Polymorphisms in Human Homeobox HLX1 and DNA Repair RAD51 Genes Increase the Risk of Therapy-Related Acute Myeloid Leukemia.

It is generally accepted that genetic low penetrance susceptibility and resistance genes within the general population make a significant contribution to therapy-related acute myeloid leukaemia (t-AML) risk. The risk of malignant transformation is defined by the number of mutations required and the mutation rate, but one additional theoretical risk factor is the target cell frequency, as this defines the number of genomes at risk. Whilst the genetic control of target cell frequency is unknown in humans, a leading candidate gene is the HLX1 homeobox transcription factor gene. HLX1 is specifically expressed on CD34⁺ cells and plays a vital role in hemopoietic development. In addition the HLX1 gene lies within a proposed human t-AML susceptibility locus on chromosome 1, as defined by murine radiation-induced AML genetic studies. We have examined the distribution of a C/T-3′UTR polymorphism in HLX1 in 189 AML patients, including 42 cases of t-AML, and 169 control subjects. The presence of a variant HLX1-3′UTR T allele resulted in a significant increase in the risk of t-AML (OR= 3.36, 95% CI 1.65–6.84); to our knowledge, this is the first time that a gene which is implicated in target cell biology has been associated with an increased risk of malignant transformation. In addition we examined polymorphisms in genes involved in DNA repair (RAD51 and XRCC2), carcinogen detoxification (HYL1) and methionine metabolism (MTHFR and MS). There was no difference in the distribution of the XRCC2, HYL1, MTHFR or MS polymorphisms in the control and AML cohorts studied. The RAD51 homologous recombination DNA repair gene polymorphism (135G/C-5′UTR) has been shown to result in enhanced promoter activity with consequent elevated mRNA expression and this polymorphism has previously been demonstrated to increase t-AML risk (OR=2.66, 95% CI 1.17–6.02). When combined analysis was performed on RAD51 (135G/C- 5′ UTR) and HLX1 (C/T -3′UTR) a 9.5-fold increase in the risk of t-AML was associated with the presence of variant alleles of both genes (OR= 9.50, 95% CI 2.22–40.64). The synergistic genetic interaction between the HLX1-C/T (3′- UTR) and RAD51-135 G/C polymorphisms demonstrates that target cell biology together with an increased homologous recombination DNA repair capacity, significantly increases the risk of t-AML. We suggest that this is because the HLX1-C/T (3′- UTR) polymorphism results in a higher number of stem cells; hence during genotoxic therapy (for a primary malignancy) there is an increased target cell number for genotoxic damage and potential malignant transformation. An optimum repair system is necessary to ‘deal’ with this burden of repair, however, the presence of altered repair capacity due to the RAD51-135C/T polymorphism may result in mis-repair and together with lack of apoptosis (due to a over-efficient repair system) will encourage the perpetuation of the oncogenic transformation and AML susceptibility.