Background
There is an unmet need for developing faithful animal models for preclinical evaluation of immunotherapy. The current approach to generate preclinical models for immunotherapy evaluation has been to transplant CD34+ cells from umbilical cord blood into immune-deficient mice followed by implantation of patient derived tumor cells. However, current models are associated with high tumor rejection rate secondary to the allograft vs. tumor response from human leukocyte antigen (HLA) mismatches. We herein report the first development of a novel, humanized patient-derived xenograft (PDX) model using autologous CD34+ cells from bone marrow aspirate obtained from a patient with metastatic clear cell renal cell carcinoma (mRCC) from whom a PDX had been developed. Case description: This is a 68-year-old Caucasian man diagnosed with mRCC with metastasis to the liver in 2014. He was treated with sunitinib +/- AGS-003 and underwent a cytoreductive right nephrectomy, left adrenalectomy and partial liver resection. PDX was generated using resected nephrectomy specimen. After surgery, patient received multiple lines of standard of care therapy including sunitinib, axitinib, bevacizumab, everolimus and cabozantinib. While progressing on cabozantinib, he was treated with nivolumab. Seven years after initiation of nivolumab, and 4 years after stopping systemic therapy, he remains in complete remission. To generate autologous PDX model, bone marrow aspirate was performed and CD34+ hematopoietic stem/progenitor cells (HSPCs) were isolated and injected into 150 rad irradiated non-obese diabetic scid gamma null (NSG) mice. At 11 weeks post-transplant, the matched patient PDX was injected subcutaneously into the humanized mice and the mice were treated with nivolumab. Conclusions: Our case represents successful therapy of nivolumab in mRCC. Furthermore, HPSCs obtained from a single bone marrow aspirate were able to reconstitute an immune system in the mice that allowed nivolumab to inhibit the tumor growth of PDX and recapitulated the durable remission observed in the patient with nivolumab. We observed the reconstitution of human T cells, B cells and natural killer (NK) cells and unlike the humanized mouse model using cord blood, our model system eliminates the tumor rejection from mis-matched HLA. Our autologous humanized renal cell carcinoma (RCC) PDX model provides an effective tool to study immunotherapy in a preclinical setting.
Conclusions
Our case represents successful therapy of nivolumab in mRCC. Furthermore, HPSCs obtained from a single bone marrow aspirate were able to reconstitute an immune system in the mice that allowed nivolumab to inhibit the tumor growth of PDX and recapitulated the durable remission observed in the patient with nivolumab. We observed the reconstitution of human T cells, B cells and natural killer (NK) cells and unlike the humanized mouse model using cord blood, our model system eliminates the tumor rejection from mis-matched HLA. Our autologous humanized renal cell carcinoma (RCC) PDX model provides an effective tool to study immunotherapy in a preclinical setting.