Metabolic Data Research Articles

Integrating Machine Learning in Metabolomics: A Path to Enhanced Diagnostics and Data Interpretation.

Metabolomics, leveraging techniques like NMR and MS, is crucial for understanding biochemical processes in pathophysiological states. This field, however, faces challenges in metabolite sensitivity, data complexity, and omics data integration. Recent machine learning advancements have enhanced data analysis and disease classification in metabolomics. This study explores machine learning integration with metabolomics to improve metabolite identification, data efficiency, and diagnostic methods. Using deep learning and traditional machine learning, it presents advancements in metabolic data analysis, including novel algorithms for accurate peak identification, robust disease classification from metabolic profiles, and improved metabolite annotation. It also highlights multiomics integration, demonstrating machine learning's potential in elucidating biological phenomena and advancing disease diagnostics. This work contributes significantly to metabolomics by merging it with machine learning, offering innovative solutions to analytical challenges and setting new standards for omics data analysis.

Genomic factors shape carbon and nitrogen metabolic niche breadth across Saccharomycotina yeasts.

Organisms exhibit extensive variation in ecological niche breadth, from very narrow (specialists) to very broad (generalists). Two general paradigms have been proposed to explain this variation: (i) trade-offs between performance efficiency and breadth and (ii) the joint influence of extrinsic (environmental) and intrinsic (genomic) factors. We assembled genomic, metabolic, and ecological data from nearly all known species of the ancient fungal subphylum Saccharomycotina (1154 yeast strains from 1051 species), grown in 24 different environmental conditions, to examine niche breadth evolution. We found that large differences in the breadth of carbon utilization traits between yeasts stem from intrinsic differences in genes encoding specific metabolic pathways, but we found limited evidence for trade-offs. These comprehensive data argue that intrinsic factors shape niche breadth variation in microbes.

Machine learning enables identification of an alternative yeast galactose utilization pathway

How genomic differences contribute to phenotypic differences is a major question in biology. The recently characterized genomes, isolation environments, and qualitative patterns of growth on 122 sources and conditions of 1,154 strains from 1,049 fungal species (nearly all known) in the yeast subphylum Saccharomycotina provide a powerful, yet complex, dataset for addressing this question. We used a random forest algorithm trained on these genomic, metabolic, and environmental data to predict growth on several carbon sources with high accuracy. Known structural genes involved in assimilation of these sources and presence/absence patterns of growth in other sources were important features contributing to prediction accuracy. By further examining growth on galactose, we found that it can be predicted with high accuracy from either genomic (92.2%) or growth data (82.6%) but not from isolation environment data (65.6%). Prediction accuracy was even higher (93.3%) when we combined genomic and growth data. After the GALactose utilization genes, the most important feature for predicting growth on galactose was growth on galactitol, raising the hypothesis that several species in two orders, Serinales and Pichiales (containing the emerging pathogen Candida auris and the genus Ogataea, respectively), have an alternative galactose utilization pathway because they lack the GAL genes. Growth and biochemical assays confirmed that several of these species utilize galactose through an alternative oxidoreductive D-galactose pathway, rather than the canonical GAL pathway. Machine learning approaches are powerful for investigating the evolution of the yeast genotype-phenotype map, and their application will uncover novel biology, even in well-studied traits.

Comparative physiological, biochemical, metabolomic, and transcriptomic analyses reveal the formation mechanism of heartwood for Acacia melanoxylon

Acacia melanoxylon is well known as a valuable commercial tree species owing to its high-quality heartwood (HW) products. However, the metabolism and regulatory mechanism of heartwood during wood development remain largely unclear. In this study, both microscopic observation and content determination proved that total amount of starches decreased and phenolics and flavonoids increased gradually from sapwood (SW) to HW. We also obtained the metabolite profiles of 10 metabolites related to phenolics and flavonoids during HW formation by metabolomics. Additionally, we collected a comprehensive overview of genes associated with the biosynthesis of sugars, terpenoids, phenolics, and flavonoids using RNA-seq. A total of ninety-one genes related to HW formation were identified. The transcripts related to plant hormones, programmed cell death (PCD), and dehydration were increased in transition zone (TZ) than in SW. The results of RT-PCR showed that the relative expression level of genes and transcription factors was also high in the TZ, regardless of the horizontal or vertical direction of the trunk. Therefore, the HW formation took place in the TZ for A. melanoxylon from molecular level, and potentially connected to plant hormones, PCD, and cell dehydration. Besides, the increased expression of sugar and terpenoid biosynthesis-related genes in TZ further confirmed the close connection between terpenoid biosynthesis and carbohydrate metabolites of A. melanoxylon. Furthermore, the integrated analysis of metabolism data and RNA-seq data showed the key transcription factors (TFs) regulating flavonoids and phenolics accumulation in HW, including negative correlation TFs (WRKY, MYB) and positive correlation TFs (AP2, bZIP, CBF, PB1, and TCP). And, the genes and metabolites from phenylpropanoid and flavonoid metabolism and biosynthesis were up-regulated and largely accumulated in TZ and HW, respectively. The findings of this research provide a basis for comprehending the buildup of metabolites and the molecular regulatory processes of HW formation in A. melanoxylon.

The Obese Taste Bud study: Objectives and study design.

Taste modifies eating behaviour, impacting body weight and potentially obesity development. The Obese Taste Bud (OTB) Study is a prospective cohort study launched in 2020 at the University of Leipzig Obesity Centre in cooperation with the HI-MAG Institute. OTB will test the hypothesis that taste cell homeostasis and taste perception are linked to obesity. Here, we provide the study design, data collection process and baseline characteristics. Participants presenting overweight, obesity or normal weight undergo taste and smell tests, anthropometric, and taste bud density (TBD) assessment on Day 1. Information on physical and mental health, eating behaviour, physical activity, and dental hygiene are obtained, while biomaterial (saliva, tongue swap, blood) is collected in the fasted state. Further blood samples are taken during a glucose tolerance test. A stool sample is collected at home prior to Day 2, on which a taste bud biopsy follows dental examination. A subsample undergoes functional magnetic resonance imaging while exposed to eating-related cognitive tasks. Follow-up investigations after conventional weight loss interventions and bariatric surgery will be included. Initial results show that glycated haemoglobin levels and age are negatively associated with TBD, while an unfavourable metabolic profile, current dieting, and vegan diet are related to taste perception. Olfactory function negatively correlates with age and high-density lipoprotein cholesterol. Initial findings suggest that metabolic alterations are relevant for taste and smell function and TBD. By combining omics data from collected biomaterial with physiological, metabolic and psychological data related to taste perception and eating behaviour, the OTB study aims to strengthen our understanding of taste perception in obesity.

Impact of Distinct Antiandrogen Exposures on the Plasma Metabolome in Feminizing Gender-affirming Hormone Therapy.

The plasma metabolome is a functional readout of metabolic activity and is associated with phenotypes exhibiting sexual dimorphism, such as cardiovascular disease. Sex hormones are thought to play a key role in driving sexual dimorphism. Gender-affirming hormone therapy (GAHT) is a cornerstone of transgender care, but longitudinal changes in the plasma metabolome with feminizing GAHT have not been described. Blood samples were collected at baseline and after 3 and 6 months of GAHT from transgender women (n = 53). Participants were randomized to different anti-androgens, cyproterone acetate or spironolactone. Nuclear magnetic resonance-based metabolomics was used to measure 249 metabolic biomarkers in plasma. Additionally, we used metabolic biomarker data from an unrelated cohort of children and their parents (n = 3748) to identify sex- and age-related metabolite patterns. We identified 43 metabolic biomarkers altered after 6 months in both anti-androgen groups, most belonging to the very low- or low-density lipoprotein subclasses, with all but 1 showing a decrease. We observed a cyproterone acetate-specific decrease in glutamine, glycine, and alanine levels. Notably, of the metabolic biomarkers exhibiting the most abundant "sex- and age-related" pattern (higher in assigned female children and lower in assigned female adults, relative to assigned males), 80% were significantly lowered after GAHT, reflecting a shift toward the adult female profile. Our results suggest an anti-atherogenic signature in the plasma metabolome after the first 6 months of feminizing GAHT, with cyproterone acetate also reducing specific plasma amino acids. This study provides novel insight into the metabolic changes occurring across feminizing GAHT.

A prognostic risk model based on lactate metabolism and transport-related lncRNAs for gastric adenocarcinoma

Background Increased lactate levels and metastasis in tumours are strongly associated with dismal outcomes. But prognostic value of lactate metabolism and transport-related lncRNAs in gastric adenocarcinoma (GA) patients remains unaddressed. Methods Gene expression data of GA were provided by The Cancer Genome Atlas. Lactate metabolism and transport-related gene data were accessed from GSEA. LncRNAs related to lactate metabolism and transport were identified by correlation analysis. A prognostic model was built by regression analysis. Validity of prognostic model was confirmed through survival analysis and receiver operating characteristic (ROC) curve. Immunity of each risk group was evaluated by immune correlation analysis .LncRNA-mRNA network was built by correlation analysis using Cytoscape software. Results A 12-gene prognostic model based on lactate metabolism and transport-related lncRNAs was built in GA. Median riskscore was utilized to classify GA samples into high- and low-risk groups. Survival analysis and ROC curves demonstrated validity of prognostic model. Most immune checkpoint molecules and TIDE scores were lower in the low-risk group. LINC01303 and LINC01545 may be the key prognostic factors in patients with GA. Conclusion This study successfully built a prognostic model of lactate metabolism and transport-related lncRNAs in GA. The findings guide prognostic management of GA patients.

Transcriptional, metabolic, physiological and developmental responses to nitrogen limitation in switchgrass (Panicum virgatum)

Sustainable switchgrass production (Panicum virgatum) as a bioenergy crop hinges partly on efficiently using soil macronutrients, especially nitrogen (N). In this study, switchgrass plants were grown under four different N conditions (6.0 mM KNO3, N-replete control; 2.0 mM, mild limitation; 0.6 mM, moderate limitation; and 0.2 mM, severe limitation). Subsequently, the physiological, metabolic, and transcriptomic responses of switchgrass roots and shoots to N limitation were determined.Moderate N limitation marked a tipping point for large changes in plant growth, root-to-shoot ratio, root system architecture (root surfaces and specific root length - SRL) and total nitrogen content. Integrating metabolic data and transcriptomic analyses revealed that N limitation reduced switchgrass photosynthetic capacity and carbon(C)-fixation activities. Our results indicate that switchgrass balances C-fixation with N-assimilation and N transport and recycles N-compounds by rerouting C-flux from glycolysis, the oxidative branch of the pentose phosphate pathway (OPPP), and from the tricarboxylic acid (TCA) cycle in an organ-specific manner. The energy and reduction power so generated, and C-skeletons appear to be directed towards N uptake, biosynthesis of energy storage compounds with high C/N ratio such as sucrose, non-N-containing lipids, and various branches of secondary metabolism. Taken together, the findings of this study provide new insights into how switchgrass plants adapt to N limitation.

Quantifying the physical demands undertaken by offshore wind technicians during a working day and casualty evacuations.

Offshore wind technicians (WTs) have been identified as having an occupation with high physical demands. To characterise the physical demands of WTs, during the working day and when performing casualty rescues. Data collection consisted of two components. Component A, 14 WTs (n = 13 male, 1 female) were monitored over three consecutive working days. The data collected consisted of: a questionnaire; heart rate and movement data; and accelerometer data. Component B, 5 WTs (n = 5 male) were assessed undertaking two simulated casualty evacuations. Heart rate and metabolic data were collected. A working day lasted 7.5 hrs to 9 hrs, with WTs covering, on average, 4.7 km on the turbine. WTs spent 28% of their time undertaking "light activity", 69% "moderate activity" and 3% working "vigorously". The greatest amount of time in a day was spent working on manual handling tasks such as torque and tensioning (up to 4 hours in total). Both rescues were performed largely aerobically, with similar cardiac and metabolic demands (14.56 (4.84) mL.kg-1.min-1 vs. 17.07 (3.54) mL.kg-1.min-1). Higher RPE values were reported for the rescue from within the Hub (median (range), 13 (12-14)) compared to down the ladder (11 (7-12)), likely due to a greater requirement to manoeuvre the casualty in tight spaces. The results presented characterise the physical requirements of WTs during the working day and two casualty evacuations. These data play an important role in supporting the job task analysis for WTs working in the offshore wind industry.

In Vitro Characterization and Rescue of VX Metabolism in Human Liver Microsomes.

Venomous agent X (VX) is an organophosphate acetylcholinesterase (AChE) inhibitor, and although it is one of the most toxic AChE inhibitors known, the extent of metabolism in humans is not currently well understood. The known metabolism in humans is limited to the metabolite identification from a single victim of the Osaka poisoning in 1994, which allowed for the identification of several metabolic products. VX has been reported to be metabolized in vitro by paraoxonase-1 and phosphotriesterase, although their binding constants are many orders of magnitude above the LD50, suggesting limited physiologic relevance. Using incubation with human liver microsomes (HLMs), we have now characterized the metabolism of VX and the formation of multiple metabolites as well as identified a Food and Drug Administration-approved drug [ethylenediaminetetraacetic acid (EDTA)] that enhances the metabolic rate. HLM incubation alone shows a pronounced increase in the metabolism of VX compared with buffer, suggesting that cytochrome P450-mediated metabolism of VX is occurring. We identified a biphasic decay with two distinct rates of metabolism. The enhancement of VX metabolism in multiple buffers was assessed to attempt to mitigate the effect of hydrolysis rates. The formation of VX metabolites was shown to be shifted with HLMs, suggesting a pathway enhancement over simple hydrolysis. Additionally, our investigation of hydrolysis rates in various common buffers used in biologic assays discovered dramatic differences in VX stability. The new human in vitro VX metabolic data reported points to a potential in vivo treatment strategy (EDTA) for rescue in individuals that are poisoned though enhancement of metabolism alongside existing treatments. SIGNIFICANCE STATEMENT: Venomous agent X (VX) is a potent acetylcholinesterase inhibitor and chemical weapon. To date, we do not possess a clear understanding of its metabolism in humans that would assist us in treating those exposed to it. This study now describes the human liver microsomal metabolism of VX and identifies ethylenediaminetetraacetic acid, which appears to enhance the rate of metabolism. This may provide a potential treatment option for human VX poisoning.

Genetics and diet shape the relationship between islet function and whole body metabolism.

Despite the fact that genes and the environment are known to play a central role in islet function, our knowledge of how these parameters interact to modulate insulin secretory function remains relatively poor. Presently, we performed ex vivo glucose-stimulated insulin secretion and insulin content assays in islets of 213 mice from 13 inbred mouse strains on chow, Western diet (WD), and a high-fat, carbohydrate-free (KETO) diet. Strikingly, among these 13 strains, islets from the commonly used C57BL/6J mouse strain were the least glucose responsive. Using matched metabolic phenotyping data, we performed correlation analyses of isolated islet parameters and found a positive correlation between basal and glucose-stimulated insulin secretion, but no relationship between insulin secretion and insulin content. Using in vivo metabolic measures, we found that glucose tolerance determines the relationship between ex vivo islet insulin secretion and plasma insulin levels. Finally, we showed that islet glucose-stimulated insulin secretion decreased with KETO in almost all strains, concomitant with broader phenotypic changes, such as increased adiposity and glucose intolerance. This is an important finding as it should caution against the application of KETO diet for beta-cell health. Together these data offer key insights into the intersection of diet and genetic background on islet function and whole body glucose metabolism.NEW & NOTEWORTHY Thirteen strains of mice on chow, Western diet, and high-fat, carbohydrate-free (KETO), correlating whole body phenotypes to ex vivo pancreatic islet functional measurements, were used. The study finds a huge spectrum of functional islet responses and insulin phenotypes across all strains and diets, with the ubiquitous C57Bl/6J mouse exhibiting the lowest secretory response of all strains, highlighting the overall importance of considering genetic background when investigating islet function. Ex vivo basal and stimulated insulin secretion are correlated in the islet, and KETO imparts widescale downregulation of islet insulin secretion.

Teaching a hands-on session on simultaneous Positron Emission Tomography and Magnetic Resonance image acquisition and quantification in a clinical setting

Positron Emission Tomography/Magnetic Resonance (PET/MR) imaging, is a novel imaging modality that combines the capabilities of two powerful imaging techniques in a single acquisition. This unique integration allows for simultaneous acquisition of both metabolic and structural data within a single imaging session. In this sense, PET/MR offers a comprehensive and innovative approach to medical imaging, but accompanied by its intrinsic physics and engineering complexity, involving intricate synchronization of high-performance detectors, electromagnetic shielding, and sophisticated correction algorithms, among others. In this hands-on session performed during the 6th INFIERI summer school (Madrid, Spain), we described an introduction to simultaneous PET/MR image acquisition and quantification in a real clinical to students from diverse backgrounds in physics and engineering. We first presented the fundamental aspects of PET/MR image acquisition, including the main theoretical characteristics of PET/MR physics and electronics, visiting the PET/MR facilities at Hospital Universitario HM Puerta del Sur (HM Hospitales, Móstoles, Madrid), and discussing the main PET/MR clinical and research applications. Then, we followed with a practical session on PET/MR image quantification, preparing the students to understand and solve examples on image reconstruction, multimodal data visualization, inhomogeneity correction, image registration, and attenuation correction exploiting the simplicity and usefulness of 3D Slicer.

Development and External Validation of Machine Learning Models for Diabetic Microvascular Complications: Cross-Sectional Study With Metabolites.

Diabetic kidney disease (DKD) and diabetic retinopathy (DR) are major diabetic microvascular complications, contributing significantly to morbidity, disability, and mortality worldwide. The kidney and the eye, having similar microvascular structures and physiological and pathogenic features, may experience similar metabolic changes in diabetes. This study aimed to use machine learning (ML) methods integrated with metabolic data to identify biomarkers associated with DKD and DR in a multiethnic Asian population with diabetes, as well as to improve the performance of DKD and DR detection models beyond traditional risk factors. We used ML algorithms (logistic regression [LR] with Least Absolute Shrinkage and Selection Operator and gradient-boosting decision tree) to analyze 2772 adults with diabetes from the Singapore Epidemiology of Eye Diseases study, a population-based cross-sectional study conducted in Singapore (2004-2011). From 220 circulating metabolites and 19 risk factors, we selected the most important variables associated with DKD (defined as an estimated glomerular filtration rate <60 mL/min/1.73 m2) and DR (defined as an Early Treatment Diabetic Retinopathy Study severity level ≥20). DKD and DR detection models were developed based on the variable selection results and externally validated on a sample of 5843 participants with diabetes from the UK biobank (2007-2010). Machine-learned model performance (area under the receiver operating characteristic curve [AUC] with 95% CI, sensitivity, and specificity) was compared to that of traditional LR adjusted for age, sex, diabetes duration, hemoglobin A1c, systolic blood pressure, and BMI. Singapore Epidemiology of Eye Diseases participants had a median age of 61.7 (IQR 53.5-69.4) years, with 49.1% (1361/2772) being women, 20.2% (555/2753) having DKD, and 25.4% (685/2693) having DR. UK biobank participants had a median age of 61.0 (IQR 55.0-65.0) years, with 35.8% (2090/5843) being women, 6.7% (374/5570) having DKD, and 6.1% (355/5843) having DR. The ML algorithms identified diabetes duration, insulin usage, age, and tyrosine as the most important factors of both DKD and DR. DKD was additionally associated with cardiovascular disease history, antihypertensive medication use, and 3 metabolites (lactate, citrate, and cholesterol esters to total lipids ratio in intermediate-density lipoprotein), while DR was additionally associated with hemoglobin A1c, blood glucose, pulse pressure, and alanine. Machine-learned models for DKD and DR detection outperformed traditional LR models in both internal (AUC 0.838 vs 0.743 for DKD and 0.790 vs 0.764 for DR) and external validation (AUC 0.791 vs 0.691 for DKD and 0.778 vs 0.760 for DR). This study highlighted diabetes duration, insulin usage, age, and circulating tyrosine as important factors in detecting DKD and DR. The integration of ML with biomedical big data enables biomarker discovery and improves disease detection beyond traditional risk factors.

Long-term pasireotide therapy in acromegaly: extensive real-life experience of a referral center

PurposePasireotide is a novel therapeutic option for patients with acromegaly resistant to first-generation somatostatin receptor ligands. To date, real-life data are still scant, therefore, the aim of the current study is to evaluate the impact of long-term pasireotide therapy on disease control, pituitary tumor size, gluco-insulinemic and lipid profile in a real-life setting.MethodsRetrospective study of data prospectively collected, evaluating hormonal, tumoral, and metabolic data of 28 patients with acromegaly administered with pasireotide in a pituitary tertiary referral center.ResultsWithin the first 12 months of treatment, 70.4% of patients achieved normal IGF-I levels, which was maintained at 36-month evaluation in these responders patients. Patients who started with pasireotide 60 mg monthly exhibited significantly lower IGF-I levels after 36 months (p = 0.05) as compared to patients administered first with pasireotide 20 or 40 mg monthly. The maximal tumoral diameter was significantly decreased after 12 months of pasireotide (p < 0.001) and a further reduction was registered throughout the following months, with 41.2% of patients achieving a significant reduction (> 25% of baseline measurement) after 36 months of treatment. Fasting glucose significantly increased during the first 6 months (p < 0.001) with a gradual rise in diabetes prevalence during the following months, resulting diabetes prevalence after 36 months of pasireotide significantly increased compared to baseline (p = 0.003), although with glycated hemoglobin levels within the normal range. Diabetes was managed using oral glucose-lowering drugs or glucagon-like peptide 1 agonists, with no patient requiring insulin therapy. Pasireotide improved lipid profile, mainly during the first 12 months of treatment, by increasing HDL and decreasing triglycerides levels.ConclusionPasireotide is effective and safe in the long-term. Hyperglycemia is a common event and is manageable even without insulin treatment.

Abstract 4428: Integrated multi-omics analysis reveals systemic and localized metabolic disruptions in colorectal cancer

Abstract Despite technological advances in molecular medicine over the last 30 years, no single approach has proven to be sufficient to meet the diagnostic needs in cancer care. Here we use a multiomic approach leveraging proteomic, metabolomic, and transcriptomic technologies to study the metabolic shifts that occur during colorectal cancer (CRC) progression, Our integrated analysis identified systemic changes to beta oxidation pathways and localized changes to tyrosine metabolism within the tumor, creating a mechanistic, actionable description of the core CRC metabolic program. We conducted untargeted proteomics and metabolomics profiling on serum samples from CRC patients (n=10) and healthy controls (n=10). Our serum proteomics data is generated through sample enrichment on the Seer Proteograph system followed by LC/MS analysis with the Thermo Orbitrap Astral. Our metabolomics approach employs LC/MS assays for unbiased profiling of the serum metabolome, exposome, and lipidome. We supplemented this discovery work with both targeted serum and tumor-specific approaches including targeted proteomics to quantify select targets, inflammatory profiling with Alamar proteomics, public transcriptomics data from TCGA, and in vitro metabolic flux data. These large and diverse datasets were then integrated through joint pathway analysis and network integration. The global serum proteomics and metabolomics profiles generated show distinct separation between healthy and diseased individuals. Joint pathway analysis of these discovery datasets highlighted enrichment in beta oxidation pathways and tyrosine metabolism, among others. Interestingly, the Alamar inflammatory panel revealed that many of the analytes in the tyrosine metabolism pathway were well correlated with inflammatory status, while the beta oxidation signature had a lower correlation. To determine the relationship of these systemic findings from serum to tumor metabolism itself, we integrated tumor-specific transcriptomics data and found alterations to tyrosine metabolism with differences in tyrosine aminotransferase expression. The beta oxidation related genes, on the other hand, were not concordant with the serum findings. However, our in vitro metabolic flux studies have shown beta oxidation in CRC cells is upregulated to provide additional fuel for oxidative phosphorylation. This result suggests that beta oxidation may not be transcriptionally regulated in the tumor but rather a consequence of organismal metabolic rewiring. The results of this work, which is currently expanding into a larger and more diverse patient population, underscores the power of multiomic profiling for enhancing our understanding of metabolic dysregulation in CRC. Ultimately, a comprehensive model of the molecular alterations in CRC will yield a better understanding of tumor phenotypes and inform better diagnostics and therapies. Citation Format: Gary Patti, Ethan Stancliffe, Adam Richardson, Ashima Mehta, Monil Gandhi, Kevin Cho. Integrated multi-omics analysis reveals systemic and localized metabolic disruptions in colorectal cancer [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 4428.

Abstract 2536: Multimodal stratification of predictive biomarkers in head and neck cancers: A focus on cytokine-based immunotherapy

Abstract Background Immunotherapy, particularly cytokine-based therapy, has been gaining traction in the treatment of head and neck cancers. Cytokines are small proteins that play a crucial role in the immune response to cancer. However, not all patients respond to cytokine-based immunotherapies, and some may experience severe side effects. The ability to predict which patients are most likely to benefit from these therapies could greatly improve the efficacy and tolerability of treatment. In this context, the study of predictive biomarkers, such as the expression of immune checkpoint molecules like PD-L1 and TIGIT, and the profiling of cytokines within the tumor microenvironment, has become crucial. These biomarkers could provide valuable information about the patient's immune response to the tumor and their likelihood of responding to cytokine-based immunotherapies. Method A multimodal approach to stratify predictive biomarkers in head and neck cancers was used. The methodology was founded on combining metabolic, transcriptomic, and proteomic data to offer a comprehensive understanding of potential biomarkers. The data from different modalities were integrated using bioinformatics and machine learning algorithms. This comprehensive dataset was analyzed to identify multimodal biomarker signatures that could predict patient responses to cytokine-based therapies. Result Preliminary finding showed significant heterogeneity in the expression of PD-L1, TIGIT, and various cytokines across the tumor microenvironment. This underscored the complexity of head and neck cancers. Certain multimodal biomarker signatures, which included specific patterns of immune checkpoint molecule expression, cytokine profiles, and immune cell infiltration, correlated robustly with patient responses to cytokine-based therapies. Conclusion Our study demonstrates the considerable potential of a multimodal stratification approach in deciphering the complexity of head and neck cancers and predicting responses to cytokine-based immunotherapies. By integrating metabolomic, transcriptomic, and proteomic data, we identified unique biomarker signatures that strongly correlated with therapy responses. This work underscores the invaluable role of integrated predictive biomarker profiles in enhancing the precision of immunotherapy. Citation Format: Corinne Ramos, Richard Ruez, Sandra Delebecq, Victor Senechal. Multimodal stratification of predictive biomarkers in head and neck cancers: A focus on cytokine-based immunotherapy [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 2536.

Abstract 1823: In vitro evaluation of novel pyrimidine nucleosides for the treatment of pancreatic cancer

Abstract The purpose of the study is to synthesize and characterize novel 5-FU analogs (AGY1 and AGY2) and investigate their anticancer activity against pancreatic cancer cell lines. Methods: 5-FU derivatives were synthesized and characterized using Nuclear magnetic resonance (NMR), high-performance liquid chromatography (HPLC), and elemental analysis to determine the presence of new bond formation and the purity of the derivatives or analogs. 2D cell viability, 3D models, cell migration, and cell cycle studies were performed on MiaPaCa-2 (MP-2) cells and PANC-1 cells. Apoptosis was performed on PANC-1 cells by staining with Acridine Orange and Ethidium Bromide. The metabolic stability of the analogs was determined by using human liver microsomes. PARP, p53, and BAX expressions in PANC-1 were determined by Western Blot. Results: NMR data showed the conjugation of alkyl and tetrahydrofuran groups to the 5-FU ring confirming the formation of AGY1 and AGY2. The elemental analysis of AGY1 and AGY2 showed 99.6% and 99.5 % purity respectively. The IC50 value of the AGY1 against 2D MP-2 was found to be 12.2 ± 1 µM, while the IC50 value of AGY2 was determined to be 2.9 ± 0.2 µM. Both AGY1 and AGY2 anticancer activities were significantly higher than 5-FU (18.5 ± 1.1 µM). However, AGY2 was more effective compared to AGY1. A similar trend was observed for AGY2 analog where the IC50 value for treated 2D PANC-1 was 3.1 ± 0.29 µM and demonstrated a higher anticancer activity compared to AGY1 (6.4 ± 0.38 µM) or 5-FU (7.5 ± 1.1 µM). For cell migration studies, AGY2 (43 ± 5.5) exposed MP-2 cells showed significant arrest towards the wound gap compared to AGY1 (74 ± 18.5) and 5-FU (86 ± 15.6). A similar result of cell migration arrests AGY2 was observed in PANC-1 treated-AGY1, AGY2, and 5-FU. Organoid and spheroid models demonstrated high and comparable anticancer activity of AGY1 and AGY2 compared to 5-FU treated organoid and spheroid models. The cell-cycle analysis showed that AGY2-treated MP-2 cells at 1μΜ concentration showed a higher G1 population arrest (64.25%) compared to AGY1 (54.2%) and 5-FU (55.75%). Similarly, in PANC-1 cells, AGY2 at 1μΜ concentration showed a higher S-phase arrest (53.86%) compared to AGY1 (50.76%) and 5-FU (36.14%). Metabolic stability data showed higher stability of AGY2 (95% intact) compared to AGY1 (85% intact) compared to 5-FU (55% intact) after 2 hour of human liver microsomal incubation. The expressions of p53 and BAX proteins in PANC-1 cells were significantly upregulated and PARP was downregulated at higher concentration when treated with AGY1 and AGY2 compared to 5-FU. Conclusion: We demonstrate that AGY2 showed more promising anticancer activity compared to AGY1 and 5-FU against pancreatic cancer cells. AGY2 has the potential to improve 5-FU’s metabolic stability and enhance its therapeutic efficacy. Citation Format: Raviteja Bulusu, Joy Okoro, Esther Frimpong, Bo Han, Sherise Rogers, Xue Zhu, Edward Agyare. In vitro evaluation of novel pyrimidine nucleosides for the treatment of pancreatic cancer [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 1823.

Abstract 3512: Joint probabilistic modeling of multimodal metabolic data with UnitedMet

Abstract Metabolites play a crucial role in the functioning of biological systems. Measurements of metabolite abundances by metabolomics and metabolic pathway activity by isotope-tracing are powerful approaches to study metabolic phenotypes. However, large-scale measurements of metabolite levels and pathway activity remain extremely scarce due to the technical challenges. Motivated by the strength of RNA-metabolite covariation, we present UnitedMet, a Bayesian probabilistic method for end-to-end joint modeling of RNA-sequencing and metabolic datasets that is capable of dimensionality reduction, data integration and metabolic modality prediction. UnitedMet demonstrates high accuracy predicting metabolite levels and isotopologue distributions from gene expression data in human tumor samples. To evaluate UnitedMet’s efficacy in clinical applications, we inferred metabolomic profiles and isotope labeling patterns from the RNA sequencing data of clear cell renal cell carcinoma (ccRCC) patients enrolled in TCGA and clinical trials. Correlation analysis between genetic mutations and predicted metabolic phenotypes revealed that BAP1 mutations were associated with increased contribution of glucose to TCA cycle in ccRCC. Increased TCA cycle activity was found to be associated with disease progression and poorer clinical outcome. UnitedMet, therefore, provides a solution that could be broadly applicable to enhance multimodal metabolomic datasets and facilitates comprehensive study of interplays between metabolic profiles, molecular alterations and human phenotypes. Citation Format: Amy X. Xie, Casey Bradshaw, Christopher Tosh, Wesley Tansey, Ed Reznik. Joint probabilistic modeling of multimodal metabolic data with UnitedMet [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 3512.

Effective development of intensified perfusion culture based on shake tube semi-continuous bioreactors and optimal osmolarity constraint

The development of perfusion strategies supporting ultra-high cell growth and productivity for the intensified perfusion cultures (IPC) is very challenging. In this study, a novel approach was developed to address this issue by utilizing the shake tube semi-continuous bioreactors (ST) combining with a concentrated feed from the traditional fed-batch (TFB) platform. First, the impact of operating conditions in the ST were evaluated to identify those that best reflected the cell performance for the benchtop perfusion bioreactors (STR). Subsequently, the optimization of several key process parameters was performed in the ST along with the metabolic data obtained for the design of the IPC. Further, an empirical model to predict culture osmolarity was developed, verified, and subsequently utilized to determine the best combinations of the basal and feed perfusion rates ensuring the optimal osmolarity at the targeted viable cell density for the IPC process. Finally, the designed perfusion strategies were successfully validated in two IPC processes operated at 80 and 100×106 cells/m in the 3-L STR with an approximate 9-fold increase in cumulative titers compared to the TFB process. Overall, this study offers a novel and prospective method for the effective development of the IPC process to achieve significant process intensification.

O167: 3D-TIPS Printed Nanohybrid Composite Vascular Grafts for Evaluation of Human Umbilical Vein Endothelial Cell Response

Abstract Introduction Cardiovascular disease is a global killer. For progressive disease coronary artery bypass grafting is used, which is limited by suboptimal vasculature and patency. Alternatively, tissue engineered scaffolds are cell seeded structures supporting neovascularization. Poly(ethylene-diethylene glycol succinate)urea-urethane-polyhedral oligomeric silsesquioxane [P(EDS)UU-POSS] shows potential for its elastic and degradation properties. In this study, we manufacture a P(EDS)UU-POSS scaffold and assess human umbilical vein endothelial cell (HUVEC) attachment. Method 3D printed thermally induced phase separated (3DTIPS) scaffolds were manufactured. Mechanical testing was employed to compare scaffold stress-strain curves with those of currently utilised vessels and synthetic grafts. HUVEC were seeded onto scaffolds coated with type I collagen and peptide hydrogel respectively. Surface functionalization in this way, is hypothesised to optimise cell attachment. Assessment of cell viability was with PrestoBlueTM and total DNA assay. Confocal imaging was also performed. Results Tensile properties of 3D-TIPS scaffolds match the greater saphenous vein. 3D-TIPS scaffolds exhibit higher elasticity than synthetic PCL grafts, meeting stretch and strength requirements of vessels. Metabolism data shows an initial increase followed by a decrease for both types of coating. This indicates a decrease in overall metabolism but not cell death, attributed to cellular adaptation to the scaffold microenvironment, with collagen coating performing better. Visualisation of type I collagen samples demonstrates increases in cellular attachment. Conclusions There is potential for optimised 3DTIPS P(EDS)UU-POSS scaffolds with suitable mechanical properties. Further study of HUVEC and smooth muscle cell co-culture will better mimic in vivo environments, assessing sustained potential as vascular grafts.