International Journal of Hematology

, Volume 97, Issue 5, pp 581–598

Indications and outcomes of reduced-toxicity hematopoietic stem cell transplantation in adult patients with hematological malignancies

Authors

    • Department of Medicine, Cell Therapy Centre, Faculty of MedicineUniversiti Kebangsaan Malaysia Medical Centre (UKMMC)
Review Article

DOI: 10.1007/s12185-013-1313-0

Cite this article as:
Wahid, S.F.A. Int J Hematol (2013) 97: 581. doi:10.1007/s12185-013-1313-0

Abstract

Hematopoietic stem cell transplantation (HCT) utilizing non-myeloablative (NMA) and reduced-intensity conditioning (RIC) regimens (collectively referred to as reduced-toxicity HCT, RT-HCT) has become a viable therapeutic option for patients with hematological malignancies who are ineligible for standard myeloablative conditioning transplantation (MA-HCT). RT-HCT has been shown to induce stable engraftment with low toxicity, and to produce similar overall and progression-free survival (PFS) when compared to MA-HCT in acute myeloid leukemia and myelodysplastic syndrome. The best results for RT-HCT have been reported for patients with disease that is in remission, indolent and chemosensitive, and with a strong graft-versus-malignancy effect. Chronic graft-versus-host disease seems to correlate with a lower relapse rate and better PFS. RT-HCT is inferior when performed in poor risk or advanced disease, due to high relapse rates. A search for novel strategies that includes the most appropriate conditioning regimens and post-transplant immunomodulation protocols with more intensive anti-malignancy activity but limited toxicity is in progress. This review provides an update on the results of clinical studies of RT-HCT, and discusses possible indications and investigative strategies for improving the clinical outcomes of RT-HCT for the major hematological malignancies.

Keywords

Reduced-intensity conditioning (RIC)Non-myeloablative (NMA) regimenHematological malignancies (HM)Reduced-toxicity hematopoietic stem cell transplantation (RT-HCT)Graft-versus-malignancy (GvM)Transplantation-related mortality (TRM)Transplant-related toxicities (TRT)

Introduction

Hematopoietic stem cell transplantation (HCT) utilizing less toxic conditioning regimens (RT-HCT) including non-myeloablative (NMA) and reduced-intensity conditioning (RIC) regimens have been developed to reduce the high rate of transplantation-related mortality (TRM) associated with standard myeloablative allogeneic transplant (MA-HCT) [1]. The concept of RT-SCT is to deliver adequate immunosuppression to allow successful engraftment of donor hematopoietic cells and to eradicate the underlying malignancy via graft-versus-malignancy (GvM) effect, while minimizing transplant-related toxicities (TRT). Accordingly, most RT-SCT regimens combine fludarabine and low-dose alkylating agents or total body irradiation (TBI) with or without T cell depleting agents. The addition of in vivo T cell depletion with anti-thymocyte globulin (ATG) or alemtuzumab is thought to be important in limiting the development and severity of graft-versus-host disease (GvHD) and to facilitate engraftment. However, there are potential toxicities associated with its use including increased risk of serious viral infection, EBV-driven post-transplant lymphoproliferative disease, delay in immune reconstitution and disease relapse [24]. To date, there has been no phase 3 randomized controlled trials to address the role of ATG in the RT-SCT setting. Existing data (mostly retrospective) showed that ATG at intermediate doses are able to reduce the incidence of acute GvHD and chronic GvHD and to limit of existing data [57]. A recent review suggests that a dose of 4.5–6 mg/kg of rabbit ATG (rATG) administered on the last few days before HCT is the best schedule to incorporate ATG into RIC regimen (grade 2C), and this dose should be used in future prospective studies to confirm the potential benefit of rATG in RIC for GvHD the type, dose and the timing of ATG still need to be defined [8].

To overcome the toxicities with the use of ATG in RIC regimen, several groups have investigated the impact of a non-ATG-containing RIC regimen on survival, engraftment, GvHD and relapse among patients with HM undergoing CBT. A recent prospective study [9] showed that non-ATG containing RIC regimen provides a low incidence of graft failure without increasing regimen-related toxicity.

What are the indications for RT-HCT?

The indications for performing RT-HCT are determined by two main factors: the patients’ characteristics and the nature of the underlying disease. Commonly cited indications for RT-HCT include patients who are elderly (considered variably in different studies as above 45, 50, 55 or 60 years) and those with poor performance status and poor organ function [10, 11]. Other indication includes failure of a previous autologous HCT [12, 13]. As RT-HCT have relied mainly on the immune-mediated graft-versus-malignancy (GvM) effects for tumor eradication, patients with diseases with a strong GvM effects and a more indolent course (e.g., chronic myeloid leukemia [14, 15], chronic lymphocytic leukemia [16], mantle cell lymphoma [12], and low-grade lymphoma and in remission at the time of transplant would potentially benefit the most from this type of transplant [17].

Table 1 shows the proposed indications for RT-HCT based on recent reports analyzing the efficacy of RIC versus MAC transplants as treatment of specific hematological malignancies (HM).
Table 1

Proposed Indications of RT-HCT versus MA-HCT for adult patients with HM

Disease

RT-HCT

MA-HCT

Comments

AML

Can be considered as standard of care for MAC-HCT ineligible pts in remission preferably in CR1 [23]

Standard of care in younger pts in CR1 except those with favorable cytogenetics and pts with advanced disease (early relapse, CR2, CR3, refractory disease) [68]

RT-HCT is as effective as MA-HCT in AML pts aged 40–60 years in remission and pts with intermediate-risk karyotype in CR1 [7, 23]

May be considered in younger AML pts with good/intermediate cytogenetics in CR1 [7]

 

MAC-HCT ineligible pts or high-risk older pts with good performance status should probably be offered reduced-toxicity full-intensity regimens for better anti-leukemic activity [7, 15, 25, 56]

Consider utilizing “myeloablative” doses of alkylating agents in combination with fludarabine (“full-intensity reduced-toxicity conditioning regimens”) especially in pts beyond CR1 [15, 25]

A clinical option for molecular persistence APML [68]

The outcome of RT-HCT in younger AML patient need to be determined in a prospective randomize study before it can be recommended in this cohort of pts

MDS

A clinical option for MAC-HCT ineligible pts preferably before disease progression and patient who had less than 10 % marrow blast prior to allografting [8, 22, 68]

Standard of care for MDS pts (RA, RAEB, RAEBt, sAML in CR1 or CR2) [68]

Delayed allogeneic HCT is recommended for low & intermediate-1 risk pts, whereas immediate HCT for intermediate-2 and high-risk disease [79]

Consider chemotherapy or azacytidine prior to HCT for MDS pts with excess of blast to reduce the risk of relapse [68]

ALL

Developmental approach for pts aged ≥45 years in CR and ineligible for MAC-HCT [32]

Standard of care in CR1 for high-risk pts (Ph+, hypodiploidy, t(4,11), delayed remission), CR2 and incipient relapse [68]

Allo-HCT for standard risk pts in CR1 can be considered if a MSD is available and the risk of HCT is low

Imatinib given after RT-CST may be particularly relevant for Ph+ ALL

 

Two large retrospective studies showed similar survival in older pts (median age 45–56 years) given RIC and MAC regimens [31, 32]

There are currently insufficient data to routinely recommend RIC for pts who not eligible for MAC-HCT

CML

A clinical option in older pts (>50 years) and younger pts with comorbidities in CP1 with suboptimal or failure of response to TKIs

Standard of care in pts in CP1 with suboptimal or failure of response to TKIs, and controlled accelerated phase (AP) or blast crises (BC) [68]

Patients in CML-AP or -BC should receive a TKI to obtain CP2 before allografting to improve outcome [35]

Consider pre-HCT debulking with TKI

 

Relapse of CML after RIC remains an important problem, which could be successfully treated with post-transplant TKI and DLI [36, 42]

Not recommended for pts with advanced disease (AP/BC) [35]

RIC is currently being explored in older patients of patients with comorbidities who have advanced disease or younger patients wishing to preserve fertility

CLL

A clinical option in poor risk disease who is ineligible for MA-HCT [18, 47]

Standard of care in younger pts with poor risk disease (non-response or early relapse after fludarabine, p53 mutation (del17p13) [45, 68]

CLL is remarkably susceptible to GvM effect, hence a significant proportion of poor risk pts might be cured by RT-HCT [18, 44]

Debulking therapy to obtain optimal disease control prior to RT-HCT is recommended

 

Fludarabine-refractory and bulky disease at the time of transplant were associated with inferior PFS after allografting [48]

Prospective studies are required to determine the timing of allografting during the disease course and select most appropriate conditioning regimen

NHL

Important options for NHL pts (DLBCL, FL) relapsing after auto-HCT or those unsuitable for auto-HCT (failed SC mobilization or extensive BM involvement) [54]

A clinical option in young pts with B-NHL with chemosensitive relapse and ≥CR2 and T-NHL with ≥CR1 [68]

TRM and relapse rate after RT-HCT were low and the long-term outcome seemed favorable in indolent B-NHL and chemosensitive aggressive NHL [14, 51]

Experimental in MCL and PTCL [68]

 

Best results for RT-HCT in younger pts and those without prior autograft or active disease[52] and if rituximab was included in their treatment regimen [53]

The role and long-term outcome of RT-HCT after failure of autograft need to be determined in prospective studies

HL

RIC allo-HCT may be an effective salvage therapy for pts with good risk features who relapse after auto-HCT [11, 12, 60]

A clinical option as salvage therapy in selected young pts with chemosensitive relapse and ≥CR2 [68]

Allo-HCT has mainly been used as salvage therapy for multiple relapsed or refractory HL

 

Patients relapsing after autograft could be salvaged by RT-HCT [11, 12, 60]

Chemosensitivity and disease status at transplant are important predictors of outcome after RT-HCT [11, 12]

The long-term outcome of RT-HCT in relapse HL and its role in patients failing autograft have to be prospectively evaluated.

MM

RIC allografting after an auto-HCT (tandem auto/allo transplant) is an alternative treatment to a second auto-HCT (double autograft) in pts not achieving CR after the first auto-HCT and in high-risk pts (poor cytogenetics or those with progressive disease during induction therapy) [63, 64]

Its use is limited due to high TRM. This approach should be considered in the setting of a clinical trial (for example, testing new conditioning regimen [62]

Allo-HCT is curative in MM but it is associated with considerable TRM which can be reduced with RT-HCT [62]

RIC allografting after autografting can be considered for pts with chemosensitive disease and with low tumor burden [13]

The definite role of tandem auto/allo transplant vs double autograft after failure of first autograft needs to be confirmed in well-controlled prospective studies

Induction therapy with novel agents followed by auto/RIC-allo tandem transplant represents an attractive approach in high-risk MM pts

MF

RT-HCT is the preferred choice of HCT for older pts with intermediate/high-risk score [71, 74, 75]

Standard of care for younger pts with primary or secondary MF with an intermediate/high Lille score [68]

Allo-HCT is the only curative option for MF but its use is limited due to high TRM associated with standard MA-HCT

Pts should be included in prospective clinical trials

 

RT-HCT could result in prolonged survival with a low TRM in older pts (median age 50–60 years) [70, 71]

JAK 2V617F mutation is useful as a post allo-HCT marker of MRD and to guide preemptive DLI [77, 78]

AML acute myeloid leukemia, APML acute promyelocytic leukemia, MDS myelodysplastic syndrome, RA refractory anemia, RAEB refractory anemia with excess blast, RAEB-t refractory anemia with excess blast in transformation, sAML secondary AML, ALL acute lymphoblastic leukemia, CML chronic myeloid leukemia, CLL chronic lymphocytic leukemia, NHL non-Hodgkin lymphoma, DLBCL diffuse large B cell lymphoma, FL follicular lymphoma, MCL mantle cell lymphoma, PTCL peripheral T cell non-Hodgkin’s lymphoma, HL Hodgkin’s lymphoma, MM multiple myeloma, MF myelofibrosis, pts patients, CR1, CR2 first, second complete remission, CP1 first chronic phase, CP2 second chronic phase, AP accelerated phase, BC blastic crises, CR1, 2 first, second complete remission, TRM transplant-related mortality, TKIs tyrosine kinase inhibitors, MSD matched sibling donor, DLI donor lymphocyte infusion, GvM graft-versus-malignancy, PFS progression-free-survival

What are the outcomes of RT-HCT in HM?

Previous studies have suggested that outcomes with RT-HCT are affected by the underlying disease type, disease burden/phase at transplantation, performance status, comorbidity, and the intensity of the conditioning regimen [1726]. It has been historically difficult to generate a reproducible and easy patient comorbidity models for predicting mortality and non-relapse mortality (NRM) after RT-SCT. The ability of this index to predict survival and NRM varies between studies, and in some studies, its predictive potential has not been confirmed [27, 28]. The HCT-specific comorbidity index (HCT-CI) has been developed to identify patients at high risk of mortality after an allograft. Farina et al. [29] showed that HCT-CI significantly predicted NRM (hazard ratio = 1.6, p < 0.03), overall survival (OS) (hazard ratio = 1.62, p < 0.001), and progression-free survival (PFS) (hazard ratio = 1.43, p = 0.002) in a cohort of 203 patients with lymphoma and myeloma after RIC/NMA regimens. In contrast, a recent study [30] reported that HCT-CI on its own did not independently predict NRM or survival but when taken together with age and disease status at transplantation, it can be utilized to further delineate RIC allograft recipients into groups with different outcomes.

The value of 3 risk scoring systems (CCI, HCT-CI using the flexible stratification, and the PA < score) in predicting transplant outcome was recently evaluated among 194 allo-RIC recipients. This retrospective single-center study showed that only the flexible HCT-CT is a good predictor of 2-year NRM and survival after an allo-RIC [31]. Thus, a flexible HCT-CI appears to be a strong independent predictor of outcome in allo-RIC cohort and continual international efforts to develop robust comorbidity scores are warranted.

Retrospective analyses showed that RT-HCT has been associated with lower transplantation-related mortality (TRM) but a higher relapse rate leading to similar overall and progression-free survivals compared to MAC-HCT [11]. There are currently no completed prospective randomized controlled trials comparing outcomes of RT-HCT to MA-HCT as treatment of specific HM. Retrospective comparisons of the outcome of RT-HCT vs MA-HCT are always confounded by the fact that the two patient groups are not matched for factors such as age, disease, or comorbidity that are likely to affect the outcome directly. Currently available data consist primarily of large registry-based survey comparing outcome based on regimen intensity. These studies suggest the following general observations about RT-HCT compared to MA-HCT: (1) they are capable of providing comparable rates of engraftment, (2) lower TRM, (3) higher risk of relapse, and (4) similar overall survival (OS). However, due to either small numbers of patient or heterogeneity of patient population (age, comorbidity, stage of disease) and conditioning regimens, it is not possible to draw definitive conclusion and formulate recommendation from these studies. In fact, it is quite possible that with elimination of patient selection bias, RT-HCT may be superior to MA-HCT [10]. Also, there have been no head-to-head comparisons between the various RIC regimens; hence, to date, there is no consensus as to which of these regimens are optimal for a particular hematological malignancy. Randomized controlled trials are ongoing to define which patients might benefit from RT-HCT.

Limited information is available on quality of life (QOL) among recipients of RT-SCT. Because of anticipated lower TRT associated with RT-SCT, these patients may have better health-related quality of life (HRQOL) in the immediate post-transplant period; however, owing to older age and presence of comorbidities, recipients of RT-SCT are more vulnerable to develop other TRT, namely infections and GvHD. Two studies have evaluated longitudinal QOL in patients with hematological and nonhematological malignancies undergoing RIC and compared with MAC [32, 33]. Both studies reported similar patterns of HRQOL in the 2 cohorts. A recent prospective study [34] comparing HRQOL in patients with myeloid malignancies undergoing RIC (64 patients) and MAC (51 patients) transplant showed that the HRQOL and survival outcomes at 1 year in patients receiving of RIC are not inferior to MAC regimens.

The feasibility and efficacy of RIC regimens has been documented in multiple studies, mostly using matched sibling donors [3540] Recently, single and multi-institution reports demonstrating the feasibility of RIC regimens using unrelated and mismatched donors have also been reported [36, 4042]. Giralt et al. [43] performed a retrospective analysis of transplant outcomes in patients receiving an RIC regimen as reported to the National Marrow Donor Program (NMDP) to describe the long-term outcomes of patients treated with an RIC regimen followed by URD (unrelated donor) HCT, and to define potential prognostic factors for outcome. RIC recipients were older (53 vs. 33 years) and had a higher likelihood of having advanced disease (81 vs. 51 %) when compared to patients undergoing a MAC regimen during the same time period. The 5-year survival rate is 23 %, whereas the 5-year incidence of progression/relapse is 43.4 %. Prognostic factors for better overall survival on multivariate analysis were earlier disease stage, longer time to transplant from diagnosis, better HLA match, ≥90 % performance score, and use of peripheral blood stem cells. This analysis demonstrates that long-term survival and disease control can be obtained with URD HCT after RIC conditioning (RIC-URD). However, only prospective trials will define the impact of donor type in RIC/NMA transplants. Therefore, RIC-URD should continue to be explored in the context of clinical trials.

RIC regimens in cord blood transplant (CBT) are increasingly utilized for patients with high-risk HM who cannot tolerate conventional MAC regimen, and who lack a suitable related or unrelated donor. The optimal conditioning regimen for RIC-CBT has not been established and graft failure remains the greatest challenge. The largest series reported to date by the Minnesota group included 110 patients with various HM [44]. Median neutrophil engraftment was 12 days, grades III and IV aGvHD and cGvHD were 22 and 23 %, respectively, while TRM, OS and EFS at 3 years were 26, 45 and 38 %, respectively. These findings support the use of RIC-CBT as an effective therapy for extending the availability of SCT to older patients. Future directions for optimization of RIC-CBT should include improvement in the post-transplant immunosuppression to minimize graft failure and NRM, as well as to improve immune reconstitution and to prevent relapse.

What are the roles and results of RT-HCT in specific HM?

Table 2 shows recent reports analyzing the outcomes of RT-HCT as treatment of the major HM.
Table 2

Summary of outcome of RT-HCT in Common hematological malignancies

Disease, Type of study, no. of patients, (Ref)

Median (range) age; Disease status at transplant

Conditioning regimen

Donor source

aGvHD; cGvHD

NRM; TRM

Relapse

Survival

AML, Retro, 274 pts [6]

1998–2008

Median follow-up:

38 mths (6–122 mths)

60 yrs (5–74)

CR1 = 160 pts, CR2 = 71 pts, advanced disease = 43 pts

2 Gy TBI ± Flu 90 mg/m2

MRD = 43 %,

MUD = 45 %;

Mismatched = 12 %

5-yr aGvHD (II–IV) = 52 %, cGvHD = 44 %

5-yr NRM = 26 %

5-yr relapse = 42 %

5-yr OS: 33 %,

CR1 = 39 %, CR2 = 41 %, >CR2/refractory = 52 %

AML, MDS; Retro, CIBMTR; 4499 pts [23]

1997–2004

Median follow-up:

RIC BM = 38 mths (4–124 mths)

RIC PBSC = 38 mths (3–90 mths)

MAC = 58 mths (3–128 mths)

RIC: 56 yrs (18–70)

CR1 = 24 %, ≥CR2 = 16 %

MAC: 42 yr (18–68),

CR1 = 32 %, ≥CR2 = 15 %

RIC (CIBMTR definition)

RIC: 768,

MAC: 3731

RIC: MSD = 44 %, MUD = 29 %

MAC: MSD = 42 %, MUD = 27 %

RIC: 100-days aGvHD

(II–IV) = 41.5 %

1-year cGvHD = 40–42 %

MAC: aGvHD = 39 %, cGvHD = 36–39 % (p = ns)

3-month TRM

RIC: 12 %

MAC 18 %

(p = ns)

5-year relapse:

RIC = 39 %;

MAC:32 %

(p = ns)

5-year DFS, OS;

RIC : 33, 33 %;

MAC: 43, 34 %;

(p = ns)

AML, Retro, 93 pts; (RIC: 37, MAC: 56) [7]

1999–2007

Median follow-up:

RIC = 23 mths (2–122 mths)

MAC = 39 mths (1–133 mths)

RIC: 51 yrs (19–68

MAC: 34 yrs(17–66

CR1 (All)

RIC:

Flu 180 mg/m2 + Bu 8 mg/kg + ATG 40 mg/kg

MAC: Cy + TBI

RIC:

MRD = 59.5 % MUD = 35.1 %

MAC: MRD = 66.1 %,

MUD = 28.6 %

RIC: aGvHD (II–IV) = 16.2 %, cGvHD = 56.8 %

MAC: aGvHD (II–IV) = 42.8 % (p = 0.003), cGvHD = 44.6 % (p = n.s)

1-yr, 2-yr TRM

RIC: = 9, 16 %

MAC:

1-yr, 2-yr TRM = 17, 17 %,

(p = n.s.)

Relapse

RIC = 29.7 %,

MAC = 26.8 %

(p = n.s.)

2-yr DFS & OS; 5-yr DFS & OS:

For all pts:

RIC: 2-yr = 57, 65 %;

5-yr = 53, 61 %

MAC: 2-yr = 62, 68 %;

5-yr = 54, 56 %

(All p values = n.s.)

2-yr, 5-yr OS

For intermediate risk karyotype:

RIC: 2-yr = 68 %; 5-yr = 68 %

MAC: 2-yr = 69 %; 5-yr = 66 %

MDS; Retro, EBMT, 836 pts [8]

1997–2001

Median follow-up: RIC = 50 mths MAC: 38 mths

RIC:56 yrs(27–72);

Int-1 = 31 %, Int-2 = 25 %, High = 44 %

MAC: 45 yrs (18–67);

Int-1 = 34 %, Int-2 = 27 %, high = 39 %

RIC:

Flu + Bu/Mel/Thio or Flu + TBI ± ATG/Alem;

MAC: Cy + TBI > 8 Gy, or Cy + Bu 16 mg/kg

(RIC = 215, MAC = 621)

MSD (all)

RIC: 100 days- aGvHD (II)–IV = 43 %, 1-yr cGvHD = 45 % (28 % exrtensive);

MAC: 58 % (p < 0.01),

52 % (25 % extensive)

3-mths, 3-yr NRM

RIC = 15, 22 %;

MAC = 20, 32 %

(p = 0.04)

3-yr relapse:

RIC = 45 %;

MAC = 27 %

(p < 0.01)

3-yr DFS:

RIC = 33 %, MAC = 41 %;

3-yr OS:

RIC = 41 %, MAC: 45 %

(all p values < 0.05).

MDS; Prosp, 31 pts [22]

2006–2009

Median follow-up:

35 mths (6–54.9 mths)

39 yrs (19–63)

Standard risk (RA/RCMD): 20 pts;

High risk (RAEB-1/RAEB-2) = 11 pts;

IPSS Group

INT-1 = 77 %; INT-2 = 12.9), High = 9.7 %

Flu 150 mg/m2 + Bu 8 mg/kg + TBI 4 Gy + ATG 40 mg/kg

MSD: 61.3 %,

MMSD 3.2 %;

MUD: 16.1 %;

MMUD 19.4 %

3-yr

aGvHD (II–IV) = 39.2 %; 3-yr cGvHD = 44.6 %

3-yr TRM = 20.5 %

3-yr relapse = 11.4 %

3-yr OS & EFS = 67.6 and 63.2 %

MDS;

Retro, 36 pts [15]

Median follow-up: 737 days

44 yrs (19–61)

Standard risk: 10 pts

High risk of relapse: 26 pts

Flu 120 or 160 mg/m2 + BU 12.8 mg/kg + ATG 7 mg/kg

PBSC = 28, BM = 8

MRD = 17 patients

URD = 19 patients

Of all pts: 1-yr aGvHD (II–IV) = 19 %

1-yr cGvHD=

37 %

High risk: 1-yr aGvHD (II–IV) = 18 %

1-yr cGvHD = 43 %

TRM:

Standard risk = 10 %;

High risk = 19 %

70-days relapse

Standard risk = 20 %

High risk = 46 %

737-days EFS & OS

Standard risk = 70 and 80 %

High risk = 31, 35 %

EFS: (p = 0.03)

OS: (p = 0.01)

AML/MDS; Retro, 133 pts [25]

1998–2003

Median follow-up:

58 mths

55.6 yrs (23–73)

MDS/sAML/AML = 81,

MPS = 20, lymphoid malignancies = 32

CR1/CP1or MDS RA/RARS = 27 pts;

CR2/CP2 = 106 pts

FBM:

Flu 150 mg/m + BCNU 400 mg/m2 + Mel 140 mg/m2

MRD = 51.1 %

MUD = 44.4 %

MMUD = 4.5 %

100-day aGVHD (III–IV) = 44.4 %,

1-yr, 2-yr Extensive cGvHD = 32.3, 33.1 %

100-days, 1-yr, 5-yr NRM = 15.8, 26.3, 33.9 %

100-days, 1-yr, 5-year relapse = 1.5, 12, 20.1 %

(p = n.s)

1 yr, 3-yr, 5-yr EFS & OS

56.4 and 61.7 %,

46.4 and 53 %,

41.9 and 46.1 %

EFS: (p = 0.07)

OS: (p = 0.3)

Acute leukemia Retro, 64 pts [28]

1999–2005

Median follow-up:

AML = 26 mths (12–26 mths)

ALL = 48 mths (12–73 mths)

AML: 46 yrs (20–60)

ALL: 31 yrs (18–63)

Flu 250 mg/m2 + Bu 12.8 mg/kg + TBI 2 Gy + rATG 4.5 mg/kg

AML : MRD = 44 %; URD = 56 %,

ALL :

MRD = 54 %; URD = 46 %,

MRD: aGvHD (II–IV) = 11 %,

2-yr cGvHD = 40 %;

URD: aGvHD = 35 %,

2-yr cGvHD = 66 %

(p = ns)

TRM = 3 % (MRD + URD)

4-yr relapse:

AML = 17 %;

ALL = 29 %

3-yr DFS & OS;

AML: 83, 83 %

ALL = 65, 78 %,

MRD = 77 and 87 %,

URD = 71 and 74 %

ALL,

Retro, CIBMTR, 1521 pts [31]

1995–2006

Median follow-up:

54 mths (3–166 mths)

RIC: 45 yrs (17–66);

CR1 = 59 %, CR2 = 38 %;

MAC: 28 yrs (16–62),

CR1 = 52 %, CR2 = 48 %

RIC: Flu + Bu ≤ 9 mg/kg, Flu + Mel ≤ 150 mg/m2, Flu + TBI 2 Gy;

MAC: Cy + TBI, Bu + Cy

RIC = 93, MAC = 1428

RIC: MSD = 32 % MRD = 39 %;

MAC: MSD = 41 % MRD = 27 %

RIC: 100-days aGvHD (II–IV) = 39 %, 3-yr cGvHD = 34 %

MAC: aGvHD = 46 %, cGvHD = 42 %

(p = n.s.)

RIC:

1-yr TRM 1- = 26 %, 3-yr 32 %,

MAC: 1- yr = 29 %, 3- yr = 33 %

(p = n.s.)

RIC: 1- yr Relapse = 26 %, 3- yr = 35 %,

MAC: 1- yr = 19 %, 3- yr = 26 %

(p = n.s.)

3-yr OS & DFS :

RIC: 38 and 32 %,

MAC:

43 and 41 %

(p = n.s.)

ALL, Retro, EBMT; 576 pts [32]

1997–2007

Median follow-up:

16 mths (1–119 mths)

RIC = 56 yrs (45–73),

CR1 = 83 %, CR2 = 17 %;

MAC: 50 yrs (45–68),

CR1 = 87 %, CR2 = 13 %

RIC: Flu + Bu ≤ 8 mg/kg, Flu + Mel, Flu + TBI ≤ 6 Gy

MAC: Cy/TBI, Bu/Cy

RIC = 127 MAC = 449

MSD = all pts

RIC: aGvHD (II–IV) = 29 %, cGvHD = 38 %

MAC: aGvHD (II–IV) = 37 %, cGvHD = 36 %

(p = ns)

RIC: 2-yr NRM = 21 ± 5 %,

MAC: 1- yr = 29 ± 2 %,

(p < 0.01)

RIC: 2-yr Relapse = 47 ± 5 %,

MAC: 31 ± 2 %,

(p = 0.03)

2-yr OS & LFS,:

RIC : 48 ± 5 and 32 ± 6 %,

MAC: 45 ± 3 and 38 ± 3 %,

(p = ns)

ALL,

Prosp, 37 pts [33]

2000–2007

Median follow-up:

36 mths (12–96 mths)

45 yrs (15–63)

CR1 = 30 pts, CR2 = 7 pts

Flu 150 mg/m2 + Mel 140 mg/m2

MSD = 27 pts

URD = 10 pt

aGvHD (II–IV) = 43.2 ± 8.3 %,

3-yr cGvHD = 65.6 ± 8.7 %

3-yr NRM = 17.7 ± 6.9 %,

3-yr relapse = 19.7 ± 6.9 %,

3-yr OS & DFS = 64.1 ± 8.6 and 62.6 ± 8.5 %

CML, Retro, 64 pts [10]

1996–2005

Median follow-up : 7 yrs (0.8–9.8 yrs)

52 yrs (17–72)

CP = 46.9 % AP = 45.3 % BC = 7.8 %

Flu + Mel, or Fu + Ara-C + Idarubicin

MRD = 46.9 %;

MUD = 46.9 %;

mismatched = 6.2 %

aGvHD (II–IV) = 31 %,

100-days cGvHD = 31 % (20 % extensive)

1-, 2-, 5-yr TRM = 38, 39, 48 %

relapse = 34.4 %

2-, 5-yr PFS:

CP = 47, 31 %;

AP/BC = 15, 11 %;

2-, 5-yr OS: CP = 66, 48 %; AP/BC = 32, 19 %

CML, Retro, 28 pts [42]

26-yrs (17–49)

All in CP

Flu 180 mg/m2 + Bu PO 8 mg/kg IV 6.4 mg/kg + ATG 20 mg/kg + Imatinib pre-HCT and post HCT

MSD = 13 patients

aGvHD (II) = 7 %,

1-yr cGvHD = 47.6 %

3-yr NRM = 15.4 %

1-yr relapse 32 % (8 patients; 5 patients cytogenetic relapse, 2 patients molecular relapse)

3-yr OS & DFS: 81 and 67 %

CLL, Retro, 28 pts [48]

2001–2008

Median follow-up:

25.1 mths (5.8–81.5 mths)

52.5 yrs (38–68)

Induction failure = 3.7 %,

PR = 59.3 %, PD = 33.3 %, SD = 3.7 %

Flu + Mel ± Rituximab (77.8 % patients), Flu + Cy + Rituximab (18.5 %), Flu +TBI (3.7 %)

MRD = 55.6 %,

MUD = 44.4 %

aGvHD (II–IV) = 29.6 %,

1-yr cGvHD = 47.6 %

2-yr NRM = 24.7 %

2-yr Relapse = 15.4 %

1-, 2-yr OS & PFS:

OS = 80, 64 %

PFS = 72.8, 62.4 %

NHL; Prosp, 123 pts [14]

2002–2008

Median follow-up:

2.5 yrs (0.3–6.6 yrs)

57 yrs (23–70)

Aggressive NHL = 27 patients; Indolent NHL = 8 patients

Flu 200 mg/m2 + Cy 50 mg/kg + TBI 2 Gy + ATG 90 mg/kg

MR = : 92 %,

MMRD = 8 %

100 days-aGvHD (II–IV) = 38 %,

2-yr cGvHD = 50 %

1-yr TRM

Aggressive = 7 %; Indolent = 13 %

1-yr and 4-yr relapse: Aggressive = 15, 32 %; Indolent = 0, 0 %

1- and 4-yr OS:

Aggressive = 81, 58 %, Indolent = 88, 73 %

1- and 4-yr PFS:

Aggressive = 70, 45 %; Indolent = 88, 73 %;

Indolent NHL; Prosp, 47 pts [51]

2001–2006

Median follow-up: 4.6 yrs

53 yrs (39–68)

FL = 36, MCL = 9 %, CLL = 34 %

Flu 150 mg/m2 + Cy 3 g/m2

MSD = all pts

aGvHD (II–IV) = 29 %,

1-yr cGvHD = 48 %

6 mo-TRM = 2.4 %, 3-yr TRM = 9 %

 

3-yr OS & EFS

FL:, 81 and 75 %

CLL/PLL 71 and 59 % others: 64 and 55 %

Relapsed/refractory DLBCL; Prosp, 48 pts [54]

1998–2006

Median follow-up:

52 mths (18–89 mths)

46 yrs (23–64)

Prior auto-HCT = 69 %, chemorefractory = 17 %

Flu 150 mg/m2 + Mel 140 mg/m2 + Alemtuzumab 100 mg

MSD = 62 %,

MUD = 38 %

Mismatched = 19 %

aGvHD (II–IV) = 17 %,

extensive cGvHD = 13 %

4-yr NRM = 32 %

4-yr relapse = : 33 %

4-yr OS & PFS = 47 and 48 %

Relapsed HL, Prosp, 49 pts [59]

1997–2003

Median follow-up:

967 days (102–2232 days)

32 yrs (18–51)

44 pts (90 %) had previous autograft

Flu 150 mg/m2

Mel 140 mg/m2

Alemtuzumab 100 mg

MRD = 63 %

MUD = 37 %

365-days, 730-days aGVHD (II–IV)=

21, 32.6 %

730-days cGvHD = 22.6 %

100-days, 730-days NRM = 4.1, 16.3 %

Relapse = 43 %

4-yr OS & PFS = 55.7, 39 %

HL,

Retro, 285 pts [11]

1995–2005

Median follow-up:

26 mths (3–94 mths)

31.2 yrs (18–57)

CR1 = 2 %

CR ≥ 2 = 14 %

Chemosensitive = 43 %, Chemoresistant = 25 %

Flu + Mel, Flu + Bu, Flu + Cy, Flu + Cy + Thio,

TBI + -Flu

MSD = 60 %

MUD = 33 %

MMUD = 4 %

aGvHD (II–IV) at 100 days = 30 %

cGvHD at 3 yr = 42 %

1-yr NRM=

19.5 %;

3-yr NRM = 21.1 %

6-mth relapse=

147 pts

1-yr, 2-yr, 3-yr PFS & OS = 52 and 67 %; 39 and 43 %; 25 and 29 %

Relapsed/refractory HL, Retro, EBMT; 168 pts [12]

1999–2001

Median follow-up: 75 mths (12–120 mths)

RIC: 30 yrs (9–64)

Chemosensitive = 44.9 %, Chemorefractory = 55.1 %

MAC: 31 yrs (12–61)

Chemosensitive = 45.6 %, Chemorefractory = 54.4 %

RIC:

BCNU 300 mg/m2 +Eto 600-800 mg/m2 +cytarabine 800-1600 mg/m2 + Mel 100-140 mg/m2/Flu + Bu/Mel/Cy/Thio/TBI (2 Gy)

MAC:

Cy + TBI (> 8 Gy)/Bu ((RIC = 89 pts, MAC = 79 pts)

RIC:

MSD = 86.5 %

URD = 13.5 %

MAC:

MSD = 88.6 %

URD = 11.4 %

RIC:

100-days aGvHD (I–IV) = 44 %,

1-yr cGvHD = 38 %

MAC:

100-days aGvHD (I–IV) = 53 %,

1-yr cGvHD = 33 %

3-mth, 1-yr, NRM

RIC = 15, 23 %

MAC = 28, 46 %

6-mth relapse

RIC = 57.3 %

MAC = 30.4 %

5-yr OS & PFS

RIC = 28 and 18 %,

MAC = 22, 20 %

MM, Prosp, 100 pts [63]

1999–2005

Median follow-up: 5 yrs

54 yrs (30–65)

Stage III = 67 %, II = 29 %;

Chemosensitive = 48 %

Mel 200 mg/m2 + TBI 2 Gy

MSD = all pts

41-days aGvHD (II–IV) = 38 %,

120-days cGvHD = 50 %

Overall TRM = 11.4 %

Overall relapse of 4–5 yrs = 44 %

2-yr, 5-yr Disease related mortality = 5.2, 20.5 %

After median follow-up of 5 yrs:

Whereas median EFS was 3.1 yrs (2.6–4.5 yrs), median OS was not reached

MM, Prosp, 357 pts [65]

2001–2005

Median follow-up:

61-mths

69 yrs (31–69)

CR = 27 pts

PR = 282 pts

SD = 48 pts

Auto:

Mel 200 mg/m2

Auto-allo:

TBI 2 Gy + Flu 90 mg/m2

Auto-allo = 108 pts,

Auto = 249 pts

(Single auto = 145, Tandem Auto = 104)

MSD = 108 pts

MUD = 249 pts

100-days aGvHD (II–IV) = 20 %

100-days cGvHD = 54 %

2-yr, 5-yr NRM

Auto-allo = 12, 16 %

Auto alone = 3, 4 %

(p < 0.01)

5-yr relapse:

Auto-allo = 49 %

Auto alone = 78 %

(p = 0.003)

5-yr PFS & OS:

Auto-allo = 39 and 65 %

Auto alone = 19 and 58 %

OS: (p = 0.006)

MF, Prosp, (EBMT), 103 pts [71]

2002–2007

Median follow-up: 33 mths (12–76 mths)

55 yrs (32–68)

Lille risk profile: low 18 %, intermediate 58 %, high 7 %

Flu 180 mg/m2 + Bu PO 10 mg/kg 6.4 mg/kg + ATG 20 mg/kg

MSD = 33 pts

MUD = 70 pts

aGvHD (II–IV) = 2 %,

cGvHD = 48 %

1-yr NRM 16 %

(Full matched donor = 12 %)

3-yr relapse = 22 % (low risk Lille Score = 14 %; intermediate = 22 %, high = 34 %)

(p = 0.02)

5-yr DFS & OS = 51 and 67 %

MF, Retro (GITMO), 100 pts (RIC = 52, MAC = 48) [74]

1986–2006

Median follow-up:

5 yrs

49 yrs (21–68)

90 % had intermediate/high risk Dupriez score

RIC: Thio10 mg/kg + Cy 100 mg/kg,

Flu + Bu

Flu + TBI,

Flu + Mel

MAC:,

Bu + Cy,

TBI + Cy,

Thio 15 mg/kg + Cy 120 mg/kg ± Mel

MSD = 78 %

MMUD = 22 %

100 days-aGvHD (II–IV) = 41 %

2-yr cGvHD = 43 %

(p = ns)

1-yr TRM = 35 %,

3-yr TRM = 43 %

(p = ns)

2-yr relapse = 41 %;

2-yr, 5-yr DFM = 5.2, 20.5 %

(p = ns)

3-yr OS & DFS = 42 and 35 %

(p = ns)

RT-HCT reduced-toxicity stem cell transplantation, aGvHD acute graft-versus-host disease, cGvHD chronic graft-versus-host disease, AML acute myeloid leukemia, sAML secondary AML, MPS myeloproliferative syndromes, MDS myelodysplastic syndrome, RA refractory anemia, RARS refractory anemia with ringed sideroblasts, RCMD refractory cytopenia with multilineage dysplasia, RAEB 1, 2 refractory anemia with excess blast 1, 2, MPD myeloproliferative disorder, ALL acute lymphoblastic leukemia, CML chronic myeloid leukemia, CLL chronic lymphocytic leukemia, FL follicular lymphoma, MCL mantle cell lymphoma, NHL non-Hodgkin lymphoma, DLBCL diffuse large B cell lymphoma, HL hodgkin lymphoma, MF myelofibrosis, IPSS international prognostic scoring system, Retro retrospective study, Prosp prospective study, pts patients, mth month, yr year, CR1, CR2 first, second complete remission, CP1 first chronic phase, AP accelerated phase, BC blastic crises, PD progressive disease, SD stable disease, RIC reduced-intensify conditioning, MAC myeloablative conditioning, Flu fludarabine, Bu busulfan, Cy cyclophosphamide, TBI total body irradiation, Mel melphalan, BCNU 1,3-bis(2-chloroethyl)-1-nitrosourea, or Carmustine, Thio thiotepa, ATG anti-thymocyte globulin, r-ATG rabbit anti-thymocyte globulin, Alem alemtuzumab, MSD HLA-identical sibling donor, MMSD mismatched sibling donor, MRD matched related donor, URD unrelated donor, MUD full-matched unrelated donor, MMUD mismatched unrelated donor, NRM non-relapse mortality, TRM transplant-related mortality, DFM disease related mortality, OS overall survival, DFS disease-free survival, EFS event-free survival, ns non-significant, EBMT European Group for Blood and Marrow Transplant, CIBMTR Centre for International Blood and Marrow Transplant Research

Acute myeloid leukemia (AML)/myelodysplastic syndromes (MDS)

Patients with AML older than 60 years of age have a poor prognosis with conventional chemotherapy and are usually not candidates for curative standard MA-HCT [18, 45]. Hence, RT-HCT may be the best treatment option in many of such AML patients. In support of this notion, a recent large retrospective analysis [18] showed that HCT using NMA/RIC regimens from related or unrelated donors can result in long-term remissions in older or medically infirm patients with de novo or secondary AML by virtue of graft-versus-leukemia (GvL) effect.

The composition and the intensity of RIC regimens that have been used in myeloid malignancies vary considerably [11]. The least intensive, best-tolerated regimen consists of 2 Gy TBI and fludarabine [18], while the more intensive RIC regimen that are commonly used include busulfan 9 mg/kg or less or melphalan 140 mg/m2 or less combined with fludarabine 160 mg/m2 or less [17, 19, 4648].

Most previous RT-HCT studies performed specifically in AML report a 2- or 3-year leukemia-free survival (LFS) ranging from 27 to 60 % and an OS rate of 30 to 60 % [17, 18, 39, 49, 50]. Factors shown to be associated with survival included disease burden at the time of HCT, cytogenetic risk and chronic graft-versus-host disease (cGvHD). RT-HCT is more likely to be effective if performed when the disease is in complete remission (CR) and of limited benefit for patients with advanced or refractory disease [17, 18]. A large prospective study on RT-HCT [18] showed that the 5-year OS was 37 % for AML patients transplanted in first CR (CR1), 34 % for those in second CR (CR2) and 18 % for those with more advanced disease (Table 2). Thus, patients with active disease at the time of transplant should not be considered for RT-HCT unless a morphological remission can be achieved with salvage chemotherapy. Likewise, patients with secondary AML (AML evolving from MDS, or therapy-related AML) may be at high risk of relapse following RT-HCT [20]. Other factors associated with survival included cytogenetic risk (5-year OS is 40 % with good/intermediate cytogenetics and 19 % with poor cytogenetics) and cGvHD that was associated with better survival due to 50 % reduction in relapse rate [18]. The latter observations clearly demonstrate the protective effect of cGvHD on relapse, indicating that AML is highly susceptible to graft-versus leukemia (GvL) effects.

Several retrospective studies comparing RT-HCT and MA-HCT in AML and MDS [45, 51, 52] suggest that RT-HCT can yield satisfactory survival for patients who are ineligible for MA-HCT. To better address the important question of whether the intensity of the conditioning regimens correlates with survival, Luger et al. [49] reported the largest comparative analysis of outcomes of transplants among AML/MDS patients aged 40–60 years receiving MAC (n = 3731), RIC (n = 1041) and NMA (n = 407) regimens. NMA conditioning resulted in inferior disease-free-survival (DFS) and OS, but there was no difference in DFS and OS between RIC and MAC regimens. Similar findings have been reported by 2 other studies demonstrating that RIC is as effective as MA-HCT regimens when the disease is in remission at the time of transplant while patients with active disease could only be salvaged with MA-HCT [17, 20]. A subset analysis showed that RIC is as effective as MAC regimens in AML patients with intermediate-risk karyotype transplanted in CR1 [17] (Table 2). Prospective studies comparing RIC regimens are needed to confirm which approach is best for different risk groups of AML patients.

Allogeneic HCT remains the only potential cure for de novo MDS and offers a good chance of long-term DFS, if the treatment is performed before disease progression. However, since most patients with MDS are older than 60 years, few are candidates for conventional MA-HCT: RT-HCT becomes an important treatment option for many MDS patients. Data from retrospective studies demonstrated reduced non-relapse mortality (NRM) [19, 47], comparable survival times, but higher relapse rate in the RT-HCT group [19] (Table 2).

Although RT-HCT has been increasingly used in AML/MDS to reduce TRM, relapse remains a common problem particularly patients with unfavorable cytogenetics, high-risk MDS and low CD34+ cells infused [19]. Therefore, recent studies have focused on the development of regimens that could reduce relapse without increasing toxicity. A few studies involving AML and ALL patients showed that the addition of low-dose TBI (4 Gy) to the standard RIC regimen reduced risk of relapse (4-year relapse rates of 17–29 %) without increasing TRM [48, 53, 54]. A possible explanation is that, directed radiotherapy could improve the efficacy of tumor immunotherapy by causing apoptosis of the tumor cells and enhancing antigen presentation by MHC class I molecules [54]. A recent study [48] on a small cohort of MDS patients showed that the combination of Flu/Bu/TBI 4 Gy resulted in a lower relapse rate and encouraging survival data, therefore this regimen may represent a viable therapeutic option for treating patients with de novo MDS with a high risk of relapse (Table 2).

Another approach is to reduce the leukemia burden prior to allografting by (1) administering sequential chemotherapy followed by RIC in high-risk AML and MDS patients [55] or by (2) increasing the dose intensity of the conditioning regimen to increase its anti-leukemic activity. Conditioning regimen incorporating “myeloablative” doses of alkylating agents (12.8 mg/kg IV busulfan or 400 mg/m2 BCNU plus 140 mg/m2 melphalan) in combination with fludarabine have been shown to produce effective disease control with low NRM even in advanced cases of AML/MDS [26, 50] (Table 2). Recent retrospective study on 79 AML/MDS patients older than 55 years showed good outcomes, including OS, EFS, TRM and GvHD rates following conditioning with IV busulfan (520 mg/m2) plus IV fludarabine (160 mg/m2) and the authors recommended the systematic use of myeloablative reduced-toxicity conditioning in high-risk AML/MDS patients up to age 65, unless they suffer serious comorbidities [56].

URD transplants are in general only performed in patients with high-risk AML. RIC regimens have been increasingly used in an attempt to reduce the TRT associated with URD transplants. To date, there is limited data where RIC and MAC have been compared in AML patients receiving grafts from URD. In a study comparing URD with HLA-identical sibling donors in patients with AML receiving NMA, LFS was the same [57]. In another study among AML patients given RIC, 2-year LFS was 40 % using related donors, compared to 17 % using URD [58]. A large retrospective EBMT study involving 1555 AML patients (RIC: 401, MAC: 1154) [59] showed that RIC-URD transplants are associated with higher relapse in patients younger than 50 years and decreased NRM in those above 50 years compared with MAC-URD, while LFS was similar after both conditioning regimens, regardless of age. This study illustrates that the relative impact of dose intensity may depend on age, and encourages the use of RIC-URD in AML patients in particular older patients who lack suitable related donor.

Several clinical trials have shown the feasibility of RIC in haploidentical HCT (RIC-Haplo) in adult patients with HM. The high engraftment rates as well as the low GvHD rates observed after haploidentical SCT using RIC containing ATG is intriguing [60, 61]. A recent study showed that RIC-haploidentical conditioning with fludarabine/busulfan/ATG in 83 patients with AML and MDS produced consistent donor cell engraftment with low rates of GvHD and TRM [62]. The incidence of grade II–IV aGvHD, cGvHD and TRM were 20, 34 and 18 %, respectively, while EFS and OS were 56 and 46 %, respectively, for patients with AML in remission; and 53 and 53 % in MDS. Thus, this approach may be used when a suitable HLA-matched donor is not available or when allogeneic HCT is urgently needed in particular for elderly patients or patients with comorbidities.

Acute lymphoblastic leukemia (ALL)

In contrast to myeloid malignancies, there have been less published experience of RT-HCT in ALL. Preliminary data from two large registry studies (CIBMTR [63] and EBMT registries [64] comparing RIC and MAC-HCT in ALL have shown similar DFS and OS despite RIC patients being older than MAC-HCT patients (Table 2). Multivariate analysis in the CIBMTR ALL study showed that conditioning intensity did not affect TRM, relapse rate or survival, and improved survival was observed in Philadelphia (Ph)-negative ALL patients who were younger than 30 years, in CR1, those who had better performance status, well-matched donors and conditioning with TBI [63]. In contrast to the CIBMTR study, the EBMT study revealed that RIC patients experienced lower NRM and higher relapse rates; however, similar to the CIBMTR study, regimen intensity had no impact on OS and LFS. These encouraging results confirm the existence of graft-versus-ALL effect. A prospective study [65] demonstrated better outcomes among patients transplanted in CR1 than CR2 and among patients who developed cGvHD (Table 2). A recent prospective study on 51 high-risk ALL patients (21 Ph+ ALL) in CR1 showed that fludarabine/2 Gy TBI with post-allografting imatinib for Ph+ ALL patients resulted in improved OS (3-year OS was 62 % for Ph-positive and 52 % for Ph-negative patients) [66]. Despite the need for prospective randomized trials, current data show that RT-HCT is a potential therapeutic option for older ALL patients (aged ≥ 45 years) who are in CR and are not eligible for MA-HCT. Strategies aimed at reducing relapse by reducing leukemia cell burden pre-transplant and augmenting GvL effects post-transplant using tyrosine kinase inhibitors (TKIs) for Ph+ ALL and prophylactic DLI are currently being investigated.

Chronic myeloid leukemia (CML)

Allogeneic HCT remains the only therapeutic option for a minority of patients who fail TKI therapy or progress to advanced disease. Several studies have demonstrated GvL effects for patients with CML, making this disease an ideal candidate for allogeneic HCT using RIC regimen [15]. RT-HCT therefore represents important options for CML patients with imatinib failure who are older (>50 years) and have comorbidities. To date, single and multi-institution studies comparing RT-HCT and MA-HCT have not provided consistent data with regard to OS, relapse rates and frequency of GvHD [21, 6771]. Chen et al. [69] reported a higher 5-year OS for the RIC group (70 vs 56 %), better DFS, higher incidence of cGvHD (80 vs 66 %) and a similar incidence of acute GvHD (14 vs 18.7 %). The timing of the HCT and stages of disease significantly influence the transplant outcome with patients in chronic phase (CP) having improved OS than those with advanced disease [accelerated phase (AP) and blast crisis (BC)] [21, 72] (Table 2). The poor outcome after RT-HCT in advanced disease suggests that the larger tumor burden may outpace the GvL effect and efforts should be made to restore a second CP (CP2) with AML-like chemotherapy or second generation TKI before proceeding to HCT. The benefit of RT-HCT for young CML patients is unclear in view of the small number of patients’ analyzed and short follow-up. Two studies involving young patients (median age 34–35 years) with CML in CP1 who received a similar RIC regimen (fludarabine/busulfan/ATG) reported different OS rate (85 % at 37 months [14] vs 35 % at 30 months [73].

Relapse of CML after RT-HCT remains an important problem, which could be resolved by post-transplant imatinib and/or DLI [74]. However, the use of DLI in the early post-transplant period is associated with life-threatening GvHD and pancytopenia thus making imatinib an attractive alternative for early relapse after RT-HCT. Imatinib could delay the requirement of DLI and synergized with DLI [68, 74]. RIC regimen combined with pre- and post-transplant imatinib for patients in CP1 showed encouraging results resulting in high molecular remission and survival rates [74] (Table 2). Importantly, this study confirmed that the majority of patients relapsing after RT-HCT could successfully achieve complete molecular remission (CMoR) with imatinib, with 21 patients remaining in CMoR 30 months after withdrawal of imatinib. It is currently believed that imatinib does not deplete leukemic stem cells and its discontinuation usually leads to relapse. However, when combined with allogeneic HCT, the maintenance of durable response is possible, even after discontinuation of imatinib [74] hence providing evidence that the administration of RIC/imatinib combination in early CP could provide a definitive cure for CML patients.

Chronic lymphocytic leukemia (CLL)

Chemoimmunotherapy which is the first line treatment for CLL has limited efficacy in patients with poor cytogenetic features (17p and 11q deletion) [75]. Allogeneic HCT remains the only potentially curative treatment in CLL and is currently indicated for patients with poor response to chemoimmunotherapy. There is strong evidence that the therapeutic efficacy of allogeneic HCT in CLL derived mainly from a GvL activity even in poor risk patient. GvL effectiveness in CLL is indicated by a reduced relapse rate in the presence of cGvHD [76], increased relapse rate after a T cell depleted allogeneic HCT and efficacy of DLI [77].

Earlier studies of MA-HCT in younger patients with CLL resulted in durable remissions and a survival plateau; however, the TRM was exceptionally high (25–50 %) [78]. Due to advanced age of patients with CLL, RT-HCT has been exploited to reduce TRM and increase the availability of allogeneic HCT. RT-HCT regimens seem to provide a high overall response rates that are comparable to those with MA-HCT implicating the effect of GvL in long-term disease control. The OS and progression-free survival (PFS) at 2 years ranges from 50 to 70 % and, 45 to 72 %, respectively. The addition of rituximab to fludarabine/cyclophosphamide conditioning regimen has increased the response rates to 90 % and 2-year OS to 100 % [79].

Importantly, fludarabine-refractory disease and bulky lymphadenopathy at the time of transplant were associated with inferior PFS [80] implying that fludarabine-refractory disease could predict intrinsic resistance to GvL activity or could be a marker of aggressive disease that progresses too rapidly to be controlled by even an active immune response [80] (Table 2). For patients with less than optimal disease control at the time of transplant, potential ways to improve the efficacy of RT-HCT that are being investigated include aggressive debulking prior to transplant, immunomodulation with monoclonal antibodies, DLI post-transplant and incorporation of novel agents with activity against CLL in the conditioning regimen [80].

Non-Hodgkin lymphoma (NHL)

Autologous HCT is still considered the standard therapy for patients with chemosensitive relapse of diffuse large B cell lymphoma (DLBCL) [81, 82]. Compared to autologous HCT, a lower relapse rate and longer DFS has been reported after MA-HCT but the OS was not improved because of high TRM [82]. RT-HCT therefore represents important options for NHL patients failing or unsuitable for autologous HCT due to inadequate stem cell mobilization or extensive bone marrow infiltration. Prospective studies of RT-HCT revealed superior OS and PFS in patients with follicular, mantle cell lymphoma and DLBCL [25, 83] highlighting the importance of graft-versus-lymphoma effect in indolent and chemosensitive aggressive lymphomas (Table 2).

The major predictors of outcome after RT-HCT are disease state at the time of transplant, chemosensitivity, a previous failed autograft and age. The best results for RT-HCT have been reported for younger patients and those without prior autograft or active disease. Corradini et al. [84] reported that TRM was lower in patients <55 years than >55 years (5 vs 28 %), and 2-year OS was higher in chemosensitive than refractory disease (90 vs 35 %), and in patients who had not received an autograft than those who had prior autograft (75 vs 60 %). Recent reports showed that patients with NHL who relapse after autograft may still benefit from RT-HCT if their disease is chemosensitive and rituximab is included in their treatment regimen [85, 86] (Table 2). Future studies will be required to determine whether any subset of patients with relapsed NHL should be considered for RT-HCT versus autologous HCT as first line salvage therapy. The role of RT-HCT as front-line therapy after debulking therapy for newly diagnosed patients with poor prognostic features (e.g., mantle cell lymphoma with blastoid morphology [85] or highly aggressive lymphoblastic lymphoma) merits further investigation.

Hodgkin lymphoma (HL)

Similarly to high-grade NHL, autologous HCT is the current standard of care for patients with HL in first chemosensitive relapse or second CR [87]. Allogeneic HCT has mainly been used as salvage therapy for multiple relapsed or refractory HL [88]. Conventional MA-HCT carries a high risk for TRM and should therefore be considered in selected young patients. Data from prospective and retrospective studies on small groups of HL patients with a relatively short follow-up suggest that RT-HCT is a feasible, tolerable and can induce durable clinical remission in relapse/refractory even if they were heavily pretreated [89, 90]. Peggs et al. [91] reported encouraging survival rates with low NRM even in multiple relapses HL treated with RIC-allograft and DLI from unrelated and related donors (90 % of patients had progressive disease after previous auto-HCT) indicating the existence of a clinically relevant graft-versus-HL (Table 2). A recent retrospective analysis [13] performed on 24 heavily pretreated patients with relapsed HL (20 patients had prior autograft) who had received RIC (fludarabine/melphalan) allogeneic HCT showed that NRM and OS at 2 years were 13 and 60 %, respectively. Retrospective analysis [22] have been performed on large cohort of patients in an attempt to identify prognostic factors predicting the outcome after RT-HCT (Table 2). TRM and PFS were significantly worse with chemoresistant disease and extent of prior therapy. The use of unrelated donor had no impact on NRM. Relapse within 6 months of a prior autologous transplant was associated with a higher relapse rate and a lower PFS while acute or chronic GvHD by 9 months post-HCT was associated with a lower relapse rate. The LWP of the EBMT [23] performed a study comparing RT-HCT and MAC-HCT in patients with relapsed/refractory HL have shown decreased NRM and increased PFS and OS in the RT-HCT group (Table 2). The development of cGvHD significantly decreased the relapse rate after transplant which is translated into a better PFS indicating the existence of a graft-versus-HL effect [23]. Chemosensitivity and CR status at transplant appeared to be the most important predictors to improve survival after RT-HCT, thus suggesting that pre-transplant disease control is one of the main determinants of long-term outcome. Taken together, current available data suggest that RT-HCT may be an effective salvage therapy for patients with good risk features who relapse after autografting, while patients with chemorefractory disease or poor performance status had a poor outcome and it is difficult to recommend RIC allograft for these groups of patients [92]. For patients deemed to be at high risk of failing an autograft, a RIC allograft may represent a more effective therapy and its role in this setting should be prospectively evaluated.

Multiple myeloma (MM)

Despite remarkable advances in its treatment, allografting remains the only potential curative approach which may offer long-term DFS via its well-documented graft-versus-myeloma (GvM) effects [93]. However, its use remains controversial especially in newly diagnosed patients owing to the high TRM following MA-HCT [94]. In contrast, RT-HCT can result in sustained remissions with relatively low TRM. However, substantial cytoreduction pre-allografting is often necessary because of a variable graft-versus-myeloma effect. In MM patients, the best results are achieved when RIC is part of a planned tandem transplant (autograft followed by RIC allografting). This tandem transplant approach allows prolonged survival and long-term disease control in patients with reduced tumor burden at the time of allografting [95] (Table 2). The use of RIC allograft immediately after autologous HCT provides a temporal separation between tumor reduction by a high-dose melphalan-based autograft and the GvM of a non-myeloablative allograft. There are currently a number of prospective trials attempting to address the issue of whether this strategy leads to decrease in relapse and/or improvement in OS as compared with double autologous transplants. Two prospective studies [95, 96] showed that auto/RIC-allo tandem transplants resulted in longer OS and event-free survival (EFS) in newly diagnosed MM younger than 65 years as compared with double autologous transplants. A recent analysis [97] involving the largest cohort and the longest follow-up of MM patients demonstrated that the long-term superiority in terms of OS and relapse of auto/RIC tandem transplants over autograft alone, implying that the antitumour effect of RIC allograft is more important for the outcome than TRM associated with RIC allografting (Table 2). Unfortunately, similar to autografting, relapse remains the major cause of treatment failure after RT-HCT in particular those with active disease at the time of allografting [98]. Profound cytoreduction (CR or very good partial response) before allografting was associated with longer response duration and EFS [24]. To improve treatment results with allografting, consideration should be given to incorporating immunomodulatory drugs and targeted treatments to enhance pre-transplant remission status [94] as post-transplant maintenance therapy, or in combination with DLI for refractory or relapsed disease. Sequential treatment approach consisting of induction therapy with novel agents followed by auto/RIC-allo tandem transplant represents an attractive approach in high-risk MM pts including patients with poor cytogenetics [t(4:14), t(14:16), del(13), complex karyotype or hypodiploidy] or those with progressive disease during induction therapy. Studies exploring these strategies are ongoing [99].

Myelofibrosis (MF)

Allogeneic HCT is the only curative treatment in MF and has been generally performed in primary MF patients with intermediate and high-risk prognostic score [100]. Nevertheless, its use is limited by a high TRM associated with conventional MA-HCT [101]. RT-HCT has now become a preferred therapeutic option particularly in older patients due to lower TRM [102, 103] while maintaining the potential for eradicating the disease based on the graft-vs-MF effect. Evidence of the existence of a graft-vs-MF came from two reports [104, 105] in which patients with relapse after allogeneic HCT achieved regression of marrow fibrosis after DLI.

A prospective study involving a large cohort of MF patients treated with RT-HCT revealed a favorable outcome with a 16 % 1-year TRM, 22 % 3-year relapse rate, 67 % 5-year OS, and 51 % 5-year DFS [103] (Table 2). Age above 55 years and HLA-mismatched transplantation significantly predicted inferior OS and a high-risk Lille score was independent factor for relapse and decreased DFS. Several retrospective comparisons between RT-HCT and MA-HCT have shown that the intensity of the conditioning regimen does not significantly influence the outcome, even though RT-HCT recipients were older and had more comorbidities, they had less TRM and tended to survive [106, 107] (Table 2) and importantly relapse rates at 3 years did not differ between RT-HCT and MA-HCT groups [107]. Taken together, the above data provide strong evidence that RT-HCT leads to a better outcome than MA-HCT, at least in the older MF patients and it is probable that the same result would apply for young patients. Future prospective studies should address strategies to further improve the outcome of RT-HCT including the use of the novel JAK-thyrosine kinase inhibitors as remission inducing drugs pre-HCT [108] and the use of JAK2 V617 mutation level as a marker for residual disease post-HCT to guide adoptive immunotherapy with DLI [109, 110].

Conclusion

The advent of HCT incorporating less toxic conditioning regimens permits the delivery of a potentially curative GvM effect with less TRT to the majority of patients with HM whose outcome with conventional chemotherapy would be dismal. Hence, RT-HCT has become a reasonable curative option for patients with HM who are not suitable for MA-HCT on the grounds of advanced age, poor performance status and presence of comorbidities. AML and MDS patients who are not suitable for MA-HCT are now considered standard indications for RT-HCT as a number of studies have demonstrated durable long-term remissions after RT-HCT. RT-HCT continues to be an important strategy for a significant proportion of patients with chronic leukemias with better outcomes noted when performed early in the disease and if optimal disease control prior to allografting. The role of RT-HCT in younger ‘standard risk’ adults with acute/chronic myeloid and lymphoid leukemias remains to be addressed in prospective trials. RT-HCT has become the preferred HCT procedure for MF patients. RT-HCT is feasible and tolerable and can induce durable clinical remission in relapse/refractory NHL and HL even if they were heavily pretreated: therefore, RT-HCT represents an important option for lymphoma patients failing or unsuitable for autologous HCT due to inadequate stem cell mobilization or extensive bone marrow infiltration. RT-HCT can induce durable remission in myeloma with acceptable toxicities and the best results are achieved when RT-HCT is performed after autograft (tandem transplant).

Recent prospective studies have shown HCT using CB or HLA-haploidentical family donors was feasible using RIC regimens. The high engraftment rates as well as the low GvHD rates observed after haploidentical SCT using RIC containing ATG are intriguing. In the setting of unrelated donor HCT, the use of RIC containing anti-T cell antibodies appears to negate the adverse effects of HLA disparity on engraftment, GvHD and TRM [111, 112]. Thus, this transplant approach may be considered when a suitable HLA-matched donor is not available or when allogeneic HCT is needed urgently. Disease stage, performance status, stem cell source, HLA matching and timing of transplant emerged as the most important prognostic factors for survival after RIC-URD transplant, and should be considered when planning and designing future trials with this treatment modality. Although RT-HCT may have expanded the numbers of patients who can benefit from allogeneic HCT, relapse is currently the most common cause of treatment failure as the regimens used may not have conferred enough cytoreduction for adequate disease control. Chemosensitivity and CR status at transplant appeared to be the most important predictors to improve survival after RT-HCT, thus suggesting that pre-transplant disease control is one of the main determinants of long-term outcome. Patients with disease that is sensitive to GvM effects and do not possess a rapid turnover would be ideal candidates for RT-HCT. As there are major differences among malignancies in their susceptibility to GvM effects, it would be necessary to perform prospective comparisons of higher and lower intensity regimens for specific disease states in order to find the best compromise between efficacy and toxicity. The role of the comorbidity models to predict survival and NRM after RT-SCT has not been extensively studied and their validation in independent cohorts of allo-RIC recipients is required. This information can be used to guide patient selection for RT-SCT and to appropriately counsel patients of the risks and benefits of this treatment. Since CR status is the most important predictor of a favorable outcome after RIC allograft, future studies should be aimed at integrating intensified or targeted therapy into the overall treatment algorithm to pursue the best response before allografting. In addition, survival outcome can also be improved by incorporating targeted treatment and immunomodulatory drug as post-transplant preemptive or maintenance therapy after RIC/NMA transplants.

Acknowledgments

The author would like to thank Aqilah Md Pazil and Siti Nor Fatimah Mohd Zain for their technical assistance in the preparation and submission of the manuscript and the Dean of Faculty of Medicine, UKM Medical Center for his continuous support.

Conflict of interest

The author declares that she has no conflict of interest.

Copyright information

© The Japanese Society of Hematology 2013