Back

Combined radiomics and liquid biopsy reflect tumor biology towards multimodal non-invasive prostate cancer risk stratification

Riediger, A. L.; Meinzer, C.; Janscho, D.; Eickelschulte, S.; Janke, F.; Lazareva, O.; Huebschmann, D.; Stegle, O.; Froehling, S.; Bonekamp, D.; Schlemmer, H.-P.; Sueltmann, H.; Goertz, M.

2025-09-09 oncology
10.1101/2025.09.06.25335220 medRxiv
Show abstract

Accurate risk stratification is essential to prevent over- and undertreatment in newly diagnosed prostate cancer (PCa). In a pilot study with 49 men (36 PCa patients, 13 controls), we integrated non-invasive data modalities collected in a real-world clinical setting: radiomics from multiparametric magnetic resonance imaging (T2/ADC index-lesion features), routine and extended serological parameters, multimodal liquid biopsy features (copy number variations, chromosomal instability, methylation, fragmentation in plasma/urinary cfDNA), plus clinical and lifestyle factors. Pairwise Spearman analyses revealed significant inter-modality correlations. Our holistic PCa characterization showed that imaging phenotypes reflect tumor stage and molecular aggressiveness: lower ADC metrics correlated inversely with genomic cfDNA features, while ADC/T2 lesion volumes correlated positively with molecular signals. These results indicate that routine imaging and serological data may capture systemic tumor biology. Our proof-of-concept demonstrates feasibility of combining multiple data modalities in real-world clinical workflows and supports developing integrative models for non-invasive PCa risk stratification.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.