Background: Allogeneic hematopoietic stem-cell transplantation (HSCT) is curative for adult sickle-cell disease (SCD). Yet donor source (matched related [MRD] or unrelated [MUD], mismatched related [haplo] or unrelated [MMUD]), conditioning regimen (combination of in vivo T-cell depletion with single agent busulfan, fludarabine-melphalan, or TBI), and GVHD prophylaxis (post-transplant cyclophosphamide [PTCy], ex vivo T-cell depletion [e.g, CD34 selection], or sirolimus) each shape bilateral tolerance and immune reconstitution differently. Systematic, high-dimensional comparisons of immune recovery after HSCT in SCD remain scarce.

Aims: Computationally map lymphoid and myeloid recovery in SCD over time across conditioning platforms, benchmarking patterns against healthy controls (HC).

Methods: High-dimensional flow cytometry was performed on 70 peripheral-blood samples from 13 of 22 transplanted adults with SCD, alongside three HC specimens, using a 43-marker panel. After FlowCT quality control, data were biexponentially transformed, meta-clustered with FlowSOM, visualized with UMAP, and manually curated into lineages; subset frequencies and absolute counts were compared with Wilcoxon tests. Conditioning/GVHD-prophylaxis groups were 4 Gy + PTCy (MUD/MMUD, n = 3), FluMel + CD34-selected graft + sirolimus (MMUD/MUD/Haplo, n = 4), and 3 Gy TBI + sirolimus (MRD, n = 6). Timepoints were classified as baseline, early (< 30 d), intermediate (60–90 d), and late (> 180 d). Clinical variables, including engraftment kinetics and lineage chimerism, were analyzed in the full 22-patient cohort, while immunophenotypic trajectories were derived from the 13-patient flow subset.

Results: At baseline, SCD exhibited a B-cell skew, with naïve B-cell frequency at 11.6% of CD45⁺ (IQR 7.1–15.1%) vs. 4.0% in HC (P = 0.02), and trends toward higher plasma cells and monocytes.

Engraftment kinetics differed by platform: CD34-selected grafts yielded the fastest ANC recovery (median 19 d, range 13–21) vs. TBI (median 24 d, 12–30) and PTCy (26 d, 22–31) (P = 0.005), whereas platelet recovery was quickest with TBI (7 d, 7–26) compared with CD34 (20.5 d, 16–38) and PTCy (29 d, 22–35) (P = 0.03).

In line with the rapid engraftment and neutrophil recovery, CD34 selection drove a myeloid-dominant landscape—monocytes at 62.7% (53.6–67.4%) vs. TBI at 50.0% and PTCy at 45.7% (P = 0.049) during the early time period (<30 d), accompanied by an innate lymphoid cell surge (7.7% vs ≤ 2.6%; P = 0.03). PTCy uniquely expanded plasma cells (26.8 × 10³ cells L⁻¹; 17.4–36.1) during this period.

At 60–90 d, CD34 recipients showed rapid B-cell reconstitution: naïve B-cells 30.8% (18.7–39.7; 419 × 10³ cells L⁻¹) vs. 4.7% (48 cells L⁻¹) with TBI and 0.9% (6 cells L⁻¹) with PTCy (P ≤ 0.017). In contrast, PTCy favored plasma-cell (1.6%) and NK-cell (16.7%) dominance (P < 0.025). Day +100 chimerism was uniformly high—myeloid 99%, total PB 96%, B-cells 95%—with variable T-cell chimerism (CD34 50%, TBI 33%, PTCy 19%; P= 0.24).

Beyond 180 d, CD34 grafts maintained higher T-cell numbers (CD4 116 vs 75 cells L⁻¹, P = 0.049; CD8 134 vs 21, P = 0.026) and preserved unconventional subsets (double-negative T 8.2 vs 0.6 cells L⁻¹, P = 0.01; innate-like T 40.6 vs 8.6, P = 0.049), whereas 3 Gy TBI favored memory B-cells (1.5% vs 0.3%, P = 0.049); late PTCy samples were unavailable.

At one year, T-cell chimerism differed significantly (CD34 77%, TBI 58%, PTCy 35%; P = 0.047). Myeloid (99%) and B-cell (94%) chimerism remained high with no inter-group differences, while total PB chimerism trended lower with PTCy (55%) than with TBI (91%) or CD34 (98%) (P = 0.26).

Conclusions: CD34-selected grafts accelerate early myeloid and naïve-B-cell rebound and foster robust CD4/CD8 T-cell reconstitution later. PTCy favors early NK- and plasma-cell expansion, whereas low-dose TBI promotes dendritic and memory-B recovery. These divergent trajectories mirror lineage-specific mixed chimerism, as full donor engraftment is uncommon across all compartments. Sub-lineage chimerism assays are therefore critical to dissect platform-specific immune recovery, bilateral tolerance, and graft-rejection risk.

This content is only available as a PDF.
2025
Sign in via your Institution