INTRODUCTION — Graft-versus-host disease (GVHD) can develop after allogeneic hematopoietic cell transplantation (HCT), when immune cells from the transplant donor (the graft) initiate an immune reaction against the transplant recipient (the host). Acute GVHD and chronic GVHD are multisystem disorders that are distinguished by their clinical findings, according to the widely-accepted National Institutes of Health (NIH) consensus criteria; an overlap syndrome with features of both acute GVHD and chronic GVHD is also recognized.
Prevention of GVHD is discussed in this topic.
Other aspects of GVHD are discussed separately:
●(See "Clinical manifestations, diagnosis, and grading of acute graft-versus-host disease".)
●(See "Clinical manifestations and diagnosis of chronic graft-versus-host disease".)
●(See "Treatment of acute graft-versus-host disease".)
●(See "Treatment of chronic graft-versus-host disease".)
●(See "Pathogenesis of graft-versus-host disease (GVHD)".)
OVERVIEW OF GVHD AND ITS PREVENTION — GVHD can develop after allogeneic hematopoietic cell transplantation (HCT), when immune cells from an immunologically non-identical donor (the graft) initiate an immune reaction against the transplant recipient (the host). The incidence and severity of GVHD vary with the degree of immunologic mismatch, aspects of the transplantation procedure, and patient risk factors. Effective prophylaxis can decrease the incidence of GVHD, but it does not eliminate it entirely.
Acute GVHD (aGVHD) and chronic GVHD (cGVHD) are multisystem disorders that share some features, but they can be distinguished by clinical findings, affected organs, and underlying pathophysiology. The syndromes are classified and severity is graded according to the widely-accepted National Institutes of Health (NIH) consensus criteria; an overlap syndrome with features of both aGVHD and cGVHD is also recognized [1].
●Acute GVHD – aGVHD classically presents in the early post-transplantation period (eg, first 100 days), although it can present later. It is initiated by tissue damage to the gastrointestinal (GI) tract caused by the conditioning regimen, which enables translocation of intestinal flora and causes a T lymphocyte-mediated/cytokine-driven inflammatory syndrome that is bolstered by macrophages, neutrophils, and dendritic cells. Clinical manifestations primarily affect skin, GI tract, and liver. Greater severity of aGVHD is associated with an increased risk of mortality, but despite routine administration of prophylaxis, clinically significant aGVHD develops in up to one-half of transplant recipients, especially when the donor and recipient have greater immunologic differences.
●Chronic GVHD – cGVHD is a major cause of late morbidity and mortality after allogeneic HCT. It generally presents ≥100 days after transplantation with fibrosis and chronic inflammation of the skin, lungs, GI tract, and soft tissues; the protean manifestations of cGVHD resemble immune-mediated disorders. cGVHD has a complex pathophysiology that involves early inflammation from the conditioning regimen, activation of donor T cells, and injury of vascular endothelium that facilitates migration of donor-derived T and B lymphocytes into host tissues. Aberrant repair mechanisms foster activation of fibroblasts, collagen deposition, and fibrosis that can lead to irreversible end-organ injury and dysfunction. More than one-third of patients who undergo allogeneic HCT develop clinically significant cGVHD, but the incidence and severity vary according to the degree of immunologic mismatch, transplant conditions, and other risk factors.
The immunologic mechanisms that cause GVHD are also responsible for the graft-versus-tumor (GVT) effect, which is important for eradicating an underling malignancy. Treatments that limit GVHD can diminish the GVT effect and may increase the likelihood of recurrent disease. As such, the application of GVHD prophylaxis must balance the benefit of reducing GVHD with the potential harm of decreasing the GVT effect.
●Methods of prophylaxis – Prophylaxis for GVHD focuses on inhibition of donor T cells using pharmacologic agents, with or without in vivo or ex vivo T cell depletion.
•Pharmacologic inhibition – The mainstay of pharmacologic GVHD prophylaxis is a combination of a calcineurin inhibitor (CNI; ie, cyclosporine or tacrolimus) plus an antimetabolite (eg, methotrexate or mycophenolate mofetil [MMF]). (See 'Immunosuppressive backbone' below.)
In many cases, other agents may be added to this backbone or substituted, as described below. (See 'Selection of a GVHD prophylaxis regimen' below.)
•T cell depletion (TCD) – T lymphocytes are essential components of GVHD pathophysiology. TCD can be accomplished with in vivo or ex vivo techniques, as described below. (See 'T cell depletion (TCD)' below.)
Preferred approaches for prophylaxis are guided by features of the transplantation technique and risk factors, as described below. (See 'GVHD prophylaxis' below.)
GVHD PROPHYLAXIS — Preferred approaches for GVHD prophylaxis vary with transplant conditions (eg, degree of immunologic match, conditioning regimen, graft source, and other factors), institutional practice, and clinical trial protocol [2]. Nevertheless, each institution must have guidelines for the prevention and management of GVHD that are approved by international accrediting organizations for transplant centers (Joint Accreditation Committee for ISCT Europe and EBMT [JACIE] and Foundation for the Accreditation of Cell Therapy [FACT]). Guidelines for GVHD prophylaxis have been proposed by the European Group for Blood and Marrow Transplantation (EBMT) and European LeukemiaNet (table 1) [3]. Our approach is generally consistent with these guidelines.
Immunosuppressive backbone — For GVHD prophylaxis, we suggest a calcineurin inhibitor (CNI) plus an antimetabolite, rather than a CNI alone or an antimetabolite alone. This suggestion is based on superior survival and less acute GVHD (aGVHD) with the combination, compared with either agent alone.
Most often, either tacrolimus (Tac) or cyclosporine (CsA) is the CNI, while the antimetabolite is either methotrexate (MTX) or mycophenolate mofetil (MMF) (table 1).
Evidence that supports the use of a combination of a CNI plus an antimetabolite include:
●CNI plus an antimetabolite is superior to a CNI alone – The combination of a CNI plus an antimetabolite achieves superior survival and a lower incidence of aGVHD, compared with a CNI alone.
A phase 3 trial reported that overall survival (OS) was higher and the incidence of aGVHD was lower with CsA plus MTX, compared with CsA alone [4]. The trial of 93 patients undergoing myeloablative conditioning (MAC) allogeneic hematopoietic cell transplantation (HCT) using matched sibling donor (MSD) grafts reported that 15-month OS was higher with CsA/MTX (80 versus 55 percent) compared with CsA alone; this difference corresponds to a 2.4-fold relative risk (RR) for death with CsA alone. CsA/MTX was also associated with less grade ≥2 aGVHD (33 versus 54 percent); the RR for CsA alone was 2.06 (95% CI 1.08-3.88); there was no grade 4 aGVHD with CsA/MTX, compared with grade 4 aGVHD in seven patients who received CsA alone.
●CNI plus an antimetabolite is superior to an antimetabolite alone – The combination a CNI plus an antimetabolite is associated with less aGVHD and a trend toward improved survival, compared with an antimetabolite alone.
A trial that randomly assigned 46 patients undergoing MAC HCT to CsA/MTX versus MTX alone reported better outcomes with the combination regimen [5]. Compared with MTX alone, there was a trend toward improved two-year OS with the combination (82 versus 60 percent) and there was less grade 2-4 aGVHD (18 versus 53 percent), fewer infections, and similar rate of chronic GVHD (cGVHD). No patients given CsA/MTX prophylaxis developed grade 3-4 aGVHD, while MTX alone was associated with grade 3 aGVHD in three patients and grade 4 aGVHD in six patients. One patient in each trial arm did not successfully engraft. With longer follow-up (three to six years), neither four-year OS (73 percent with CsA/MTX versus 58 percent with MTX alone) nor incidence of cGVHD (58 versus 38 percent, respectively) differed significantly between trial arms [6].
Other studies also support the conclusion that a combination of a CNI plus an antimetabolite is superior to monotherapy for GVHD prophylaxis [7-10].
Specific prophylaxis regimens vary according to the degree of immunologic match, conditioning regimen, graft source, and recipient factors, as follows:
●Immunologic match:
•Matched sibling (related) donor – (See 'Matched sibling donor (MSD)' below.)
•Matched unrelated donor – (See 'Matched unrelated donor (MUD)' below.)
•Haploidentical (haplo) donor – (See 'Haploidentical donor' below.)
•Mismatched unrelated donor or umbilical cord blood – (See 'Other alternative donors' below.)
●Conditioning regimen:
•Myeloablative conditioning – (See 'Myeloablative conditioning (MAC)' below.)
•Non-MAC regimens (ie, reduced intensity conditioning [RIC] or non-myeloablative [NMA] conditioning) – (See 'Non-MAC regimens' below.)
●Peripheral blood stem/progenitor cell (PBSPC) graft – (See 'Peripheral blood grafts' below.)
●Other considerations – Special consideration may apply for patients with high-risk for GVHD (eg, greater immunologic mismatch, older age, sex disparity between donor and recipient, prior donor alloimmunization, cytomegalovirus [CMV] serostatus disparity), intolerance for certain agents (eg, use of MTX in patients with limited kidney function or fluid collections), or a need for rapid engraftment (eg, patients with aspergillosis). (See 'Other considerations' below.)
Selection of a GVHD prophylaxis regimen
Immunologic match — Preferences for GVHD prophylaxis vary according to the degree of immunologic match.
Matched sibling donor (MSD) — For HCT with a human leukocyte antigen (HLA)-matched sibling/related donor, we suggest MTX plus either Tac or CsA. This suggestion is based on similar survival with MTX plus either CNI, but Tac is generally associated with a lower incidence of aGVHD, according to randomized trials and retrospective studies [11-14].
The choice of Tac versus CsA is largely guided by institutional preference, as they have similar efficacy and toxicity. Studies that compared Tac versus CsA are described below. (See 'Choice of CNI' below.)
Matched unrelated donor (MUD) — For HCT using an MUD (ie, ≥9/10 or ≥7/8 HLA alleles), we suggest adding antithymocyte globulin (ATG) to MTX plus a CNI, rather than MTX/CNI without ATG. In randomized trials, survival was similar with or without added ATG, but patients who received ATG had a lower incidence of cGVHD and required less immunosuppressive therapy.
Administration of ATG is discussed below. (See 'Antithymocyte globulin (ATG)' below.)
Trials of GVHD prophylaxis using MUD grafts include:
●Addition of ATG to standard prophylaxis for MUD grafts – Addition of ATG to standard prophylaxis (ie, CNI plus an antimetabolite) does not affect OS, but it reduces cGVHD and enables more patients to discontinue immunosuppressive drugs.
•A systematic review and meta-analysis of six randomized trials (845 total patients) concluded that including ATG in the GVHD prophylaxis regimen decreased GVHD and enabled more patients to discontinue immunosuppressive drugs, but it did not improve survival [15]. There was no significant difference in rates of OS, engraftment, relapse, or nonrelapse mortality (NRM), based on use of ATG. However, ATG was associated with less aGVHD (RR 0.75 [95% CI 0.65-0.88]); subset analysis noted this association for grade 2-4 and grade 3-4 aGVHD and with both bone marrow (BM) and peripheral blood (PB) graft sources. ATG use was also associated with less cGVHD (RR 0.54 [95% CI 0.44-0.66]) and a lower incidence of extensive cGVHD (RR 0.32 [95% CI 0.22-0.45]) with both BM and PB graft sources. There was more reactivation of CMV and Epstein-Barr virus (EBV) with ATG, but no difference in bacterial infections, rate of engraftment, or post-transplant lymphoproliferative disorder (PTLD). ATG was associated with modest delays in neutrophil and platelet engraftment (median of 2.7 and 7.5 days, respectively).
•An earlier Cochrane meta-analysis of six randomized controlled trials (568 participants) also reported that ATG did not affect OS, but it was associated with less treatment-requiring or grade 2-4 aGVHD (RR 0.68 [95% CI 0.55-0.85]) [16].
•Examples of phase 3 trials that evaluated outcomes with ATG include:
-A multicenter phase 3 trial randomly assigned 161 patients to CsA/MTX, with or without ATG, for MAC HCT in acute leukemia [17]. With median follow-up of two years, ATG resulted in less cGVHD (32 versus 69 percent) and a higher percentage of patients who could discontinue CsA (91 versus 39 percent). The two cohorts had similar rates of OS, relapse-free survival (RFS), aGVHD, reactivation of CMV and EBV, and there were no cases of PTLD.
-In a multicenter trial of 203 patients undergoing MAC or non-MAC HCT from an unrelated donor (UD) for a hematologic malignancy, patients assigned to ATG were more likely to be free from immunosuppressive treatment at 12 months (37 versus 16 percent) and had less cGVHD, but there was no difference between trial arms in OS, NRM, relapse, graft failure, or severity of aGVHD [18]. EBV reactivation was more common with ATG (20 versus 2 patients) and one patient who received ATG died from EBV reactivation.
-Similar findings were reported in another phase 3 trial; longer-term follow-up noted that the lower incidence of cGVHD and less need for immunosuppressive therapy that persisted after eight years [19,20].
●Survival is similar with Tac versus CsA for MUD grafts – In a phase 3 trial of 180 patients receiving MUD grafts, MTX plus Tac versus CsA did not differ regarding OS, RFS, toxicity, infections, or leukemia relapse, but Tac was associated with less grade 2-4 aGVHD (56 versus 74 percent) [11].
Haploidentical donor — For haploidentical (haplo) HCT, we suggest adding post-transplant cyclophosphamide (PTCy) to a CNI plus MMF, rather than other immunosuppressive regimens. This suggestion is based on less GVHD, fewer relapses, less NRM, and a trend toward improved survival with PTCy.
Treatment with PTCy is described below. (See 'Post-transplant cyclophosphamide (PTCy)' below.)
PTCy was superior to antithymocyte globulin (ATG) in a retrospective cooperative group study of haplo HCT in patients with acute myeloid leukemia (AML) in remission [21]. The study compared 193 patients who received PTCy-based prophylaxis versus 115 who were treated with ATG-based prophylaxis; approximately half of patients received MAC HCT. Patients in the PTCy group also received MMF and CsA (62 percent) or MMF and Tac (28 percent), while those in the ATG group also received three to five other drugs, including CsA, MTX, and/or basiliximab. Compared with ATG-based prophylaxis, patients treated with PTCy had less grade 3-4 aGVHD (5 versus 12 percent, respectively) and in multivariate analysis, PTCy was also associated with superior GVHD-free/RFS (GRFS; hazard ratio [HR] 1.5 [95% CI 1.0-2.0]), leukemia-free survival (LFS; HR 1.5 [95% CI 1.0-2.1]), and NRM (HR 1.8 [95% CI 1.1-2.9]); there was a trend toward improved two-year OS (HR 1.43 [95% CI 0.98-2.09]) with PTCy.
Studies with PTCy for haplo HCT using both MAC and non-MAC (RIC or NMA conditioning) regimens include:
●Non-MAC (ie, RIC or NMA):
•A retrospective analysis of 271 adults (50 to 75 years) who received PTCy for non-MAC, T cell-replete haplo HCT reported aGVHD and cGVHD in 33 and 10 percent of patients, respectively; 3 percent grade 3-4 GVHD; and 8 percent NRM at six months [22]. Rates of severe GVHD, NRM, progression-free survival (PFS), and OS did not differ by age.
•A single-institution study of 67 patients with hematologic malignancies who were treated with NMA conditioning, haplo grafts, and PTCy-based prophylaxis reported 36 percent two-year OS, 26 percent two-year event-free survival (EFS), 15 percent one-year NRM, and 51 percent one-year relapse [23]. Other outcomes included 87 percent engraftment, 34 percent grade 2-4 aGVHD, and 6 percent grade 3-4 aGVHD, and low rates of cGVHD.
●MAC – Historically, haplo grafts using MAC HCT without PTCy were associated with excessive NRM and GVHD. However, since the inclusion of PTCy in the prophylaxis regimen, MAC HCT using haplo grafts is associated with rates of GVHD that are similar to those with standard prophylaxis (CNI plus an antimetabolite) with MSD HCT:
•A multi-institution study of PTCy for MAC HCT (using busulfan and fludarabine) in 92 adults (45 with related donors and 47 with unrelated donors) reported two-year rates of OS and disease-free survival (DFS) were 67 and 62 percent, respectively; 51 percent grade 2-4 aGVHD; and 15 percent grade 3-4 aGVHD [24]. Among patients who survived >2 years, the cumulative incidence of cGVHD was 14 percent.
•In a single-institution study, transplantation of 30 patients using total body irradiation (TBI)-based MAC reported 78 percent two-year OS, 73 percent DFS, 3 percent NRM, and 24 percent relapse [25]. Other outcomes included 43 percent grade 2-4 aGVHD, 23 percent grade 3-4 aGVHD, and 56 percent cGVHD (severe in 10 percent). These outcomes were similar to outcomes in contemporaneous patients who received MAC HCT using MUD grafts.
•A registry study of 92 patients who underwent haplo HCT (77 percent MAC) reported 52 percent four-year OS, 18 percent day 1000 treatment-related mortality (TRM), and 18 percent relapse rate [26]. Other outcomes included 14 percent grade 2-4 aGVHD, 4 percent grade 3-4 aGVHD, and 15 percent cGVHD. The proportion of patients who were off CsA by day 180 was 68 percent and was 81 percent after one year. OS with haplo HCT was similar to that of 176 patients who received MSD grafts, while the incidence of grade 2-4 aGVHD was significantly lower with haplo HCT compared with MSD HCT.
Tac/MMF/PTCy is currently being compared with Tac/MTX in a phase 3 trial.
Other alternative donors — There is no consensus GVHD prophylaxis regimen for other alternative donors, such as mismatched unrelated donor (mMUD) or umbilical cord blood (UCB) grafts. However, the risk for GVHD increases with greater immunologic differences, so we consider donor-graft discordance, conditioning regimen, and risk of relapse in choosing a prophylaxis regimen.
Conditioning regimen
Myeloablative conditioning (MAC) — For MAC allogeneic HCT, we suggest GVHD prophylaxis using a CNI plus MTX, rather than a CNI plus MMF, based on similar survival but less severe GVHD, according to meta-analyses and retrospective studies. However, some institutions favor MMF to avoid the higher rate of mucositis associated with MTX.
Treatment with MTX is described below. (See 'Methotrexate (MTX)' below.)
For patients with a contraindication to MTX or who need rapid engraftment (eg, patients with aspergillosis), MMF can be used instead of methotrexate. (See 'Mycophenolate mofetil (MMF)' below.)
Meta-analyses reported that, compared with MTX, MMF is associated with similar survival, faster engraftment, less mucositis, but more grade 3-4 aGVHD:
●A meta-analysis of three randomized controlled trials (RCT; 174 patients analyzed) reported no difference in survival, relapse, aGVHD, cGVHD, or neutrophil engraftment between MMF and MTX [27]. The analysis also noted that there was low-quality evidence that MMF was associated with faster platelet engraftment, a lower incidence of severe mucositis, and less need for pain control and parenteral nutrition.
●A meta-analysis that included RCTs and other studies also reported no difference in survival outcomes, but MMF was associated with less mucositis (RR 0.35 [95% CI 0.25–0.49]) and faster engraftment (mean difference 3.6 days), but more grade 3-4 aGVHD (RR 1.61 [95% CI 1.18–2.30]) [28].
Other informative studies that compared MTX and MMF for MAC HCT include:
●A phase 3 trial in patients undergoing MAC HCT with MSD grafts was terminated early because CsA plus MMF was associated with faster hematopoietic engraftment and less mucositis than CsA plus MTX [29]. Compared with 19 patients who received CsA plus MTX, the 21 patients who received MMF had a lower incidence of severe mucositis (21 versus 65 percent) and a shorter time to neutrophil engraftment (11 versus 18 days); day 100 OS and incidence of aGVHD were similar in both arms.
●Retrospective analysis of patients who underwent MAC HCT using MSD grafts reported no difference in survival with CsA/MMF versus CsA/MTX, but MMF was associated with more rapid marrow recovery, shorter hospital stays, and more grade 3-4 aGVHD [30]. There were no significant differences in OS, NRM, or relapse, but MMF was associated with faster neutrophil recovery (median 11 versus 19 days) and platelet recovery (median 19 versus 25 days), less severe mucositis (19 versus 53 percent), and shorter length of hospital stay (median 25 versus 36 days). However, multivariate analysis detected increased risk of grade 3-4 aGVHD (HR 2.92 [95% CI 1.2-7.15]), but there were no differences in aGVHD or grade 2-4 cGVHD.
●A randomized phase 2 study reported no difference in OS, RFS, NRM, relapse, or time to neutrophil recovery among 47 patients treated with Tac plus MTX versus 42 patients treated with Tac plus MMF [31]. Patients who received MMF were less likely to experience severe mucositis, require narcotic analgesia or parenteral nutrition, and had earlier hospital discharge. Grade 2-4 aGVHD at day 100 was similar, but there was more grade 3-4 aGVHD with MMF (19 versus 4 percent); this was predominantly seen with MUD grafts (26 versus 4 percent), as the effects with MSD grafts did not differ significantly (11 versus 4 percent). Platelet recovery was earlier in patients treated with MMF.
Non-MAC regimens — For adults undergoing NMA or RIC allogeneic HCT, we suggest a CNI plus MMF, rather than a CNI plus MTX. No phase 3 trials have directly compared MMF versus MTX in this setting, but a CNI plus MMF has been widely adopted for prophylaxis with NMA or RIC HCT, primarily because it is associated with faster hematologic recovery and less mucositis than MTX and because GVHD is less of a concern with non-MAC regimens compared with MAC HCT.
PTCy is also being evaluated for non-MAC HCT, as discussed below. (See 'Post-transplant cyclophosphamide (PTCy)' below.)
Studies that compared MTX and MMF for non-MAC HCT include:
●A meta-analysis reported no differences in OS, DFS, or cGVHD among 1564 adults who underwent RIC allogeneic HCT for leukemia or myelodysplastic syndrome (MDS) [32]. The study compared four cohorts of patients based on GVHD prophylaxis: MMF/Tac, MMF/CsA, MTX/Tac, and MTX/CsA. For recipients of MSD grafts, there were no differences in outcomes, including rates of GVHD. However, for recipients of MUD grafts, compared with MTX/Tac, patients treated with MMF/CsA had a higher rate of NRM (HR 1.48) and a trend toward lower relapse rate (HR 0.53), which did not affect OS. MMF was also associated with more grade 2-4 aGVHD (RR 1.78) and grade 3-4 aGVHD (RR 1.93) but faster neutrophil recovery.
●Meta-analyses (which included both non-MAC and MAC HCT) that described similar survival, faster engraftment, less mucositis, but more grade 3-4 aGVHD with MMF compared with MTX are described above [27,28]. (See 'Myeloablative conditioning (MAC)' above.)
●A multicenter randomized phase 2 study reported that for patients who underwent RIC HCT, those who received PTCy had superior GRFS compared with other GVHD prophylaxis regimens [33]. Patients were randomly assigned (1:1:1) to: Tac/MMF/PTCy, Tac/MTX/bortezomib, or Tac/MMF/maraviroc; each group was compared separately to a control group (a contemporary cohort of 224 patients who met the same eligibility criteria but were treated with Tac/MTX at other institutions). Compared with control patients, those who received PTCy had superior GRFS (HR 0.72 [90% CI 0.54-0.94]), while GRFS for the other treatments did not differ from the control group. Adverse events (AEs) in patients treated with PTCy were similar to the other regimens and were primarily hematologic; there were 13 percent grade 3 and 73 percent grade 4 AEs. Other studies of PTCy for non-haplo HCT are described below. (See 'Post-transplant cyclophosphamide (PTCy)' below.)
Peripheral blood grafts — For transplantation using a PB graft source, we consider treatment with either Tac or CsA plus either MTX or MMF acceptable.
In general, there are no significant differences in outcomes based on the immunosuppressive regimen, although MMF may be associated with more GVHD but faster hematologic recovery:
●A registry study compared outcomes in patients who underwent RIC HCT using MSD or UD PB grafts for leukemia or MDS from 2000 to 2013; the study divided the 1564 adult patients into four cohorts: MMF/Tac, MMF/CsA, MTX/Tac, and MTX/CsA [32]. There was no difference in OS, DFS, or cGVHD among any of the GVHD prophylaxis regimens. However, for the UD group, MMF/CsA was associated with more NRM (HR 1.48) and increased risk of grade 2-4 aGVHD (RR 1.78) and grade 3-4 aGVHD (RR 1.93), compared with MTX/Tac.
●A retrospective study compared outcomes in 456 patients who received MTX plus either Tac or CsA after MAC HCT using granulocyte colony-stimulating factor (G-CSF)-mobilized PB grafts from MSD or MUD donors for hematologic malignancies [12]. Multivariate analysis reported no differences between Tac and CsA for OS, DFS, relapse, NRM, aGVHD, or cGVHD.
Other considerations — Other considerations in choosing a GVHD prophylaxis regimen include:
●Limited kidney function or other contraindication to MTX – For patients with limited renal function or a fluid collection that precludes use of MTX, MMF and/or sirolimus can be considered for prophylaxis. (See 'Mycophenolate mofetil (MMF)' below and 'Sirolimus' below.)
●Need for rapid engraftment – For patients in whom rapid engraftment is a high priority (eg, patients with aspergillosis), MMF can be used instead of MTX. (See 'Mycophenolate mofetil (MMF)' below.)
CALCINEURIN INHIBITORS (CNI) — The primary pharmacologic strategy to prevent GVHD is inhibition of the cytoplasmic enzyme, calcineurin. T cell activation is highly dependent on calcineurin and calcium-dependent signal transduction pathways downstream of the T cell receptor (TCR) (figure 1). The CNIs, tacrolimus (Tac) and cyclosporine (CsA), are structurally distinct, but they have similar mechanisms of action:
●TCR signaling – In T lymphocytes, the activation cascade downstream from the TCR includes binding of calcium to calmodulin, which in turn, leads to binding of calmodulin to calcineurin. Activated calcineurin dephosphorylates the cytoplasmic unit of nuclear factor of activated T cells (NF-AT), which enables translocation of NF-AT from the cytoplasm into the nucleus and renders it competent to activate transcription of genes that encode key lymphocyte cytokines, including interleukin (IL)-2, tumor necrosis factor (TNF)-alpha, IL-3, IL-4, CD40L, granulocyte-macrophage colony-stimulating factor, and interferon-gamma [34,35].
●Mechanism of action – Both CsA and Tac bind cyclophilin-related proteins (ie, cyclophilin and FK binding proteins [FKBP], respectively). The CNI-cyclophilin complex binds and inhibits calcineurin, which blocks nuclear translocation and transcriptional activation of NF-AT, and proliferation of T lymphocytes [34,35].
Additional details of CNI activity are presented separately. (See "Pharmacology of cyclosporine and tacrolimus", section on 'Mechanism of action'.)
Choice of CNI — Tac and CsA have similar clinical properties, efficacy, and toxicity. Selection of a CNI for GVHD prophylaxis in allogeneic hematopoietic cell transplantation (HCT) is largely guided by institutional preference.
Common adverse effects (AEs) of CNIs include hypomagnesemia, hyperkalemia, hypertension, and nephrotoxicity. Rarely, these CNIs are associated with life-threatening complications of transplant-associated thrombotic microangiopathy (TA-TMA) and neurotoxicity. (See "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS".)
Monitoring and management of CNI levels in blood are important for optimal efficacy and toxicity; other nephrotoxic drugs should be avoided, when possible, to enable delivery of the CNI at the target dose. For patients with limited kidney function or if nephrotoxicity occurs with a CNI, alternative agents can be employed. (See 'Other considerations' above.)
Phase 3 trials and registry studies that compared prophylaxis with an antimetabolite plus either Tac or CsA reported similar outcomes, but Tac is generally associated with less acute GVHD (aGVHD).
●In a phase 3 trial of HCT using unrelated donor grafts, the outcomes, efficacy, and toxicity were similar when patients were treated with methotrexate (MTX) in combination with either Tac or CsA [13]. CNI doses were adjusted to achieve Tac (15 ng/mL) or CsA (500 ng/mL); >90 percent of patients achieved serum concentrations of 80 to 120 percent of the target concentration in the first four weeks after HCT. Overall survival (OS), disease-free survival (DFS), incidence of relapse, nonrelapse mortality (NRM), and organ toxicities were equivalent with either CNI. Rates of aGVHD were also similar; Tac was associated with 33 percent grade 2-4 aGVHD and 9 percent grade 3-4 aGVHD, compared with 40 percent and 8 percent, respectively, with CsA.
●In a phase 3 trial of 180 patients undergoing allogeneic HCT with matched unrelated donor (MUD) grafts, outcomes did not differ with MTX/Tac versus MTX/CsA [11]. There were no significant differences in OS, relapse-free survival (RFS), toxicity, infections, or leukemia relapse. However, Tac was associated with less grade 2-4 aGVHD (56 versus 74 percent) and less need for glucocorticoids to manage GVHD.
●A phase 3 trial of 45 patients with hematologic malignancies who received matched sibling donor (MSD) grafts reported that MTX/Tac was associated with less aGVHD, but MTX/CsA was associated with better survival; however, interpretation of the outcomes is limited because more patients in the MTX/CsA had advanced disease [14]. Compared with CsA, patients who received Tac had less grade 2-4 GVHD (32 versus 44 percent), but a similar incidence of grade 3-4 aGVHD (17 versus 13 percent); incidence of chronic GVHD (cGVHD) was similar (56 versus 49 percent), but CsA was associated with more clinically extensive cGVHD. Patients treated with MTX/CsA had better two-year OS (57 versus 47 percent) and two-year DFS (50 versus 41 percent), but these differences were largely due to inferior outcomes in patients with advanced disease; OS was 25 percent with Tac versus 42 percent with CsA and DFS was 20 percent with Tac versus 28 percent with CsA.
●A retrospective study of 456 patients reported no differences in OS, DFS, NRM, relapse, aGVHD, or cGVHD for patients who received Tac versus CsA [12].
Cyclosporine (CsA) — CsA is a neutral hydrophobic cyclic peptide composed of 11 amino acids that was originally extracted from soil fungi. (See "Pharmacology of cyclosporine and tacrolimus".)
●Administration – The initial dose of CsA is 3 mg/kg/day, given as a short intravenous (IV) bolus infusion in two daily doses beginning on day –2 or –1 (table 1). It is usually given IV for the first several weeks to allow healing of the oral mucosa and gastrointestinal (GI) tract after conditioning therapy. CsA prophylaxis is continued for six months in the absence of GVHD (table 1).
CsA administration is changed to the oral route when eating and drinking occur without significant difficulties. CsA is available as a solution and as soft gelatin capsules. CsA is administered in two daily oral doses; the first oral dose is twice the IV dose. Food (especially high fat content foods) may influence CsA bioavailability.
●Monitoring – CsA levels are monitored to guide dosing until approximately day +100. Trough levels are measured 12 hours after a dose. Many drugs can increase or decrease CsA levels. CsA can also leach into plastic catheters, so it is important that the lumen used to measure plasma levels is separate from the lumen through which cyclosporine is infused; otherwise, serum levels are likely to be falsely elevated.
•Target levels – The target concentration for CsA varies over time from transplantation:
-First three to four weeks after HCT – A target concentration of 200 to 300 mg/mL is used during the first three to four weeks.
-Subsequent monitoring – If there is no GVHD after three to four weeks, the target concentration is decreased to 100 to 200 mg/mL until three months after transplantation; CsA is then tapered further, as described below.
•Dose adjustment – The dose of CsA should be increased if levels are below the therapeutic threshold. If levels are modestly elevated in an asymptomatic patient, no dose adjustment is needed. However, if levels are >2 times the upper limit of the therapeutic range, the dose should be reduced to avoid seizures and other neurotoxicity.
●Tapering – Tapering of CsA varies according to the degree of immunologic match and recipient age; a slower taper may be preferable in older patients.
The dose of CsA should not be tapered if there are signs of aGVHD or if cGVHD exceeding mild skin disease develops. If GVHD develops, the higher target level (ie, 200 to 300 mg/mL) is maintained.
•MSD graft – For an MSD graft, if the patient has no evidence of GVHD, CsA may be tapered at approximately day +90 and discontinued by the end of six months.
•Other graft donors – For MUD grafts or mismatched related donor grafts, CsA may need to be continued for longer, sometimes even for years.
It is important to monitor patients carefully for development of cGVHD during the taper; the onset of cGVHD can be fulminant at times. The same concern applies to patients who receive prednisone (either as a component of the prophylaxis regimen or for the treatment of aGVHD). (See "Clinical manifestations and diagnosis of chronic graft-versus-host disease".)
●Adverse effects (AEs) – The major clinical toxicities of CsA are renal insufficiency and elevated bilirubin levels; both AEs are dose-related.
CsA is highly lipophilic and it is extensively metabolized with subsequent biliary excretion; there is limited urinary excretion. Some of the >15 metabolites of CsA have immunosuppressive activity, while others are nephrotoxic. The clinical efficacy and interactions between CsA and other drugs may be confounded by its metabolites.
•Kidney – Acute nephrotoxicity is due to vasoconstriction and ischemia of the afferent arterioles in the kidney [36]. These acute changes are reversible, but they can lead to irreversible interstitial injury with glomerular thrombosis, causing permanent azotemia. It is important to distinguish CsA-associated nephrotoxicity from AEs of other drugs (eg, amphotericin, MTX, aminoglycosides) and from GVHD, itself. (See "Pharmacology of cyclosporine and tacrolimus", section on 'Side effects'.)
•Liver – Cytochrome p450 IIIA enzymes (eg, HLp and PCN1) can inhibit or induce CsA metabolism. This interaction is important to keep in mind, especially when patients are treated with antifungal medications (eg, voriconazole), which may significantly increase CsA levels.
•Neurotoxicity – CsA can cause neurotoxicity, which may be more prominent with concomitant hypomagnesemia or hypocholesterolemia. Neurologic AEs can involve all levels of the neuraxis, including cortical blindness. Associated abnormalities include elevated cerebrospinal fluid protein and pleocytosis, electroencephalographic abnormalities, and central and extrapontine myelinolysis with imaging. In most patients, these events were reversible with dose reduction or withdrawal of CsA.
•Other AEs – Other AEs include hypertension, hyperglycemia, headaches, and hirsutism. Rarely, gum hypertrophy, brittle nails, acne, nausea, and vomiting develop.
Calcium channel blockers are the drugs of choice if hypertension develops. However, some such agents (eg, verapamil) can affect CsA levels, which may require adjusting the CsA dose. (See "Pharmacology of cyclosporine and tacrolimus", section on 'Side effects'.)
Tacrolimus (Tac) — Tacrolimus (FK506) is a highly lipophilic macrolide that was originally extracted from a soil fungus. It is structurally distinct from CsA, which is a cyclic peptide, but its mechanism of action (immunosuppression through inhibition of signaling via the T cell receptor) and efficacy are similar to CsA.
●Administration – Prophylaxis is initiated with Tac at 0.03 to 0.04 mg/kg/day by continuous infusion. The target trough range for Tac is 15 ng/mL. Tacrolimus can leach into plastic catheters, so it is important that the lumen used to measure serum levels is separate from the lumen through which tacrolimus is infused; otherwise serum levels are likely to be falsely elevated.
Tac administration can be converted to the oral route (0.15 mg/kg/day, in two divided doses) when eating and drinking occur without significant difficulties.
Tac is almost completely metabolized in the liver (through demethylation and hydroxylation) prior to its elimination. The half-life is approximately nine hours, but it is longer in patients with liver dysfunction. GI absorption of Tac does not seem to depend on the presence of bile salts, which distinguishes it from CsA.
●Toxicity – Tac can cause neurotoxicity, renal toxicity, hyperkalemia, and hyperglycemia. The AEs are similar to those associated with CsA, but there is less hypertension associated with Tac [37]. Hypertension should be managed with calcium channel blockers (as described for CsA above). (See 'Cyclosporine (CsA)' above.)
Central nervous system (CNS) effects can occur, including headaches, tremors, paresthesias, photophobia, mental status changes, and coma [38]. Other AEs include dyspnea, musculoskeletal pain, itching, GI complaints (eg, anorexia, nausea, vomiting, abdominal pains), and fatigue.
Tac has also been associated with TMA. (See "Drug-induced thrombotic microangiopathy (DITMA)", section on 'Immunosuppressive agents'.)
ANTIMETABOLITES
Choice of antimetabolite — The choice of methotrexate (MTX) versus mycophenolate mofetil (MMF) as an antimetabolite varies with the conditioning regimen, risk factors for GVHD, and institutional preference.
MTX and MMF are generally associated with similar survival outcomes, but MTX is associated with less acute GVHD (aGVHD), while MMF is associated with faster engraftment and less mucositis, as described above. (See 'Conditioning regimen' above and 'Other considerations' above.)
Methotrexate (MTX) — MTX inhibits dihydrofolate reductase (DHFR), the intracellular enzyme that converts folic acid to tetrahydrofolates, which are responsible for transport of single carbon groups needed for purine and thymidylate synthesis. MTX induces immune tolerance by blocking proliferation of activated T cells.
●Precautions – Patients with a known third-space fluid collection (eg, ascites, pleural effusions) should not receive MTX. The MTX dose is usually reduced in patients who have hyperbilirubinemia, severe mucositis, or renal insufficiency; care should be exercised to avoid other factors that might predispose to severe MTX toxicity.
Patients should be well hydrated and have their urine alkalinized with sodium bicarbonate to aid MTX excretion and minimize its precipitation in renal tubules, which can lead to acute renal failure. This can be achieved by the administration of roughly 3 liters per day of dextrose in water to which 44 to 66 mEq of sodium bicarbonate has been added. MTX should not be administered until the urine pH is ≥8.0.
Ascites and pleural effusions can accumulate high levels of MTX that slowly leach into the circulation long after the initial dose; this can result in prolonged drug elimination and severe delayed toxicity, particularly if renal function is impaired. MMF can be considered as an alternative antimetabolite for patients with impaired kidney function or fluid collections. (See 'Mycophenolate mofetil (MMF)' below.)
Prevention and management of MTX toxicity is discussed in greater detail separately. (See "Therapeutic use and toxicity of high-dose methotrexate", section on 'Prevention and management of high-dose methotrexate toxicity'.)
●Administration:
•Initial dose – The first dose of MTX is 15 mg/m2, given intravenously on day +1, followed 24 hours later by leucovorin rescue (15 mg orally every six hours for three doses) (table 1); leucovorin can be given intravenously, instead of orally, if there is severe mucositis.
•Subsequent doses – Three additional doses of MTX (10 mg/m2) are given on days +3, +6 and +11. Leucovorin rescue is given (15 mg orally every six hours for four doses) after each dose of MTX. Full target doses of MTX may not be achievable in patients who already have grade 3-4 mucositis (eg, from the conditioning regimen). The day +11 dose should be decreased or omitted in case of any grade ≥2 adverse effects (AEs).
●Adverse effects – The most common AEs when MTX is used for GVHD prophylaxis are renal, hepatic, and gastrointestinal toxicity (especially severe mucositis), delayed hematopoietic recovery, and organ toxicity [31,39-41].
Patients commonly have an elevation in creatinine and bilirubin levels following MTX administration. As these are also the AEs associated with CNIs, dose attenuation of MTX, Tac, or CsA may be necessary. (See "Hepatotoxicity associated with chronic low-dose methotrexate for nonmalignant disease" and "Methotrexate-induced lung injury" and "Major side effects of low-dose methotrexate", section on 'Common toxicities'.)
Mycophenolate mofetil (MMF) — MMF is an ester prodrug of mycophenolic acid and a known inhibitor of inosine monophosphate dehydrogenase. MMF selectively inhibits de novo purine biosynthesis in activated lymphocytes; it has less effect on neutrophils and this selectivity seems to spare the higher risk of acute infections.
MMF is generally used for non-myeloablative conditioning (MAC) hematopoietic cell transplantation (HCT), for patients receiving MAC with a contraindication to MTX (eg, fluid collection), or for patients who need rapid engraftment (eg, those with aspergillosis) [42]. (See 'Non-MAC regimens' above and 'Other considerations' above.)
●Administration – MMF dose is 1 gram orally twice daily, beginning on day +1; the dose can be increased to 1 gram three times daily [43]. MMF dose should be adjusted according to toxicity.
For children, the MMF dose is 30 mg/kg/day, given orally in two doses (table 1).
•Matched sibling donor (MSD) grafts – For MSD grafts, MMF prophylaxis is for one month.
•Other donor grafts – For other donor sources, (eg, mismatched sibling or unrelated donors), MMF is given for three months.
●Adverse effects – Hematologic AEs (primarily neutropenia) are the most common toxicity, but gastrointestinal AEs are reported occasionally. Doses up to 1.5 gram twice daily had no nephrotoxicity, myelosuppression, or other serious side effects in patients with rheumatoid arthritis. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Mycophenolate mofetil'.)
Rarely, MMF has rarely been associated with progressive multifocal leukoencephalopathy (PML) and opportunistic infections. (See "Progressive multifocal leukoencephalopathy (PML): Epidemiology, clinical manifestations, and diagnosis".)
T CELL DEPLETION (TCD) — TCD can be via in vivo or ex vivo methods.
In vivo TCD
Antithymocyte globulin (ATG) — ATG is a polyclonal immune globulin prepared by injecting various cellular preparations into animals. ATG has been used as a component of transplant conditioning regimen (to decrease the incidence of graft rejection) or after allogeneic hematopoietic cell transplantation (HCT) to ameliorate GVHD, especially for graft sources that are associated with increased risk for GVHD (eg, unrelated donors, haploidentical grafts). (See 'Matched unrelated donor (MUD)' above and 'Haploidentical donor' above.)
Rabbit antithymocyte globulin (rATG) is the preferred product for GVHD prophylaxis. It is more efficacious than horse antithymocyte globulin (hATG) and there is less lot-to-lot variation in potency. Use of ATG as a component of GVHD prophylaxis does not confer a survival advantage; it is associated with less acute GVHD (aGVHD) and less need for other medications to control GVHD, but it is associated with an increased risk for post-transplant lymphoproliferative disorders (PTLD).
●Administration – The recommended dose of rATG is dose is 15 mg/kg [44]. While the measurable half-life of the antibody is approximately nine hours, the functional half-life is not known.
Corticosteroids, acetaminophen, H1 and H2 blockers are frequently used to prevent or treat the symptoms associated with the infusion of ATG. (See "Treatment of acute graft-versus-host disease", section on 'Anti-thymocyte globulin (ATG)'.)
If hATG is used, doses are generally in the range of 10 to 30 mg/kg per day.
●Toxicity – Since ATG is both a foreign xenogeneic protein and an antibody, serum sickness can occur after its use. Diagnosis and management of serum sickness are discussed separately. (See "Serum sickness and serum sickness-like reactions".)
The use of ATG after HCT has been associated with an increase in PTLD [45]. (See "Treatment and prevention of post-transplant lymphoproliferative disorders".)
●Outcomes – Addition of ATG to standard prophylaxis is associated with similar survival and less chronic GVHD (cGVHD) but an increased risk for viral reactivation, as described above. (See 'Matched unrelated donor (MUD)' above.)
Post-transplant cyclophosphamide (PTCy) — PTCy can lower the incidence of both aGVHD and cGVHD after allogeneic HCT. The mechanism of action is not certain, but PTCy may deplete or suppress alloreactive T cells in the host and/or support development of regulatory T cell (Treg)-mediated tolerance [46,47]. (See "HLA-haploidentical hematopoietic cell transplantation".)
PTCy is a standard approach for haploidentical (haplo) HCT, but it also has been used for matched sibling donor (MSD) and unrelated donor grafts [48,49]. It is administered after infusion of the graft to remove alloreactive T cells and stimulate Treg recovery [23]. PTCy can induce immune tolerance without causing global myelosuppression, because hematopoietic stem cells express high levels of aldehyde dehydrogenase (ALDH), which confers cellular resistance to cyclophosphamide (Cy) [50].
●Administration – PTCy uses a brief course of high-dose Cy (50 mg/kg per day on days +3 and +4) as a single agent or together with mycophenolate mofetil (MMF) and tacrolimus (Tac) or cyclosporine (CsA).
Cy is not currently approved by the US Food and Drug Administration (FDA) for GVHD prophylaxis.
●Toxicity – Adverse effects (AEs) are related to myelosuppression, nausea/vomiting, and hemorrhagic cystitis. Pulmonary injury is rare, but it can present with early acute pneumonitis or chronic progressive pulmonary fibrosis. Cardiotoxicity may include myocarditis, arrhythmias, and/or congestive heart failure. Cy can cause secondary malignancies and embryo-fetal toxicity. (See "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines", section on 'Cyclophosphamide'.)
When used in high doses, Cy metabolites can cause hemorrhagic cystitis. (See "Chemotherapy and radiation-related hemorrhagic cystitis in cancer patients", section on 'Cyclophosphamide'.)
Drugs that alter cytochrome P450 (eg, alcohol, barbiturates, phenytoin) may accelerate metabolism of Cy into its active metabolites, increasing both pharmacologic and toxic effects of the drug. Conversely, drugs that inhibit cytochrome P450 (eg, corticosteroids, allopurinol, tricyclic antidepressants) may slow conversion of Cy into its metabolites and consequently reduce therapeutic and toxic effects.
Cy is metabolized, primarily in the liver, to 4-OH Cy, acrolein, and the active component, phosphoramide mustard, which forms irreversible intra- and inter-strand cross-linkages at guanine N-7 positions in DNA and leads to apoptosis [51]. Many tissues, including bone marrow, liver, and intestinal epithelium have abundant ALDH, which lessens toxicity on these organs. Cy metabolites are primarily excreted unchanged in the urine; drug dosing should be adjusted in the setting of renal dysfunction [52].
●Outcomes – Studies of PTCy for haploidentical (haplo) HCT are presented above. (See 'Haploidentical donor' above.)
PTCy has also been examined for non-haplo graft sources:
•A multicenter randomized phase 2 study reported that for reduced intensity conditioning (RIC) HCT, patients who received PTCy had superior GVHD-free/relapse-free survival (GRFS) compared with other GVHD prophylaxis regimens [33]. (See 'Non-MAC regimens' above.)
•A retrospective study of PTCy for matched unrelated donors (MUD) reported 60 percent two-year overall survival (OS), 56 percent leukemia-free survival (LFS), 16 to 20 percent nonrelapse mortality (NRM), and relapse in approximately one-quarter of patients [53]. Day 100 grade 2-4 and grade 3-4 aGVHD were 28 percent and 10 percent, respectively, and extensive cGVHD was 21 percent.
•Preliminary results of a phase 3 trial, reported in abstract form, found that PTCy was associated with improved GRFS compared with conventional immunosuppression (CsA plus MMF) for MSD or MUD grafts [54].
Other studies describe PTCy as safe and effective for preventing graft rejection and reducing aGVHD and cGVHD following myeloablative or nonmyeloablative conditioning with T cell-replete grafts from partially mismatched related donors [23,24,55-58].
Ex vivo TCD — Eliminating T lymphocytes ex vivo could reduce the incidence of GVHD. Most such methods have been associated with no improvement in OS, but they may increase graft failure and disease relapse and delay immune reconstitution. There is no standard approach, and use of ex vivo TCD varies among institutions.
●Methods – Methods to deplete T lymphocytes from donor bone marrow include:
•Ex-vivo treatment of the donor bone marrow with monoclonal antibodies with broad reactivity (eg, anti-CD52, anti-CD2, anti-CD3, and anti-CD5 antibodies) or more restricted reactivity (eg, anti-CD8 and anti-CD25).
•Physical separation techniques include density gradients, selective depletion with lectins, treatment with cytotoxic drugs, and the use of anti-T cell sera or monoclonal antibodies.
●Outcomes – In studies comparing TCD with pharmacologic therapy for the prevention of GVHD, TCD was associated with lower rates of severe aGVHD but higher rates of graft failure, relapse, infections, and other complications.
As examples:
•A multicenter phase 3 trial compared TCD (using two different methods) plus CsA versus CsA/MTX in >400 patients receiving unrelated donor bone marrow HCT for hematologic malignancies [59,60]. TCD was associated with less grade 3-4 aGVHD (18 versus 37 percent), but no significant difference in cGVHD (29 versus 34 percent), transplant-related mortality (TRM), or three-year disease-free survival (DFS). Patients TCD grafts had a higher risk of relapse (20 versus 9 percent) and a higher risk of cytomegalovirus infection (28 versus 17 percent).
•In a multicenter phase 2 study, ex vivo CD34+ cell selection was the sole method of GVHD prophylaxis in 44 patients undergoing MSD HCT for acute myeloid leukemia (AML) in first or second remission [61]. When compared with a similar population of 84 patients receiving T cell-replete grafts and treated with calcineurin inhibitor (CNI)-based GVHD prophylaxis, TCD was associated with similar rates of grade 2-4 aGVHD (23 versus 39 percent), lower rates of cGVHD (19 versus 50 percent), and a lower percentage of patients requiring immunosuppression at one year (12 versus 54 percent).
•Depletion of naïve T cells from human leukocyte antigen (HLA)-matched peripheral blood (PB) graft sources was associated with infrequent, mild GVHD in a study of 138 patients who received myeloablative conditioning (MAC) HCT [62]. Donors were treated with granulocyte colony-stimulating factor (G-CSF) for five days and positive selection of CD34+ progenitor cells was followed by depletion of CD45RA+ cells from the CD34-negative fraction. Grade 3 and grade 4 aGVHD were reported in 4 percent and 0 percent of patients, respectively; only two patients required treatment beyond corticosteroids. Three-year cumulative incidence of mild, moderate, and severe cGVHD were 6, 1, and 0 percent, respectively. Other outcomes at three years included OS (77 percent), relapse-free survival (69 percent), and 64 percent GRFS.
OTHER TREATMENTS
Abatacept — Abatacept is a selective inhibitor of T cell costimulation that may reduce acute GVHD (aGVHD) when added to a calcineurin inhibitor (CNI) plus methotrexate (MTX).
A phase 2 study reported that adding abatacept to a CNI/MTX was associated with less aGVHD compared with a CNI/MTX [63]. The study included two strata: a randomized, double-blind, placebo-controlled cohort of 8/8 human leukocyte antigen (HLA)-matched unrelated donors (MUD) and a second cohort in which patients with 7/8 HLA-allele MUD grafts who received abatacept plus a CNI/MTX were compared with historical control patients treated with a CNI/MTX:
●In the 8/8 HLA-matched cohort, abatacept/CNI/MTX was associated with a lower incidence of day 100 grade 2-4 aGVHD (43 versus 62 percent; HR 0.53 [95% CI 0.39-0.71]), but there was no difference in two-year overall survival (OS), relapse-free survival (RFS), nonrelapse mortality (NRM), grade 3-4 aGVHD, or chronic GVHD (cGVHD), compared with patients randomly assigned to CNI/MTX (no abatacept).
●In the 7/8 HLA-matched cohort, abatacept was associated with improved two-year OS (74 to 77 percent versus 45 percent; HR 0.28 [95% CI 0.16-0.48]), superior two-year NRM and RFS, less grade 2-4 aGVHD at day 100 and day 180, but no difference in cGVHD, compared with historical controls.
An ongoing multicenter phase 2 study (#NCT03924401) is evaluating abatacept for prevention of cGVHD.
Abatacept is the first drug approved by the US Food and Drug Administration (FDA) for aGVHD prophylaxis.
Antibiotics (to alter gut flora) — Observational studies suggest that the diversity and composition of intestinal bacteria (ie, gastrointestinal [GI] microbiome) plays a role in the development of GVHD involving the lower GI tract. GI GVHD has been associated with the expansion of certain pro-inflammatory bacteria (eg, Enterobacteriaceae), a decrease in anti-inflammatory bacteria (eg, Clostridia spp.), and less bacterial diversity overall [64-67].
An international study of >1300 patients reported that alterations of the intestinal microbiome were associated with increased mortality from GVHD [68]. Loss of bacterial diversity and domination by single taxa before and after transplantation were independently associated with increased GVHD-associated mortality and transplantation-related mortality (TRM). It is not clear at present if this association reflects a causal relationship or if the microbiome can be manipulated to influence outcomes. (See "Pathogenesis of graft-versus-host disease (GVHD)", section on 'Microbiome'.)
In a phase 3 trial, 134 patients undergoing allogeneic HCT were randomly assigned to receive either ciprofloxacin or ciprofloxacin plus metronidazole for five weeks following transplantation [69]. There was no difference in five-year OS or incidence of cGVHD, but among patients who received matched sibling donor (MSD) grafts, combination antibiotic treatment was associated with less grade 2-4 aGVHD (18 percent compared with 54 percent with ciprofloxacin alone). The incidence of aGVHD did not differ for recipients of partially matched related or unrelated grafts (36 versus 46 percent).
Some experts begin an antibiotic (eg, a quinolone) on the day before initiating the conditioning regimen and continue it until patients have engrafted or are started on intravenous antibiotics. (See "Prevention of infections in hematopoietic cell transplant recipients", section on 'Antibacterial prophylaxis'.)
Prednisone — Prednisone is used for treatment of aGVHD, but it is not part of standard prophylactic regimens because of the toxicity of long-term treatment.
Prednisone can be used if there are concerns about administering MTX or in patients at high risk of hepatic sinusoidal obstructive syndrome (SOS; also called veno-occlusive disease) (table 1). (See "Treatment of acute graft-versus-host disease" and "Hepatic sinusoidal obstruction syndrome (veno-occlusive disease) in adults", section on 'Risk factors'.)
Prospective trials have evaluated the use of prednisone in combination with other agents (eg, cyclosporine [CsA], MTX, or cyclophosphamide) for the prevention of GVHD [70-78]. While a meta-analysis suggested that the addition of glucocorticoids significantly reduced the risk of both mild and severe aGVHD, there were no alterations in OS, disease-free survival (DFS), or the incidence of relapse [79].
Long-term follow-up (median >24 years) from one phase 3 trial reported that, compared with patients who received CsA plus MTX, patients who received methylprednisolone plus CsA and MTX had superior OS (55 versus 20 percent, respectively) and RFS (49 versus 15 percent) [78]. There was more late nonrecurrence mortality in the group not given methylprednisolone (11 versus 0 deaths >15 years after transplantation), but no differences in rates of disease recurrence or secondary malignancies.
Rituximab — There is no demonstrated role for rituximab (CD20-directed monoclonal antibody) as prophylaxis against GVHD.
Addition of rituximab to standard GVHD prophylaxis was associated with no improvement in OS or grade 2-4 GVHD in a randomized open-label phase 2 study that included 84 patients with aggressive lymphomas [80].
Sirolimus — Sirolimus (Sir; also called rapamycin) is a lipophilic macrolide isolated from a soil sample from Easter Island (Rapa Nui).
Sir is not currently approved by the US FDA for use against GVHD.
●Mechanism of action – The major effect of Sir is to interfere with the CD28 signaling pathway. Secondary signals required for T cell activation and proliferation depend on ligation of the B7 family with CD28; an absence of the secondary signal results in T cell anergy. Sir inhibits progression of cells from G1 into the S phase and interrupts signal transduction pathways that mediate specific cytokine responses (eg, interleukin [IL]-1-driven interferon-gamma production) [81,82].
Sir binds to the same family of intracellular receptors as tacrolimus (Tac; ie, FK-binding proteins [FKBP]), but the two drugs have distinct mechanisms of action; Tac (and CsA) inhibit T cell proliferation by altering signaling downstream of the T cell receptor (TCR). (See 'Calcineurin inhibitors (CNI)' above.)
●Administration – Sir is administered orally and monitored twice weekly to maintain a target plasma level of 8 to 14 ng/mL for the first two months and 5 to 8 ng/mL until discontinuation. Sir has been combined with Tac or with post-transplant cyclophosphamide (PTCy) and mycophenolate mofetil (MMF) as non-MTX-containing regimens (eg, to avoid mucositis) [83,84].
●Adverse effects (AEs) – Elevated liver function tests, diarrhea, hypertriglyceridemia, and cytopenias are the most common AEs; blood pressure changes, headaches, nausea, mucous membrane irritation, and infections also occur. Nephrotoxicity is rarely encountered with Sir, although it may potentiate the nephrotoxicity of CsA.
Sir has been associated with hepatic SOS following myeloablative conditioning (MAC) regimens, especially when myeloablative doses of busulfan were employed [85]. Myeloablative doses of busulfan should not be used with sirolimus-based immunosuppression [86]. (See "Hepatic sinusoidal obstruction syndrome (veno-occlusive disease) in adults".)
●Outcomes – Sir has been evaluated in several settings:
•MSD grafts – A phase 3 trial in 304 patients undergoing MSD HCT reported that Tac plus Sir versus Tac plus MTX were associated with similar survival and incidence of GVHD, but Sir was associated with faster engraftment and less mucositis [83]. Compared with Tac/MTX, Tac/Sir was associated with similar two-year OS (59 versus 63 percent) and RFS (53 versus 54 percent) and similar rates of grade 2-4 aGVHD (26 versus 34 percent) and cGVHD (53 versus 45 percent).
•Haploidentical HCT – Sir-based GVHD prophylaxis (using PTCy and MMF; ie, a CNI-free regimen) for haploidentical MAC HCT was associated with a low rate of GVHD, a high rate of engraftment, acceptable rates of NRM and GVHD, but a high incidence of viral reactivation (eg, CMV, EBV, HHV6) and other infections [84]. Among the 40 patients, one-year OS and DFS were 56 and 48 percent, respectively, one-year NRM was 17 percent, and 35 percent relapsed. There were no cases of hepatic SOS or thrombotic microangiopathy (TMA), but one case of posterior reversible encephalopathy syndrome (PRES).
•Peripheral blood (PB) grafts – Sirolimus plus Tac (ie, without MTX) for GVHD prophylaxis in PB MAC HCT was associated with rapid engraftment, a low incidence of aGVHD, minimal transplant-related toxicity, and good survival; hepatic SOS occurred in 8 percent of patients [87]. At day 100, TRM was 5 percent; transplant-related complications included seven patients with hepatic SOS (four died of this complication) and TMA in six patients. Outcomes were similar for recipients of MSD and unrelated donor grafts, and included 72 percent two-year OS, 68 percent two-year RFS, 21 percent grade 2-4 aGVHD, 5 percent grade 3-4 aGVHD, and 59 percent cGVHD. Day 30 TRM was 0 percent and day 100 TRM was 5 percent.
•HCT for lymphomas – In a retrospective study of 190 patients, compared with a CNI plus MTX, treatment that included Sir was associated with improved OS and decreased risk of disease progression in lymphoma patients undergoing allogeneic HCT [88]. The benefit was restricted to patients undergoing reduced intensity conditioning (RIC) HCT: 66 percent three-year OS compared with 38 percent for non-Sir group. Patients who received Sir had a similar incidence of NRM but decreased incidence of disease progression (42 versus 74 percent).
•Children – A phase 3 trial in 146 children reported that addition of Sir to Tac/MTX was associated with no difference in two-year OS, event-free survival (EFS), or grade 3-4 aGVHD, less grade 2-4 aGVHD (18 versus 31 percent), but more treatment-related complications [89]. Grafts included MSD, MUD, and umbilical cord blood (UCB) donors. Compared with Tac/MTX, there was more hepatic SOS in patients receiving Tac/MTX/Sir (21 percent versus 9 percent) and more TMA (10 versus 1 percent).
●Incidence of hepatic SOS – A retrospective review of 488 patients reported that Sir was associated with an increased incidence of hepatic SOS (16 versus 7 percent for patients who did not receive Sir) [85]. Rates of SOS varied according to the prophylaxis regimen: Tac/MTX (7 percent), Tac/Sir (11 percent), and Tac/Sir/MTX (21 percent). Compared with Tac/MTX, the odds ratio (OR) for hepatic SOS with Tac/Sir/MTX was 3.23 [95% CI 1.61-6.69]). Death occurred in at least one-third of SOS-affected patients. Use of an unrelated or mismatched donor further increased the risk of SOS. When combined with busulfan-based conditioning, the risk of SOS was still higher (OR 8.8).
Diagnosis and management of hepatic SOS in adults and children are discussed separately. (See "Hepatic sinusoidal obstruction syndrome (veno-occlusive disease) in adults" and "Hepatic sinusoidal obstruction syndrome (veno-occlusive disease) in children".)
Sitagliptin — Sitagliptin, which is approved for treatment of type 2 diabetes mellitus, is an inhibitor of T cell costimulation that may be effective against aGVHD in combination with Tac and Sir. Sitagliptin is an inhibitor of dipeptidyl peptidase 4 (DPP-4; also known as CD26), a homodimeric transmembrane receptor expressed on T cells. DPP-4 interacts with caveolin on antigen-presenting cells (APC), which upregulates CD86 on APCs and results in costimulation [90,91].
In a phase 2 study of 36 patients who underwent myeloablative allogeneic HCT, sitagliptin in combination with Tac/Sir was associated with 5 percent day 100 grade 2-4 aGVHD [92]. At one year, 46 percent of patients were GVHD-free/relapse-free and no toxic effects were attributed to sitagliptin. Compared with previous studies using Tac/Sir, addition of sitagliptin was associated with decreased GVHD but no increase in relapses, suggesting no loss of graft-versus-leukemia effect. Sitagliptin was previously shown to be safe and effective in reducing aGVHD after UCB transplantation [93,94].
Tocilizumab — There is no proven benefit for adding tocilizumab (Toci), an inhibitor of interleukin-6 (IL-6), to standard GVHD prophylaxis. Involvement of IL-6 in the pathogenesis of aGVHD is discussed separately. (See "Pathogenesis of graft-versus-host disease (GVHD)".)
In a phase 3 trial, addition of Toci to MTX and CsA did not alter median survival, but it was associated with trends toward less aGVHD and improved GVHD-free survival, with little incremental toxicity [95]. In this multicenter trial, 145 patients were treated with CsA/MTX plus either Toci (8 mg/kg to a maximum 800 mg as a single intravenous infusion over 60 minutes) or placebo on day -1. Compared with placebo, Toci was associated with a lower incidence of grade 2-4 aGVHD at day 100 (36 versus 27 percent) and at day 180 (40 versus 29 percent), but neither difference was statistically significant. There were similar levels of liver toxicity and infectious complications in both trial arms.
Single-arm prospective studies had also suggested that Toci was associated with a reduced incidence of GVHD, compared with historical controls [96,97].
SUMMARY AND RECOMMENDATIONS
●Graft-versus-host disease (GVHD) – GVHD occurs when immune cells transplanted from a nonidentical donor (the graft) recognize the transplant recipient (the host) as foreign, thereby initiating an immune reaction in the recipient. The pathogenesis of GVHD is a complex, multistep process, but it is primarily initiated by T lymphocytes. (See "Pathogenesis of graft-versus-host disease (GVHD)".)
●Immunosuppressive backbone – For GVHD prophylaxis in allogeneic hematopoietic cell transplantation (HCT), we suggest a calcineurin inhibitor (CNI) plus an antimetabolite, rather than a CNI alone or an antimetabolite alone (Grade 2B). CNIs include tacrolimus (Tac) or cyclosporine (CsA), while antimetabolites include either methotrexate (MTX) or mycophenolate mofetil (MMF) (table 1). (See 'Immunosuppressive backbone' above.)
●Selection of a GVHD prophylactic regimen varies according to the degree of immunologic match, transplant conditioning regimen, graft source (ie, bone marrow versus peripheral blood), and other risk factors:
•Matched sibling donor (MSD) graft – For HCT with a human leukocyte antigen (HLA)-matched sibling/related donor, we suggest use of CNI plus MTX rather than CNI plus MMF (Grade 2C). (See 'Matched sibling donor (MSD)' above.)
•Matched unrelated donor (MUD) graft – For HCT using an MUD (ie, ≥9/10 or ≥7/8 HLA alleles), we suggest the addition of antithymocyte globulin (ATG) to a CNI plus MTX, rather than other regimens (Grade 2C). (See 'Matched unrelated donor (MUD)' above.)
•Haploidentical (haplo) HCT – For haplo HCT, we suggest adding post-transplant cyclophosphamide (PTCy) to a CNI plus MMF, rather than other immunosuppressive regimens (Grade 2C). (See 'Haploidentical donor' above.)
•Myeloablative conditioning (MAC) – For MAC allogeneic HCT, we suggest a CNI plus MTX, rather than a CNI plus MMF (Grade 2C). (See 'Myeloablative conditioning (MAC)' above.)
•Non-MAC – For adults undergoing nonmyeloablative conditioning (NMA) or reduced intensity conditioning (RIC) allogeneic HCT, we suggest a CNI plus MMF, rather than a CNI plus MTX (Grade 2C). (See 'Non-MAC regimens' above.)
•Peripheral blood stem/progenitor cell (PBSPC) grafts – For transplantation using a peripheral blood graft source, we consider treatment with either CNI (Tac or CsA) plus either MTX or MMF acceptable. (See 'Peripheral blood grafts' above.)
●Selection of a CNI – Tac and CsA are structurally distinct, but they have similar clinical properties, efficacy, and toxicity. Selection of a CNI for GVHD prophylaxis is largely guided by institutional preference. (See 'Choice of CNI' above.)
●Selection of an antimetabolite – The choice of MTX versus MMF varies with the conditioning regimen, patient factors, and institutional preference. (See 'Conditioning regimen' above and 'Other considerations' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟