Abstract
Allogeneic hematopoietic stem cell transplantation (HSCT) is a potentially curative treatment for chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) and is performed in patients (pts) with high-risk features at diagnosis (del17p/p53 mutations) or advanced disease. Although approximately 30% of pts have matched siblings, alternative stem cell sources such as umbilical cord blood, extend the use of HSCT to pts lacking a conventional donor. However, little is known about outcomes after umbilical cord blood transplantation (UCBT) for CLL/SLL.
We analyzed 68 pts (50 males) who underwent a single (n=16) or double (n=52) HLA-mismatched UCBT between 2004 and 2012 in 34 EBMT centers. Median age at UCBT was 57 years (yrs) (range, 27-68). At diagnosis, 56 pts had CLL/SLL, 8 prolymphocytic leukemia (4 B-cell and 4 T-cell) and 4 Richter transformations. Cytogenetic was available in 70% (48/68) of pts. Seventy-three percent had abnormal karyotype (36% del17p/p53, 17% del13q, 8% del11q and 12% others) and 27% normal. Median time from diagnosis to UCBT was 54 months (range, 3-358). Thirty-seven pts were in partial remission (PR) at UCBT, 23 in complete remission (CR) and 8 had refractory disease (RD). The hematopoietic stem cell comordibity index (HCT-CI) at UCBT was 0 in 60% of pts, 1-2 in 25% and 3-4 in 15%. Sixty-five percent of pts received ≥3 chemotherapy lines prior to UCBT, 28% were refractory to purine analogues and 16% underwent previous autologous stem cell transplantation (ASCT). Sixty patients received low-dose TBI (2-4 Gy) from which the Minnesota RIC regimen (TBI-Cyclophosphamide-Fludarabine: TCF) was given to 57 pts; 15 pts received ATG and GVHD prophylaxis was CsA+MMF in 61 pts. Median TNC collected was 3.7x107/Kg (1.8-7.1) for single and 5x107/Kg (2.0-9.7) for double UCBT. Units were HLA matched to the recipient at 5-6 loci in 30% of pts and at ≤4 in 70%.
Median follow-up was 37 months (range, 3-98). OS and PFS at 3 yrs were, respectively, 53±7% and 45±7%. The cumulative incidences (CI) of neutrophils and platelets engraftment were 84±5% and 72±6%, respectively with a median time for engraftment of 21(range, 5-67) and 43 (range, 1-189) days, respectively. CI of acute graft-versus-host disease (aGVHD) at 100 days was 43%±6 for grade II-IV and 19±5% for grade III-IV with a median time of onset of 23 days (9-95). Three yrs CI of chronic GVHD (cGVHD) was 32±6% (12 limited and 6 extensive) with a median time of onset of 130 (range, 101-393) days. CI of relapse and NRM at 3 yrs were,16±5% and 39±6%, respectively.
In a univariate analysis the use of TCF conditioning was associated with improved OS (62% vs 15%, p<0.001), PFS (52% vs 15%, p<0.001) and lower NRM (34% vs 66%, p=0.008). CMV negative serology prior UCBT was associated with higher OS (71% vs 41%, p=0.02). Fludarabine sensitive disease at transplantation was associated to improved OS (70 vs 21%, p<0,001), PFS (57% vs 21%, p=0.001 – figure 1) and lower NRM (30% vs 58%, p=0.009). Patients who received ATG had lower incidence of aGVHD grades II-IV (14% vs 45%, p=0.02). The HCT-CI ≤2 was associated with improved PFS (51 vs 25%, p=0.04). Patients who received ASCT prior to UCBT had higher rates of engraftment (91% vs 82%, p=0.03). No risk factors were associated with cGVHD or relapse. Twelve patients relapsed or progressed after UCBT, of which 8 were transplanted in PR or had RD at time of transplant. Median time to relapse or progression was 19 months (2-89).
The use of TCF (HR: 0.34 (0.14-0.82), p=0.02), fludarabine sensitive disease at transplantation (HR: 0.41 (0.20-0.81), p=0.01) and HCT-CI ≤2 (HR: 0.38 (0.17-0.87), p=0.02) were associated with improved PFS in multivariate analysis. Acute GVHD grade III-IV was associated with lower OS (HR: 3.4 (1.6-7.4), p=0.002) and PFS (HR: 2.8 (1.4-5.8), p=0.005) in a time-dependent model. Overall, 32 patients died; 7 died of relapse and 25 of transplant related causes (10 infections, 4 post-transplant lymphoprolypherative diseases, 3 aGVHD, 3 toxicities, 1 secondary lung cancer and 4 other causes).
In conclusion, RIC-UCBT appears to be a valid option for CLL/SLL patients. Strategies to optimize infection prophylaxis, use of low-dose TBI RIC conditionings and to perform the UCBT before development of fludarabine resistance may improve overall outcome.
No relevant conflicts of interest to declare.
