Which method better evaluates the molecular response in newly diagnosed chronic phase chronic myeloid leukemia patients with imatinib treatment, BCR-ABLIS or log reduction from the baseline level?

Abstract

The molecular response of chronic myeloid leukemia (CML) patients to tyrosine kinase inhibitor treatment can be evaluated either by BCR-ABL mRNA levels on international scale (IS) or by log reduction from the baseline level of the laboratory. Both methods were compared in 248 newly diagnosed chronic phase CML patients treated with imatinib. The major molecular responses (MMR) obtained by both methods predict progression-free survival (PFS, all P<0.0001). Thirty-six patients, who were identified as MMR patients by the IS method but as non-MMR patients by the log reduction method, had the same PFS as MMR patients identified by both methods. The molecular responses of patients at 3 and 6 months, as evaluated by the two methods, have similar predictive values on their cytogenetic responses at 12 months and on their molecular responses at 18 months. Both ≤ 10%(IS) and ≥ 1 log reduction at 3 months and ≤ 1%(IS) at 6 months were significantly associated with PFS (P=0.0011, 0.0090, and 0.0064). The percentages of patients with BCR-ABL(IS) of ≤ 1%, >1-10%, and of >10% at 3 months and 6 months in the German CML Study IV were similar with those with corresponding BCR-ABL(IS) in our center, but was significantly different with those evaluated by the log reduction method. Therefore, the molecular response evaluated by BCR-ABL(IS) has similar trends in PFS and in response prediction, but can better differentiate patients than that by the log reduction method. Furthermore, the IS method allows comparison among molecular response results from different laboratories.