Combination of EUTOS score and 3-months BCR-ABL transcript level identifies a distinct subgroup of ECP-CML patients with high risk of non optimal response and imatinib discontinuation.

Background: Response to TKI is considered the strongest predictor of long-term outcome in CML patients. Effective treatment overcomes the negative impact of most prognostic factors, including Sokal and Hasford scores and recent data demonstrated that an early molecular response is strictly related to the outcome. In fact, missing the 10% BCR-ABL landmark at 3 months predicts inferior long-term survival. The ‘EUTOS Score’ combines baseline spleen size and peripheral blood basophils to foresee the achievement of complete cyto- genetic remission (CCyR) and progression-free survival (PFS). Since its efficacy is still matter of debate, the predictive power of EUTOS in discriminating poor risk patients could be improved if associated with early molecular response assessment. Aims: We tested whether a combination of EUTOS score and 3-months BCR-ABL transcript level identifies a poor prognosis population of CML patients. Methods: 148 ECP CML patients treated with front-line standard dose imatinib (400 mg daily) at 5 major hematological centres in the north-eastern area of Italy were evaluated. Partial cytogenetic response (PCyR) and complete cytogenetic response (CCyR) were defined as 1–35% and 0% Ph+ metaphases, respectively; major molecular response (MMR) was defined as BCR-ABL <0.1%IS. For the purpose of the analysis patients were stratified into 3 subgroups according to EUTOS and 3-months molecular response (“good”: low EUTOS score and ≤10%IS BCR-ABL transcript level; “intermediate”: high EUTOS score or >10%IS BCR-ABL; “poor”: high EUTOS score and >10%IS BCR-ABL). TTF was measured from the start of imatinib to the date of any of the following events: progression to accelerated or blastic phase, death for any cause at any time, primary or secondary hematologic or cytogenetic resistance leading to imatinib discontinuation. PFS was measured from the start of imatinib to the date of pro- gression to accelerated or blastic phase or death for any cause at any time. Survival probabilities were estimated by the Kaplan-Meier method and compared by log rank test; differences among variables were evaluated by the Chi- squared test or by Cochran–Mantel–Haenszel test. Results: The median age was 55 years (range 19–84), with 80 males and 68 females. The median follow-up was 42.7 months (range 11–109). The median time from diagnosis to imatinib therapy was 0.9 months (range 0-7.9). The dis- tribution according to the EUTOS score was: 132 patients (89.2%) low risk; 16 patients (10.8%) high risk ; 3-months BCR-ABL transcript was ≤10% in 114 (79.2%) patients, >10% in 30 (20.8%). Patients with “good risk”, “intermediate risk” and “poor risk” profile were 72.1%, 22.1% and 4.7%, respectively. The “optimal response” endpoints to imatinib were: 6 months PCyR: 91.7% vs 76.7% vs 42.9%; 12 months CCyR: 91.8% vs 65.4% vs 28.6%; 18 months MMR: 76.7% vs 27.3 vs 25% (differences among groups were significant except for intermediate vs poor profile). Imatinib discontinuation rate for failure was significantly different between poor risk group compared with intermediate and low risk (71.4% vs 21.2% and 13%; P=0.008, P<0.0001 respectively), as well as PFS and TTF. Remarkably, all progressions to blast crisis were observed in “poor risk” group. Summary / Conclusion: Combination of High EUTOS score and 3-months BCR-ABL transcript level higher than 10% IS identifies a subpopulation of CML patients with significant probability of treatment failure and poorer outcome. Early switch to alternative treatment strategies (including allo-BMT) should be considered for these patients.