In Vivo Biosafety Model To Assess Risk of Adverse Events from Retroviral and Lentiviral Vectors.

Abstract

Due to a heightened awareness of the potential for adverse events during human gene therapy trials, it is important to document the safety of the vectors and to examine the potential for malignancy arising from insertional mutagenesis from integrating vectors, using detailed in vivo analyses. In the current studies, we assessed the potential for generation of replication-competent retrovirus (RCR) and insertional mutagenesis causing adverse events, in human cells transplanted into immune deficient mice. Both retroviral and lentiviral vectors were assessed, in a total of 630 recipient mice. Human hematopoietic and mesenchymal stem cells carrying two different Moloney-based vectors were co-transplanted into beige/nude/XID and NOD/SCID mice that are unable to reject cells that become transformed. A total of 481 mice transplanted with human cells transduced by Moloney-based vectors, and an additional 149 mice that had received human cells transduced by HIV-based lentiviral vectors were monitored daily for adverse events for 2–18 months post-transplantation. An “adverse event” was logged in the experiment, the mouse was immediately sacrificed, and an autopsy was performed if ANY signs of ill health were observed, including any of the following: weight loss, hunching, lethargy, rapid breathing, skin discoloration or irregularities, bloating, hemi-paresis, enlarged lymph nodes or visible solid tumors under the skin. The presence of tumors and organ or lymph node abnormalities, or progression to monoclonal expansion in hematopoietic cells were specifically sought. The organs, marrow, blood, and serum were banked. Tissue sections were prepared from all organs (and visible tumor, if present). DNA was isolated from marrow, organs, blood, and any tumors to determine whether vector elements were present. Of the 630 recipient mice, 117 had some evidence of adverse effects ranging from development of leukemia to scarred skin or discolored organs detected upon autopsy. Human leukemic cells were found infiltrating the murine liver, spleen, blood and lymph nodes in 4 of these 117 mice. No vector DNA was present. 37 of the mice had developed solid tumors, determined to be of murine origin with no vector DNA present or RCR detected in their serum or tumor tissue. Despite the injection of 459 million transduced cells throughout the experiments, the engineered human MSC never gave rise to solid tumors. None of the serum samples or affected tissues from these mice had vector integrants or produced RCR in the marker rescue assay. In summary, 4 of the 117 total adverse events were due to the spontaneous development of human leukemia and the remainder of the events were due to development of cancerous states in the murine tissues, with no evidence of retroviral vector involvement. There was no evidence of insertional mutagenesis causing human leukemia or solid tumors in any mouse. No RCR were detected in 117 serum samples analyzed by vector rescue assay. No adverse events were caused by the vectors and no mouse tested had HIV p24 capsid protein in their serum. The current studies provide an in vivo system to assess potential risk from RCR and from insertional mutagenesis when retroviral or lentiviral vectors are considered as candidates for human gene therapy.