High Proportions of Regulatory T Cells in Peripheral Blood Stem Cell Grafts Are Associated with Improved Total Lymphocyte, CD8⁺ T Cell and CD19⁺ Cell Recovery Following Allogeneic Haematopoietic Stem Cell Transplantation

Co-infusion of donor regulatory T-cells (Tregs) with conventional T-cells (Tcons) in allogeneic haematopoietic stem cell transplantation (HSCT) may improve immune reconstitution and suppress graft-versus-host disease (GvHD) without inhibiting the graft-versus-tumour (GvT) response. While phase I/II studies have focused on co-infusion of ex-vivo isolated Tregs, few have examined the impact of Tregs within peripheral blood stem cell (PBSC) grafts. In previous work, we observed that high proportions of Tregs (CD4⁺CD25⁺FOXP3⁺CD127^-/dim T-cells; expressed as percentage of CD4⁺ T-cells) in PBSC grafts were associated with improved myeloid and platelet recovery and decreased non-relapse mortality (NRM). In this current study, we therefore hypothesised that patients transplanted with high proportions of Tregs within allogeneic PBSC grafts will also have improved lymphocyte recovery.

Ninety-four allogeneic HSCT transplants were recruited from Oxford University Hospitals NHS Trust. The median age of the recipients was 48 years (range, 18-74). Forty-seven (50%) of the transplants were performed for acute leukaemia and 80 (85%) were performed using a reduced intensity conditioning regimen. Thirty-eight (40%) PBSC grafts were obtained from sibling donors (37 HLA-matched; 1 single HLA-mismatch (9/10)) and 56 (60%) from unrelated donors (39 HLA-matched (10/10); 17 single HLA-mismatch (9/10)). The cellular contents of the grafts were analysed using multi-colour flow cytometry. The median total nucleated cell (TNC), CD34⁺, CD3⁺, CD19⁺ and CD3^-CD56⁺ cell dose per kg recipient body weight was 10.1 x 10⁸/kg (range, 2.7-37.6), 6.3 x 10⁶/kg (range, 1.1-19.8), 280 x10⁶/kg (range, 87-801), 62 x 10⁶/kg (range, 13-245) and 35 x10⁶/kg (range, 8-101) respectively. Tregs accounted for a median of 2.96% of CD4⁺ T-cells (range, 0.81-8.56%) within the grafts with a median Treg dose of 4.7 x 10⁶/kg (range, 0.8-20.6). There were no significant correlations between the proportion of Tregs in the grafts and the TNC, CD34⁺, CD3⁺, CD19⁺ or CD3^-CD56⁺ cell dose.

The cumulative incidence of total lymphocyte (>1.0 x 10⁹/l), CD3⁺ (>0.7 x 10⁹/l), CD19⁺ (>0.1 x 10⁹/l) and CD3^-CD56⁺ (>0.1 x 10⁹/l) cell recovery by 6 months was 71.3% (95% CI, 60.8-79.4%), 51.2% (95% CI, 40.6-62.3%), 62.4% (95% CI, 50.8-72.0%) and 91.0% (95% CI, 82.0-95.7%) respectively. The median time to total lymphocyte, CD3⁺, CD19⁺ and CD3^-CD56⁺ recovery was 47 days (range, 11-455), 40 days (range, 33-265), 62 days (range, 27-262) and 34 days (range, 27-173) respectively. Patients transplanted with PBSC grafts containing higher proportions of Tregs (>2.96% CD4⁺ T-cells (median)) had improved lymphocyte recovery (P<0.001), with a cumulative incidence at 6 months of 83.0% (95% CI, 68.1-91.3%) compared to 59.6% (95% CI, 43.9-72.2%) for those receiving less. In multivariate analysis, high proportions of Tregs in the graft (%CD4⁺ T-cells) (HR 1.95 (95% CI, 1.24-3.06), P=0.004), use of Alemtuzumab during conditioning (HR 0.31 (95% CI, 0.17-0.56), P<0.001) and use of female donors (HR 1.65 (95% CI, 1.01-2.56), P=0.03) were independent predictors of total lymphocyte recovery. When analysing the major lymphocyte populations, high proportions of Tregs in the grafts were significantly associated with improved recovery of CD8⁺ T-cells (HR 1.98 (95% CI, 1.12-3.50), P=0.02) and CD19⁺ cells (HR 2.02 (95% CI, 1.23-3.06), P=0.004). In relation to clinical markers of immune recovery, there was no significant difference in the rates of CMV reactivation (P=0.89) between the two groups. However, less acute GvHD (grade II-IV) was observed in those patients receiving higher proportions of Tregs in the graft (23.4% (95% CI, 12.5-36.3%) vs 38.3% (95% CI, 24.5-51.9%) at 100 days), although this did not reach statistical significance (P=0.11). NRM at 3-years was lower (P=0.02) in patients receiving high proportions of Tregs in the grafts (12.8% (95% CI, 5.1-24.0%) [GvHD (n=1), infection (n=2), other (n=3)] vs 34.7% (95% CI, 21.2-48.5%) [GvHD (n=9), infection (n=4), other (n=3)]). The proportion of Tregs in the graft (HR 0.30 (0.11-0.85), P=0.02) was an independent predictor of NRM.

These data support the hypothesis that higher proportions of donor Tregs in PBSC grafts may also enhance lymphoid reconstitution following allogeneic HSCT. However, larger studies will be required to understand the potential mechanisms involved and to confirm our findings.