Resistance Relapse Research Articles

Abstract 28: Cell penetrating proteins targeting Mcl-1 induce in vitro and in vivo on-target cancer cell killing of Mcl-1 dependent cell lines

Abstract The pro-survival protein Myeloid Cell Leukaemia-1 (Mcl-1) plays an essential role in survival of numerous cancers. MCL-1 gene amplifications occur in a variety of human cancers and overexpression of the Mcl-1 protein is often associated with chemotherapeutic resistance and disease relapse. Complix has developed Cell Penetrating Alphabodies (CPAB), a novel and unique therapeutic class of proteins engineered to efficiently enter cells and inhibit proteins including Mcl-1. High affinity Alphabodies (ABs) targeting Mcl-1 were engineered by a combination of rational design and phage library screening. In affinity assays, these ABs were shown to bind to Mcl-1 with picomolar affinities whilst binding to Bcl-XL and Bcl-2 was below the detection limit of the assay. In vitro, CPAB uptake was shown to occur rapidly with cytosolic levels reaching up to 1 μM within 2 hours of CPAB exposure. Uptake was associated with cell death of the Mcl-1 dependent multiple myeloma (MM) cell line NCI-H929 (IC50=0.5 µM) and killing was correlated with caspase-3/7 activation. Anti-Mcl-1 CPAB were also shown to disrupt Mcl-1-Bak and Mcl-1-Bim complexes in H929 cells and induced dose-dependent Bak activation. In a panel of MM cell lines, anti-Mcl-1 CPAB induced cell death with a median IC50 of 0.96 μM and cell killing was not restricted to a specific subset of MM cell lines (CCDN1, MAF, MMSET). Gene expression analysis revealed that the anti-Mcl-1 CPAB cell killing potency correlates with MCL-1 gene expression but correlates best with the MCL-1:BCL-2 gene expression ratio. The same gene expression correlation analysis of the Bcl-2 targeting agent Venetoclax revealed an inverse pattern to that achieved with the Mcl-1 specific CPAB. In vivo, CPAB conferred with an albumin binding domain for extension of half-life, showed a serum half-life in mice of more than 1 hour and associated tumor concentrations of more than 1 µM. Immunohistochemistry and direct detection of fractionated tumor tissue confirmed the intracellular presence of the CPAB in the tumor cells. When given daily IV at 20 mg/kg, anti-Mcl-1 CPAB induced tumor growth inhibition of 50% versus control in two MM xenograft models (H929 and MOLP-8). Tumor growth remained significantly inhibited even two days after the last treatment in the MOLP-8 model and tumor growth inhibition was associated with increased staining of cleaved caspase-3 as compared to vehicle treated tumors. In summary, anti-Mcl-1 CPAB efficiently kill Mcl-1 dependent cancer cell lines by on-target effects as demonstrated by (1) disruption of Mcl-1-Bak and Mcl-1-Bim complexes, (2) Bak activation and (3) correlation of potency with MCL-1:BCL-2 gene expression ratio. These CPAB induced a robust reduction in tumor growth in mouse models and represent a best-in-class cell penetrating protein therapeutics for tackling intracellular PPI critical to diseases with unmet medical need. Citation Format: Sabrina Deroo, Sophie Thiolloy, Johan Desmet, Franky Baatz, Karen Vandenbroucke, Eric Lorent, Paula Henderikx, Philippe Alard, Stefan Loverix, Ignace Lasters, Yvonne McGrath. Cell penetrating proteins targeting Mcl-1 induce in vitro and in vivo on-target cancer cell killing of Mcl-1 dependent cell lines [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 28. doi:10.1158/1538-7445.AM2017-28

Abstract 5649: Cell-intrinsic PD-L1 directs tumor initiating cell fate and rapamycin and interferon-γ response

Abstract Tumor initiating cells (TIC) foster treatment resistance and tumor relapse. We show here that in mouse B16 (melanoma) and ID8agg (ovarian carcinoma), and human ovarian cancer ES2, PD-L1 knockdown (shRNA) reduced TIC numbers and functions. Data shown are from 2 PD-L1lo clones/cell type. Sorted, control CD44+CD133+CD24+ (B16), CD44+CD24+ (ID8agg) and ALDHhi (ES2) TIC formed 50-75% more and bigger spheres vs. PD-L1lo TIC in vitro, consistent with reduced TIC self-renewal in PD-L1lo. In vivo, PD-L1lo B16 and ID8agg TIC exhibited reduced tumorigenicity in wild type mice, consistent with poor TIC function, but immune effects could not be excluded. B16 and ES2 PD-L1lo TIC had significantly reduced tumorigenicity versus control TIC in immune deficient NSG mice consistent with cell-autonomous, immune-independent TIC promotion by PD-L1. In further confirmation, we found reduced serial transplantability of PD-L1lo B16 TIC in NSG mice. The canonical stemness genes nanog and pou5f1 (oct4) were significantly higher in control vs. PD-L1lo TIC for B16 and ID8agg. The canonical stemness gene SOX2 was significantly higher in control vs. PD-L1lo ES2 TIC. Altogether, these data validate effects in human cells, and are consistent with PD-L1 driving TIC through stemness gene control, although specific genes differed in distinct tumors (likely due to differing mutational landscapes). mTORC1 signaling linked to stemness (assessed by raptor) was lower but mTORC2 was unaffected (rictor) in PD-L1lo vs control TIC from all 3 tumors, suggesting mTORC1 control of PD-L1-mediated TIC generation. In support, the mTORC1 inhibitor rapamycin reduced control TIC significantly (>50% vs. untreated). Strikingly, rapamycin significantly increased PD-L1lo TIC in all 3 tumors, suggesting mTORC1 effects differ by tumor PD-L1 status. Interferon (IFN)-γ similarly reduced control TIC and augmented PD-L1lo TIC numbers, suggesting the novel concept that PD-L1 alters IFN-γ signals. PD-L1 sensitization of TIC to IFN-γ and rapamycin is the first example of a molecule enhancing TIC treatment response to our knowledge. We recently reported that αPD-L1 directly reduces B16 and ID8agg proliferation in vitro and in vivo (accession no. 27671674). Preliminary data now show that αPD-L1 (and αPD-1) directly reduce TIC proliferation, and these effects differ from non-TIC. We challenge the paradigm that tumor PD-L1 is primarily an immune escape molecule and that αPD-L1 works primarily to block T cell PD-1 interactions. Citation Format: Harshita B. Gupta, Curtis A. Clark, Bin Yuan, Gangadhara Sareddy, Srilakshmi Pandeswara, Alvaro S. Padron, Vincent Hurez, Jose-Conejo Garcia, Ratna Vadlamudi, Rong Li, Tyler J. Curiel. Cell-intrinsic PD-L1 directs tumor initiating cell fate and rapamycin and interferon-γ response [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 5649. doi:10.1158/1538-7445.AM2017-5649

Abstract 5136: Upregulation of Salmonella motility in tumors improves dispersion, colonization and intracellular invasion of the bacteria

Abstract Diffusion limitations in tumors prevent small molecule therapies from reaching cancer cells located distally from vasculature. This is one reason why some cancers become drug resistant and relapse. To circumvent these diffusional limitations, Salmonella can be used as a drug delivery vehicle. The facultative anaerobe preferentially colonizes tumor tissue over normal tissue at ratios greater than 10,000:1. Moreover, flagella allow the bacteria to penetrate deep into tumor tissue. Finally, genetically modified Salmonella can invade cancer cells and deliver a wide range of intracellularly functioning and patient specific therapies (DNA, RNA, and inhibitory peptides). But, Salmonella must distribute homogeneously within tumors and invade a significant number of live cancer cells in order to deliver therapies uniformly within tumors. To enable this, we hypothesized that up-regulating the master motility regulator, flhDC, in Salmonella would improve intratumoral penetration, uniform colonization and intracellular invasion of cancer cells. Salmonella utilize the transcription factor complex, flhDC, to regulate flagella synthesis and thus, motility. This transcription factor complex also positively influences type three secretion system (T3SS) assembly, which, Salmonella use to intracellularly invade epithelial cells. To test this hypothesis, we created a set of synthetic gene circuits that use arabinose for inducible expression of flhDC, the Plac promoter for constitutive expression of DSRed and the PSSEJ promoter to express GFP when Salmonella invade tumor cells. We analyzed aqueous bacterial motility with video microscopy. Finally, we administered the motility inducible Salmonella into a tumor-on-a-chip device to examine spatial and temporal tumor colonization characteristics of the bacteria. In an aqueous environment, the flhDC induced Salmonella swam approximately 33% faster than a Salmonella control (P<.01). Induction of flhDC increased the motile fraction (15-30 micrometers/second) of bacteria two-fold (P<.05) while also decreasing the non-motile fraction (0-15 micrometers/second) six-fold (P<.05) when compared to a Salmonella control. In a microfluidic device, flhDC induced Salmonella exhibited increased colonization and growth in tumor tissue located far away from channels that were meant to resemble tumor microvasculature (P<.05), when compared to a control. Finally, flhDC induction increased tumor cell invasion approximately two-fold in tissue located both proximal and distal to the micro-channels within the microfluidic device (P<.05), when compared to a control. Highly motile Salmonella could reduce the occurrence of drug resistant cancer relapse by delivering targeted therapies uniformly to cancer cells that would otherwise remain untreated with conventional therapy. Citation Format: Vishnu Raman, Nele V. Dessel, Owen O'Connor, Neil S. Forbes. Upregulation of Salmonella motility in tumors improves dispersion, colonization and intracellular invasion of the bacteria [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 5136. doi:10.1158/1538-7445.AM2017-5136

The potential of gemcitabine, cisplatin, and bevacizumab combination therapy for patients with ovarian clear cell carcinoma.

e17065 Background: Patients with advanced ovarian clear cell carcinoma (OCC) have poor prognosis, and response rate of chemotherapy for relapsed OCC is less than 10%. The effective standard treatment has not been established yet. The synergic effect between gemcitabine and cisplatin in ovarian cancer patients has been reported. And bevacizumab added on chemotherapy has some evidences of progression-free survival improvement at primary, platinum sensitive relapse and platinum resistant relapse situation than chemotherapy only. Therefore, we conducted a retrospective charts review to examine the safety and efficacy of gemcitabine, cisplatin, and bevacizumab combination (GPB) therapy in OCC patients treated in our hospital. Methods: Intravenous gemcitabine (1,000 mg/m2, day 1, 15), cisplatin (40 mg/m2, day 1, 15), and bevacizumab (15 mg/kg, day 1) were administrated every 28 days. This combination therapy was continued until disease progression or appearance of intolerable toxicity. This retrospective study was approved by our institutional ethics board. Results: Eight patients were treated with this GPB therapy in our institution between April 2013 and August 2015. Six patients received this therapy due to presence of relapse, and 2 patients were administered GPB as post-operative adjuvant therapy. Four patients were not treated with bevacizumab due to their medical conditions. The median treatment cycle was 5.5 (range: 1-15). Six patients were evaluable and their responses were partial response (PR) 1, stable disease (SD) 2, and progressive disease (PD) 3. The overall response rate was 13% and the disease control rate was 50%. Four patients needed dose reduction or treatment delay. Grade 3 adverse events observed were anemia (25%), nausea (13%), and elevation of G-GTP (13%). Conclusions: GPB therapy is feasible and moderately effective for patients with OCC. We are currently conducting the prospective multi-institutional phase II trial (UMIN 000023097) and expect that GPB therapy will be used as the standard treatment regimen for OCC, in future.

A study on intima- media thickness of carotid artery in children with nephrotic syndrome: a cross sectional study

Introduction: Atherosclerosis is one of the complications of Nephrotic syndrome. It is due to frequent exposure to hyperlipidemia, hypertension, and hypoproteinemia, steroid and immunosuppressive therapy may also contribute to the risk for cardiovascular diseases. Objective: To evaluate intima- media thickness of carotid artery in children with nephrotic syndrome. Design: A cross sectional study. Setting: The study was done in Nehru hospital of B.R.D Medical College Gorakhpur (U.P.) from June 2014- March 2015. Material & Methods: Total 150 children divided in to study group (n1=50) and control group II (n2=100). (n1=50) children with history of nephrotic syndrome enrolled in this study. The Inclusion criteria were nephrotic syndrome with normal serum complement, being on therapy for nephrotic syndrome (continuous or interrupted) for at least one year, glomerular filtration rate more than 90 ml/ min/ 1.73 m2 and age above two years at the time of study. (N2=100) healthy age, sex weight, height and RBS matched children considered as control group. The cIMT was evaluated in nephrotic children. Chi square test was used to test difference among study and control groups. A P value less than 0.05 was considered statistically significant. Result: In the study group (n1=50) steroid resistant, steroid dependent, steroid sensitive and infrequent relapse nephrotic syndrome included one-fourth each. The mean cIMT (mm) in nephrotic children was 0.42(±0.14) while mean cIMT in control was 0.34 (±0.06) and (p <0.0014). Subsequently, the factor that influenced on thickness of cIMT were BMI, SBP, DBP, Total cholesterol, VLDL, LDL, TGs, serum protein ((p<0.05) and duration of disease (r=0.83). Conclusion: cIMT was thicker in nephrotic children, were influenced by the factor that increases thickness of cIMT were hyperlipidemia, hypertension, hypoproteinemia and duration of disease.

Presence of novel compound BCR-ABL mutations in late chronic and advanced phase imatinib sensitive CML patients indicates their possible role in CML progression

ABSTRACTBCR-ABL kinase domain (KD) mutations are well known for causing resistance against tyrosine kinase inhibitors (TKIs) and disease progression in chronic myeloid leukemia (CML). In recent years, compound BCR-ABL mutations have emerged as a new threat to CML patients by causing higher degrees of resistance involving multiple TKIs, including ponatinib. However, there are limited reports about association of compound BCR-ABL mutations with disease progression in imatinib (IM) sensitive CML patients. Therefore, we investigated presence of ABL-KD mutations in chronic phase (n = 41), late chronic phase (n = 33) and accelerated phase (n = 16) imatinib responders. Direct sequencing analysis was used for this purpose. Eleven patients (12.22%) in late-CP CML were detected having total 24 types of point mutations, out of which 8 (72.72%) harbored compound mutated sites. SH2 contact site mutations were dominant in our study cohort, with E355G (3.33%) being the most prevalent. Five patients (45%) all having compound mutated sites, progressed to advanced phases of disease during follow up studies. Two novel silent mutations G208G and E292E/E were detected in combination with other mutants, indicating limited tolerance for BCR-ABL1 kinase domain for missense mutations. However, no patient in early CP of disease manifested mutated ABL-KD. Occurrence of mutations was found associated with elevated platelet count (p = 0.037) and patients of male sex (p = 0.049). The median overall survival and event free survival of CML patients (n = 90) was 6.98 and 5.8 y respectively. The compound missense mutations in BCR-ABL kinase domain responsible to elicit disease progression, drug resistance or disease relapse in CML, can be present in yet Imatinib sensitive patients. Disease progression observed here, emphasizes the need of ABL-KD mutation screening in late chronic phase CML patients for improved clinical management of disease.

Abstract P3-07-10: The role of CBP/FOXM1 in triple negative breast cancer

Abstract Background: Accurate assessment of prognostic and predictive biomarkers (estrogen receptor, progesterone receptor, HER2) plays a critical role in the clinical management of breast cancer. Triple negative breast cancers (TNBCs) lack the expression of all three targets, and no targetable molecular pathways have been identified to date. Hence, TNBCs are treated with non-targeted, cytotoxic chemotherapeutic agents (e.g. paclitaxel), and are characterized by high rates of drug resistance and metastatic relapse. CREB binding protein (CBP) has been implicated in cell growth and malignant transformation in various cancers. CBP is an important co-activator in the β-catenin driven transcription, including the Wnt signaling pathway, which has been implicated in TNBC biology. The Kahn lab has developed a specific CBP-binding small molecule inhibitor, ICG-001. We hypothesized that CBP-signaling plays an important role in TNBC biology and may provide a novel therapeutic target. Methods: We used TOP-flash assay to quantify Wnt signaling activity in TNBC. ICG-001 treatment combined with RNA Seq was used to characterize the role of CBP in TNBC cell line models. Co-immunoprecipitation (CoIP), protein and gene expression studies, as well as gene knock-down were used for validation. In vitro drug resistant cell line models as well as in vivo cell line (n=40 mice) and patient derived xenografts (PDX) in NOD scid gamma mouse models (2 patients, n=40 mice per patient) (treatmenent groups: control, paclitaxel, ICG-001, paclitaxel+ICG-001, n=5 mice per condition, primary and secondary implantation) were used to establish the effect of CBP inhibition via ICG-001 in TNBC on drug resistance and metastasis. We used the TCGA breast cancer data set to substantiate the experimental results. Results: We demonstrated that gene expression in TNBC is CBP, but not Wnt signaling dependent, and can be disrupted via ICG-001. RNA Seq analysis of TNBC cells treated with ICG-001 revealed Forkhead box M1 (FOXM1) as a potential downstream regulator. CoIP demonstrated that CBP binding to the FOXM1/β-catenin transcriptional complex. Treatment with ICG-001 revealed that CBP/FOXM1 binding, but not FOXM1/ β-catenin binding, is critical for FOXM1 expression. The PDX mouse models demonstrated that FOXM1 expression correlates with response to chemotherapy and disease recurrence in vivo. Treatment with ICG-001 sensitized FOXM1 high tumors to chemotherapeutic treatment and statistically significantly reduced tumor growth in serial transplantation experiments. Comparison of clinical data with FOXM1 expression in tumor samples from patients indicated that high levels of FOXM1 were associated with disease relapse and poor survival outcomes. Conclusion: CBP/FOXM1 binding is critical for FOXM1 expression. Targeting CBP/FOXM1 binding via ICG-001 could provide a novel therapeutic strategy in TNBC. The use of clinically annotated tissue microarrays containing a total of 430 breast tissue cores (51 TNBC cases) is currently pending to correlate nuclear protein expression of CBP and FOXM1 with survival outcomes in TNBC. FOXM1 and CBP could potentially be of value as predictive biomarkers in TNBC. These results could provide a clinical-translational rational for patient stratification based on CBP and FOXM1 expression for clinical trials exploring the therapeutic potential of FOXM1 inhibition via ICG-001 in combination with chemotherapy. Citation Format: Ring A, Nguyen C, Lenz H-J, Tripathy D, Lang JE, Kahn M. The role of CBP/FOXM1 in triple negative breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-07-10.

Sustained Overall Survival Benefit with Lenalidomide Plus Dexamethasone Versus No Treatment in Patients with Smoldering Myeloma at High Risk of Progression to Myeloma: Long Term Analysis

Background: For patients with smoldering multiple myeloma (SMM), the standard of care is observation. However, high-risk patients may benefit from early intervention.Methods: In this phase 3 trial, 119 patients with high-risk SMM were randomized to treatment or observation. The high risk populationwas defined by the presence of both PC_ 10% and MC_ 3g/dl or ifonly one criterion was present, patients must have a proportionof aPC within the total PCBM compartment by immunophenotypingof 95% plus immunoparesis. Patients in the treatment group received nine 4-week induction cycles (lenalidomide at a dose of 25 mg per day on days 1 to 21, plus dexamethasone at a dose of 20 mg per day on days 1 to 4 and days 12 to 15), followed by maintenance (lenalidomide at a dose of 10 mg per day on days 1 to 21 of each 28-day cycle) up to 2 years. The primary end point was time to progression (TTP) to myeloma. Secondary end points were overall survival (OS), response rate and safety.Results: After a median follow-up of 75 months (range: 57-100), there was a 57% reduction in the risk of death for the early treatment with lenalidomide-dexamethasone versus not treatment (hazard ratio, 0.43; 95% confidence interval, 0.2 to 0.9; P=0.02). Median overall survival has not been reached in either group, but 86% and 62% of patients are alive at 6 years in the early treatment and observation arms, respectively (Figure 1). The benefit in TTP is also highly sustained (hazard ratio: 0.24 (95% confidence interval, 0.14 to 0.41; P<0.0001). Progression to MM occurred in 53 out of the 62 patients (86%) in the abstention arm while only 22 out of 57 patients (38%) in the len-dex arm. At the time of progression patients received optimized treatments: bortezomib-based combinations were administered to thirteen out of 22 patients (59%) in the len-dex arm and to 23 out of 53 patients (43%) in the observation arm; lenalidomide-based combinations to 3 out of 22 patients (14%) in the experimental and to 8 out of 53 patients (15%) in the control arm; two out of 22 patients in the len-dex arm (9%) received bortezomib plus immunomodulatory agents whilst 16 out of 53 patients (30%) in the observation group received this combination; four out of 22 patients (18%) and six out of 53 patients (11%) in the len-dex and observation groups, respectively, were treated with chemotherapy; four patients (18%) in the experimental arm and 15 (28%) in the observation groups received an ASCT. Most patients responded to rescue therapies in both arms, resulting in overall response rates of 78% (17/22) and 86% (45/53) in the experimental and control arm, respectively. We compared survival from start of subsequent therapy in the patients population who progressed to active disease; the outcome was similar in both arms: at 6 years, 62% (16/22) of the patients in the len-dex arm remain alive and 49% (31/53) in the observation arm (P=0.50; Fig. 2C). The survival benefit observed was independent of the classification model used for defining high risk SMM ( Mayo Clinic and Spanish model)Conclusion: This long term follow-up analysis confirms that early treatment with lenalidomide-dexamethasone for high-risk SMM translates into a significant benefit in TTP but also in a sustained significant prolongation of the OS. The early exposure to lenalidomide-dexamethasone does not induce more resistant relapses. [Display omitted] DisclosuresMateos:Celgene: Honoraria; Janssen: Honoraria; Takeda: Honoraria; Amgen: Honoraria. De La Rubia:Amgen, Bristol Myers, Celgene, Janssen: Consultancy. Paiva:Celgene: Honoraria, Research Funding; Janssen: Honoraria; Takeda: Honoraria, Research Funding; Sanofi: Consultancy, Research Funding; EngMab: Research Funding; Amgen: Honoraria; Binding Site: Research Funding. Oriol:Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Relapse Prediction By Flow Cytometric Minimal Residual Disease Assessment in Infants with Acute Lymphoblastic Leukemia

Abstract Acute lymphoblastic leukemia (ALL) in infants (less than 1 year old) is a unique tumor with distinct biological features and poor outcome even when modern treatment schemes are used. Rearrangementsof MLL-gene, located in 11q23 region, are detected in approximately 75-80% of infants with ALL and lead to treatment resistance and high relapse rate. Nevertheless patients with germline MLL also demonstrate inferior outcome compared to ALL in older children. Thus additional risk factors implementation is one of the crucial points in infant ALL management. Minimal residual disease (MRD) measurement by multicolor flow cytometry (FCM) or various PCR technics is a well-standardized method of treatment response evaluation in childhood ALL, although in infants MRD data is not so widely applied. The aim of the study was to evaluate the relapse prediction feasibility in infant ALL by FCM MRD assessment during remission induction of MLL-Baby protocol. Methods. Totally 89 infants aged from 5 days to 11 months were enrolled in the present study. In 23 cases (25.8%) MLL gene was germline (MLL-g group), 33 patients (37.1%) had MLL-AF4 fusion while in remaining 33 cases (37.1%) other types of MLL-rearrangements were found. All patients were diagnosed as B-cell precursor ALL and all were treated by well established in Russia and Belarus MLL-Baby protocol, which is specially designed for infant ALL management. MRD was measured by 6-10-color FCM in bone marrow (BM) samples obtained at day 15 and at the end of remission induction (day 36). The availability of samples at at least one of these time-points was the only criteria for study group completion. In 43 patients with known types of MLL-rearrangements MRD was also assessed by fusion gene transcript (FGT) detection in RQ-PCR after first consolidation or first high risk block (for intermediate risk and high risk groups respectively). Relapse risk was investigated by cumulative incidence of relapse (CIR) estimation. Median of follow-up was 3 years 10 months. Results. Finally at day 15 MRD was studied in 71 cases. 17 patients (23.9%) were tested MRD-negative while remaining ones displayed various levels of MRD-positivity: 8 cases (11.3%) - from 0.01% to 0.1%; 14 (19.7%) - from 0.1% to 1%; 22 (31.0%) - from 1% to 10%; 10 (14.1%) - more than 10%. Proportion of MRD-positivity was lower in MLL-g group compared to MLL-rearranged (MLL-r) patients (58.8% and 81.5% respectively, p=0.06). Prognostic impact of day 15 MRD differed due to MLL-status. In MLL-r group significant differences between MRD(+) and MRD(-) patients were observed (n=10, CIR 0.28(0.18) and n=37, CIR 0.67(0.08) respectively, p=0.025). At the same time in MLL-g group these outcome differences were not significant (n=7, CIR 0 and n=9, CIR 0.22(0.14) correspondingly, p=0.197). Interestingly, in patients carrying MLL-AF4 fusion, known to be one of the most adverse types of MLL-rearrangements, day 15 MRD-negativity predicted low relapse incidence (n=5, CIR 0 and n=16, CIR 0.68(0.12) respectively, p=0.045). Thus day 15 MRD-negativity allows to detect low-risk MLL-r infants but it is not applicable in MLL-g group. It was previously shown that any detectable level of FGT after first consolidation or first high risk block predicts very poor outcome in MLL-r cases (G. Tsaur et al, ASH-2011). In current series RQ-PCR data in patients, FCM MRD(+) at day 15, distinguished groups of intermediate and very high relapse risk (n=17, CIR 0.51(0.13) and n=18, CIR 0.84(0.09) respectively, p=0.017). At day 36 FCM MRD was assessed in 82 infants. Among them 35 (42.7%) were tested negative while remaining 47 (57.3%) were MRD(+) at various levels. Prognostic value of day 36 FCM MRD data in MLL-r group was not significant: MRD(+) patients (n=23) had CIR of 0.71(0.08) while in MRD(-) cases (n=33) CIR was 0.49(0.12), p=0.153. Conversely, in MLL-g group low end-induction MRD (less than 0.1%) lead to excellent outcome compared to patients with higher MRD (n=14, CIR 0 and n=7, CIR 0.22(0.20) respectively, p=0.010). Conclusions. Thus FCM MRD data could distinguish infants with low risk of ALL relapse, but in MLL-r and MLL-g groups different time-points are prognosticaly significant. In MLL-g patients tandem application of FCM at early time-point and RQ-PCR later could help to define groups with low, intermediate and high relapse risk. MRD data could be added to MLL-Baby protocol risk group stratification, which is currently based on type of MLL-rearrangement. Disclosures No relevant conflicts of interest to declare.

Final Analysis of Overall Survival from the First Trial

Final Analysis of Overall Survival from the First Trial

A vicious cycle of metabolic interaction between leukemia stem cells and adipose tissues: implications in cachexia and drug resistance

Compelling evidences from recent studies suggest that leukemia stem cells (LSCs), a small subpopulation of malignant cells with long-term self-renewal capacity, play a key role in leukemia development, drug resistance, and disease relapse and recurrence (1). It has also been known for some time that tumor microenvironment or tissue niches may exert profound effect on the fates and viability of LSCs. A recent study by Ye et al . revealed a metabolic interaction between LSCs and adipose tissues, and showed that LSCs could be segregated into two metabolically and functionally distinct subgroups by their expression of the fatty acid (FA) transporter CD36 (2). The authors also suggested that gonadal adipose tissue (GAT) might provide a preferred niche to support CD36 + LSCs and promote drug resistance in a mouse model of blast crisis chronic myeloid leukemia (CML).

P01.08 Targeting DNA repair mechanisms in glioblastoma: from basic mechanisms to pre-clinical aspects and personalized therapy

AbstractIntroduction:Despite surgical resection and genotoxic treatment with ionizing radiation (IR) and the DNA alkylating agent temozolomide (TMZ), glioblastoma remains one of the most lethal cancers, due in great part to the action of complex DNA repair mechanisms that drive resistance and tumour relapse. One such mechanism involves the DNA-repair protein O6-methylguanine-DNA methyltransferase (MGMT) which mediates the direct removal of O6-methylguanine, the most highly cytotoxic lesion induced by TMZ. Importantly, silencing of MGMT through promoter methylation, which is observed in about 40% of GBM patients, confers a small but significant survival benefit in patients treated with TMZ and IR compared to IR only. In addition to MGMT, several DNA repair pathways have been involved in the repair of TMZ-induced lesions. Understanding the molecular details of these mechanisms and identifying potential pharmacological targets have emerged as vital tasks to improve treatment. At the same time, deciphering the genetic and epigenetic alterations that shape the “DNA repair makeup” of GBM cells should help tailor therapy to individual patients.Approach and Results:We have undertaken complementary approaches to understand the molecular basis behind chemoresistance, tumour progression and relapse in GBM patients. Firstly, we have measured the mRNA expression levels of a selection of DNA repair and cell cycle factors in paired, primary and recurrent GBM biopsies from patients treated or not with TMZ. Classification of the deregulated genes led to the identification of a gene signature that segregated 3 groups of biopsies. Inspection of these groups suggests that one route to tumour progression is associated with profound alterations in cell cycle genes as well as genes encoding crucial DNA repair factors. We have further exploited our differential gene expression analysis to learn how cancer cells modulate the expression of specific pathways in response to glioblastomagenesis and genotoxic treatment, and propose novel therapeutic strategies based on the inhibition of DNA repair factors. Lastly, we have carried out a large-scale shRNA screen of DNA repair/chromatin factors to identify those gene silencings that sensitize GBM cells to TMZ, as well as synthetic lethal interactions with MGMT. We are currently validating the most promising candidates in vitro as well as in vivo, using orthotopic xenograft models of GBM.

Utility of NSE, ProGRP and LDH in Diagnosis and Treatment in Patients with Small Cell Lung Cancer

背景与目的小细胞肺癌（small cell lung cancer, SCLC）是一种生长迅速、具有神经内分泌特性的肿瘤。血清神经元特异性烯醇化酶（neuron specific enolase, NSE）、胃泌素释放肽前体（pro-gastrin-releasing peptide, ProGRP）和乳酸脱氢酶（lactic dehydrogenase, LDH）已在SCLC的诊断和治疗中起到一定的辅助作用，本研究旨在通过治疗前后SCLC患者NSE、ProGRP和LDH的变化探讨标志物在肿瘤分期、疗效评价及预测复发方面的价值。方法纳入中国医学科学院肿瘤医院的SCLC初治病例，回顾性分析其临床数据，包括临床特征、治疗前及2周期化疗后的血清NSE、ProGRP及LDH，疗效及无进展生存期。结果治疗前，广泛期（extensive disease, ED）患者NSE、ProGRP及LDH均高于局限期（limited disease, LD）（P < 0.005）；LD患者的NSE水平随淋巴结分期的升高而明显增加（P=0.010）；有体重下降的患者NSE及LDH均高于无体重下降者（P=0.032, P=0.014）。化疗2周期后，有效患者的NSE及ProGRP下降程度明显高于疗效为稳定或无效的患者（P=0.015, P=0.002）。LD组化疗周期数 > 4个及治疗后ProGRP下降明显的患者较化疗周期数≤4个及ProGRP下降不明显的患者复发风险低；而远处转移数目≤2个、疗前LDH正常及治疗后ProGRP的明显下降，提示ED患者的近期复发风险低。此外，肿瘤复发类型（敏感复发、耐药复发、难治复发）与化疗后ProGRP下降程度呈负相关（P=0.044）。多因素分析结果提示治疗周期数是LD组SCLC近期复发的独立影响因素，远处转移数目及治疗后ProGRP的下降程度是ED组SCLC近期复发的独立影响因素。结论血清肿瘤标志物升高的程度与肿瘤负荷相关，ProGRP在治疗后的下降程度可能预测疗效及复发风险。

Abstract LB-119: Single-cell live imaging of paclitaxel-resistant cancer cell generation

Abstract Polyploid giant cancer cells (PGCCs) are abundant in cancers after chemotherapy. PGCCs have generally been considered nonviable because of their inability to execute mitosis. However, our recently reported studies challenged this paradigm by showing that PGCCs were capable of tumor initiation. To understand the role of PGCCs in chemoresistance, we monitored paclitaxel-induced PGCCs by conventional and live-cell fluorescence time-lapse recording. We found that paclitaxel induced massive cell death but also initiated endoreplication to form PGCCs. These PGCCs continued to grow via endoreplication, endomitosis, or cytofission before generating new mononuclear or multinucleated daughter cells via nuclear fragmentation or budding. These daughter cells resumed mitosis with a new stable karyotype and exhibited drug resistance. Budding from PGCCs was also observed in chemotherapy-treated ovarian cancer specimens from patients. On the basis of our findings, we propose the existence of an amitotic giant cell cycle composed of four phases: initiation, maintenance, termination, and speciation. We believe this is the first report of fluorescent live-cell imaging of the birth of resistant cancer cells at the single-cell level in response to paclitaxel treatment. Our findings provide strong evidence that a stress-induced amitotic giant cell cycle may be a programmed event underlying chemotherapy resistance and cancer relapse. Citation Format: Na Niu, Jie Zhang, Ning Zhang, Imelda Mercado-Uribe, Fangfang Tao, Zhenbo Han, Sen Pathak, Asha Multani, Jian Kuang, Jun Yao, Robert Bast, Anil Sood, Mien-Chie Hung, Jinsong Liu. Single-cell live imaging of paclitaxel-resistant cancer cell generation. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr LB-119.

Abstract 380: The dual PI3K/MTOR inhibitor PQR309 is active in mature B cell lymphoma cell lines bearing resistance to the PI3K-delta inhibitor idelalisib and specific gene expression features

Abstract Introduction The PI3K-delta idelalisib is among the most promising novel agents for lymphomas, but certain patients present a primary resistant disease or relapse after an initial response. Here, we studied the dual PI3K/MTOR inhibitor PQR309 in lymphoma cell lines that were resistant to idelalisib. Methods IC50 values were calculated with the MTT assay (72h) for 40 lymphoma cell lines. Gene expression profiling was done with Illumina HumanHT12 Expression BeadChips, data mining with Limma, GSEA and LINCSCLOUD. Results PQR309 and idelalisib showed only a partially overlapping pattern of activity (R = 0.6). While no cell line was sensitive to idelalisib and not to PQR309, 10 cell lines were sensitive to both compounds and 7 were sensitive only to PQR309. Idelalisib median IC50 was 28 nM (95%CI; 13-147) in the PQR309/idelalisib group and 13.4 μM (7.2-16.7 μM) in the PQR309-only (P 0.0006). PQR309 median IC50 was 113 nM (33-166 nM) in PQR309/idelalisib sensitive and 193 nM (111-372 nM) in the PQR309-only sensitive cells (P 0.03). No differences between the two groups were observed in terms of histology, diffuse large B cell lymphoma cell of origin, MYC deregulation, TP53 inactivation, BCL6 or BCL2 translocations. The gene expression signature of the PQR309/idelalisib sensitive cell lines was enriched of genes involved in IFN signaling, IL6/Jak/Stat3 signaling, oxidative phosphorylation, PI3K/MTOR signaling, chemokines signaling and proteasome (GSEA FDR &amp;lt;0.05; core enrichment genes including among others: PIK3CD, PTPN6, CCL3, CXCR4, CXCL10, JAK2, MYD88, STAT3, BAX, CDKN1A, AKT1, CDK2, PRKCB, PSMC4, PSMC2, MX1, MX2). Conversely, gene expression profiles of the PQR309-only sensitive cell lines were enriched of genes involved in hypoxia, IL2/STAT5 and glycolysis (GSEA FDR &amp;lt;0.05; core enrichment genes including, among others, ENO2, AKAP12, PRKCA, VEGFA, XBP1). The gene expression signature derived from genetic silencing of SOX4 in different cancer cells overlapped with the differences observed between our groups (connectivity score = -0.5) and the transcription factor itself, a positive-regulator of the PI3K/AKT pathway, was more expressed in the PQR309-only cells (log ratio 0.99 P 0.036). The cell surface marker CD24, involved in PI3K signaling, was the most up-regulated gene in the PQR309/idelalisib sensitive group (log ratio -2.1 P 0.02). Transcripts more expressed in the PQR309-only sensitive cells overlapped with genes down-regulated by PI3K and MTOR inhibitors in cancer cell lines (connectivity scores &amp;lt; -0.55). Conclusions. The dual PI3K/MTOR inhibitor PQR309 was active in a subset of mature B cell lymphoma cell lines bearing a primary resistance to the PI3K-delta inhibitor idelalisib and specific gene expression features. Expression of SOX4 and CD24 appear worth of validation on clinical specimens derived from prospectively treated patients. Citation Format: Tarantelli Chiara, Eugenio Gaudio, Ivo Kwee, Alberto Arribas, Andrea Rinaldi, Elena Bernasconi, Luciano Cascione, Petra Hillmann, Anastasios Stathis, Laura Carrassa, Massimo Broggini, Georg Stussi, Doriano Fabbro, Emanuele Zucca, Vladimir Cmiljanovic, Francesco Bertoni. The dual PI3K/MTOR inhibitor PQR309 is active in mature B cell lymphoma cell lines bearing resistance to the PI3K-delta inhibitor idelalisib and specific gene expression features. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 380.

Abstract 1222: Fluorinated N, N-diarylureas treatment as effective strategy to target colorectal cancer stem cells

Abstract Colorectal (CRC) cancer is the second leading cause of cancer deaths in the US; treatment of metastatic CRCs is limited due to drug resistance and eventual relapse. Cancer stem cells and PI3K mutation (i.e., pik3ca mutation) have been implicated in the relapse and lack of treatment response for many cancers including CRCs; therefore, it is critical to develop a therapeutic strategy that eliminates both the fast-growing cancer cells and the more resistant cancer stem cells. The purpose of the present study was to evaluate the anticancer activity of recently-described, novel AMPK activators, fluorinated N, N-diarylureas (FNDs), against CRC metastatic cell lines, pik3ca mutant cell lines, and CRC stem cells. METHODS. i) First, to assess AMPK activation, metastatic CRC cell lines (HT29 and KM20 cells), HCT116 pik3ca wild-type (WT) and mutant (MUT) cell lines, and CRC stem cell lines (from Celprogen) were treated with varying concentrations of 8 FNDs (4a, 4b, 4h, 4j, 4k, 4z, 4aa, 4bb). AMPK activation and activation of the upstream regulator, LKB1, was assessed by western blot. ii) Next, we determined the effect of the FNDs and, for comparison, metformin (an AMPK activator) on induction of cell cycle suppression and induction of apoptosis as assessed by cyclin D1 expression and PARP cleavage, respectively; β-actin antibody was used as a loading control. RESULTS. i) Treatment with the FNDs resulted in AMPK activation in HT29 and KM20 cells and CRC stem cells at concentrations as low as 10 μM with a peak activation at 12 h. We observed LKB1-independent AMPK kinase activation. Decreased LKB1 phosphorylation after FND treatment was dose dependent (i.e., decreased by ∼50% at 10 μM and undetectable at 50 μM). ii) All 8 FNDs (at a 25 μM dosage) completely suppressed cyclin D1 expression in HT29 and KM20 cells; a similar cyclin D1 suppression was noted with metformin at a 500x higher dosage (i.e., 10mM). Five of the 8 FNDs (4b, 4j, 4z, 4aa, 4bb) increased PARP cleavage in HT29 and KM20 cells. We used these 5 FNDs (at a dosage of 50 μM each) to treat the CRC stem cell lines; cyclin D1 suppression was noted for all 5 FNDs but strong induction of apoptosis was only identified using the 4b FND. Finally, we treated HCT116 pik3ca WT and MUT cell lines with 4b FND and found that cyclin D1 expression was completely suppressed at 40 μM and induction of apoptosis was noted at a dosage of 20 μM. CONCLUSIONS. We demonstrate cell cycle suppression and apoptosis in CRC cells and stem cells using the recently-developed FND compounds with the 4b compound as the most effective. These compounds appear to have considerable promise as a targeted cancer stem cell-specific agents and offer a potentially novel strategy for the treatment of CRC metastases. Citation Format: Piotr Rychahou, Dasha Kenlan, Vitaliy M. Sviripa, David S. Watt, B. Mark Evers. Fluorinated N, N-diarylureas treatment as effective strategy to target colorectal cancer stem cells. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1222.

Abstract 2520: IRS-1/LC3 nuclear structures and glioblastoma drug resistance

Abstract Drug resistance and frequent tumor relapses are the major obstacles in glioblastoma therapy, and recurrent tumors are practically incurable. We previously reported that insulin receptor substrate 1 (IRS-1), which is a typical signaling molecule for insulin and insulin-like growth factor 1 receptors, can translocate to nucleus, and that nuclear IRS-1 (nIRS-1) was found in different tumor cells, including glioblastomas. To unravel its function, we employed glioblastoma cell culture, animal models, and clinical samples. Using confocal imaging, molecular cloning, subcellular fractionation, mass spectrometry, gene expression analysis, and different approaches to verify protein-protein interactions, we demonstrate for the first time that nIRS-1 can form complex nuclear structures in a restricted number of cancer cells in glioblastoma biopsies and in intracranial glioblastoma xenografts. We also demonstrated the formation of highly organized ring-like structures in several cell lines, following ectopic expression of IRS-1 cloned in frame with nuclear localization signal (NLS-IRS-1). In these nuclear structures IRS-1 localizes at the periphery, and the core of the structure harbors a key autophagy protein, LC3; however, other autophagy proteins or biological membranes were not detected. In living cells expressing NLS-IRS-1-GFP fusion protein, IRS-1/LC3 structures are highly dynamic. They rapidly exchange IRS-1 molecules with nucleoplasm and interact with other nuclear complexes including BMI1-positive Polycomb bodies, PML bodies and Cajal bodies. Importantly, clones and mixed populations of cells expressing the NLS-IRS-1 and capable of forming the IRS-1/LC3 ring-like structures undergo extensive remodeling of gene expression, which suggests a transition to stem-like phenotype and associated resistance to several different anticancer drugs, including temozolomide. This is the first demonstration of IRS-1/LC3 nuclear complexes, which are highly dynamic and may play a role in epigenetic remodeling of glioblastoma cells towards stemness. Further studies are required to determine detailed molecular composition and to explain how these new nuclear structures function. Citation Format: Adam Lassak, Dorota Wyczechowska, Anna Wilk, Adriana Zapata, Mathew Dean, Luis DelValle, Jann N. Sarkaria, Augusto Ochoas, Francesca Peruzzi, Krzysztof Reiss. IRS-1/LC3 nuclear structures and glioblastoma drug resistance. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2520.

Abstract 2454: Omega-3 fatty acids DHA and EPA decrease oncogenic PGE2 in medulloblastoma in vitro and in vivo

Abstract Background: The embryonic brain tumor medulloblastoma is a highly malignant tumor of the cerebellum and the most common CNS malignancy of childhood. By combining neurosurgery, chemotherapy and radiation overall survival is about 70% but the intensive treatment regimen often leads to severe neurological sequelae and the risk of resistant relapses is significant. Hence, there is a need to improve treatment to reduce and alleviate late side effects. Inflammation plays an important role in the tumor microenvironment by suppressing immune response, promoting angiogenesis, facilitating tumor invasion, and metastasis. Prostaglandin E2 (PGE2) is a pro-inflammatory lipid mediator derived from the omega-6 fatty acid arachidonic acid (AA) through conversion by the key enzymes COX-2 and microsomal prostaglandin E synthase-1 (mPGES-1). Previous work from our group showed that medulloblastomas express high levels of COX-2 and mPGES-1 as well as all four PGE2 receptors (EP1-4) and most importantly that PGE2 promotes tumor proliferation. The ω-3 fatty acids docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) show anti-inflammatory characteristics in general and exert anti-tumoral properties in several cancer types. Here we investigate the potential role of DHA and EPA in treatment of medulloblastoma and their effect on prostaglandins in vitro and in vivo. Methods: In six medulloblastoma cell lines the cytotoxic activity of DHA and EPA was analyzed using cell viability assays. PGE2 levels were assessed in three medulloblastoma cell supernatants after 24 hrs of DHA treatment by ELISA. The prostaglandins were evaluated in human medulloblastoma xenografts in nude mice treated with DHA or a combination of DHA and EPA for 30 days from time of tumor take. Control mice were fed with normal chow. The levels of prostaglandins in the tumor tissue were analyzed with LC-MS/MS. Furthermore, the incorporation of fatty acids in tumors, erythrocytes and brain tissue was analyzed with GC-MS/MS. Results: Both DHA and EPA exerted cell toxicity in a dose dependent manner in vitro. DHA reduced the PGE2 production in all three cell lines investigated. In vivo, the treatment with DHA alone and the combination of DHA and EPA significantly increased the omega-3 index in tumors, erythrocytes and normal brain. The PGE2, and prostacyclin was significantly reduced in both treatment groups while thromboxane levels remained unchanged. The in vivo treatment was non-toxic. Conclusions: Lowering the pro-inflammatory PGE2 levels could mitigate the ongoing inflammation in the tumor microenvironment. We show that the omega-3 fatty acids DHA and EPA exert toxic effects on medulloblastoma cells in vitro and reduce PGE2 both in vitro and in vivo. We propose that DHA and EPA are interesting candidates to improve current medulloblastoma therapy. Citation Format: Linda Ljungblad, Filip Bergqvist, Teodora Andonova, Per-Johan Jakobsson, John Inge Johnsen, Per Kogner, Malin Wickström. Omega-3 fatty acids DHA and EPA decrease oncogenic PGE2 in medulloblastoma in vitro and in vivo. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2454.

A Whole Genome In Vivo CRISPR Screen in Primary ALL Predicts the Relapse

Introduction: Investigating gain of drug resistance in ALL is complicated by questions about a suitable model, cell lines are by and large not drug naive and patient material does not grow well in vitro. To this end we have sought to show that CRISR Cas9 genome editing technology can be used in patient derived ALL material in vivo to study emergence of drug resistance. Methods: We used primografted material from patient L707, who initially presented with a Dexamethasone (DEX) sensitive t(17;19) ALL, but relapsed 5 months later. Presentation cells were transduced with the full genome GeCKOv2 CRISPR library before transplantation into immunodeficient NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) mice. Mice were subsequently treated with DEX by oral gavage (15 mg/kg for 5 weeks, 10 mg/kg thereafter). DNA from was used for next gen sequencing and changes in pool complexity were analysed. By far the most significantly enriched sgRNAs were those targeting NR3C1, the glucocorticoid receptor. Results: SNP 6.0 analysis of the matched L707 relapse revealed a 5q deletion in the relapse, spanning 5 genes including NR3C1. Microarray analysis confirms that of the 5 genes, only 2 are expressed at any significant level in the presentation, NR3C1 and HMHB1. Growth of the relapse material in vivo and in vitro was slower than the presentation in a competitive situation, but with DEX treatment the relapse phenotype began to emerge with a small percentage of cells showing a heterozygous deletion of NR3C1. These data further support that NR3C1 deletion is the main driver of DEX resistance and relapse in this patient. Congruence between the genetic changes from presentation to relapse and the CRISPR screen imply that the results both robust and that the screen is effectively able to predict the major genetic alterations at relapse. Conclusion: These results demonstrate the power of in vivo CRISPR screens, we are able to predict the mechanism of gain of drug resistance and subsequent relapse in this patient. CRISPR genome editing is still a relatively new technique but we have shown it can be performed in patient derived material, and that the results from the screen correspond with the matched relapse material.

Abstract IA01: Epigenetic mechanisms of tumor initiation and evolution

Abstract The paramount importance of epigenetic regulation in cancer is now well established by a convergence of data from cancer genome sequencing, epigenomics and the transcriptional and molecular characterization of tumors and models. Our laboratory research is focused on characterizing epigenetic mechanisms of tumor initiation, as well as on understanding how epigenetic changes and intratumoral heterogeneity contribute to drug resistance and tumor relapse. We are particularly interested in brain tumors as clinically compelling models for investigating epigenetic mechanisms of tumorigenesis. I will present two ongoing projects in this area. The first project is focused on genetic mutations that affect epigenetic regulators and processes, and are associated with gliomagenesis. Here I will discuss our progress towards delineating specific mechanisms, regulatory targets and programs that explain the tumorigenicity of individual mutations. The second project is focused on epigenetic mechanisms that enable glioblastoma stem cells to adapt to environmental stressors or drug pressures. Here we have identified a specific role for histone demethylases in resetting the epigenetic landscapes of glioblastoma stem cells, and thereby allowing them to transition to a more primitive and refractory state. These respective studies exemplify ongoing research in our lab as well as in the broader research community, and have general relevance for cancer biology, diagnosis and therapy. Citation Format: Bradley E. Bernstein. Epigenetic mechanisms of tumor initiation and evolution. [abstract]. In: Proceedings of the AACR Special Conference on Chromatin and Epigenetics in Cancer; Sep 24-27, 2015; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2016;76(2 Suppl):Abstract nr IA01.