New and emerging therapies for acute and chronic graft versus host disease

Abstract

Graft versus host disease (GVHD) remains a major cause of morbidity and mortality following allogeneic hematopoietic stem-cell transplantation (HSCT). Despite the use of prophylactic GVHD regimens, a significant proportion of transplant recipients will develop acute or chronic GVHD following HSCT. Corticosteroids are standard first-line therapy, but are only effective in roughly half of all cases with ~50% of patients going on to develop steroid-refractory disease, which increases the risk of nonrelapse mortality. While progress has been made with improvements in survival outcomes over time, corticosteroids are associated with significant toxicities, and many currently available salvage therapies are associated with increased immunosuppression, infectious complications, and potential loss of the graft versus leukemia (GVL) effect. Thus, there is an unmet need for development of newer treatment strategies for both acute and chronic GVHD to improve long-term post-transplant outcomes and quality of life for HSCT recipients. Here, we provide a concise review of major emerging therapies currently being studied in the treatment of acute and chronic GVHD.

Introduction

Allogeneic hematopoietic stem-cell transplantation (HSCT) is a potentially curative treatment for both benign and malignant hematologic conditions. Unfortunately, graft versus host disease (GVHD) remains a major cause of morbidity and mortality following HSCT. Despite prophylactic treatment, acute GVHD (aGVHD) affects 30–70% of recipients and chronic GVHD (cGVHD) occurs in 20–50% of recipients depending on the type of transplant, patient characteristics, and GVHD prophylaxis regimen.^1–5 GVHD is a systemic inflammatory condition primarily mediated by the transplanted immune system that can lead to severe multiorgan damage. The need for increased and prolonged immunosuppression to treat GVHD, in addition to the immunosuppressive effects of the disease itself, increases the risk of: infection, organ impairment, poor quality of life and ultimately, mortality.

A large registry analysis by Khoury and colleagues recently reported a 20% decline in the proportion of grade III–IV acute GVHD between 1999 and 2012.⁶ Overall survival (OS) and treatment-related mortality (TRM) were also noted to have improved during this time frame for patients treated with tacrolimus-based prophylactic regimens. This is encouraging data for both clinicians and HSCT patients, as despite increasing age of transplant recipients, increasing use of alternative donors, and use of more reduced-intensity conditioning regimens over time, there has been improvement in transplant outcomes.^7,8 As we continue to investigate potentially more effective preventive and treatment strategies, hopefully we can continue to make a meaningful impact on transplant outcomes.

We briefly discuss current knowledge of emerging mechanistic targets for treatment of aGVHD and cGVHD, and novel therapies that are showing promising efficacy in both upfront and steroid-refractory (SR) settings. However, most of these agents are still in early-phase clinical studies and have yet to be evaluated in large, late-phase randomized controlled trials. Many of these agents are also being investigated in preventive trials; however, that is beyond the scope of this review.

Acute graft versus host disease

aGVHD is an immunologically mediated process due to mature donor T cells interacting with host and donor antigen presenting cells (APCs), leading to release of pro-inflammatory cytokines, which in turn results in the proliferation and homing of activated T cells to aGVHD target tissues, ultimately causing host tissue damage.^9–11 Recent comprehensive reviews of aGVHD biology have be performed by Magenau and colleagues¹² and Zeiser and colleagues,¹³ and are beyond the scope of this review. A number of major histocompatibility complex (MHC)-independent mechanisms have recently been implicated in GVHD pathogenesis based on findings in murine models. These will be discussed in more detail below.

Corticosteroids (steroids) remain the first-line of therapy for both aGVHD and cGVHD despite suboptimal response rates of 40–60%.^14–16 The likelihood of response to treatment in aGVHD decreases with increasing severity of disease.² For patients who develop SR disease, the long-term prognosis is very poor, with a mortality rate of approximately 70–80%,^14,17 as response rates with second-line treatments are low.^2,18,19 To date, no second-line therapy has been proven superior to another²⁰ for the treatment of SR-aGVHD, and choice of therapy is often based on patient specific characteristics, side-effect profile, and physician preference.

Such suboptimal outcomes underscore the need for new treatment strategies. One newly proposed treatment paradigm is risk stratifying patients for treatment based on clinical staging of aGVHD and blood biomarkers. This has been proposed in an attempt to spare those likely to respond to steroids from excessive toxicity, and to identify those who are less likely to respond and require aggressive upfront therapy to improve nonrelapse mortality (NRM). The aGVHD risk score, which classifies patients into high risk or standard risk categories at the time of diagnosis, was developed by the Minnesota group to identify patients unlikely to respond to upfront treatment with steroids.²¹ Patients with high risk (HR) GVHD were less likely to respond to therapy and had a twofold increased risk of TRM.

The risk score was recently refined, based on validated data from a multicenter cohort totaling 1723 patients. Standard risk was defined as single-organ involvement [stage 1–3 skin or stage 1–2 gastrointestinal (GI)] or two-organ involvement (stage 1–3 skin plus stage 1 GI or stage 1–3 skin plus stage 1–4 liver); all others were defined as high risk (Table 1). Standard-risk patients had a day 28 overall response rate (ORR) rate of 69% versus 43% in high-risk patients (Table 2). The 6-month NRM was 22% versus 44%, respectively.

Table 1.

Graft versus host disease organ staging categories and graft versus host disease risk.¹⁶

GVHD categories grouped by day 28 CR/PR	n	Day 28 CR/PR	OR of day 28 CR/PR (95% CI)	p value	Relative risk of mortality (95% CI)	p value	Relative risk of TRM (95% CI)	p value	GVHD risk
Stage 1–3 skin only**	901	68%	1.0	1.0	1.0				StdR
Upper GI only	115	78%	1.6 (1.0–2.6)	0.04	0.6 (0.4–0.9)	0.02	0.5 (0.3–0.9)	0.03	StdR
Stage 1–2 lower GI only	100	73%	1.3 (0.8–2.1)	0.29	1.2 (0.8–1.7)	0.41	1.6 (1.0–2.3)	0.04	StdR
Stage 1–3 skin + upper GI	90	69%	1.0 (0.6–1.6)	0.89	0.9 (0.6–1.4)	0.64	0.9 (0.6–1.5)	0.71	StdR
Stage 1–3 skin + stage 1 lower GI	71	61%	0.7 (0.4–1.2)	0.16	1.1 (0.7–1.8)	0.60	1.3 (0.8–2.1)	0.35	StdR
Stage 1 lower GI + upper GI	64	64%	0.8 (0.5–1.4)	0.42	1.2 (0.7–1.9)	0.54	1.4 (0.9–2.4)	0.18	StdR
Stage 1–3 skin + stage 1–4 liver	51	71%	1.1 (0.6–2.1)	0.71	1.3 (0.8–2.0)	0.30	1.4 (0.9–2.3)	0.17	StdR
Stage 1–3 skin + stage 1 lower GI + upper GI	62	61%	0.6 (0.4–0.9)	0.12	0.6 (0.3–1.1)	0.11	0.6 (0.3–1.3)	0.18	StdR
Stage 1–2 lower GI + stage 1–3 liver	12	50%	0.4 (0.1–1.4)	0.15	2.9 (1.4–6.3)	0.005	3.2 (1.5–7.0)	0.003	HR
Stage 1–3 skin + (stage 1–2 lower GI or upper GI) + stage 1–3 liver	23	35%	0.2 (0.1–0.6)	0.001	3.0 (1.7–5.1)	<0.001	3.2 (1.8–5.6)	<0.001	HR
Stage 3 lower GI only	65	55%	0.5 (0.3–0.9)	0.02	2.1 (1.4–3.1)	0.003	2.7 (1.8–4.2)	<0.001	HR
Stage 1–3 skin + stage 2 lower GI	54	52%	0.5 (0.3–0.9)	0.01	1.5 (1.0–2.3)	0.08	1.6 (1.0–2.6)	0.06	HR
Stage 3–4 lower GI + (stage 1–3 skin or liver stage 1–4)	55	36%	0.2 (0.1–0.4)	<0.001	2.5 (1.7–3.6)	<0.001	3.0 (2.0–4.5)	<0.001	HR
Stage 1–4 liver alone	25	48%	0.3 (0.2–0.8)	0.01	1.0 (0.5–2.3)	0.96	1.7 (0.8–3.8)	0.19	HR
Stage 1–3 skin + stage 3–4 lower GI + stage 1–4 liver	13	8%	0.1 (0.01–0.3)	0.002	4.3 (2.3–8.2)	<0.001	7.4 (3.6–15.2)	<0.001	HR
Stage 4 skin only	13	38%	0.3 (0.1–0.8)	0.02	0.8 (0.2–3.1)	0.71	2.2 (0.7–6.9)	0.16	HR
Stage 4 lower GI only	22	22%	0.1 (0.03–0.7)	0.02	3.5 (1.5–7.9)	0.003	3.1 (1.2–8.1)	0.02	HR

GVHD, graft versus host disease; CR, complete response; PR, partial response; CI, confidence interval; TRM, treatment-related mortality; GI, gastrointestinal; StdR, standard risk; HR, high risk.

Table 2.

Day 28 response rate and 6-month nonrelapse mortality by risk scoring systems.

Refined Minnesota aGVHD Risk Score	Day 28 RR		6-month NRM		n
	CR	PR	%	95% CI
StdR	48%	21%	22	20–24	1454
HR	27%	16%	44	38–50	269
Ann Arbor GVHD Risk Score	CR/PR
1		81%	8	3–16	74
2		68%	27	20–34	165
3		46%	46	33–58	61

aGVHD, acute graft versus host disease; CR, complete response; CI, confidence interval; HR, high risk; NRM, nonrelapse mortality; PR, partial response; RR, response rate; StdR, standard risk.

Levine and colleagues also recently validated the Ann Arbor biomarker risk score based on plasma levels of tumor necrosis factor receptor-1 (TNFR-1), regenerating islet-derived 3-alpha (REG3α) and suppression of tumorigenicity 2 (ST2).²² A risk score of 3 demonstrated a 46% ORR compared with AA1/AA2 patients who had an 81% and 68% response rate, respectively. A risk score of 3 was also associated with a less durable response (defined as a complete response for ⩾6 months without recurrence of symptoms), higher likelihood of later developing SR lower GI GVHD and had an NRM of 46% at 6 months regardless of clinical severity of GVHD (Table 2). By utilizing new therapies based on risk stratification models, we may be able to improve outcomes, while minimizing toxicity.

Steroid-free acute graft versus host disease therapy

Sirolimus

Sirolimus (also known as rapamycin) is a mammalian target of rapamycin (mTOR) inhibitor that has immunosuppressive activity in addition to its antineoplastic properties.²³ Inhibition of mTOR impairs T-cell signaling, and its use as a prophylactic agent and as a second-line agent for SR-GVHD have been previously reported.^24–27 However, a recent retrospective analysis evaluated the use of sirolimus monotherapy as upfront treatment in 32 patients with newly diagnosed aGVHD who were poor candidates for steroids due to older age, high risk of toxicity, or at high risk for relapse.²⁸ A total of 27 patients were identified as Minnesota standard risk and five were high risk. Of note, no patients with grade IV aGVHD were included in this cohort. A total of 16 (50%) patients achieved a complete response (CR) with sirolimus versus 59% of patients from a matched cohort treated with steroids. These data indicate that use of sirolimus as a steroid-free primary therapy would potentially be safe and effective for aGVHD. The Bone Marrow Transplant Clinical Trials Network (BMT CTN) 1501 trial is a phase II trial investigating the use of single-agent sirolimus versus prednisone for frontline treatment of standard-risk aGVHD by refined Minnesota risk criteria, with further confirmation based on the Ann Arbor biomarker risk score [ClinicalTrials.gov identifier: NCT02806947].

Kinase inhibitors

Janus kinases

The Janus kinase (JAK) family is composed of four tyrosine kinases that function as signal transducers of cytokine-mediated pathways to control cellular proliferation, survival and differentiation.²⁹ JAKs associate with a variety of cytokine receptors, and activation of JAK leads to phosphorylation of signal proteins of the signal transducers and activators of transcription (STAT) family, which act as transcription factors for pro-inflammatory genes.³⁰ JAK1 and JAK2 are both widely expressed, and their inactivation leads to impairment of immune function via cytokine-induced intracellular signaling of lymphocytes.³¹

While inhibition of JAK1/2 (nonspecific inhibition) significantly reduces T-cell function, recent evidence suggests that specific inhibition of JAK1 may be the primary mediator of response.³² More specific targeting may ameliorate off-target effects such as myelosuppression which is primarily mediated by inhibition of JAK2-associated receptors required for hematopoietic cell development and proliferation.³³ There is also evidence that JAK inhibition impairs the differentiation and function of dendritic cells, the most important APCs, causing reduced T-cell activation.^34,35 These data support the potential role of JAK inhibition for treatment of GVHD.

Ruxolitinib

Ruxolitinib is a JAK 1/2 inhibitor that has been shown to have a significant role in signaling pathways mediating inflammation in acute and chronic GVHD,³⁶ in addition to roles in adaptive and innate immune responses. It has been shown that interferon gamma receptor (INFγR) signaling is mediated via JAK1/JAK2 and is upregulated in activated Tcells.³⁷ T cells deficient in INFγR cause significantly less GVHD compared with wild-type T cells, and pharmacologic inhibition of JAK1/2 in mouse models leads to less GVHD, in addition to preserving anti-leukemic effects previously reported.^37,38 Spoerl and colleagues showed that JAK 1/2 inhibition led to reduction of pro-inflammatory cytokines and an increase in the frequency of regulatory T cells (Tregs) in mouse models with improved survival in mice who developed acute GVHD and reduced histopathological grading.³⁰ Based on this observation, they trialled six patients with heavily treated SR-aGVHD with ruxolitinib. Two patients with GI GVHD had complete resolution of symptoms during treatment, one patient with liver GVHD had complete normalization of bilirubin, and all four patients with skin GVHD had a response to ruxolitinib therapy. All patients also had a reduction in serum levels of pro-inflammatory cytokines [interleukin-6 (IL-6) and IL-2R], and the amount of corticosteroids required.

In a multicenter survey study done in centers throughout Europe and the US, 54 patients with SR-aGVHD (grade III or IV) were treated with off-label use of ruxolitinib³⁹ at a dose of 5–10 mg orally twice daily. The ORR was 81.5% (44/54) with 25 (46.3%) patients achieving a CR. The median time to response was 1.5 (range 1–11) weeks, and only 3 patients experienced a GVHD-relapse, suggesting durable response was achievable. The 6-month survival was 79%. Cytomegalovirus (CMV) reactivation and cytopenias were the most common adverse events (AEs) observed. CMV reactivation occurred in 33% of patients, all of which responded to antiviral treatment. Grade III/IV cytopenias, a previously established side effect of ruxolitinib, occurred in 33% of patients. Several phase II and III trials are currently ongoing for SR-aGVHD (Table 3).

Table 3.

Clinical trials of tyrosine kinase inhibitors (registered through ClinicalTrials.gov).

Trial number	Study design	Diagnosis	Intervention	Location	Status
Ruxolitinib
NCT02953678	Phase II	SR-aGVHD	Ruxolitinib + CCS	USA	Recruiting
NCT03147742	Open label, expanded access	SR-aGVHD / SR-cGVHD	Ruxolitinib	USA	Recruiting
NCT03112603	Phase III	SR-cGVHD	Ruxolitinib versus BAT	USA	Not yet recruiting
NCT02997280	Phase II	SR-aGVHD / SR-cGVHD	Ruxolitinib	Russia	Recruiting
NCT02396628	Phase II	SR-aGVHD	BAT +/–ruxolitinib	Germany	Recruiting
NCT02913261	Phase III	SR-aGVHD	Ruxolitinib versus BAT	Europe, Asia, Australia, Canada	Recruiting

SR, steroid refractory; aGVHD, acute graft versus host disease; cGVHD, chronic graft versus host disease; BAT, best available therapy; CCS, corticosteroids.

Itacitinib

Results from a phase 1 trial of selective JAK1 inhibitor itacitinib (INCB039110) in patients with aGVHD were reported at the 2016 American Society of Hematology Meeting (ASH).⁴⁰ Patients with grades II–IV aGVHD were randomized 1:1 to a dose of 200 or 300 mg orally once daily of INCB039110 combined with steroids, and were stratified based on treatment-naïve (n = 14) versus SR (n = 17) disease. Preliminary results showed a day 28 ORR of 88.3% when used in the first-line setting as treatment for aGVHD, and 64.7% for SR disease. GRAVITAS-301 is a phase III trial currently recruiting for front-line treatment of aGVHD with itacitinib versus placebo in combination with steroids [ClinicalTrials.gov identifier: NCT03139604]. A phase II trial of upfront use of a JAK1 inhibitor or placebo in combination with steroids is currently being planned through the BMT CTN.

Proteasome inhibitors

Proteasome inhibitors (PIs) are drugs that inhibit the proteasome, which are enzyme complexes that are responsible for regulating protein turnover within cells. They have been shown to have immune-modulating effects in murine models of aGVHD and cGVHD, including deletion of alloreactive T cells, inhibition of APCs, inhibition of IL-6, increased survival of Tregs, and decrease in levels of B-cell activating factor (BAFF).⁴¹ Several PIs have been US Food and Drug Administration (FDA) approved for the treatment of multiple myeloma.

Bortezomib

Bortezomib is a first-generation, reversible PI that directly induces apoptosis of tumor cells by suppressing tumor-survival pathways and arresting cell growth.⁴² It is a potent inhibitor of nuclear factor-kappa B, which regulates the transcription of multiple pro-inflammatory genes, and is important for T-cell proliferation, activation and survival.⁴³

Several preclinical and early-phase studies have shown bortezomib to be effective in preventing aGVHD,^44–47 although in one murine study, delayed or prolonged administration was associated with severe colonic toxicity.⁴⁵ Given the critical role of T cells and pro-inflammatory cytokines in the development of aGVHD, bortezomib would be a potential therapeutic candidate.

Results of a small phase II study [ClinicalTrials.gov identifier: NCT00408928] evaluating the use of bortezomib for SR-aGVHD, the largest to date, were recently reported by Wagner and colleagues.⁴⁸ Eleven patients were treated on study with eight patients being evaluable for response. Two patients (25%) achieved a CR, and two patients had a partial response (PR). However, 10 patients died, with the majority of deaths attributed to infection (60%), followed by progressive disease (20%). One patient also died of hemorrhage and pulmonary complication, respectively. The authors were unable to determine if any of the deaths were related to the use of bortezomib. While bortezomib appears to have some activity in treating aGVHD, additional studies are needed to better determine the safety and toxicity profile.

Cytokine modulation

Alpha-1 antitrypsin

Alpha-1 antitrypsin (AAT) is a serine protease inhibitor that is able to modulate immune and inflammatory functions through alteration of cytokine profiles.⁴⁹ Low AAT plasma levels in human donors were found to be associated with a higher rate of aGVHD in recipients⁵⁰ and GVHD severity increased with decreasing AAT levels as well. An inverse relationship between AAT levels and GVHD was noted, with high levels of ATT exerting a protective effect against GVHD while maintaining, or even enhancing, graft versus leukemia (GVL) activity⁵⁰ and decreasing mortality in mouse models.⁵¹ Donors treated with ATT had an increase in Tregs, as well as a 50-fold increase of IL-10, favoring an anti-inflammatory profile.

Based on these observations, a prospective phase I/II dose-escalation study was performed to evaluate the use of ATT in the treatment of SR-aGVHD. Twelve patients were enrolled in the first two cohorts (all grade III or IV GVHD with stage 3–4 lower gastrointestinal (LGI) involvement), and eight patients achieved a response, with four patients achieving a CR.⁵² There are currently two clinical trials underway to further evaluate the use of ATT for GVHD. One is a phase II/III study for treatment of newly diagnosed LGI aGVHD in combination with steroids [ClinicalTrials.gov identifier: NCT02956122], and the other is an early-access trial for patients with SR-aGVHD [ClinicalTrials.gov identifier: NCT03172455]. A pilot study for use of ATT in SR-aGVHD has completed enrollment, but results are not yet available [ClinicalTrials.gov identifier: NCT01700036].

Interleukin-22

IL-22 is a cytokine produced by cells from both the adaptive and the innate system including CD4⁺ T cells, CD8⁺ T cells, natural killer cells, and gamma-delta (γδ) T cells.⁵³ It serves an essential role in host defense against extracellular pathogens, strengthens epithelial barrier function by acting upon intestinal stem cells (ISCs), and helps with tissue repair and wound healing. In addition to its protective properties, it has pro-inflammatory properties when produced in excess.^53,54 Studies of murine models transplanted with IL-22-deficient grafts showed loss of ISCs, development of more severe aGVHD and increased mortality compared with wild-type models.^55,56 Lindemans and colleagues showed that treatment of murine models with IL-22 in vivo reduced intestinal pathology and mortality from GVHD.⁵⁷ There is also a potential relationship between IL-22 and the microbiome, as microbacteria has been shown to induce IL-22 secretion that then induces the production of antimicrobial proteins.^57–59 There is currently a phase I/II clinical trial underway [ClinicalTrials.gov identifier: NCT02406651] evaluating the safety and feasibility of use of IL-22 in combination with corticosteroids for the treatment of newly diagnosed grade II–IV acute LGI GVHD.

Monoclonal antibodies

Natalizumab

Natalizumab is a humanized monoclonal antibody (mAb) against α4-integrin containing adhesion molecules which are widely expressed on leukocytes, primarily lymphocytes.⁶⁰ Based on studies done in inflammatory bowel disease (IBD), it has been recognized that the endothelium plays a key role in the pathogenesis of inflammation and in mucosal immune trafficking.⁶¹ Natalizumab inhibits adhesion molecules, preventing leukocyte migration from the circulation into inflamed gut mucosa.^60,62 It is currently FDA approved for use in multiple sclerosis and Crohn’s disease. Despite its proven efficacy in these diseases, use has been limited due to the possible side effect of progressive multifocal leukoencephalopathy (PML) – a potentially life-threatening demyelinating neurologic disease. There has been a reported incidence of 4.2 cases per 1000 patients treated.^63,64

Acute GI GVHD is often propagated after intestinal damage from transplant-conditioning regimens, leading to significant inflammation in the gut lining. Therefore, it would seem plausible that natalizumab might be able to mitigate GI GVHD by preventing homing of leukocytes to the GI tract. Patients are currently being recruited for two phase II studies evaluating the safety and efficacy of natalizumab for treatment of GVHD. One study is evaluating its use in combination with steroids for initial treatment of acute GI GVHD [ClinicalTrials.gov identifier: NCT02176031]. The second study is evaluating the combination for treatment of high-risk aGVHD (based on Ann Arbor risk score) with the primary outcome being rate of CR at day 28 [ClinicalTrials.gov identifier: NCT02133924].

Vedolizumab

Vedolizumab is a mAb that specifically inhibits α4β7 integrins located on activated T cells from interacting with MAdCAM-1 (located on gut epithelium) preventing homing to the intestinal mucosa.^65,66 It is currently FDA approved for the treatment of Crohn’s disease and ulcerative colitis. Several preclinical murine studies have shown that α4β7 integrins play an important role in the development of intestinal GVHD.^67–69 A recent case series describing the off-label use of vedolizumab for treatment of SR-aGVHD of the GI tract showed promising results.⁷⁰ Five patients with clinical grade IV GI GVHD were treated; four patients had isolated GI manifestations and one patient had multiorgan involvement.

Four patients were able to discontinue systemic steroids after receiving three doses of vedolizumab, and two patients were off all immunosuppressant medications at the time of last follow up. Clinical responses were noted within 7–10 days with patients receiving a variable number of infusions.⁷⁰ The median time to follow up was 10 months, with four patients alive and receiving outpatient care. No associated toxicities were mentioned. Safety and toxicity, in addition to clinical efficacy will need to be confirmed in phase II/III trials. A multicenter phase II study recently opened and is recruiting [ClinicalTrials.gov identifier: NCT02993783].

Brentuximab vedotin

Brentuximab vedotin (BV) is an anti-CD30 antibody–drug conjugate composed of the antimicrotubule agent monomethyl auristatin E (MMAE) and a human CD30 antibody. It has been approved for the treatment of relapsed classical Hodgkin lymphoma, systemic anaplastic large cell lymphoma and for postautologous HSCT consolidation of those CD30⁺ diseases. CD30 is a tumor necrosis factor receptor family member and is highly expressed on activated lymphocytes with minimal expression on normal tissues.⁷¹ Infiltrating lymphocytes in chronic autoimmune and inflammatory diseases such as rheumatoid arthritis and systemic sclerosis have also been shown to express high levels of CD30.^72,73 In preclinical studies, Chen and colleagues demonstrated that patients with aGVHD had a higher percentage of CD30 expressing CD8⁺ T cells, significantly higher plasma levels of soluble CD30, and increased CD30⁺ lymphocytes in affected intestinal tissue compared with patients without aGVHD.⁷⁴

Based on these results, a multicenter phase I trial was initiated to investigate the use of BV for the treatment of SR-aGVHD.⁷⁵ Thirty-four patients were treated in the dose-escalating study to determine maximum tolerated dose (MTD). The study initially had a schedule of weekly dosing for 3 weeks followed by maintenance dosing every 3 weeks for four additional doses. However, after treatment of six patients with death of two patients from neutropenic sepsis, the study was revised to dosing every 2 weeks for four doses with cohorts of five patients per dose level (0.6 mg/kg, 1.0 mg/kg, and 1.2 mg/kg). The dose-limiting toxicity (DLT) was defined at 0.8 mg/kg after one patient developed sepsis and multiorgan failure. Grade III toxicities included neutropenia, headache, hypoxia, ileus and elevated bilirubin. No peripheral neuropathy of any grade was observed.

The day 28 ORR was 38.2% with five (14.7%) patients achieving a CR and eight (23.5%) patients achieving a very good PR (VGPR). At day 56, seven additional patients had achieved CR; three of whom had previously achieved VGPR. Responses were not significantly different for individual organ involvement: intestinal (n = 10; 42%), liver (n = 5; 45%), and skin (n = 5; 28%), however the study was not designed to assess efficacy. The OS at 6 and 12 months follow up was 41% and 38%, respectively. Of those patients (n = 16) who did not respond to BV, only one remained alive at 12 months. Causes of death in the other patients were: GVHD (n = 11), infection (n = 3), and disease relapse (n = 1). Larger randomized trials are needed to confirm efficacy.

Multiple mAbs targeting different pathways in the pathophysiology of aGVHD have been evaluated in the past. The responses and toxicities of several of the older agents compared with newer aAbs are summarized in Table 6. However, cross comparison of these studies is limited due to multiple factors, including differences in study design (the majority of reports are retrospective series), varying definitions of SR-aGVHD, heterogeneous patient populations, variation in time-to-response assessments, and use of successive salvage agents.

Table 6.

Responses and toxicities of conventional therapies compared with new monoclonal antibodies.

Drug/treatment	Study design	Line of tx	n	ORR	CR	OS	Toxicities	Reference
Anti-CD52
Alemtuzumab	Retrospective	SR-aGVHD Grade II–IV	18	83%	33%	55%	Infections 78% CMV 61% Neutropenia 43%	Gómez-Almaguer et al.168
	Retrospective	GI SR-aGVHD Grade III–IV	20	70%	35%	50% (1 year)	Bacterial 40% Aspergillosis 35% EBV 15% CMV frequent	Schnitzler et al.169
	Phase II	SR-aGVHD Grade III–IV	10	55%	20%	0%	Infections 80% CMV 70% Cytopenias 60%	Martínez et al.170
	Phase II	SR-aGVHD Grade III–IV	18	99%	28%	35%	Bacterial 61% Viral 44% Fungal 22% CMV 56% EBV 11%	Schub et al.171
Anti-TNF
Infliximab	Retrospective	SR-aGVHD Grade II–IV	21	67%	62%	38% (21 mos)	Bacterial 81% Fungal 48% Viral 67%	Couriel et al.172
	Retrospective	SR-aGVHD Grade II–IV	32	59%	19%	68%	Infections 72% CMV 41%	Patriarca et al.173
Etanercept	Retrospective	SR-aGVHD Grade II–IV SR-cGVHD	13 8	46% 62%	31% 12%	69%	Bacterial 14% Fungal 19% CMV 48%	Busca et al.174
	Phase I/II	SR-cGVHD	10	60%	10%	–	Cause of death: infection 10%	Chiang et al.175
Immunmodulatory
Mycophenolate mofetil	Phase II	SR-aGVHD Grade II–IV SR-cGVHD	13 13	31% 77%	15% –	33% (2 years) 54% (2 years)	Gastrointestinal 27% Infections 23% CMV 11%	Kim et al.176
	Retrospective	SR-aGVHD Grade II–III SR-cGVHD	10 11	60% 64%	0% –	70% –	Infection 67% Hematologic 29%	Krejci et al.177
	Phase II	SR-aGVHD Grade II–IV	19	47%	31%	16%	Cause of death: Infection 32%	Furlong et al.178
Sirolimus	Pilot	SR-aGVHD Grade III–IV	21	57%	24%	33%	TCP 33% Neutropenia 19% TG 38% TMA 23%	Benito et al.26
	Retrospective	SR-aGVHD	34	76%	44%	44%	TG 71% TMA 21%	Hoda et al.179
	Phase II	SR-cGVHD	35	63%	17%	57% (3 years)	Infections 77% TMA 11% Renal 66% TG 18%	Couriel et al.24
Pentostatin	Phase I	SR-aGVHD Grade II–IV	23	77%	63%	26%	TCP 4% Infection 9%	Bolaños-Meade et al.180
	Retrospective	SR-aGVHD Grade II–IV	12	50%	33%	10%	Bacterial 50% Fungal 25% Viral 8% CMV 25%	Pidala et al.181
	Phase II	SR-cGVHD	58	55%	–	70% (2 years)	Infection 20%	Jacobsohn et al.182
Cellular photoimmunotherapy
Extracorporeal photopheresis	Pilot	SR-aGVHD Grade II–IV	21	67%	60%	57%	Anemia 90% TCP 71% CMV 52%	Greinix et al.183
	Phase I/II	SR-aGVHD Grade II–IV SR-cGVHD	9 14	78% 64%	56% 29%	56% (8 months) 79% (36 months)	Hypotension Anemia CMV reactivation	Salvaneschi et al.184
	Retrospective	SR-aGVHD Grade II–IV	23	–	52%	48%	Hypotension Anemia TCP	Perfetti et al.185
	Retrospective	SR-cGVHD	71	61%	20%	53%	Anemia Thrombocytopenia	Couriel et al.186
Anti-α4β7 integrin
Vedolizumab	Case series	SR-aGVHD GI grade IV	6	100%	–	4/6 pts alive at median f/u 10 months	None reported	Floisand et al.70
Anti-CD30
Brentuximab vedotin	Phase I	SR-aGVHD	34	38%	15%	38%	Neutropenia Elevated bilirubin Headache Ileus	Chen et al.75
Anti-CTLA4
Abetacept	Phase I	SR-cGVHD	16	44% (PR)	–	–	Pulmonary infection Diarrhea/gastritis Rash	Nahas et al.132

aGVHD, acute graft versus host disease; cGVHD, chronic graft versus host disease; CMV, cytomegalovirus; CR, complete response; EBV, Epstein-Barr virus; f/u, follow up; GI, gastrointestinal; ORR, overall response rate; OS, overall survival; PR, partial response; PTLD, post-transplant lymphoproliferative disease; pts, patients; ref, reference; SR, steroid refractory; TCP, thrombocytopenia; TG, triglyceridemia; TMA, thrombotic microangiopathy; TNF, tumor necrosis factor; tx, treatment.

Adoptive cell therapy

Mesenchymal stromal cells

Mesenchymal stromal cells (MSCs) are pluripotent stem cells that are able to differentiate into multiple cell lineages of mesenchymal origin including osteoblast, chondrocytes, and adipocytes.^76–79 They have been shown to inhibit B- and T-cell activation, block the function of APCs, inhibit natural killer cells, and increase Tregs.^78,80,81 MSCs can be isolated and expanded ex vivo from bone marrow, umbilical cord blood, adipose tissue, and placenta.^76,78,82 MSCs do not express class II human histocompatibility leukocyte antigens (HLA) and therefore do not provoke immunologic responses in HLA-mismatched recipients.

MSC therapies have been studied in phase I and II trials for the treatment of SR-aGVHD with overall responses ranging from 60–75%.^83–86 Results from two multicenter, randomized phase III trials led by Osiris Therapeutic were reported, in abstract form only, on the use of MSCs for treatment of de novo aGVHD and SR-aGVHD.^82,87 Neither trial reached its primary endpoint of durable CR ⩾28 days. However, patients with SR liver and GI GVHD treated with Prochymal, the Osiris product, were reported to have significantly improved response rates (76% versus 47% and 82% versus 68%, respectively; p = 0.03 for both).

Hashmi and colleagues performed a systematic review and meta-analysis to assess response and survival in patients with SR-aGVHD treated with MSCs.⁸⁸ Thirteen nonrandomized studies including 336 patients were included, and six studies (n = 119) provided data on primary outcome of 6-month survival following treatment with MSCs. The weighted 6-month survival was 63% [confidence interval (CI) 50–74%] compared with the weighted average of 49% from 29 non-MSC studies of American Society of Blood and Marrow Transplantation (ASBMT) that were used to formulate guidelines for recommended second-line therapies. However, this must be interpreted with caution as there were many limitations of the study, including differences in timing of administration of MSCs, heterogeneity of study populations, variable definitions of SR disease, and variability in the dose and quality of the MSC product. Thus, direct comparison of MSCs with other second-line agents in prospective, randomized trials are needed to reach a definitive conclusion on survival advantage. There are currently four studies for upfront use and nine studies for SR-aGVHD underway utilizing MSCs for treatment of aGVHD (Table 5).

Table 5.

Clinical trials utilizing mesenchymal stem cells (registered through ClinicalTrials.gov).

Trial identifier	Study design	Diagnosis	Intervention	Location	Status
NCT00603330	Phase II	SR-aGVHD	MSC	Germany	Recruiting
NCT02770430	Phase II	SR-aGVHD	MSC	Brazil	Recruiting
NCT02055625	Phase I/II	Oral cGVHD	MSC	Sweden	Recruiting
NCT02379442	Phase I/II	aGVHD	MSC + CCS	NIH	Recruiting
NCT03158896	Phase I	HR or SR-aGVHD	MSC	USA	Not yet Recruiting
NCT02923375	Phase I	SR-aGVHD	MSC	Australia, UK	Recruiting
NCT02241018	Phase I/II	SR-aGVHD	Anti-CD25 + CNI +/– MSC	China	Recruiting
NCT02291770	Phase III	cGVHD	First-line tx +/– MSC	China	Recruiting
NCT02687646	Phase I/II	SR-aGVHD	MSC	Spain	Recruiting
NCT02032446	Phase I/II	SR-aGVHD	MSC	Italy	Recruiting
NCT02359929	Phase I	SR-aGVHD / SR-cGVHD	MSC	Emory	Recruiting
NCT02336230	Phase III	SR-aGVHD	MSC	USA	Recruiting

SR, steroid refractory for NCT03158896; aGVHD, acute graft versus host disease; cGVHD, chronic graft versus host disease; HR, high risk; CCS, corticosteroids; CNI, calcineurin inhibitor; MSC, mesenchymal stem cells; tx, treatment; NIH, National Institutes of Health.

Microbiome restoration

Fecal microbiota transplantation

Alteration of intestinal homeostasis, via disruption and damage to the gastrointestinal epithelium, has emerged as having a critical role in the development and severity of acute GVHD involving the GI tract.^89,90 Dysbiosis, or loss of intestinal diversity, has been associated with increased TRM.^91–93 Obligate anaerobes, particularly members of the Clostridiales order, have been identified as important mediators of intestinal homeostasis by upregulation of intestinal Tregs and inhibition of inflammation by producing anti-inflammatory cytokines.⁹⁴ Increased abundance of intestinal Blautia (a member of the Clostridia class) has been associated with reduced GVHD mortality and improved OS.⁹⁵

Two recent small case series utilized fecal microbiota transplants (FMTs) in an attempt to restore normal intestinal flora. The first study evaluated the safety and efficacy of FMT in patients with SR or steroid-dependent aGVHD. Out of a total of four patients, three had achieved complete resolution of symptoms at day 28 GVHD assessment, and one had a PR.⁹⁶ In a study evaluating three patients with SR-acute GI GVHD who received FMT on a compassionate use basis; two patients had resolution of all signs and symptoms of GVHD and one patient had partial resolution.⁹⁷ No significant severe AEs, including infection, were attributed to the procedure in either study. In the latter study, all three patients ultimately died from non-GVHD causes with duration of response lasting 8–9 weeks at time of death. Early-phase trials are currently in development at MD Anderson Cancer Center and Memorial Sloan Kettering to determine if these results are reproducible.

Chronic graft versus host disease

While aGVHD is driven mainly by mature donor T cells, cGVHD has been discovered to involve a more complex immune reaction with both T and B cells contributing to the underlying pathology. Although donor antibodies to recipient antigens are known to play a role in cGVHD, the exact mechanisms of how these cells contribute to the underlying pathology are still being investigated. BAFF has been identified as a key regulator of B-cell homeostasis, and high levels have been shown to rescue self-reactive B cells from peripheral deletion,⁹⁸ promoting survival and differentiation. Sarantopoulos and colleagues showed that patients with active cGVHD had significantly elevated levels of BAFF and increased signaling through the ERK and AKT pathways (both of which are activated via BAFF) which was associated with decreased apoptosis of activated B cells.^99,100

In addition, distinct subsets of B cells known as regulatory B cells (Bregs) have been identified¹⁰¹ to have immunosuppressive effects via IL-10 production that inhibit CD4⁺ T-cell proliferation and cytokine production.¹⁰² Sarvaria and colleagues demonstrated that patients with cGVHD had reduced levels of circulating Bregs and impaired IL-10 production compared with patients without cGVHD^102,103 suggesting Breg therapy might be a potential therapeutic target. In-depth reviews of cGVHD biology have been recently published by Im and colleagues⁴¹ and Reddy and colleagues.¹² A better understanding of potential mechanisms contributing to the pathophysiology of cGVHD has led to development of targeted therapies that will be discussed below.

Chronic GVHD is a multiorgan disease associated with significant immunodeficiency which makes treatment with immunosuppressive medications challenging due to the increased risk of severe, life-threatening infections.^3–5 As previously mentioned, steroids are the most widely used first-line therapy for treatment of moderate-to-severe cGVHD with or without the addition of other immunosuppressive agents.⁴ If patients fail to respond or have progressive disease with steroid therapy, second-line treatment is required. The response rate to second-line therapy is reported to be 25–50% based primarily on phase II trials, with no single therapy being better than any other.^104,105 Choice of therapy is generally based on patient comorbidities, organs involved, and physician experience.

Kinase inhibitors

Janus kinase inhibitors

In addition to mechanisms previously elucidated above, it has been demonstrated that patients with cGVHD have upregulated IFN-inducible genes and elevated levels of INF-induced chemokines which are involved in CXCR3⁺-mediated lymphocyte trafficking.¹⁰⁶ JAK1/JAK2 are mediators of INFγR signaling and inhibition of the JAK pathway results in decreased expression of CXCR3, reduced GVHD, and improved survival in murine models.³⁷ Data from preclinical studies and observational human studies have identified JAK-STAT inhibition as an exciting therapeutic target.

Ruxolitinib

In the previously mentioned multicenter survey study, 41 patients with moderate-to-severe SR-cGVHD were also included. The ORR was 85.4% with 32 (78%) patients achieving a PR and three patients achieving a CR. The median time to response was 3 (range 1–25) weeks, with responses noted in all involved organ systems. Only two patients (5.7%) experienced a GVHD relapse, and 14.6% (6/41) of patients had no response. CMV reactivation was and cytopenias were observed in 17% and 14.6% of patients, respectively. However, caution must be taken when interpreting these results due to potential recall bias, as the reported response rates were based on retrospective chart review. Evaluation and assessment of responses in cGVHD, based on the National Institutes of Health (NIH) consensus criteria,¹⁰⁷ requires extensive examination and detailed documentation in order to capture accurate responses. It often takes 4–6 weeks to note objective improvement in cGVHD following initiation of therapy, although they reported observing improvements as early as 1 week after treatment was started.

Larger, prospective randomized controlled trials are needed to determine the true response rate based on standardized grading of responses, and to determine the optimal dosing schedule and duration of treatment required for long-term control of GVHD. Several phase II and III trials are currently ongoing (Table 3).

Baricitinib

Baricitinib is an oral JAK1/2 inhibitor developed for the treatment of rheumatoid arthritis (RA), atopic dermatitis, and systemic lupus erythematosus (SLE). It received its first global approval in February 2017 for refractory RA in the European Union. There is a phase I/II study underway evaluating its use in the treatment of SR-cGVHD [ClinicalTrials.gov identifier: NCT02759731].

Ibrutinib

Ibrutinib is an irreversible inhibitor of Bruton’s tyrosine kinase (BTK) and IL-2 inducible T-cell kinase (ITK). BTK is primarily responsible for the activation of signaling pathways involved in B-cell proliferation, trafficking and adhesion, while ITK is involved in the secretion of IL-2 and T-helper 2 cytokines. While the exact involvement of BTK and ITK in the pathogenesis of cGVHD is not known, given the role of T- and B-cell dysregulation in the development of cGVHD,^108–110 there have been investigations into targeting the B-cell receptor (BCR) and T-cell receptor (TCR) signaling pathways.

In studies of murine models of cGVHD, treatment with ibrutinib was shown to delay disease progression, increase GVHD progression-free survival, and improve both clinical and pathological findings in the T-cell mediated model. Improved pulmonary function and tissue immunoglobulin deposition was noted in an alloantibody-driven model that induced bronchiolitis obliterans (BO).¹¹¹ In contrast to murine studies by Schutt and colleagues that showed prophylactic ibrutinib to be effective in preventing cGVHD in mice,¹¹² Dubovsky and colleagues did not see any reduction in cGVHD in their mouse model when treated with ibrutinib prophylactically starting day –2 through day 28.¹¹¹

Ibrutinib was recently granted FDA approval, the first drug to ever be approved for treatment of cGVHD, based on results from a multicenter phase II study evaluating the safety and efficacy of its use in patients with cGVHD who had failed one to three lines of prior systemic therapy.¹¹³ Forty-two patients with steroid-dependent or -resistant cGVHD were treated with a daily dose of 420 mg of ibrutinib. The ORR was 67% (28/42 patients; 21% CR; 45% PR). A total of 71% of responders had a sustained response of at least 20 weeks. Approximately one third of patients who responded were able to reduce the dose of steroids to ⩽0.15 mg/kg/day during the course of the study, and five patients were able to completely discontinue steroid therapy. Improvement occurred in multiple organ systems, with 42% (5/12) of patients with at least three involved showing a response in at least three organs, which were associated with improvement in quality of life.

Importantly, the drug was not without toxicity, with the most common AEs observed being fatigue (57%), diarrhea (36%), muscle spasms (29%), nausea (26%) and bruising (24%). Serious AEs occurred in 22 patients (52%) with ⩾grade III occurring in 17 patients (40%). Two deaths were reported due to multilobular pneumonia and bronchopulmonary aspergillosis. There is currently a phase III trial [ClinicalTrials.gov identifier: NCT02959944] underway evaluating the efficacy of corticosteroids plus ibrutinib versus placebo for the frontline treatment of new-onset moderate-to-severe cGVHD.

Spleen tyrosine kinase inhibitors

Spleen tyrosine kinase (Syk), predominantly expressed in hematopoietic cells, has recently been identified as being necessary for both TCR signaling activation,¹¹⁴ and activation of BCR signaling pathways leading to B-cell proliferation and survival.¹¹⁵ Syk activation also leads to activation of multiple immune cells (mast cells, macrophages, neutrophils) leading to production and release of pro-inflammatory cytokines. Based on these early observations, Syk inhibitors were initially tested in murine models and early stage clinical trials of various autoimmune diseases such as SLE, RA, and IgA nephropathy with varying degrees of success.^116–120 Also, there are currently several phase I/II trials evaluating the safety and efficacy of two different orally available Syk inhibitors, fostamatinib and entospletinib (GS-9973), for the treatment of relapsed/refractory B-cell malignancies.

Syk inhibitors have also shown efficacy in animal and human studies of both acute and chronic GVHD. Leonhardt and colleagues showed decreased aGVHD, improved survival and reduced inflammatory cytokine profile (IL-6, INF-γ, MCP-1) in mice treated with fostamatinib compared with those treated with cyclosporine A ¹¹⁴ without affecting T-cell cytolytic function. In a murine model of sclerodermatous cGVHD, administration of fostamatinib led to decreased fibrosis and a reduction in severity of sclerodermatous changes.¹²¹ Deletion of the Syk gene in murine donor bone marrow cells and in vivo inhibition with fostamatinib both showed reversibility of chronic lung GVHD manifestations in a mouse model with BO.¹¹⁵ Inhibition of Syk also led to apoptosis in vitro of human B cells from patients with active cGVHD.

Based on results from early-phase clinical trials in autoimmune diseases and preclinical data in cGVHD, entospletinib is currently being evaluated in a phase II trial in combination with systemic corticosteroids for first-line treatment of cGVHD [ClinicalTrials.gov identifier: NCT02701634]. Fostamatinib is currently being tested for prevention of cGVHD [ClinicalTrials.gov identifier: NCT02611063].

Rho kinase inhibitors

Rho-associated protein kinase (ROCK) is a serine-threonine kinase primarily involved in regulating cytoskeleton and cellular functions via phosphorylation of various downstream substrates.^122,123 Two isoforms have been identified: ROCK1 and ROCK2, sharing more than 90% homology within their kinase domain. Recent studies suggest that the ROCK2 isoform may play a specific role in the development of autoimmunity, as inhibition of ROCK2 in mice and human T-cells was effective in decreasing production of IL-17 and IL-21, both pro-inflammatory cytokines that have been identified as mediators of autoimmune disorders such as RA and SLE.^123–127

KD025 is an orally available selective ROCK2 inhibitor that downregulates phosphorylation of STAT3 which is a transcription factor necessary for the induction and expression of IL-17 and IL-21.^127–129 Inhibition of ROCK2 was also shown to increase regulatory T cells via phosphorylation of STAT5.^123,127 Recently, Flynn and colleagues were able to show reversal of cGVHD manifestations in two murine models after targeted inhibition of ROCK2, and further validated the KD025-mediated effects on STAT3 and STAT5 phosphorylation shifting the balance from a pro-inflammatory balance.¹³⁰ Based on these promising preclinical findings, there is currently a phase II trial underway to evaluate the safety, tolerability, and activity of KD025 for the treatment of SR-cGVHD [ClinicalTrials.gov identifier: NCT02841995].

Immune checkpoint blockade

Cytotoxic T-lymphocyte associated protein-4

Cytotoxic T-lymphocyte associated protein-4 (CTLA-4) is a negative regulator of T-cell immune function. T-cell activation requires multiple costimulatory signals, one of which is the binding of CD28 on T-cells with CD80 or CD86 on APCs. CTLA-4 is homologous to CD28, but binds with a greater affinity and avidity than CD28. It is expressed on activated T cells and binds to CD80/CD86, thereby blocking the interaction with CD28, and inhibits T-cell proliferation, differentiation and survival.¹³¹ Abatacept is a selective costimulation modulator composed of human CTLA-4 and a fragment of the fragment crystallizable (Fc) domain of human immunoglobulin-G1. Given the role that activated T cells play in GVHD, inhibition of T-cell activation via blockade of costimulatory molecules, distinct from direct inhibition of CTLA-4, could be a potential therapeutic option.

Results from a phase I trial evaluating the use of abatacept in patients with SR-cGVHD were presented at the 2016 ASH conference.¹³² A total of 17 patients were enrolled and 16 were treated with a planned total of six doses of abatacept. In all, 7/16 (44%) evaluable patients had a PR based on the NIH consensus criteria. Prednisone dosage was decreased by ~51% by 1 month following completion of the sixth dose. Severe AEs included one grade IV pulmonary infection and three grade III cases, all of which resolved. Other low-grade events included gastritis, diarrhea, fatigue, rash, and skin pain. The trial is currently ongoing. Overall, abatacept was well tolerated and based on these promising results, a phase II trial is being planned. There are also several trials underway using it in the preventive setting.

Cytokine modulation

Interleukin-2

IL-2 is a cytokine necessary for Treg development, expansion, activity, and survival.^133–135 As previously mentioned, Tregs play an important role in preventing cGVHD. Koreth and colleagues administered daily low-dose subcutaneous IL-2 to 29 patients with SR-cGVHD at three different dose levels (0.3 × 10⁶; 1 × 10⁶ or 3 × 10⁶ IU/m²) for a total of 8 weeks.¹³⁶ In all, 12 of 23 evaluable patients had major responses in multiple sites. Tregs were increased in all patients with minimal effect on conventional CD4⁺ T cells.^136,137 The highest dose level of 3 × 10⁶ IU/m² was not well tolerated due to persistent grade I constitutional symptoms (fever, malaise and arthralgias) requiring 50% dose reduction.¹³⁶ Overall, the therapy was safe and well tolerated at the lower dose levels.

In a phase II study by the same authors, 35 patients with SR-cGVHD were treated with daily IL-2 for 12 weeks with 61% of evaluable patients having a clinical response in multiple organ sites, including liver, skin, GI tract, lung, and joint/muscle/fascia.¹³⁸ The Treg cell numbers in the blood were again noted to be significantly elevated compared with baseline, with no significant change in conventional T cells. Several clinical trials are now recruiting to evaluate the use of IL-2 alone or in combination with Tregs or extracorporeal photopheresis (ECP) for treatment of SR-cGVHD (Table 4).

Table 4.

Clinical trials of interleukin-2 and regulatory T cells (registered through ClinicalTrials.gov).

Trial number	Study design	Diagnosis	Intervention	Location	Status
NCT01937468	Phase I	SR-cGVHD	LD IL-2 + Tregs	Dana Farber	Recruiting
NCT02340676	Phase II	SR-cGVHD	LD IL-2 + ECP	Dana Farber	Recruiting
NCT03007238	Phase II	SR-cGVHD	LD IL-2 + ECP	City of Hope	Recruiting
NCT02318082	Phase II	SR-cGVHD	Dose-escalated IL-2	Dana Farber	Recruiting
NCT02749084	Phase I/II	SR-cGVHD	Donor Treg DLI	Italy	Recruiting
NCT02385019	Phase I/II	SR-cGVHD	Donor Treg	Portugal	Recruiting
NCT02519816	Phase II	SR-cGVHD	TCD + Treg Expansion	Canada	Recruiting

SR, steroid refractory; cGVHD, chronic graft versus host disease; DLI, donor lymphocyte infusion; ECP, extracorporeal photopheresis; LD IL-2, low-dose interleukin-2; TCD, T-cell depletion; Tregs, regulatory T cells.

Proteasome inhibitors

Bortezomib

As previously mentioned, bortezomib inhibits nuclear factor-kappa B which is important for B-cell development, activation, and survival in addition to T cells.¹³⁹ Prior studies have shown that bortezomib has inhibitory effects on activated B cells and plasma cells,^140,141 and case reports of multiple myeloma patients treated with bortezomib for relapsed disease or maintenance therapy following HSCT showed encouraging improvement in cGVHD symptoms.^142–144 A phase II trial evaluating use of bortezomib in combination with steroids for initial therapy of cGVHD reported an ORR at week 15 of 80% with two (10%) complete responses and 14 (70%) PRs in 20 evaluable patients.¹³⁹ Responses were seen in most organs, and it was possible to taper the median steroid dose from 50 mg/day to 20 mg/day by week 15 (p < 0.001). However, progression of cGVHD was observed after cessation of bortezomib. While these results are encouraging, lack of a comparator arm makes it difficult to discern the true impact of adding bortezomib to steroids. Larger, well-designed randomized trials are needed to determine if there is a true benefit. There is currently a phase II trial underway for upfront treatment of bronchiolitis obliterans [ClinicalTrials.gov identifier: NCT01163786].

In a single-arm pilot study of bortezomib in patients with steroid-dependent, -intolerant, or refractory cGVHD, five of six (83%) patients achieved a PR defined as a decrease in ⩾1 point on a 0–3-point scale per NIH consensus criteria.¹⁴⁵ Most responders were also able to reduce the dose or number of immunosuppressive therapies being used during treatment with bortezomib. Flow cytometry revealed a significant decrease in B cells and an increase in Tregs. These findings suggest that proteasome inhibition may be a potential targeted therapy but additional studies with longer follow up are needed. A phase II trial is underway for SR-cGVHD [ClinicalTrials.gov identifier: NCT01158105].

Carfilzomib

Carfilzomib is an irreversible, second-generation PI that potently inhibits the 26S proteasome. It has been shown to have fewer off-target side effects, specifically, less peripheral neuropathy, compared with bortezomib.¹⁴⁶ It is currently being studied for treatment of SR-cGVHD in a phase II study [ClinicalTrials.gov identifier: NCT02491359].

Ixazomib

Ixazomib is a reversible second-generation oral PI that inhibits the 20S proteasome.¹⁴⁶ It has been well tolerated in patients treated with multiple myeloma, with the main toxicity being mild GI effects. It is being tested in a phase II trial for SR-cGVHD [ClinicalTrials.gov identifier: NCT02513498].

Anti-CD20 monoclonal antibodies

The role of B-cell dysregulation in the underlying pathophysiology of cGVHD, as described above, has led to the identification of B-cells as a therapeutic target in treating cGVHD.

Rituximab is a chimeric mouse/human IgG anti-CD20 antibody that mediates B-cell lysis via antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), and induction of apoptosis. It is currently FDA approved for treatment of CD20 positive B-cell malignancies. Most data reported on its use in cGVHD are based on case reports, small early-phase trials, or retrospective studies; with ORRs ranging from 50% to 86% reported.¹⁴⁷ The best response was seen primarily in patients with cutaneous and oral involvement. Rituximab has also been demonstrated to be effective for GVHD prophylaxis in a phase II study that showed a reduction in steroid-requiring cGVHD and improved OS in patients receiving peripheral blood grafts.¹⁴⁸ However, randomized trials are still needed to confirm these findings.

Ofatumumab

Ofatumumab is a fully humanized IgG1 kappa second-generation monoclonal antibody against CD20. Compared with rituximab, it has more potent CDC leading to increased B-cell destruction due to a higher binding affinity.¹⁴⁹ Ofatumumab is currently FDA approved for treatment of chronic lymphocytic leukemia. Pidala and colleagues recently demonstrated safety of ofatumumab in combination with steroids for the treatment of newly diagnosed moderate-to-severe cGVHD.¹⁵⁰ Twelve patients were enrolled and two infusion reactions (grade II and III, respectively) were observed. No grade IV infusion reactions, constitutional symptoms, or ⩾ grade III organ toxicities were observed; and hence no dose-limiting toxicities occurred. A phase II study is currently in progress to assess efficacy as primary therapy in cGVHD [ClinicalTrials.gov identifier: NCT01680965].

Obinutuzumab

Obinutuzumab is a second-generation anti-CD20 monoclonal antibody currently FDA approved for the treatment of follicular lymphoma. Based on the promising activity of B-cell depletion for prevention of cGVHD with rituximab, a phase II study using obinutuzumab for prevention of cGVHD following peripheral blood stem-cell transplantation is currently recruiting [ClinicalTrials.gov identifier: NCT02867384].

Adoptive cell therapy

Regulatory T cells

Tregs are a subset of peripheral CD4⁺ T cells (~5–10%) that are imperative for the development of immune tolerance in healthy individuals.^151–154 Reduction or loss of this population of T cells is associated with loss of immune tolerance and development of autoimmunity. Rezvani and colleagues reported that donor Treg content is predictive of aGVHD post-transplant, with a high Treg content associated with a lower risk of GVHD following HLA-matched allogeneic HSCT.¹⁵⁵ Several phase I/II studies have also shown that infusion of native donor Tregs and ex vivo-expanded umbilical-cord-blood-derived Tregs are associated with a lower incidence of aGVHD and improved immune reconstitution compared with historical controls.^156–158 In one study, no patients who received Treg infusion developed cGVHD at 1 year versus 14% in the control group.¹⁵⁸ Parmar and colleagues were able to show in animal models that infusion of fucosylated Tregs to improve engraftment and homing lead to longer in vivo persistence and decreased aGVHD at a lower cell dose compared with untreated Tregs.¹⁵⁹

Studies of Tregs in the setting of cGVHD have shown reduced frequency of circulating Tregs compared with patients without cGVHD and normal controls; however, the function of the cells remains normal.^154,160,161 The first in-human use of adoptive transfer of ex vivo-expanded Tregs was performed in two patients; one with SR-aGVHD and the other with SR-cGVHD.¹⁶² The patient with cGVHD had significant improvement in GVHD symptoms and was able to reduce the amount of immunosuppressive therapy required. Theil and colleagues evaluated use of adoptive Treg transfer in five patients with SR-cGVHD; two patients had clinical improvement in GVHD symptoms while the other three had stable disease as the best response.¹⁶³ No significant toxicity or GVHD exacerbations were observed.

One potential limiting factor of widespread use of adoptive Treg therapy is the need to manufacture sufficient numbers of Tregs for infusion without contaminating the product with other T-cell subsets, such as effector T cells, which could affect the response. This requires laboratory expertise to perform complex methods for both cell purification and expansion.¹⁶⁴ Larger prospective randomized studies are still needed to determine the benefit and long-term outcomes. See Table 4 for ongoing clinical trials.

Mesenchymal stromal cells

Response rates in treatment of SR-cGVHD have been less robust,^86,165,166 except notably in patients with sclerodermatous GVHD. The best reported response rate was seen in four patients with extensive skin scleroderma and ulceration who had significant improvement after four to eight treatments with MSCs.¹⁶⁷ Lack of clear benefit in cGVHD may be due to significant heterogeneity in multiple aspects between studies.⁸¹ Well-designed randomized studies with standardized methods of source procurement, expansion techniques, and dosing schedules are needed to further clarify the role of MSCs in the treatment of GVHD. There are currently three trials underway evaluating MSCs for treatment of cGVHD (Table 5).

Conclusion

As the number of allogeneic HSCT continues to increase each year, the importance and need for effective therapies to treat aGVHD and cGVHD and thereby improve outcomes for this potentially curative therapy cannot be over-emphasized. Responses to primary therapy with corticosteroids are often suboptimal and treatment can be complicated by significant side effects, including diabetes, hypertension, osteoporosis, myopathy, and avascular necrosis. Available second-line treatments have even lower response rates, leading to poor outcomes in acute GVHD, and prolonged morbidity with delayed immune recovery with chronic GVHD. Currently, there is a desperate need for new treatment options for this morbid complication of transplant. It would be ideal to identify targeted therapies that are not as broadly immunosuppressive, which can be used prior to the development of severe and irreversible tissue damage, and can provide durable responses without impacting graft versus tumor (GVT) effects. Continued understanding of the underlying immune mechanisms and pathways that are involved in the pathophysiology of GVHD should yield development of more effective therapeutics.

While these novel therapies have been tested in preclinical models and have shown promising results in early phase studies, larger multicenter studies are needed to determine true efficacy, preferably in comparison with current available therapies and long-term outcomes. Many of these agents are also being studied for prevention of GVHD, which would be ideal as long as GVT is not compromised. These therapies should help to improve symptoms and prevent progression of GVHD, to enhance patient quality of life and ultimately improve OS.