Berend Snijder

Genomic stratification leaves most patients with colorectal cancer (CRC) without opportunities for targeted therapy. Seeking tumour state-linked drug sensitivities, we combined image-based ex vivo drug profiling with matched transcriptomics in 127 prospectively collected patient biopsies across CRC stages and molecular subtypes. Network-level integration identified a ciliogenic cell state defined by upregulation of cilia related genes with selective sensitivity to the FGFR inhibitor erdafitinib. This ciliogenic signature was associated with poor progression-free survival (hazard ratio 2.28; adjusted p=0.0098), even in early-stage colorectal cancer. Furthermore, knockdown of the transcriptional regulators RFX2 and RFX3, upstream of the ciliogenic cell state, attenuated erdafitinib sensitivity. Orthogonal bulk and single-cell analyses across normal, primary and metastatic tissues contextualised the ciliogenic programme along a transition from canonical epithelial states towards fetal and non-canonical tumour states, concomitant with Wntrelated remodelling, FGFR1/FGFR2 upregulation, and loss of intestinal epithelial identity. Together, these findings define a tumour-intrinsic ciliogenic cell state as a feature of CRC progression, revealing a patient-stratified repurposing opportunity for targeted therapy beyond the current CRC treatment paradigms.

Tumour heterogeneity presents a major challenge for precision oncology, as genetically and phenotypically distinct tumour clones may respond differently to therapy. To address this, we introduce scClone2DR, a probabilistic multi-modal framework that predicts drug responses at the level of individual tumour clones by integrating single-cell DNA and RNA sequencing with ex-vivo drug-screening data. In simulations, scClone2DR substantially outperforms alternatives in recovering true drug-effects and clonal sensitivities. Applied to 60 melanoma and 21 acute myeloid leukaemia patient samples, the method identifies heterogeneous clonal responses, yields biologically meaningful feature rankings, highlights clones that may be resistant to treatment, and improves the prediction of clinical outcomes compared to models ignoring clonal structure. These results demonstrate that modelling tumour evolution and clonal diversity is crucial for accurate drug-response prediction and provides a foundation for more effective, clone-aware precision oncology.

Supplementary Data containing the statistical analysis plan, Supplemental Figures 1-6 and Supplemental Tables 1 and 2

Table ST2: Summary of the clinical data corresponding to the primary cell cultures used in this publication

Tumor Profiler Consortium author list

Table ST6: Differential expressed lipids and their FDRs in sensitive vs resistant cells (negative values are upregulated in sensitive cells).

S4: Neocuproine is a copper-dependent ROS inducer

S1: Evaluation of HST screening.

S15: Tumor response curves in mice for ionophoric copper-chelators in combination with binimetinib.