Abstract The National Cancer Institute has developed a Patient-Derived Models Repository (NCI PDMR; https://pdmr.cancer.gov) comprised of quality-controlled, early passage, and clinically-annotated patient-derived tumor xenografts (PDXs), organoids (PDOrgs), cell cultures (PDCs), and cancer associated fibroblasts (CAFs) available with genomic data to the extramural community for research use. Models are developed by the NCI PDMR in decreasing order of complexity, specifically 1) patient material can be used to develop PDXs, PDOrgs, PDCs, and CAFs, 2) PDX material for PDOrgs and PDCs, and 3) PDOrg material for PDCs, but a PDC is never used to develop a PDOrg or PDX. Eleven pairs of 22 matched PDCs have been developed in parallel from either patient, PDX, patient-derived organoid, or PDX-derived organoid tumor material and sequenced by WES and RNASeq. Genetic stability was assessed using multiple approaches including microsatellite instability (MSI) generated from MSISensor2, percentage of genomic loss of heterozygosity (LOH) using a set of ~800k heterozygous SNPs from a population level genomic database (gnomAD), pairwise Spearman correlation based on BIN level copy number (CN)/RNA expression profiles, and OncoKB annotated oncogenic/likely oncogenic variants. No systematic differences were observed within PDC pairs derived from different origins or compared to their patient and/or PDX material in MSI, LOH% and RNA expression profile but pairwise Spearman correlation (0.66-0.88) in CNV profiles were somewhat variable, likely due to low sequencing depth. In one PDC pair (299254), 3 out of 12 OncoKB annotated Indels and CNV showed opposite level of variant allele frequencies/CN when comparing a model derived from patient material to one developed from a PDX-derived organoid, possibly driven by a lineage-specific subclonal outgrowth when compared to patient and PDX data. Phenotypic characteristics of matched PDCs also overall show no major differences, though variability in growth rates and the ability to form spheroids in serum-free medium were noted. In one pair (919269), the PDC derived from patient material was able to form a cell line xenograft (CLX) in NSG mice but not the PDC developed from a patient-derived organoid. Overall, these models demonstrate a high degree of concordance at the genetic and phenotypic level when compared to the originating patient and/or PDX tumor. Though further characterization (e.g., preclinical drug testing) may be needed to define differences between matched PDC pairs, lack of access to patient tissue or failure to generate tumor cell cultures from one source of material should not hamper development of preclinical in vitro models from other patient-derived model types as long as the source-of-origin is clearly defined. Funded by NCI Contract No. HHSN261200800001E Citation Format: Cindy R. Timme, Ting-Chia Chang, Sergio Y. Alcoser, Gareth Bliss, Carrie Bonomi, Suzanne Borgel, John Carter, Alice Chen, Li Chen, Kevin Cooper, Biswajit Das, Kelly Dougherty, Lindsay Dutko, Marion Gibson, Michelle M. Gottholm-Ahalt, Tara Grinnage-Pulley, Shahanawaz Jiwani, Keegan Kalmbach, Chris Karlovich, Kimberly Klarmann, Tiffanie Chase, Michael Mullendore, Matthew Murphy, Kevin Plater, Gloryvee Rivera, Jessica Steed, Luke Stockwin, Yvonne A. Evrard, Mickey Williams, Dianne L. Newton, Melinda G. Hollingshead, James H. Doroshow. Comparing twenty-two matched patient-derived cell lines developed from either patient, PDX, or organoid tumor cell material [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 36.
Read full abstract