New Bayesian Method Detects Drug-Resistant Cancer Earlier by Analyzing Faint Circulating Tumor DNA Signals
Researchers developed a statistical method called Span that detects emerging drug resistance in metastatic breast cancer by analyzing patterns of faint and undetectable circulating tumor DNA (ctDNA) signals over time, rather than treating each test independently. The method uses censored-Poisson Bayesian modeling to identify growth trends below the assay's detection limit, potentially catching resistance months before imaging shows it. This approach could enable earlier treatment adjustments and improve outcomes for patients with hormone receptor-positive, HER2-negative metastatic breast cancer.
Researchers introduced Span, a Bayesian change-point detection algorithm designed to identify emerging drug resistance in metastatic breast cancer by analyzing longitudinal patterns of circulating tumor DNA (ctDNA). The key innovation is treating undetectable results as left-censored observations rather than negative signals, allowing the method to extract actionable information from sequences of faint detections and non-detections over time. In synthetic simulations of HR+/HER2- breast cancer patients on CDK4/6-inhibitor therapy, Span roughly doubled the detection rate of impending progressions three months ahead compared to snapshot-based approaches (25% vs 11% at matched 10% false-alarm rates). The method employs a sequential generalized-likelihood-ratio statistic with calibrated false-alarm control and contains no learned weights, eliminating overfitting risk. Validation on real breast cancer datasets (GBSG-2 and PBC2) confirmed the survival predictions matched established baselines and appropriately declined to outperform on datasets where the mechanism was not applicable, supporting the method's specificity to the intended clinical regime.
What's missing
The study uses entirely synthetic ctDNA trajectories for the main results, which limits direct clinical validation. The authors acknowledge this limitation but do not provide a timeline for prospective clinical validation or discuss regulatory pathways for clinical implementation. Additionally, the method's performance on other cancer types or treatment regimens remains unexplored.
What different sources said
- arXiv cs.LGCenter
Seeing Below the Limit of Detection: A Censored-Poisson Bayesian Latent-Growth Change-Point Detector (the Span Detector) for Serial ctDNA in HR+/HER2- Metastatic Breast Cancer
Related
Genetic Drift, Not Selection, Drives Rapid Feather Color Evolution in Island Bird Radiation
A new study of an island bird radiation found that rapid evolution of feather coloration is driven primarily by genetic drift in small populations rather than sexual or ecological selection. The research integrated whole-genome data with detailed plumage measurements across complete species sampling to test whether signaling trait evolution correlates with speciation rates. The findings suggest that neutral demographic processes play a central role in generating phenotypic diversity during island radiations, challenging assumptions about the mechanisms driving rapid evolution.
New AI Model Improves Prediction of Therapeutic Peptide Function from Protein Sequences
Researchers developed a lightweight CNN classifier that predicts whether peptide sequences have therapeutic properties, trained on a database of 54,655 peptides across 48 functional categories. The model uses a novel negative sampling strategy to reduce false positive rates from over 60% in previous approaches to 2.1%. This advancement could accelerate drug discovery by enabling faster computational screening of peptide candidates before expensive experimental testing.
Study Shows Different Metabolic Stress Models Produce Distinct Effects on Human Neuronal Networks
Researchers tested three common in vitro metabolic stress models on human-derived neuronal networks and found each produced different patterns of neuronal activity and cell damage. The models tested were hypoxia alone, oxygen-glucose deprivation (OGD), and hypoxia combined with glutamate exposure. The findings suggest that choice of experimental model significantly affects results and that combining electrophysiological and structural analyses is important for accurately assessing metabolic stress in stroke research.