Abstract
T cell exhaustion is characterized by coordinated expression of a series of negative checkpoint receptors such as programmed death-1 (PD-1), 2B4, CD160 and TIGIT, resulting in T cell dysfunction and immune evasion. Under physiological states, these inhibitory molecules maintain self-tolerance and prevent autoimmunity by applying a break on cytotoxic T cells. In cancer, T-cells exhibit features of T-cell exhaustion including increased expression of PD-1, 2B4 and CD160, coupled with reduced T cell proliferation, altered synapse formation and impaired cytotoxicity. Although the role of the PD-1/PD-L1 axis in mediating T cell defects in chronic lymphocytic leukemia (CLL) is well-studied, the contribution of other checkpoint molecules such as 2B4, CD160 and TIGIT in mediating tumor-induced immune dysfunction remains to be determined.
Checkpoint inhibitors have provided a paradigm-shifting approach to cancer treatment. We hypothesized that the expression levels of checkpoint receptors on T-cells, as well as the "fitness" of the T cell compartment may provide a prognostic stratification system to predict response to checkpoint inhibitors in CLL. To determine if the number of inhibitory receptors per cell and their expression level may identify patient-to-patient differences that may not be easily deciphered using conventional research tools, we performed a detailed single-cell analysis of the T-cell repertoire, using 40-parameter mass cytometry (CyTOF) in 12 untreated CLL and 12 healthy controls.
Consistent with previous reports, we found that expression of 2B4 (43.7% vs 30.8%), PD1 (28.8% vs 21%) and CD160 (17% vs. 9.7%) was significantly higher on CLL CD8+ T cells compared to healthy controls. In addition, CD8+T cells in CLL expressed higher levels of TIGIT (48.2% vs 25.2%), CD57 (43.9% vs 17.9%) and KLRG1 (49.5% vs. 29.7%). We clearly distinguished 2 patterns of exhaustion marker distribution in CLL. In one group of patients, the expression of checkpoint receptors was similar to that seen in healthy controls, whereas in the second group, CD8+ T-cells expressed higher levels of PD1, 2B4, TIGIT, CD160 as well as markers of terminal differentiation such as CD57 and KLRG1.
Compared to healthy donors, CLL was characterized by an inversion in the CD4:CD8 ratio. Interestingly, CD8+ T cells in patients with a low CD4:CD8 ratio (defined as <2.5) expressed significantly higher levels of 2B4 (56.6% vs 31.25%), TIGIT (62.9% vs 37%), CD160 (22.8% vs 12.6%), CD57 (57% vs 28.7), PD-1 (34.6% vs 24.5%) and KLRG1 (62.3% vs 36.3%). In contrast, the expression levels of PD-1, 2B4 and CD160 in CLL patients with a CD4:CD8 ratio of ≥2.5 were similar to that seen in healthy controls, suggesting that the CD4:CD8 ratio may be a valuable marker of T cell exhaustion in CLL.
Next, we compared the number of checkpoint molecules expressed per CD8+ T-cells in CLL patients versus healthy donors. Whereas a similar proportion of CD8+ T-cells in CLL (mean 19.56%, range 18.34-31.73%) and healthy donors (mean 22.13%, 14.17-41.19%) expressed one inhibitory receptor, a significantly higher proportion of CLL patients expressed 2 and more inhibitory receptors (mean 28.4, range 10.52-48.78%) compared to healthy controls (mean 15.38%, range 9.67-21.94%). PD-1 was mostly co-expressed with TIGIT, although TIGIT+PD-1+CD4+ and CD8+ T-cells were higher in CLL compared to healthy controls (12.9% vs 7.1%). Interestingly the predominant population of PD-1+CD8+ T cells in CLL was also positive for 2B4 and TIGIT, whereas expression of TIGIT was more diverse and was seen in association with PD-1, 2B4, KLRG1 or CD57.
Taken together, our findings indicate a remarkable heterogeneity in the expression patterns of inhibitory molecules on CD8+ and CD4+ T-cells in CLL. While CLL patients with a normal CD4:CD8 ratio expressed comparable levels of inhibitory molecules to that seen in healthy controls, a low CD4:CD8 ratio was indicative of higher expression of checkpoint molecules. On a per cell basis, CLL CD8+ T cells expressed more inhibitory receptors compared to healthy controls, suggesting that certain patients may benefit from combinational use of checkpoint molecules. A more detailed data and analysis, including transcription and functional profile of exhausted CLL T cells, will be presented in the meeting.
Disclosures
Wierda: Abbvie: Research Funding; Novartis: Research Funding; Acerta: Research Funding; Gilead: Research Funding; Genentech: Research Funding. Jain:Incyte: Research Funding; Pharmacyclics: Consultancy, Honoraria, Research Funding; Abbvie: Research Funding; Infinity: Research Funding; BMS: Research Funding; Genentech: Research Funding; Pfizer: Consultancy, Honoraria, Research Funding; ADC Therapeutics: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Honoraria; Seattle Genetics: Research Funding; Celgene: Research Funding; Servier: Consultancy, Honoraria; Novimmune: Consultancy, Honoraria.
A subpopulation of CD8+ T cells specific for melanocyte differentiation antigens expresses killer inhibitory receptors (KIR) in healthy donors: evidence for a role of KIR in the control of peripheral tolerance