Chemokines and graft-versus-host disease - Immune Biology of Allogeneic Hematopoietic Stem Cell Transplantation - page 411

were evaluated and compared to 8 healthy controls. Fifty percent of the 14
patients post-HCT developed acute GVHD. A cut-off value of 48 pg/ml of
CCL8 in serum distinguished all of those with acute GVHD from the healthy
controls and transplanted patients who did not develop GVHD.

The most comprehensive biomarker analyses have been performed by the
group at the University of Michigan [84] . The initial analysis evaluated 21
patients with severe acute GVHD and compared them to matched control
patients that had undergone HSCT without the development of GVHD. Thirty-
five biomarkers were initially identified, 23 of which were further evaluated
based on the differences between groups, the availability of reagents, and
their immunological relevance. Eight proteins including CXCL8 were associ-
ated with GVHD. The researchers further refined their biomarker analysis
using separate training and validation sets of patients. Using logistic regres-
sion, they found that four biomarkers, CXCL8, IL-2Rα (CD25), TNFR1, and
human growth factor, predicted best for acute GVHD. Additionally, this panel
of biomarkers predicted for non-relapse mortality, and the extent of increase
in these biomarkers predicted for GVHD severity and overall survival.

As mentioned throughout this chapter, another approach commonly
employed to interrogate the function of a protein is to determine if a natural
SNP is present that affects the expression level of the protein. The presence
of an SNP allows investigators to determine the relationship between the
SNP and a clinical outcome. To this end, a number of studies have focused
on the possible importance of genetic polymorphisms in the chemokine
family to outcomes following HCT, including CCL2 (TRM), CCL5 (cGVHD),
CCR9 (aGVHD), DARK (duffy antigen receptor for chemokines), CCR5 (sur-
vival and aGVHD), and CCR9 (skin GVHD) [91,108,109] .

416

Based upon an emerging body of preclinical and clinical data regarding the
role of CCR5 in the development of GVHD, several groups have evaluated
the impact of CCR5 polymorphism on the incidence of GVHD and out-
comes following allogeneic HCT [110] . A natural deletion mutation of CCR5
(CCR5Δ32) is present in Northern European individuals with a frequency
of approximately 2% for individuals homozygous for the mutation and 13%
for those heterozygous for this mutation. The corresponding 32-bp deletion
results in a markedly truncated nonfunctional protein. Homozygosity for
this mutation dramatically associates with graft survival in renal allograft
recipients [55] and in protection from disease in multiply-exposed or HIV-
infected patients [111] . Bogunia-Kubik and colleagues [112] typed 186
recipients and 163 donors for CCR5 polymorphisms and found that recipi-
ents carrying the Δ32 mutation had a statistically significant decreased risk
of grade II-IV and III-IV acute GVHD, but an equivalent risk of cGVHD.
While the absence of CCR5 on donor cells did not influence the incidence
or severity of acute GVHD, the combined absence of CCR5 on donor and
recipient cells was associated with a profound decrease in the incidence of
grade I-IV GVHD (0/11 vs 70/151, p = 0.002). By contrast, the Seattle group
has shown that patients receiving HCT (with bone marrow as the stem cell
source) from CCR5Δ32 homozygous donors are at lower risk of developing
skin GVHD compared to controls [113] .

In a separate report, McDermott and colleagues [114] evaluated the impact
of another CCR5 haplotype termed CCR5 H1 homozygosity, which results in