Recombinant Thrombomodulin for the Treatment of Transplantation-Associated Coagulopathy after Allogeneic Hematopoietic Stem Cell Transplantation

Background: Thrombotic microangiopathy (TMA) and sinusoidal obstruction syndrome (SOS) as well as acute GVHD are major complications after allogeneic hematopoietic stem cell transplantation (alloHSCT) to overcome to achieve better overall survival. These complications are closely relevant to inflammation and coagulation involving endothelium, which are named as transplantation-associated coagulopathy (TAC). Recombinant thrombomodulin (rTM) is a new drug for treating DIC approved by Japanese Ministry of Health, Labour and Welfare in 2008. Membranous TM has many aspects to inhibit inflammation via activated Protein C (APC), lectin-like domain, and activated thrombin-activatable fibrinolysis inhibitor as well as to inhibit coagulation by binding factor IXa and Xa through APC. Biomarkers such as inflammatory cytokines and endothelial molecules are expected to be clue in elucidating TAC. We established SIGHT study group to explore above complications following alloHSCT. In the present study, we investigated the effects of rTM on levels of the biomarkers and whether we could prevent TAC using rTM among registered patients in the SIGHT study group.

Patients and methods: SIGHT study group covered nationwide institutes more than 70 in Japan. Patients whoever were eligible for alloHSCT to treat hematopoietic disorders by physicians decision were permitted and written informed consent was requested to register in the study. Blood samples were collected from the patients before and after transplantation through day 28. Levels of cytokines (IL-6, TNF-α, HMGB1, MCP-1, RANTES) and soluble molecules (VCAM-1, E-selectin, PAI-1, PDMP) were measured by ELISA. rTM was administered as a prophylactic therapy for TAC in day 4-14 after alloHSCT. Control group was received heparin or no anti-coagulation therapy as prophylaxis during same schedule as rTM group. Patients were not randomized to these three arms but allowed to select one by their physicians decision. The primary endpoint was a comparison of fluctuation of these biomarkers in three arms. Secondary endopoint was a comparison of the rate of developing TAC.

Results: The subjects were 293 patients who underwent alloHSCT in SIGHT study group between June 2010 and February 2014. 271 patients out of them were analyzed to measure level of these biomarkers. There was no significant difference in all characteristics but GVHD prophylaxis between two groups. rTM group received more cyclosporine than control group (p=0.05). MCP-1 and IL-6 exhibited more significant elevations on day 7 after alloHSCT, while HMGB-1 did on the day of alloHSCT. In contrast, the levels of TNF-α, E-selectin, VCAM-1, PAI-1 and PDMP exhibited increment since around day 7 after alloHSCT. Significant improvements in TNF-α, E-selectin, VCAM-1, HMGB-1, PAI-1 and PDMP was shown after rTM-treatment, but not shown in control group. Regarding complications following alloHSCT, the rate of developing acute GVHD and SOS was significantly lower in rTM group than without it (36.5% vs 62.2%, p=0.000; 12.5% vs 24.5%, p=0.032, respectively), while TMA did not differ (8.4% vs 10.2%, p=0.961). Stepwise multivariate logistic regression analyses revealed that anti-coagulation therapy without rTM and the level of PAI-1 on day 28 after HSCT were independent risk factor for acute GVHD (p=0.000, odds ratio=3.006; p=0.000, odds ratio=1.068, respectively). Also anti-coagulation therapy without rTM and the level of HMGB-1 on day 28 after HSCT were significantly predictive for risk of SOS (p=0.015, odds ratio=2.650; p=0.001, odds ratio=1.433, respectively). Tacrolimus was significantly good risk of SOS (p=0.022, odds ratio=0.404). The level of VCAM-1 on day 28 after HSCT was significantly associated with risk of TMA (p=0.040, odds ratio=1.001).

Conclusions: We identified predictive biomarkers for TAC following alloHSCT, which were PAI-1 for acute GVHD, HMGB-1 for SOS, and VCAM-1 for TMA, respectively in this study. The present findings suggest that rTM has the possibility to play a therapeutic role for acute GVHD and SOS induced by suppression on these biomarkers. Endothelial and pro-coagulant biomarkers were released delayed compared with inflammatory biomarkers. TMA is reported to develop around day 44 in median after alloHSCT. Further studies are needed whether rTM should administer later or longer to improve TMA than the present duration after alloHSCT.