Establishment Of A Predictive Patient-Derived Xenograft Model For Renal Cell Carcinoma

Renal Cell Carcinoma (RCC) is classically considered a difficult malignancy to diagnose and treat. Although the surgery is a resolving approach for many local tumors, the metastatic disease generally is characterized by poor outcomes. The lack of valid preclinical cancer models has hampered the discovery of valuable diagnostic, prognostic and predictive biomarkers to develop effective therapeutic options. In order to create preclinical models, we established an orthotopic patient derived xenograft (PDX) murine model using an enriched stem cell like-heterogeneous bulk obtained from surgery patients’ specimens. To predict effective conventional and innovative therapeutic treatment, the stem cell like-heterogeneous bulks were analyzed before injection for key total and phosphorylated protein expression by Reverse Phase Protein Array (RPPA). METHODS: Using specific enzymatic dissociation and culture conditions, we first isolated enriched stem cell like-heterogeneous bulks from fresh surgery specimens derived from 30 patients who underwent nephrectomy. After one week, bulks were orthotopically injected in immunocompromised mice (NSG mice). Cells were infected with lentiviral vector (Tween-Luc) and in vivo monitored by IVIS imaging system. The percentage of engrafting bulks, sizes and distal spreading capacity were evaluated. Several xenografts were further dissociated, serially inoculated and propagated for up to eighth generations. Bulks were analyzed before injection by RPPA. This technique performs the detection of total and phosphoproteins allowing the analysis of hundreds of proteins. This RPPA platform includes specific antibodies for key proteins belonging or involved in targeted therapy signaling cascade. RESULTS: Approximately 67% of the implanted samples engrafted. We observed a correlation between the engraftment success and the aggressiveness of the parental tumor. Moreover, the PDXs obtained from more aggressive tumors showed increased size. Hematoxylin and Eosin staining showed that PDX displayed similar histological architectures to parental tumors. Furthermore, immunohistochemical analysis demonstrated that PDX retained CD10 and PAX 8 expression, two typical RCC biomarkers, at similar level than parental tumors. Finally, we obtained maps highlighting the specific activated proteins which are key candidates of targeted therapy pathways by RPPA. CONCLUSIONS: Our established in vitro and PDX models recapitulate tumor histology and molecular characteristic creating new source for ameliorating diagnosis, prognosis and therapy sensitiveness prediction. Furthermore, we were able to propagate PDXs from the same patient for up to eight generations. Since in average tumor uptake is 4-5 weeks, it would be theoretically possible to test in parallel therapeutics in murine models while the corresponding metastatic patient is under therapy. The ability to generate large PDX animal cohorts is very promising and could be exploited to obtain valuable preclinical platforms for innovative drug testing. Moreover, the possibility to map a wide variety of intracellular pathways and biological functions in enriched stem cell like-heterogeneous bulks enable us to explore protein-protein interaction networks with potential impact on drug activity and to identify deregulated circuits related to pharmacological inhibition. Citation Format: Simona Di Martino, Gabriele De Luca, Ludovica Grassi, Giulia Federici, Laura De Salvo, Anna Laura Di Pace, Antonio Addario, Giovanni Muto, Manuela Costantini, Mauro Biffoni, Michele Signore, Steno Sentinelli, Michele Milella, Michele Gallucci, Desiree Bonci, Ruggero De Maria. Establishment of a predictive patient-derived xenograft model for renal cell carcinoma. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr LB-040.