Abstract 514

JMM and XQ contributed equally to this work.

Regulatory T cells (Treg) play an important role in the maintenance of tolerance after bone marrow transplantation (BMT) in experimental models. However, the relationship between Treg and acute graft-versus-host disease (GVHD) in allogeneic BMT recipients is not well established. We conducted a prospective analysis of Treg frequency in 215 BMT patients (125 allogeneic, 90 autologous) at the University of Michigan. Fresh peripheral blood samples were acquired prior to therapy within 24 hours of GVHD onset and at equivalent time points in patients without GVHD. Treg frequency was measured in triplicate by three color cytometry analysis to determine the frequency of CD4+CD25hiFOXP3+ cells within total lymphocytes. There were no significant differences between patients with and without GVHD for median age, nonmalignant disease, conditioning intensity, and median day of sample acquisition. Recipients of grafts from donors who were not family members or who were not HLA-matched were overrepresented in the GVHD group. Autologous BMT patients (N=90) and allogeneic BMT patients without GVHD (N=65) had the same mean Treg frequency (1.09% ± 0.10 vs. 1.09% ± 0.11, p = 0.54), showing that the use of calcineurin inhibitors did not affect Treg frequency. Patients with GVHD (N=60) had 40% fewer Treg (0.66% ± 0.07, p < 0.001) than those without GVHD. Frequencies of Tregs decreased in a linear fashion with each increasing grade of GVHD at onset, and were significantly reduced in patients with ≥ grade II compared to patients without GVHD (p < 0.001) (Figure 1). The calculated Receiver Operating Characteristic (ROC) curve for Treg frequency as an independent biomarker of GVHD was 0.69 (95%CI, 0.55-0.83). Treg frequency also correlated with the eventual maximum overall grade of GVHD (p < 0.001), suggesting a prognostic value for this measurement. Therefore, we evaluated whether Treg frequency would correlate with non relapse mortality (NRM) in the 60 patients with GVHD. Patients with low Treg frequency (<0.5%, N=30) had a significantly greater NRM (41% vs. 8%, p = 0.03) than patients with high Treg frequency (≥0.5%, N=30), which resulted in an inferior survival at two years (38% vs. 63%, p = 0.03) (Table 1). Acute GVHD accounted for the majority of NRM in the low Treg frequency group. Relapse mortality was similar between groups (p = 0.9) (Table 1). This difference in survival remained significant after adjusting for other important prognostic factors such as age, degree of HLA-match, and conditioning intensity (p = 0.05). In this set of sixty patients, frequency of CD4+CD25hiFOXP3+ Tregs at onset of GVHD correlates with GVHD severity, eventual maximum grade and NRM. Treg frequency thus has important diagnostic and prognostic value as biomarker for acute GVHD.

Figure 1.

Mean Treg frequencies by grade of GVHD at onset

Figure 1.

Mean Treg frequencies by grade of GVHD at onset

Close modal
Table 1

Clinical outcomes according to CD4+CD25hiFOXP3+ frequency at GVHD onset.

OutcomeTreg Frequency ≥ 0.5 %
Treg Frequency <0.5%
p value
n=30n=30
1 year Non-Relapse Mortality 8% (95% CI, 2-27) 41% (95% CI, 23-61) 0.03 
2 year Relapse Mortality 25% (95% CI, 15-40) 20% (95% CI, 11-35) 0.9 
2 year Overall Survival 63% (95% CI, 43-91) 38% (95% CI, 23-63) 0.03 
OutcomeTreg Frequency ≥ 0.5 %
Treg Frequency <0.5%
p value
n=30n=30
1 year Non-Relapse Mortality 8% (95% CI, 2-27) 41% (95% CI, 23-61) 0.03 
2 year Relapse Mortality 25% (95% CI, 15-40) 20% (95% CI, 11-35) 0.9 
2 year Overall Survival 63% (95% CI, 43-91) 38% (95% CI, 23-63) 0.03 
Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution