Very Early Molecular Responses During The First Two Months Of Therapy Are Highly Predicitive Of Deep Molecular Responses In Newly-Diagnosed Chronic Myeloid Leukemia In Chronic Phase( CML-CP) Patients Treated Upfront With Nilotinib. The Spanish Substudy Of The ENEST1st Trial

Objectives To analyze the molecular response during the first trimester of nilotinib therapy in newly diagnosed CML-CP patients. Hypothesis The values of BCR-ABL ratios during the first trimester of nilotinib treatment, and the kinetic of their descent, could be predictive of molecular response thereafter. Patients ENEST1st ([NCT01061177][1]) is an open-label study of nilotinib 300 mg twice daily in adults with newly diagnosed BCR-ABL+ CML-CP. Imatinib pretreatment was not allowed. Methods BCR-ABLIS and BCR-ABL/GUS ratios were measured previously to nilotinib, and fortnightly thereafter until the 3rd month (m), and at 6, 12 and 18 m. Sokal, Euro and Eutos scores were calculated with data at diagnosis. BCR-ABL values were centrally measured in an ELN-EUTOS certified laboratory. Molecular response was classified by ELN2013 recommendations. As the linearity of values of BCR-ABL using ABL as control is questionable, when ratios are higher than 10%, only baseline BCR-ABL/GUS ratios were used when analyzing the molecular response using GUS as control. The kinetic of the descent was calculated using the ratio of a given time compared with that of an earlier time, and measuring slopes. Logistic regressions and ROC analysis have been used, calculating positive and negative predictive values (PPV and NPV) Results 61 patients were included. 1 patient was excluded of the analysis because of lack of molecular data (baseline). Out of 60 patients, 10 abandoned during the first 18 m because of AE’ s. Those patients have been classified as non-responders after the time they went off-study. Risk distribution: Sokal (L, I, H): 57%, 32%, 11.7% Euro: 52%, 45%, 3%) Eutos (L, H): 92%, 8%). Outcomes and Molecular response: No patient died or transformed during the follow-up. MR4.5 at 18 M has been obtained in 30% of the patients. The ELN 2013 molecular milestones for optimal response at 3,6,12 and thereafter were obtained in 97%, 93%, 83%, and 70%, respectively.([Table][2]) View this table: Predictive variables of response: ([Table][2]). Major molecular response (²0.1%) (BCR-ABL/ABL, BCR-ABL/GUS). At 3MBaseline BCR-ABL/GUS ratios were significantly different between responders and non-responders (22,1±23,1 vs 41,5± 38,1, p=0.05). For both control genes, the multivariate analysis disclosed that the independent and significant variable was the ratio at 45d. The ROC analysis disclosed a cut-off of 3.28 (PPV: 83% NPV: 80%). [OR: 20(4.4- 90) p<0.00001]. For BCR-ABL/GUS it was 3.5 (PPV: 87% NPV: 73%). At 6M and at 12 M: For both control genes, only the ratio at 3M and the ratio at 6 M were independently associated, respectively. At 18 M; With BCR-ABL/ABL, only the ratio at 12M was the independent and significant variable. Variables associated with MR4.5 at 18 m (BCR-ABL/ABL). The ratios at 1M, 1.5 M, and 2 M were significantly associated with the MR4.5 at 18 M. In the multivariant analysis the only independent variable associated with this response was the BCR-ABL ratio at 2M. The ROC curve disclosed a cut-off of 1.52 (PPV: 91% NPV: 59%). When including this variable as dichotomic, in a multivariant analysis, was the only significant one [OR: 14(3.2-60) P=0.0004)]. The correspondent cut-off at 3M was 0.1 (PPV: 84% NPV: 60%); [OR: 7,7(2.2-27) P=0.001]. Discussion Our results show that, in newly diagnosed patients, nilotinib obtained responses very quickly, and the proportions of patients having an optimal ELN2013 response at 3M (BCR-ABLIS² 10%), and at 6M (BCR-ABL IS ² 1%) were 97% and 93%. For obtaining a MMR at 3M, the only significant cut-off was at 45 d. For subsequent MMR, the only independent variable was the ratio at the most immediate earlier point. Besides, our results show a similar proportion of responses when using GUS as control gene, and similar predictive cut-offs in the ROC analysis of MMR at 3 and 6 M. Nilotinib treatment obtained 30% of MR4.5 at 18M. Ratios at 1M, 1.5 M, and 2 M were significantly associated with this response, but the only independent variable was the BCR-ABL ratio at 2M. At 2 M, having a ratio higher than 1.52 % will be linked with a 91% probability of not obtaining a MR4.5 at 18 M. Disclosures: Steegmann: Pfizer: Honoraria, Research Funding, Speakers Bureau; BMS: Honoraria, Research Funding, Speakers Bureau; Novartis: Honoraria, Research Funding, Speakers Bureau. Casado Montero: Pfizer: Honoraria, Research Funding, Speakers Bureau; BMS: Honoraria, Research Funding, Speakers Bureau; Novartis: Honoraria, Research Funding, Speakers Bureau. Echeveste: Celgene: Consultancy; Novartis: Consultancy. Garcia-Gutierrez: Novartis: Membership on an entity’s Board of Directors or advisory committees; BMS: Membership on an entity’s Board of Directors or advisory committees; Pfizer: Membership on an entity’s Board of Directors or advisory committees. Ruiz: Novartis: Employment. Walasek: Novartis: Employment. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01061177&atom=%2Fbloodjournal%2F122%2F21%2F5190.atom [2]: #T1