Duvelisib treatment is associated with altered expression of apoptotic regulators that helps in sensitization of chronic lymphocytic leukemia cells to venetoclax (ABT-199)

Abstract

Duvelisib, an oral dual inhibitor of PI3K-δ and PI3K-γ, is in phase III trials for the treatment of chronic lymphocytic leukemia (CLL) and indolent non-Hodgkin's lymphoma. In CLL, duvelisib monotherapy is associated with high iwCLL (International Workshop on Chronic Lymphocytic Leukemia) and nodal response rates, but complete remissions are rare. To characterize the molecular effect of duvelisib, we obtained samples from CLL patients on the duvelisib phase I trial. Gene expression studies (RNAseq, Nanostring, Affymetrix array and real-time RT-PCR) demonstrated increased expression of BCL2 along with several BH3-only pro-apoptotic genes. In concert with induction of transcript levels, reverse phase protein arrays and immunoblots confirmed increase at the protein level. The BCL2 inhibitor venetoclax induced greater apoptosis in ex vivo-cultured CLL cells obtained from patients on duvelisib compared with pre-treatment CLL cells from the same patients. In vitro combination of duvelisib and venetoclax resulted in enhanced apoptosis even in CLL cells cultured under conditions that simulate the tumor microenvironment. These data provide a mechanistic rationale for testing the combination of duvelisib and venetoclax in the clinic. Such combination regimen (NCT02640833) is being evaluated for patients with B-cell malignancies including CLL.

Trial registration: ClinicalTrials.gov NCT01476657.

Conflict of interest statement

Conflict-of-interest: KB/VG received sponsored research agreement from Infinity Pharmaceuticals, Inc. MD, TT, EYX, JLK, HMS are employees and shareholders of Infinity Pharmaceuticals, Inc. Other authors declare no competing financial interests.

Figures

Figure 1

Effect of oral duvelisib therapy on peripheral blood absolute lymphocyte count (ALC) and white blood cell (WBC count). Patients with CLL received oral duvelisib therapy. Peripheral blood samples were obtained prior to therapy (baseline) and 28 days after start of therapy. Absolute lymphocyte count and WBC were measured. (A) Changes in ALC and (B) WBC counts from 17 CLL patients on oral duvelisib therapy were plotted prior to treatment (C1D1; cycle 1 day 1) and on day 28 (C2D1; cycle 2 day 1).

Figure 2

Base-line expression levels of anti-apoptotic and pro-apoptotic Bcl-2 family transcripts in primary CLL lymphocytes. (A) Schema of BCL2 sub-families reflecting anti-and pro-apoptotic molecules. (B) Anti-apoptotic Bcl-2 family members (C) Pro-apoptotic multi-domain and (D) Pro-apoptotic BH3 only protein transcripts in peripheral blood CLL cells. CLL cells were obtained prior to any therapy and transcript levels were measured in isolated RNA through RNAseq method as described under Methods and Supplementary Material. The transcripts that are not shown were not in the available assay.

Figure 3

Effect of oral duvelisib therapy on expression of transcripts of BCL2 family in CLL lymphocytes after 7 days (C1D8; cycle 1 day 8) and 28 days (C2D1; cycle 2 day 1) of oral duvelisib intake. (A) Real-Time Quantitative Reverse Transcription PCR (qRT-PCR) using microfluidic chip analysis showing fold change in BCL2 family mRNA expressions before and after 28 days of duvelisib therapy from 5 patient samples. (B) RNAseq analysis and (C) Nanostring quantitation showing log2 fold change in BCL2 family mRNA expressions in CLL B cells isolated from patients (n=31) before and after 7 days of duvelisib therapy. CLL cells were obtained before and after duvelisib therapy and transcript levels were measured in isolated RNA through different methods as described under Methods and Supplementary Material. Detailed information regarding median value and p values is provided in the Supplemental Table 4. The transcripts that are not shown were not available in the assay.

Figure 4

Effect of oral duvelisib therapy on expression of BCL2 family proteins in CLL lymphocytes after 28 days (C2D1; cycle 2 day 1) of oral duvelisib intake. (A–C) Reverse phase protein array (RPPA) analysis showing log2 difference in BCL2 family proteins expressions in CLL B cells isolated from patients before and after 28 days of duvelisib therapy (n=16). CLL cells were obtained pre and post duvelisib therapy and protein was extracted. Protein extracts were analyzed through reverse-phase protein array as described in the Methods section. Proteins that are not included were not available in the assay. Detailed information regarding median value and p values is provided in the Supplemental Table 4.

Figure 5

Effect of oral duvelisib therapy on expression of BCL-2 family proteins in CLL lymphocytes before therapy (C1D1) and after 28 days (C2D1) of oral duvelisib intake. (A and B) Immunoblots of expression level of BCL2, Bim, Puma, and Bax proteins in CLL cells from patients before and after 28 days of oral duvelisib intake (n=15). Actin was used as normalization protein. These proteins were selected based on their changes as determined by the RPPA assay, available antibodies, and availability of amount of CLL samples. (C) Quantitation of these blots for all the patient samples analyzed (n=16). Statistical analyses were conducted to obtain p values that are included in the text.

Figure 6

Ex-vivo pharmacological intervention in duvelisib treated CLL cells with venetoclax (ABT-199). (A) Schematic model representing mechanistic rationale for combining duvelisib and venetoclax. (B) % cell death of lymphocytes obtained from patients (n=14) after 28 days of oral duvelisib intake when incubated with different pharmacological agents (ibrutinib, ABT-737, venetoclax and idelalisib) in RPMI + 10% HS media for 24 hour. (C) % cell death of leukocytes obtained from patients (n=5) before and after 28 days of oral duvelisib intake when incubated with venetoclax (10 nM) in RPMI + 10% HS media for 24 hour. Each pre-treatment and on-treatment sample was normalized to endogenous cell death. Statistical analyses were conducted to obtain p values that are included in the text.

Figure 7

Effect of in vitro combination of duvelisib and venetoclax on CLL cell death. In-vitro incubations of CLL cells obtained from previously untreated CLL patients with duvelisib (4 days) and venetoclax (12–16 hours) either alone or in sequential combinations in suspension culture, IgM-stimulated cells, and in media containing cytokine cocktail. (A) Schema to represent processing and culture conditions for in vitro incubation of CLL cells. (B–D) Cell death measurement of CLL cells after incubation with vehicle (DMSO), duvelisib alone, venetoclax alone, and combination of venetoclax and duvelisib in three different culture conditions; (B) suspension culture, (C) soluble IgM-stimulation, (D) cytokine cocktail-stimulation. Statistical analyses were conducted to obtain p values that are included in the text.