Dataset Information

Automated image-based profiling of complex drug induced phenotypes in patient-derived organoids

ABSTRACT: We established a platform for high-throughput confocal fluorescence microscopy and automated image analysis to perform large-scale drug response profiling with patient derived organoids (PDOs). Using PDO lines newly established from colorectal cancer patients with a standardized workflow, we performed chemical perturbations with more than 500 experimental and clinically used compounds. Subsequently, we captured approx. six million images by confocal microscopy. To analyze drug-induced PDO re-organization on a single organoid level we developed SCOPE, a software framework for automated PDO high-throughput image analysis. We observed a rich variety of divergent and reoccurring compound induced phenotypes in organoids. Perturbation of cellular processes, including signaling by MEK, GSK3β, cyclin dependent kinases (CDKs) and others, led to recurring and distinct architectural changes. Highlighting the clinical relevance, PDO viability was heterogeneous after perturbation with anticancer drugs and matched clinical response of corresponding colorectal cancer patients. Our work opens new avenues for image-based profiling for accelerated drug discovery and clinical decision-making.

PROVIDER: EGAS00001003140 | EGA |

REPOSITORIES: EGA

ACCESS DATA

Similar Datasets

Project description:Precision oncology aims to improve clinical outcome of patients by offering personalized treatment through identifying druggable genomic aberrations within their tumors. This is particularly valid when it comes to offering alternative treatment options for patients with advanced tumors that are chemo-refractory. Patient-derived organoids (PDOs) are 3 dimensional tumoroids that can be expanded ex vivo and are both pheno- and genotypically identical to patients’ tumors. Observational studies have shown that PDO-based drug screens can predict treatment response with high sensitivity and specificity. Vlachogiannis G. reported a living biobank of patient-derived organoids (PDOs) from patients with advanced GI cancers enrolled in clinical trials. PDOs can recapitulate patients’ clinical response to chemotherapeutic agents. In 19 tumor organoids, the group performed molecular profiling and drug screens and then compared ex vivo organoid responses to anticancer drugs. Drug response to PDO based orthotopic mouse tumor xenografts correlated to the drug response of the patient in clinical trials. Further to the study, there were other retrospective validation studies utilizing PDOs from patients enrolled in clinical trials such as the TUMOROID, CinClare to predict clinical response. Ooft studied PDOs from patients with metastatic colorectal cancers enrolled in the TUMOROID study to predict response to irinotecan-based therapies. Yao generated a organoid biobank of 80 locally advanced rectal cancers. These patients were derived from a phase III study (CinClare) that compared neoadjuvant chemo-radiation using either capecitabine or CAPIRI. Response to chemoradiation in patients matched to that of rectal cancer organoids (sensitivity 78% and specificity 91.9%). In a systematic analysis of 17 studies (9 on advanced GI and pancreatic cancers, one on renal cell cancer and others on miscellaneous cancers), the pooled sensitivity and specificity for discriminating patients with a clinical response through PDO-based drug screen was 0.81 (95%CI 0.69-0.89) and 0.74 (95%CI 0.64-0.82) respectively. Within 4-6 weeks, PDO-based drug screen creates a true personalised platform by predicting patient-specific drug response with high accuracy. Recent technical advancements in growing these PDO ‘avatars’ from biopsies have made it possible to test suitable anticancer drugs in patients with advanced inoperable tumors, and explore the new possibilities for treatment options that otherwise would be missed by standard conventional therapies. In 2019, our group embarked on PDO research; investigators obtained tissues from patients with advanced/ inoperable solid tumors, and performing drug screens on these PDOs ex vivo. In several patients, investigators were able to identified drugs not otherwise used through sequencing data, and observed remarkable clinical response in patients with PDO responsive tumors. Investigators illustrate with cases that underwent PDO culture and drug screens. [ See appendix ] In the literature, the clinical utility of treatment based on PDO informed drug options has however not been fully established. Investigators therefore propose a phase 2 proof-of-concept clinical trial to evaluate efficacy of NGS/ PDO guided treatment in patients with inoperable or metastatic solid tumors..

Dataset Information

Automated image-based profiling of complex drug induced phenotypes in patient-derived organoids

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets