CTCL Cell Lines Research Articles

Abstract 2907: Therapeutic targeting miR-146a and miR-21 induce malignant cell death and regulate CD8+ T-cell function in mycosis fungoides with large-cell transformation

Abstract Large cell transformation of mycosis fungoides (LCT-MF) occurs in 20-50% of advanced MF and is associated with an aggressive clinical course and poor survival not to overcome by any standard treatment regimen. We have previously identified a distinct miRNA expression profile in LCT-MF from that of non-transformed MF with significant upregulation of miR-21 and miR-146a. Our analyses demonstrated the involvement of genes for immune checkpoint pathways such as ICOS-ICOSL and PD1-PDL1 signaling (Di Raimondo C et al. Cancers 2021). Here, we aimed to investigate the efficacy of antagomiR-146a and -21 (amiR-146a, and -21) on the tumor growth and CD8+ tumor infiltrating lymphocyte exhaustion in LCT-MF. The amiR-146a, and -21 were synthesized in our DNA/RNA Synthesis Core by linking CpG-D19. In vitro, CTCL cell lines (Myla and HH) were treated with amiR-146a and -21, the cell viability was assessed by the 2,5-diphenyl-2H-tetrazolium bromide assay and cell apoptosis was evaluated using apoptosis assay. We found that amiR-146a, and -21 synergistically inhibited the proliferation of MyLa and HH cells due to the activation of apoptosis through Caspase3/7 pathways and inducing cell cycle arrest by blocking STAT3/CDK1/Cyclin B1 pathway. Our RNA-seq data indicated that the exhausted CD8+ T cells express elevated amounts of STAT3, IRF4, and BATF in LCT-MF compared with non-LCT MF. To evaluate the functional importance of amiR-146a, and -21 on CD8+ T cell exhaustion, we induced an exhausted state of CD8+ T cells with high level of immune checkpoints and dysfunctional cytokine production by continues anti-CD3/DC28 beads and culture supernatant (MyLa or HH cell) exposure. Our data revealed that amiR-146a and -21 attenuated the CD8+ T cell exhaustion by blockade of immune checkpoints and STAT3/IRF4/BATF pathway to trigger the cytotoxic immune response. Collectively, the findings of our study suggest that targeting miR-146a and -21 is a promising and novel therapeutic strategy for LCT-MF. Citation Format: Zhen Han, Piotr Swiderski, Xiwei Wu, Yate-Ching Yuan, Jun Wu, Chingyu Su, Hanjun Qin, Steven Rosen, Christiane Querfeld. Therapeutic targeting miR-146a and miR-21 induce malignant cell death and regulate CD8+ T-cell function in mycosis fungoides with large-cell transformation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2907.

Abstract 5720: Targeting CDK9 in cutaneous T-cell lymphoma using patient-derived xenograft models

Abstract Cutaneous T-cell lymphoma (CTCL) is a rare form of non-Hodgkin lymphoma. Mycosis fungoides and Sézary syndrome (MF/SS) are the most common subtypes in CTCL. Early-stage MF/SS has a favorable prognosis. The advanced-stage CTCL has limited treatment options and a median survival of <5 years, highlighting the urgent need to develop targeted therapies in CTCL. Preclinical studies of CDK9 inhibitors have shown anti-tumor activity in various subtypes of hematologic malignancies. Several CDK9 inhibitors have entered clinical trials for lymphomas and leukemias. However, whether targeting CDK9 in CTCL is a therapeutic option is little known and has resulted in CTCL not being included in ongoing clinical investigations of CDK9 inhibitors. AZD4573 is a highly selective and potent CDK9 inhibitor currently under clinical investigation for Hodgkin and non-Hodgkin lymphomas. It was shown that AZD4573 induces tumor cell death through depletion of MCL-1. Overexpression of MCL-1 has been shown in CTCL cell lines and skin biopsy specimens of CTCL patients. MCL-1 expression appears to be particularly elevated in patients with advanced stage disease. Here, we investigated the ex vivo anti-tumor activity of AZD4573 using patient-derived CTCL xenograft models (PDXs). Primary tumor cells harvested from CTCL PDX mice were activated with CD2/CD3/CD28 beads and incubated in hIL-2 and hIL-7 at an optimal cell density. Subsequently, cells were exposed to AZD4573 at various concentrations for 8 hr, then treated cells were harvested, washed, and placed in fresh media for another 48 hr. Growth inhibition was measured by the Real-Time Glo cell viability assay, and apoptotic activity was assessed by flow cytometry analysis of Caspase 3/7 activity and Annexin V + propidium iodide (PI) double staining. Additionally, we conducted RNA-seq analysis for MCL-1 expression in CTCL PDX primary tumor cells. AZD4573 inhibited cell proliferation in a dose-dependent manner in primary tumor cells from three CTCL PDX mice (PRS-1, PRS-2 PRS-3), with GI50 of 43.5 nM, 175.4 nM, and 113.5 nM, respectively. Furthermore, AZD4573 significantly increased caspase 3/7 activity and apoptosis in PRS-1 and PRS-3, but not PRS-2 primary tumor cells. In addition, our RNA-seq analysis showed that compared with normal controls (normal PBMC from healthy donors), the MCL-1 gene expression is increased 10-fold (P-value <0.0001) in PRS1 cells. The MCL-1 gene expression in PRS2 and PRS3 cells is the same as normal control. Furthermore, PRS-1, which has the highest MCL-1 expression level, is most sensitive to AZD4573. In conclusion, AZD4573 demonstrated dose-dependent growth inhibition and apoptosis induction in primary CTCL tumor cells. In addition, the expression level of MCL-1 appeared to be correlated with the ex vivo anti-tumor activity of AZD4573. Our findings provide a rationale for clinical investigation of targeting CDK9 in CTCL. Citation Format: Yan-Jin Liu, Chi-Heng Wu, Laura Pincus, Weiyun Z. Ai. Targeting CDK9 in cutaneous T-cell lymphoma using patient-derived xenograft models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5720.

GATA-3 Dependent Gene Transcription Is Impaired upon HDAC Inhibition

Background: GATA-3 is a lymphoid-lineage specific zinc-finger transcription factor that plays a pivotal role in the development and homeostatic survival of thymic T-cell progenitors and subsets of their mature progeny. Diverse T-cell lymphomas (TCL), derived from conventional (non-malignant) T cells expressing GATA-3, highly express GATA-3 and a GATA-3 dependent transcriptional program, and are generally associated with poor outcomes. Upon binding gene enhancers and promoters, GATA-3 regulates gene transcription, including those with cell-autonomous (e.g. oncogenes, including c-Myc) and non-cell-autonomous (e.g. type 2 cytokines) functions in TCL. Therefore, GATA-3, and the transcriptional program it instigates, is a bona-fide proto-oncogene in these lymphomas. GATA-3 acetylation is dynamically regulated, and its acetylation controls its DNA binding affinity and target gene expression. Therefore, we sought to examine the extent to which HDAC inhibition may alter GATA-3 acetylation and functionally regulate its DNA binding capacity and target gene expression in CTCL. Methods: Integrated GATA-3 ChIP-seq and RNA-seq was performed in CTCL cell lines and primary specimens treated with clinically available HDACi. Complementary strategies were utilized to examine GATA-3/HDAC binding and GATA-3 acetylation in these cells. Results: In order to investigate early transcriptional changes upon HDACi treatment, a well characterized CTCL cell line (H9) and primary Sezary Syndrome (SS) specimens (n=3) were transcriptionally profiled. Gene set enrichment analysis (GSEA) was performed using the differentially expressed genes we identified, and a convergence on relevant T-cell growth and survival pathways was observed. GATA-3 target genes were differentially expressed upon HDACi treatment, and binding sites for GATA-3, and transcription factors (TF) it transcriptionally regulates, were enriched among HDACi responsive genes. Among the 5,510-6,811 transcripts that were differentially expressed upon either belinostat or romidepsin treatment, respectively, ≈6-7% of transcripts were GATA-3 targets. GATA-3 dependent transcripts were significantly altered upon treatment with belinostat or romidepsin in primary SS specimens. We next examined the association between GATA-3 and class I HDACs in CTCL cell lines. HDAC1/HDAC2 and GATA-3 binding was observed, suggesting that HDAC inhibition may lead to GATA-3 hyperacetylation. Treatment with the three clinically available HDACi led to a significant increase in GATA-3 acetylation. As GATA-3 acetylation regulates its ability to bind DNA and regulate transcription, we examined GATA-3 DNA binding using orthogonal approaches, including GATA-3 ChIP-seq, and a significant reduction in GATA-3 binding peaks was observed, culminating in significant changes in the GATA-3 dependent transcriptome. Conclusions: Collectively, these findings demonstrate that GATA-3 hyperacetylation upon HDAC inhibition significantly impairs GATA-3 DNA binding and GATA-3 dependent gene transcription. Impact: GATA-3 and its target genes identify clinically and transcriptionally distinct T-cell lymphomas, including CTCL, that are resistant to conventional chemotherapeutic agents and generally associated with dismal outcomes. Therefore, GATA-3 is a rational, albeit challenging, therapeutic target in these T-cell lymphomas. We have previously explored “upstream” strategies to target GATA-3 by exploiting transcription factors that regulate its expression, and have more recently explored “downstream” strategies by targeting its target genes, including ITK. To the best of our knowledge, the work described here is the first - and a clinically approved - strategy to therapeutically target GATA-3 directly, as HDACi significantly increase GATA-3 acetylation, and by doing so, impair its ability to bind DNA and regulate transcription of its gene targets. Consequently, these findings have significant implications for the interpretation of existing clinical trial datasets and for the development of future combinatorial strategies utilizing an HDACi-containing backbone in GATA-3 driven T-cell lymphomas.

First-in-Class HAT Activator (YF2) Combined with JAK/STAT Inhibitor (ruxolitinib) Unveils Potential Novel Treatment Approach for HDAC Inhibitor-Resistant CTCL

T-cell lymphomas are both aggressive and challenging to treat. This is in-part due to multiple epigenetic derangements found in T-cell lymphomas. One treatment strategy includes HDAC inhibitors; however, the duration of response is short for most patients which is consistent with many treatments for T-cell lymphoma. As a result, alternative treatments are required. YF2, a first-in-class histone acetyltransferase (HAT) activator, was invented at Columbia University. YF2 increases acetylation of histones by augmenting the activity of HAT enzymes. The JAK/STAT pathway contributes to CTCL lymphomagenesis through a number of cellular processes including cell cycle, proliferation, apoptosis, etc. JAKs activate STAT3, which activates the transcription of the anti-apoptotic Bcl-2-like protein, Bcl-XL. Additionally, HATs acetylate the pro-apoptotic p53 protein, increasing its stability and activity. Given the dependence on both the JAK/STAT pathway and epigenetic derangements in CTCL, we hypothesize that if the JAK inhibitor, ruxolitinib, induces apoptosis via modulation of Bcl-XL, and YF2 induces apoptosis via acetylation and activation of p53, then dual targeting of these complementary pathways could lead to synergistic apoptosis in CTCL. To study this hypothesis, we used two CTCL cell lines: HH [JAK wildtype (JAK WT)], and H9 [JAK1/3 gain-of-function mutated (JAK1/3 GOF)]. In addition, we generated a belinostat-resistant H9 cell line (H9-belino-R) by incrementally exposing H9 to increasing concentrations of belinostat. H9-belino-R resistance was observed at a 20-fold (IC 50 = 896.7nM, SEM ± 185.3) increase over that of the parental H9 (IC 50 = 43.8nM, SEM ± 4.61) ( figure 1). H9-belino-R retained significant resistance to other HDAC inhibitors such as romidepsin [(H9: IC 50 = 0.97nM (SEM ± 0.030), H9-belino-R: IC 50 = 1.38nM (SEM ± 0.028)] and panobinostat [H9: IC 50 = 3.11nM (SEM ± 0.19), H9-belino-R: IC 50 = 9.00nM (SEM ± 1.55)] as measured by the CellTiter-Glo Viability Assay. Interestingly, the IC 50 of the HAT activator, YF2, was retained between the resistant (14.84µM, SEM ± 0.46) and parental (12.45µM, SEM ± 0.27) cell lines ( figure 1). HH (JAK WT) showed significant cytotoxicity (63.5% live, 30.7% apoptotic, and 5.4% dead) after 4-day exposure to 10µM YF2 compared to limited cytotoxicity (93.3% live, 4.6% apoptotic, and 2.0% dead) in H9 (JAK GOF) as measured by flow cytometry with Annexin V. At baseline, active STAT3 (pSTAT3) was not detected in HH possibly contributing to its susceptibility to YF2 as the absence of pSTAT3 reduces the transcription Bcl-XL. As illustrated by figure 2, western blot analysis confirmed ruxolitinib-induced inhibition of active STAT3 (pSTAT3) with associated reduction in Bcl-XL expression in H9 and H9-belino-R. This may explain the observed synergistic cytotoxicity with YF2 and ruxolitinib in HH, H9 and H9-belino-R (Excess over Bliss scores: 4.85, 14.23, and 5.38 respectively) as determined by flow cytometry. In conclusion, the novel first-in-class HAT activator, YF2, retains activity in HDAC inhibitor-resistant CTCL. YF2's activity may be blunted by JAK1/3 GOF mutation; however, when combined with the JAK-inhibitor ruxolitinib, the response mirrored that of the HH (JAK WT) cell line. The combination of HAT activator and JAK inhibitor led to enhanced reduction of the anti-apoptotic protein Bcl-XL, potentially explaining synergistic apoptosis in CTCL. These preliminary findings provide us with evidence that suggests that the combination of YF2 and ruxolitinib can serve as a novel treatment combination for CTCL. Further study is planned to explore this treatment in murine models of disease.

In vitro evaluation of Neosetophomone B inducing apoptosis in cutaneous T cell lymphoma by targeting the FOXM1 signaling pathway

BackgroundCutaneous T cell lymphoma (CTCL) is a T cell-derived non-Hodgkin lymphoma primarily affecting the skin, with treatment posing a significant challenge and low survival rates. ObjectiveIn this study, we investigated the anti-cancer potential of Neosetophomone B (NSP-B), a fungal-derived secondary metabolite, on CTCL cell lines H9 and HH. MethodsCell viability was measured using Cell counting Kit-8 (CCK8) assays. Apoptosis was measured by annexin V/PI dual staining. Immunoblotting was performed to examine the expression of proteins. Applied Biosystems' high-resolution Human Transcriptome Array 2.0 was used to examine gene expression. ResultsNSP-B induced apoptosis in CTCL cells by activating mitochondrial signaling pathways and caspases. We observed downregulated expression of BUB1B, Aurora Kinases A and B, cyclin-dependent kinases (CDKs) 4 and 6, and polo-like kinase 1 (PLK1) in NSP-B treated cells, which was further corroborated by Western blot analysis. Notably, higher expression levels of these genes showed reduced overall and progression-free survival in the CTCL patient cohort. FOXM1 and BUB1B expression exhibited a dose-dependent reduction in NSP-B-treated CTCL cells.FOXM1 silencing decreased cell viability and increased apoptosis via BUB1B downregulation. Moreover, NSP-B suppressed FOXM1-regulated genes, such as Aurora Kinases A and B, CDKs 4 and 6, and PLK1. The combined treatment of Bortezomib and NSP-B showed greater efficacy in reducing CTCL cell viability and promoting apoptosis compared to either treatment alone. ConclusionOur findings suggest that targeting the FOXM1 pathway may provide a promising therapeutic strategy for CTCL management, with NSP-B offering significant potential as a novel treatment option.

HZ-R061, a Selective STAT3 Degrader with Activity in Preclinical Model of Hematologic Malignances

The Janus kinase/Signal transducer and activator of transcription (JAK-STAT) pathway is one of the core cancer pathways that integrates signals from cytokines, hormones and growth factors to induce or repress gene expression in cells . The pathway consists of four JAK kinases and seven STAT transcription factors (STAT1, STAT2, STAT3, STAT4, STAT5A, STAT5B and STAT6). STAT3 is the best-studied family member of the JAK-STAT pathway in cancer, and is a known oncogene in various types of tumor. So STAT3 is an attractive cancer therapeutic target, great efforts have been made to discover effective STAT3 inhibitor. However most of reported STAT3 inhibitors got failed for further development due to the low selectivity, weak binding affinity or poor drugability. STAT3 is considered as an undruggable target for long time. Recently, the proteolysis targeting chimeras (PROTAC) strategy has gained momentum for its promise in the discovery and development of an entirely new type of therapeutics for the treatment of human diseases. PROTAC molecule is a bifunctional small molecule designed to recruit cellular degradation machinery to induce targeted protein degradation in a catalytical way. PROTAC can be a new strategy to effectively target undruggable targets such as STAT3. HZ-R061 is a novel STAT3 selective degrader which is discovered and developed through our DaTProD® platform which was set up to promote PROTAC development. HZ-R061 can selective and effective degrade STAT3 in various types of cells including MOLM16, A549, SU-DHL1, karpas299, LY3, THP1 and human PBMC with DC50 values at nanomolar level. In in-vitro anti-proliferative assay, HZ-R061 exhibits potent anti-proliferative activity against CTCL and AML cell lines. Furthermore, MOLM16, SU-DHL1 and Karpas299 xenograft models were used for in-vivo anti-tumor evaluation. After IV administration of HZ-R061 (QW), the growth of tumor was significantly suppressed in all three models. As STAT3 is a master regulator in tumor microenvironment, inhibition of STAT3 could remodel TME towards in favor of anti-cancer. The combination of HZ-R061 with PD-(L)1 mAb was also conducted in CT26 model. HZ-R061 exhibits significantly synergistic effect in combination with PD-(L)1 mAb. Additionally, HZ-R061 also potently decreases phosphorylation level of STAT3 in IL-6 stimulated THP-1 cells with IC50 value of 3.8nM, which is 50 folds stronger than JAK inhibitor Ruxolitinib. HZ-R061 also shows strong effect in in-vitro Th17 differentiation assay and EAE model. These results suggest that HZ-R061 is also a promising drug candidate for various Th17 driven auto-immune diseases including MS, RA, SSc and IPF. In conclusion, we discovered a pre-clinical effective STAT3 degrader HZ-R061, which has potential for further development in tumor and auto-immune diseases.

Combination of Resminostat with Ruxolitinib Exerts Antitumor Effects in the Chick Embryo Chorioallantoic Membrane Model for Cutaneous T Cell Lymphoma.

Simple SummaryThe combination of Resminostat (HDACi) and Ruxolitinib (JAKi) exerted cytotoxic effects and inhibited proliferation of CTCL cell lines (MyLa, SeAx) in vitro. The aim of the present study was to validate their antitumor effects in vivo using the chick embryo chorioallantoic membrane (CAM) model, which allows quick and efficient monitoring of tumor growth, migration, invasion, and metastatic potential. The drug combination exhibited a significant inhibition of primary tumor size, and inhibited intravasation and extravasation of tumor cells to the liver and lung. It also exerted an inhibitory effect in the migration and invasion of tumor cells and significantly reduced key signaling pathway activation. Our data demonstrate that the CAM assay could be employed as a preclinical in vivo model in CTCL for pharmacological testing, and that the combination of Resminostat and Ruxolitinib exerts significant antitumor effects in CTCL progression that need to be further evaluated in a clinical setting.The combination of Resminostat (HDACi) and Ruxolitinib (JAKi) exerted cytotoxic effects and inhibited proliferation of CTCL cell lines (MyLa, SeAx) in previously published work. A xenograft tumor formation was produced by implanting the MyLa or SeAx cells on top of the chick embryo chorioallantoic membrane (CAM). The CAM assay protocol was developed to monitor the metastatic properties of CTCL cells and the effects of Resminostat and/or Ruxolitinib in vivo. In the spontaneous CAM assays, Resminostat and Ruxolitinib treatment inhibited the cell proliferation (p < 0.001) of MyLa and SeAx, and induced cell apoptosis (p < 0.005, p < 0.001, respectively). Although monotherapies reduced the size of primary tumors in the metastasis CAM assay, the drug combination exhibited a significant inhibition of primary tumor size (p < 0.0001). Furthermore, the combined treatment inhibited the intravasation of MyLa (p < 0.005) and SeAx cells (p < 0.0001) in the organs, as well as their extravasation to the liver (p < 0.0001) and lung (p < 0.0001). The drug combination also exerted a stronger inhibitory effect in migration (p < 0.0001) rather in invasion (p < 0.005) of both MyLa and SeAx cells. It further reduced p-p38, p-ERK, p-AKT, and p-STAT in MyLa cells, while it decreased p-ERK and p-STAT in SeAx cells in CAM tumors. Our data demonstrated that the CAM assay could be employed as a preclinical in vivo model in CTCL for pharmacological testing. In agreement with previous in vitro data, the combination of Resminostat and Ruxolitinib was shown to exert antitumor effects in CTCL in vivo.

Targeting CD70 in cutaneous T-cell lymphoma using an antibody-drug conjugate in patient-derived xenograft models

AbstractCD70 is a member of the tumor necrosis factor receptor superfamily. Emerging data indicate that CD70 may be a suitable target for various malignancies. We investigated the expression of CD70 in cutaneous and systemic T-cell lymphomas and conducted preclinical studies of SGN-CD70A, a CD70-directed antibody-drug conjugate (ADC), using patient-derived xenograft cutaneous T-cell lymphoma (CTCL PDX) models. CD70 expression was examined by immunohistochemical (IHC) stains in 49 diagnostic specimens of T-cell lymphomas. The activities of SGN-CD70A in growth inhibition and apoptosis induction were examined in CTCL cell lines and primary CTCL tumor cells. Using previously established CTCL PDXs, we conducted a dose-finding trial followed by a phase 2-like trial to evaluate the optimal dosing and the efficacy of SGN-CD70A in tumor-bearing PDX animals. The therapeutic efficacy of SGN-CD70A was measured by tumor-associated cell-free DNA (cfDNA) and survival of treated PDXs. We found that CD70 is highly expressed in T-cell lymphomas, especially in CTCL. SGN-CD70A inhibited cell growth and induced apoptosis in CD70-expressing CTCL cell lines and primary tumors cells. Additionally, SGN-CD70A at 100 μg/kg and 300 μg/kg prolonged the survival of PDXs in a dose-dependent manner. Finally, treatment with 3 doses of SGN-CD70A at 300 μg/kg was superior to a single-dose treatment in survival prolongation (median survival: 111 days vs 39 days; P = .017). Most importantly, multiple dosing of SGN-CD70A induced complete eradication of established tumors in PDXs measured by cfDNA. Our results demonstrated marked antitumor activity of SGN-CD70A in CTCL PDXs, providing compelling support for its clinical investigation.

Chimerized Anti-ICOS 314.8 Monoclonal Antibodies Inhibit Tumor Cells and Regulatory T Cells in Patients with Sézary Syndrome

Background: In a previous study, we reported the strong expression of Inducible T-cell costimulatory (ICOS) by Sézary cells, and presented the excellent preclinical efficacy results of anti-ICOS antibody drug conjugates (ADCs) in both Sézary syndrome (SS) and angioimmunoblastic T-cell lymphoma. Although exerting a potent direct action on the tumor cell, ADCs have the disadvantage of being associated with a cumulative toxicity, related to the conjugated drug. The development of antibody-dependent cell-mediated cytotoxicity/phagocytosis (ADCC/ADCP)-inducing anti-ICOS monoclonal antibodies (mAbs) is therefore of great importance to ensure long-term maintenance therapy.Methods: We first determined which anti-ICOS clone was the best candidate to induce an ADCC effect on ICOS + cell lines (MyLa, MJ and HUT78), using Mouse FcγRIII ADCC Bioassay (Promega®). The selected mAb was then chimerized and afucosylated (GlymaxX® technology, Evitria®).Secondly, we evaluated in vitro ADCC and ADCP effect of the chimeric anti-ICOS mAb against ICOS + CTCL cell lines, compared to both positive controls (mogamulizumab (moga) and alemtuzumab) and negative controls (IgG1 isotype control, and rituximab). To perform ADCC experiments, we co-incubated target cells with mAbs and allogenic NK lymphocytes from healthy volunteers (HV). Cellular apoptosis was measured by flow cytometry using the Caspase 3/7 assay (Promega®). For ADCP, monocytes were sorted from HV blood samples and treated with M-CSF. Target cells were labeled with PKH67 (Sigma-Aldrich®) and after co-incubation, CD14 +CD11b +PKH67 + monocytes were analyzed by flow cytometry.Finally, we verified the ADCC/ADCP potency of anti-ICOS mAbs on primary Sézary cells isolated from 16 moga-naïve SS patients, and 6 patients who had developed resistance to moga. We also questioned whether anti-ICOS mAbs were able to promote the autologous apoptosis of Sézary cells and T regulatory cells (Tregs) when directly incubated with peripheral blood mononuclear cells (PBMCs) from patients with SS.Results: Among 7 different anti-ICOS clones, 314.8, 92.17 and 293.1 clones had the higher ADCC activity against MyLa, MJ and HUT78. Of these 3 clones, 314.8 had the best affinity to the receptor, and the best ability to inhibit binding between ICOS receptor and a recombinant ICOS ligand protein. Anti-ICOS 314.8 mAb was then chosen to be chimerized and glyco-engineered.ICOS and CCR4 were strong and similarly expressed on MyLa and MJ. HUT78 had only mild expression of ICOS and CD52. Anti-ICOS mediated potent ADCC of cell lines (respectively 39.1±5% and 52.6±2.4% for MyLa and MJ cells), without significant difference when compared to mogamulizumab. In HUT78 cells, anti-ICOS induced a specific apoptosis of 35.7±5% versus 15.6±5.6% with alemtuzumab (p=0.02; CI95%: 4.1-36.1).Moreover, phagocytosis induced by anti-ICOS was significantly increased than that induced by negative controls. On MyLa cells, anti-ICOS had a greater phagocytosis activity than moga (59.4±5.2% vs 39.4±5.1%, p=0.031).Expression of ICOS by circulating tumor cells was found in all the 16 moga-naïve patients. The expression was strong, as 61±6% of tumor cells expressed ICOS vs 20±8% of non-tumoral CD4 + cells. CCR4 was more expressed than ICOS on both Sézary cells and non-tumoral CD4 + cells (91±6%, and 44±9% respectively). Anti-ICOS induced the apoptosis of 57.1±4.7% Sézary cells, compared to 16.9±2.2% with rituximab (p<0.01; CI95%:-53–27). The efficacy was better than with alemtuzumab, but there was no significant difference with moga. The ADCP effect induced by anti-ICOS did not differ with moga. Interestingly, anti-ICOS were effective in 6 moga-resistant patient, as they induced 38.9±5.9% of apoptosis, compared to 12.4±4.7% with moga (p<0.001).Ex vivo, anti-ICOS allowed 39.4±19.9% and 70.1±20.1% lysis of Sézary cells and Tregs respectively, with no difference with moga and alemtuzumab. However, the depletion of non-tumoral CD4 + and total PBMCs was significantly lower with anti-ICOS mAbs than with moga and alemtuzumab.Discussion: In a recent study, we showed that Treg cells of patients with SS have a high expression of ICOS. Here, we demonstrate that anti-ICOS mAbs induce Tregs depletion, which may improve immune profiles and emphasize Sézary cells killing. Our data suggest robust anti-tumor activity of anti-ICOS mAbs in SS, and xenograft experiments are underway to confirm these findings. DisclosuresLopez: Emergence Therapeutics: Current holder of individual stocks in a privately-held company. Bagot: Takeda: Membership on an entity's Board of Directors or advisory committees. Olive: Alderaan Biotechnology: Current holder of individual stocks in a privately-held company, Membership on an entity's Board of Directors or advisory committees; ImCheck Therapeutics: Current holder of individual stocks in a privately-held company, Membership on an entity's Board of Directors or advisory committees; Emergence Therapeutics: Current holder of individual stocks in a privately-held company, Membership on an entity's Board of Directors or advisory committees.

Abstract 1844: A CD70 antibody-drug conjugate is highly active and induces long term remission in patient-derived xenograft mouse models of cutaneous T cell lymphoma

Abstract Cutaneous T cell lymphoma (CTCL) is a type of non-Hodgkin's lymphoma in which malignant T cells initially home to the skin followed by systemic dissemination as the disease progresses. CD70 is rarely expressed in normal tissues but is overexpressed in activated T and B lymphocytes. From archival diagnostic specimens of 37 patients, we demonstrate that CD70 is highly expressed in subtypes of mature TCL, especially CTCL. SGNCD70A is a new anti-CD70 antibody-drug conjugate combining a monoclonal antibody to CD70 with a DNA-damaging agent. SGNCD70A inhibited cell growth and induced apoptosis of CD70+ CTCL cell lines and primary tumors from patient-derived xenograft (PDX) mice, but had no inhibitory effect on CD70- lymphoblastic TCL cells and normal T cells. Next, we examined the anti-tumor activity of SGNCD70A in CTCL PDX model. One week after tumor inoculation, mice were treated intraperitoneally with single-dosing of PBS, 100 μg/kg (SGN70A100) or 300 μg/kg (SGN70A300) of SGN-CD70A. SGN-CD70A prolonged survival of treated mice in a dose-dependent fashion compared to untreated mice. Median survival was 21 days, 27.5 days (P=0.041), and 57 days (p&amp;lt;0.01) for the PBS control, SGN70A100 and SGN70A300 group, respectively. To confirm that survival was dictated by disease progression, rather than drug toxicity of SGN-CD70A, we measured tumor burden by cell-free DNA (cfDNA) throughout the course of treatment. We demonstrated that cfDNA paralleled the disease progression and peaked before death of the animals. In contrast, in surviving mice, cfDNA remained at the background level. Importantly, we discovered relapse occurred in SGN70A300 group (3 of 4, 75%) due to persistent disease, not due to acquired resistance to SGN-CD70A. Thus, we investigated whether multiple dosing could lead to complete regression of tumors in PDX model. We treated PDX mice with SGN70A300 weekly (SGN70A300x3) for 3 consecutive weeks. In the SGN70A300x3 group, three of four mice survived beyond day 90 (median &amp;gt;85 days), resulting in markedly prolonged survival compared to the single-dose- and PBS-treated mice with a median survival of 39 days, (P&amp;lt;0.01) and 20 days (P&amp;lt;0.01), respectively. Compared to PBS group, both SGN70A300 and SGN70A300x3 treated mice had delayed tumor growth and cfDNA rise. Furthermore, in SGN70A300 group, cfDNA started to rise seven days before the endpoint of the mice in a range of 30-42 days, while three surviving mice in SGN70A300x3 remained at background cfDNA levels until day 85, indicating that multiple doses of SGN-CD70A decreased overall tumor burden and the occurrence of relapse in CTCL. Our results showed that CD70 is highly expressed in mature TCL, especially in CTCL. Using CTCL PDX models, SGN-CD70A has marked anti-tumor activity, leading to long-term survival of treated mice. Our results provide a rationale for clinical investigation of SGN-CD70A in patients with TCL. Citation Format: Chi-Heng Wu, Chen-Yen Yang, Linlin Wang, Ryan Gill, Frank McCormick, Weiyun Z. Ai. A CD70 antibody-drug conjugate is highly active and induces long term remission in patient-derived xenograft mouse models of cutaneous T cell lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1844.

Edo-S101, a Bendamustine (BDM)/ Histone-Deacetylase Inhibitor (HDACi) Fusion Molecule, Demonstrates Potent Preclinical Activity Against T-Cell Malignancies and Overcomes BDM-Resistance

Introduction: T-cell and natural killer (NK)-cell lymphomas encompass a heterogeneous spectrum of malignancies with variable degrees of biological and clinical aggressiveness and different patterns of nodal/extra nodal dissemination. Despite advances in understanding of their genetic landscape, treatment options for these tumors are limited and survival remains dismal especially in patients (pts.) who relapse/progress after frontline treatment. Therefore, development of new active agents is a pressing medical need in this setting. The unconventional alkylator BDM and HDACi are clinically effective in pts. with recurring disease and the combination of these agents was suggested as a way to overcome early resistance and improve response and survival rates in T-cell tumors. Since EDO-S101 is a fusion molecule that combines the active structures of BDM and the pan-HDACi vorinostat to simultaneously damage DNA and block damage repair, we investigated its antitumor activity in preclinical models of human T- and NK- lymphomas.Methods: We assessed the patterns of EDO-S101 cytotoxicity and its regulatory effects on genes involved in DNA-damage/repair, apoptosis and drug resistance on a large panel of cell lines representative of most T-lymphoma subtypes, including precursor T-cell lymphoma/leukemia (pTCL; MOLT-4, KE-37, Jurkat, CCRF-CEM), anaplastic large T-cell lymphoma (ALCL; SR-786, SUDHL1) and cutaneous T-cell lymphoma (CTCL; HUT78, MYLA, HH, PB2B, MAC-2A). As a further model, we exploited an ALCL cell line (SUDHL1-R100) selected for resistance to BDM. SUDHL1-R100 display a growth pattern indistinguishable from parental cells in the presence of BDM (100 μmol/L).Results: EDO-S101 induced a significant time- and dose-dependent inhibition of growth and survival in all T-cell lines, with a IC50 at 48 hrs (0.33 to 0.68 μmol/L), that was at least 90% and 50% lower than IC50 of BDM (10.5 to 23.67 μmol/L) and vorinostat (0.42 to 1.17 μmol/L) in the same cell lines. No significant differences emerged across various cell lines as to sensitivity to EDO-S101. Notably, growth inhibitory effects of EDO-S101 were 35% to 40% higher than those achieved by the concomitant exposure of tumor cells to BDM and vorinostat at equimolar concentrations. A significant induction of the DNA double strand breaks marker pγH2AX, along with decreased levels of Thr-1989 pATR and Polo-like kinase 1 (PLK1) were consistently observed in pTCL, ALCL and CTCL cell lines, indicating that EDO-S101 triggered inappropriate mitotic entry with unrepaired DNA damage. In addition, qRT-PCR results showed upregulation of the proapototic genes NOXA, p21 and/or p27 in pTCL, CTCL and ALCL cells. Among DNA repair genes, EDO-S101-induced downregulation of ATM, ATR and EXO1 was paralleled by an increase of MGMT transcripts. Most intriguingly, BDM-resistant ALCL cells (SUDHL1-R100) were highly sensitive to EDO-S101, with an IC50 of only 0.72 μmol/L, but less responsive to vorinostat (IC50: 2.17 μmol/L) than parental cells (IC50: 1.46 μmol/L). Exposure to EDO-S101 also resulted in transcriptional regulation of components of the PI3K pathway, mainly the PI3Kα and δ isoforms and mTOR. These findings, prompted experiments showing that EDO-S101 at sublethal concentrations was highly synergic with the predominant PI3Kα/δ inhibitor copanlisib in inducing cell death of tumor T- and NK-cells. Finally, exposure to EDO-S101 (1.0 μmol/L) for 48 hrs resulted in a sustained upregulation of surface PD1 and PD-L1 in most tumor T- and NK-cell lines, as a possible result of its HDACi component.Conclusions: We have shown for the first time that the BDM/vorinostat fusion molecule EDO-S101 is very active in preclinical models of T/NK cell lymphomas. It shows a superior activity towards different types of tumor T- and NK-cells than its single components administered either individually or in combination by also overcoming acquired resistance to BDM. Data indicate that EDO-S101 simultaneously damages DNA and interferes on its repair pathways in tumor T-cells. We also showed synergy between EDO-S101 and PI3Kα/δ inhibitors possibly due to regulatory effects on specific components of the PI3K pathway as well as a potential for this agent to enhance PD1-blockade strategies. A clinical trial is ongoing to evaluate efficacy and safety of EDO-S101 in pts. with relapsed/refractory hemopoietic malignancies including T-cell lymphomas. DisclosuresMehrling:Mundipharma-EDO Gmbh, Basel, CH: Employment. Pinto:Celgene: Honoraria; Merck Sharp Dome: Honoraria; Millenium Takeda: Research Funding; Roche: Honoraria; Bristol Myers Squibb: Honoraria; Gilead: Honoraria; Helssin: Honoraria; Mundipharma EDO: Speakers Bureau.

Ruxolitinib with resminostat exert synergistic antitumor effects in Cutaneous T-cell Lymphoma.

The combination of JAK/STAT and HDAC inhibitors exerted beneficial effects in haematological malignancies, presenting promising therapeutic CTCL targets. We aim to investigate the efficacy of JAK1/2i ruxolitinib in combination with HDACi resminostat in CTCL in vitro. Non-toxic concentrations of ruxolitinib and/or resminostat were administered to MyLa (MF) and SeAx (SS) cells for 24h. Cytotoxicity, cell proliferation and apoptosis were estimated through MTT, BrdU/7AAD and Annexin V/PI assay. Multi-pathway analysis was performed to investigate the effect of JAK1/2i and/or HDACi on JAK/STAT, Akt/mTOR and MAPK signalling pathways. Both drugs and their combination were cytotoxic in MyLa (p<0.05) and in SeAx cell line (p<0.001), inhibited proliferation of MyLa (p<0.001) and SeAx (p<0.001) at 24h, compared to untreated cells. Moreover, combined drug treatment induced apoptosis after 24h (p<0.001) in MyLa, and SeAx (p<0.001). The combination of drugs had a strong synergistic effect with a CI<1. Importantly, the drugs' combination inhibited phosphorylation of STAT3 (p<0.001), Akt (p<0.05), ERK1/2 (p<0.001) and JNK (p<0.001) in MyLa, while it reduced activation of Akt (p<0.05) and JNK (p<0.001) in SeAx. The JAKi/HDACi combination exhibited substantial anti-tumor effects in CTCL cell lines, and may represent a promising novel therapeutic modality for CTCL patients.

JAK inhibition synergistically potentiates BCL2, BET, HDAC, and proteasome inhibition in advanced CTCL

AbstractCutaneous T-cell lymphoma (CTCL) is a malignancy of skin-homing T lymphocytes that is more likely to involve the peripheral blood in advanced stages. For such patients with advanced disease, there are few available systemic treatment options, and prognosis remains poor. Exome sequencing studies of CTCL have suggested therapeutic targets, including within the JAK/STAT pathway, but JAK inhibition strategies may be limited by patient-specific mutational status. Because our recent research has highlighted the potential roles of single and combination approaches specifically using BCL2, bromodomain and extra-terminal domain (BET), and histone deacetylase (HDAC) inhibition, we aimed to investigate the effects of JAK inhibition on CTCL cells and established CTCL cell lines when paired with these and other targeting agents. Peripheral blood malignant CTCL isolates exhibited differential responses to JAK inhibition, with JAK2 expression levels negatively correlating to 50% inhibitory concentration (IC50) values. Regardless of single-agent sensitivity, JAK inhibition potentiated malignant cell cytotoxicity in combination with BCL2, BET, HDAC, or proteasome inhibition. Combination inhibition of JAK and BCL2 showed the strongest potentiation of CTCL cytotoxicity, driven by both intrinsic and extrinsic apoptosis pathways. JAK inhibition decreased expression of BCL2 in the high-responder samples, suggesting a putative mechanism for this combination activity. These results indicate that JAK inhibition may have major effects on CTCL cells, and that combination strategies using JAK inhibition may allow for more generalized cytotoxic effects against the malignant cells from patients with CTCL. Such preclinical assessments help inform prioritization for combination targeted drug approaches for clinical utilization in the treatment of CTCL.

Synthesis and Investigation of Therapeutic Potential of Isoform-Specific HDAC8 Inhibitors for the Treatment of Cutaneous T Cell Lymphoma.

The available treatment option for any type of cancer including CTCL is chemotherapy and radiation therapy which indiscriminately persuade on the normal cells. One way out for selective destruction of CTCL cells without damaging normal cells is the use of histone deacetylase inhibitors (HDACi). Despite promising results in the treatment of CTCL, these HDACi have shown a broadband inhibition profile, moderately selective for one HDAC class but not for a particular isotype. The prevalence of drug-induced side effects leaves open a narrow window of speculation that the decreased therapeutic efficacy and observed side effects may be most likely due to non specific HDAC isoform inhibition. The aim of this paper is to synthesis and evaluates HDAC8 isoform specific inhibitors. Based on the preliminary report on the design and in silico studies of 52 hydroxamic acid derivatives bearing multi-substituent heteroaromatic rings with chiral amine linker, five compounds were shortlisted and synthesized by microwave assisted approach and high yielding synthetic protocol. A series of in vitro assays in addition to HDAC8 inhibitory activity was used to evaluate the synthesised compounds. Inhibitors 1e, 2e, 3e, 4e and 5e exerted the anti-proliferative activities against CTCL cell lines at 20- 100 µM concentrations. Both the pyrimidine- and pyridine-based probes exhibited μM inhibitory activity against HDAC8. The pyrimidine-based probe 1e displayed remarkable HDAC8 selectivity superior to that of the standard drug, SAHA with an IC50 at 0.1µM. Our study demonstrated that simple modifications at different portions of pharmacophore in the hydroxamic acid analogues are effective for improving both HDAC8 inhibitory activity and isoform selectivity. Potent and highly isoform-selective HDAC8 inhibitors were identified. These findings would be expedient for further development of HDAC8-selective inhibitors.

Abstract LB-115: Preclinical assessment of retinoid X-receptor agonist, UAB110, for potential treatment of cutaneous T-cell lymphoma

Abstract Cutaneous T-cell lymphomas (CTCLs) are a group of non-Hodgkin diseases characterized by localization of neoplastic T-cells in the skin. The most common forms of CTCLs are mycosis fungoides (MF) and Sézary syndrome (SS). Bexarotene (Targretin®) is currently the only FDA approved retinoid X receptor (RXR) -selective agonist (rexinoid) for the treatment of CTCLs. Despite its low toxicity compared to retinoids and other chemotherapeutic agents, it induces hypothyroidism and markedly elevates serum triglycerides (TGs). Thus, improved rexinoids for the treatment of this indolent disease is necessary. A potential candidate is the novel rexinoid, UAB110, which does not elevate TGs in animal studies and effectively prevents mammary cancer development. Compared to bexarotene, it has higher binding affinity for RXRα receptor and transcriptional activation activity. The purpose of this study is to assess preclinical efficacy, mechanism of action of UAB110, and compare its action with bexarotene and the rexinoid 9-cisUAB30 (UAB30) using human-derived CTCL cell lines, MyLa and HuT78, which are representative of MF and SS, respectively. Cell proliferation analysis and cell-cycle related markers were determined by hemocytometer and Western blot analysis, respectively. UAB110 is as effective as UAB30 and bexarotene in suppressing cell proliferation. Furthermore, it significantly (p &amp;lt; 0.05 relative to DMSO control) suppresses the protein expression of SKP2 (S-phase kinase-associated protein 2, a F-box protein that targets p27kip1 for degradation by 26S proteasome) and PSMA7 (proteasome subunit alpha type 7, a component of the 20S proteolytic complex in 26S proteasome). The decrease in SKP2 resulted in a concomitant up-regulation of p27kip1 protein, the major cell cycle inhibitor. The downregulating of PSMA7 impaired the 20S proteasome activity in both cell lines. These findings indicate that the novel rexinoid is effective in inhibiting cell proliferation of CTCL cells, in part, by suppressing SKP2 expression and the 20S proteasome activity resulting in increasing p27kip1 protein. In sum, UAB110 and also UAB30 with minimal effect in elevating serum TGs are both possible alternatives to bexarotene for the treatment of CTCLs. Citation Format: Chu-Fang Chou, Jason Esla, Venkatram Atigadda, Donald Muccio, Clinton Grubbs, Elmets Craig, Pi-Ling Chang. Preclinical assessment of retinoid X-receptor agonist, UAB110, for potential treatment of cutaneous T-cell lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-115.

PO-507 Deregulated mechanisms downstream of PLCG1 promoting cutaneous T cell lymphomaprogression

IntroductionCutaneous T Cell Lymphoma (CTCL) is a heterogeneous group of non-Hodgkin lymphomas with unfavourable prognosis at advanced stages. The malignant network that control the development of CTCL is now starting to be characterised. NGS-derived studies have identified multiple molecular alterations in specific genes and signalling pathways including PLCG1 and JAK-STAT. It has been shown that up to 50% of CTCL cases develop deregulated PLCG1-downstream signalling as shown by activating PLCG1 mutations and PRKCQ amplifications. Also, frequent mutations in JAK1 and JAK3 and in members of the STAT family have been identified, thus revealing a potentially important role of these signalling pathways in CTCL. We aimed at exploring the interactions between deregulated PLCG1 and JAK-STAT activity in CTCL using pre-clinical models and in samples from CTCL patients at different stages of the disease.Material and methodsHEK-BlueTM IL-6 and CTCL cell lines were used to study PLCG1 downstream signalling towards NFAT and STAT3 activation. Immunohistochemistry (IHC) of NFAT, pSTAT1, pSTAT3, pSTAT5, p50, p52, p65 and RelB wasperformed in 71 patient samples. The status of PRKCQ was analysed in 45 samples by fluorescent in situhybridization (FISH).Results and discussionsDeregulated PLCG1 activity elicited NFAT and STAT3 activation ex vivo. Downstream of PLCG1, pharmacological and genetic inhibition of PRKCQ negatively affected NFAT and STAT3 activation whereas a constitutively active PRKCQ mutant (A148E) was sufficient to trigger their transcriptional activity. Our data shows that downstream of PLCG1, PRKCQ can activate transcription mediated by both NFAT and STAT proteins. We also analysed nuclear accumulation of NFAT, STAT and NF-κB family proteins by IHC in patient samples. Interestingly positive P-STAT3 staining correlated with advanced diseases. Moreover, most CTCL-tumour stage cases studied were positive for PRKCQ amplification and/or 10q23 polysomy.ConclusionOur results support PLCG1 playing a mayor role at controlling CTCL progression towards advanced stages. Downstream of PLCG1 or due to genetic amplification, PRKCQ can mediate the activation of NFAT and STAT3. Thus, mutations in PLCG1, PRKCQ amplifications and/or nuclear NFAT and STAT3 accumulation can serve as diagnostic markers for CTCL. Moreover, these can also provide rational to develop specific therapies targeting PRKCQ alone or in combination with calcineurin or JAK-STAT inhibitors.

2325 Targeting isoform-specific PI3 kinase signaling for treatment of cutaneous T cell lymphoma

OBJECTIVES/SPECIFIC AIMS: (1) Determine the anti-proliferative effect of Copanlisib (α/δ PI3 kinase inhibitor) in CTCL cell lines and synergy with other anti-tumor drugs such as HDAC inhibitors. (2) Determine the effect of Copanlisib treatment on downstream targets of the PI3K/AKT/mTOR pathway. METHODS/STUDY POPULATION: We will test the anti-proliferative effect of Copanlisib on the cell lines H9 and HH, which are well characterized for evaluating new therapies for CTCL (Netchiporouk et al., 2017). We will also test the anti-proliferative effect of Copanlisib in combination with HDAC inhibitors on CTCL cell lines. Cell Titer Glo (Promega) will be used for the proliferation assay. Briefly, Cell-Titler Glo will be added after 24, 48, and 72 hours and luminescence will be measured as a % of maximal growth. Inhibitory effect will be determined by comparison to % growth of the control cultures without Copanlisib treatment. Our next objective is to determine the effect of Copanlisib treatment on downstream targets of the PI3K/AKT/mTOR pathway using Western blot analysis. In brief, 30 μg of each lysate will be subjected to 4%–12% gradient SDS-PAGE gel eletrophoresis. All primary antibodies were purchased from Cell Signaling Technology. Membranes will be washed with TBST and incubated with 1:10,000 dilution of IRDye-conjugated secondary antibody (Licor) for 1 hour. Results will be expressed as relative intensity: the intensity of each band adjusted to that of GAPDH. Experiments will be done in triplicate and 1-way analysis of variance followed by multiple comparison test will be applied to compare the cell proliferation between different treatment groups. RESULTS/ANTICIPATED RESULTS: Previous results from the Ai lab have demonstrated the importance of the PI3 kinase signaling cascade using high-throughput proliferation assays and siRNA knockdown of individual and double PI3 kinase isoforms (unpublished data). So far I have successfully established culture of HH and H9 cells in 96 well plates (100,000 cells/200 μL). We are titrating Copanlisib at doses from 20 nm to 20 μM. Initial results suggest a promising anti-proliferative effect and currently we are optimizing the most effective pharmacologic dose. Thus we anticipate that Copanlisib will exhibit a potent anti-proliferation activity in CTCL cell lines. H9 and HH cell lysates have been collected and preserved for Western blot analysis. We anticipate that Copanlisib treatment will significantly decrease the phosphorylation of AKT and 4EB-P1, both downstream targets of PI3 kinase signaling. DISCUSSION/SIGNIFICANCE OF IMPACT: These findings will elucidate the importance of the PI3 kinase/AKT/mTOR pathway in tumor proliferation in CTCL. Identifying the importance of specific isoforms of PI3 kinase in CTCL will allow for more targeted selection of treatment. Copanlisib targets α and δ isoforms of PI3K and is newly approved by the FDA for low grade B cell lymphoma. Our results seek to quantify the anti-proliferative effect of Copanlisib and determine an on-target mechanism of action by investigating the drug’s effects on downstream signaling molecules of the PI3 kinase pathway. This project will elucidate the disease process of CTCL and provide important insight for its management.

197 Loss of 5-hydroxymethylcytosine is an epigenetic hallmark of disease progression in cutaneous T cell lymphoma

DNA hydroxymethylation at the 5 position of cytosine (5-hmC) is a product of the ten-eleven translocation (TET) family of DNA hydroxylases. Accumulating pieces of evidence showed that loss of 5-hmC is critical for various biological and pathological processes. However, its expression and function in cutaneous T cell lymphoma remain largely unknown. Here, with a large cohort of patients, we report that loss of 5-hmC is an epigenetic hallmark of cutaneous T cell lymphoma, with diagnostic and prognostic implications. Immunohistochemistry staining on 91 mycosis fungoides (MF) cases display a significant decrease of 5-hmC staining in CD4+ T cells, compared to benign inflammatory dermatoses, including lupus erythematosus (N=15), lichen planus (N=15), and psoriasis (N=12). 5-hmC staining level is further decreased with disease progression and demonstrates remarkable loss of staining in the large cells of large cell transformed MF cases, regardless of the CD30 positivity. Furthermore, 5-hmC decreases are correlated with poorer overall survival in our patient cohort. Compared to healthy CD4+ T cells, 5-hmC levels also decreased in 9 CTCL cell lines. Pharmacological augment of global 5hmC with L-ascorbic acid [L-AA] and 2-phospho-L-ascorbic acid (P-AA) in CTCL cell lines lead to a remarkable 5-hmC accumulation and induce specific apoptosis in CTCL cell lines, as well as primary CTCL cells. In conclusion, 5-hmC is an epigenetic marker of disease progression in MF. Reversing 5-hmC levels in MF may serve as a novel therapeutic regimen in cutaneous T cell lymphoma.

PAK1 overexpression promotes cell proliferation in cutaneous T cell lymphoma via suppression of PUMA and p21

BackgroundCutaneous T cell lymphoma (CTCL) comprises a heterogeneous group of skin-homing T cell tumors. The small guanosine triphosphate effector p21-activated kinase 1 (PAK1) plays an important role in many fundamental cellular functions, including cell motility, proliferation, and apoptosis. The expression of PAK1 is up-regulated in several types of human cancers. However, little is known about the role of PAK1 in the pathogenesis of CTCL. ObjectiveThe aim of this study was to evaluate the expression pattern and underlying mechanism of PAK1 in CTCL. MethodsQuantitative real-time polymerase chain reaction(qRT-PCR) was used to detect PAK1 mRNA expression in the peripheral blood mononuclear cells (PBMCs) of patients with CTCL. The expression of PAK1 protein in CTCL tumor tissues was determined by immunohistochemistry. CTCL cell lines were treated with a small molecule inhibitor of PAK1, p21-activated kinase inhibitor III (IPA3), at concentrations of 2, 3.5 and 5 μM for 24 h. Hut 78 and HH CTCL cells were transfected with lentiviral-based PAK1 gene knockdown vectors. We determined the effects of PAK1 knockdown on cell proliferation and apoptosis in CTCL cells by MTS assay and flow cytometry. Animal experiments were performed to investigate the effects of PAK1 knockdown on the growth of tumors in vivo. Transcriptomic sequencing was performed to detect the direct downstream targets of PAK1 silencing. Reverse transcription polymerase chain reaction and western blot analysis were applied to verify the results of the transcriptomic analysis. ResultsWe detected PAK1 overexpression in PBMCs and skin lesions from patients with CTCL compared with benign inflammatory dermatoses (BID). Knockdown of PAK1 inhibited cell proliferation and promoted spontaneous apoptosis. In addition, the inhibitory effect of IPA3 was validated in the CTCL cell lines. Additionally, mice injected with PAK1-silenced cells presented with a decreased rate of tumor growth compared with the control groups. Moreover, the mRNA and protein expression of PUMA (BBC3) and p21 (CDKN1A) were increased in PAK1-silenced Hut 78 and HH cells. ConclusionsOur data indicated that PAK1 is upregulated in CTCL. PAK1 silencing induced apoptosis and inhibited cell growth by stimulating the expression of PUMA and p21. Thus, PAK1 may be a potential tumor marker and therapeutic target of CTCL.

Abstract 481: Diminished microRNA-29b results in overexpression of BRD4 and BRD4-regulated oncogenes in cutaneous T-cell lymphoma

Abstract Cutaneous T-cell lymphoma (CTCL) is a CD4+ T-cell malignancy that affects the skin but may disseminate systemically, and there are no long-term effective therapeutic strategies. Here, we characterized the therapeutic efficacy of manipulating the tumor suppressor microRNA-29b (miR-29b) and its target bromodomain-containing protein 4 (BRD4) in CTCL pathogenesis. Using primary CD4+ T-cells we determined that miR-29b expression is significantly decreased in patients compared to healthy controls (0.007±0.002, n=9 vs 1.008±0.052, n=6, p&amp;lt;0.0001). We utilized miR-29b-/- mice and bortezomib, a proteasome inhibitor known to increase miR-29b levels, to confirm the inverse relationship between miR-29b and BRD4 (Mishra, A. et al. Cancer Cell, 2012). Diminished miR-29b level resulted in increased BRD4 protein expression (1.87±0.29, p=0.014), while increase in miR-29b, as in bortezomib-treated CTCL cell lines, results in nearly undetectable BRD4 protein. We also observed increased genome-wide occupancy of BRD4 at regulatory regions in CTCL patients, an effect that is reversed with BRD4 inhibitor, JQ1, to similar levels as observed in healthy control CD4+ T-cells. Specifically, oncogene loci of NOTCH1 and RBPJ demonstrated enhanced BRD4 binding in CTCL patients, with resultant increases in mRNA in patient vs normal donors (4.16±0.98, p=0.024; 3.02±0.54, p=0.012). Further, in vivo treatment of interleukin-15 (IL-15) transgenic mice (Mishra, A. et al. Cancer Discovery, 2016) with JQ1 prevented CTCL development and decreased lesion severity (3.37±0.49, n=8 vs 6.0±0.45, n=5, p=0.004). Skin from JQ1-treated mice demonstrated decreased protein expression of BRD4, NOTCH1, and RBPJ, consistent with our findings in patients and CTCL cell lines. Bortezomib treatment of CTCL mice yielded similar results to JQ1 in vivo, while miR-29b levels were elevated in treated mice vs control animals. Since IL-15 signals through its receptor complex, we evaluated and observed increased BRD4 binding at IL-15 receptor αβγ gene loci in CTCL patients, which was reversed with both JQ1 and bortezomib treatment. Thus, we conclude that diminished miR-29b results in increased expression and subsequent binding of BRD4 at regulatory regions of known oncogenes in CTCL cells including NOTCH1, RBPJ, and the IL-15 receptor complex. We confirm the potential therapeutic utility of targeting this pathway in vivo, by direct displacement of BRD4 by JQ1, or by rescue of miR-29b expression by bortezomib. We therefore describe a novel targetable oncogenic pathway featuring IL-15, miR-29b, and BRD4 in CTCL. Citation Format: Rebecca Kohnken, Jing Wen, Bethany Mundy-Bosse, Max Yano, Leah Grinshpun, Kathleen McConnell, Ashleigh Keiter, Alex Hartlage, James Bradner, Michael Caligiuri, Pierluigi Porcu, Anjali Mishra. Diminished microRNA-29b results in overexpression of BRD4 and BRD4-regulated oncogenes in cutaneous T-cell lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 481. doi:10.1158/1538-7445.AM2017-481