Molecular monitoring of residual disease in chronic myeloid leukemia by genomic DNA compared with conventional mRNA analysis

Abstract

Translocation t(9;22), which produces the BCR-ABL gene, is pathognomonic of chronic myeloid leukemia. For clinical purposes, the amount of chimeric transcript is considered proportional to the leukemic clone; thus, mRNA is commonly used for molecular monitoring of patients. However, there is no consensus regarding the degree of increase in mRNA that should cause concern or whether the absence of transcript indicates a "cure." In this study, we analyzed 57 samples from 10 chronic myeloid leukemia patients undergoing imatinib treatment. For each sample, we compared BCR-ABL mRNA levels with the actual proportion of leukemic cells, which were measured through a novel genomic approach based on the quantitative amplification of DNA breakpoints. The two approaches gave similar patterns of residual disease, and the majority of patients were still positive after an average treatment period of 2 years. Nevertheless, in one of two patients with confirmed undetectable levels of chimeric transcript, DNA still revealed the persistence of leukemic cells at 42 months. These findings appear to justify the clinical practice of maintaining imatinib treatment indefinitely. However, the absence of leukemic DNA (observed in 1 of 10 patients) could be used to identify possible candidates for drug discontinuation. In conclusion, DNA analysis proved to be a reliable index of residual disease with potential applications in the field of clinical diagnostics and research.

Figure 1

Molecular follow-up of patient 8. A: Comparison of DNA and mRNA results during the 42 months of IM therapy. UND, undetectable levels. B: The amplification of 300 ng of peripheral blood DNA extracted after 42 months of therapy still reveals disease eradication. C: The same amount of DNA from a healthy participant was positive for the control (BCR) but not for the BCR-ABL sequence. D: Amplification of 300 ng of DNA from patient 8, subsequently diluted 1:10 into the DNA from a healthy participant, demonstrates a sensitivity greater than the minimum value of 10⁻⁴ dilutions.