Tumor Xenografts of Human Clear Cell Renal Cell Carcinoma But Not Corresponding Cell Lines Recapitulate Clinical Response to Sunitinib: Feasibility of Using Biopsy Samples

Abstract

Background Parallel development of preclinical models that recapitulate treatment response observed in patients is central to the advancement of personalized medicine. Objective To evaluate the use of biopsy specimens to develop patient-derived xenografts and the use of corresponding cell lines from renal cell carcinoma (RCC) tumors for the assessment of histopathology, genomics, and treatment response. Design, setting, and participants A total of 74 tumor specimens from 66 patients with RCC were implanted into immunocompromised NOD-SCID IL2Rg-/- mice. Four cell lines generated from patients' specimens with clear cell pathology were used for comparative studies. Outcome measurements and statistical analysis Preclinical models were established and assessed. Engraftment rates were analyzed using chi-square testing. Analysis of variance (two-way analysis of variance) was conducted to assess tumor growth. Results and limitations Overall, 33 RCC mouse xenograft models were generated with an overall engraftment rate of 45% (33 of 74). Tumor biopsies engrafted comparably with surgically resected tumors (58% vs 41%; p = 0.3). Xenograft tumors and their original tumors showed high fidelity in regard to histology, mutation status, copy number change, and targeted therapy response. Engraftment rates from metastatic tumors were higher but not more significant than primary tumors (54% vs 34%; p = 0.091). Our engraftment rate using metastases or biopsies was comparable with recent reports using resected primary tumors. In stark contrast to corresponding cell lines, all tested xenografts recapitulated patients' clinical response to sunitinib. Conclusions Patient-derived xenograft models can be effectively established from tumor biopsies. Preclinical xenograft models but not matched cell lines reflected clinical responses to sunitinib. Patient summary Matched patient-derived clear cell renal cell carcinoma xenografts and cell lines from responsive and refractory patients treated with sunitinib were established and evaluated for pharmacologic response to anti–vascular endothelial growth factor treatment. Both models accurately reflected the genetic characteristics of original tumors, but only xenografts recapitulated drug responses observed in patients. These models could serve as a powerful platform for precision medicine.

Modeling biological and genetic diversity in upper tract urothelial carcinoma with patient derived xenografts

“…We have complied with all relevant ethical regulations. All surgical specimens were transported in RPMI medium or PBS on ice, washed with cold PBS, cut into small pieces (2 × 2 to 3 × 3 mm 3 ) and then implanted subcutaneously into one or two sides of the flank of 6-8-week-old male immunocompromised NOD-SCID IL2Rg −/− (NSG) mice (Jackson Laboratory) via a 4-5 mm skin incision over the mid-lumbar spine 46,47 . PDX from UCC11 was generated from a frozen cell pellet after a collagenbased digestion process.…”

Patient-Derived Xenograft Models in Urological Malignancies: Urothelial Cell Carcinoma and Renal Cell Carcinoma

“…Various agents have been tested in the pre-clinical setting using PDX models to investigate tumor response. Importantly, it has been shown that PDX models for RCC can better correlate with clinical response to sunitinib than cancer cell lines [57]. In one of the largest studies to date, Sivavand et al performed validation of targeted therapies in 94 different RCC patient tumor samples, 35 of which yielded viable tumor xenografts, and 16 were able to be serially passed successfully.…”

Clonal heterogeneity influences drug responsiveness in renal cancer assessed by ex vivo drug testing of multiple patient‐derived cancer cells

“…Functional studies based on model systems to recapitulate genetic and phenotypic characteristics of the patient profile such as patient‐derived xenograft (PDX)/tissue grafts (TG), organoids and patient‐derived cultures that have recently been established from RCC . TG models were generally well‐characterized based on histology, gene expression and genomics, and are known to retain regional ITH .…”

“…TG models were generally well‐characterized based on histology, gene expression and genomics, and are known to retain regional ITH . TG models only allow for small scale drug testing studies, however, may recapitulate the drug responsiveness in patients . A study on a single patient's PDX models derived from a primary and a lung metastatic sample showed noticeable variation in drug response profiles inferred from the single cell RNA‐sequencing (scRNA‐seq) of the paired samples, and underlined the impact of ITH in RCC .…”