Allogeneic Stem Cell Transplantation Versus B-Cell-Receptor Inhibitors in 17p Deletion and/or Refractory Chronic Lymphocytic Leukemia: A Retrospective Comparative Analysis of 'Real Life' Approaches to High Risk Patients, on Behalf of Rete Ematologica Lombarda (REL) and Gruppo Italiano Trapianto Di Midollo Osseo (GITMO)

Introduction: chronic lymphocytic leukemia (CLL) is an indolent disease, but 17p deleted (del) and/or patients who experience early relapse or not responding to chemoimmunotherapy have a very poor outcome. In these patients an allogeneic stem cell transplantation (alloSCT) is indicated, whenever possible. B-cell receptor inhibitors (BCRi) have shown high efficacy with a low toxicity, making the choice of an alloSCT challenging even in high risk patients. The aim of the study is to highlight the outcome of different clinical approaches available in the era of the new drugs.

Method: this is a multicenter retrospective analysis on 88 high risk patients treated in 8 Italian Centers. Inclusion criteria were: i) age ≤ 70 years; ii) responding to one of the EBMT (European Bone Marrow Transplantation) criteria for elegibility to alloSCT: relapse within one year after purine-analogues or two years after chemioimmunoterapy or autologous-SCT (ASCT); de novo or acquired 17p del; iii) alloSCT from 2001. Patients who received BCRi before alloSCT or ineligible to alloSCT due to comorbidities were excluded. Patients were assigned to alloSCT or BCRi based on physician and/or patient choice or unavailability of a donor. The analysis started from the transplant date or the start date of BCRi therapy.

Results: 50 patients (M/F: 42/8) received an alloSCT, and 38 (M/F 30/8) were treated with BCRi (ibrutinib n=28, rituximab-idelalisib n=10). Median age was 55 (range, 34-68) in alloSCT and 60 years (42-69) in BCRi (p=0.06). Time from diagnosis to alloSCT or BCRi was 59,4 (range, 5-210) and 86,3 months (1-211) (p=0.15), respectively. Fluorescence in situ hybridization (FISH) data were available in 34/50 (68%) alloSCT patients: 17p del was positive in 21 (62%) (de novo n=12, acquired n=8, unknown n=1). Elegibility criteria for alloSCT group were: early relapse n=11, refractory n= 7, ASCT relapse n=11, 17p del n=21 (of which 11 were also chemorefractory). FISH data were available in all BCRi patients: 17p del was positive in 26 (68%) (de novo n=8, acquired n=14, unknown n=4). Elegibility criteria for BCRi group were: early relapse n=2, refractory n= 9, ASCT relapse n=1, 17p del n=26 (of which 16 were also chemorefractory). AlloSCT group had more patients in early relapse or with a failed ASCT (p=0.03 and p=0.01). Heavy chain gene rearrangement was available in 26/50 (52%) alloSCT and 35/38 (92%) BCRi patients and was unmutated in 87% and 86%, respectively. The median number of previous therapy was 2 in both groups (alloSCT: range 1-7; BCRi: range 0-8, p=0.25). Reduced-intensity conditioning regimen was used in 48/50 alloSCT patients, and donor type was sibling in 19, matched unrelated in 26 and haploidentical in 5 cases. Disease status before alloSCT was complete remission (CR)=20, partial remission (PR)=17, stable/progressive disease (SD/PD)=13. The median follow-up was 33 months (1-134) and 14 months (3-32), respectively for alloSCT and BCRi group (p=0.0008). Two-year OS was 59% vs 79% in alloSCT and BCRi, respectively (p=0.32). The cumulative incidence of relapse at 2 years was 30% in alloSCT and 23% in BCRi, with a non-relapse mortality of 20% after alloSCT. Median PFS was 18,6 months (range 1-134) and 12,6 (range 3-24,6) in alloSCT and BCRi, respectively. Two-year PFS was 54% in alloSCT and 77% in BCRi (p=0.19). The main cause of treatment failure was disease progression. In the BCRi group, 8 patients achieved CR at the last follow-up, 5 patients relapsed and died of progressive disease (2 with Richter's transformation), 1 patient died of second cancer. In the alloSCT group, 17 patients were alive and in CR, 11 CR patients died of transplant-related-mortality (6 graft-versus-host disease, 2 infections, 1 embolism, 2 second cancers), 19 patients died in SD/PD at the last follow-up.

Conclusions: these retrospective data showed that so far no significant difference in the outcome of 17p del and/or refractory CLL patients have been observed after either alloSCT or BCR inhibitors. The significant different follow-up of the two groups implies some limits: i) BCR inhibitors still have to show long term responses; ii) alloSCT results may improve over the next future by a better selection of eligible patients and the improvement of the transplant procedures. Hopefully, the combination of the two strategies will increase the chance of cure of poor risk CLL patients.