Murine stroma adopts a human-like metabolic phenotype in the PDX model of colorectal cancer and liver metastases

Abstract

Cancer research is increasingly dependent of patient-derived xenograft model (PDX). However, a major point of concern regarding the PDX model remains the replacement of the human stroma with murine counterpart. In the present work we aimed at clarifying the significance of the human-to-murine stromal replacement for the fidelity of colorectal cancer (CRC) and liver metastasis (CRC-LM) PDX model. We have conducted a comparative metabolic analysis between 6 patient tumors and corresponding PDX across 4 generations. Metabolic signatures of cancer cells and stroma were measured separately by MALDI-imaging, while metabolite changes in entire tumors were quantified using mass spectrometry approach. Measurement of glucose metabolism was also conducted in vivo using [F]-fluorodeoxyglucose (FDG) and positron emission tomography (PET). In CRC/CRC-LM PDX model, human stroma was entirely replaced at the second generation. Despite this change, MALDI-imaging demonstrated that the metabolic profiles of both stromal and cancer cells remained stable for at least four generations in comparison to the original patient material. On the tumor level, profiles of 86 water-soluble metabolites as well as 93 lipid mediators underlined the functional stability of the PDX model. In vivo PET measurement of glucose uptake (reflecting tumor glucose metabolism) supported the ex vivo observations. Our data show for the first time that CRC/CRC-LM PDX model maintains the functional stability at the metabolic level despite the early replacement of the human stroma by murine cells. The findings demonstrate that human cancer cells actively educate murine stromal cells during PDX development to adopt the human-like phenotype.

Challenges in Stratifying the Molecular Variability of Patient-Derived Colon Tumor Xenografts

“…Recent work showed that PDX propagation causes changes in DNA copy number, allowing clones with a minor representation in the xenograft to obtain a fitness advantage over the course of PDX passage [6]. On the other hand, murine stroma cells adapt a human-like metabolome profile in PDXs [30]. Given the divergent and common features of PDX models and primary tumors, an extended PDX-tailored classifier that includes multiple molecular features should be developed and trained on the growing number of thoroughly characterized CRC PDX samples.…”

The fidelity of cancer cells in PDX models: Characteristics, mechanism and clinical significance

“…Second, human tumor stromal cells and extracellular matrix (ECM) that are transplanted into immunodeficient mice are likely replaced by mouse tissue, which can significantly impair studies of the tumor microenvironment (TME) and cancer immunotherapy in PDX models . To some extent, mouse stromal cells in the TME share common features with matched human cells in parental tumors, but the exact impact of murine stroma on human cancer cells in PDX models remains unknown. Third, the engraftment of a PDX can cause lymphoma but not the expected tumors in host mice if the donor patient has a history of Epstein Barr virus infection …”

Patient-Derived Urothelial Cancer Xenograft Models: A Systematic Review and Future Perspectives

“…As summarized above, in general UC PDX maintain the genetic background of their donor patient tumors and can also retain patient stromal cells at early passages. In addition, a recent report in colorectal cancer PDX models showed that after the early replacement of the human stroma by murine cells, the murine stroma adopts a human-like metabolic phenotype suggesting that the PDX stroma recapitulates primary tumor stroma in important ways [51]. If a similar adaptation take place in UC PDX, it will also be a great advantage for future translational studies aiming at assessing therapies and concepts that involve both cancer cells and stroma.…”