Abstract 2500: Technical feasibility of CRISPR-based functional profiling for solid tumor precision medicine
Karl Hodel,Andreas Varkaris,Nicholas Chevalier,Maci Meyers,Allison Kehlmann,Stephane Corbel,Anastasia Stavridi,Peter Kilfeather,Leonardo Hagmann,Kathleen Tran,Yali Li,Christian Schmedt,Ioannis Sanidas,Srinath Sampath,Srihari Sampath,Dejan Juric
DOI: https://doi.org/10.1158/1538-7445.am2024-2500
IF: 11.2
2024-03-22
Cancer Research
Abstract:Abstract Introduction: Next generation sequencing (NGS) has revolutionized precision oncology by enabling identification of cancer patients harboring actionable mutations who may benefit from corresponding targeted therapies. However, sequencing-based biomarkers are not detected for most patients and imperfectly predict response to therapy even when present. To address this unmet need, we have developed a CRISPR-based diagnostics platform for ‘personalized functional genomics’, allowing identification of patient-specific therapeutic vulnerabilities irrespective of mutation status. We have previously applied this approach to primary hematologic malignancies, and here report application to solid tumor-derived samples. Methods: Patient-derived solid tumor samples from several tumor types including breast, ovarian, colon and lung were obtained from 3 sources: fresh tissue (obtained from patients via core biopsy, surgical resection, or malignant fluid sampling), primary Patient-Derived Cell Lines (PDCLs, generated from surgical material and adapted for ex vivo culture), and commercially obtained viably frozen dissociated tumor cells (DTCs). Samples were dissociated to single cell suspensions and transduced with a lentivirus encoding Cas9 and an sgRNA library covering a custom targeted oncology gene panel. The panel consists of genes encoding targets of FDA-approved oncology drugs, as well as reference genes. After a short culture period, transduced cells were harvested and relative sgRNA distribution was measured by amplicon NGS. Changes in sgRNA abundance were collapsed into gene-level log2-fold changes, and significantly depleted genes scored as functional dependencies. For select PDCLs, cells were treated with cognate small molecule inhibitors over the course of 4 days. Cell viability was read out using CTG assay and a calculated IC50 lower than 500nM inferred as inhibitor sensitive. Results: In a double-blind study using primary PDCLs, functional dependencies identified by CRISPR perfectly discriminated alpelisib sensitive (2/2) from resistant samples (6/6). Using fresh primary samples from varying cancer types and specimen sources, we achieve a technical success rate >70%. Among gene dependencies identified, targets of FDA-approved oncology drugs and those under late-stage clinical investigation were both observed including BRAF, CDK4/6, VEGFR1/2 and PARP1/2. Conclusion: We demonstrate technical proof of concept of the FX platform for CRISPR-based personalized functional genomics. This approach is feasible when applied to primary patient samples from a variety of tumor types and tissue sources, and results are validated by comparison with orthogonal methods. CRISPR-based functional profiling represents a powerful adjunct to traditional genomic profiling, for the first time allowing drug target dependency mapping in a patient-specific and mutation-agnostic manner. Citation Format: Karl Hodel, Andreas Varkaris, Nicholas Chevalier, Maci Meyers, Allison Kehlmann, Stephane Corbel, Anastasia Stavridi, Peter Kilfeather, Leonardo Hagmann, Kathleen Tran, Yali Li, Christian Schmedt, Ioannis Sanidas, Srinath Sampath, Srihari Sampath, Dejan Juric. Technical feasibility of CRISPR-based functional profiling for solid tumor precision medicine [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2500.
oncology