AML and high risk MDS
In the early 2000s a series of reports established the efficacy of non-myeloablative (NMA) fludarabine and melphalan (FluMel) based stem cell (HCT) conditioning regimens. The MD Anderson group assessed the utility of FluMel HCT regimens in older patients with associated co-morbidities. Giralt et al compared the outcomes of 86 patients with a variety of high risk haematological malignancies randomised to receive either FluMel 180 mg/m2 (n=66) or dose reduced for renal impairment or extensive prior therapy to 140 mg/m2 (n=12) or cladribine and melphalan (CdaMel) 180 mg/m2, (n=8).r Only 7 patients were in first complete remission (CR) at time of transplant. Early treatment related mortality (TRM) was reported as 37.5% in FluMel versus 87.5% in CdaMel. 10 grade 3-4 toxicities and 3 toxic deaths were reported in the FluMel arm. 2-year disease free survival (DFS) and overall survival (OS) probabilities were 0.28 (95% CI 0.20-0.39) and 0.23 (95% CI 0.15-0.34, respectively.
Subsequent studies showed similar outcomes between higher (180 mg/m2) and lower (140 mg/m2) doses of melphalan. de Lima reported a retrospective analysis of fludarabine/cytarabine/idarubicin (n=32) versus FluMel 180 mg/m2 or 140 mg/m2 (n=62) in acute myeloid leukaemia (AML) or high risk myelodysplastic syndrome (MDS).r Outcomes were reported as similar for the two doses of melphalan. Despite lower rates of CR at time of HCT (44% vs 16%, p=0.006) in the FluMel cohort, this combination offered superior disease control with superior engraftment and lower relapse related mortality (p=0.029). However, FluMel was associated with greater TRM (p=0.036) and higher rates of grade 3-4 toxicity (p=0.002) with similar rates of OS. Oran et al reported follow up data on 112 patients, median age 55 (22-74) receiving FluMel 100 mg/m2 (n=13), 140 mg/m2 (n=46) or 180 mg/m2 (n=53) for AML or high risk MDS.r 47.3% had an unrelated donor (MUD). Graft versus host disease (GVHD) prophylaxis was tacrolimus and methotrexate (MTX) and antithymocyte globulin (ATG) in those receiving a MUD. The 2-year incidences of disease progression and OS were 25% and 44% with a cumulative incidence of NRM of 54% for the whole cohort. Survival outcomes were similar with Mel-140 versus Mel-180 but with higher nonlethal toxicity reported with Mel 180.
Follow on studies confirmed the efficacy of FluMel Conditioning with in AML and MDSrrr and some reported on long term outcomes. Popat et al reported long term outcomes of 36 patients with AML in CR, who received FluMel (58% Matched sibling donor (MSD) and 42% MUD).r With a median follow-up of 52 months (range, 34-103), 4-year OS and progression free survival (PFS) were 71% and 68%, respectively. At 4 years, the cumulative incidence of TRM was 20% and relapse mortality was 8%. Neither OS nor PFS were affected by older age (>60 years), unrelated donor, melphalan dose, or comorbidity score. At last follow up, of the 24 surviving patients, 88% had performance status (ECOG) of 0 without any active chronic GVHD requiring steroids. This study indicated that FluMel yielded durable long-term remission in older patients. The Australasian Bone Marrow Transplant Recipient Registry (ABMTRR) also reported long-term outcomes in a large retrospective study assessing the outcome of FluMel between 1998 and 2008 with a median follow-up of 3.4 years.r 344 patients with a median age of 54 years (range, 18-68) were included, 234 patients had myeloid malignancies, with AML (n=166) the commonest indication. TRM at day 100 was 14% with no significant difference between the groups. 3-year OS and DFS were similar between myeloid and lymphoid patients (57 and 50%, respectively). There was no difference in cumulative incidence of relapse or GVHD between groups. Multivariate analysis revealed four significant adverse risk factors for DFS: donor other than MSD, not in remission at HCT, previous autologous transplant and recipient CMV positive. Chronic GVHD was associated with improved DFS in multivariate analysis predominantly due to a marked reduction in relapse (HR 0.44, P=0.003). This study confirmed that FluMel yielded durable long-term remissions in both myeloid and lymphoid malignancies.
The only comparative study of fludarabine and busulfan FluBu versus Flu Mel in AML was reported by the Acute Leukaemia Working Party of the European Group for Blood and Marrow Transplantation (9). Baron et al reported outcomes in a cohort of 394 AML patients with MSD comparing FluBu (n = 218) and FluMel (n = 176).r Patients given manipulated grafts or T-cell-depleting agents were not excluded. All patients were in CR (68% in CR1, 18% in CR2 or later and 15% had advanced disease). The incidences of acute and chronic GVHD were similar in the 2 groups. The 2-year RR, TRM, DFS, and OS were 31% (+/-3%), 18% (+/-3%), 51% (+/-4%), and 54% (+/-4%), respectively, for FluBu patients and 20% (+/-3% P=0.007), 20% (+/-3% P = 0.4), 60% (+/-4% P=0.08), and 62% (+/-4% P=0.2), respectively, for FluMel. In multivariate analyses, FluMel was associated with lower relapse (hazard ratio [HR], 0.5; P =0.01) and a trend toward higher TRM (HR, 1.6; P= 0.1) with similar DFS (HR, 0.8; P=0.2) and OS (HR, 0.9; P=0.6). The results indicated that although FluMel provides better AML control than FluBu as a NMA regimen, the 2 regimens provide similar survival.
Initial reports on the use of Flu-Mel in myelofibrosis (MF) were small case series.rrr Later the Myeloproliferative Diseases–Research Consortium reported a prospective study on outcomes of FluMel conditioning in 66 patients with primary or secondary myelofibrosis with MSD versus MUD.r Patients with MUDs received ATG. Patient characteristics in the 2 cohorts were similar. Engraftment occurred in 97% with MSD and 76% with MUD and secondary graft failure occurred in 3% and 12%, respectively. With a median follow-up of 25 months the OS was 75% in the MSD (median not reached) and 32% in the MUD group (median OS: 6 months, 95% CI: 3, 25) (HR 3.9, 95% CI: 1.8,8.9) (P < .001). TRM was 22% in MSD and 59% in MUD. Survival correlated with type of donor, but not with the degree of histocompatibility match, age, or JAK2(V617F) status. It appeared that in patients with MSD FluMel was effective, but in those with MUD it led to a high rate of graft failure and limited survival. Robin et al reported a study comparing FluBu (n = 105) and FluMel (n = 55) NMA conditioning in 160 patients with MF.r Using complex statistical modelling weighting and adjusting to allow comparison between these 2 groups, the incidence of a GVHD was 62% with FluMel and 31% with FluBu (P = .001), and the corresponding incidences of chronic GVHD was 49% and 53%, respectively. The 7-year PFS was 52% with FluMel vs 33% with FluBu, and the 7-year OS was 52% vs 59% respectively. TRM was 43% with FluMel vs 31% with FluBu. Multivariable analyses revealed no significant differences in PFS between the 2 groups; however, the relapse rate was significantly lower with FluMel (HR, 9.21; P = 0.008), whereas a trend toward reduced TRM was seen with FluBu (HR, 0.51; P= 0.068). Both regimens appear to be effective in disease control and while FluMel had a higher TRM, this lead to comparable OS due to augmented disease control.
Acute lymphoblastic leukaemia
Eom et al compared long-term outcomes between NMA and myeloablative (MA) conditioning regimens in adult high-risk acute lymphoblastic leukaemia (ALL) patients in CR.r This study compared long-term outcomes of 60 consecutive NMA FluMel HCTs with 120 MA (total body irradiation (TBI) plus cyclophosphamide) HCT. The FluMel group had older patients (46 years vs. 33 years, P < 0.001) and a higher proportion of peripheral blood transplants (PBSC) (93.3% vs. 13.3%; P < 0.001) but otherwise the groups were similar. After a median follow-up of 67 months, in multivariate analysis FluMel had a comparable TRM (21.2% vs. 24.3%) and DFS (50.8% vs. 54.9%) to MA transplants, but RR was higher (34.2% vs. 26.4%; HR, 2.07; P=0.019). The negative impact of FluMel on RR was seen only for Philadelphia-positive ALL (32.7% vs. 19.6%; HR, 3.46; P=0.020), with no difference in Philadelphia-negative ALL.). Raida et al recently reported outcome in 13 patients with high-risk ALL in CR1 receiving NMA conditioning with fludarabine (150 mg/m2) and melphalan (140 mg/m2) with ATG (4.5 mg/kg).r TRM was compensated by early withdrawal of prophylactic immunosuppression aimed at maintaining effective immunological antileukaemic control. The median post-HCT follow-up was 23 months (range10-65). Acute GVHD was observed in 15.4% and chronic GVHD was not noted. 15.4% relapsed. The 2-year DFS was 76.9% (95% CI 51-100%). Although these are small studies FluMel appears to be a feasible NMA conditioning regimen in ALL.
Non Hodgkin lymphoma
Rodriguez et al reported a retrospective analysis comparing MA conditioning with NMA FLuMel regimens in non-Hodgkin's lymphoma.r The relapse rate was significantly lower in the MA group compared to the FluMel group (13% versus 28%; P=0.05). There were no significant differences in TRM, OS and PFS. Hertzberg et al reported on outcomes of NMA conditioning in 36 patients with various chronic B-lymphoproliferative disorders: follicular lymphoma (n=17), mantle cell (n=9), small lymphocytic lymphoma (n=2), and chronic lymphocytic leukaemia (n=8).r The median age was 51 years (range, 30-66) and time from diagnosis was 3.4 years (range, 0.3-9.5). 28% were in CR, 36% were in PR and 36% were chemorefractory. Conditioning included fludarabine and either cyclophosphamide (n=27) or melphalan (n=9). TRM at d100 was 8%, at 1 year 17.5% and 2 years 20%. The cumulative incidence of grade II-IV acute GVHD was 58% and of limited and extensive chronic GVHD was 25 and 56%, respectively. At a median follow-up of 48 months no patient relapsed or progressed and OS was 80% (95% CI, 67-93%). In the ABMTRR study survival was similar in both myeloid and lymphoid malignancies (8). 3-year OS and DFS in lymphoid malignancies were 59% and 48%, respectively. Death was due to relapse in 27% and progressive disease in 5%, infection in 25%, GVHD in 17% and organ toxicity in 13%.
Diffuse large B-cell lymphoma (DLBCL): Thomson et al reported on the efficacy of FluMel with alemtuzumab in 48 patients with relapsed/refractory DLBCL.r4-year TRM was 32% and RR 33% with PFS and OS rates of 48% and 47% respectively and 55% and 54% in those with chemosensitive disease, respectively. In the ABMTRR study patients with DLBCL only had a 3-year OS of 33%.r
Follicular lymphoma (FL): Pinana et al reported the results of a prospective multicentre GELTAMO trial of FluMel conditioning in 37 patients with FL.r 25/37 patients received FluMel after a median of three prior therapies. 4-year OS for patients in CR at HCT was 71%, with low RR but high TRM and poor OS for those with progressive disease. Ono et al retrospectively evaluated outcomes of FluMel in FL in 19 patients with relapsed or refractory FL. Both the 5-year OS and PFS were 84.2% (95% CI: 67.7-100%).r In the ABMTRR study patients with CLL or follicular lymphoma had a 3-year OS of 64%.r
Mantle cell lymphoma (MCL): Cruz et al reported long-term outcomes of NMA FluMel MSD PBSC HCT in 21 patients with MCL in CR.r Early toxicity included mild/moderate mucositis (43%), febrile neutropenia (33%) and bacterial infections (19%). With a median follow up of 48 months, TRM was 19.5%. Grade III-IV acute GVHD occurred in 15% and chronic GVHD in 78% (extensive in 39%). The 5-year PFS and OS was 80% (95% CI: 63-97%) for both. Age was the only prognostic factor for OS, which was 43% for those > 60 years and 100% for those younger (p < 0.001).
T-cell lymphoma (TCL): Duvic et al reported the results of conditioning using total electron beam radiotherapy with FluMel in 16 patients with advanced cutaneous TCL.r Overall intent to treat response was 68% with CR rate of 58%. Follow up was short but 11/13 patients were in CR at a median of 19 months. Delioukina et al reported a single centre retrospective analysis of 27 patients who received NMA FluMel for mature TCL.r The majority of patients had advanced disease. At a median follow-up of 36 months the 2-year OS was 55%, PFS was 47%, RR was 30% and TRM was 22%. Cutaneous TCL patients had a 2-year PFS of 45% and TRM of 27% compared with patients with other histologies, who had a PFS of 62% and NRM of 19%.
Alvarez et al reported the efficacy to NMA FluMel in refractory or relapsed Hodgkin lymphoma (HL) in a prospective Spanish cooperative study, which included 40 patients (50% in resistant relapse).r TRM at d100 and 1 year were 12% and 25%. 2-year OS and PFS were 48% (+/-10%) and 32% (+/-10%), respectively. Chemotherapy refractoriness was the only adverse prognostic factor for OS. In patients with failure of a prior autologous stem cell transplantation, results were especially good for those who experienced late relapses (=12 months: 2-year OS and PFS were 75%(+/-16%) and 70%(+/-18%), respectively). Anderlini et al reported outcomes in 58 patients with HL (refractory relapse (n=28) or sensitive relapse (n=30).r Projected 2-year OS, PFS and RR were 64% (49-76%) and 32% (20-45%) and 55% (43-70%), respectively. Chen et al also confirmed responses in their single centre retrospective study of 24 heavily pretreated patients with relapsed HL. At a median follow-up of 39 months the 2-years OS was 60% (95% CI 42, 72) and PFS was 27% (95% CI 22, 32), TRM was 13.1% (95% CI 5.1, 31.4).r
The UK and Spanish Collaborative Group reported improved PFS in a similar cohort of poor risk HL in those receiving CsA/MTX prophylaxis together with alemtuzumab with durable responses to donor lymphocyte infusion (DLI) contributing to an improved outcome.r Chen et al reported on the addition of brentuximab vedotin prior to NMA FluMel transplantation in a retrospective study with a historical control. After a median follow-up of 29.9 months the addition of brentuximab was associated with improvements in 2-year PFS (59.3% versus 26.1%) and relapse/progression (23.8% versus 56.5%).
Initially results in patients with multiple myeloma (MM) were not positive. Giralt et al reported poor 2-year OS 30% (+/-11%) and EFS 19(+/- 9%) with a d100 TRM of 19(+/-10%) in a group of 22 patients, 60% of whom had refractory disease at HCT using FluMel-140 (n=18) or FluMel-180 (n=4).r The European Group for Blood and Marrow Transplantation reported the outcomes of FluMel in 49 patients with MUD for relapsed MM post autograft.r 5-year PFS and OS were 20% and 26%. Outcomes were significantly poorer in patients with HLA mismatched recipients with a 1-year TRM of 53% versus 10% for those receiving a MUD (p= 0.001).
A number of studies reported longer term outcomes with FluMel in MM. Kikuchi et al reported a retrospective single centre study of 23 patients with MM using FluMel for HCT.r With a median follow-up 73.2 months (range 46.0-158.9) the 5-year OS and PFS rates were 38.6% (95% CI 19.3-57.7%) and 5.4% (95% CI 0.4-21.6%), respectively. Disease recurrence or progression was the primary causes of death in 88% of patients. Shimoni et al reported long term outcomes in a retrospective study of 53 patients.r With a median follow-up of 6.4 years (range, 5-7.9), OS, PFS and TRM were 34%, 26% and 26%, respectively. Adverse prognostic factors for survival included not in remission and long duration of disease (>5 years from diagnosis). Sahebi et al reported 60 patients who either received a FluMel allogeneic HCT or tandem regimen consisting of high-dose melphalan conditioning for autograft followed by low-dose TBI conditioning for allogeneic HCT.r With a median follow-up of 9.8 years, 7-year OS and PFS were 60% and 31%, respectively. Remission status was significantly associated with improved OS. However 10% of patients relapsed late, between 6 and 12 years post HCT.
The optimal dose of melphalan is not well established. Bashir et al reported a randomized phase II trial of FluMel-100 versus FluMel-140 in 50 patients who received MUD HCT for newly diagnosed, relapsed, or primary refractory MM.r There were no significant differences between FluMel-100 and FluMel-140 in time to engraftment, acute or chronic GVHD, response rate, TRM (13% versus 15%), median PFS, 11.7 versus 8.4 months, or median OS, 35.1 versus 19.7 months. Cumulative incidence of disease progression with FluMel-100 and FluMel-140 were 43% and 70%, respectively (P = 0.08). Recurrent disease was the most common cause of death for both FluMel-100 (26%) and FluMel-140 (44%), P = 0.24.
Kroger et al compared the outcomes of patients receiving alemtuzumab or ATG as GVHD prophylaxis in 73 patients enrolled in UK and German multicentre trials of FluMel prior to unrelated HCT.r Most patients had advanced disease with 50% experiencing relapse post autograft. Melphalan 140 mg/m2 was given day -2 together with Fludarabine 150 mg/m2 given over 5 days in the UK study. In the German study, a median melphalan dose of 140 mg/m2 (100-150 mg/m2) was given over days -3 and -2 together with a median fludarabine dose of 120 mg/m2 (90-180 mg/m2) between day -7 to -3 and -5 to -3. 1-year TRM was 27% with no significant difference in estimated 2-year OS (54% versus 45%) or PFS (30% versus 36%) between the ATG and alemtuzumab groups, but less grade II-IV acute GVHD and higher RR was observed in the alemtuzumab group.
Cooper et al reported 12 patients with hemophagocytic lymphohistiocytosis (HLH) who received haploidentical or MUD HCT with FluMel with additional busulfan in haploidentical donors.r 75% were alive at a median follow up of 30 months (range, 9-73). Hamidieh et al reported good outcomes in a prospective study of 10 paediatric patients with primary HLH using FluMel in combination with ATG.r With a median follow-up of 39 months 8 patients were alive with no evidence of disease, 2 died of sepsis and chronic GVHD.
Benign haematological disorders
There have been a number of reports of small patient numbers using FluMel conditioning in benign conditions. The majority of these had positive outcomes when used with alemtuzumab. Kharbanda, et al. reported 6 patients: 4 with sickle cell disease (SCD) 2 with transfusion-dependent thalassemia major using FluMel and alemtuzumab.r GVHD prophylaxis consisted of CsA/MMF. One patient had neutropenic graft failure, 2 had autologous hematopoietic recovery, and 3 had hematopoietic recovery with complete chimerism. The 2 SCD patients with autologous hematopoietic recovery were alive. The remaining 4 died from opportunistic infection, GVHD, or intracranial haemorrhage and because of the lack of improved engraftment and unacceptably high TRM, the study was prematurely terminated. Matthes-Martin et al. reported 8 patients (median age, 9 years) with symptomatic SCD who received MSD HCT using FluMel and either thiotepa or total lymphoid irradiation plus ATG or alemtuzumab.r All patients had mixed chimerism on day +28. After a median follow-up of 4 years 3/8 patients had mixed leucocyte chimerism and all patients had 100% donor erythropoiesis. Other studies also had more positive outcomes with the addition of alemtuzumab.rrrrr
Various GVHD prophylaxis strategies
Others studies explored various GVHD prophylaxis regimens with FluMel. Nakamura et al reported on the use of tacrolimus/sirolimus-based GVHD prophylaxis with FluMel transplantation in 59 patients with MDS.r The 2 year OS, EFS, and RR were 75.1%, 65.2%, and 20.9%, respectively. The cumulative incidence of TRM at 100 days, 1 year, and 2 years was 3.4%, 8.5%, and 10.5%, respectively. The incidence of grade II-IV acute GVHD was 35.4%; grade III-IV was 18.6%. Forty of 55 evaluable patients developed chronic GVHD; of these 35 were extensive grade.
Langston et al explored the role of ATG in FluMel in a single centre retrospective analysis of 85 patients who received unrelated donor transplants for hematologic malignancies, with or without rabbit ATG (6 mg/kg).r ATG was given to patients receiving HLA-mismatched grafts. With a median follow-up of 36 months, those receiving ATG and a mismatched graft had similar rates of acute and chronic GVHD, relapse, and similar OS compared with those receiving HLA-matched grafts without ATG. In a multivariate analysis, HLA-mismatched donor was not associated with a lower OS indicating the efficacy of this regimen with ATG in mismatched unrelated transplants. Please refer to the antithymocyte globulins for use in blood and marrow transplant (BMT) document for more information on the use of ATG.
In vivo T cell depletion with alemtuzumab has been used in combination with FluMel to reduce the incidence of severe acute and chronic GVHD. Ven Besien et al reported specifically on the outcomes of 95 patients treated with FluMel and alemtuzumab (100 mg total dose) vs 59 patients treated Flu Mel alone at two separate institutions respectively. Both groups had similar patient and donor characteristics and there were no significant differences in TRM, relapse, survival, and DFS between the 2 groups.r The incidence of acute graft-versus-host disease aGVHD grade II-IV (RR 5.5, P < 0.01) and cGVHD (RR 6.6, P < 0.01) were significantly lower in patients receiving alemtuzumab. Other studies reported an increased risk of relapse.rr Alemtuzumab is generally well tolerated other than infusional reactions (fever, rash, hypotension), which can be reduced by using steroid premedication. There is also a marked delay in immune reconstitution with alemtuzumab and increased incidence of opportunistic infection, often viral and particularly CMV and low incidence of graft rejection of <3% but is more frequent in mismatched unrelated donors.r The optimal dose of alemtuzumab remains unknown. Numerous studies have explored de-escalation of Alemtuzumab dosing.rrr Doses as low 30 mg provide a low risk of GVHD, with no increase in TRM and improved lymphocyte recovery at 12 months in the setting of HLA-sibling transplantation for various haematological malignancy. Further dose reduction to 20 mg have been shown to result in higher rates of severe GVHD (grade 2-4 5% vs 16%, p=0.012). Using related and unrelated donors for RIC transplantation in AML/MDS, dose reduction of alemtuzumab from 100 mg to 50 mg (10 mg daily for 5 days) resulted in similar outcomes relating to survival, relapse and GVHD.
© Bone Marrow Transplantation 2014
© Bone Marrow Transplantation 2014
Most early reports of Flu/Mel based conditioning were associated with similar high rates of NRM, toxicity and acute and chronic GVHD.rrrr Rates of grade 3-4 oral mucositis in Flu/Mel regimens, particularly when combined with MTX based GVHD prophylaxis have been reported to be up to 40%.r A retrospective analysis of 151 consecutive pts with haematological malignancy reported by Shimoni et al compared outcomes using Flu/Bu (n=72) based conditioning with Flu/Mel 140 (n=79).r Grade III-IV organ toxicity (53% vs 31%, p=0.005) and grade II-IV aGVHD (53 vs 31%, p=0.01) was higher in those receiving Flu/Mel. 2yr OS was 34% with disease status at transplant the strongest predictor of outcome. Flu/Mel was associated with higher NRM and poorer OS in pts in CR vs Flu/Bu (36% vs 72%, p=0.03), but improved disease control in those transplanted with active disease. Regimen related toxicity data from study by de Lima is shown in Table 1.r
© Blood 2004