Graft Versus Host Disease: Failures and Future Opportunities

In September 2009, Osiris Therapeutics, Inc. (OSIR) reported preliminary results from two Phase III trials evaluating its Prochymal product candidate for the treatment of acute Graft-versus-Host disease [GVHD].  Unfortunately, neither trial reached its primary endpoint, sending shares from $14 to a 52-week low of $5.35 by November 2009.  The results were the latest in a number of late-stage, clinical trial setbacks for the treatment of GVHD [see table 1], prompting us to review this very complicated disease in search of future treatment opportunities.

What is GVHD? 

When a cancer patient with myeloma, lymphoma, or other blood-borne diseases does not respond to traditional pharmacological therapies, hematopoietic stem cell transplantation [HCT] is often used as a last line of defense.  HCT is the transplantation of blood stem cells derived from the bone marrow or peripheral blood to the patient.  There are two types of HCT:

  • Autologous : stem cells come from patient’s own blood or bone marrow
  • Allogeneic: stem cells come from another person

While HCT remains a risky procedure with many possible complications, technological advances have resulted in diminished transplant-related deaths.  As a result, the number of allogeneic HCTs performed continues to rise, with greater than 25,000 procedures currently performed annually and the number is expected to double within five years [ref 1].

One of the major complications associated with allogeneic HCTs is GVHD.  GVHD is an immunological disorder that affects many organ systems including the gastrointestinal [GI] tract, skin, liver, and lungs.  If it happens within 3 months, it is called acute GVHD.  If it happens after 3 months, it is called chronic GVHD and may take as long as 3 years to go away.

GVHD arises when donor immune cells attack defined proteins on the host cells resulting in an array of clinical features ranging from mild [grade 1] to very severe [grade 4].  Severe GVHD [grade 3] has poor prognosis, with 25% long-term survival and only 5% for very severe GVHD [ref 1].

Within the GI tract, GVHD usually presents as bleeding, diarrhea, vomiting, anorexia, and abdominal pain.  The clinical symptoms of GI GVHD are those of typical inflammatory mechanisms and thus can be summarized into three sequential steps: 1) activation of antigen-presenting cells, 2) immune cell proliferation, and 3) target tissue destruction.

The primary pharmacological strategy to prevent GVHD is the use of cyclosporine and tacrolimus in combination with other immunosuppressants.  Despite these prophylactic therapies, GVHD still develops in 30%-80% of patients in the second month following HCT [ref 2].  Steroids, including prednisone, remain the gold standard therapy for GVHD treatment but only 25% to 41% of patients treated have complete GVHD remission [ref 3].  In addition, systemic treatment with prednisone or other steroids can lead to severe side effects such as opportunistic infections, electrolyte imbalances, and lymphoproliferative disease.

Currently, no therapies are approved by the United States Food and Drug Administration [FDA] for either prevention or treatment of GVHD [ref 4].  GVHD represents a growing problem due to an increasing number of HCT procedures and HCT survival with no change in the morbidity and mortality of this complication [ref 4].  As a result there is a great-unmet medical need to find therapies for this disease.

Past Failures

While several companies have brought new therapies into late-stage clinical development, no FDA approved treatments are currently available for the treatment or prevention of GVHD.  There are three primary reasons for this:

  1. One of the difficulties in finding new GVHD therapies is due to a lack of understanding of the pathophysiology of the syndrome.  Many different triggers can cause immunological diseases and the best therapeutic target has not been determined.  However, the very complexity of GVHD affords the opportunity to treat it by attacking its many different levels.
  2. The second reason for the high number of failures relates to the delicate balance between the harmful consequences of GVHD and the beneficial effects incurred when donor T cells attack malignant cells, a process referred to as the graft versus leukemia effect [GvL] and the underlying reason for performing HCT.  With the use of immunomodulatory agents to treat GVHD, consideration must be given to the need to control the immune response of GVHD without compromising the disease-fighting role of the donor immune cells.
  3. Thirdly, there are inherent difficulties in conducting human clinical trials.  The difficulty of demonstrating clinical benefits from objective parameters, such as survival and morbidity, and the subjectivity of grading acute GVHD, emphasize the need for blinded assessments in clinical trials of GVHD treatment [ref 5].

Regardless of the specific cause, there have been many high-profile, late-stage clinical trial failures, several of which are summarized below in reverse chronological order:

Osiris Therapeutics, Inc. (OSIR)

In September 2009, Osiris Therapeutics announced preliminary results from its two Phase III trials evaluating Prochymal for the treatment of GVHD.  The active ingredient of Prochymal is adult mesenchymal stem cells [MSCs].  The cells are from normal healthy adult volunteer donors and are not derived from a fetus, embryo or animal.  Studies have shown that these cells are universally compatible.  Similar to Blood Type O, these MSCs may be used without tissue type matching for specific patients.  Prochymal is produced in a controlled setting and is tested for possible infectious agents [such as viruses, bacteria, etc.] before it is given by infusion into a vein.

While Prochymal showed improvements in response rates for GI GVHD, neither trial achieved its primary endpoint.  There was no statistical difference between Prochymal and placebo for the steroid-refractory (35% vs. 30%, n=260) or first-line GVHD trials (45% vs. 46%, n=192).  However, in patients with steroid-refractory liver GVHD, treatment with Prochymal significantly improved response (76% vs. 47%, p=0.026, n=61) and durable complete response (29% vs. 5%, p=0.046). Osiris plans to meet with the FDA to evaluate the trial and discuss the next steps for moving forward with Prochymal.

SangStat Medical Corporation (SANG) and Abgenix (ABGX)

In February 2003, SangStat [subsequently acquired by Genzyme Corporation (GENZ)] and Abgenix [subsequently acquired by Amgen, Inc. (AMGN)] presented data from their Phase II/III study for evaluating ABX-CBL in patients with steroid-resistant GVHD.  The data presented showed that patients treated with ABX-CBL, an anti-CD147 monoclonal antibody, were similar to the control arm [antithymocyte globulin].  The primary endpoint of this study was to demonstrate superior survival with ABX-CBL, thus the primary endpoint was not met.  Further development of ABX-CBL for GVHD is not expected.

Roche (OTCQX: RHHBY) and Protein Design Labs, Inc. (PDLI) 

In 1995, Roche and Protein Design Labs presented the results of a study using Zenapax™ [daclizumab], a humanized monoclonal antibody against the interleukin-2 [IL-2] receptor, for the prevention of GVHD following bone marrow transplantation. The 102 patient study was halted after a planned interim analysis showed a significantly worse 100-day survival in the group receiving corticosteroids plus daclizumab (77% vs. 94%; p=0.02).  Overall survival at 1 year was also inferior in the combination arm (29% vs. 60%; p=0.002).  Both relapse- and GVHD-related mortality contributed to the increased mortality in the combination group.  Roche concluded that Zenapax is not effective in reducing the incidence of GVHD in the population of patients included in this study.


In December 1994, results of a Phase III trial of Xoma’s CD5 Plus™, a CD5-specific immunotoxin, for the treatment of GVHD were presented at the annual meeting of the American Society of Hematology.  A total of 243 patients were included in the trial, all of whom developed acute GVHD after allogeneic bone marrow transplantation. The trial compared CD5 Plus and a standard steroidal immunosuppressant therapy [methylprednisolone] versus placebo and steroids as initial therapy. The primary endpoint was defined as no evidence of acute GVHD at day 43 post starting treatment.

While the percentage of patients with no evidence of GVHD was higher in the CD5 Plus group than in the placebo group during the entire 43-day period of observation, at 6 weeks the difference was not statistically significant [44% of patients assigned to the CD5 group had complete response as compared with 38% in the placebo group, p=0 .36].  Xoma discontinued development of CD5 Plus.

New Opportunities

Despite the challenges in developing GVHD therapies, several companies are continuing with both preclinical and clinical programs.  The approaches range from novel, locally acting steroids [lower risk] to more complex immunomodulatory agents and cell cycle regulators [higher risk].  Several companies with promising programs in the GVHD space along with their technology and clinical development status are as follows:

Novel, Locally Acting Steroids

Soligenix, Inc. (SNGX) is developing orBec® [oral beclomethasone dipropionate] for the treatment of acute GI GVHD.  Beclomethasone [BDP] is a corticosteroid with potent topical activity used for inflammatory disorders affecting mucosal surfaces, such as the GI tract.  BDP’s mechanism of action is similar to other corticosteroids acting as an inhibitor of inflammatory cytokine production of immune cells.  One of the clear advantages of BDP versus other steroids is that oral BDP does not enter into the circulation thus avoiding many of the side effects associated with systemic steroid uses.  Pharmacokinetic studies have shown that BDP is metabolized in the intestinal mucosa and only the inactive metabolite is found in the circulation.  Additionally, the safety profile of BDP is well studied as it is approved by the FDA for three indications:

  • Becloforte®: inhalant marketed by GlaxoSmithKline plc (GSK) and used to treat asthma
  • Beconase®: nasal spray marketed by GlaxoSmithKline for rhinitis
  • Propaderm®: topical cream for psoriasis

Formulated for oral administration as a single product, orBec is a single product consisting of two separate pills.  One tablet is intended to release BDP in the proximal portions of the GI tract, and the other tablet is intended to release BDP in the distal portions of the GI tract.  This novel delivery system ensures that BDP is delivered to the entire GI tract.

Soligenix recently completed a 129-patient randomized, double blind, placebo-controlled, multicenter trial Phase III trial for orBec.  While the primary endpoint of this trial, time-to-treatment failure through day 50, did not reach statistical significance [p=0.1177], orBec did meet statistical significance for all of the secondary endpoints, such as the proportion of patients free of GVHD at Day 50 (p=0.05) and Day 80 (p=0.005) and the median time-to-treatment failure through Day 80 (p=0.0226).

Importantly, among all of the late-stage clinical trials for GVHD listed in Table 1, orBec is the only product to demonstrate a reduction in mortality.  In the Phase III trial, there was a 66% reduction in mortality among patients randomized to orBec at 200 days post-transplant with only 5 patient [8%] deaths in the orBec group compared to 16 patient [24%] deaths in the placebo group (p=0.0139).  At one year post-randomization in the Phase III trial, 18 patients [29%] in the orBec group and 28 patients [42%] in the placebo group died within one year of randomization [46% reduction in mortality, hazard ratio 0.54, 95% CI: 0.30, 0.99, p=0.04, stratified log-rank test].

Soligenix received a special protocol assessment [SPA] for a confirmatory, pivotal Phase III clinical trial cleared by the FDA.  The European Medicines Agency [EMEA] also agreed that should the new confirmatory Phase III study produce positive results, the data would be sufficient to support a marketing authorization approval in all 27 European Union [EU] member states.  Importantly, the primary endpoint for this study, treatment failure rate at day 80, was statistically significant in the prior Phase III trial [p=0.005].  The trial is enrolling patients and Soligenix has partnered with Sigma-Tau Pharmaceuticals, Inc. for commercialization of orBec.

Immunomodulatory Agents

In July 2009, results were published from a randomized, 4-arm, Phase II trial of 180 patients designed to identify the most promising agent(s) for initial therapy for acute GVHD [ref 3].  Patients were randomized to receive methylprednisolone 2 mg/kg per day plus etanercept [Enbrel® by Amgen, Inc. and Wyeth Pharmaceuticals], mycophenolate mofetil [MMF, CellCept® by Roche], denileukin diftitox [denileukin, Ontak® by Eisai Co., Ltd.], or pentostatin [Nipent® by Hospira, Inc.].  Day 28 complete response rates were etanercept 26%, MMF 60%, denileukin 53%, and pentostatin 38%.  Corresponding 9-month overall survival was 47%, 64%, 49%, and 47%, respectively.  Cumulative incidences of severe infections were as follows: etanercept 48%, MMF 44%, denileukin 62%, and pentostatin 57%.  Efficacy and toxicity data suggest the use of Roche’s MMF plus corticosteroids is the most promising regimen to compare against corticosteroids alone in a definitive Phase III trial.  The Phase II study is registered at [identifier NCT00224874].

Cell Cycle Regulators

GVHD is initiated when host antigen-presenting cells are detected by donor T-cells leading a cascade of cellular signaling events resulting in the expansion of donor immune cells and release of cytokines and chemokines, resulting in physiological damage to the GI tract and presentation of GVHD clinical symptoms.

Cyclacel Pharmaceuticals, Inc. (CYCC) is developing seliciclib [CYC202 or R- roscovitine] for the treatment of acute GVHD.  Seliciclib is an inhibitor of cyclin-dependent kinases [CDKs], such as CDK2, CDK7 and CDK9.  Although seliciclib is in preclinical development for GVHD, the product is also in Phase II trials for nasopharyngeal cancer and non-small cell lung cancer.

CDKs have been shown to be central kinases involved in the regulation and progression of the cell cycle.  Inhibition of CDK2 leads to cell cycle arrest and apoptosis and CDK7 and CDK9 to suppression of transcription in aberrantly proliferating cells.  Investigators from Harvard Medical School have recently published in vitro and in vivo evidence that CDK inhibition by seliciclib suppressed activation and expression of T cells and resulted in protection from acute GVHD [ref 6].  Seliciclib’s mechanism had three primary consequences in the context of GvHD:

  1. Inhibition of CDKs resulting in blocking the cell cycle
  2. Inhibition of RNA polymerase 2 resulting in apoptosis
  3. Prevention of TNF alpha dependent NFkB activation, a pathway shown to be activated in GVHD

Together, seliciclib may be able to specifically target the allo-reactive T cells preventing the progression of GVHD without targeting the immune cells needed for GvL or other pathogens.


A number of novel agents have been investigated to target various aspects in GVHD.  The majority of previous clinical trial setbacks have been immunomodulatory agents, which may favor lower-risk, steroid-sparing approaches in the short-term given the complexity of GVHD.  Ultimately, there appears to be potential synergies between different therapies, as they all possess different mechanisms and targets.  Future results from an ongoing pivotal trial and additional clinical results could provide optimism for both patients and investors in the GVHD space.

Table 1: Late-Stage, Completed GVHD Trials

Company Therapy Stage GVHD population Primary endpoint Result
Soligenix orBec® [oral beclomethasone dipropionate] in conjunction with steroidal therapy [prednisone] Phase III Post-bone marrow transplant patients presenting with Grade II intestinal GVHD Time to treatment failure, defined as the need for additional therapies due to uncontrolled signs or symptoms of GVHD, at study day 50 No statistical difference in primary endpoint, but secondary endpoint of time to treatment failure at Day 80, as well the treatment failure rate at Day 80, each achieved statistical significance [p-values 0.0226 and 0.0048, respectively].  Source: Blood. 2007 May 15;109(10):4557-63NOTE: In October 2009, the company initiated enrollment in a confirmatory Phase III trial using treatment failure rate at study day 80 as the primary endpoint.
Osiris Prochymal™ stem cell therapy in conjunction with standard of care Phase III Acute Steroid-Refractory Durable complete response defined as complete resolution of GvHD for a duration of at least 28 days No statistical difference. Source: Company press release dated 10/9/09, results not yet published
Osiris Prochymal™ stem cell therapy in conjunction with steroid therapy Phase III Newly diagnosed acute GvHD, grades B-D Proportion of patients surviving at least 90 days that achieve a complete response No statistical difference. Source: Company press release dated 10/9/09, results not yet published
SangStat Medical Corporation and Abgenix Murine anti-CD147 monoclonal antibody, ABX-CBL Phase II/III Acute Steroid-Refractory Demonstrate superior survival rate at 180 days compared to antithymocyte globulin [ATG], the study’s control arm Survival rate similar to ATG Source: Blood. 2007 Mar 15;109(6):2657-62
Xoma CD5 Plus™ with standard steroidal therapy [methylprednisolone] Phase III Acute GVHD following bone marrow transplantation No evidence of acute GVHD at day 43 after starting treatment Differences between the active and placebo groups were statistically significant at days 22, 29 and 36 but not at day 43. Source: Blood. 1996 Aug 1;88(3):824-30
Roche and Protein Design Labs, Inc. Zenapax™ [daclizumab] Phase II/III Newly diagnosed acute GVHD Proportion of patients in each treatment arm who experienced a decrease of acute GVHD overall severityby at least one grade on study day 42 without failing treatment Study halted after interim analysis showed a significantly worse 100-day survival in the group receiving corticosteroids plus daclizumab [77% vs. 94%; p=0.02]. Overall survival at 1 year was also inferior in the combination arm [29% vs. 60%; p=0.002].  Source: Blood. 2004 Sep 1;104(5):1559-64



  1. Lancet 2009; 373: 1550–61
  2. Expert Opin Investig Drugs. 2008 Sep;17(9):1389-401
  3. Blood. 2009 Jul 16;114(3):511-7
  4. Blood. 2005 Jun 1;105(11):4200-6
  5. Biol Blood Marrow Transplant. 2009 Jul;15(7):777-84.
  6. Cell Cycle 8:11, 1794-1802; 1 June 2009

7 Replies to “Graft Versus Host Disease: Failures and Future Opportunities”

  1. Pingback: pentostatin

Leave a Reply

Your email address will not be published. Required fields are marked *