IMiDs like lenalidomide have immunostimulatory effects and therefore the potential to reduce relapse after allogeneic haematopoietic cell transplant (AHCT) by increasing graft-versus-tumour (GvT) effects. However, early clinical experience using IMiDs after AHCT has been limited by induction of graft-versus-host disease (GvHD). Although lenalidomide has been shown to augment mitogen-stimulated T cell responses, the effects of this drug on T cell alloresponses that mediate both GvT and GvHD have not been well defined. Better understanding of the immune mechanisms involved would facilitate tracking and manipulation of lenalidomide-potentiated alloresponses and could reveal ways to use the drug to maximise GvT without excess GvHD. Therefore we used an HLA-mismatched in vitro model to analyse in depth the effects of lenalidomide on functional human T cell alloresponses.
Materials and Methods
We cocultured CFSE-labelled PBMC from healthy donors with irradiated allogeneic PBMC in the presence of 1μM lenalidomide, vehicle control or following pre-incubation with 1μM lenalidomide for 24 hours. Functional alloresponses were quantified after 7-9 days of allo-coculture by flow cytometry. In addition, allo-coculture responders were flow-sorted into alloproliferative or non-proliferative fractions and extracted RNA used for gene expression profiling.
Addition of lenalidomide to allo-cocultures increased the total number of responder cells (p<0.001) due primarily to increased proliferation (74% median increase) of allospecific responder CD8 (alloCD8) T cells (p<0.001). Proliferation kinetic analysis showed that lenalidomide did not increase the number of cell divisions of alloCD8 cells, but increased the CD8 allospecific precursor frequency within the responder cell pool (from a median of 2.6% to 10%, p<0.001) consistent with lowering the activation threshold of alloCD8 cells. A significant enrichment for effector memory phenotype was observed in these cells (median 48% increased to 59%, p<0.001). Addition of lenalidomide to allo-cocultures also increased the proportion of alloCD8 cells secreting TNF-α, IFN-γ and expressing CD107a, as well as polyfunctional effector cells (Fig. 1A). Although lenalidomide did not increase proliferation of CD4 cells, TNF-α production by proliferative CD4 T cells was increased suggesting they may contribute indirectly to CD8 alloresponses. Pre-treatment of stimulators, responders or both prior to allo-coculture did not result in increased alloCD8 proliferation, indicating that the drug must be present in the co-culture to exert an effect. Finally to assess whether lenalidomide exerted effects via potentiation of intrinsic alloproliferative pathways or by qualitatively different pathways we performed gene expression profiling of CD8 T cells sorted from allo-cocultures. As expected, alloCD8 cells from untreated allo-cocultures demonstrated >2-fold altered expression of >500 genes mostly associated with DNA synthesis and cellular proliferation when compared to non-proliferative CD8 cells. Lenalidomide-treated alloCD8 cells showed further increases in expression of many of these genes; however treatment also resulted in significant changes in expression of additional genes in alloCD8 cells compared to untreated alloCD8 cells (Fig 1B). These included >8 fold increases in expression of genes reported to potentiate T cell immune responses in other settings including PFKFB4,Pirin, and SOCS2 (part of the E3 ubiquitin ligase complex with cereblon), and >5 fold decreases in genes which can suppress T cell activation and memory differentiation including FAIM3 and PMCH.
We have shown for the first time that lenalidomide potentiates human alloresponses primarily by increasing alloproliferation of effector memory CD8 T cells. This likely results from altered expression of (i) multiple genes common to the intrinsic CD8 alloproliferative response and (ii) additional genes involved in the control of T cell activation and differentiation specific to lenalidomide-potentiated CD8 alloresponses. Furthermore treatment enhances the functional capacity of these cells by conferring greater polyfunctional effector potential. These findings could enable tracking of CD8 alloresponses induced by lenalidomide after AHCT and could inform novel clinical strategies for the use of the drug to augment GvT effects.
Gribben:Celgene: Research Funding; Pharmacyclics: Honoraria; Roche: Honoraria.
Asterisk with author names denotes non-ASH members.
This icon denotes a clinically relevant abstract