THE EUTOS LONG-TERM SURVIVAL SCORE IS PREDICTIVE FOR RESPONSE AND OUTCOME IN CML PATIENTS TREATED FRONTLINE WITH NILOTINIB-BASED REGIMENS

Background: The outcome of BCR-ABL+ chronic myeloid leukemia (CML) has been significantly improved by the introduction of tyrosine kinase inhibitors (TKIs) and, actually, many patients die of reasons unrelated to CML. A new scoring system, the EUTOS long-term survival (ELTS) score, based on the analysis of a large cohort of CML patients treated frontline with IM and aimed to discriminate the probability of dying of CML, has been recently published by the European LeukemiaNet (ELN). The relevant variables included were: age, peripheral blasts, spleen size and platelet count (Pfirrmann et al. Leukemia 16). Nilotinib (NIL) is a 2nd generation TKI approved as frontline therapy of CML in many countries. Aims: To investigate the prognostic value of the ELTS score in a cohort of CML patients in early chronic phase treated with NIL-based regimens as firstline therapy. Methods: Three hundred and forty-five adult patients were included. The patients were enrolled in three multicenter studies (ClinicalTrials.gov NCT00481052, NCT00769327 and NCT01535391) conducted by the GIMEMA CML WP, or treated at the University Hospital of Bologna, Italy. The initial treatment was NIL 300 mg BID or NIL 400 mg BID. The intentionto- treat population of each study was analyzed. Definitions: risk scores were defined according to Sokal, EUTOS and ELTs formulations; major molecular response (MMR): BCR-ABL1IS ratio <0.1%; MR4.0: BCR-ABL1IS ratio <0.01% with >10,000 ABL1 copies; progression: transformation to advanced phases according to ELN criteria; leukemia-related death (LRD): death after progression. Results: The median age was 53 years (range 18-86). The patient distribution according the different scoring systems was as follows: 41% low, 40% intermediate and 19% high Sokal score, 94% low and 6% high EUTOS score, 59% low, 30% intermediate and 10% high ELTS score, respectively. The median follow-up was 59 months (range: 24-82 months). The cumulative incidence of MMR and MR4.0 was 83% and 69%, respectively; the 6-year progressionfree survival (PFS), overall survival (OS) and cumulative incidence of LRD were 91%, 92% and 4%, respectively. All the 3 scores were associated with significantly different probabilities of MMR (cumulative incidence of MMR according to ELTS score: 90%, 77% and 61% in low, intermediate and high risk patients, respectively; p<0.001), but only the ELTS score was able to predict the achievement of MR4.0 (cumulative incidence of MR4.0: 75%, 63% and 53% in low, intermediate and high ELTS score patients, respectively; p=0.013). Interestingly, both the Sokal score and the ELTS score, but not the EUTOS score, predicted the OS (p=0.021 and p=0.037, respectively), while only the ELTS score predicted a significantly different PFS (p=0.038); the 6-year cumulative incidence of progression was 2%, 7% and 6% in patients with low, intermediate and high ELTS score, respectively. Probably due to the low number of deaths related to CML, all the 3 scores failed to predict LRD, but comparing patients with low ELTS score versus patients with intermediate and high ELTS risk the difference became significant. Summary/Conclusions: In a cohort of CML patients treated with NIL-based regimens as frontline therapy, the prognostic predictive ability of ELTS score resulted superior to Sokal and EUTOS score.