There is compelling evidence that increased levels of Tissue-factor (TF) and TF-bearing microparticles (MPs) as well as P-selectin and its receptor P-selectin glycoprotein ligand 1 (PSGL-1) play an important role in disease progression, angiogenesis and cancer-related coagulopathy in solid and hematologic malignancies. We therefore hypothesized that numbers of circulating TF- and PSGL-1 bearing MPs might have substantial influence on outcome in allogeneic stem cell transplantation (alloSCT).

Blood samples were obtained from 33 patients with different hematologic diseases (AML=19, ALL=5, CML/MPS=2, CLL/lymphoma=2, myeloma=2, aplastic anemia=2) during the course of alloSCT at distinct time points: at admission (“start”), during conditioning therapy before anti-thymocyte globulin (ATG) infusion (“pre-ATG”), on day 3 of ATG infusion (“during-ATG”), before transplantation (“pre-Tx”), 1 h after transplantation (“post-Tx), on day 1 after transplantation (“d+1post-Tx”) and on day 1 after neutrophil engraftment (“engraftment”). Conditioning regimens included either standard therapies (n=17), FLAMSA (n=11) or other (n=5). GvHD prophylaxis consisted of either CsA/MTX/ATG (n=19), CsA/MMF/ATG (n=11) or CsA/Steroids (n=3). Typical patient and transplant risk factors were included in statistical analysis.

Platelet-free plasma (PFP) was obtained by three-step centrifugation, snap-frozen in liquid nitrogen and stored at -80°C. After thawing, 50μl of PFP were resuspended in Annexin V binding or control buffer and labeled with 1μl Annxin V-Cy5 and 0.5μg of CD142-FITC (TF), CD162-FITC (PSGL-1) or isotype control. In flow cytometry MPs were gated by size (<1μm) and Annexin V positivity and Trucount tubes® were used for MP enumeration.

Mean follow-up time was 759 (10-1305) days. 17 (51.5%) patients died, 9 due to TRM (27.3%), and 12 (36.4%) had recurrence or progression during follow-up. Median overall survival (OS) was 815 days, median disease-free or progression survival (DFS/PFS) 440 (10-1272) days and median time to progression (TTP) 365.5 (130-1183) days.

MP counts are shown in table 1. There was a significant reduction (p<0.05) for Annexin V- and PSGL-1-MPs from “start” to “pre-Tx” and an increase after transplantation while TF-MPs had significant higher numbers at “d+1post-Tx” compared to “start” values.

Univariate Cox analysis identified TF-MPs at “start” (p=0.011, HR=1.006 per 1 MP/μl) and “pre-ATG” (p=0.004, HR=1.011) to be significantly associated with OS. When corrected for known risk factors, TF-MPs “pre-ATG” remained the only independent predictive parameter for OS in multivariate Cox analysis s(p=0.036, HR 1.022). A cut-off value was determined for stratification of patients by calculating the AUC of a ROC–curve (p=0.030, AUC 0.762). Thus, Patients with TF-MP values >140/μl showed a significantly shorter survival time (p=0.007; mean OS: 469.4±160.0 vs. 1038.5±110.3 days). Same was done for DFS/PFS and again TF-MP values >140/μl were predictive of worse DFS/PFS (p=0.020, HR=3.61) in multivariate Cox regression with a shortened DFS/PFS (p=0.013; mean: 350.9±113.8 vs. 852.5±129.6 days). These findings were caused by an increased treatment-related mortality (TRM) with a cumulative incidence at 1 year of 55.5% vs. 13.3% for corresponding TF-MP groups (p=0.029). TRM was caused by infection (n=4), VOD (n=2) or other reasons (n=3). There was no significant association with relapse or acute GvHD.

Furthermore, we found PSGL-1-MP counts below 250/μl “post-Tx” (and on “engraftment”) to be a significant predictor of relapse/progression (p=0.044, HR=4.91) with a cumulative incidence at 3 years of 81% (95%-CI: 61.9-100) vs. 11.1% (95%-CI: 1.8-70.5) and a mean TTP of 602±106.1 vs. 1140±104.5 days (p=0.027).

TF-bearing MPs before and during conditioning therapy are prognostic markers of OS and DFS/PFS due to increased TRM while PSGL-1 bearing MPs after transplantation and at engraftment predict relapse/progression and TTP. Course of MP counts suggests increased TF-MP release by monocytes, endothelial and stromal cell while significant differences in PSGL-1-MPs might rather be graft-related.

Table 1

MP counts during transplantation course

View Large

Trummer:CSL Behring: Research Funding.