AUTHORS
Pamela J. Sung, MD PhD, Elizabeth A Griffiths, MD2, Winny Chan3, Joseph Hanson4, Aram Oganesian, PhD5, Harold Keer, MD, PhD6, Michael Nemeth, PhD7
AFFILIATIONS
1 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
2 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
3 Taiho Oncology, Inc., Pleasanton, California, United States
4 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
5 Taiho Oncology, Inc., Pleasanton, California, United States
6 Taiho Oncology, Inc., Pleasanton, California, United States
7 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
ABSTRACT
Background: The backbone of standard care for patients with higher-risk MDS is epigenetic therapy with the hypomethylating agents (HMAs) azacitidine or decitabine. As single agents, HMAs improve survival, decrease leukemic transformation and transfusion dependence and improve quality of life. However, less than 50% of patients show clinical responses to HMA therapy and responses are typically of limited duration (usually less than 2 years). Developing evidence supports the hypothesis that the clinical mechanism of action for HMAs is not solely due to re-expression of hypermethylated tumor suppressor genes. HMAs have been shown to elicit anti-tumor responses via induction of immune signaling, raising the possibility that HMAs could be activating anti-MDS immune responses as a part of their mechanism of action. Many of these concepts have been studied in the context of pre-clinical models but have yet to be effectively translated to the clinical setting. Since MDS patients receive HMAs as standard-of-care, they are an ideal population for the clinical translation and mechanistic study of these concepts. Dendritic cell (DC) populations control anti-tumor immune responses through interactions with anti-tumor immune cells (e.g. cytotoxic CD8+ T and natural killer (NK) cells). We have shown in patients with MDS that dysregulated production of DC populations is associated with reduced overall survival. We hypothesized that the deficiency of DC populations in patients with MDS contributes to the limited duration of response to HMAs. We further predict that increased numbers of DCs will be associated with increased number of cytotoxic CD8+ T and NK cells that are needed for anti-MDS immunity.
Methods: To test this hypothesis, we performed correlative studies in collaboration with the phase 2-3 portion of the ASTX030 development program (A Multi-phase Study of ASTX030 (Azacitidine and Cedazuridine) in Myeloid Neoplasm Alone or in Combination with Venetoclax in AML (AZTOUND Study; NCT04256317). Serial bone marrow aspirate samples were collected from MDS patients at baseline and during treatment with oral ASTX030/parenteral azacitidine monotherapy on day 1 of cycles 3, and 7. Where cell number permitted, flow cytometry analysis was performed to quantify 1) myeloid DC populations (CD141Hi cDC1 and cDC1+ cDC2); 2) activated and exhausted CD8+ T cells (based on CXCR3 and PD-1 expression); and activated NK populations (based on expression of CXCR4, Nkp30, and NKG2D).
Results: We have observed a 5.5 fold increase in the cDC1 population from baseline to cycle 7 (0.004 + 0.006% of CD45+ cells versus 0.022 + 0.034% at C7D1; n = 16 paired samples; p = 0.03). By contrast, we did not observe a comparable increase in the related cDC2 population. In agreement with prior studies, we observed evidence suggesting that between baseline and cycle 7, HMAs increased numbers of PD-1+ CD8+ T cells; both of CXCR3+ PD-1+ (3.0 + 1.9 versus 4.3 + 2.3%) and more exhausted CXCR3- PD-1+ T cells (2.3 + 2.2% versus 3.4 + 3.4%; n = 7 paired samples; p = 0.016 and p = 0.08 respectively). While there were no changes in the frequency of mature or terminal NK populations, a 1.5-fold increase in expression of the immune activating receptor NKG2D was seen on both populations during therapy (n = 8 paired samples pooled from C3D1 and C7D1 timepoints, p < 0.05 for both populations).
Conclusions: To date, the major observations of this initial analysis are that HMA monotherapy results in expansion of the cDC1 population, increased numbers of active and exhausted PD-1+ CD8+ T cells, and increased expression of the activating NK receptor NKG2D. While analyses of additional samples are required, our initial observations suggest that HMA monotherapy results in increased numbers of cDC1s and subsequent activation of CD8+ T cells and NK cells. These results support our hypothesis that HMAs modulate the myeloid immune microenvironment to enable activation of pro-inflammatory immune responses. This work will inform the development of new immunotherapeutic strategies that can leverage the reshaped immune microenvironment.
DISCLOSURES
Pamela J. Sung, MD PhD: Daiichi-Sankyo - Honoraria; Daiichi-Sankyo - Consultancy (Includes expert testimony)