Background: Placental growth factor (PlGF) is a member of the vascular endothelial growth factor family of proteins involved in angiogenesis, with direct effects on endothelial cells as well as a chemotactic effect on monocytes. Outside of normal pregnancy, pathologically high levels of PlGF have been described in the plasma of patients with inflammatory diseases (i.e. sickle cell, inflammatory bowel disease), and increased PlGF tissue expression in rheumatoid inflammation. We have recently identified that circulating levels of PlGF are significantly elevated in patients receiving allogeneic hematopoietic cell transplantation (HCT), with even higher levels seen in those with acute graft-versus-host disease [(aGVHD), Holtan SG et al, ASH 2014]. Given prior reports describing a correlation between higher PlGF tissue expression and inflammatory disease, as well as our findings of higher circulating PlGF levels in the transplant setting and with aGVHD, we sought to determine whether organ tissue expression of PlGF correlates with GVHD organ involvement and severity.
Patients and Methods: Immunohistochemical (IHC) staining for PlGF as well as common leukocyte antigen CD45 was performed on tissue biopsies of aGVHD target organs obtained during routine patient care, as part of an IRB-approved clinical research study. Patients with classic or late-onset acute GVHD were eligible for enrollment. Samples from normal human skin, colon, and duodenum collected during routine care and made available through an institutional tissue sample repository were also evaluated as controls. Slides were reviewed by two independent reviewers blinded to the subject’s condition. Expression of PlGF was graded on a semi-quantitative scale of 0-3 with 3+ being the strongest tissue expression. IHC results were then analyzed in the context of the patient’s clinical characteristics including underlying disease, type of transplant, status of the underlying disease (remission, relapse), organ involvement and severity of GVHD, and response to GVHD treatment.
Results: 13 patients had biopsies obtained as part of their work-up for aGVHD [9 skin biopsies and 15 gastrointestinal (GI) biopsies] available for review. Of the 9 skin biopsies, 7 biopsies were consistent with aGVHD by clinical pathology review. PlGF was present in moderate to high levels (2-3+) in all 7 cases of acute skin GVHD while the two cases found not to have aGVHD (vasculitis and engraftment rash), as well as normal skin, had little to no PlGF staining (0-1+). All acute skin GVHD patients responded to systemic corticosteroid therapy, with no correlation between the degree of PlGF staining and severity of aGVHD. In the 15 biopsies of the GI tract, 14 were consistent with aGVHD. One biopsy specimen not classified as aGVHD instead had documented Helicobacter pylori infection in the upper GI tissues, with infiltration of CD45+ cells but without increased expression of PlGF relative to control tissues. In contrast to the skin findings, normal colon and duodenal tissue in our controls showed higher baseline PlGF staining (3+ and 2+, respectively). The tissue expression of PlGF in patients with aGVHD was less intense than control in 11 of the 15 cases. Of the 4 patients with more intense PlGF expression than control, 3 had GVHD in the setting of relapsed leukemia. The lowest PlGF scores in colonic mucosa were seen in two patients who ultimately died of steroid refractory aGVHD.
Conclusion: These results provide the first evidence of tissue PlGF expression occurring in the setting of HCT and aGVHD. In our cohort, patients diagnosed with aGVHD of the skin had more intense expression of PlGF relative to control (2-3+). In contrast, patients diagnosed with aGVHD of the GI tract had lower expression of PlGF than control GI tissue; and patients with the lowest GI expression of PlGF had refractory aGVHD. These results suggest that tissue PlGF levels may correlate with organ-specific GVHD outcomes. Further studies are ongoing to determine the role of PlGF in neovascularization and tissue repair in aGVHD in an organ-specific manner.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.