Extracellular Vesicles as Potential Biomarker for Acute Graft-Versus-Host-Disease

INTRODUCTION: Graft-versus-Host-Disease (GVHD) is the main cause of non-relapse mortality after Hematopoietic Stem Cells Transplantation (HSCT). Identyfing potential biomarkers of GVHD could be crucial to define patients at high risk of GVHD development and to better assess GVHD grading. One potential attractive biomarker may be represented by extracellular vesicles (EVs). EVs are membrane-enclosed structures secreted by many cell types and may be represented by exosomes, shedding vesicles (microvesicles; MVs) and apoptotic bodies. EVs are detectable in body fluids, in particular peripheral blood (PB) and urine; EVs play a key role in the regulation of physiological and pathological processes. The aim of this study was to investigate EVs in PB of allotransplanted patients (EVs count, size and phenotype) and analyze a potential correlation with acute and chronic GVHD (aGVHD, cGVHD).

METHODS: At our center, between 2000 and 2008, 41 multiple myeloma patients underwent an allograft. Median age of the patients was 53 years (range 34-65). Donors were HLA-identical siblings in 83% of the transplants. Conditioning regimen was mostly non-myeloablative (32/41,78%) and PB stem cells were used as source in all patients. GVHD Prophylaxis consisted of cyclosporine and mycophenolic acid in 34/41 patients (83%). Disease status was partial or less than partial remission in 95% of the patients. For standard policy, serum samples were collected before and monthly after transplant up to 6 months or disease relapse. EVs extraction was feasible due to high EVs stability in frozen serum samples. EVs were extracted using a protamine-based precipitation method and analyzed by flow-cytometry (Guava EasyCyte Flow Cytometer) without using latex-activated beads. We investigated a panel of fluorescence antibodies against specific membrane proteins (CD44, CD138, CD146, KRT18, CD120a, CD8, CD30, CD106, CD25, CD31, CD144, CD86, and CD140a). At each time-point, we determined for each sample: total EVs concentration, fluorescence distribution and percentage of positive EVs for a given marker. Data were computed by logistic regression analysis; Odds Ratio (OR) was calculated as proportional change as compared with pre-transplant baseline level of each marker.

RESULTS: In our cohort, the cumulative incidence of aGVHD and cGVHD was 56% (95% 40.7-71.8%) and 71% (95% CI 56.3-85.2%) at day 100 and 24 months, respectively.

Four biomarkers (CD146, CD25, CD106, CD31) showed a potential correlation with acute GVHD development. Two biomarkers (CD146 and CD25) were associated with an increased risk of developing aGVHD (CD146 fluorescence, OR 2.94, p=0.040 and CD25 fluorescence, OR 1.61, p<0.001). Furthermore, CD106 positive concentration (OR 0.23, p=0.077) and CD31 fluorescence and positive concentration (OR 0.24, p=0.052; OR 0.50, p=0.067, respectively) were associated with a decreased risk of aGVHD. All the biomarkers associated with aGVHD showed a proportional change from baseline in signal level before GVHD onset (an increase in case of CD146 and CD25, a reduction with CD106 and CD31, respectively).

No statistically significant association was observed between our panel of biomarkers and cGVHD.

CONCLUSIONS: In this study, we observed a potential association between 4 biomarkers expressed on EVs' surface and aGVHD. CD146, CD31 and CD106 belong to Cell Adhesion Molecule family (MCAM-1, PECAM-1 and VCAM-1, respectively), which are crucial for endothelium and immune cells interaction. CD25 is IL2 Receptor and is a marker of immune activation and inflammation. All these proteins mentioned above, could have a role in acute GVHD pathogenesis. A perspective study is currently ongoing at our Center to validate these data.