Very Early Effects of Ibrutinib on Tumor and Immune Cells in Blood and Lymph Nodes in Relapsed or Refractory Chronic Lymphocytic Leukemia (CLL) Patients

In this study we analyzed the very early effects of the Bruton's tyrosine kinase (BTK) inhibitor ibrutinib on tumor and immune cells in 7 symptomatic, relapsed or refractory CLL patients during the first four weeks of treatment. Five of the patients had Rai stage IV at study entry and four patients had 17p deletion or TP53 mutation. Median number of previous treatment regimens was 2 (range 1-4). Peripheral blood (PB) samples were collected before (< 1 week) treatment start and at six different time points during the treatment (9 hours after treatment start on day 1; on day 2; day 4; day 8; day 15 and day 29). Fine needle aspiration of two pathological lymph nodes (LN) identified by ultrasound was performed before (< 1 week) treatment start and at day 2, day 8 and day 29. Flow-cytometry was performed to analyze the changes in the peripheral blood mononuclear cell populations in PB including CLL cells, Natural Killer (NK) cells, monocytes, dendritic cells (DC) and T cells memory subsets, helper subpopulations (Ths) and regulatory T cells (Tregs). Moreover, changes in expression of 18 B-cell activation and migration markers on CLL cells as well as T-cell activation and proliferation markers, were analyzed in paired LN and PB samples. Finally, plasma levels of 92 inflammation-related protein biomarkers were assessed by Multiplex Proximity Extension Assay (PEA).

In six out of seven patients the size of the LN decreased gradually during the four weeks of observation, achieving complete clinical remission in three patients and partial remission in three other. In 5 evaluated patients, the CLL cell counts in PB increased already 9 hours after treatment start (fold increase 1.36-2.84). In all the patients, CLL cells were higher at day 2 and 8 compared to baseline (p=0.02), and decreased by day 29 to levels not significantly different from the baseline. Over the four weeks period CD4⁺ cells increased (p=0.03), while CD8⁺, NK, and NKT cells remained stable. The distribution of the CD4⁺ and CD8⁺ memory cell subsets remained unchanged. Th1 cells increased (p=0.01) while Th2, Th17 and Tregs were stable.

Expression of Programmed cell death protein 1 (PD-1) and Cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) in T cells was unaffected. No effect was seen on CLL apoptosis. Proliferating (Ki67⁺) CLL cells, which at baseline varied between 1.2-10.8%, were not detected in blood after 1 week of treatment (p=0.0006). Ki67⁺ T cells were also eliminated (p=0.0007), but with a delayed kinetics (4 weeks).

Paired LN and PB samples were screened for changes in 18 surface markers relevant for B-cell activation and migration. With the exception of CD23, whose expression was lower in PB compared to LN (p=0.02), no difference was observed in the expression of the other analyzed surface markers on CLL cells from the two different tumor compartments at baseline. CD5 expression remained stable during treatment, albeit decreasing significantly by day 29 in PB (p=0.02). Expression of CD20 decreased in both compartments at day 8 (p=0.02) and CD40 decreased at day 8 only in PB (p=0.02). CD69 expression (MFI) decreased both in PB and LN by day 8 (p=0.02), while the percentage of positive cells significantly dropped already at day 2 in PB (p=0.02) and from day 8 in LN (p=0.03). CD23 MFI decreased in the LN compartment already at day 2 (p=0.02), but the percentage of positive cells was significantly lower only at day 8 (p=0.03). Compared to LN, CD23 expression was lower in PB before treatment and therefore the significant drop both in MFI and percentage of positive cells was seen only after 4 weeks (p=0.02). No change in the expression of CD49d was observed.

In 5 evaluable patients, the plasmacytoid DCs, undetectable at baseline, increased during treatment (p=0.06), while CD16⁺SLAN⁺monocytes decreased (p=0.06).

Finally, plasma levels of 50 molecules were significantly changed at ≥ 1 time point during treatment. With the exception of CST5 and IL-17C, which were increased, the residual 48 proteins were all down-regulated, some of them (e.g. CCL3 and CCL4) already by 9 hours. The majority of the cytokines with significantly reduced levels are pro-inflammatory molecules.

In conclusion, these data indicate that ibrutinib causes major changes both in CLL and bystander cells as well as in the levels of several inflammation-related protein biomarkers already shortly after treatment initiation.