Current landscape of vector safety and genotoxicity after hematopoietic stem or immune cell gene therapy

Abstract

Malignant transformation of gene modified haematopoietic stem cells caused anxiety following adverse events in early clinical trials using gamma-retroviral vectors (γRV) to correct haematopoietic stem cells (HSC) in monogenic immune disorders. Adoption of HIV-derived lentiviral vectors (LV) with SIN (self-inactivating) configurations greatly reduced risks and subsequently hundreds of patients have been dosed with HSC gene therapy for blood, immune and metabolic conditions. Nevertheless, as experience builds, it's now well recognised that vector integration can drive clonal expansions and these may carry long term safety risks. Documented cases of haematological malignancy after SIN-LV gene therapy have recently emerged, in particular where heterologous retroviral promoters were employed and there are concerns around certain insulator elements and other possible contributors to clonal expansions. Similarly, tens of thousands of subjects have now received engineered T cell products, and longstanding dogma that mature T cells cannot be transformed is being questioned, with reports of a small number of malignant transformation events and wider concerns around secondary malignancies in some groups of patients. We summarize current clinical information and revisit genotoxicity risks following ex-vivo gene modification of HSC and T cells.

Fig. 1. Risk factors for genotoxicity & transformation.

This figure summarises potential risk factors for the development of malignancy after possible transformation events in engineered human cells, noting that underlying plasticity and stemness may be a critical factor, with more differentiated populations such as T cells less likely to transform than haematopoietic stem cells or pluripotent stem cells: (i) Host Factors: underlying genetic pre-dispositions may involve chromosomal changes or specific genetic mutations in pathways controlling cell signalling, proliferation, apoptosis or particular checkpoint pathways. Such changes may constitute an additional ‘hit’, operating in synergy with other events as they arise to drive clonal expansion; Defective immune surveillance due to underlying conditions or following immunosuppressive chemotherapy may result in a failure to recognise or effectively eliminate transformed populations; Cumulative toxicities may deliver ‘multiple hits’ following intensive chemotherapy, radiotherapy or after allogeneic stem cell transplantation; Underlying DNA fragility or DNA repair defects, are recognised as predisposing to cancer development, often involving multiple lineages beyond the haematopoietic system. (ii) Engineering Platforms: Integrating gamma retroviral and lentiviral vector platforms favour gene rich regions with predispositions towards transcription start sites and active genes respectively. Transactivation of proto-oncogenes, disruption of miRNA binding sites and other mechanisms may lead to insertional effects, and may be compounded by pre-existing mutations in certain circumstances. These risks may be influenced by the choice of endogenous promoter, with concerns around strong retroviral promoter and enhancer elements. There may also be risks related to high vector copy numbers or manufacturing processes with multiple rounds of gene transfer. The inclusion of nucleases for genome editing effects, including site-directed gene insertions, may be associated with predictable or unpredictable chromosomal changes. Alternative base editing or prime editing platforms may also mediate changes beyond those predicted at the DNA level.