Immunotherapy for the prevention and treatment of relapse following allogeneic hematopoietic cell transplantation

INTRODUCTION — Most patients who undergo allogeneic hematopoietic cell transplantation (HCT) for a hematologic malignancy attain a complete remission, but relapse occurs in 10 to 40 percent of patients.

Donor lymphocyte infusion (DLI), immune checkpoint inhibitors, and chimeric antigen receptor T cells (CAR-T cells) are important immunotherapy approaches for managing relapse after allogeneic HCT. Various cell therapies and vaccination are among the experimental immunotherapeutic strategies that are undergoing investigation in this setting.

This topic discusses immunotherapy to prevent and/or treat relapse following allogeneic HCT.

Management of relapse of specific hematologic malignancies following HCT is discussed separately:

●(See "Treatment of relapsed or refractory acute myeloid leukemia", section on 'Relapse after HCT'.)

●(See "Treatment of relapsed or refractory acute lymphoblastic leukemia in adults".)

●(See "Treatment of acute lymphoblastic leukemia/lymphoma in children and adolescents".)

●(See "Diffuse large B cell lymphoma (DLBCL): Second or later relapse or patients who are medically unfit".)

●(See "Treatment of relapsed or refractory classic Hodgkin lymphoma", section on 'Second and subsequent relapses'.)

Principles that underlie immunotherapy of graft-versus-tumor effect after HCT are presented separately. (See "Biology of the graft-versus-tumor effect following hematopoietic cell transplantation".)

RATIONALE — The rationale for using immunotherapy to prevent and/or treat disease relapse after HCT is based on the following observations:

●Graft-versus-tumor (GVT) effect reduces the risk of relapse following allogeneic HCT, as illustrated by higher rates of leukemia relapse in patients who have received identical twin (syngeneic) transplants compared with HLA-identical sibling transplants, despite identical cytotoxic treatment [1]. (See "Donor selection for hematopoietic cell transplantation", section on 'Identical twin donors'.)

●T cell-depleted grafts are associated with a higher incidence of disease relapse. (See "Prevention of graft-versus-host disease", section on 'T cell depletion (TCD)'.)

●Patients who develop acute or chronic graft-versus-host disease have a lower risk of leukemia relapse [2,3].

●Donor lymphocyte infusion (DLI) into patients who have relapsed following an allogeneic HCT induces GVT and can achieve remission with some diseases. (See 'Donor lymphocyte infusion (DLI)' below.)

●Grafts with higher numbers of certain natural killer T cells (eg, iNKT cells) are associated with improved survival [4-6].

●Relapse of acute myeloid leukemia after allogeneic HCT is associated with dysregulation of immune function pathways, including downregulation of MHC class II genes, which are involved in antigen presentation [7].

DONOR LYMPHOCYTE INFUSION (DLI) — Donor lymphocyte infusion (DLI) can induce a graft-versus-tumor (GVT) effect and achieve complete and durable remissions in some patients who relapse after allogeneic HCT. DLI demonstrates the important role of GVT against hematologic malignancies. (See "Biology of the graft-versus-tumor effect following hematopoietic cell transplantation".)

Mechanism of action — DLI is primarily mediated by immune effector cells (table 1), including CD4+ T cells, CD8+ T cells, regulatory T cells (T regs), natural killer (NK) cells, and antigen presenting cells. Only rarely does reduction of immunosuppressive therapy without DLI achieve clinically meaningful GVT, and this effect may be limited to treatment of chronic myeloid leukemia (CML) [8,9].(See "Biology of the graft-versus-tumor effect following hematopoietic cell transplantation".)

DLI is thought to mediate GVT primarily via reversal of T cell exhaustion in resident CD8+ T cells (a state of reduced effector function and proliferation by T cells associated with chronic antigen exposure) [10,11]. DLI is also associated with normalization of the T cell receptor repertoire and clonal expansion of allogeneic T cells and improved coordination of T and B cell immunity [12,13].

The targets of the GVT response are uncertain. Potential targets include disease-specific antigens (eg, BCR::ABL1 in chronic myeloid leukemia [CML], other leukemia-specific antigens, idiotypic immunoglobulins in plasma cell disorders), and nondisease-specific targets (eg, minor histocompatibility antigens), as listed in the accompanying table (table 2).

There is controversy about whether the GVT effect can be separated from graft-versus-host disease (GVHD). Experimental immunotherapy approaches that seek to separate GVT from GVHD are discussed separately. (See "Biology of the graft-versus-tumor effect following hematopoietic cell transplantation", section on 'Possible separation of GVT from GVHD'.)

Time course of DLI response — Patients who respond to DLI usually demonstrate a clinical response within two to three months, but a full response may take one year or longer [14]. Responses to DLI can be durable, with reports of responses lasting ≥20 years [15]. As an example, for patients with CML who achieved a molecular complete response (CR) in response to DLI, 95 percent were alive and in continuous CR at three years [16]. Similar results have been reported by others [17-19].

Efficacy of DLI — Efficacy of DLI varies with the type and amount of the underlying malignancy, dose and type of lymphocytes that are infused, and factors related to the initial allogeneic HCT.

Disease effects — Hematologic malignancies differ in their intrinsic sensitivity to GVT but, in general, relapse rates are lower in proportion to the severity of GVHD. A study that included nearly 50,000 patients who underwent allogeneic HCT reported the greatest sensitivity to GVHD for CML and BCR::ABL1-negative myeloproliferative neoplasms (eg, polycythemia vera, essential thrombocythemia, myelofibrosis), intermediate sensitivity for myelodysplastic neoplasms/syndromes (MDS), acute myeloid leukemia (AML), and lymphoproliferative disorders, and least sensitivity for acute lymphoblastic leukemia (ALL) and plasma cell disorders [20].

A review of responses to DLI based on the type of hematologic malignancy included the following estimates [14]:

●CML – Nearly 100 percent response in patients with molecular relapse of chronic phase CML, >75 percent response with hematologic relapse of chronic phase CML, and lower response rates (eg, 12 to 35 percent) in accelerated or blast phase of CML.

●AML – CR rates of 15 to 42 percent with two-year overall survival (OS) 15 to 20 percent at two years.

●ALL – CR rates of 18 to >50 percent (although most patients also received chemotherapy) and two-year OS 5 to 20 percent.

●Lymphomas – CR rates of 42 to 85 percent and two-year OS 35 to 70 percent.

●Multiple myeloma – CR rates of 42 to 85 percent and two-year OS >40 percent at two years.

DLI is effective for both clinical relapse and in the setting of lower disease burden (ie, molecular relapse). As an example, in a prospective study of allogeneic HCT for acute leukemia, DLI was administered to patients with measurable residual disease (MRD) following transplantation [21]. Those who received DLI achieved comparable clinical outcomes to patients who remained MRD negative, and both groups had superior outcomes compared to MRD-positive patients who were treated with IL-2 because there was no available donor for DLI.

Use of DLI for CML and AML is discussed separately. (See "Hematopoietic cell transplantation in chronic myeloid leukemia", section on 'Donor lymphocyte infusion' and "Treatment of relapsed or refractory acute myeloid leukemia", section on 'Donor lymphocyte infusion'.)

Dose and type of DLI — DLI efficacy is influenced by the number and nature of infused cells.

●Dose – Published studies have reported doses ranging from 0.01 to 8.8 x 10⁸ T cells/kg [14]. Cell doses <0.01 x 10⁸ T cells/kg appear to be suboptimal and doses >4.5 x 10⁸ T cells/kg do not appear to improve response and likely carry more risk for GVHD. (See 'Clinical administration of DLI' below.)

●Type of infusion – Unmanipulated leukocytes can achieve significant clinical responses, particularly in patients with CML [22]. Donor source (ie, DLI from HLA-matched related donors versus matched unrelated donors) had no effect on the rates of durable molecular remission [16].

The effect of donor source and manipulation of the infused cells on GVHD is discussed below. (See 'GVHD' below.)

Experimental methods of manipulating cells prior to infusion are discussed below. (See 'Manipulated DLI grafts' below.)

Host effects — Efficacy of DLI is influenced by host effects, including the intensity and type of conditioning, donor-recipient alloreactivity, T cell depletion of the host, timing of DLI, and post-transplantation immunosuppression therapy [23,24]. However, the direction and magnitude of these contributions to efficacy of DLI are not well defined and difficult to utilize clinically.

As an example, in one uncontrolled study, T cell depletion of the host with alemtuzumab conditioning or anti-thymocyte globulin was associated with a low rate of relapse [25]. In contrast, in another study, lymphodepletion of the host was associated with increased GVHD [26]. The presence of host-derived regulatory T cells was reported to be associated with decreased GVT effect [27].

DLI has been used successfully with both myeloablative and nonmyeloablative/reduced-intensity conditioning [25,28].

Toxicity of DLI — The major complications of DLI are GVHD and bone marrow hypoplasia.

GVHD — GVHD is the most important adverse effect of DLI. Clinical manifestations of acute and chronic GVHD are described separately. (See "Clinical manifestations and diagnosis of chronic graft-versus-host disease" and "Clinical manifestations, diagnosis, and grading of acute graft-versus-host disease".)

Approximately 60 to 70 percent of patients receiving a DLI will develop GVHD [29]. DLI is likely to exacerbate ongoing GVHD. (See 'Cautions regarding DLI' below.)

Factors that may contribute to the incidence and severity of GVHD following DLI include:

●Dose – Doses >1 x 10⁸ T cells/kg are associated with higher rates of GVHD [14,30].

●Schedule – One study reported a lower rate of GVHD when using an escalating DLI cell dose regimen versus a single bulk infusion, but the two approaches achieved similar rates of remission [16].

●Manipulation of the graft – Depletion of CD8+ T cells or enrichment of CD4+ T cells may increase the rate of GVHD [31,32]. Depletion of naïve T cells and infusion of memory CD8+ cells may reduce the risk of GVHD; however, the impact on GVT reactions is less clear.

●Lymphodepletion – Lymphodepletion of the host prior to DLI (ie, use of a lymphodepleting HCT conditioning regimen) is associated with more acute GVHD and greater GVHD lethality [26].

●Disease – The rate of GVHD appears to be independent of the underlying malignant disorder.

Other factors that may influence GVHD include donor-recipient HLA status, intensity of the conditioning regimen, degree of donor chimerism, and post-transplantation immunosuppression therapy [24]. (See "Clinical manifestations, diagnosis, and grading of acute graft-versus-host disease", section on 'Risk factors' and "Clinical manifestations and diagnosis of chronic graft-versus-host disease", section on 'Risk factors'.)

Hypoplasia — Bone marrow hypoplasia and/or aplasia after DLI has been reported in 20 to 40 percent of patients. Patients with aplasia have an overall mortality of about 5 percent due to infections or bleeding [14].

Aplasia has been reported primarily in patients with little or no evidence of ongoing donor hematopoiesis (measured by the presence of donor CD34+ cells) at the time of DLI [33]. Bone marrow chimerism should be evaluated prior to the possible use of DLI, as discussed below. (See 'Clinical administration of DLI' below.)

Enhancement of chimerism — DLI may enhance donor engraftment in the setting of mixed chimerism following HCT. In a single center study, DLI after T cell depletion with alemtuzumab or anti-thymocyte globulin achieved full donor chimerism in 67 percent of patients with mixed donor chimerism [25]. However, other experiences are not as favorable and in our experience DLI alone rarely results in improved chimerism. An alternative is to combine pentostatin with DLI, although definitive studies are lacking [34].

We suggest not using DLI solely to enhance engraftment in the setting of mixed chimerism, unless there is evidence of relapsed disease, as discussed below. (See 'Cautions regarding DLI' below.)

Clinical administration of DLI — No prospective studies have directly compared DLI regimens for treatment of relapsed hematologic malignancies. There is no consensus regarding the optimal dose or timing of DLI, or the method for collection or processing of infusion products. Clinical application of DLI varies in different transplant centers.

Outside of the context of a clinical trial, we favor administration of DLI upon detection of molecular relapse (ie, MRD) rather than waiting for emergence of hematologic/clinical relapse. Selection of dose should consider the clinical setting (ie, clinical versus molecular relapse) and the doses that were used in trials of DLI in patients with that particular malignancy. As examples, for molecular relapse we treat with ≥1 x 10⁶ CD3+ cells/kg, while for clinical relapse we generally treat with 1 to 3 x 10⁷ CD3+ cells/kg. We treat with unmanipulated T cells, collected by apheresis, and administered in a single infusion. Use of DLI for CML and AML is discussed separately. (See "Hematopoietic cell transplantation in chronic myeloid leukemia", section on 'Donor lymphocyte infusion' and "Treatment of relapsed or refractory acute myeloid leukemia", section on 'Donor lymphocyte infusion'.)

Patients who respond to DLI usually demonstrate a clinical response within two to three months, although a full response may take one year or longer [14]. A second DLI infusion may be administered if no clinical response is seen within three months.

We favor administration of DLI upon detection of measurable residual disease (MRD), rather than waiting for emergence of hematologic/clinical relapse. In theory, DLI may be more effective when there is less disease burden, thereby permitting a lower dose DLI, which has been associated with reduced risk of GVHD. Some experts advocate the use of DLI upon detection of MRD [21], while others await clinical relapse [35]. A decision regarding the timing of DLI is influenced by the responsiveness of the underlying disease to DLI, the risk of GVHD, institutional practice, and goals and fears of the patient and clinician.

In the setting of hematologic/clinical relapse, we generally treat with salvage chemotherapy to reduce the burden of disease prior to DLI. We do not treat with concurrent chemotherapy and DLI, because it is associated with increased toxicity, excess treatment-related mortality, and no evidence of improved survival [35]. Chemotherapy regimens to reduce the burden of relapsed disease are discussed separately. (See "Treatment of relapsed or refractory acute myeloid leukemia" and "Treatment of relapsed or refractory acute lymphoblastic leukemia in adults".)

Techniques to deplete or enrich specific lymphocyte subsets or to manipulate the donor infusion product are considered experimental and we suggest not using them outside of the context of a clinical trial. Such experimental approaches are discussed below. (See 'Investigational approaches' below.)

Cautions regarding DLI — Toxicity of DLI is significant, and certain cautions should be considered prior to proceeding, including:

●GVHD – DLI should be avoided in patients with steroid-resistant GVHD (ie, ≥grade 2) because of the significant risk of further exacerbating this condition. (See "Clinical manifestations, diagnosis, and grading of acute graft-versus-host disease" and "Clinical manifestations and diagnosis of chronic graft-versus-host disease".)

●Hypoplasia – Chimerism studies should be performed before DLI because of the risk of bone marrow hypoplasia in patients who have converted to host (rather than donor) chimerism. (See 'Hypoplasia' above.)

DLI should not be used (or used only with extreme caution) in patients who have converted to host chimerism (ie, graft failure). There is no consensus, but various experts suggest ≥10 percent to ≥50 percent donor chimerism as a threshold for use of DLI.

●Graft failure – We suggest not using DLI solely for the purpose of enhancing engraftment in the setting of mixed chimerism, unless there is evidence of relapsed disease, because of the risk of exacerbating GVHD.

CHECKPOINT INHIBITORS — Inhibition of negative regulators of T cell-mediated immunity (eg, PD-1/PD-L1, CTLA-4) can achieve remission of hematologic malignancies that relapse after HCT. Regulation of the immune response after HCT is discussed separately. (See "Biology of the graft-versus-tumor effect following hematopoietic cell transplantation".)

As an example, the checkpoint inhibitor ipilimumab achieved complete response (CR) in 23 percent, partial response (PR) in 9 percent, and decreased tumor burden in 27 percent of 22 patients with relapsed hematologic malignancies; four patients sustained a response for more than one year [36]. Immune-related adverse events, including one death, were observed in 21 percent, and GVHD that precluded further administration of ipilimumab was observed in 14 percent. In another study, treatment with PD-1 blockade after allogeneic HCT was frequently complicated by severe and treatment-refractory GVHD [37].

Blockade of the PD-1/PD-L1 interaction may relieve T cell "exhaustion" (inhibition) exerted by cancer cells on CD8+ T cells against minor histocompatibility antigens. Blockade of this pathway has been combined with an experimental vaccine strategy for multiple myeloma [38]. Use of PD-1 blockade in Hodgkin lymphoma is discussed separately. (See "Treatment of relapsed or refractory classic Hodgkin lymphoma", section on 'PD-1 blockade'.)

CHIMERIC ANTIGEN RECEPTOR T CELLS — Chimeric antigen receptor T cells (CAR-T cells) are a form of genetically modified autologous immunotherapy that can be directed at tumor antigens.

CAR-T cells are effective for relapsed leukemias and lymphomas after HCT. Efficacy and toxicity of tisagenlecleucel and axicabtagene ciloleucel for treatment of relapsed acute lymphoblastic leukemia and diffuse large B cell lymphoma are discussed separately. (See "Treatment of relapsed or refractory acute lymphoblastic leukemia in adults", section on 'CAR-T' and "Diffuse large B cell lymphoma (DLBCL): Second or later relapse or patients who are medically unfit", section on 'Chimeric antigen receptor T cell therapy'.)

CAR-T cells are undergoing investigation for other hematologic malignancies such as chronic lymphocytic leukemia (CLL). (See "Treatment of relapsed or refractory chronic lymphocytic leukemia", section on 'Investigational therapies'.)

INTERLEUKIN-2 — We advise that the cytokine interleukin-2 (IL-2) should not be used for relapse after HCT outside of the context of a clinical trial. IL-2 is approved by the US Food and Drug Administration to promote survival, expansion, and activation of lymphocytes, but there are concerns about its potential suppression of graft-versus-tumor (GVT) because it preferentially stimulates Tregs.

No prospective controlled studies of IL-2 have demonstrated improved clinical outcomes after HCT, and it may worsen toxicity [39-42]. IL-2 was associated with inferior rates of disease-free survival and overall survival at three years in a prospective study in which patients with measurable residual disease (MRD) after allogeneic HCT for acute leukemia were treated with either DLI or IL-2 (if they did not have an available donor) [21].

INVESTIGATIONAL APPROACHES — Various novel strategies are under investigation to enhance donor-mediated graft-versus-tumor (GVT) effect (to prevent or control disease relapse) and/or reduce graft-versus-host disease (GVHD) in patients who relapse after allogeneic HCT.

While many of these approaches are promising, we suggest not using them at present, outside of the context of a clinical trial.

DLI plus targeted therapy — Donor lymphocyte infusion (DLI) has been combined with targeted agents to control residual disease after allogeneic HCT.

As examples:

●AML/CML:

•In older patients with relapse of acute myeloid leukemia (AML) or chronic myeloid leukemia (CML), 5-azacitidine followed by DLI achieved responses in 66 percent, including patients with adverse-risk cytogenetic findings [43].

•DLI in combination with the tyrosine kinase inhibitor sorafenib for FLT3-ITD mutant AML was associated with substantial responses but significant GVHD, cytopenias, and hand-foot syndrome [44].

•Low dose DLI (0.5 to 1 x 10⁷ mononuclear cells/kg) in combination with interferon alfa achieved remissions with acceptable levels of GVHD for chronic phase CML that relapsed after allogeneic HCT [45].

●Multiple myeloma – Bortezomib and lenalidomide have been administered together with DLI in patients with residual multiple myeloma after allogeneic HCT and appeared to enhance responses without exacerbating GVHD [46,47].

Manipulated DLI grafts — Techniques for manipulation of the donor product (eg, enrichment, depletion, activation) are considered experimental and require further study before they can be widely applied for DLI.

Lymphocyte depletion/enrichment — Enrichment and/or depletion of specific lymphocyte subsets to enhance GVT or lessen GVHD are undergoing investigation.

Examples include:

●T lymphocyte subsets – Fractionation of the donor product has been used to enrich CD4+ T cells or deplete CD8+ T cells. As an example, in a study of patients who underwent reduced intensity conditioning (RIC) allogeneic HCT using alemtuzumab, DLI with CD8-depleted product accelerated immune reconstitution, improved donor engraftment, and was not associated with significant GVHD [48]. Other studies have also shown safety of CD4-enriched and/or CD8-depleted DLI treatment [31,49].

DLI with CD25/Treg-depleted product was capable of inducing GVT tumor responses without excessive GVHD [50].

Enrichment of memory T cells (CD8+, CD44^high) is undergoing investigation as a means of establishing an in vivo reservoir of functional antigen-specific T cells [24,51].

●Natural killer (NK) cells – NK cells exert potent GVT effects without initiating GVHD, and act through a variety of cytokine receptors, and activating or inhibitory receptors [52]. Alloreactive haploidentical NK cells can be safely administered and may reduce leukemia progression without an increase in GVHD or transplantation-related mortality [53-57].

Ex vivo activation — Activation of DLI effectors prior to infusion may enhance GVT.

Examples of clinical studies of ex vivo activated lymphocytes include:

●Cytokine-induced killer (CIK) cells – Ex vivo culture of lymphocytes with interferon gamma, IL-2, and anti-CD3 creates CIKs, which are cytotoxic T cells that recognize targets through the C-type lectin receptor, NKG2D [58]. CIKs can kill autologous tumor targets in patients with a variety of malignancies [59,60]. Allogeneic CIK cells achieved response rates of 30 percent with acceptable toxicity in patients with relapsed AML [61]. CIKs have also been examined as DLI after autologous and allogeneic HCT for CML and lymphomas [62-64]. These cells have been used for construction of chimeric antigen receptor (CAR)-T cells, with encouraging results [65].

●Natural killer (NK) cells – IL-15 is a cytokine that can increase NK and T cell proliferation and activation without stimulating Tregs, but it has a short half-life in vivo [66]. ALT-803 is an engineered IL-15 superagonist complex with an extended half-life that can mimic the physiologic action of IL-15 [67,68]. In a phase I study in patients with relapsed hematologic malignancies after allogeneic HCT, ALT-803 achieved clinical responses in 19 percent of evaluable patients (including one complete remission lasting seven months) and was well tolerated with no dose-limiting toxicities or exacerbation of GVHD [69]. ALT-803 resulted in NK cell expansion and activation with minimal changes in CD4+ T cells and Tregs. NK cells have also been used for generation of CAR-NK cells with encouraging results in early phase clinical trials [70].

Modified leukocytes — Modification of T cells by creation of chimeric antigen receptor T cells (CAR-T cells) is discussed above. (See 'Chimeric antigen receptor T cells' above.)

Donor T cells have been genetically altered to contain a fusion caspase 9 protein that could induce apoptosis upon exposure to a synthetic dimerizing drug [71]. Administration of the dimerizing drug to four patients who developed GVHD eliminated 90 percent of the modified T cells within 30 minutes and resulted in resolution of the GVHD within 48 hours.

HLA-mismatched HCT — In most centers, a complete match at the HLA-A, HLA-B, HLA-C, and HLA-DR loci is preferred for allogeneic donors, because mismatches are associated with a higher risk of severe GVHD, despite the possible induction of GVT. (See "Donor selection for hematopoietic cell transplantation".)

Transplantation across HLA barriers in the setting of a nonmyeloablative (NMA) conditioning regimen (with or without DLI) may create mixed chimerism with potent GVT without induction of severe GVHD [72,73]. In one study of five patients with refractory non-Hodgkin lymphoma, the use of such a strategy was associated with complete and partial clinical remission in two patients and neither patient developed GVHD [72]. (See "Donor selection for hematopoietic cell transplantation", section on 'Mismatched related donors'.)

Haploidentical grafts should trigger alloreactivity based on HLA class I disparity between donor and recipient via killer immunoglobulin-like receptors (KIR) on NK cells [74]. NK cells can persist and expand in vivo in the setting of haploidentical HCT [53], but a retrospective analysis identified no clear advantage associated with KIR ligand incompatibility [75].

Cancer vaccines — Cancer vaccines may provide a strategy to disrupt tumor-associated tolerance and activate and/or selectively expand tumor specific immunity. The post-transplant setting is a promising platform for vaccination because greater immune responsiveness may result from depletion of inhibitory accessory cells.

Dendritic cells (DC) are antigen-presenting cells that play a major role in the induction of an immune-mediated response [76]. DCs can be obtained from the peripheral blood and expanded ex vivo.

Whole cell-based strategies have created fusions of DCs with tumor cells. As an example, one study treated six patients with multiple myeloma with reduced intensity conditioning (RIC) allogeneic HCT and a partial T cell-depleted graft followed by DLI and DC vaccination [77]. DC vaccination was associated with limited toxicity, and none of these patients developed GVHD.

In a phase I clinical trial, patients with high-risk AML or MDS were immunized with irradiated, autologous, granulocyte-macrophage colony-stimulation factor (GM-CSF)-secreting tumor cells early after allogeneic NMA HCT [78]. Vaccination elicited local and systemic reactions without an increase in GVHD, and 9 of 10 subjects who completed vaccination achieved durable complete remissions (median follow-up 26 months).

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Management of Hodgkin lymphoma".)

SUMMARY

●Immunotherapy takes advantage of the graft-versus-tumor (GVT) effect to prevent or treat relapse of hematologic malignancies following allogeneic hematopoietic cell transplantation (HCT). (See 'Introduction' above.)

●Donor lymphocyte infusion (DLI) can induce GVT and achieve durable remissions in some patients who relapse after allogeneic HCT. DLI is mediated by immune effector cells (table 1), including various T cells and natural killer (NK) cells, primarily by relieving T cell "exhaustion." (See 'Donor lymphocyte infusion (DLI)' above.)

●Efficacy of DLI varies with the type and amount of the underlying malignancy, dose and type of lymphocytes that are infused, and factors related to the initial allogeneic HCT. Response rates are highest with chronic myeloid leukemia (CML), but other types of leukemia and lymphomas also respond to DLI. (See 'Efficacy of DLI' above.)

●Patients who respond to DLI usually demonstrate a clinical response within two to three months, but a full response may take one year or longer. Responses to DLI can be durable, with some responses lasting for many years. (See 'Time course of DLI response' above.)

●Optimal dose and timing of DLI are not well defined and clinical practice varies by institution. Outside of the context of a clinical trial, we favor administration of DLI upon detection of molecular relapse, rather than waiting for emergence of hematologic/clinical relapse. Selection of dose should consider the clinical setting (ie, clinical versus molecular relapse) and doses that were used in trials of DLI in patients with that particular malignancy, as described above. (See 'Clinical administration of DLI' above.)

●The major complications of DLI are graft-versus-host disease (GVHD) and myelosuppression/bone marrow hypoplasia. (See 'Toxicity of DLI' above.)

For some patients, ongoing GVHD or increased risk of marrow hypoplasia (eg, due to loss of donor chimerism) may preclude use of DLI. (See 'Cautions regarding DLI' above.)

●Other immunotherapy approaches for relapse after HCT include checkpoint inhibitors and chimeric antigen T cell receptors (CAR-T). (See 'Checkpoint inhibitors' above and 'Chimeric antigen receptor T cells' above.)

We advise that interleukin-2 (IL-2) should not be used for relapse after HCT, outside of the context of a clinical trial, as discussed above. (See 'Interleukin-2' above.)

●Experimental approaches to enhance donor-mediated GVT and/or reduce GVHD include combinations of DLI with targeted agents, manipulation of the DLI product before infusion, HLA-mismatched allogeneic HCT, and vaccination strategies, as described above. (See 'Investigational approaches' above.)