Influence of polyfunctional Tbet+ T cells on specific clinical events in chronic lymphocytic leukaemia.

INTRODUCTION

T-cell dysfunction is a hallmark of chronic lymphocytic leukemia (CLL), but the extent to which individual CD4+ or CD8+ T-cell subpopulations influence specific clinical events remains unclear. To address this knowledge gap, we utilised high-dimensional mass cytometry to profile circulating CD4+ and CD8+ T-cells in pre-treatment samples from a well-defined cohort of CLL patients undergoing initial therapy as part of a clinical trial.

METHODS

Pre-treatment blood samples from 138 CLL patients receiving initial chemoimmunotherapy containing bendamustine or chlorambucil in the NCRI RIAltO trial (NCT01678430; EudraCT 2011-000919-22) were subjected to deep immunophenotyping by mass cytometry using a bespoke panel of 37 antibodies. T-cell clusters were identified through unsupervised clustering and related to treatment outcomes. Additionally, a randomly selected cohort of 30 CLL patients underwent T-cell stimulation with anti-CD3/CD28 microbeads, followed by cytokine analysis using a separate 36-antibody panel, which included seven cytokines.

RESULTS

Seventeen CD4+ and 22 CD8+ T-cell clusters were identified in a discovery cohort of 79 patients. Three of these clusters, measured as a proportion of their parental CD4+ or CD8+ populations, correlated with a reduced risk of grade ≥3 infection, grade ≥3 second primary malignancy (SPM) and death, respectively. Three corresponding T-cell subpopulations prospectively defined by non-redundant markers and Boolean gating (ICOS+HLA-DR+PD1+TIGIT+Tbet+CD4+ T-helper cells; CD27+CD28-PD1+Tbet+Eomes+CD8+ cells; and CD27+CD28-GrymB+Tbet+Eomes+CD8+ terminal effector cells) showed the same clinical correlations as the clusters on which they were based. With the exception of SPM for which there were insufficient events, these correlations were confirmed in a separate validation cohort of 59 patients. In-vitro stimulation of a subset of CLL patients in the discovery cohort showed an enrichment of primed and polyfunctional cells in all three Tbet+ T-cell subpopulations of interest.

CONCLUSION

Our study provides new insights into the potential for Tbet+ T-cell subpopulations to influence and predict specific clinical events in CLL. This, in turn, raises the possibility that these respective subpopulations could play an important role in controlling infection, solid tumours and CLL itself.

CLINICAL TRIAL REGISTRATION: https://www.clinicaltrials.gov/, identifier NCT01678430; https://www.isrctn.com/ISRCTN09988575, identifier EudraCT 2011-000919-22.