Carlson, Brett L.

Abstract BACKGROUND/PURPOSE The Mayo Clinic has developed a large panel of patient-derived xenografts (PDX) from patients with glioblastoma multiforme (GBM). Here we report initial molecular profiling of the Mayo PDX panel and a comparison of patient and PDX molecular profile and response to therapy. METHODS & MATERIALS Clinical data was retrieved by retrospective chart review. Animal data from all PDX investigations conducted between 2004-2014 were retrieved from experimental logs and subsequently consolidated into a database for analysis. RESULTS From 1999 to 2014, 182 patient tumor samples of varying histological grades were attempted for xenograft with 73 resulting viable PDX lines. Viable xenografts were only produced from WHO grade IV tumor specimens, yielding an overall success rate of 49% for these tumors. GBM patients that produced viable xenografts compared to those that did not exhibited a trend for decreased overall survival (P = 0.18). There was no significant association between successful xenografting and whether tissue was from newly diagnosed (45/93%) or recurrent (20/40; 50%) tumors. Patient age > 45 at diagnosis was correlated with increased PDX viability in GBMs (p = 0.05). Of the viable PDX models analyzed, EGFR mutation was identified in 17 lines, TERT mutation was found in 13 lines, IDH mutation in 1 line, and MGMT hypermethylation in 25 lines. RNAseq was performed on orthotopic tumor samples from 53 PDX models. After excluding contaminating murine sequence reads, expression analysis demonstrated 32 models with a mesenchymal phenotype. Array comparative genomic hybridization was performed on 9 patient samples and derivative early, mid and late passage PDX tumors. Using unbiased hierarchical clustering, there was a high concordance between patient and xenograft models. Within the PDX panel, fractionated radiation (RT) alone and RT combined with temozolomide (TMZ) was tested in orthotopic tumors in 38 lines. The overall median survival benefit (ratio of median survival for treated vs. placebo) in PDX lines treated with RT only was 1.6 (range: 0.9-2.5) and with RT/TMZ was 2.5 (range: 1.1 - 8.9). There was a positive association between observed patient survival and the corresponding survival benefit in the PDX for subjects treated with RT/TMZ (r = 0.2; n = 17). Response to adjuvant TMZ was evaluated in 42 tumor lines, and response to bevacizumab was tested in 33 tumor lines, and correlations with clinical treatment response are being evaluated. CONCLUSIONS: The Mayo GBM PDX panel is widely used in the neuro-oncology community. The initial molecular analysis suggests a good correlation between patient and PDX at the level of genotypic characteristics and therapeutic sensitivity. Citation Format: Dioval A. Remonde, Brett L. Carlson, Mark A. Schroeder, Brock Armstrong, Sen Peng, Lisa Evers, Paul A. Decker, Jeanette Eckel Passow, Michael E. Berens, Nhan L. Tran, Robert B. Jenkins, Jann N. Sarkaria. Characterization of patient-derived xenograft (PDX) models to evaluate clinical and therapeutic responses of glioblastoma multiforme. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1468. doi:10.1158/1538-7445.AM2015-1468

Abstract Glioblastoma multiforme (GBM) is an aggressive and fatal disease commonly treated with radiation and temozolomide (TMZ), but emergence of therapy resistance invariably limits treatment efficacy. Inhibition of Poly-ADP ribose polymerases 1 and 2 (PARP1/2) sensitizes GBM cells, and inhibitors of PARP are emerging as combination partners with TMZ. However, limited brain penetration precludes use of several promising PARP inhibitors in GBM. BGB-290 is a relatively new and potent PARP inhibitor that crosses the blood brain barrier with a brain to plasma ratio of ∼0.2. The focus of the present study was to evaluate antitumor effects of BGB-290, as single agent or in combination with TMZ, in patient derived GBM xenografts and glioma cell lines. In a biochemical assay, low doses (30-100 nM) of BGB-290 suppressed PARP activity, while single agent anti-tumor activity of BGB-290 was modest and restricted to concentrations above 3 μM. Combinations of 0.1 μM BGB-290 with 10 μM TMZ significantly diminished primary neurosphere formation in a patient-derived xenograft line, GBM12, decreasing relative neurosphere count to 3.1 ± 1.3% as compared to 9.2 ± 3.1% with TMZ alone, p = 0.03. Consistent with a mechanism involving disruption of DNA repair, treatment of GBM12 cells with TMZ and BGB-290 combination resulted in robust γH2AX foci formation and a synergistic increase in DNA damage signaling as compared to TMZ or BGB-290 alone. TMZ-sensitizing effects of BGB-290 were heterogeneous in TMZ-resistant (GBM12TMZ) sublines that were developed previously by in vivo selection of parental GBM12 with cyclical TMZ therapy. The average neurosphere formation across 4 resistant sublines lacking MGMT expression after 100 μM TMZ only was 75.3 ± 9.0% versus 14.9 ± 7.4% with TMZ and 0.1 μM BGB-290, p<0.01. In contrast, 2 MGMT overexpressing sublines were much less sensitive to the combination with relative neurospheres 91.9 ± 5.2% versus 73.2 ± 9.2% respectively, p< 0.01. Next Gen Sequencing revealed discrete non-synonymous somatic mutations in the mismatch repair pathway of GBM12TMZ sublines lacking MGMT, which led us to examine the relationship between mismatch repair and sensitizing effects of PARP inhibition. Silencing Msh2 resulted in marked TMZ resistance in U251 cells (94.7± 5.4% relative growth with 100 μM TMZ), while co-treatment with 0.1 μM BGB-290 and 100 μM TMZ decreased growth to 23.5 ± 9.6% (p<0.01), which is similar sensitivity to TMZ alone in U251 parental cells. In an initial in vivo study with orthotopic GBM12 xenografts, administration of BGB-290 at 3 mg/kg twice per day combined with 25 mg/kg TMZ extended survival of animals beyond 300 days (80% survival at day 320), while in the group treated with TMZ alone 80% animals were moribund before day 150. In conclusion, BGB-290 is a promising PARP inhibitor with high potency, cytotoxicity and favorable in vivo efficacy in TMZ sensitive and possibly a subset of recurrent TMZ resistant GBM. Citation Format: Shiv K. Gupta, Brett L. Carlson, Mark A. Schroeder, Katrina K. Bakken, Ann C. Tuma, Jann N. Sarkaria. Inhibition of PARP activity by BGB-290 potentiates efficacy of temozolomide in patient derived xenografts of glioblastoma multiforme. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3505. doi:10.1158/1538-7445.AM2015-3505

Abstract GNE-317 is a dual PI3K/mTOR inhibitor with excellent blood-brain barrier penetration. In this study, the efficacy of GNE-317 was evaluated in the Mayo panel of primary glioblastoma multiforme (GBM) xenografts. Based on the concept of oncogene addiction, we hypothesized that GNE-317 would be most efficacious in EGFR-hyper-activated and/or PTEN-deficient lines. In a CyQUANT cell proliferation assay, the inhibitory concentration of 50% cell number (IC50) for GBM6, 10, 22, and 84 were 0.59±0.50, 0.72±0.40, 0.26±0.14, and 3.49±1.64 microM, respectively. In these selected lines, there was a poor correlation between EGFR or PTEN status and in vitro response. The efficacy of GNE-317 also was evaluated in 10 lines in an orthotopic xenograft model. Mice with established tumors were randomized to daily dosing with GNE-317 (30 mg/kg) or placebo until moribund. Similar to in vitro data, there was a spectrum of response to GNE-317 (Table 1). The survival benefit ratio (ratio of median survival) did not correlate with EGFR amplification (P=0.78), EGFR mutation (P=0.25), or PTEN (P=0.55) status when analyzed singly. However, GNE-317 did extend survival in 4 of 5 lines in which either EGFR or PTEN were deregulated. To evaluate potentially clinically relevant combinations, the efficacy of GNE-317 combined with bevacizumab (BEV) was tested in the same orthotopic model in GBM8, 10, and 59. In GBM10, GNE-317/BEV provided a significant increase in survival compared to the most effective single agent (P<0.0001). In GBM8, 10, and 59, the ratio of median survival for GNE-317/BEV vs. placebo were 1.47, 1.75, and 1.74, respectively (all P≤0.01). In conclusion, there may be a correlation between deregulation of the EGFR/PTEN signaling network and GNE-317 efficacy, and the efficacy of GNE-317 in some models can be extended when combined with bevacizumab. Table 1:Molecular Characteristics and Efficacy of GNE-317 in Mayo Primary GBM XenograftsEGFR AmplifiedEGFR vIII MutatedPTEN MutatedGNE-317 Responder*Benefit Ratio** of GNE-317GBM6YesYesWTYes1.35GBM8YesNoYes, HDYes1.19GBM10NoNoYes, HDYes1.47GBM22NoNoWTNo0.85GBM59YesYesYes, HDYes1.27GBM84YesNoYes, del Exon 1No1.02GBM115NoNoTBDNo0.86GBM116NoNoTBDNo1.05GBM117NoNoTBDYes1.38GBM122YesNoTBDNo0.99Abbreviations: HD=homozygous deleted; del=deletion in; TBD=to be determined; WT=wild type.* If in vivo survival difference reached P<0.05 by Log-rank Kaplan-Meier method between the GNE-317 treated and placebo groups, the GBM line was classified as a responder to GNE-317.** The benefit ratio equals to median survival of the GNE-317 treated group, divided by that of the placebo group. Note: This abstract was not presented at the meeting. Citation Format: Terence T. Sio, Jenny L. Pokorny, Ann C. Mladek, Brett L. Carlson, Mark A. Schroeder, Dennis O. Iyekegbe, Katrina Bakken, Laurent Salphati, Heidi Phillips, Jann N. Sarkaria. Preclinical efficacy of GNE-317, a dual PI3K/mTOR inhibitor, with or without bevacizumab in EGFR and/or PTEN-mutated glioblastoma multiforme. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2617. doi:10.1158/1538-7445.AM2014-2617

Abstract Purpose: Wee1 regulates key DNA damage checkpoints, and in this study, the efficacy of the Wee1 inhibitor MK-1775 was evaluated in glioblastoma multiforme (GBM) xenograft models alone and in combination with radiation and/or temozolomide. Experimental Design: In vitro MK-1775 efficacy alone and in combination with temozolomide, and the impact on DNA damage, was analyzed by Western blotting and γH2AX foci formation. In vivo efficacy was evaluated in orthotopic and heterotopic xenografts. Drug distribution was assessed by conventional mass spectrometry (MS) and matrix-assisted laser desorption/ionization (MALDI)-MS imaging. Results: GBM22 (IC50 = 68 nmol/L) was significantly more sensitive to MK-1775 compared with five other GBM xenograft lines, including GBM6 (IC50 >300 nmol/L), and this was associated with a significant difference in pan-nuclear γH2AX staining between treated GBM22 (81% cells positive) and GBM6 (20% cells positive) cells. However, there was no sensitizing effect of MK-1775 when combined with temozolomide in vitro. In an orthotopic GBM22 model, MK-1775 was ineffective when combined with temozolomide, whereas in a flank model of GBM22, MK-1775 exhibited both single-agent and combinatorial activity with temozolomide. Consistent with limited drug delivery into orthotopic tumors, the normal brain to whole blood ratio following a single MK-1775 dose was 5%, and MALDI-MS imaging demonstrated heterogeneous and markedly lower MK-1775 distribution in orthotopic as compared with heterotopic GBM22 tumors. Conclusions: Limited distribution to brain tumors may limit the efficacy of MK-1775 in GBM. Clin Cancer Res; 21(8); 1916–24. ©2015 AACR.

Abstract Wee1 is a key regulator of both G2/M transition and DNA replication fork stability. In this study, the efficacy of the Wee1 inhibitor MK1775 was evaluated in the Mayo primary glioblastoma multiforme (GBM) xenograft model. Based on western blot expression of Wee1, GBM22, GBM12 and GBM6 were selected for study based on low, intermediate and high-level expression, respectively. Wee1-mediated phosphorylation of Y15-Cdc2 was robustly suppressed in vitro in both GBM22 and GBM6 at 300 nM MK1775. At this concentration, MK1775 had marked anti-proliferative effects on GBM22 with 80±2.5% suppression of absorbance in a CyQUANT cell proliferation assay, while similar treatment of GBM6 had minimal effect on growth (26±21% suppression of absorbance). In a similar primary neurosphere formation assay, 300 nM completely inhibited neurosphere formation in GBM22 and GBM12, while having no effect on GBM6 neurosphere formation. Consistent with the cytotoxicity data, single agent treatment of GBM22 with MK1775 resulted in a marked S-phase arrest 24 hours after treatment (60% vs. 40% cells in S with MK1775 vs. control, respectively) and subsequent accumulation of cells in G2/M 48 hours after treatment (95% vs. 9.5%, respectively). In contrast, relatively minimal cell cycle perturbations were observed with MK1775 in GBM6 (24 hour S-phase accumulation 27% vs. 21%, respectively). Combination studies with MK1775 and temozolomide (TMZ) in a neurosphere formation assay demonstrated enhancement of treatment benefit only in GBM22 and GBM12 but not GBM6 (relative neurosphere numbers for TMZ, MK1775, or combined treatment compared to untreated were 34%, 16% and 7%, respectively for GBM22, 41%, 95% and 15%, respectively, for GBM12; while the same combination had no effect on GBM6 neurosphere formation). In anticipation of evaluating treatment efficacy in animal models, pharmacokinetic modeling of MK1775 brain delivery was performed in nude mice. Following a single MK1775 dose of 50 mg/kg, time to maximum blood concentration was 1.0 hours, AUC was 29.8 microM·h, and Cmax was 7.8 microM in whole blood and 0.24 microM in brain parenchyma. To evaluate whether limitations in brain delivery affected efficacy, parallel in vivo GBM12 heterotopic and orthotopic xenograft efficacy studies were performed with a 5 day dosing regimen of TMZ and MK1775 repeated every 28 days for 3 cycles. In the orthotopic model, MK1775 provided no survival benefit with median survival being 16, 15, 88 and 92 days for placebo, MK1775 alone, TMZ alone, or MK1775 + TMZ treatment, respectively. In contrast, there is some suggestion of MK1775 efficacy in the flank tumor model with median time to reach 1000 mm3 being 45, 56, 101 and 112 days, respectively. Collectively, these data suggest that despite promising in vitro activity for MK1775, limited drug delivery into the brain may reduce the efficacy of MK1775 therapy in an orthotopic GBM xenograft model. Citation Format: Jenny Pokorny, Vincenzo Pucci, Stuart Shumway, Bennett Ma, Katrina Bakken, Mark Schroeder, Brett Carlson, Jann N. Sarkaria. Pre-clinical investigation of the Wee1 inhibitor, MK1775, in glioblastoma multiforme. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3240. doi:10.1158/1538-7445.AM2013-3240

Abstract Purpose: Effective sensitizing strategies potentially can extend the benefit of temozolomide (TMZ) therapy in patients with glioblastoma (GBM). We previously demonstrated that robust TMZ-sensitizing effects of the [poly (ADP-ribose) polymerase] (PARP) inhibitor veliparib (ABT-888) are restricted to TMZ-sensitive GBM xenografts. The focus of this study is to provide an understanding for the differential sensitization in paired TMZ-sensitive and -resistant GBM models. Experimental Design: The impact of veliparib on TMZ-induced cytotoxicity and DNA damage was evaluated in vitro and in vivo in models of acquired TMZ resistance (GBM12TMZ-mgmtHigh, GBM12TMZ-mgmtLow, and U251TMZ), inherent TMZ resistance (T98G), and TMZ-sensitive (U251 and GBM12). In vivo drug efficacy, pharmacokinetics, and pharmacodynamics were analyzed using clinically relevant dosing regimens. Results: Veliparib enhanced TMZ cytotoxicity and DNA-damage signaling in all GBM models in vitro with more pronounced effects in TMZ-resistant lines at 3 to 10 μmol/L veliparib. In vivo, combined TMZ/veliparib, compared with TMZ alone, significantly delayed tumor growth and enhanced DNA-damage signaling and γH2AX levels in the sensitive GBM12 xenograft line but not in the resistant GBM12TMZ lines. The pharmacokinetic profile of veliparib was similar for GBM12 and GBM12TMZ tumors with Cmax (∼1.5 μmol/L) in tissue significantly lower than concentrations associated with optimal in vitro sensitizing effects for resistant tumors. In contrast, robust suppression of PARP-1 expression by shRNA significantly increased TMZ sensitivity of U251TMZ in vitro and in vivo. Conclusions: In vitro cytotoxicity assays do not adequately model the therapeutic index of PARP inhibitors, as concentrations of veliparib and TMZ required to sensitize TMZ-resistant cancer cells in vivo cannot be achieved using a tolerable dosing regimen. Clin Cancer Res; 20(14); 3730–41. ©2014 AACR.

Abstract Heat shock protein (HSP) 90 is a molecular chaperone that modulates stability of several client proteins, many of which are key regulators of signaling pathways in glioblastoma multiforme (GBM). The potential for simultaneously targeting multiple effectors makes HSP90 an appealing therapeutic target. Here, we evaluated the biological activity of HSP990, a novel HSP90 inhibitor (Novartis Pharma AG) in a panel of adult glioma primary xenograft lines (n=15). HSP990 treatment resulted in a dose-dependent growth inhibition with IC50 values in the low nM range in each of the primary lines. HSP990 caused G2/M cell cycle arrest and apoptosis-induction within 72 hours of treatment. In addition, treatment with HSP990 resulted in up-regulation of HSP70, and decrease in activity of client proteins AKT, CDC2, CDC25C and activation of caspase 8 and 3, in tested cell lines. A range of sensitivity based on EGFR status was observed. EGFR-amplified cells showed depletion of client proteins and growth inhibition at lower dose of HSP990 (10nM) than wildtype EGFR cells (30nM). Furthermore, neurosphere formation assays revealed that HSP990 was effective against the growth of glioblastoma stem cells (GSCs) in 6 tested glioblastoma xenograft lines (IC50 less than 10nM).These results suggest that that both cell cycle effects and induction of apoptosis are mechanisms governing HSP990 activity against glioblastoma. Further studies are warranted to characterize the mechanisms by which HSP990 exerts its anti-glioma effects which in turn may provide valuable insights in developing HSP990 as a novel personalized GBM treatment based on specific molecular signatures. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2898. doi:10.1158/1538-7445.AM2011-2898

Abstract Purpose/Background: The first-line chemotherapeutic temozolomide (TMZ) does not offer any therapeutic benefit in some cases of glioblastoma (GBM) and emergence of resistance leads to tumor progression in almost all patients. While the expression of O-6-methylguanine-DNA-methyltransferase (MGMT) correlates well with TMZ resistance, its expression varies among GBM tumors. The goal of this study is to identify additional genes with potential roles in TMZ resistance. Materials and Methods: Five primary GBM xenografts were generated from patient samples, and subsequent secondary models for resistance were developed by subjecting tumor-bearing mice to escalating doses of TMZ. Triplicate samples from parental and TMZ-resistant xenografts were analyzed for gene expression using the Illumina BeadChip microarray platform. The top 100 differentially expressed genes for each line were considered candidates. Genes were selected for qRT-PCR validation after ontological and pathway analysis. A lentivirus-mediated shRNA knockdown was performed to validate the role of several candidate genes in TMZ resistance in U87 and primary GBM12 cell lines. Knockdown was confirmed with qRT-PCR and western blots. TMZ responsiveness was assessed using CyQUANT and Neurosphere assays. Results: Upregulated MGMT expression was found in 3 of the 5 xenografts (GBM12TMZ, GBM14TMZ and GBM28TMZ) and was mechanistically linked with TMZ resistance in these lines using O6-benzylguanine to reverse the resistance phenotype. Differential expression of other DNA damage response genes included CCND2, CDKN1A, GADD45B, UBB as well as apoptosis regulators NFKB2 and BIRC3. In lines with unchanged MGMT expression (GBM22TMZ and GBM39TMZ), changes in anti-apoptosis genes NFKB2 and MAP3K1, transcription factor SRF and signal transduction gene PRKCDBP were validated by qRT-PCR. In GBM12TMZ, in addition to MGMT upregulation, expression of Ubiquitin B (UBB) was found to be the most significantly down-regulated gene both on microarray (Log2FC = −4.55, Fold change = 0.043) and qRT-PCR (Log2RQ =-4.04, RQ = 0.061). To test a possible role of UBB in TMZ resistance, knockdown was achieved in U87 (knockdown efficiency=91.2%) and in primary xenograft line GBM12 (knockdown efficiency=84.4%). The CyQUANT assay for U87UBB-KD and U87NT treated with escalating doses of TMZ revealed no significant difference in cell survival (10 M TMZ relative survival= 89.7% for U87UBB-KD and 79.2% for U87NT; p-value=0.53). Similarly, a neurosphere assay for GBM12UBB-KD and GBM12NT also yielded no significant difference in cell survival (10 M TMZ relative survival= 25.13% for GBM12UBB-KD and 23.35% for GBM12NT; p-value=0.83). Conclusions: We have used a TMZ resistance xenograft model to identify candidate TMZ resistance genes. The current data provides evidence on the role of MGMT in the development of acquired TMZ resistance while ruling out UBB as potential TMZ resistance gene in GBM. Additional studies are underway to assess the functional significance of other candidate genes, particularly those with differential expression in the setting of constant MGMT expression. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B68.

Abstract Heat shock protein (HSP) 90 is a molecular chaperone that modulates stability of several client proteins, many of which are key regulators of signaling pathways in glioblastoma multiforme (GBM). The potential for targeting multiple effectors makes HSP90 an appealing therapeutic target for GBM. The biological activity of HSP990, a novel HSP90 inhibitor (Novartis Pharmaceuticals) as single agent was evaluated in a panel of short-term explant cultures derived from primary GBM xenografts (n=10). HSP990 treatment resulted in a dose-dependent growth inhibition in the all xenograft lines, with an IC50 ranging from 5 to 30 nM. A 72 h exposure to HSP990 (10–30 nM) resulted in a dose-dependent decrease in the HSP90 client proteins Akt, Cdc2 and Cdc25C, a decrease in associated post-translational modifications p-Akt (S473), p-CDC2 (Y15), p-CDC25 (S216), and increased expression of HSP70. Additionally, a dose-dependent increase in HSP990-induced apoptosis (Annexin-V positivity), was observed within 24–72 hours of drug treatment. GBM tumors harbor a subpopulation of glioma stem cells that grow as neurospheres in neurobasal media and are considered important for tumor maintenance and progression. The effects of HSP990 were tested in neurospheres derived from 4 of the xenografts tested above (GBM12, GBM39, GB59, GBM12TMZ). These cultures demonstrated potential for tri-lineage differentiation upon serum challenge and had robust secondary neurosphere formation, demonstrating two hallmark features of cancer stem cells. In a bulk-culture neurosphere formation assay, co-incubation with 5 to 10 nM HSP990 was highly effective at suppressing neurosphere formation in all 4 xenograft lines. Notably, in GBM12 neurosphere cultures maintained in either serum-free media or challenged with serum, 30 nM HSP990 was equally effective at promoting up-regulation of HSP70, but exhibited less effect on Akt or Cdc2 in serum-free cultures as compared to cultures supplemented with serum. Incubation of serum-free neurosphere cultures with 5 nM HSP990 (5 days) resulted in increased glioneuronal differentiation exhibiting elevated GFAP, Oct-4 and beta-tubulin expression compared to untreated controls. To test ‘self renewal’ GBM12 neurospheres were treated with HSP990 (10, 30 nM) for 7 days followed by dissociation and plating equal numbers of viable cells at low density in fresh drug-free neurobasal media. Interestingly, HSP990 treatment did not affect the number of secondary neurospheres (surviving fraction: control = 1.00, 10 nM = 1.30, 30 nM = 1.05). Treatment of GBM12TMZ flank tumor xenografts with weekly HSP990 (13 mg/kg P.O.) resulted in a modest 9 day delay in tumor re-growth (40.3 and 49.3 respectively) compared to placebo treatment. In conclusion, HSP990 (a) had anti-proliferative, pro-apoptotic effect on differentiated serum-supplemented xenograft cultures (b) a pro-differentiation effect on serum-free GBM stem cell cultures (c) markedly suppressed neurosphere formation with prolonged drug exposure but not with more limited drug exposure (d) inhibited GBM growth in vivo. Thus, HSP990 may inhibit GBM growth both through direct growth inhibition and through pro-differentiation effects on GBM stem-like cells. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B100.