2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Saturday, December 12, 2026
04:00 p.m. - 05:30 p.m.
MCCNO - 388-390
Treatment Considerations, Biological Therapies, Transplantation, Adverse Events, Biological Processes, Pathogenesis, Immune Mechanism, Diseases, Immunology, GVHD, Lymphoid Malignancies, Myeloid Malignancies, Research, Translational Research, Clinical Research, Clinical Trials, Real-World Evidence, Study Population, Human
PAPER

440 HLA class I peptide-binding motif mismatch status is a predictor of cytokine release syndrome in haploidentical and mismatched donor transplantation with post-transplant cyclophosphamide

Nuttavut Sumransub, MD, Sandra Lee, PhD2, Shruthi Manjunath3, Remy Dulery, MD, PhD4, Corey Cutler, MD,FRCPC,MPH5, Vincent Ho, MD6, John Koreth, MD,MBBS,PhD,DPhil7, Amar Harry Kelkar, MD, MPH8, Mahasweta Gooptu, MD9, Rizwan Romee, MD10, Sarah Nikiforow, MD, PhD11, Francesco Malfona, MD, PhD12, Nicoletta Cieri, MD, PhD13, Catherine J. Wu, MD14, Jerome Ritz, MD15, Joseph Antin, MD16, Robert J Soiffer, MD17, Roman Shapiro, MD18
1 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States, University of Minnesota, Minneapolis, Minnesota, United States
2 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
3 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States, Frank H. Netter School of Medicine at Quinnipiac University, North Haven, Connecticut, United States
4 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
5 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
6 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
7 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
8 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
9 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
10 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
11 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
12 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
13 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
14 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
15 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
16 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
17 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States
18 Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, United States

Background: Cytokine release syndrome (CRS) is an early inflammatory complication following HLA-mismatched allogeneic hematopoietic cell transplantation (HCT) with post-transplant cyclophosphamide (PTCy). However, many patients with HLA mismatch never develop CRS, suggesting that conventional HLA matching alone does not fully capture CRS risk. We therefore investigated the association of HLA-related predicted alloreactivity and graft composition with CRS incidence and severity in HCT recipients receiving PTCy-based GVHD prophylaxis.

Methods: We analyzed 417 allogeneic HCT recipients, including 211 haploidentical (HAPLO), 119 mismatched unrelated donor (MMUD; 6/8 or 7/8 HLA-matched), and 87 matched unrelated donor (MUD; 8/8 HLA-matched), who underwent HCT with PTCy-based GVHD prophylaxis between 2015 and 2022 at Dana-Farber Cancer Institute. CRS was defined and graded according to the American Society for Transplantation and Cellular Therapy (ASTCT) consensus criteria. HLA-related alloreactivity mediated through killer-cell immunoglobulin-like receptor (KIR) interaction was predicted using the International Immunogenetics/Human Leukocyte Antigen (IMGT/HLA) database, whereas peptide-binding motif (PBM) matching, HLA-B leader status, and DPB1 T-cell epitope (TCE) permissiveness were predicted using National Marrow Donor Program (NMDP) bioinformatics tools. Univariate associations between CRS and categorical variables were assessed using Fisher’s exact test.  Multivariate logistic regression models were developed to adjust for other variables. Odd ratios (OR) were estimated and presented with a 95% CI. All analyses were reported using the 2-sided p-values.

Results: As expected, CRS incidence increased markedly with the degree of HLA disparity, occurring in 70.6% of HAPLO recipients compared with 37% of MMUD and 10.3% of MUD recipients (p<0.001). Peripheral blood stem cell (PBSC) grafts were also associated with significantly higher CRS incidence compared with bone marrow grafts (54.1% vs 26.2%, p<0.001).

Among 287 recipients with at least one class I HLA mismatch evaluated for PBM matching, 16 recipients had no mismatched PBM (PBM-matched group), whereas 271 had at least one mismatched PBM (PBM-mismatched group). PBM mismatch was strongly associated with both higher incidence (62.4% vs 25.0%, p=0.003) and greater severity of CRS. PBM-mismatched recipients experienced grade 1-4 CRS rates of 43.9%, 17.3%, 0.7%, and 0.4%, respectively, whereas PBM-matched recipients developed only grade 1 CRS (25%) with no grade ≥2 events. CRS incidence did not differ significantly according to individual PBM-A, PBM-B, or PBM-C mismatch categories. However, increasing graft-versus-host (GvH) directionality burden of PBM mismatch was associated with progressively higher CRS incidence (55.9% for 1 GvH mismatch count, 72.1% for 2 GvH mismatches, and 100% for 3 GvH mismatches; p<0.001).

Because donor type and graft source strongly influenced CRS risk, PBM mismatch was further evaluated among HAPLO and MMUD recipients receiving PBSC grafts (N=221). After adjusting for age, disease risk index (DRI), and conditioning intensity, PBM mismatch remained independently associated with CRS (OR 4.15, 95%CI 1.17-14.7, p = 0.027). Each additional PBM mismatch count further increased CRS risk (OR 2.4, 95%CI 1.7-3.6, p<0.001). These associations remained significant after further adjustment for graft composition, including CD3, CD4, CD8, NK, and CD19 cell counts, for both PBM mismatch status (OR 10.9, 95% CI 2.1-57.2, p = 0.002) and PBM mismatch burden (OR 2.6, 95% CI 1.7-3.9, p<0.001).

Among the HLA-related biomarkers evaluated, PBM mismatch emerged as the only independent predictor of CRS, whereas DPB1 TCE permissiveness, KIR mismatch, and HLA-B leader status showed no association with CRS.

Conclusions: This is the first study to identify PBM mismatch as the most consistent predictor of incidence and severity of CRS following allogeneic HCT with PTCy-based GVHD prophylaxis, independent of clinical factors and graft immune cell composition. These findings are biologically consistent with the established role of donor T-cell alloreactivity in the pathogenesis of post-HCT CRS and highlight the contribution of HLA-related alloreactivity beyond conventional matching. PBM matching may provide a clinically useful biomarker for CRS risk stratification and could inform prophylactic strategies, monitoring intensity, and patient selection for ambulatory HCT.

Nuttavut Sumransub, MD: No relevant conflicts of interest to declare.