The Bruton tyrosine kinase (BTK) inhibitor ibrutinib has substantially improved therapeutic options for chronic lymphocytic leukemia (CLL). While ibrutinib is not curative, it has a profound effect on CLL cells and may create new pharmacologically exploitable vulnerabilities. To identify such vulnerabilities, we developed a systematic approach that combines epigenome profiling (charting the gene-regulatory basis of cell state) with single-cell chemosensitivity profiling (quantifying cell-type-specific drug response) and bioinformatic data integration. Applying our method to a cohort of matched patient samples collected before and during ibrutinib therapy, we identified characteristic ibrutinib-induced changes that provide a starting point for the rational design of ibrutinib combination therapies. Specifically, we observed and validated preferential sensitivity to proteasome, PLK1, and mTOR inhibitors during ibrutinib treatment. More generally, our study establishes a broadly applicable method for investigating treatment-specific vulnerabilities by integrating the complementary perspectives of epigenetic cell states and phenotypic drug responses in primary patient samples.

Deregulation of NOTCH2 signaling is implicated in a wide variety of human neoplasias. The current concept of targeting NOTCH is based on using gamma secretase inhibitors (GSI) to regulate the release of the active NOTCH intracellular domain. However, the clinical outcome of GSI remains unsatisfactory. Therefore we analyzed human solid tumor derived cell lines for their nuclear NOTCH activity and evaluated the therapeutic potential of the NOTCH2 transactivation inhibitor gliotoxin in comparison to the representative GSI DAPT. Electrophoretic mobility shift assays (EMSA) were used as a surrogate method for the detection of NOTCH/CSL transcription factor complexes. The effect of gliotoxin on cell viability and its clinical relevance was evaluated in vitro and in a melanoma xenograft mouse model. Cell lines derived from melanoma (518A2), hepatocellular carcinoma (SNU398, HCC-3, Hep3B), and pancreas carcinoma (PANC1) express high amounts of nuclear NOTCH2. Gliotoxin efficiently induced apoptosis in these cell lines whereas the GSI DAPT was ineffective. The specificity of gliotoxin was demonstrated in the well differentiated nuclear NOTCH negative cell line Huh7, which was resistant to gliotoxin treatment in vitro. In xenotransplanted 518A2 melanomas, a single day dosing schedule of gliotoxin was well tolerated without any study limiting side effects. Gliotoxin significantly reduced the tumor volume in early (83 mm3 vs. 115 mm3, p = 0.008) as well as in late stage (218 mm3 vs. 576 mm3, p = 0.005) tumor models. In conclusion, NOTCH2 appears to be a key target of gliotoxin in human neoplasias and gliotoxin deserves further evaluation as a potential therapeutic agent in cancer management.

NOTCH signaling represents a promising therapeutic target in chronic lymphocytic leukemia (CLL). We compared the anti-neoplastic effects of the nuclear NOTCH2 inhibitor gliotoxin and the pan-NOTCH γ-secretase inhibitor RO4929097 in primary CLL cells with special emphasis on the individual roles of the different NOTCH receptors. Gliotoxin rapidly induced apoptosis in all CLL cases tested, whereas RO4929097 exerted a variable and delayed effect on CLL cell viability. Gliotoxin-induced apoptosis was associated with inhibition of the NOTCH2/FCER2 (CD23) axis together with concomitant upregulation of the NOTCH3/NR4A1 axis. In contrast, RO4929097 downregulated the NOTCH3/NR4A1 axis and counteracted the spontaneous and gliotoxin-induced apoptosis. On the cell surface, NOTCH3 and CD23 expression were mutually exclusive, suggesting that downregulation of NOTCH2 signaling is a prerequisite for NOTCH3 expression in CLL cells. ATAC-seq confirmed that gliotoxin targeted the canonical NOTCH signaling, as indicated by the loss of chromatin accessibility at the potential NOTCH/CSL site containing the gene regulatory elements. This was accompanied by a gain in accessibility at the NR4A1, NFκB, and ATF3 motifs close to the genes involved in B-cell activation, differentiation, and apoptosis. In summary, these data show that gliotoxin recovers a non-canonical tumor-suppressing NOTCH3 activity, indicating that nuclear NOTCH2 inhibitors might be beneficial compared to pan-NOTCH inhibitors in the treatment of CLL.

Chronic lymphocytic leukemia (CLL) is characterized by clonal proliferation and accumulation of malignant B lymphocytes in the blood, bone marrow, spleen, and lymph nodes. This process is associated with constitutively activated B cell receptor (BCR) signaling, and interference with BCR signaling provides therapeutic benefit. Specifically, the Bruton's Tyrosine Kinase (BTK) inhibitor ibrutinib prevents BTK tyrosine phosphorylation and thereby interferes with pathways downstream of BCR. It has shown high clinical response rates in patients with relapsed and refractory CLL, including patients with adverse cytogenetic profiles. Despite the high response rates achieved by ibrutinib, the drug has important limitations. Ibrutinib treatment induces a redistribution of CLL cells from protected niches to the peripheral blood, and the cellular response to ibrutinib is slow and often incomplete. Further, there is no evidence that a cure can be achieved, and drug discontinuation (e.g., due to toxicity) is associated with rapid progression. Even among patients that tolerate long-term treatment with ibrutinib, a considerable percentage develops drug resistance (e.g., due to mutations in the BTK gene), BTK independent disease progression, or Richter's transformation, indicating drug synergies with ibrutinib may increase prognosis. Recent studies have explored the combined use of ibrutinib with the proteasome inhibitor carfilzomib, the BCL2 inhibitor venetoclax, and the HDAC inhibitor abexinostat in preclinical models, which has shown promising initial results. However, these approaches were largely empirical, and little is known about the gene-regulatory effects of ibrutinib treatment in CLL cells. Here, we charted the ibrutinib-induced chromatin regulatory landscape of CLL, and in parallel mapped targetable pathways for synergistic combination therapies that could potentially improve disease control. Therefore, peripheral blood from 24 fully characterized CLL cases, including the clinical and hematological parameters, disease stage, cytogenetics (13q del, 11q del, 17p del, trisomy 12), p53 and IGHV mutation status were collected before and during therapy with ibrutinib. Chromatin accessibility was measured by ATAC-seq, a powerful assay for mapping the genome-wide regulatory landscape of cell. In addition, the ex vivo chemosensitivity to >140 drugs on paired CLL samples collected before and during ibrutinib treatment was measured using a novel method for ex vivo single-cell drug cytotoxicity profiling in primary samples (Snijder et al. submitted; NCT03096821). We bioinformatically analyzed the changes in chromatin accessibility and differential drug responses before and during ibrutinib treatment, thereby establishing a comprehensive picture of the cellular responses to the drug. As a proof of principle for this approach, ex vivo results confirmed strong clinical activity of BCL2 inhibitors with and without ibrutinib. We identified increases in chromatin accessibility and chemosensitivity in proteasome, inflammatory NF-κB/TNF signaling, CoA biosynthesis, PI3K/Akt, caffeine metabolism pathways, along with changes affecting genes such as FOXO3 and IκBa. Integration of chromatin accessibility and differential chemosensitivity data revealed robust ibrutinib-induced signatures, which we exploited to prioritize approved drugs and drug candidates for synergistic treatment of CLL cells in co-culture models using primary bone marrow stromal cells. Cell viability assays revealed that ibrutinib treatment in vivo and in vitro sensitizes CLL cells to compounds such as the proteasome inhibitor bortezomib, the JAK inhibitor ruxolitinib, the PLK1 inhibitor volasertib, and the bisphosphate zoledronate. In summary, our results show that the synergistic combination and bioinformatic integration of chromatin profiling with functional drug screening is a powerful tool to identify targetable pathways in CLL. This approach may be useful for designing personalized therapies as well as the rational planning of clinical studies. Our approach is directly transferable to other leukemic diseases in which malignant cells can be obtained for chromatin profiling and drug sensitivity analysis, thus providing a widely applicable tool for the rational development of combination therapies. Figure Figure. Disclosures Vladimer: Allcyte Gmbh: Equity Ownership. Snijder: Allcyte: Equity Ownership. Krall: Allcyte Gmbh: Equity Ownership. Hoermann: Gilead: Honoraria, Research Funding; Amgen: Honoraria; Novartis: Honoraria; Ariad: Honoraria. Staber: Morphosys: Membership on an entity's Board of Directors or advisory committees; Roche: Honoraria, Membership on an entity's Board of Directors or advisory committees; Abbie: Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Honoraria; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria; Gilad: Honoraria, Membership on an entity's Board of Directors or advisory committees; MSD: Honoraria. Superti-Furga: Allcyte GmbH: Equity Ownership. Jaeger: Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel, Accommodations, Expenses; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Roche: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel, Accommodations, Expenses; Novartis Pharmaceuticals Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.