International, prospective study comparing nilotinib versus imatinib with early switch to nilotinib to obtain sustained treatment-free remission (TFR) in patients with chronic myeloid leukemia: The tfr rate at the end of follow-up

Abstract Background Today, an important final endpoint in the management of chronic phase chronic myeloid leukemia (CP-CML) patients is the achievement of a sustained and deep molecular response (DMR) for subsequent discontinuation of therapy (treatment free remission - TFR). The selection of first-line TKI when the goal of therapy is TFR is still controversial, particularly in elderly patients with comorbidities in whom therapy with a more potent second-generation TKI could be associated with toxicities due to medium- or long-term off-target effects. The SUSTRENIM study is an international, prospective study comparing not only the rate of DMR but, more importantly, the rate of TFR in newly diagnosed CML patient randomized to receive NIL vs IM frontline followed by systematic switching to NIL in case of non-optimal response (Clinical Trial number 02602314). Methods Patients are stratified at randomization according to the Sokal risk score (high versus intermediate/low risk) and country. All the patients who obtained a reduction greater than 4.0 logs of residual disease (MR4.0) within the first three years of treatment and maintained or improved this level of response in all the subsequent tests up to the end of the fourth years of therapy qualified for treatment discontinuation phase of the study. In the present analysis we compared, in the two arms of the trial, the rate of patients eligible to discontinuation after at 48 months of therapy and of those who showed a successful TFR after 12 months after TKI withdrawal. In addition, the rate of TFR was also analyzed for the patient features (i.e. age, gender, ELTS and Sokal risk, BCR::ABL1 isoform) using Fisher's exact test or Wilcoxon according to variable type. Results Overall, 448 patients consecutively enrolled in the study were randomized to the NIL (N = 228) and IM (N = 220) arms of the study. Of these, 258 patients (58%) were male and 190 (42%) were female, and the median age was 54 years (range, 19-86) with 20% and 28% aged >65 years in the NIL and IM arms, respectively. Overall, the percentage of high-risk patients was 16% according to the Sokal score and 9% for the ELTS score. Of the 448 randomized patients, 151 were eligible for TFR entry after the first 48 mo. of observation (median 58 mo.) at the time of the present analysis and were evenly distributed between the two arms of the trial (77 in the IM and 74 in the NIL arm, respectively). After 6 and 12 mo. of follow-up, 72.1% (95% CI: 64.8-80.3) and 65.2% (95% CI: 57.3-74.1) of these patients were still in TFR. Interestingly, multivariate analysis showed that only two independent factors predict successful TFR: treatment arms (IM vs NIL) and ELTS score (Low vs Int/High) but not the Sokal risk nor age or others. Consistently, 75.0% (95% CI: 65.1-86.5) of patients randomized in the NIL arm were in TFR at 12 mo. vs 55.2% (95% CI: 44.0-69.3) of those in the IMA arm (p<0.02) at K-M analysis. Still more impressive the differences between the TFR rate at 12 mo. between the ELTS low risk patients of the two arms of the study. Indeed, 67.5% of ELTS low patients of the IMA arm vs 86.1% of the NIL arm remained in TFR at 6 months and 58.6% vs 84.0% at 12 months of follow-up (p=0.0041). On the other hands, no differences were observed between the two arms of treatments when we selected only ELTS low/int patients. Noteworthy, severe non-cardiovascular events were rather infrequent and well balanced between the two arms of the study while either vascular and cardiac events were more frequently observed in patients randomized to NIL arm. However, we observed only two cardiac-related deaths, one in the patient treated with NIL and the other in the patient treated with IM. Conclusions This study demonstrates that, if the final goal is TFR an aggressive therapeutic strategy can be pursued in all patients. First-line therapy with second-generation TKIs is feasible and more effective to achieve a deep and early molecular response and, consequently, a higher TFR rate compared to IM despite the optimization with early switch to NIL in case of suboptimal response. This appears to be particularly true in case of low ELTS risk patients, regardless of their age, gender or other baseline features. On the other hand, in patients at intermediate or high ELTS risk, the benefit of NIL frontline is negligible if the aim of the therapy is TFR achievement.