Early Breakthrough T Cell Proliferation Despite CNI/MTX Prophylaxis Is a Harbinger of Acute Gvhd (AGVHD) and Is Controlled By Abatacept: Mechanism of Success of the ABA2 Agvhd Prevention Trial

Background: Abatacept was recently FDA-approved for AGVHD prevention, based in part on the results of the ABA2 Phase 2 GVHD prevention trial. ABA2 demonstrated a significant decrease in Grade 2-4 AGVHD after both 8/8 and 7/8 URD HCT in patients prophylaxed with CNI/MTX + abatacept versus those receiving CNI/MTX alone. The randomized double-blind, placebo-controlled design of the 8/8 arm has enabled a rigorous determination of the immunologic drivers of the success of abatacept in preventing AGVHD.

Methods: We studied hematopoietic and immune reconstitution in patients receiving 8/8 HLA-matched URD HCT enrolled on ABA2, who received either CNI/MTX + placebo (PBO) or CNI/MTX + abatacept (ABA) for GVHD prophylaxis. Flow cytometry using 16 ten-color panels evaluated T cell reconstitution, T cell naïve/effector cell balance, T cell proliferation and cytotoxicity. Comparisons were made between healthy controls (HC, n = 23), PBO (n = 69) and ABA (n = 73) patients, and between patients developing Gr 2-4 AGVHD (n=43 PBO and n = 31 ABA) versus Gr 0-1 AGVHD (n=26 PBO and n = 42 ABA). RNASeq was performed on CD4+ and CD8+ T cells purified by flow cytometry (within 3 hours of blood collection) at day 21 and 28 post-HCT at 2 central lab sites, from patients who developed Gr 2-4 AGVHD (PBO, n=87 total samples ; ABA, n=67 total samples) or Gr 0-1 AGVHD (PBO, n=45 total samples; ABA, n=89 total samples). RNASeq was performed using the Clontech SMART-Seq v4 kit, Illumina NexteraXT DNA library prep, and the Illumina HiSeq3000. Statistical modeling was performed using DESeq2, followed by differential expression and Gene Set Enrichment Analysis (GSEA).

Results: Combined flow cytometric and transcriptomic analysis provided compelling evidence that T cell allo-proliferation was a harbinger of Gr 2-4 AGVHD onset in PBO patients, and that abatacept was able to control this proliferative escape. Thus, using flow cytometry, we detected early (Day +28) breakthrough CD4+ T cell proliferation (as measured by Ki67) in PBO patients who developed Grade 2-4 AGVHD compared with those who developed Gr 0-1 AGVHD (mean 20.4% vs 12.6% Ki67+ CD4 T cells) (p<0.05).In contrast, there was no increase in Ki67+ CD4 T cells in ABA patients with AGVHD (5.8% vs 6.7% Ki67+ CD4 T cells (p = ns). Concordant with flow cytometry results, RNA-Seq detected prominent Cell Cycle-associated transcriptomic signatures in PBO AGVHD samples via Gene Set Enrichment Analysis (GSEA) in both CD4+ and CD8+ T cells, with abatacept successfully controlling this proliferative escape (Fig.1). Importantly, despite the difference in Gr 2-4 AGVHD-associated breakthrough T cell proliferation between PBO and ABA patients, there was no difference in the longitudinal reconstitution of total T cells, total CD4+ T cells and CD8+ T cells in the two cohorts (Watkins et al, JCO 2021). Moreover, both PBO and ABA patients were able to mount homeostatic proliferative responses, with upregulated Ki67 compared to HC by flow cytometry (Day 28 Ki67+ CD4 and CD8 T cells higher in PBO vs HC and ABA vs HC, with p values <0.01 for all comparisons to HC). Notably, both PBO and ABA CD4 and CD8 T cells also demonstrated enrichment in RNASeq proliferation signatures compared to HC, as measured by GSEA (not shown). PBO and ABA patients also demonstrated equivalent T cell repertoire reconstitution as measured by longitudinal TCRSeq clonality measurements (Fig. 2), further substantiating intact T cell reconstitution in these patients. Unlike the proliferation-dominated breakthrough AGVHD profile in PBO patients, RNA-Seq revealed a unique AGVHD signature in ABA patients. Thus, patients who developed Gr 2-4 AGVHD despite abatacept exhibited a GSEA gene expression profile in both CD4+ and CD8+ T cells enriched for naïve T cell gene sets, prominently including IL6 signaling pathway transcripts.

Conclusions: These data provide compelling evidence for early (Day 21-28) T cell allo-proliferative escape being a key mechanism driving breakthrough AGVHD after standard CNI/MTX prophylaxis, and control of this escape with abatacept. This analysis further documented retained overall T cell reconstitution with abatacept despite control of AGVHD-associated proliferation, consistent with the reassuring safety profile of abatacept observed in ABA2. Together, these results provide new insights into the central role that allo-proliferation has in breakthrough AGVHD, and its control with CD28:CD80/86 costimulation blockade.

Qayed:Novartis: Honoraria; Vertex: Honoraria. Kean:Bristol Myers Squibb: Patents & Royalties: clinical trial, Research Funding; HiFiBio: Consultancy; Bluebird Bio: Research Funding; Novartis: Research Funding; Mammoth Biosciences: Current equity holder in private company, Current holder of stock options in a privately-held company; Magenta: Research Funding; EMD-Serono: Research Funding; Vertex: Consultancy.