Impact of Baseline BCR-ABL Transcript Levels on the Response to Imatinib In CP CML Patients

Abstract 1240

Use of reliable predictable factors in newly diagnosed chronic phase (CP) CML patients could ensure the most appropriate selection of therapy in individual patients, and the Sokal score which was developed in the pre-interferon era still retains prognostic values in patients treated with imatinib. In addition to the Sokal score, patients with a high BCR-ABL transcript level at diagnosis might respond differently to therapy. However, to our knowledge, there has been no distinct study to investigate the association between BCR-ABL transcript level at diagnosis and the response to therapy. It has been generally agreed that baseline BCR-ABL transcript levels are not relevant to predict the response to therapy, and the use of international scale (IS) is also based on the concept in which 100%IS is defined as the standardized baseline and 0.1%IS corresponds to major molecular response (MMR) in all CML patients. In this study, cytogenetic and molecular responses to the therapy was investigated in patients with high BCR-ABL transcript levels at diagnosis and patients with low BCR-ABL transcript levels at diagnosis to assess if baseline BCR-ABL transcript levels could be the reliable predictable factor in CP CML patients. Between May 2001 and Aug 2008, total 756 CML patients were treated with imatinib in St. Mary's Hospital of the Catholic University of Korea, and among them, 113 CP patients have been treated with imatinib at a dose of 400 mg/day for more than 6 months without interferon or stem cell transplantation prior to imatinib therapy. BM or PB samples were obtained at regular intervals from diagnosis for hematologic response (HR), cytogenetic response (CyR) and molecular response (MR) monitoring. For additional statistical analysis, 102 patients with Sokal scores available at the time of diagnosis were divided into 3 groups: low, intermediate and high Sokal groups. The median baseline BCR-ABL transcript level for 113 patients was 103.79%(IS), and the median BCR-ABL transcript levels were compared between 57 patients in a group of high baseline BCR-ABL levels and 56 patients in a group with low BCR-ABL levels by 3, 6, 12 and 18 months of imatinib therapy (Table 1). No difference was observed when the median values at month 3 and month 6 were considered, but from month 12 to month 18 after initiation of imatinib, the median BCR-ABL levels in patients with high baseline BCR-ABL levels was at least twice as high as median BCR-ABL levels in patients with low baseline BCR-ABL levels. However, differences cannot be considered to be significant due to the small number of patients under analysis. Regarding the CCyR rate by 3, 6, 12 and 18 months of imatinib therapy, no significant difference was observed between 2 groups of high and low baseline BCR-ABL levels, and there was also no statistically significant difference in CCyR and MMR rates depending on the baseline BCR-ABL levels even after grouping the patients according to Sokal scores.

In this study, no evidence of the correlation of the baseline BCR-ABL levels and the response to therapy was observed, and BCR-ABL level at diagnosis does not seem to give additional prognostic information to predict the response to therapy. Thus, the value of baseline BCR-ABL levels is still questionable to be used to identify patients who require higher dose of imatinib or more potent tyrosine kinase inhibitors, and further study with larger cohort and longer follow-up would be necessary to confirm whether baseline BCR-ABL levels are relevant to predict the response to therapy.