In Vivo Genotoxicity and Selection Associated with Thiopurine Treatment in Subjects with Inflammatory Bowel Disease

Thiopurines azathioprine and 6-mercaptopurine are effective immune-modulators and cytotoxic agents widely used in the treatment of autoimmune diseases, graft rejection, and cancer. Of concern is the growing epidemiologic evidence that chronic exposure to these agents increases the risk of mesenchymal and solid tumors. Although in vitro and in vivo animal studies suggest that these agents are mutagenic and have the potential to select for mismatch repair deficient mutants, their in vivo mechanism(s) of carcinogenicity in humans remain(s) unclear. We investigated the in vivo mutagenicity of thiopurines in a cross-sectional study of 119 children and adults with inflammatory bowel disease by determining the frequency and spectra of somatic mutations at the HPRT locus using the T-cell cloning assay. We observed a significant increase in the frequency of somatic mutations that was related to total dose (p<0.001) and duration of treatment (p<0.001) among 56 thiopurine treated subjects compared to 63 untreated subjects. By mutation spectra comparison of 1020 mutant isolates, we also observed a significant dose-dependent increase in the proportion of G:C to A:T transitions (p<0.001) that is consistent with the in vitro mutagenic signature of these drugs. Although in vivo selection for HPRT mutants may have confounded the absolute mutation frequency since this is a ratio value, the significant thiopurine dose-dependent correlations with frequency and mutation spectra shift strongly suggest that thiopurine treatment is mutagenic in vivo. We also investigated the prevalence of in vivo clonal proliferation by the combined analyses of HPRT mutations and TCRβ CDR3 regions from mutant isolates and observed a higher prevalence of cell proliferation in association with chronic thiopurine therapy. Thiopurine treated subjects displayed expansions of precursor or committed progenitor cells and post-β rearrangement thymocytes or mature T-cells while untreated adult subjects showed only post-β rearrangement expansions of thymocytes or mature T-cells. Thiopurine treated subjects also uniquely displayed clonal expansions of “hyper-mutable” thymocytes or mature T-cells that gave rise to progenies with independent mutations either through extensive proliferation or from genetic instability. This pattern predominated in children and exhibited the highest occurrences of G:C to A:T mutations. G:C to A:T transitions were also observed in both early precursors or committed progenitors and post-β rearrangement thymocytes or mature T-cells from thiopurine treated subjects compared to untreated subjects who displayed a low frequency of G:C to A:T mutations unrelated to thiopurine therapy in post-β rearrangement thymocytes or mature T-cells. These data provide direct evidence that thiopurine treatment is mutagenic in vivo by mutation induction that may be enhanced through its effects on cell proliferation, and provide the first mechanistic evidence for the carcinogenicity of these therapeutic agents in humans. Our observations also suggest the potential risks to proliferating precursor and mature cells. In children, we may be capturing a subpopulation of “hypermutable” T-cells with a higher propensity to progress to lymphoma, which occur at highest frequency following transplantation.