State-of-the art selection methods fail to identify weak but cumulative effects of features found in many high-dimensional omics datasets. Nevertheless, these features play an important role in certain diseases. We present Netboost, a three-step dimension reduction technique. First, a boosting-based filter is combined with the topological overlap measure to identify the essential edges of the network. Second, sparse hierarchical clustering is applied on the selected edges to identify modules and finally module information is aggregated by the first principal components. We demonstrate the application of the newly developed Netboost in combination with CoxBoost for survival prediction of DNA methylation and gene expression data from 180 acute myeloid leukemia (AML) patients and show, based on cross-validated prediction error curve estimates, its prediction superiority over variable selection on the full dataset as well as over an alternative clustering approach. The identified signature related to chromatin modifying enzymes was replicated in an independent dataset, the phase II AMLSG 12-09 study. In a second application we combine Netboost with Random Forest classification and improve the disease classification error in RNA-sequencing data of Huntington's disease mice. Netboost is a freely available Bioconductor R package for dimension reduction and hypothesis generation in high-dimensional omics applications.

Background: Hypoxia is an important risk factor and indicator for the declining health of inpatients. Predicting future hypoxic events using machine learning is a prospective area of study to facilitate time-critical interventions to counter patient health deterioration. Objective: This systematic review aims to summarize and compare previous efforts to predict hypoxic events in the hospital setting using machine learning with respect to their methodology, predictive performance, and assessed population. Methods: A systematic literature search was performed using Web of Science, Ovid with Embase and MEDLINE, and Google Scholar. Studies that investigated hypoxia or hypoxemia of hospitalized patients using machine learning models were considered. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. Results: After screening, a total of 12 papers were eligible for analysis, from which 32 models were extracted. The included studies showed a variety of population, methodology, and outcome definition. Comparability was further limited due to unclear or high risk of bias for most studies (10/12, 83%). The overall predictive performance ranged from moderate to high. Based on classification metrics, deep learning models performed similar to or outperformed conventional machine learning models within the same studies. Models using only prior peripheral oxygen saturation as a clinical variable showed better performance than models based on multiple variables, with most of these studies (2/3, 67%) using a long short-term memory algorithm. Conclusions: Machine learning models provide the potential to accurately predict the occurrence of hypoxic events based on retrospective data. The heterogeneity of the studies and limited generalizability of their results highlight the need for further validation studies to assess their predictive performance.

Background: Large language models (LLMs) like ChatGPT 4.0 hold promise for enhancing clinical decision-making in precision oncology, particularly within molecular tumor boards (MTBs). This study assesses ChatGPT 4.0’s performance in generating therapy recommendations for complex real-world cancer cases compared to expert human MTB (hMTB) teams. Methods: We retrospectively analyzed 20 anonymized MTB cases from the Comprehensive Cancer Center Augsburg (CCCA), covering breast cancer (n = 3), glioblastoma (n = 3), colorectal cancer (n = 2), and rare tumors. ChatGPT 4.0 recommendations were evaluated against hMTB outputs using metrics including recommendation type (therapeutic/diagnostic), information density (IDM), consistency, quality (level of evidence [LoE]), and efficiency. Each case was prompted thrice to evaluate variability (Fleiss’ Kappa). Results: ChatGPT 4.0 generated more therapeutic recommendations per case than hMTB (median 3 vs 1, P = .005), with comparable diagnostic suggestions (median 1 vs 2, P = .501). Therapeutic scope from ChatGPT 4.0 included off-label and clinical trial options. IDM scores indicated similar content depth between ChatGPT 4.0 (median 0.67) and hMTB (median 0.75; P = .084). Moderate consistency was observed across replicate runs (median Fleiss’ Kappa = 0.51). ChatGPT 4.0 occasionally utilized lower-level or preclinical evidence more frequently (P = .0019). Efficiency favored ChatGPT 4.0 significantly (median 15.2 vs 34.7 minutes; P < .001). Conclusion: Incorporating ChatGPT 4.0 into MTB workflows enhances efficiency and provides relevant recommendations, especially in guideline-supported cases. However, variability in evidence prioritization highlights the need for ongoing human oversight. A hybrid approach, integrating human expertise with LLM support, may optimize precision oncology decision-making.

Background/Objectives: Locally advanced rectal cancer is treated with neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision (TME). As this approach achieves complete pathologic remissions (pCR) in approximately 30% of patients, it raises the question of whether surgery is always necessary. Non-surgical strategies, such as “watch and wait” (W&W), have shown similarly promising outcomes. However, there is an unmet need for reliable biomarkers predicting pCR. Analysis of circulating tumor DNA (ctDNA) has shown potential for monitoring treatment response and detecting minimal residual disease. We hypothesized that monitoring ctDNA changes during nCRT might facilitate the identification of individuals who achieve pCR. Methods: In the prospective single-center NEORECT trial, the plasma of forty rectal cancer patients was collected before, during, and after nCRT and before TME. Informative somatic mutations were identified in tissue biopsies by NGS and subsequently used for ctDNA quantification by dPCR. Results: The results identified three distinct ctDNA patterns: increase, decrease, and absence. Remarkably, undetectable DNA was observed in good responders, while a tenfold ctDNA increase was associated with the emergence of new metastases. Despite these insights, ctDNA alone demonstrated low specificity, with no significant correlation to pCR or long-term prognosis. A multimodal approach incorporating routinely available clinical parameters remains inadequate for accurately predicting pCR prior to TME. Conclusions: In conclusion, the NEORECT trial establishes the feasibility of ctDNA-based personalized monitoring for rectal cancer patients undergoing nCRT. However, the utility of ctDNA in enhancing pCR prediction for a W&W strategy warrants further investigation. Larger studies integrating multi-gene analyses and expanded clinical datasets are essential in the future.

Molecular tumor boards (MTB) are interdisciplinary conferences involving various experts discussing patients with advanced tumors, to derive individualized treatment suggestions based on molecular variants. These discussions involve using heterogeneous internal data, such as patient clinical data, but also external resources such as knowledge databases for annotations and search for relevant clinical studies. This imposes a certain level of complexity that requires huge effort to homogenize the data and use it in a speedy manner to reach the needed treatment. For this purpose, most institutions involving an MTB are heading toward automation and digitalization of the process, hence reducing manual work requiring human intervention and subsequently time in deriving personalized treatment suggestions. The tools are also used to better visualize the patient’s data, which allows a refined overview for the board members. In this paper, we present the results of our thorough literature research about MTBs, their process, the most common knowledge bases, and tools used to support this decision-making process.

Tumor heterogeneity encompasses genetic, epigenetic, and phenotypic diversity, impacting treatment response and resistance. Spatial heterogeneity occurs both inter- and intra-lesionally, while temporal heterogeneity results from clonal evolution. High-throughput technologies like next-generation sequencing (NGS) enhance tumor characterization, but conventional biopsies still do not adequately capture genetic heterogeneity. Liquid biopsy (LBx), analyzing circulating tumor DNA (ctDNA), provides a minimally invasive alternative, offering real-time tumor evolution insights and identifying resistance mutations overlooked by tissue biopsies. This study evaluates the capability of LBx to capture tumor heterogeneity by comparing genetic profiles from multiple metastatic lesions and LBx samples. Eight patients from the Augsburger Longitudinal Plasma Study with various types of cancer provided 56 postmortem tissue samples, which were compared against pre-mortem LBx-derived circulating-free DNA sequenced by NGS. Tissue analyses revealed significant mutational diversity (4–12 mutations per patient, VAFs: 1.5–71.4%), with distinct intra- and inter-lesional heterogeneity. LBx identified 51 variants (4–17 per patient, VAFs: 0.2–31.1%), which overlapped with mutations from the tissue samples by 33–92%. Notably, 22 tissue variants were absent in LBx, whereas 18 LBx-exclusive variants were detected (VAFs: 0.2–2.8%). LBx effectively captures tumor heterogeneity, but should be used in conjunction with tissue biopsies for comprehensive genetic profiling.