Lowering GVHD in Hematologic Malignancies

The combination of Zolinza (vorinostat) with tacrolimus and methotrexate can be effective in mitigating graft-versus-host disease (GVHD) in patients undergoing a matched unrelated donor myeloablative conditioning hematopoietic stem cell transplant (HSCT).

In a phase 2 single-arm study, the cumulative incidence of grade 2-4 acute GVHD at day 100 with the Zolinza combination was 22 percent, which met the primary endpoint of the study (target incidence of less than 28 percent). The historical incidence of acute GVHD despite standard immunosuppressive prophylaxis in patients receiving HSCT is 50 percent, said Israel Henig, M.D., who presented the data at the 2017 BMT Tandem Meetings. In the study, the relapse rate at one year post-transplant was also low, at 19 percent.

“The reduction in acute GVHD is much more than we expected,” said Henig, a research fellow at the University of Michigan Comprehensive Cancer Center, Ann Arbor. “We’re excited. We’re looking at data for relapse because vorinostat is used in many clinical trials now as an anti-cancer drug. The relapse rate in our short-term follow-up was encouraging. The acute leukemia population is not an easy population. Most of our patients had acute myelogenous leukemia (AML) or myelodysplastic syndrome (MDS), and these are patients for which you would expect higher rates of relapse. We need longer follow-up to know.”

Acute GVHD remains a significant barrier to a more widespread application of allogeneic HSCT for patients with hematologic malignancies. Zolinza is a histone deacetylase inhibitor (HDAC). Inhibiting HDAC has been shown to regulate GVHD in experimental models of HSCT. In these models, Zolinza suppressed pro-inflammatory cytokines, regulated antigen-presenting cells through induction of indoleamine 2,3-dioxygenase, and enhanced T-regulatory functions while preserving graft-versus-leukemia responses.

An earlier, first-in-human study by Henig’s group demonstrated that Zolinza was safe to administer and resulted in a low incidence of acute GVHD following related donor-reduced intensity conditioning HSCT.

In the phase 2 trial, 37 patients 18 years or older with a hematologic malignancy for which myeloablative HSCT was appropriate were paired with 8/8 HLA-A, -B, -C unrelated donors. GVHD prophylaxis consisted of intravenous (IV) tacrolimus at 0.03 mg/kg/day starting three days before transplant through 180 days after along with IV methotrexate at 5 mg/m² once daily on days one, three, six and 11 posttransplant. Zolinza at 100 mg orally twice daily was initiated 10 days before transplant and continued through day 100 posttransplant.

Mean age of the patients was 56 years; 27 had AML, 8 had MDS, one had chronic myelogenous leukemia, and one had acute undifferentiated leukemia. Twenty one of the 37 patients (56.8 percent) had low disease status at baseline.

All patients engrafted. Zolinza was safe and tolerable with no excessive toxicity or death attributable to Zolinza. At median follow-up of one year, 10 patients died: four from relapse and three from acute GVHD-related infection.

The cumulative incidence of grade 2-4 acute GVHD was 22 percent at day 100 and 22 percent at day 180. The cumulative incidence of grade 3-4 acute GVHD was 8 percent and 11 percent at day 100 and day 180, respectively.

The cumulative incidence of chronic GVHD at one year was 29 percent. The cumulative incidence of relapse and nonrelapse mortality at one year posttransplant was 19 percent and 18 percent, respectively. Estimated overall survival at one year was 76 percent.

Blood samples from study subjects were obtained at day 30 to perform correlative pharmacodynamics studies and biomarker analyses. Acetylated histone (H3) levels were significantly higher at day 30 in peripheral blood mononuclear cells from Zolinza-treated patients compared with similarly treated patients who did not receive Zolinza. Day 30 plasma levels of inflammatory cytokines interleukin (IL)-6 and IL-10 were significantly lower in Zolinza patients compared with historical samples, as were two biomarkers used to predict acute GVHD: Reg3-alpha and ST2. “This is biological confirmation that the drug is working,” said Israel.

HDAC inhibition as a strategy for GVHD prevention should be further investigated in a multicenter phase 3 randomized study.