquity has led to rapid developments in portable diagnostics. While successful, such platforms are predominantly optics-based, using the smartphone camera as the sensing interface. By contrast, magnetics-based modalities exploiting the smartphone compass (magnetometer) remain unexplored, despite inherent advantages in optically opaque, scattering or auto-fluorescing samples. Here we report smartphone analyte sensing utilizing the built-in magnetometer for signal transduction via analyte-responsive magnetic-hydrogel composites. As these hydrogels dilate in response to targeted stimuli, they displace attached magnetic material relative to the phone’s magnetometer. Using a bilayer hydrogel geometry to amplify this motion allows for sensitive, optics-free, quantitative liquid-based analyte measurements that require neither any electronics nor power beyond that contained within the smartphone itself. We demonstrate this concept with glucose-specific and pH-responsive hydrogels, including glucose detection down to single-digit micromolar concentrations with potential for extension to nanomolar sensitivities. The platform is adaptable to numerous measurands, opening a path towards portable, inexpensive sensing of multiple analytes or biomarkers of interest.

Sensitive detection of multiple blood biomarkers via immunomagnetic exosomal PCR for the diagnosis of Alzheimer's disease.

Blood exosomes are emerging as potential biomarkers for diagnosing brain diseases such as Alzheimer's disease (AD). There is currently a lack of an ultrasensitive technology for identifying core AD biomarkers in blood exosomes to optimize the utility of biomarkers in clinical practice. Here, an immunomagnetic exosomal polymerase chain reaction (iMEP) platform was developed using DNA-conjugated antibodies for the rapid detection of amyloid-β (Aβ1-40 and Aβ1-42) and phosphorylated tau (p-tau396,404 and p-tau181) in clinical blood exosomes. The toehold shift-mediated DNA affinity pulldown eliminates the high detection background, which allows the detection of biomarkers at concentrations down to 10 femtograms per milliliter. With the iMEP assay, exosomal Aβ1-42 was more accurate in differentiating patients with AD from healthy individuals compared with exosomal p-tau181 and p-tau396,404, with a sensitivity of 95.0% and a specificity of 95.0%. The iMEP technique is also adept at quantifying the levels of different exosomal biomarkers associated with disease pathogenesis.

Food insecurity, inflammation, and immune function among older US adults: Findings from the health and Retirement study

BackgroundFood insecurity (FI) is a pressing public health concern among older adults and has been associated with adverse cardiovascular outcomes. Greater systemic inflammation may provide a pathway to explain these associations, but few studies have examined the link between FI and markers of inflammation. Thus, the objective of the present study was to evaluate the associations between FI and multiple inflammatory and immune functioning biomarkers using a nationally representative study of US adults aged > 50 years. MethodParticipants (n = 3,924) were drawn from the longitudinal Health and Retirement Study (HRS). Household FI was assessed using the six-item Short Form Food Security Survey Module from the 2013 HRS Health Care and Nutrition Study. Markers of inflammation (neutrophil–lymphocyte ratio, albumin, hs-CRP, IL6, IL10, IL-1Ra, sTNFR-1, and TGFβ-1) and immune functioning (CMV) were collected during the 2016 HRS Venous Blood Study. Multivariate logistic and linear regression models were used to evaluate associations between household FI and inflammatory and immune functioning biomarkers, adjusting for individual and household sociodemographic characteristics. ResultsThe weighted prevalence of FI was 18.8 %. Age and sex-adjusted mean showed that FI was associated with higher levels of inflammation and impaired immune functioning (Ps-value < 0.05). Older adults with FI had higher mean levels of albumin, hs-CRP, IL6, IL10, IL-1Ra, TGFß-1, and CMV seronegative and borderline (Ps-value < 0.05). Multivariate-adjusted regression model showed that FI was associated with high-risk categories of hs-CRP (OR 1.34, 95 % CI 1.06, 1.68), IL-6 (OR 1.66, 95 % CI 1.28, 2.14), IL-1Ra (OR 0.67, 95 % CI 0.48, 0.93), TGFß-1 (OR 1.87, 95 % CI 1.45, 2.42), seronegativity for CMV (OR 0.48, 95 % CI 0.35, 0.64). ConclusionIn this nationally representative sample of older adults, FI was positively associated with multiple markers of systemic inflammation and impaired immune functioning. Public health efforts that directly work to reduce FI among older adults are warranted and may result in further improvements in their health and well-being.

Assessment and Risk Prediction of Chronic Kidney Disease and Kidney Fibrosis Using Non-Invasive Biomarkers.

Effective management of chronic kidney disease (CKD), a major health problem worldwide, requires accurate and timely diagnosis, prognosis of progression, assessment of therapeutic efficacy, and, ideally, prediction of drug response. Multiple biomarkers and algorithms for evaluating specific aspects of CKD have been proposed in the literature, many of which are based on a small number of samples. Based on the evidence presented in relevant studies, a comprehensive overview of the different biomarkers applicable for clinical implementation is lacking. This review aims to compile information on the non-invasive diagnostic, prognostic, and predictive biomarkers currently available for the management of CKD and provide guidance on the application of these biomarkers. We specifically focus on biomarkers that have demonstrated added value in prospective studies or those based on prospectively collected samples including at least 100 subjects. Published data demonstrate that several valid non-invasive biomarkers of potential value in the management of CKD are currently available.

Enhancing Diabetic Macular Edema Treatment Outcomes: Exploring the ESASO Classification and Structural OCT Biomarkers.

This study assessed the European School of Advanced Studies in Ophthalmology (ESASO) classification's prognostic value for diabetic macular edema (DME) in predicting intravitreal therapy outcomes. In this retrospective, multicenter study, patients aged > 50years with type 1 or 2 diabetes and DME received intravitreal antivascular endothelial growth factor (anti-VEGF) agents (ranibizumab, bevacizumab, and aflibercept) or steroids (dexamethasone). The primary outcome was visual acuity (VA) change post-treatment, termed as functional response, measured 4-6weeks post-third anti-VEGF or 12-16weeks post-steroid injection, stratified by initial DME stage. Of the 560 eyes studied (62% male, mean age 66.7years), 31% were classified as stage 1 (early), 50% stage 2 (advanced), 17% stage 3 (severe), and 2% stage 4 (atrophic). Visual acuity (VA; decimal) improved by 0.12-0.15 decimals in stages 1-2 but only 0.03 decimal in stage 3 (all p < 0.0001) and 0.01 in stage 4 (p = 0.38). Even in eyes with low baseline VA ≤ 0.3, improvements were significant only in stages 1 and 2 (0.12 and 0.17 decimals, respectively). Central subfield thickness (CST) improvement was greatest in stage 3 (-229µm, 37.6%, p < 0.0001), but uncorrelated with VA gains, unlike stages 1 and 2 (respectively: -142µm, 27.4%; - 5µm, 12%; both p < 0.0001). Stage 4 showed no significant CST change. Baseline disorganization of retinal inner layers and focal damage of the ellipsoid zone/external limiting membrane did not influence VA improvement in stages 1 and 2. Treatment patterns varied, with 61% receiving anti-VEGF and 39% dexamethasone, influenced by DME stage, with no significant differences between therapeutic agents. The ESASO classification, which views the retina as a neurovascular unit and integrates multiple biomarkers, surpasses single biomarkers in predicting visual outcomes. Significant functional improvement occurred only in stages 1 and 2, suggesting reversible damage, whereas stages 3 and 4 likely reflect irreversible damage.

Dual-Encoded Affinity Microbead Signature Combinatorial Profiling for Acute Myocardial Infarction High-Sensitivity Diagnosis.

The early diagnosis of acute myocardial infarction (AMI) is dependent on the combined feedback of multiple cardiac biomarkers. However, it remains challenging to precisely detect multicardiac biomarkers in complex blood early due to the lack of sensitive and specific diagnostic indicators and the low abundance and small size of associated biomarkers with high specificity (such as microRNAs). To make matters worse, spectral overlap significantly limits the multiplex analysis of cardiac biomarkers by fluorescent probes, leading to bias in the diagnosis of myocardial infarction. Herein, we developed a method for simultaneous detection of miRNAs and protein biomarkers using size- and color-coded microbeads that carry signature for target capture. We also constructed a microfluidic chip with different spacer arrays that segregate these microbeads in different chip regions according to their size to produce signature signals, indicating the level of different biomarkers. The signals on the microbeads were hugely amplified by catalytic hairpin assembly and rolling circle amplification. Notably, this strategy enables the simultaneous and in situ sensitive profiling of six kinds of biomarkers via adding two different fluorescent labels, removing the limitations of spectral overlap. We envision that the strategy has great potential for application in clinical diagnosis for AMI.

Flexible and Stretchable Electrodes for In Vivo Electrophysiological and Electrochemical Monitoring†

Comprehensive SummaryIn vivo monitoring of bioelectrical and biochemical signals with implanted electrodes has received great interest over the past decades. However, this faces huge challenges because of the severe mechanical mismatch between conventional rigid electrodes and soft biological tissues. In recent years, the emergence of flexible and stretchable electrodes offers seamless and conformable biological‐electronic interfaces and has demonstrated significant advantages for in vivo electrochemical and electrophysiological monitoring. This review first summarizes the strategies for electrode fabrication from the point of substrate and conductive materials. Next, recent progress in electrode functionalization for improved performance is presented. Then, the advances of flexible and stretchable electrodes in exploring bioelectrical and biochemical signals are introduced. Finally, we present some challenges and perspectives ranging from electrode fabrication to application.Key ScientistsIn 2001, a seminal work by Kipke et al. first showed flexible polyimide‐based intracortical electrode arrays.[1] This electrode was further expanded to 252‐channel using microelectromechanical systems technology by Stieglitz et al. in 2009 and achieved large‐scale cortical recordings.[2] Later, Lieber et al. created mesh electronics that allow for seamless and minimally invasive three‐dimensional interpenetration with nerve tissues, opening up unique applications for flexible electronics.[3] Subsequently, Rogers et al. described bioresorbable electronics for transient electrical activity recordings in 2016.[4] And Frank et al. proposed polymer electrode arrays capable of resolving single neurons in 2019.[5] It wasn't until 2020 that a significant breakthrough in biochemical signals monitoring by Peng et al. demonstrated functionalized carbon nanotube fibre bundles for multiple disease biomarkers monitoring.[6] Later on, Mooney et al. established the first fully viscoelastic electrode arrays for neural recordings from the brain and heart in 2021.[7] Recently, Bao et al. presented tissue‐mimicking, stretchable neurotransmitter interfaces for monitoring the brain and gut.[8]

AI-Aided Diagnosis For Neurodegenerative Diseases: Prospects And Challenges

Neurodegenerative disease (ND) represents a chronic disease characterized by loss of neuron function and death, such as Alzheimer's disease, Parkinson's disease, etc. It’s difficult and complex to diagnose ND, due to the requirement for the synthesis of multiple biomarker data, such as genes, proteins, images, etc. The advent of artificial intelligence (AI) technology, particularly through machine learning (ML) and deep learning (DL) algorithms, introduces novel methodologies and tools for ND diagnosis. These methodologies extract pertinent information and patterns from massive, multidimensional, and non-linear data, assisting doctors to render more precise, prompt, and objective assessments. This article provides an overview of the updated implementation status of AI in ND diagnosis, assesses the advantages and contributions of AI to ND diagnosis, as well as the existing limitations and challenges, and offers insights to the future development direction of AI in ND diagnosis.

Abstract 3428: Association of adipokines and inflammation biomarkers with cancer risk in men

Abstract Using pre-diagnosis fasting plasma samples from 4,040 participants of the Shanghai Men’s Health Study which enrolled 61,464 men ages 40-70 during 2001-2006, we prospectively evaluated the associations of adipokines and inflammation biomarkers with cancer risk. Levels of leptin, resistin, adiponectin, plasminogen activator inhibitor type-1 (PAI-1), interleukin-6 (IL-6), interleukin-8 (IL-8), tumor necrosis factor-alpha (TNFa), monocyte chemoattractant protein-1 (MCP-1), and insulin were measured by Luminex assays, and C-reactive protein (CRP), measured by ACE Clinical Chemistry System. Biomarker levels were categorized by tertile distribution. The Cox regression model was applied to derive hazard ratios (HR) and 95% confidence intervals (CI) for cancer risk in association with biomarkers with adjustment for potential confounders. Principle component analysis (PCA) was performed to assess associations with multiple biomarkers, and tertile scores for each component were used in analyses. During 15-years of follow-up, 318 cancer cases were identified, including 85 lung, 49 gastric, 17 liver, and 61 colorectal cancer (CRC) cases. Analyses adjusted for age, education and income showed that 3rd vs 1st tertile of TNFa (HR=1.7, 95%CI=1.0-3.0) and adiponectin (HR=4.0, 95%CI=1.1-14.4) was associated with lung and liver cancer risk, respectively. High levels of leptin, insulin, and CRP were each associated with CRC risk: respective HRs were 2.3 (95%CI=1.1-4.7), 2.1 (95%CI=1.0-4.2), and 2.0 (95%CI=1.1-3.9). After further adjustment for known cancer risk factors, TNFa-lung cancer (HR=1.8, 95%CI=1.1-3.1); leptin-CRC (HR= 3.2, 95%CI=1.3-7.7); insulin-CRC (HR=2.1, 95%CI=1.0-4.5); and CRP-CRC (HR=2.0, 95%CI=1.0-4.0) associations remained statistically significant. The adiponectin-lung cancer association lost significance. PCA identified 4 principal components (PC). High PC1 score (main loadings: TNFa, leptin, and insulin) was associated with CRC risk (HR=2.3, 95%CI=1.1-5.0 for 3rd vs 1st tertile ptrend=0.02). High PC2 score (main loadings: adiponectin, IL-8, MCP-1) and PC3 (main loadings: resistin and PAI-1) were associated with liver cancer risk: HR=5.2 (95%CI=1.1-24.3) for the 3rd vs 1st tertile, ptrend=0.03 and HR=4.7 (95%CI=1.0-21.7), for the 3rd vs 1st tertile, ptrend=0.03, respectively. High PC4 score (main loadings: IL-6 and CRP) was associated with lung cancer risk (HR=1.7, 95%CI=1.0-3.0, for the 3rd vs 1st tertile, ptrend=0.04). These associations, except the PC1-CRC association, lost significance after further adjustment of known cancer risk factors. Our study found that high plasma levels of insulin, CRP, and their combination were associated with CRC risk, high resistin and adiponectin levels with liver cancer risk, and high TNFa levels with lung cancer risk. Some biomarker-cancer associations may be explained by their correlations with known cancer risk factors, especially smoking. Citation Format: Douglas DeMoulin, Hui Cai, Qiuyin Cai, Wei Zheng, Xiao-Ou Shu. Association of adipokines and inflammation biomarkers with cancer risk in men [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3428.

Abstract 4924: Assessing microsatellite instability dominance in colorectal cancer phenotype: A multi-study initiative using multi-target transformers for genomic biomarker prediction

Abstract We used deep learning (DL) to predict multiple genetic biomarkers of colorectal cancer (CRC) from routine histopathology images, aiming to identify morphological changes associated with the mutational profile of tumors. A new Transformer-based prediction model was developed which can simultaneously predict multiple biomarkers from a single histopathology image. We compared the performance of the multi-target approach to conventional single-target models. We analyzed five patient cohorts from the Genetics and Epidemiology of CRC Consortium (GECCO) with colorectal cancer (N=1,385 patients). Our model was trained on an internal data set (N=739) to predict the presence of &amp;gt;200 genetic alterations, focusing on genes with non-silent mutations and mutational signatures, which were assessed using targeted sequencing. We used a 7-fold cross validation and validated the model on an independent, external data set (N=646). Morphological features of microsatellite instability (MSI) are detectable by DL models in histopathology images, making MSI a potential confounding factor. Thus, we also assessed whether our model can predict genetic alterations regardless of MSI status, or whether the model identifies the MSI phenotype. For the majority of biomarkers, the multi-target transformer reached a higher prediction performance as measured by the mean Area Under the Receiver Operating Characteristic (AUROC) Curve and lower standard deviation than single-target DL models. For the external validation set, the mean AUROC of the multi-target versus the single-target model was 0.78 (+/- 0.01) versus 0.72 (+/- 0.06) for BRAF mutations, 0.88 (+/- 0.01) versus 0.86 (+/- 0.03) for hypermutation status, 0.94 (+/- 0.01) versus 0.91 (+/- 0.02) for MSI, 0.86 (+/- 0.01) versus 0.80 (+/- 0.05) for RNF43 mutations and 0.72 (+/- 0.02) versus 0.69 (+/- 0.05) for TP53 mutations, respectively. However, our analysis of the individual prediction scores suggests that the model does not identify the phenotype of the alterations themselves. Instead, the model seems to determine the status of specific genetic alterations based on their correlation with MSI. Consequently, biomarkers which are associated with microsatellite stability have prediction scores negatively correlated with predicted MSI scores. Conversely, biomarkers which are associated with MSI have prediction scores positively correlated with predicted MSI scores. Our results demonstrate potential benefits of multi-target DL models in improving the predictive power and efficacy of histology-based biomarker extraction compared to single-target DL models. However, our results crucially highlight the importance of considering potential confounders, such as MSI status in CRC, and emphasize the need for extensive analysis beyond AUROC in studies focusing on DL-based biomarker detection. Citation Format: Marco Gustav, Marko van Treeck, Zunamys I. Carrero, Chiara M. Loeffler, Nic G. Reitsam, Bruno Märkl, Asier Rabasco Meneghetti, Lisa A. Boardman, Amy J. French, Ellen L. Goode, Andrea Gsur, Stefanie Brezina, Marc J. Gunter, Neil Murphy, Paul Limburg, Stephen Thibodeau, Sebastian Foersch, Robert Steinfelder, Tabitha Harrison, Ulrike Peters, Amanda Phipps, Jakob N. Kather. Assessing microsatellite instability dominance in colorectal cancer phenotype: A multi-study initiative using multi-target transformers for genomic biomarker prediction [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4924.

Abstract 6191: Simultaneous prediction of tumor microenvironment biomarkers from pathology slides using multi-task deep regression

Abstract Understanding the Tumor Microenvironment (TME) is crucial for tailoring personalized cancer treatments and improving patient outcomes. Using deep learning, we can predict properties of the TME from standard Hematoxylin and Eosin (H&amp;E) stained slides, including specific gene expression signatures. While past models were limited to single signature predictions, we explore the potential of multi-task deep learning to identify multiple biomarkers at once, allowing more complex relationships and interactions to be modeled simultaneously. Our study focuses on the prediction of expression patterns of tumor, stromal, and immune cells in breast cancer (BC). We developed a multi-task transformer-based deep regression model, trained in a weakly-supervised setting, to predict biomarkers in the TME from routine H&amp;E-stained pathology slides of a BC cohort (n=876). The performance was evaluated using Pearson’s correlation coefficient (r) and spatial heatmaps generated via gradient-weighted class activation mapping. Our model achieved a significant correlation coefficient above 0.40 (p&amp;lt;0.0001) for all predicted TME biomarkers in the holdout test set (n=176). This includes the biomarkers lymphocyte infiltrating signature score (r=0.46, p&amp;lt;0.0001), tumor infiltrating lymphocyte regional fraction (r=0.48, p&amp;lt;0.0001), leukocyte fraction (r=0.43, p&amp;lt;0.0001), stromal fraction (r=0.42, p&amp;lt;0.0001), and tumor cell proliferation (r=0.47, p&amp;lt;0.0001). Notably, the model's heatmaps highlighted the pertinent cells and histological characteristics for its predictions. These data show that our deep learning model accurately predicts multiple TME biomarkers concurrently from routine pathology slides. This approach offers a promising avenue for cost-effective and efficient biomarker quantification in the TME, with potential applicability across various cancer types. Citation Format: Omar S. El Nahhas, Marta Ligero, Jakob N. Kather. Simultaneous prediction of tumor microenvironment biomarkers from pathology slides using multi-task deep regression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6191.

Abstract 1539: A comprehensive analytical validation framework for antibodies used in the CellScape multiplex immunofluorescence assay

Abstract Background: Multiplex immunofluorescence (mIF) is a powerful tool that enables deep phenotyping of cells within the tumor microenvironment (TME). Careful optimization of multiplex antibody panels for specificity, sensitivity and reproducibility is crucial to achieving robust data and is a precondition for assay repeatability and standardization. We describe our validation procedure for VistaPlex™ multiplex assay panels. Methods: Experiments were performed using the CellScape™ platform for precise spatial multiplexing. The platform enables automated cyclic mIF for readily available fluorophore-conjugated antibodies compatible with fresh frozen or FFPE tissues. The use of high dynamic range imaging further facilitates quantitative phenotyping, while excellent imaging resolution provides single-cell discrimination with high accuracy. To design a ready-to-use panel for immune cell phenotyping, individual antibodies were first screened for specificity and then optimized for labeling sensitivity. Signal-to-noise, inter-assay reproducibility, and intra-assay precision were evaluated across serial sections in a multi-stage testing matrix to validate each antibody in isolation and assembled into a panel. Resulting from this effort, we introduce a 15-plex ready-to-use VistaPlex immune profiling panel for human FFPE tissues. Results: Image analysis of signal-to-noise ratios demonstrated robust performance of multiplexed antibodies across multiple tissue sections and tissue types. Measurements of both signal and quantitative immune phenotyping across technical replicates showed a high degree of intra-assay precision and inter-assay reproducibility. The 15-plex panel was validated on human tonsil, lung, breast, and colon tissues. Conclusions: The CellScape platform enables simultaneous detection of multiple protein biomarkers on a single tissue section for deep immune cell profiling in the TME. Rigorously validated VistaPlex panels further streamline this spatial biology workflow and will aid in the generation of consistent and reliable results for end users. Citation Format: Charles E. Jackson, Arne Christians, Jannik Boog, Matthew H. Ingalls, Xenia Meshik, Adam Northcutt, Kayla E. Cashion, J Spencer Schwarz, Oliver Braubach. A comprehensive analytical validation framework for antibodies used in the CellScape multiplex immunofluorescence assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1539.

Abstract 2507: Drivers of immunotherapy response within the CPI3000+ cohort

Abstract Introduction: Multiple genomic and transcriptomic biomarkers have been associated with response to immune checkpoint (CPI) therapy. These markers include tumor mutational burden (TMB), immune infiltration and HLA loss of heterozygosity. Our previously published meta-analysis of biomarkers included over 1000 patients (CPI1000) across multiple cancer types and demonstrated that clonal TMB and CXCL9 expression were the strongest predictors of response pan-cancer. Biomarkers were combined into a multivariable predictor of response which out-performed TMB in three validation cohorts. However, even with an extensive systematic review, the literature and exploratory biomarkers only accounted for ~60% of the variance explained in CPI response. Approach and Results: Extending the cohort to over 3000 patients from 20 studies, across 11 tumor solid tumor types, we have utilized available genomic and transcriptomic data to further validate the multi-variable predictor of response as well as conduct additional exploratory analysis to uncover the remaining variance explained. The raw data is processed through a robust bioinformatics pipeline and is harmonized across studies. The pipeline developed to process the CPI3000+ cohort will be open-access and available to promote federated learning and collaborations. Firstly, we have expanded the systematic review to include recent biomarkers from literature. The literature review included over 1000 papers, and we identified greater than 200 unique biomarkers which can be quantified and assessed in the CPI3000+ cohort. Data presented on this extensive set of genomic and transcriptomic biomarkers across the cohorts will improve understanding of biomarkers of response to CPI therapy. Second, we will explore novel biomarkers of response including non-classical sources of antigen such as retained introns as well as pathways such as the unfolded protein response. We will also survey the effects of tumor-associated antigens such as those involved in CAR-T cell and vaccine trials in CPI response which was limited in the CPI1000 cohort. We will also be presenting data on emerging checkpoints in late-stage trials across cancer types. Finally, we will be combining the biomarkers into a novel multivariable predictor of response. Conclusions: Results from this work will highlight the varying determinants of immunotherapy response across solid tumor types, offering insight into tumor intrinsic and extrinsic drivers of immunogenicity. We have shown that sources of tumour antigen are a major group of biomarkers for CPI response, here we expand on this further in a larger cohort as well as identifying novel biomarkers to create an improved multivariable predictor of clinical value. Citation Format: Krupa Thakkar, Andrea Castro, Danwen Qian, Charles Swanton, Kevin Litchfield. Drivers of immunotherapy response within the CPI3000+ cohort [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2507.

Abstract 5212: Prognostic biomarkers for breast cancer: Metabolic and inflammatory pathways

Abstract Objective: To assess the prognostic significance of molecular biomarkers, particularly metabolic and Inflammatory pathways using samples from a prospective study of breast cancer patients undergoing adjuvant radiotherapy (RT). Methods: We used ultra-performance liquid chromatography-mass spectrometry (UPLC-MS/MS) for measuring urine metabolites. Following normalization to osmolality, log transformation, and imputation of missing values, we used Welch’s two-sample t-test to identify biochemicals that differed significantly by progression status. We measured pre- and post-RT plasma high-sensitivity C-reactive protein (hsCRP) levels. Progression-free survival (PFS) was calculated from the date of diagnosis to the date of disease progression or last follow-up. Univariable and multivariable Cox proportional hazards regression models assessed the associations between hsCRP and PFS. Results: In 120 breast cancer patients, carbohydrate metabolism, branch-chain amino acid (BCAA) metabolism, phosphatidylcholine metabolism, arginine metabolism, oxidative stress-related metabolites, androgenic steroids, and nucleotide metabolism had the most notable differences in patients with progression versus progression-free cases. In 488 breast cancer patients, RT significantly increased plasma hsCRP levels, and post-RT hsCRP levels were significantly higher in deceased vs. alive patients (p=0.031). In multivariable models, hsCRP had a prognostic value for all-stage and stage 0-II patients. Conclusions: In our study of breast cancer patients receiving adjuvant RT, multiple metabolite biomarkers and post-RT plasma CRP levels were associated with a worse prognosis. BCAAs and fatty acids regulate cell metabolism by affecting mitochondrial function and inflammation signals. Thus, our findings suggest a potential role of metabolic and inflammatory pathways in breast cancer prognosis. These molecules may have potential utility as prognostic biomarkers for breast cancer patients undergoing adjuvant RT and targets for intervention. Keywords: breast cancer, metabolomics, inflammation, prognosis Citation Format: Alexandra N. McMahon, Cristiane Takita, Jean L. Wright, Isildinha M. Reis, Eunkyung Lee, George Yang, Jennifer J. Hu. Prognostic biomarkers for breast cancer: Metabolic and inflammatory pathways [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5212.

A wearable microfluidic system for efficient sweat collection and real-time detection

Sweat is an important biofluid with rich physiological information that can evaluate human health condition. Wearable sweat sensors have received widespread attention in recent years due to the benefits of non-invasive, continuous, and real-time monitoring. Currently, an efficient device integrating sweat collection and detection is still needed. Here, a wearable sweat microfluidic system was fabricated for real-time collection and analysis of sweat. The fabricated microfluidic system consisted of four layers, including a skin adhesive layer, a microfluidic layer, an electrode layer, and a capping layer. The sweat collection rate was around 0.79 μL/min, which demonstrated efficient sweat sampling, storage, and refreshing capabilities. Simultaneous detection of multiple sweat biomarkers was achieved with a screen-printed sweat sensing array, which could realize high-precision detection of Na+, K+, and glucose. Moreover, the sensing array also showed good repeatability and stability, with a relative standard deviation of sensitivity of less than 5%. Additionally, human testing was conducted to demonstrate that this microfluidic system can continuously monitor Na+, K+, and glucose in subjects' sweat during exercise, which showed high potential for non-invasive human health monitoring.