AUTHORS
Ann Ly, BS, Bharath Dileep Kumar, BA2, Chingiz Underbayev, MD, PhD3, Christine Gruessner, MD MPH4, Layla Saleh, PhD5, Victoria Rogness, BS6, Adrian Wiestner, MD, PhD7
AFFILIATIONS
1 National Institutes of Health, Bethesda, Maryland, United States, University of Oxford, Oxford, United Kingdom
2 Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio, United States
3 National Institutes of Health, Bethesda, Maryland, United States
4 National Institutes of Health, Bethesda, Maryland, United States
5 National Institutes of Health, Bethesda, Maryland, United States
6 Massachusetts General Hospital, Boston, Massachusetts, United States
7 National Institutes of Health, Bethesda, Maryland, United States
ABSTRACT
Background:
Chronic lymphocytic leukemia (CLL) is characterized by the accumulation of mature B lymphocytes in the blood, bone marrow, lymph nodes (LN), and spleen. The LN microenvironment supports full activation and proliferation of CLL cells. A complex system of tumor-host crosstalk creates a supportive tumor microenvironment (TME) and inactivates anti-tumor immune responses. Bruton’s tyrosine kinase inhibitors (BTKi) revolutionized CLL treatment by targeting BTK-mediated signaling pathways downstream of constitutively active B-cell receptor (BCR) signaling, the oncogenic driver of CLL. While pharmacodynamic effects of BTKis have been extensively reported, their impact on the LN TME is less well characterized. Here we investigated how BTKis alter immune cell composition and signaling in the LN TME.
Methods:
LN core biopsies were obtained from peripheral nodes before and on day 2 (D2) or D4 after starting ibrutinib (n=34) or D3 or D4 on acalabrutinib (n=29). Studies were registered at clinicaltrials.gov as NCT01500733 and NCT02337829, and conducted compliant with the Declaration of Helsinki. Bulk RNA extracted from unselected biopsies was sequenced and analyzed using DeSeq2 differential gene expression analysis and pre-ranked gene set enrichment analysis (GSEA) with Hallmark and custom SignatureDB gene sets, filtering for terms with false discovery rate q value < 0.1. To determine cell type composition, we used CIBERSORTx RNA-seq deconvolution with the LM22 signature matrix. Further analyses were performed using RStudio.
Results:
We profiled 126 LNs from 63 patients who donated a biopsy before and after starting BTKi treatment (29 on acalabrutinib, 34 on ibrutinib). To define the treatment effect at the LN level, we used GSEA. BTKi treatment downregulated gene signatures of BCR, NFkB, cytokines (IL-2, IL-4, IL-6), NOTCH and CD40 signaling, proliferation (MYC), metabolism (glutamine usage, HIF1alpha) and inflammation. Treatment also significantly decreased a specific CLL activation signature originally derived from single cell sequencing of activated LN-resident CLL cells (Sun C et al. 2023; 34% decrease; p=4.00E-22).
Next, we used RNA deconvolution to estimate immune cell composition with the LNs. The mean tumor content decreased from 75.7% at baseline to 74.4% on treatment (p=0.0159). Interestingly, the B-cell fraction was divided into naïve and memory B cells, and we recorded a 7.5 percentage-point (pp) decrease in B-memory cell content on treatment (p=3.33E-06) likely consistent with inhibition of BCR and NFkB signaling.
In the TME analysis, we scaled the non-tumor cell proportions to 100%. At baseline, 44% of non-tumor cells were identified as macrophages, 43% as T cells with a CD4:CD8 ratio of 10, and 7% as T-follicular helper cells (Tfh). We observed shifts in the proportions of immune cells within the LN in response to both ibrutinib and acalabrutinib. The proportion of CD4 memory activated T cells decreased by 1.2pp (p=1.13E-04), with a corresponding increase in CD4 memory resting T cells (7.4pp; p=4.53E-07). Additional tumor-supportive TME components decreased, with Tfh by 3.8 pp (p=4.73E-06) and macrophages by 7.4pp (p=4.17E-07).
Consistent with changes in cell types, we observed statistically significant on-BTKi decreases in corresponding chemokines, e.g. CXCL13 produced by Tfh, CXCR3 ligands CXCL9/10/11 that regulate migration of activated T cells, and key transcription factor TBX21 regulating activation of T cells. Consistent with a decrease in activation state, we saw increased expression of immune checkpoint inhibitors (CTLA4, LA3, PDCD1), proliferation inhibitors (MXD1, HTR2B, PVRIG), resting B cell markers (BCL7A, VPREB8, ZNF286A), and resting CD4 T cell markers (CD4, CCL23).
Conclusion:
In addition to confirming and extending observations of direct anti-tumor effects of BTKi, we define a BTKi class effect on the CLL LN TME. BTKi reverses the supportive effect of the LN TME on CLL activation. Beyond inhibition of BCR signaling of CLL itself, BTKis profoundly reshape the TME, an effect that is already apparent after only a few days on therapy. These pronounced BTKi-specific TME changes may contribute to its clinical success.
DISCLOSURES
Ann Ly, BS: No relevant conflicts of interest to declare.