The PDX model uses the respiratory organ of the chicken embryo known as the chorioallantoic membrane (CAM). During a cytoreductive nephrectomy, several spatially distinct cores are taken from the tumor and sampling occurs for any metastases present. These cores are then subdivided into several fragments and grafted directly on the surface of the CAM and are readily accessible to either topical dosing of anti-angiogenic agents or intravascular injection. Typically, 5-6 tumor cores are gathered per primary tumor and 1-3 cores from metastatic sites. Each one of these tumor cores is subdivided into at least 36 fragments for engraftment.
Histology from a patient tumor fragment showed that the clear cell RCC subtype was maintained post-engraftment and there was good integration between the graft and host tissue. Patient #1 who presented with a 12 cm primary tumor and 3 chest wall metastases underwent a Sunitinib sensitivity assay. IMDC criteria put this patient in the poor risk group with an expected survival of 7.4 months. PDXs on this patient found that majority of the patient’s primary tumor cores, and more notably, all their metastatic lesions showed a profound response to Sunitinib therapy. This patient went on to receive TKI therapy for 11 months, well past their expected survival of 7.4 months upon which they progressed and are now on nivolumab (>50% reduction in tumor volume) and expansion of T-cells (4X fold expansion compared to vehicle treated PDXs) over 8 days of treatment. Patient #2 had a 7.3 cm primary tumor and 1 duodenal metastasis, putting them in the intermediate risk group at a 22.5 month survival under IMDC criteria. PDXs on this patient showed that half of their primary tumor cores showed a good response to Sunitinib therapy, but regrettably their metastatic lesion did not show any treatment effects and this patient progressed after only 3.4 months under systemic TKI therapy. Patient #3 had 3 lymph node metastases and was screened for nivolumab and pembrolizumab. Overall this patient was found resistant to both therapies and no TILs or TIL expansion were observed.
In conclusion, CAM PDX model can screen for patient drug response through the direct engraftment of patient tumor fragments and the tumor subtype appears to be consisted and unchanged. This PDX model also preserves tumor infiltrating lymphocytes likely due to the rapid engraftment with short study timeline, and this further permits a phenotype-based readout for patient response to PD-1/PD-L1 blockade therapies
Presented by: Matthew Lowerison, MD
Written by: Rita Derderian, Department of Urology, University of California-Irvine at the 2018 AUA Annual Meeting - May 18 - 21, 2018 – San Francisco, CA USA