Haema 2012; 3(1): 10-22
by Maria Ν. Pagoni and Ifigeneia A. Tzannou
Department: Department of Hematology and Stem Cell Transplantation Unit, Evaggelismos General Hospital, Athens, Greece
Abstract
The definition of prognostic factors is considered to be important for the selection of the optimal treatment modality and the prompt modification of treatment strategy. Two categories of prognostic factors can be established in CML. Those, that can be identified at the time of diagnosis (baseline) and those that can be assessed during treatment, and are usually relative to response. The most wide- ly used prognostic scoring systems are Sokal and Hasford scores. Even though both of these systems were established and validated before the introduction of TKIs, their predictive power has been proven in the TKI era. The EUTOS score has been recently developed for patients treated with imatinib, how- ever it still needs further evaluation in prospective clinical trials. Biological markers may be helpful in the timely recognision of those patients that are not going to respond to TKI therapy. Additional chromo- somal aberrations in the malignant clone, most importantly major-route ACAs, and gene profiling signature at the time of diagnosis may be useful in this way. Furthermore, the organic cation pump OCT-1 and P-glycoprotein expression, as well as phosphorylisation of Crkl may be predictive of the patient’s response to imatinib. However, they have not been incorporated in the clinical practice. Response to treatment in certain time milestones is the most important prognostic factor, and, among all prognostic factors, is the most powerful in predicting complete cytogenetic response (CCgR). The importance of the time and deapth of molecular response have also been evaluated. In a recent study, a single measurement of BCR-ABL transcript levels was found to be the best tool to identify patients with dismal prognosis, therefore allowing early intervention. BCR-ABL kinase domain mutations may result in suboptimal responses or failure to imatinib, and are associated with poor prognosis regarding event free and overall survival. Screening for mutations is mandatory upon first line treatment failure, in order to make the op- timal therapeutic choice. In regard with second generation TKIs, the data concerning factors predictive of responce are limited. BCR-ABL kinase domain mutations and maintenance of CCgR are concidered to be the most important prognostic parameters. The introduction of second generation TKIs in first line treatment of CML has resulted in better outcomes. As far as prognostic factors are concerned, achievement of CCgR and BCR-ABL transcript levels in certain timepoints remain crucial.