Vulnerability of Small-Cell Lung Cancer to Apoptosis Induced by the Combination of BET Bromodomain Proteins and BCL2 Inhibitors

Ten percent to 15% of all lung cancers are small-cell lung cancer (SCLC), named for the size of the cancer cells when seen under a microscope and strongly correlated with smoking (1, 2). SCLC often starts in the bronchi near the center of the chest and usually grows and metastasizes rapidly before it is diagnosed. Although a good initial response to chemotherapy and radiotherapy is achieved in most patients, almost all patients relapse within a year with resistant disease. Unfortunately, no new targeted treatment has been approved in the past 30 years for patients with SCLC (3).

Recent genome-wide characterization of SCLC identified a high frequency of genomic alterations linked to mutagen exposure in tobacco smoke. There are uniform recurrent mutations in SCLC that include loss of function of the tumor suppressors TP53 and RB1 in 75% to 90% tumors (4). Loss of TP53 allows for genomic instability that leads to accumulation of further driver mutations, such as MYC, MYCN, or MYCL amplification, and PTEN loss or mutations. Upregulation of the antiapoptotic proteins, such as BCL2, BCLxl, and MCL1, was shown in SCLC cell lines and tumors (5, 6). These antiapoptotic proteins bind proapoptotic BH3 domain proteins, such as BIM and BAD, and prevent cells from undergoing apoptosis (7).

Although multiple targeted agents are being tested in clinical trials, none of the agents are showing sustainable efficacy (3). There is increasing evidence that immune responses against SCLC cells make immunotherapy a viable therapeutic approach, but further understanding and testing is required (8). Given the unmet need in this indication, additional therapies that can extend survival are required. The BET (bromodomain and extraterminal) proteins bind acetylated histones or transcription factors and recruit protein complexes to promote transcription initiation and elongation. There are four members in this family, BRD2, BRD3, BRD4, and BRDT, that share a common structure of two N-terminal bromodomains that have high levels of sequence conservation as well as an extraterminal domain and a more divergent C-terminal recruitment domain. In hematologic cancers, BET proteins have been shown to regulate expression of MYC and other oncogenic transcription factors that drive disease pathology (9). Pharmacologic inhibition of BET binding to acetylated proteins has been shown to inhibit tumor growth in MYC-dependent cancers, such as leukemia (10, 11), multiple myeloma (9, 12), acute myeloid leukemia (AML; 13), aggressive B-cell lymphoma, such as Burkitt lymphoma (14), and neuroblastoma (15).

ABBV-075 is a potent inhibitor of all BRD family proteins (BRD2, BRD3, BRD4, and BRDT; refs. 16–18) and is currently being tested in a phase I clinical trial (ClinicalTrials.gov; NCT02391480). SCLC was shown to be sensitive to BET inhibition in preclinical models (19, 20); here, we demonstrate that SCLC cell lines are sensitive to ABBV-075 and elucidate the potential mechanism of action. Approximately 50% of SCLC cell lines are exquisitely sensitive to growth inhibition by the BET inhibitor ABBV-075, and many of them undergo BIM-mediated apoptosis. In addition, synergistic cell killing was observed when cotreating the cells with ABBV-075 and the BCL2 inhibitor, venetoclax (ABT-199; refs. 21, 22), and activity correlated with BCL2 expression. This combination was also efficacious in SCLC xenograft models. Our results identify a novel therapeutic combination for SCLC, and support a viable rationale for additional testing of the potential benefit of venetoclax with BET inhibitors.

ABBV-075 (N-[4-(2,4-difluorophenoxy)-3-(6-methyl-7-oxo-6,7-dihydro-1H-pyrrolo[2,3-c]pyridin-4-yl)phenyl]ethanesulfonamide) (23) and venetoclax (ABT-199) were synthesized at AbbVie, Inc. All siRNAs (control: 6568S; BCL2L11: 6461S; BAK: 6486S; BAX: 6321S; BAD: 6512S, and BCL2L1: 6362S) were purchased from Cell Signaling Technology.

SCLC cell lines were obtained from ATCC and The Leibniz Institute DSMZ – German Collection of Microorganisms and Cell Cultures between 2003 and 2006. The cells were tested for mycoplasma using MycoAlert Detection Kit (Lonza), authenticated using GenePrint 10 STR Authentication Kit (Promega) prior to performing experiments, and grown in media as recommended by the supplier.

siRNAs were introduced into the cells by reverse transfection using Lipofectamine RNAiMax according to the manufacturer's instructions (Invitrogen). Briefly, siRNAs were first mixed with Lipofectamine RNAiMax in Opti-MEM (Invitrogen) and dispensed in 96-well tissue culture plates. Cells were added at 1.5–2.5 × 10⁴ cells/100 μL to final concentration of 20 nmol/L siRNA. The cells were then grown in medium with ABBV-075. Forty-eight hours after transfection, cells were assayed for viability.

Cell viability and caspase-3/7 activation were measured using CellTiter-Glo or Caspase-Glo 3/7 luminescent assays, respectively, according to the manufacturer's protocol (Promega). IC₅₀ calculation was performed using GraphPad Prism software. Statistical analysis was carried out using Microsoft Excel to determine P value (two-tailed), and P < 0.05 is indicated by "*" in the figures.

Caspase inhibitor Z-VAD-FMK purchased from MPBio was preincubated with cells for 30 minutes before treatment with ABBV-075.

Cells in logarithmic growing phase were treated with compounds for designated time. Cells were harvested and subjected to cell-cycle or apoptosis analyses using PI/RNase Staining Buffer or Annexin V Apoptosis Detection Kit I, respectively, according to the manufacturer's protocols (BD Pharmingen). All flow cytometry analyses were performed on FACSCalibur (BD Biosciences).

Combination studies were performed in 20 SCLC cell lines with increasing concentration of ABBV-075 and venetoclax. Synergistic activity of ABBV-075 and venetoclax was determined using the Bliss additivity model (24), whereby the combined response C of both agents with individual effects A and B is C = A + B – (A × B), where A and B represent the fractional inhibition between 0 and 1. Combined response scores greater than 600 were considered strongly synergistic, 400–600 were considered synergistic, 250–400 were considered additive, and 0–250 were considered no combination effect.

Cell lysates were prepared in cell lysis buffer (Cell Signaling Technology) with protease inhibitor cocktail (Roche). Thirty micrograms of total protein was resolved on a 12% SDS polyacrylamide gel and probed with anti-MCL1 (Cell Signaling Technology #94296), anti-BCLxl (Cell Signaling Technology #2764), anti-BCL2 (Abcam Ab-32124), anti-BIM (Cell Signaling Technology # 2933), and anti-actin (Santa Cruz Biotechnology).

All reagents for qPCR were obtained from Life Technologies. RNA was harvested using the PureLink RNA Mini Kit. cDNA was generated with the High-Capacity cDNA Reverse Transcription Kit. qPCR was performed with TaqMan Universal PCR master mix and predesigned probes on a Quant Studio Flex6 (BCL2L1 Hs00236329_m1; BAX Hs00180269_m1; BAD Hs00188930_m1; BCL2 Hs00608023_m1; BCL2L11 Hs01076940_m1; BAK Hs00832876_g1). Relative gene expression was analyzed using the ΔΔC_t method.

QuantiGene assay was described previously (25) and performed according to the manufacture's protocol (Affymetrix/eBioscience, Inc.).

ELISA to detect BCL2–Bim complex was performed according to Phillips and colleagues (26).

Female SCID Beige mice (Charles River Laboratories) were group-housed 8 to 10 per cage and weighed 21 to 22 grams at initiation of therapy. Food and water were available ad libitum throughout the studies. All experiments were conducted in compliance with AbbVie's Institutional Animal Care and Use Committee, and the NIH Guide for Care and Use of Laboratory Animals guidelines in a facility accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care.

To generate xenografts, a suspension of viable tumors cells mixed with Matrigel (phenol red free, BD Biosciences) in a ratio of 1:1 (V/V) was injected subcutaneously into the flank of 6- to 8-week-old mice (5 × 10⁶ cells/site). Inoculated mice were randomized into groups and treatment was initiated when mean tumor volume was approximately 200 mm³. Tumor growth was assessed by measuring tumor size with calipers and calculating volume using the formula (L × W²/2; L = length, W = width). Inhibition of tumor growth was assessed at the time the vehicle-treated group was terminated by calculating the ratio of the mean volume of the test drug group to the mean volume of the vehicle-treated (control) group (T/C) and calculating percent inhibition of control [(1−T/C) × 100)].

We first investigated the sensitivity of 24 SCLC cell lines to the BET inhibitor, ABBV-075. ABBV-075 demonstrated strong growth inhibition of half of the SCLC cell lines (Fig. 1A) as defined by IC₅₀ values ≤0.1 μmol/L. In addition, the BET inhibitor induced apoptosis in half of the SCLC cell lines tested, as determined by ≥40% positive Annexin V staining (Fig. 1B). There was a significant correlation (R² = 0.59) between apoptosis and antiproliferative activity (Fig. 1C).

We next investigated the cellular response to ABBV-075 to better understand the sensitivity of SCLC cell lines. Caspases are proteases that “execute” apoptosis. To determine whether caspase-3/7 are involved in the BET inhibitor activity, we treated multiple SCLC cell lines with increasing concentration of ABBV-075. SCLC cell lines (NCI-H211, NCI-H146, H1963, and H69) that are more sensitive to ABBV-075 showed strong activation of caspase-3/7, whereas less sensitive cell lines (NCI-H1048 and H345) did not show caspase activation (Fig. 1D). The activation of caspase-3/7 was abrogated with preincubation of NCI-H211, NCI-H146 and H1963 cells with pan-caspase inhibitor prior to treatment with ABBV-075 (Supplementary Fig. S1). Together, these data suggest that inhibition of BET family proteins could be an effective treatment for SCLC tumors and that apoptosis could be induced by ABBV-075 in SCLC cell lines. We also noted that some SCLC lines were sensitive to the antiproliferative effects of ABBV-075 without evidence of apoptosis (Fig. 1B). As BET inhibitor is known to inhibit cell cycle at the G₁ phase, we tested whether nonapoptotic cell lines underwent cell-cycle arrest. Indeed, NCI-H1048 cells showed increase in G₁ arrest and decreased S-phase (Fig. 1E). Our data indicate that BET inhibitor could assert its effect through apoptosis or cell-cycle arrest.

Given that apoptosis was observed in these cells, we reasoned that components of the intrinsic apoptotic pathway may be involved in the sensitivity to BET inhibitor. We first evaluated the role of proapoptotic proteins in mediating ABBV-075–induced apoptosis by silencing the expression of proapoptotic proteins using siRNAs. siRNAs that rescued the cells from ABBV-075–induced apoptosis could serve to identify proteins that mediate this pathway. We focused on BIM and BAD as they were recently described to play a role in response to BET inhibitors (14, 27–30). BIM is an activator BH3-only protein that binds and inactivates all antiapoptotic proteins and is essential for activation of BAX- and BAK-dependent cell death program (31). BAD is a sensitizer BH3-only protein that binds and inactivates BCL2 and BCLxl. BAX and BAK initiate apoptosis by forming a pore in the mitochondrial outer membrane that allows cytochrome c to be released into the cytoplasm and activate the caspase cascade. The silencing of these proteins would act as controls for preventing apoptosis. As shown in Fig. 2A, silencing BCL2L11 (encodes BIM), BAD, BAX, and BAK led to a substantial or partial rescue from ABBV-075–induced apoptosis in H1963 or NCI-H146 cells, respectively. The level of silencing of these genes by siRNAs was shown in Supplementary Fig. S2.

An independent evaluation on the role of these proteins was performed by gene expression analysis of proapoptotic proteins (BCL2L11 that encodes BIM, BAD, BAX, and BAK) and antiapoptotic proteins (BCL2 and BCL2L1 that encode BCLxl) in NCI-H146 and H1963 cells upon treatment with ABBV-075. Expression of BCL2L11 was induced around 2-fold (Fig. 2B), while expression of BCL2 and BCL2L1 was reduced in both NCI-H146 and H1963 cells (Supplementary Fig. S3A). No change in BAD, BAX, and BAK expression was observed (Fig. 2B). In addition, ABBV-075 treatment induced BIM protein expression (Fig. 2C) and moderately reduced BCL2 and BCLxL protein levels (Supplementary Fig. S3B and S3C). The increased expression of proapoptotic BCL2L11 (BIM) together with small reduction in antiapoptotic BCLxl and BCL2 may tip the scales toward an apoptotic response to ABBV-075 in sensitive SCLC cell lines.

BCL2 overexpression and gene amplification are common in SCLC (32). Our findings that ABBV-075 induces apoptosis and modulates the expression of BH3 protein led us to speculate whether apoptosis can be further augmented by combination with a specific BCL2 inhibitor, ABT-199 (venetoclax; ref. 21). We found that there was a range of cell killing when combining ABBV-075 with ABT-199 in SCLC cell lines (Fig. 3A). Cell lines that showed strong synergy include NCI-H146, H1963, H345, and NCI-H526 (Fig. 3B). NCI-H146, H1963, and NCI-H526 were also sensitive to single-agent ABBV-075. Interestingly, a few cell lines (such as H345) that were resistant to single-agent ABBV-075 were sensitized by the combination treatment with venetoclax. Notably, the majority of the cell lines that demonstrated synergistic effect between ABBV-075 and ABT-199 express BCL2 (Fig. 3C). In Fig. 3D, quantitation of BCL2 protein expression was shown in synergy and no synergy cell lines. There were a few cell lines in which we did not observe synergy despite high BCL2 expression. NCI-H889 cells are known to have high BCL2 amplification and are solely dependent on BCL2 for survival (22); thus, the effect of adding BET inhibition is marginal. On the other hand, even though NCI-H211 and DMS53 express BCL2, these lines are very sensitive to single-agent ABBV-075 and that could explain the lack of synergy with ABT-199. Other than cell lines that are sensitive to single-agent ABBV-075 or ABT-199, all cell lines that exhibited synergy express BCL2 protein (BCL2-to-actin ratio >0.2), whereas only 1 of 5 cell lines that exhibited no synergy express BCL2 protein. Nevertheless, our data provide a rationale for treating SCLC with BET and BCL2 inhibitors in cells that express BCL2.

BIM is usually sequestered by BCLxl and/or BCL2 to prevent cells from undergoing apoptosis. As ABBV-075 induced the expression of BIM and moderately reduced BCL2 and BCLxl expression, we reasoned that free BIM could be sequestered by BCL2 to avoid cell death. The addition of ABT-199 would thus free BIM from BCL2 and induce cell death. To examine this potential mechanism of action for this drug combination, we measured the BCL2–BIM complex upon treatment of ABBV-075 with or without the addition of ABT-199 (Fig. 4A). As shown in Fig. 4B–D, ABBV-075 treatment led to increase of BIM bound to BCL2 in NCI-H146, H1963, and H345 cell lines. Importantly, ABT-199 strongly disrupted BCL2–BIM complexes in all these cells. These results indicate that the combination of ABBV-075 and ABT-199 would be more efficacious in inducing apoptosis than single agents alone.

To gain further insight into the sensitivity of SCLC cell lines to ABBV-075 alone and in combination with the BCL2 inhibitor, we determined the dependency of SCLC cell lines to BCLxl and BCL2. Silencing BCL2L1 expression with siRNA led to reduced viability in NCI-H146 and H1963 (Fig. 5), indicating their dependence on BCLxl for survival. Interestingly, the addition of ABT-199 further decreased the viability of these cells. Our data suggested that SCLC cells have a dual dependency on BCLxl and BCL2. Indeed, the majority of SCLC cell lines were sensitive to a BCL2/BCLxl dual inhibitor ABT-263 but not to BCL2-selective inhibitor ABT-199 or BCLxl-selective inhibitor A-1155463 (22).

On the basis of the in vitro results, we undertook in vivo studies to investigate the efficacy of ABBV-075 in combination with venetoclax. NCI-H146 and H1963 models were chosen as we observed strong combination effect in vitro (Fig. 6A and B). Although single-agent treatment of ABBV-075 or ABT-199 showed marginal efficacy in H1963, tumor regression was observed for the combination during the treatment phase. No obvious toxicity was observed during the treatment period (maximum weight loss for treatment groups was ≤8%). Although the tumors regrew shortly after the treatment was stopped, they regressed again with re-treatment. Similarly, single-agent activity in NCI-H146 was observed, but the combination of ABBV-075 and ABT-199 elicited a more robust effect during the treatment phase. No obvious toxicity was observed during the treatment period.

Collectively, these findings elucidate the mode of action of ABBV-075 in combination with ABT-199 in SCLC. As SCLC cell lines have dual dependence on BCLxl and BCL2, the increase of BIM and moderate decrease in BCL2L1 and BCL2 expression by ABBV-075 together with the BCL2 inhibitor, ABT-199, induce strong apoptosis in these cells (Fig. 6C).

We report here that multiple SCLC cell lines are sensitive to BET inhibition by ABBV-075. Many of the cell lines underwent apoptosis via induction of caspase-3/7. Upregulation of proapoptotic protein BIM and downregulation of antiapoptotic protein BCLxl and BCL2 were observed. In addition, silencing BCL2L11 could rescue sensitive cells from BET inhibition. Synergy was observed when cotreating BET inhibitor with BCL2 inhibitor venetoclax (ABT-199) in vitro and in vivo, and in vitro efficacy positively correlated with BCL2 expression.

We found that high percentage of SCLC cell lines underwent apoptosis upon BET inhibitor treatment. High percentage of cellular apoptosis was also detected in AML cell lines (13), and other hematologic malignancies (33). We hypothesized that sensitivity of SCLC to BET inhibitor is due to the dependency of SCLC cells to antiapoptotic proteins, such as BCL2 and BCLxl. SCLC is well known to overexpress BCL2 proteins in part due to amplification of BCL2 locus (6). In addition, most SCLC cells express BCLxl and/or MCL1 (5). These antiapoptotic proteins sequester proapoptotic proteins, such as BIM, to inhibit cell death. We and others have shown that SCLC cells have strong dependency on BCLxl and BCL2 for survival (Fig. 5; ref. 5). In particular, SCLC cell lines are sensitive to silencing BCL2L1 in combination with BCL2 inhibitor venetoclax (ABT-199). Others have shown that SCLC cell lines are sensitive to treatment with navitoclax (ABT-263), a dual inhibitor of BCL2 and BCLxl (22). In contrast, other solid tumors were shown to depend mainly on BCLxl and MCL1 but not BCL2 for survival (34). The role of MCL1 in BET inhibitor sensitivity in SCLC cells is currently under investigation. Interestingly, a recent report also revealed that the expression signatures of the apoptotic genes BCL2, BCL2L1, and BAD significantly predict response to BET inhibitor (30), suggesting the apoptotic program as a determinant of response to BET inhibitors.

The sensitivity of SCLC to BET inhibitor(s) has been recently reported (19) and shown to be mediated by downregulation of the neuroendocrine transcription factor ASCL1. We focused on ABBV-075–induced apoptosis and demonstrate modulation of BIM, BCL2, and BCLxl expression as the key mechanism of action that determines sensitivity and growth inhibition. The mechanism of BIM induction is unclear but could potentially be through FOXO3a posttranscriptional mechanisms (35). Furthermore, although SCLC tumors and cell lines have MYC, MYCN, and MYCL amplifications, association between sensitivity to BET inhibitor and overall MYC status is currently under investigation. Literature review suggests that MYC expression may play a role in leukemia (10, 11), multiple myeloma (9, 12), and hepatocellular carcinoma (27), and MYCN in neuroblastoma (15). However, the role of MYC in SCLC is uncertain. For example, one study shows that MYC does not play a role in response to BET inhibitor JQ1 in SCLC (19), while another study suggests BET inhibitor JQ1 is more active in MYC-amplified cell lines (20). In addition, BET inhibition with JQ1 occurs in the absence of modulation of MYC gene expression in aggressive B-cell lymphoma (14). Further experimentation is required to determine the role of MYC status in the response to BET inhibitor.

We showed very strong combination effect between BET inhibitor ABBV-075 and the BCL2 inhibitor venetoclax (ABT-199). This combination was recently shown to be effective in an AML cell line and in primary double hit lymphoma cells (30, 36). We showed that there is an induction of BIM by BET inhibition that could prime cells to apoptosis. In addition, we showed that both BCL2 and BCLxl are moderately reduced, and more BIM is sequestered by BCL2 upon BET inhibitor treatment. The subsequent addition of BH3 peptidomimetics would free BIM and induce apoptosis. Similar observations between BCL2 inhibitors with targeted agents that induce BIM were observed by others. For example, EGFR inhibitors are known to induce BIM in NSCLC cells, and combination was observed with ABT-737, an analogue of navitoclax (37, 38).

On the basis of these observations, a combination regimen including BET inhibitor and BCL2 inhibition would be highly desirable for treating SCLC cancer. Future studies should investigate the efficacy in patients and the utility of BCL2 expression or amplification as a stratification biomarker.

No potential conflicts of interest were disclosed.

AbbVie participated in the interpretation of data, review, and approval of the manuscript.

Conception and design: L.T. Lam, X. Lin, Z. Yang, X. Huang, D.H. Albert, Y. Shen, T. Uziel

Development of methodology: L.T. Lam, Z. Yang, X. Huang, R.J. Bellin

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L.T. Lam, X. Lin, E.J. Faivre, Z. Yang, X. Huang, D.M. Wilcox, R.J. Bellin, S. Jin, M. Mitten, T. Magoc

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L.T. Lam, X. Lin, E.J. Faivre, Z. Yang, X. Huang, R.J. Bellin, S. Jin, D.H. Albert, Y. Shen, T. Uziel

Writing, review, and/or revision of the manuscript: L.T. Lam, X. Lin, Z. Yang, X. Huang, S.K. Tahir, M. Mitten, T. Magoc, A. Bhathena, W.M. Kati, D.H. Albert, Y. Shen, T. Uziel

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L.T. Lam, Z. Yang, X. Huang

Study supervision: L.T. Lam, M. Mitten, T. Magoc, A. Bhathena, T. Uziel

We thank AbbVie members of the BET biology and apoptosis groups for helpful discussion. We thank Joel Leverson, an AbbVie employee, for critical review of the manuscript.

The design, study conduct, and financial support for this research were provided by AbbVie.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.