Luck, T. | bioRxiv Science

bioRxiv2026

Pathway-centric multi-omics and functional precision medicine reveal shared drug vulnerabilities in heterogeneous adult Wilms tumor

Wilms tumor, i.e., nephroblastoma, is rare in adults and lacks standardized treatment, complicating clinical decision-making. Within the functional precision medicine study (DEDUCER), we profiled two spatially distinct tumor regions (T1 and T2) of an adult Wilms tumor patient using integrated histopathology, whole-exome sequencing, FFPE transcriptomics, and ex vivo drug screening of short-term cultured patient-derived cancer cells (PDCs) with 528 compounds. Genomic profiling revealed a truncal ASXL1 frameshift and shared F7, UBA1, COL21A1, and ATM variants alongside region-specific alterations: a TP53 mutation and broad copy-number (CN) gains in T1, versus ARID1A and KMT2D stop-gains in copy-neutral T2. Transcriptomics of tumor areas identified convergent activation of the G2/M checkpoint, E2F targets, and mitotic spindle programs across regions, consistent with high proliferation and partially comparable biomarker signatures to those observed in an open-source pediatric Wilms tumor dataset (n = 130). Functional assays uncovered distinct and shared drug vulnerabilities: although ATM alterations were present in both tumors, T1 PDCs showed selective sensitivity to topoisomerase I and BCL-2 inhibition in the context of an additional T1-specific TP53 alteration, while broader single-agent sensitivity and stronger drug synergies were observed in T2. Pathway-centric data integration indicated that differential gene expression and copy-number gains, rather than single mutations alone, better predicted ex vivo drug responses, revealing actionable shared dependencies despite pronounced spatial heterogeneity and establishing a translational framework for individualized management in this rare disease. HIGHLIGHTS- In the adult Wilms tumor, multi-region genomics revealed a truncal ASXL1 frameshift together with F7, UBA1, COL21A1 and ATM mutations across two tumor regions (T1 and T2), as well as region-specific alterations: TP53 mutation and widespread copy-number gains in T1, versus ARID1A and KMT2D stop-gains in copy-neutral T2, illustrating spatial heterogeneity. - Transcriptomics showed convergent activation of E2F targets, G2/M checkpoint, and mitotic spindle programs in both regions, consistent with high proliferation and aligning with Wilms tumor signatures (TARGET dataset); these pathways were associated with higher ex vivo drug sensitivity scores. - Functional drug sensitivity testing of patient -derived cancer cells ex vivo uncovered distinct and shared vulnerabilities: Although both tumors shared an ATM mutation, T1-specific TP53 alteration and death-pathway/stress-response alterations may underlie selective sensitivity to topoisomerase I inhibitors and BCL-2 inhibition. - Clinically relevant combinations, including vincristine plus dactinomycin and doxorubicin plus dactinomycin, showed ex vivo synergy. These findings are consistent with the patients more than five-year relapse-free outcome following vincristine, doxorubicin, and dactinomycin treatment combined with surgery, supporting the translational relevance of the ex vivo drug testing approach. - Pathway-centric integration (copy-number gains and differential expression) predicted drug response better than single-gene biomarkers. Overall, pathway-level dependencies provide robust, actionable targets despite genomic and phenotypic heterogeneity in adult Wilms tumor.

Cancer Biology↗

bioRxiv2018

Neuroanatomical correlates of food addiction and obesity in the general population

The food addiction model suggests neurobiological similarities between substance-related and addictive disorders and obesity. While structural brain differences have been consistently reported in these conditions, little is known about the neuroanatomical correlates of food addiction. We therefore assessed whether food addiction, assessed with the Yale Food Addiction Scale (YFAS), related to obesity, personality and brain structure in a large population-based sample (n=625; 20-59 years old, 45% women). A higher YFAS symptom score correlated with obesity and disinhibited eating. In a whole-brain analysis, YFAS symptom score was not associated with cortical thickness nor subcortical gray matter volumes. Higher body mass index (BMI) correlated with reduced thickness of (pre)frontal, temporal and occipital cortex. Bayes factor analysis suggested that BMI and - to a smaller extent - YFAS symptom score contributed independently to right lateral orbitofrontal cortex thickness. Our study shows that food addiction is not associated with neuroanatomical differences in a large population-based sample, and does not account for the major part of obesity-associated gray matter alterations. Yet, food addiction might explain additional variance in orbitofrontal cortex, a hub area of the reward network. Longitudinal studies implementing both anatomical and functional MRI could further disentangle the neural mechanisms of addictive eating behaviors.

neuroscience↗

bioRxiv2016

Predicting brain-age from multimodal imaging data captures cognitive impairment

The disparity between the chronological age of an individual and their brain-age measured based on biological information has the potential to offer clinically-relevant biomarkers of neurological syndromes that emerge late in the lifespan. While prior brain-age prediction studies have relied exclusively on either structural or functional brain data, here we investigate how multimodal brainimaging data improves age prediction. Using cortical anatomy and whole-brain functional connectivity on a large adult lifespan sample (N = 2354, age 19-82), we found that multimodal data improves brain-based age prediction, resulting in a mean absolute prediction error of 4.29 years. Furthermore, we found that the discrepancy between predicted age and chronological age captures cognitive impairment. Importantly, the brain-age measure was robust to confounding effects: head motion did not drive brain-based age prediction and our models generalized reasonably to an independent dataset acquired at a different site (N = 475). Generalization performance was increased by training models on a larger and more heterogeneous dataset. The robustness of multimodal brain-age prediction to confounds, generalizability across sites, and sensitivity to clinically-relevant impairments, suggests promising future application to the early prediction of neurocognitive disorders.\n\nHighlightsO_LIBrain-based age prediction is improved with multimodal neuroimaging data.\nC_LIO_LIParticipants with cognitive impairment show increased brain aging.\nC_LIO_LIAge prediction models are robust to motion and generalize to independent datasets from other sites.\nC_LI

neuroscience↗