Highlights
- •>2,500 single-cell signaling responses from drug-treated PDOs and CAFs
- •Trellis: tree-based treatment effect method for single-cell screening analysis
- •PDOs have patient- and CAF-specific drug responses that align with PTM signaling
- •CAFs can polarize PDOs to a chemorefractory revival stem cell (revCSC) fate
Summary
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.
Introduction
Tumors are heterogeneous cellular systems comprising cancer cells, stromal fibroblasts, and various immune cells. Tumor phenotypes are regulated by cell-intrinsic mutations within cancer cells and cell-extrinsic cues from the tumor microenvironment (TME). Colorectal cancer (CRC) kills >0.9 million people per year worldwide and is characterized by high inter-patient genetic heterogeneity and patient-specific responses to therapy. Cancer-associated fibroblasts (CAFs) are one of the most profuse cell types in the CRC TME, and high CAF abundance correlates with poor overall survival and influences response to both targeted therapies and radiotherapy. Unfortunately, how CAFs regulate cancer cell therapy responses in a patient-specific manner is poorly understood.
Patient-derived organoids (PDOs) are personalized cancer models that can mimic their parent tumors’ response to chemotherapies, with several studies proposing PDOs as personalized avatars of drug response. However, epithelial PDO monocultures cannot model the influence of stromal cells on therapy response. PDOs can be co-cultured with stromal and immune cells to recapitulate elements of the TME, but how this alters PDO phenotypes and drug response mechanisms is unknown. Moreover, PDO drug sensitivity is typically measured using bulk live/dead viability assays that cannot resolve cell-type-specific data from co-cultures and provide no mechanistic insight into drug responses.
To overcome these limitations, here we developed a highly multiplexed Thiol-reactive Organoid Barcoding in situ (TOBis) mass cytometry platform to study how anti-cancer therapies regulate the post-translational modification (PTM) signaling, DNA-damage, cell-cycle, and apoptosis response of CRC PDOs in the presence or absence of CAFs at single-cell resolution across >2,500 PDO-CAF cultures. To compare single-cell drug responses from thousands of cell-type-specific datasets, we developed “Trellis,” a tree-based treatment effect analysis method that derives generalized optimal transport distances between samples after normalizing by their own controls. TOBis mass cytometry and Trellis single-cell screening revealed that drug-induced PTM signaling responses are PDO-specific and demonstrated that CAFs shift epithelial cells toward a slow-cycling revival stem cell fate to protect CRC cells from chemotherapy. CAF chemoprotection could be rationally reversed using insights from single-cell PTM data, demonstrating the utility of mechanism-focused drug screening for overcoming therapy resistance. These results illustrate the functional intertumoral heterogeneity of patient-specific drug response mechanisms and highlight the role of TME cells in regulating drug resistance plasticity in cancer.
Results
Patient- and microenvironment-specific single-cell PTM PDO-CAF drug screening
To study how CAFs regulate patient-specific drug response signaling, we established a high-throughput 3D organoid co-culture system comprising 10 CRC PDOs (Table S1) cultured either alone or with CRC CAFs.
Organoid cultures were treated in triplicate with either vehicle control or titrated combinations of clinical therapies 5-fluorouracil (5-FU), SN-38 (active metabolite of Irinotecan), Oxaliplatin, and Cetuximab (EGFR inhibitor). LGK974 (PORCN inhibitor) was also studied to investigate PDO-CAF WNT signaling and Berzosertib (VX-970), as ATR inhibition has been hypothesized to synergize with DNA-damaging agents in CRC (Figure 1A; Table S2). Following treatment, each culture was fixed in situ, stained with thiol-reactive monoisotopic TOBis barcodes, pooled, dissociated into single cells, stained with a panel of 44 rare-earth metal antibodies (identifying cell type, cell state, DNA damage, and PTM signaling; Table S3), and analyzed by mass cytometry (Figure 1B). Following multiplexed debarcoding and cell-type-specific gating, we obtained >10 million PDO cells and >15 million CAFs from 2,520 3D cultures (3,360 cell-type-specific single-cell PTM signaling datasets).
