Bridging the Gap with Evolving Kidney Cancer Preclinical Models - Beyond the Abstract

Tumor progression in clear cell renal cell carcinoma (ccRCC), the most prevalent type of kidney cancer, is a complex multistep process involving genetic alterations and tumor microenvironmental (TME) crosstalk. Understanding RCC pathophysiology through preclinical modeling has allowed for significant advances in the therapeutic armamentarium for patients. Considerable improvements in patient outcomes have been seen with a combination of anti-angiogenics and immunotherapy drugs; however, understanding factors that lead to tumor response in patients will require further exploration using advanced preclinical platforms. Our article reviews cell line, murine, and 3D organoid models relevant in ccRCC research today, each of which poses a unique set of advantages and disadvantages important for comprehensive interrogation of various aspects of ccRCC.1

Patient-derived cell line models are ideal for high-throughput drug screenings and validation studies. Much of our current understanding of the von Hippel Lindau- hypoxia inducible factor protein (pVHL-HIF) axis has been elucidated by studies using cell lines; 786-O, A-498, and ACHN, the top three cited RCC cell lines.2 Mutation in the VHL gene is a truncal event occurring in chromosome 3p that sets the stage for additional chromatin damage to accumulate over time.3 Cell line models are cost-effective and allow for manipulation of RCC relevant genes including VHL, HIF, PBRM1, SETD2, and BAP1. Cell line models alone lack the complexity needed to study tumor microenvironmental interactions and intra-tumoral heterogeneity (ITH).
Murine models in ccRCC consist of a cell line derived xenograft (CDX), patient-derived xenograft (PDX), and genetically engineered mouse models (GEMMs). CDX and PDX models allow for in vivo vascularized physiology by either orthotopic (kidney subcapsular) or heterotopic (subcutaneous) transplantation. These models are advantageous for maintaining patient tumor histology and consistency of drug responsiveness, making them ideal for drug testing in an in vivo system. However, accurate representation of the TME is limited in a CDX or PDX model because of the requisite immune suppression necessary for successful tumor implantation.

GEMM models allow for the manipulation of ccRCC relevant genes (VHL plus one or more co-drivers) in an immunocompetent animal and promote in situ tumor development. Contrary to human ccRCC, a VHL mutation in mice does not predispose to the development of RCC. In the human genome, VHL, PBRM1, SETD2, and BAP1 are all located on chromosome 3p, while VHL and commonly mutated chromatin modifying genes are located on different chromosomes in the murine genome.4 Current validated GEMMs in ccRCC exhibit a range of pathological tumor characteristics including low- and high-grade transplantable tumors, metastasis to the liver, and evidence of sub-clonal evolution. A common limitation shared by all GEMM models is the lack of sequential mutagenic events shown to occur in human ccRCC. This will be a technically challenging obstacle, though groups are implementing CRISPR/Cas9 technology with success mimicking clonal selection in other types of cancer.5

Patient-derived organoid models (PDOs) have become increasingly recognized by cancer researchers as a way to bridge the translational gap between simple cell line models and complicated murine systems. By maintaining a 3D architecture, PDOs are poised to recapitulate the stromal-inflammatory-milieu of the original ccRCC patient tumor and provide a platform for basic research and personalized medicine. One major challenge will be maintaining the ITH and hypoxic environment representative of most ccRCC tumors, while also promoting growth and survival of each cell type for TME interactions to be studied extensively.  

Model systems are imperfect but despite their deficiencies, each can add value if used in the proper context. Our work highlights the genotype and phenotypic details of these models for concise and easy access. We aim to elucidate areas in ccRCC modeling that need expansion and provide appropriate guidelines for future research to explore.

Written by: Melissa M. Wolf, PhD Student, and Kathryn E. Beckermann, MD, PhD, Department of Medicine, Division of Hematology and Oncology, Vanderbilt University Medical Center, Nashville, Tennessee


  1. Wolf, Melissa M., W. Kimryn Rathmell, and Kathryn E. Beckermann. "Modeling clear cell renal cell carcinoma and therapeutic implications." Oncogene (2020): 1-14.
  2. Sinha, Rileen, Andrew G. Winer, Michael Chevinsky, Christopher Jakubowski, Ying-Bei Chen, Yiyu Dong, Satish K. Tickoo et al. "Analysis of renal cancer cell lines from two major resources enables genomics-guided cell line selection." Nature communications 8, no. 1 (2017): 1-10.
  3. Mitchell, Thomas J., Samra Turajlic, Andrew Rowan, David Nicol, James HR Farmery, Tim O’Brien, Inigo Martincorena et al. "Timing the landmark events in the evolution of clear cell renal cell cancer: TRACERx renal." Cell 173, no. 3 (2018): 611-623.
  4. Bult, Carol J., Judith A. Blake, Cynthia L. Smith, James A. Kadin, and Joel E. Richardson. "Mouse genome database (MGD) 2019." Nucleic acids research 47, no. D1 (2019): D801-D806.
  5. Ideno, Noboru, Hiroshi Yamaguchi, Takashi Okumura, Jonathon Huang, Mitchell J. Brun, Michelle L. Ho, Junghae Suh, Sonal Gupta, Anirban Maitra, and Bidyut Ghosh. "A pipeline for rapidly generating genetically engineered mouse models of pancreatic cancer using in vivo CRISPR-Cas9-mediated somatic recombination." Laboratory Investigation 99, no. 8 (2019): 1233-1244.
Read the Abstract