Trellis tree-based analysis reveals stromal regulation of patient-derived organoid drug responses.

Patient-derived organoids (PDOs) can model personalized therapy responses; however, current screening technologies cannot reveal drug response mechanisms or how tumor microenvironment cells alter therapeutic performance. To address this, we developed a highly multiplexed mass cytometry platform to measure post-translational modification (PTM) signaling, DNA damage, cell-cycle activity, and apoptosis in >2,500 colorectal cancer (CRC) PDOs and cancer-associated fibroblasts (CAFs) in response to clinical therapies at single-cell resolution. To compare patient- and microenvironment-specific drug responses in thousands of single-cell datasets, we developed "Trellis"-a highly scalable, tree-based treatment effect analysis method. Trellis single-cell screening revealed that on-target cell-cycle blockage and DNA-damage drug effects are common, even in chemorefractory PDOs. However, drug-induced apoptosis is rarer, patient-specific, and aligns with cancer cell PTM signaling. We find that CAFs can regulate PDO plasticity-shifting proliferative colonic stem cells (proCSCs) to slow-cycling revival colonic stem cells (revCSCs) to protect cancer cells from chemotherapy.

Concentration-QT modelling in early clinical oncology settings: Simulation evaluation of performance.

AIMS: Concentration-QT modelling (C-QTc) of first-in-human data has been rapidly adopted as the primary evaluation of QTc interval prolongation risk. Here, we evaluate the performance of C-QTc in early oncology settings (i.e., patients, no placebo or supratherapeutic dose, 3 + 3 designs). METHODS: C-QTc performance was evaluated across three oncology scenarios using a simulation-estimation approach: (scen1) typical dose-escalation testing six dose levels (n = 21); (scen2) small dose-escalation testing two dose levels (n = 9); (scen3) expansion cohorts at one dose level (n = 6-140). True ΔΔQTc effects ranged from 3 ms ("no effect") to 20 ms ("large effect"). Performance was assessed based on the upper limit of the ΔQTc two-sided 90% CI against a threshold of 10 or 20 ms. RESULTS: The performance against the 10 ms threshold was limited based on C-QTc data from typical dose escalation (scen1) and acceptable performance was observed only for relatively large expansions (n ≥ 45; scen3). Performance against the 20 ms threshold was acceptable based on C-QTc data from a typical dose escalation (scen1) or dose expansion cohort n > 10 (scen3). In general, pooling C-QTc data from dose escalation and expansion cohorts substantially improved the performance and reduced the ΔQTc 90% CI width. CONCLUSION: C-QTc performance appeared limited using a 10 ms threshold, but acceptable against a 20 ms threshold. Selection of threshold may be informed by the benefit-risk balance in a specific disease area. Acceptable precision (i.e., confidence intervals) of the estimated ΔQTc, regardless of its magnitude, can be facilitated by pooling data from dose escalation and expansion cohorts.