Abstract C061: Therapeutic profiling, patient response prediction, and tumor evolution in pancreatic cancer organoids

Abstract

Abstract Pancreatic cancer (PC) is characterized by an aggressive biology and high tumor heterogeneity causing considerable variations in therapy response. Patient-derived organoids (PDOs) reflect parental tumor features and represent a powerful preclinical tool to predict drug response and harness personalized treatment. We have derived >100 PDO lines from treatment-naïve and pretreated PC patient primary tumor and metastases with a reliable efficacy and previously developed a pharmacotyping-guided prediction model to prognosticate patient therapy response (Beutel, 2021). Following up our initial feasibility trial, we now validated the model accuracy in real-life in a higher number of PC patients (43 cases). Additionally, a limited set of 7 PDO pairs were derived from the same patients at two distinct timepoints, followed by pharmacotyping and whole exome sequencing (WES). Our model allowed overall a successful drug-response prediction in naïve patients with an accuracy of 85.7% for first and second-line regimens. Prediction power was nevertheless lower in pretreated patients with a precision of 57.1% for subsequent chemotherapy lines. Intriguingly, a trend was observed towards a better performance of our system in prognosticating chemoresponsiveness vs. unresponsiveness (89.5% vs. 68.8%). Retrospective analysis of patient clinical data finally showed that the administration of a regimen predicted to be efficient ultimately translated into a longer progression-free survival. Importantly, the implementation of an automated pipetting system and the subsequent miniaturization of drug screenings enhanced our process capacity, conferring the possibility to extend our panel to a second list of approved targeted substances and significantly reducing the time before pharmacotyping. The access to longitudinal biopsies allowed to conduct WES on 14 PDOs to capture a comprehensive genetic profiling over the time of treatment, notably revealing a CHEK2-mutated patient responding over time upon PARP inhibitor maintenance therapy, in line with our PDO-based prediction, further highlighting the robustness of our method and algorithm. WES revealed a lower mutational burden and amount of nonsynonymous mutations upon chemotherapy, while tracking clonal evolution also revealed therapy-linked changes. Finally, to decipher the molecular mechanisms driving PC chemoresistance with an unprecedented level of accuracy, we took advantage of our phenotypically and therapeutically-profiled PDO living biobank and employed a true single-nucleus multiome approach to fully capture the transcriptome, as well as the chromatin accessibility landscape, of chemotherapy responsive and non-responsive PC PDOs. Overall, we report a robust and clinically-relevant preclinical tool for drug-response prediction, paving the way towards a PDO therapeutic profiling-guided true precision medicine in clinical routine. The identification of gene-regulatory networks and expression signatures driving chemoresistance may dramatically improve patient stratification and thus, foster true precision oncology. Citation Format: Johann Gout, Yazid J. Resheq, Jessica Lindenmayer, Thomas Ettrich, Lukas Perkhofer, Thomas Seufferlein, Alexander Kleger. Therapeutic profiling, patient response prediction, and tumor evolution in pancreatic cancer organoids [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Pancreatic Cancer; 2023 Sep 27-30; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(2 Suppl):Abstract nr C061.