Center For Biomarker Research In Medicine

In non-small cell lung cancer (NSCLC), non-invasive alternatives to biopsy-dependent driver mutation analysis are needed. We reviewed the effectiveness of radiomics alone or with clinical data and assessed the performance of artificial intelligence (AI) models in predicting oncogene mutation status. A PRISMA-compliant literature review for studies predicting oncogene mutation status in NSCLC patients using radiomics was conducted by a multidisciplinary team. Meta-analyses evaluating the performance of AI-based models developed with CT-derived radiomics features alone or combined with clinical data were performed. A meta-regression to analyze the influence of different predictors was also conducted. Of 890 studies identified, 124 evaluating models for the prediction of epidermal growth factor-1 (EGFR), anaplastic lymphoma kinase (ALK), and Kirsten rat sarcoma virus (KRAS) mutations were included in the systematic review, of which 51 were meta-analyzed. The AI algorithms' sensitivity/false positive rate (FPR) in predicting mutation status using radiomics-based models was 0.754 (95% CI 0.727-0.780)/0.344 (95% CI 0.308-0.381) for EGFR, 0.754 (95% CI 0.638-0.841)/0.225 (95% CI 0.163-0.302) for ALK and 0.475 (95% CI 0.153-0.820)/0.181 (95% CI 0.054-0.461) for KRAS. A meta-analysis of combined models was possible for EGFR mutation, revealing a sensitivity of 0.806 (95% CI 0.777-0.833) and a FPR of 0.315 (95% CI 0.270-0.364). No statistically significant results were obtained in the meta-regression. Radiomics-based models may offer a non-invasive alternative for determining oncogene mutation status in NSCLC. Further research is required to analyze whether clinical data might boost their performance. Question Can imaging-based radiomics and artificial intelligence non-invasively predict oncogene mutation status to improve diagnosis in non-small cell lung cancer (NSCLC)? Findings Radiomics-based models achieved high performance in predicting mutation status in NSCLC; adding clinical data showed limited improvement in predictive performance. Clinical relevance Radiomics and AI tools offer a non-invasive strategy to support molecular profiling in NSCLC. Validation studies addressing clinical and methodological aspects are essential to ensure their reliability and integration into routine clinical practice.

BackgroundP‐tau217 is emerging as the most accurate blood‐based biomarker (BBBM) for the diagnosis of Alzheimer's disease (AD) pathology, with elevated levels correlating with β‐amyloid and tau pathology, as confirmed by PET scans and cerebrospinal fluid analysis. Longitudinal increases in p‐tau217 correlate with cognitive decline, making it a promising tool for early detection. Validating remote sampling methods is essential to increase and simplify access to BBBMs for individuals with cognitive impairment.Neurogen Biomarking provides an end‐to‐end ecosystem for early AD detection, including at‐home sampling, remote cognitive assessments, and access to neurologists. A proof‐of‐concept (POC) study (N = 17) showed that p‐tau217 concentrations from samples collected with the Tasso+™ device correlated with venipuncture samples and distinguished AD patients from controls. This study aims to validate these findings in a larger cohort.MethodThis validation study includes 70 healthy volunteers and 60 amyloid‐positive AD patients recruited from multiple hospitals, including Insight Hospital, Memorial Health, Pittsburgh University, Banner Health, and Wake Forest University. Blood samples will be collected using the Tasso+™ device and standard venipuncture.Sample stability will be tested at room temperature (RT) for 24, 36, and 48 hours and samples will be shipped overnight, mimicking real‐world conditions. Samples will be processed using a proprietary method. p‐tau217 concentrations will be measured using the ALZpath assay on SimoaⓇ technology (Quanterix) and compared between sample types.ResultRecruitment is ongoing. Based on POC findings, a positive correlation between p‐tau217 concentrations from Tasso+™ and venipuncture is expected, analyzed via Spearman's rank correlation. Receiver Operating Characteristic analysis will assess differentiation accuracy between control and amyloid‐positive participants. The AUC values must meet the threshold set by CEOi guidelines for successful validation. The Youden index will determine the optimal p‐tau217 cut‐off for amyloid‐PET positivity. The validation results from this study will be first presented at AAIC2025, offering insights into Tasso+™ as a scalable, non‐invasive method for at‐home blood sampling in Alzheimer's biomarker research.ConclusionThis study aims to validate a novel, non‐invasive at‐home sampling method for p‐tau217, facilitating early AD detection and more frequent testing. By improving accessibility, this approach could enhance clinical practice and research, making early diagnosis feasible for a broader population.

Microplastic (MP) pollution is increasingly acknowledged as a critical environmental and public health issue. This study sought to establish a robust, clinically compatible method for detecting MP particles in deparaffinized formalin-fixed paraffin-embedded (FFPE) human colon tissue sections, using protocols compatible with routine clinical pathology. We employed mid-infrared photothermal (MIP) microscopy—also referred to as optical photothermal infrared (OPTIR) spectroscopy—as a non-destructive, high-resolution technique for chemical characterization and spatial mapping of polymer particles in intact FFPE samples. Following OPTIR analysis, identical sections underwent hematoxylin and eosin (H&E) staining to facilitate precise histopathological evaluation in defined regions of interest. Using this integrated workflow, we detected and localized polyethylene (PE), polystyrene (PS), and polyethylene terephthalate (PET) particles (21 PE particles, 1 PS particle, and 1 PET fiber) within distinct tissue areas. Subsequent histological assessment revealed characteristic inflammatory features near to these identified MP particles. To our knowledge, this represents the first demonstration of a diagnostic workflow that enables combined infrared spectroscopic and histopathological analysis of MPs in routinely processed human FFPE tissue. This approach offers a promising avenue to elucidate the role of microplastic accumulation in human disease and supports further investigation into potential mechanistic links between MP exposure and inflammatory processes in the colon.

Abstract BACKGROUND Use of patient derived cells (PDCs) is a new and approaching method of functional personalized medicine, exploring treatment response prediction on a patient specific level. In this regard, the multi-center, randomized, phase 2 ATTRACT trial in Austria aims to investigate the clinical potential of PDC-based drug screening in patients with newly-diagnosed, MGMT promoter unmethylated glioblastoma. While the quality of PDCs is a fundamental prerequisite for reliable and scalable drug testing, the factors impacting PDC viability as well as the comprehensive analysis of the biological resemblance to the original tumor remain poorly understood. MATERIAL AND METHODS We sought to identify factors influencing PDC quality for drug screening approaches. Patients with suspicion of glioblastoma are included in the Austrian Glioblastoma Biobank (GlioBank). Material was sampled during neurosurgical resection and transported to the CBmed laboratory overnight. Only samples with a minimum weight of 1g were processed further. The fresh tissue was mechanically and enzymatically dissociated and transferred to culture plates and cultivated in an optimized medium for 7-14 days (primary culture) and frozen for further analysis. The minimum amount of PDCs required for a drug screen with 28 drugs was defined as 4.5 x 10^5 cells. RESULTS 99 samples were obtained between May 2024 and March 2025. Of these, 81 samples were diagnosed as glioblastoma and thus were processed further. The median tumor weight used for PDC processing was 2.14 g (range 1-22g). 60/81 (74%) samples fulfilled the criteria of a full drug screen PDC culture, while in 16/81 (20%) only a compromised panel was possible. In 5/81 (6%) samples no successful PDC culture could be established. After approximately 14 days of culture in tumor cell enriching conditions, a median of 65.64 % of the cells were lost compared with the initiation of the primary culture. The weight of the initial tumor used for the generation of the PDCs significantly correlated with the cell number of the PDC culture (p<0.05), but not with the percentage of lost cells (p>0.05). The clinical factors age and gender of the patient, time between resection and placement in medium, transport time from surgery to the laboratory, histology or MGMT status did not impact the likelihood of obtaining enough cells for a full drug panel and also did not correlate with the percentage of lost cells from PDC start to end (p>0.05). Further analysis on methylation and mutational signatures and known genetic biomarkers (eg. TERT promoter alterations) are ongoing and will be reported. CONCLUSION We could establish a nation-wide program for PDC sampling enabling PDC-based drug screening within a prospective, randomized clinical trial. Clinical factors do not impact the PDC quality, while tissue weight is significantly associated with the cell number of the PDC culture.

Anorexia nervosa (AN) is characterized by high mortality, limited treatment options, altered gut microbiome composition, and impaired feedback (FB) learning. This pilot study explores the relationship between gut microbiome composition and reduced implicit FB learning in AN. Sixteen female patients with AN and fourteen healthy controls (HC) completed the Weather Prediction Task to assess implicit learning (feedback and paired-associate formats). Gut microbiome composition was analyzed via Illumina sequencing, with species differences identified using linear discriminant effect size (LEfSe) analysis. Patients with AN had lower FB learning scores compared to HC (p=.012), while no significant difference in the paired-associate task was found (p=.905). Better feedback learning was associated with greater gut microbiome diversity, indicated by a larger number of observed species (r=0.45, p=.013), and a higher Shannon index (r=0.39, p=.035). LEfSe analysis revealed an increased abundance of Akkermansia muciniphila in AN, while short-chain fatty acid-producing species (such as phylum Firmicutes, including Roseburia, Agathobacter and Faecalibacterium) were more prevalent in HC. Participants with higher FB learning exhibited an increase in Actinobacteria, particularly Bifidobacteria, while those with low FB learning had an increase in the Lachnospirales of the phylum Firmicutes. These findings highlight the gut microbiome's potential role in implicit learning in AN, which could play an important role in the success of treatments. Future research should focus on microbiome-targeted interventions, such as probiotics, particularly Bifidobacteria, to improve gut health and implicit learning in patients with AN.

Background: The novel coronavirus (SARS-CoV-2) led to gastrointestinal manifestations in up to 50% of cases, with diarrhea being common, and probiotics have been suggested as a potential treatment. Aim: This study aimed to assess changes in the microbiome and the effects of a multispecies probiotic in patients with COVID-19 in home quarantine through a fully remote telemedical approach. Methods: Thirty patients were randomized to receive either the Ecologic AAD probiotic (Winclove Probiotics, Amsterdam, The Netherlands), on the market as OMNi-BiOTiC 10 (Allergosan, Austria), or a placebo for 30 days in a 2:1 ratio. Respiratory and gastrointestinal symptoms were monitored in 2–10-day intervals via online surveys, and five stool samples were collected during the 30-day study period for microbiome and metabolomics analyses. Twenty-four healthy volunteers served as controls. Results: Of the 30 patients, 26 completed this study (10 placebo, 16 probiotic). Patients reported respiratory symptoms and a diminished gastrointestinal quality of life, both of which improved significantly during the study period, irrespective of the intervention. Compared to controls, infected patients showed significant alterations in the fecal microbiome (p = 0.002), including an increase in Bacteroidetes and decreases in Christensenellaceae, Ruminococcaceae, and Gammaproteobacteria, along with metabolomic changes. Probiotic treatment significantly modulated the patients’ microbiome beta diversity (p = 0.001) and introduced the Enterococcus faecium W54 strain. Symptoms, COVID-19-related taxa, and the fecal metabolome were not affected by the intervention. Conclusions: Patients with mild COVID-19 disease in home quarantine exhibited respiratory symptoms, a reduced gastrointestinal quality of life, and changes in the fecal microbiome and metabolome.

In the field of precision medicine, therapy is optimized individually for each patient, enhancing efficacy while reducing side effects. This involves the identification of promising drug candidates through high-throughput screening on human derived cells in culture. However, screening of drugs which have poor solubility or permeability remains challenging, especially when targeting intracellular components. Therefore, encapsulation of drugs into advanced delivery systems such as nanostructured lipid carries (NLC) becomes necessary. Here we show that the cellular uptake of NLC with different matrix compositions can be assessed in a high-throughput screening system based on acoustic droplet ejection (ADE) technology (Echo liquid handler). Our findings indicate that surface tension and viscosity of the NLC dispersions need to be tailored to enable a reliable ADE transfer. The automated NLC uptake studies indicated that the composition of the matrix, more specifically the amount of oleic acid, significantly influenced cellular uptake. The data obtained were corroborated by imaging based and spectral flow cytometry cellular uptake studies. These findings thus not only provide the basis for a screening tool to rapidly identify the efficacy of NLC uptake but also enable a next step toward precision high-throughput drug screening under consideration of an optimized drug delivery system.

Simple SummaryThis review aims to improve our understanding of gastrointestinal tract cancer and the side effects of cancer treatment by using advanced in vitro systems. Traditional models like cell cultures and animal studies provide valuable insights but have limitations in replicating the complexity of human disease and raise ethical concerns. By focusing on bioreactor-based in vitro systems, which can mimic the physical and chemical environment of the gastrointestinal tract, this study aims to provide more accurate models for studying cancer and its treatment side effects. These advancements could lead to better insights into disease mechanisms, potentially improving treatment strategies and benefiting the broader research community.Human diseases are multifaceted, starting with alterations at the cellular level, damaging organs and their functions, and disturbing interactions and immune responses. In vitro systems offer clarity and standardisation, which are crucial for effectively modelling disease. These models aim not to replicate every disease aspect but to dissect specific ones with precision. Controlled environments allow researchers to isolate key variables, eliminate confounding factors and elucidate disease mechanisms more clearly. Technological progress has rapidly advanced model systems. Initially, 2D cell culture models explored fundamental cell interactions. The transition to 3D cell cultures and organoids enabled more life-like tissue architecture and enhanced intercellular interactions. Advanced bioreactor-based devices now recreate the physicochemical environments of specific organs, simulating features like perfusion and the gastrointestinal tract’s mucus layer, enhancing physiological relevance. These systems have been simplified and adapted for high-throughput research, marking significant progress. This review focuses on in vitro systems for modelling gastrointestinal tract cancer and the side effects of cancer treatment. While cell cultures and in vivo models are invaluable, our main emphasis is on bioreactor-based in vitro modelling systems that include the gut microbiome.

We present a data visualization tool for tumor boards, merging clinical with molecular and multi-omics data to refine precision oncology decisions. The tool offers a holistic patient perspective, facilitating personalized treatment strategies. By integrating clinical and laboratory datasets, it enables intuitive navigation through complex information. Clinicians are supported in their decision-making by user-friendly visualizations. Future studies are needed to evaluate its real-world impact and usability in precision oncology settings.