Are There Specific Markers in Allografted Patients Blood Predicting GvHD?

The graft versus Host Disease (GvHD) is a major complication of allogenic hematopoietic stem cell transplantation (allo-HSCT) in term of frequency and/or mortality. The complex physiopathology is roughly characterized by 3 phases where several inflammation factors and apoptotic pathways are involved. Among them, the chemokines CCL5 and CXCL12 are mainly playing a role in the homing of hematopoietic stem cells (HSC) to bone marrow and the migration of lymphocytes to the target organs. This migration is mediated by adhesion markers like ICAM-1 (Inter Cellular Adhesion Molecule 1) or by growth factor like VEGF (Vascular Endothelial Growth Factor). The final necrotic changes in target organs due to inflammatory mechanisms involves Interleukin 2, 6 and 8 (IL2, IL6, IL8) and the activation of Nuclear Factor kappa B (NFkB). Different Single Nucleotide Polymorphisms (SNP) in several chemokines and cytokines were shown to modify the level of gene expression. The aim of this study was to identify, in the recipient before allo- HSCT, some predictive factors of occurrence of GvHD, among the molecules involved in the different steps of GvHD. After informed consent, 73 patients (26 children and 47 adults) were included. SNPs -C28G and -G403A in the CCL5 promoter and G801A in the 3′UTR region of CXCL12 were studied and expression of CCL5 was analysed by RT-QPCR. To verify their impact on corresponding gene expression, protein measurements of these chemokines (CCL5 and CXCL12) but also of their inflammatory partners IL2, IL6, IL8, VEGF, ICAM-1 and endoglin were done in 39 samples. Among the 73 patients, 41 have developed a GvHD (Grade II-IV) and 32 did not developed GvHD or developed only grade I GvHD. As shown in table 1, the global frequency of SNPs underlined an unbalanced distribution in this cohort.

Table1: Repartition of the different SNP of CCL5 and CXCL12

However, we did not observed a link between the SNP of CCL5, its expression (p=0.812) or its protein concentration (p=0.993) and the occurrence of a GvHD. Serum concentration of VEGF, CXCL12 and Endoglin were different between adults and children. In fact, in adult, the VEGF and Endoglin serum concentrations were significantly higher than in pediatric samples, with, respectively, 30.66 pg/ml [6.92;100.53] vs 9.97 pg/ml [0;31.29] and 18831.49 pg/ml [17787.7;25190.86] vs 11104.52 pg/ml [7742.54;16231.12], in opposition to CXCL12, which is higher in pediatric patients (54,76 pg/ml [43,74 ; 70,82] vs 35,1 pg/ml [22,63 ; 52,12]), but the levels of these proteins were not associated with the occurrence of GvHD. However, when considering only the adult group, significant differences were observed ( table 2), between the group of patients who developed a GvHD (II-IV) and the group of patients with no or low grade GvHD. Adult with GvHD

Table 2: Statistical difference (p<0.05) of serum concentrations (pg/mL) between adults who developped a GvHD (Grade II-IV) and adults who did not.

In conclusion, two markers ICAM-1 and VEGF show a significant prognostic value for GvHD severity in adult, when quantify in recipient patient before allografting. Further studies on a wider cohort would help to confirm these results and to precise the role of CCL5 and CXCL12 SNPs.