ALBERT

Beschreibung: Previously, we have reported that our human bone marrow (BM)-like scaffold xenograft model allows the engraftment and outgrowth of normal and malignant hematopoiesis (e.g. multiple myeloma (MM), acute myelocytic/lymphocytic leukemia (AML/ALL) and MDS (Groen et al. Blood 2012; Gutierrez et al. JCI 2014 and data not shown). Whereas the presence of osteoblasts and bone of human origin mimics a human BM-like niche more closely than the murine BM in standard xenotransplant models (e.g. NOD-SCID/NSG mice), still some essential components of the human BM niche, i.e. human blood vessels, are missing. To this end, in addition to human mesenchymal stromal cells we now incorporated cord blood-derived endothelial progenitor cells (CB-EPCs) in the hybrid scaffold production process, to create a multi-tissue compartment that "maximally humanizes" the BM-like niche of our scaffolds. Towards successful implementation of a human vascular system we compared: i) scaffold material composition (biphasic calcium phosphate (BCP) vs. tricalcium phosphate (TCP)); ii) scaffold shape (particles vs. tubes); iii) different types of matrigel for CB-EPC embedding. Histological analysis of the humanized scaffolds, eight weeks after implantation in mice, showed a large number of functional human blood vessels, as indicated by hCD31+ staining and the presence of erythrocytes within. Comparison of the composition and the shapes of the scaffolds indicated superiority of TCP and tube-shaped scaffolds in supporting the formation of vessels. Further analysis of scaffolds for CD44, CD146, LEPR and nestin-positive cells, revealed the presence of other stromal niche cells besides human osteoblasts and endothelial cells. Irradiation of mice carrying these humanized implants did not have a significant deleterious effect on the established human vessels, allowing their further functional evaluation in xenotransplantation. Additionally, mice carrying tubes with and without human CB-EPC derived vessels (on either flank) were subsequently inoculated with adult BM-derived CD34-positive cells by intracardiac injection. Upon analysis 12 weeks later, all tubes showed multi-lineage hematopoietic outgrowth. Interestingly, CB-EPC embedment resulted in increased numbers of CD45+ (2-fold), CD13+ (4-fold) and CD7+ (2-fold), while CD19+ cell numbers were equal. In contrast, in mouse BM almost only CD19+ cells could be detected. Moreover, we observed that the use of CB-EPCs in our scaffolds provides faster kinetics of in vivo engraftment and growth of both patient-derived MM or AML cells. With the addition of both human CB-EPCs and human BM stromal cells, our scaffold systems now simulate both human endosteal and vascular niches of the BM, thereby more closely recapitulating the human hematopoietic niche. Disclosures Yuan: Xpand Biotechnology BV: Employment. de Bruijn:Xpand Biotechnology BV: Employment. Mitsiades:TEVA: Research Funding; Janssen/Johnson & Johnson: Research Funding; Novartis: Research Funding. Martens:Johnson & Johnson: Research Funding. Groen:Johnson & Johnson: Research Funding.

Beschreibung: Abstract 1333 Poster Board I-355 The curative Graft versus Tumor (GvT) effect of allogeneic stem cell transplantation and donor lymphocyte infusions is mainly mediated by donor derived T cells recognizing minor Histocompatibility antigens (mHag) presented by malignant cells. Traditionally, CD8+ cytotoxic T cells (CTLs) are considered as “the” effector cells of these anti-tumor responses, whereas a sole helper role is attributed to CD4+ T cells. CD4+ T cells often possess killer capacities in vitro, raising the possibility that they may also mediate anti-tumor effects without the need for CD8+ T cells. Hence, we here explored the feasibility of adoptive immunotherapy with sole CD4+ CTLs by testing the therapeutic capacity of a mHag-specific, CD4+ CTL (3AB11) in a human GvT model. Rag2−/−γc−/− mice were inoculated with BLI detectable, mHag+ Multiple myeloma cells. Treatment of established tumors with 3AB11 but not with control CD4+ T cells by triple consecutive injections of 30-40×106 cells/day rapidly reduced the medullary growing tumors, illustrating for the first time the feasibility to establish a significant GvT effect by targeting a sole mHag recognized by CD4+ CTLs. The therapy was less effective at a higher tumor load and unsuccessful by a single injection of 20×106 cells, underscoring the critical importance of T cell dose-to-tumor load ratio to establish an efficient anti-tumor effect. In further exploration, tumors were also significantly reduced by treatment with “dual antigen-specific” T cells, which were generated by transduction of the T cell receptor (TCR) of clone 3AB11 into recall antigen (Tetanus Toxoid; TT)-specific T cells. Finally, the in vivo persisting dual-specific T cells could be boosted by administration of TT loaded mHag negative B cells, demonstrating for the first time the feasibility and potential advantages of immunotherapy with TCR-transduced “dual antigen-specific” CD4+ T cells. We conclude that potent GvT effects may be achieved in clinical trials by targeting a sole mHag antigen with original or TCR-redirected CD4+ CTLs. Disclosures No relevant conflicts of interest to declare.

Beschreibung: During the last two decades, cell lines and patient-derived samples from multiple myeloma (MM) have been extensively profiled for alterations in their genome with the anticipation that those genes with the most recurrent lesions could represent attractive novel therapeutic targets or markers for aggressive disease. Yet for many of these genes, their functional significance for MM cells has not been formally evaluated. With the advent of new CRISPR/Cas9-based functional genomics platforms, it is possible to generate in genome- or subgenome-scale direct quantitative information on the impact that perturbation of these genes exerts on tumor cell survival, proliferation or other phenotypes. We therefore examined the landscape of our CRISPR-based functional genomic data for these recurrently dysregulated genes We specifically curated information from the MMRF CoMMpass study and multiple other publicly available studies, to identify genes which are recurrently identified to harbor nonsynonymous mutation (SNV or indel), DNA copy number loss or gain, or participation in chromosomal translocations. We then examined the patterns of results for these genes in our genome scale CRISPR-based gene-editing studies for loss-of-function in n=18 MM cell lines. We identified a subset of genes (e.g. FAM46C, CDKN2C, RASA2) which are considered targets for recurrent loss-of-function events and indeed exhibit, for large fractions of the cell lines tested enrichment, of their sgRNAs in CRISPR knock-out studies, consistent with a role of these genes as suppressors of tumor cell survival or proliferation. CRISPR KO of TP53 leads to increased survival/proliferation of only a small minority (2/18 of cell lines tested thus far), which reflects the fact that the overwhelming majority of MM cell lines already harbor LOF events for this gene. Interestingly, a substantial number of genes which have been considered to harbor recurrent LOF events in MM patient samples (e.g. NF1, NF2, CYLD) do not exhibit sgRNA enrichment in CRISPR KO screens in the MM cell lines tested so far. In addition, several other recurrently mutated genes for which their loss- or gain-of-function status had not been previously evaluated with extensive functional studies in MM (e.g. SP140, LTB, EGR1, ATM, PARK2, PRKD2, RAPGEF5, DOCK5, TGDS, TNFAIP8) exhibit in the majority of cell lines tested in in CRISPR knockout studies no significant enrichment or depletion of their sgRNAs. In contrast, PTPN11, CREBBP, EP300, KMT2B, KMT2C, SETD2, SF3B1 and UBR5, are notable examples of recurrently mutated genes which represent dependencies for large fractions of MM cell lines in vitro. These results highlight the value of interpreting results from next generation sequencing studies in the context of information provided by the genome scale by use of functional genomic characterization of available cell line models. We envision that, similar sub-genome scale assays were performed at the level of patient derived samples will also provide direct information about the relevance of some of these genes. In addition, functional studies conducted with context of tumor-microenvironemtn compartment interactions and tumor interface will be needed to evaluate several genes identified in the study. Disclosures Licht: Celgene: Research Funding. Mitsiades:Takeda: Other: employment of a relative; Janssen/ Johnson & Johnson: Research Funding; Abbvie: Research Funding; EMD Serono: Research Funding; TEVA: Research Funding.

Beschreibung: TH and MW contributed equally to this work. Background Multiple myeloma is caused by an accumulation of malignant plasma cells in the bone marrow. Myeloma is characterized by an osteolytic bone disease, caused by increased bone degradation and reduced bone formation. Bone morphogenetic proteins (BMPs) are members of the transforming growth factor (TGF)-β superfamily. BMP-signaling is important for both pre- and postnatal bone formation. Additionally, several BMPs induce growth arrest and apoptosis in myeloma cells. Thus, increasing BMP-signaling in myeloma patients may reduce tumor growth and restore bone formation. We therefore explored BMP4 gene therapy in a human-mouse model of multiple myeloma. Methods Calcium phosphate scaffolds with human mesenchymal stromal cells (MSCs) were implanted in RAG2-/-GC-/- mice and the MSCs were left to differentiate in vivo for 8 weeks to create a humanized bone microenvironment. Then, adeno-associated virus (AAV), AAV8-BMP4, which has tropism for liver cells and expresses murine Bmp4 under the control of the liver specific human α1-antitrypsin (hAAT1) promoter, were administered by tail-vein injection. Empty viral vectors, AAV8-CTRL, were used for the control group. After 2 weeks, when BMP4 was detectable in circulation, we injected fluorescently labelled KJON myeloma cells in 3 out of 4 scaffolds in each mouse. The KJON cells are hyperdiploid, have a relatively slow growth rate and rely on interleukin (IL)-6 supplementation in the absence of a supporting microenvironment, thus resembling primary human myeloma cells. Tumor growth was examined by weekly imaging until end-point, 6 weeks after tumor cell injection. Results At end-point, serum levels of BMP4 in AAV8-BMP4 mice were in the range of 50-200 ng/mL, but not detectable in AAV8-CTRL mice. Strikingly, tumor growth as quantified by imaging was significantly reduced in AAV8-BMP4 mice compared with the AAV8-CTRL mice (p

Beschreibung: Heterobifunctional proteolysis-targeting chimeric compounds leverage the activity of E3 ligases (e.g. CRBN and VHL) to induce neopmorphic ubiquitination and proteasomal degradation of target oncoproteins, with potent preclinical activity against diverse neoplasias. Despite intense recent efforts to develop pharmacological "degraders" against many different oncoproteins, the mechanisms regulating tumor cell sensitivity to different classes of these "degraders" remain incompletely understood. To address this question in an unbiased manner, we performed genome-scale CRISPR/Cas9-based gene editing loss-of-function (LOF) studies in MM.1S multiple myeloma (MM) cells treated with CRBN-mediated degraders of BET bromodomain proteins (dBET6) or CDK9 (Thal-SNS-032); or with VHL-mediated degraders of BET bromodomain proteins (ARV-771 or MZ-1). We observed that MM cell resistance to any of these "degraders" does not involve genes with recurrent LOF in MM patients and association with high-risk MM (e.g. for TP53, PTEN, negative regulators of cell cycle, et.c.), suggesting that these degraders may exhibit activity against tumor cells with prognostically adverse genetic features. In tumor cells resistant to the CRBN-mediated degraders dBET6 and Thal-SNS-032, we observed significant enrichment of sgRNAs targeting CRBN itself or (to a lesser extent) other components or regulators of its cullin RING ligase (CRLCUL4A) complex, including members of the COP9 signalosome (COPS7A, COPS7B, COPS2, COPS3, COPS8, GPS1, etc.), DDB1, or the E2 ubiquitin conjugating enzyme UBE2G1. In tumor cells resistant to the VHL-mediated degraders MZ-1 and ARV-771, we observed pronounced enrichment of sgRNAs for CUL2, VHL itself, other members (e.g. RBX1, elongin B/C [TCEB1, TCEB2] of the CUL2 complex with VHL), as well as COP9 signalosome genes (COPS7B, COPS8) and UBE2R2. We also validated, using individual sgRNAs for several of these candidate genes that their CRISPR knockout can decrease tumor cell response to dBET6 and Thal-SNS-032 treatment (e.g. for CRBN, COPS7B, COPS2, or COPS8) or MZ-1 and ARV-771 (e.g. for VHL, COP7B and COPS8). Notably, the sgRNAs against COP9 signalosome genes conferred less pronounced decrease in sensitivity to VHL-, than CRBN-based, degraders, suggesting that COP9 signalosome loss has differential roles in the function of CUL4ACRBN vs. CUL2VHL and potentially other CRL complexes. Tumor cells isolated from our CRISPR knockout screens with confirmed resistance to a given degrader were then treated with other degraders operating through the same or different E3 ligase; and against the same or different oncoprotein: we observed cross-resistance between degraders operating through the same E3 ligase against different oncoproteins, but not for degraders targeting the same protein via different E3 ligase/CRLs: this result is consistent with our observation for substantial gene-level differences (despite pathway-level similarities) for resistance mechanisms for CRBN- vs. VHL-based degraders. In conclusion, our study systematically defined at genome-scale the resistance mechanisms of tumor cells against degraders which leverage the same E3 ligase against different targets; or target the same oncoprotein through different E3 ligases/CRL complexes. We observed that for multiple types of degraders, tumor cell resistance is primarily mediated by prevention of, rather than adaptation to, breakdown of the target oncoprotein. The observed pathway-level similarities and major individual gene-level differences in resistance mechanisms for CRBN- and VHL-mediated degraders likely reflects the different composition and regulation of the respective CRL complexes mediating the action of these classes of degraders Our observations suggest that preventing or delaying resistance to pharmacological degradation of oncoproteins may require concurrent or sequential/alternating use of degraders operating through different E3 ligases and ideally, different CRL complexes; while synthetic lethal strategies to prevent COP9 signalosome LOF may also be contemplated to counteract a common, but quantitatively less pronounced, potential mechanism of resistance for several different classes of degraders. Collectively, our study highlights important new directions in the development of new pharmacological degraders for blood cancers and other neoplasias. Disclosures Richardson: Karyopharm: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Jazz Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees, Research Funding; BMS: Research Funding; Oncopeptides: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees. Licht:Celgene: Research Funding. Boise:Abbvie: Consultancy; AstraZeneca: Honoraria. Gray:C4 Therapeutics: Consultancy. Mitsiades:TEVA: Research Funding; Janssen/ Johnson & Johnson: Research Funding; EMD Serono: Research Funding; Takeda: Other: employment of a relative; Abbvie: Research Funding.