Abstract Polyploid giant cancer (PGCC) state is a common response of cancer cells to various stressors including chemotherapy, irradiation, hypoxia and viral infection. Upon stress, PGCCs adopt an endoreplication in which the genome replicates, mitosis is omitted, and cells grow in size, leading to resistance. How accessible endoreplication is to a cell, therefore, directly translates to its resistance to therapy. In this study, we hypothesized that endoreplication and PGCC state are more plausible to cancer cells that already have a higher ploidy content prior to therapy. To test this hypothesis, we developed a comprehensive three-tier framework consisting of i) computational, ii) in vitro and iii) in silico components. We designed a set of ordinary differential equations (ODEs) to model how cells enter and exit the PGCC state in response to a given stressor. We engineered how this in silico approach interacts with in vitro experiments into a broadly applicable software solution called CLONEID. The software uses computer vision to monitor phenotypic changes in cell and nuclear size from standard bright-field microscopy and classify cells into PGCC and non-PGCC states. We used CLONEID to test various therapeutic agents for their ability to select for a stable near-tetraploid population in a set of near-diploid cell lines. Through spontaneous cell fusions, we also obtained tetraploid breast cancer cells matched with their parental lines. Altogether, this framework, now, enables us i) to monitor the PGCC state experimentally ii) in matched cell lines with differential DNA content iii) to model the successful entrance and exit rates to and from the endoreplication state, respectively. As the first application of this framework, we tested the ability of our matched triple-negative breast cancer (TNBC) lines (SUM-159 and MDA-MB-231) to access the PGCC state upon treatment with 18 commonly-used chemotherapy agents. We observed that gemcitabine caused continued cell growth without cell division in both tetraploid SUM-159 and MDA-MB-231 cells whereas near-diploid parental cells were hypersensitive to the treatment. Consequently, tetraploid cancer cells continued to safely grow in the presence of gemcitabine. Furthermore, these PGCCs re-entered the proliferative cell cycle and grew in cell number when treatment is terminated. Gemcitabine (GEM)-based chemotherapy is a standard treatment for patients with TNBC although its efficacy is limited mainly due to drug resistance. Moreover, such poor response rate is coupled with severe side effects that frequently leads to the failure of vital organs, serious secondary diseases, and co-morbidities. Our findings suggest that ploidy is a predictive biomarker for gemcitabine sensitivity. Thus, we expect our findings and three-component framework strategy to help stratify the TNBC patient population by their response to gemcitabine. In addition, our mathematical modelling approach has the promising potential to inform personalized dose optimization and to effectively decrease administered gemcitabine dose for a subset of patients, which would alleviate therapy-associated side effects and lethalities. Citation Format: Vural Tagal, Jackson P. Cole, Daria Miroshnychenko, Andriy Marusyk, Noemi Andor. Ploidy as a predictive biomarker for gemcitabine sensitivity in triple-negative breast cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB175.
Read full abstract