Ott MG, Schmidt M, Schwarzwaelder K, et al. Correction of X-linked chronic granulomatous disease by gene therapy, augmented by insertional activation of MDS1-EVI1, PRDM16 or SETBP1. Nat Med 2006;12:401-9.Correction of underlying hematopoietic cell disorders can be achieved by two major approaches, transplantation with HLA identical donors and genetic modification of early hematopoietic precursors in order to overcome a defined defect. The side effects of the former are well known, whereas the latter has its own set of difficulties including poor engraftment, failure to sustain the clone, and, finally, risks associated with integrating viruses that promote a growth advantage. In this manuscript, the investigators targeted chronic granulomatous disease by gene transfer in two patients. The underlying premise for this work was that this gene had no evidence of growth-promoting advantage from animal studies but would reconstitute functional NADPH oxidase activity. Their results have stirred the pot on both fronts. gp91phox, in a retroviral vector containing a spleen-focus forming virus LTR, was reconstituted into CD34+-mobilized peripheral blood cells from two patients (P1 and P2) and then introduced into those patients who were conditioned with liposomal busulfan. By day 80, peripheral blood was reconstituted with a modest but detectable proportion of virally transduced cells that expressed gp91phox. Integration preferentially occurred in gene coding regions and was skewed toward transcription start sites. Over time, the insertion sites were not stable and became less non-random, but still remained at multiple sites. After five months, it was evident that insertion occurred at three common integration sites at or near the genes encoding zinc finger proteins MDS1-EV11, PRD M1126, and SETBP1. The latter was the least frequent and occurred only in P1. In circulating blood, there was no elevation of total leukocytes or neutrophils. In P1, there was a single clone that dominated with an insertion in MDS1-EV11, whereas in P2 there was no single dominant clone. Transcripts of MDS1-EV11 and SETB11, but not PRDM16, were overexpressed; however, growth-factor-dependent growth was maintained in these clones. gp91phox expression was detected in circulating neutrophils and expression correlated with activity. However, the level of superoxide production was a fraction (1/3 to 1/7) of wild-type cells. Nonetheless, bacterial killing appeared improved, and the rate of serious infections was markedly reduced, particularly in P1 where the levels were higher.

Targeting a genetic defect in a defined cellular compartment remains attractive, but recent data that implicate vector insertion with the activation of an oncogene and leukemia have raised some concern about this approach. The true power of this study is in the detailed analysis of the vector insertion sites over time. Detailed and extensive analysis was done that demonstrated that reconstitution was due to clones that had vector insertion into three discrete genes which encode zinc-finger transcription factors. All of these genes have been associated with translocations in patients with leukemia and raise a concern about the long-term fate of the hematopoietic precursor cells in these patients. At the time of the study, it is evident that the expression of the vector is restricted to the myeloid compartment. It is clear that the insertion into these sites has conferred a selection advantage on these clones with other insertion sites disappearing with time. The major question that is raised from these findings is whether the expression and control on expansion will remain a feature of these cells with time. Expressing NADPH oxidase, even though not to the level of wild type and not in all cells, benefits patients with this disease. It may well be that the vector choice has made the difference in the insertion site. The gradual restriction to such a discrete set of clonal insertion sites that are genes associated with leukemia, and the growth of the select group of clones raises concern that a small population of cells with a capacity for self renewal may remain in the pack to repopulate with an immature population. The results so far support that insertion into this site still leads to a restricted differentiated population of cells that reconstitute a critical protein function. The data presented continue to add to the questions of targeted gene therapy, but the type of detailed analysis in this work may also be used to direct vector choices and the cell population in which the gene is introduced.

Competing Interests

Dr. Petruzzelli indicated no relevant conflicts of interest.