Comprehensive genomic profiling of urothelial carcinoma cell lines reveals hidden research bias and caveats.

Abstract

Dear Editor, Preexisting genetic mutations in cells may influence phenotypes together with other molecular biology manipulations. For instance, the bladder cancer cell RT4 harbors a TACC3-FGFR3 fusion rather than TP53 mutation (Figure 1F), whereas the other bladder cancer cell T24 have TP53 mutation but not FGFR3 mutation.1 We questioned whether those preexisting genotypes could influence research conclusions and systematically overviewed urothelial carcinoma-related 1589 articles either used or not used RT4 (Table S1). We found a significant association between target gene corresponding pathways and RT4 usage frequency. For example, the research was less likely to include RT4 (Figures 1A and 1B and Table S2) if the target gene was associate with TP53 pathways. We calculated the frequency of using RT4 and/or T24 for research on particular target gene (Table S3). Using PPI networks,2 we calculated the “interaction distance” between these genes to TP53 and FGFR3. RT4 usage frequency was increased in research of genes with closer relationship to FGFR3 compared to TP53 (Figure 1C). Those results suggested the existence of latent bias in research, that is, researcher may implicitly select consistent results by selecting carcinoma cell lines. The preexisting carcinoma cell line mutation landscape could determine its response to particular molecular biology manipulation, which can be inconsistent, even though only consistent results would be reported. Moreover, to define the genotype of carcinoma, cell line is critical to the research result. State-of-the-art practices require genotyping cell lines using short tandem repeat (STR).3 However, STR only identifies whether cell lines are cross-contaminated, but does not fully report the genomic changes of cell lines. We collected eight most widely used bladder cancer cell lines and confirmed the identity by STR. Whole-exome sequencing (WES) of the eight cell lines together with two conditionally reprogrammed cells (CRCs)4 and two bladder cancer tissues (Table S4) showed the CRCs have good consistency with tumor tissues; however, the cell lines showed significant deviation from genuine bladder cancer by wide-spread shattering copy number variation (Figures 1D and S1). Furthermore, comparing with the previous reports,5 we detected only a tiny fraction of the known driver FGFR3 fusion in SW780 (Figure 1E). Those results indicated that the carcinoma cell lines may undergone pervasive genetic drifts characterized by copy number variation and random loss of driver mutation. In conclusion, carcinoma cell line genotypes could influence research results by modifying molecular manipulation outcomes. Such influences could be implicit or explicit, and represent researcher's selection bias in research. Furthermore, the genotypes undergone neutral genetic drift, which might lead to loss of important driver mutation and gain of novel genetic identity. Thus, it is important that WES, instead of STR, should be widely used for determining carcinoma cell line genotypes. The CRCs closely resemble original tumor tissue and represent a better alternative for in vitro research to benefit the development of future precise medicine. We gratefully acknowledge excellent technical assistance provided by Ms. Yuan Zhu, Ms. Mengxue Yu, and Ms. Yayun Fang from Zhongnan Hospital of Wuhan University. This study was supported in part by grants from the Science and Technology Department of Hubei Province Key Project (2018ACA159) and Medical Science Advancement Program (Clinical Medicine) of Wuhan University (TFLC2018002). The authors declare no conflict of interest. All the data and materials are available. All the authors consent for publication. The study using clinical information and human samples (including surgical tissue specimens and primary cancer cells) was approved by the Ethics Review Committee at Zhongnan Hospital of Wuhan University (approval number: 2015029). Human sample preservation by the Zhongnan Hospital Biobank, the official member of the International Society for Biological and Environmental Repositories (https://irlocator.isber.org/details/60), was approved by the Ethics Review Committee at Zhongnan Hospital of Wuhan University (approval number: 2017038) and China Human Genetic Resources Management Office, Ministry of Science and Technology of China (approval number: 20171793). YL and LJ performed experiments. GW, CC, YW, and LC performed literature review. All authors contributed to the research. YL, LJ, YZ, YX, and XW designed the study and wrote the manuscript. All authors approved the final manuscript. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.