Renal cancer: new models and approach for personalizing therapy
Background: Clear cell RCC (ccRCC) accounts for approximately 75% of the renal cancer cases. Surgery treatment seems to be the best efficacious approach for the majority of patients. However, a consistent fraction (30%) of cases progress after surgery with curative intent. It is currently largely debated the use of adjuvant therapy for high-risk patients and the clinical and molecular parameters for stratifying beneficiary categories. In addition, the treatment of advanced forms lacks reliable driver biomarkers for the appropriated therapeutic choice. Thus, renal cancer patient management urges predictive molecular indicators and models for therapy-decision making. Methods: Here, we developed and optimized new models and tools for ameliorating renal cancer patient management. We isolated from fresh tumor specimens heterogeneous multi-clonal populations showing epithelial and mesenchymal characteristics coupled to stem cell phenotype. These cells retained long lasting-tumor-propagating capacity provided a therapy monitoring approach in vitro and in vivo while being able to form parental tumors when orthotopically injected and serially transplanted in immunocompromised murine hosts. Results: In line with recent evidence of multiclonal cancer composition, we optimized in vitro cultures enriched of multiple tumor-propagating populations. Orthotopic xenograft masses recapitulated morphology, grading and malignancy of parental cancers. High-grade but not the low-grade neoplasias, resulted in efficient serial transplantation in mice. Engraftment capacity paralleled grading and recurrence frequency advocating for a prognostic value of our developed model system. Therefore, in search of novel molecular indicators for therapy decision-making, we used Reverse-Phase Protein Arrays (RPPA) to analyze a panel of total and phosphorylated proteins in the isolated populations. Tumor-propagating cells showed several deregulated kinase cascades associated with grading, including angiogenesis and m-TOR pathways. Conclusions: In the era of personalized therapy, the analysis of tumor propagating cells may help improve prediction of disease progression and therapy assignment. The possibility to test pharmacological response of ccRCC stem-like cells in vitro and in orthotopic models may help define a pharmacological profiling for future development of more effective therapies. Likewise, RPPA screening on patient-derived populations offers innovative approach for possible prediction of therapy response.