Treatment With The Potent PI3K-δ,γ Inhibitor IPI-145 Is Associated With Rapid Decreases In Specific Cytokines, Chemokines and Matrix Metalloproteinases In The Serum Of Patients With Chronic Lymphocytic Leukemia and Indolent Non-Hodgkin Lymphoma

Phosphoinositide-3 kinases (PI3Ks) are key cellular signaling proteins that act as a central node, relaying signals from cell surface receptors to downstream mediators such as AKT. The PI3K-δ and PI3K-γ isoforms are preferentially expressed in normal and malignant leukocytes where they play critical roles in cell differentiation, migration, and proliferation. IPI-145 is a potent, oral PI3K-δ,γ inhibitor that has shown clinical activity in the Phase 1 trial (IPI-145-02) in patients with advanced hematologic malignancies (ClinicalTrials.gov NCT01476657). Cytokines, chemokines, and matrix metalloproteinases (MMPs) play key roles in the homing, migration and activation of normal immune cells, and can have similar effects on malignant leukocytes. To further explore the biological effects of IPI-145, a panel of cytokines, chemokines, and MMPs were evaluated at several time points in the serum of patients enrolled in the IPI-145-02 trial.

Serum was collected from consenting subjects at baseline, Cycle 1 Day 8 (C1D8), and Cycle 2 Day 1 (C2D1). Serum was frozen and stored at -80°C prior to analysis. Serum proteins were analyzed using Luminex xMAP(®) technology in which analytes are captured on uniquely labeled fluorescent beads, and the amount of analyte is quantified using an LED CCD camera contained in Millipore's MagPix(®) instrument. Multiplex panels of cytokines, chemokines and MMPs covering 72 analytes were evaluated in 30 chronic lymphocytic leukemia (CLL) and 19 indolent non-Hodgkin lymphoma (iNHL) subjects. Each sample was tested in duplicate, and duplicate measurements were averaged. Measurements were excluded from further analysis if duplicate readings exhibited a coefficient of variation of greater than 20% or if all values for a specific analyte and subject were below the limit of detection.

Each analyte was evaluated for evidence of a consistent change (reduction or increase) in serum levels at C1D8 and/or C2D1 compared to baseline. When data were compared between CLL subjects treated with IPI-145 at 25 mg twice daily (BID) and those treated at 75 mg BID, no clear differences in serum analyte levels were observed, although the population subgroups were relatively small. For the purposes of this analysis, all doses were pooled together (n=1 at 8 mg BID, 2 at 15 mg BID, 15 at 25 mg BID, and 13 at 75 mg BID). Likewise, the iNHL dose groups were also pooled (n=1 at 15 mg BID, 12 at 25 mg BID, 1 at 50 mg BID, and 5 at 75 mg BID). In CLL subjects, 9 of 72 analytes decreased after IPI-145 treatment compared to baseline, whereas none increased significantly. Analytes that decreased after IPI-145 treatment include CXCL13, CCL3, CCL4, IL-10, TNFα, IL-12p40, MMP-9, CCL17 and CCL22. Median serum levels of these analytes decreased by C1D8, ranging from 16% to 59% of baseline. In iNHL subjects, median serum levels of 7 analytes decreased by C1D8 (ranging from 32% to 70% of baseline), whereas none increased significantly. Of the 7 analytes that decreased in iNHL subjects, 5 also decreased in CLL subjects (CXCL13, MMP-9, TNFα, CCL17 and CCL22) and 2 were distinct (CCL1 and MMP-12). Interestingly, many of the analytes that decreased with IPI-145 treatment are involved in the communication between malignant B-cells and the microenvironment. CCL3, CCL4, CCL17 and CCL22 are expressed by malignant B-cells and may play a role in recruiting T-cells to interact with the malignant B-cells. CXCL13 is secreted by stromal cells and recruits malignant B-cells to the lymph nodes. In addition, IL-10 is produced by many normal immune cell types as well as by neoplastic B-cells. IL-10 is known to be an autocrine growth factor for B-cell lymphoma cell lines.

These pharmacodynamic data provide further evidence for biological activity of IPI-145 in patients with CLL and iNHL and suggest both similarities and differences in how these two malignancies respond to IPI-145. The association of many of these pharmacodynamic factors with the tumor microenvironment suggests a mechanistic basis for the clinical observation of lymphocytosis and nodal reduction with IPI-145 treatment of CLL subjects. Cytokine, chemokine and MMP levels from patients in IPI-145-02 are being evaluated further for associations with multiple clinical parameters to determine if there is evidence for biomarkers predictive of efficacy and tolerability.