J. L. Leiting1, M. C. Hernandez1, L. Yang2, J. R. Bergquist1, M. J. Truty1 1Mayo Clinic,Department Of Surgery,Rochester, MN, USA 2Mayo Clinic,Center For Individualized Medicine,Rochester, MN, USA
Introduction:
Patient-derived xenografts (PDX) provide clinically relevant translational models that accurately recapitulate individual patient tumor histopathologic and molecular phenotypes. Maintaining high engraftment rates is critical and loss of PDX models are due to engraftment failure or development of lymphoproliferative tumors (LTs). Previous PDX work has suggested the etiology of LTs are due to tumor associated lymphocytes and/or Epstein Barr Virus activation in an immunodeficient environment. Here, we report our efforts to decrease rates of LTs from a high volume PDX program in order to improve PDX engraftment efficiency. To address the issue of LTs, we hypothesized that routine injection of rituximab (an anti-CD20 antibody) at time of tumor implantation would maximize engraftment by reducing the rate of LTs.
Methods:
With IRB and IACUC approval, surgically resected primary patient tumors were implanted into the flanks of NOD SCID mice according to our protocol. We assessed the effect of routine rituximab injection in pre- and posttreatment groups. Implanted mice were monitored weekly for time to tumor formation (TTF) and all derived PDX models were verified by a GI cancer pathologist. Chi squared and Fisher’s Exact test were used for statistical comparison.
Results:
A total of 697 generations of PDX have been implanted with 290 individual patient tumors. These include 145 pancreatic cancers (60 treatment naïve, 85 neoadjuvant), 66 cholangiocarcinomas (CCA), 38 hepatocellular carcinomas (HCC), and 41 miscellaneous GI tumors. Of these, 308 generations received rituximab compared to 389 that did not. Overall rates of LTs were 11.3% (8.6% in primary PDX engraftments and 13.3% in subsequent generations). LT rates varied by tumor type with HCCs having the highest overall rate (19.7%), followed by naïve pancreatic cancers (11.1%) and CCAs (10.1%). Rituximab treatment decreased LT rates from 13.1% to 9% (p<0.001) in the entire cohort and as a result increased overall successful PDX engraftment. This was consistent in both primary PDX engraftment (10.5% vs 3%, p<0.001) as well as subsequent generations (16.2% vs 10%, p<0.001.) Rituximab had the most significant decrease in LTs in CCAs (13.1% vs. 5.1%, p=0.0019) and HCCs (26% vs. 13%, p<0.001).
Conclusions:
Engraftment failure due to LT formation is detrimental and leads to excessive costs and inefficiencies. Rates of LTs appear to decrease and overall engraftment rates increase with routine rituximab treatment at the time of implantation for both primary and secondary passages. Historically difficult to engraft tumors like CCAs and HCCs appear to benefit the most by rituximab treatment. Further study evaluating the etiology of LTs in PDX is warranted.
