Kitange, Gaspar J.

Abstract Background EGFR targeting antibody-drug conjugates (ADCs) are highly effective against EGFR-amplified tumors, but poor distribution across the blood–brain barrier (BBB) limits their efficacy in glioblastoma (GBM) when administered systemically. We studied whether convection-enhanced delivery (CED) can be used to safely infuse ADCs into orthotopic patient-derived xenograft (PDX) models of EGFRvIII mutant GBM. Methods The efficacy of the EGFR-targeted ADCs depatuxizumab mafodotin (Depatux-M) and Serclutamab talirine (Ser-T) was evaluated in vitro and in vivo. CED was performed in nontumor and tumor-bearing mice. Immunostaining was used to evaluate ADC distribution, pharmacodynamic effects, and normal cell toxicity. Results Dose-finding studies in orthotopic GBM6 identified single infusion of 2 μg Ser-T and 60 μg Depatux-M as safe and effective associated with extended survival prolongation (>300 days and 95 days, respectively). However, with serial infusions every 21 days, four Ser-T doses controlled tumor growth but was associated with lethal toxicity approximately 7 days after the final infusion. Limiting dosing to two infusions in GBM108 provided profound median survival extension of over 200 days. In contrast, four Depatux-M CED doses were well tolerated and significantly extended survival in both GBM6 (158 days) and GBM108 (310 days). In a toxicity analysis, Ser-T resulted in a profound loss in NeuN+ cells and markedly elevated GFAP staining, while Depatux-M was associated only with modest elevation in GFAP staining. Conclusion CED of Depatux-M is well tolerated and results in extended survival in orthotopic GBM PDXs. In contrast, CED of Ser-T was associated with a much narrower therapeutic window.

Abstract Introduction: EGFR targeted antibody-drug conjugates (ADCs) show promise as a novel treatment in a subset of glioblastoma (GBM). Two EGFR targeting ADCs include first generation Depatux-M, with an antimitotic toxin monomethyl auristatin F (MMAF), and ABBV-321, with a DNA crosslinking agent pyrrolobenzodiazepine dimer (PBD) toxin. Due to the large molecular weight, poor drug distribution across the blood-brain barrier significantly limits the efficacy in EGFR-amplified GBM. We studied whether convection enhanced delivery (CED) can be used to safely infuse these two EGFR-targeted ADCs in patient-derived xenograft (PDX) models of EGFR-amplified GBM. Methods: The efficacy of Depatux-M and ABBV-321 was evaluated in vitro and in vivo in two EGFRviii-amplified PDXs (GBM6 and GBM108). Immunofluorescence staining was used to evaluate drug distribution along with pharmacodynamics of the ADCs. CED was performed by stereotactic placement of an infusion catheter in the same location as the original tumor implantation. Immunohistochemistry was used to explore mechanisms of normal cell toxicity. Results: Despite potent activity in vitro (high ng/ml IC50), systemic administration of either ADC conferred minimal extension in survival for either GBM6 or GBM108. In contrast, CED significantly enhanced ADC delivery to tumor and peri-tumoral regions and extended survival. In a pilot study in GBM6 with n=3 mice per group, a single CED infusion of 0.002mg ABBV-321 resulted in extended survival beyond 365 days for two mice, while other doses and placebo were associated with shorter median survival (0 mg - 39 days; 0.03 mg - 105 days; 0.3 mg - 49 days). In a subsequent trial evaluating serial infusions performed every 21 days, four infusions at 0.002 mg ABBV-321 resulted in lethal toxicity. In contrast, limiting ABBV-321 to only two serial CED infusions in GBM108 was associated with extended survival of more than 300 days vs. 53 days for AB095 antibody control infusion. In contrast, serial infusion with Depatux-M was much better tolerated. In a single infusion in GBM6, a 0.06 mg dose was well tolerated and associated with a 49-day extension in median survival. Further, four serial infusions at 0.06 mg given every 21 days in GBM6 and GBM108 was associated with 100 day and more than 250-day extension in survival as compared to AB095 antibody control infusion. To investigate the mechanism of toxicity, infusion of non-tumor bearing C57Bl6 mice with ABBV-321 resulted in a marked loss in NeuN staining and elevated CD68 and GFAP staining 7 days later; Depatux-M infusion was only associated with modest elevation in GFAP without loss of NeuN staining or elevated CD68-positive cells. Conclusion: Depatux-M is well tolerated when infused into normal brain and results in extended survival in orthotopic GBM PDXs. In contrast, ABBV-321, with a distinct PBD toxin, had a much narrower therapeutic window when delivered by CED. Citation Format: Kendra A. Porath, Ann Mladek, Rachael A. Vaubel, Michael S. Regan, Sonia Jain, Danielle Burgenske, Katrina Bakken, Brett Carlson, Margaret Connors, Zeng Hu, Lihong He, Paul A. Decker, Gaspar Kitange, Shiv Gupta, Nathalie Y. Agar, Thomas M. Feldsien, Didier R. Lefebvre, Jann N. Sarkaria. Convection enhanced delivery of EGFR-targeting antibody drug conjugates ABBV-321 and Depatux-M [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 334.

Abstract The ATM inhibitor AZD1390 disrupts cellular responses to ionizing radiation (IR) and is a potent radiosensitizer being tested in clinical trials. Here, we evaluated, the effects of AZD1390 on radiation sensitivity and DNA damage repair pathways in glioblastoma (GBM) cells and patient derived xenografts (PDXs). AZD1390 (30 nM and higher) suppressed IR (5 Gy)-induced phosphorylation of ATM-Serine1981 and downstream phosphorylation sites on Kap1, Chk2 and H2AX in U251 cells and multiple PDXs. Consistent with enhanced DNA damage, AZD1390 increased IR induced G2/M arrest in U251 (80.6% with AZD1390/IR vs. 64.6% with IR, p= 0.01), GBM43 (61.9% vs. 25.7%, p= 0.01) and GBM39 (40.9% vs. 25.4%, p= 0.01). Moreover, in a clonogenic survival assay, AZD1390 sensitized U251 cells to 5 Gy IR (0.24% survival with AZD1390/IR vs. 2.3% with IR alone, p=0.01). In a reporter-based analysis of DNA repair capacity, ATM inhibition resulted in a 40 to 60% reduction in homologous recombination (HR) and modest but significant decrease in micro-homology mediated end joining (MMEJ) and gap fill-in synthesis in U251 cells but had no effect on non-homologous end joining, translesion synthesis, nucleotide or base excision repair pathways. Comparing effects of AZD1390 on repair in GBM14 (TP53-wt) and GBM43 (TP53-mutant), similar results were observed except that decreased MMEJ was seen only in GBM43 (0.03± 0.01% vs. 0.07± 0.01% in control, p=0.002). Intriguingly, RAD51 knockdown (to disrupt HR) sensitized U251 and GBM43 but not GBM14. The efficacy of AZD1390 ± IR was studied in vivo in 10 PDXs. IR was delivered to orthotopic tumors using opposed lateral 225 kVp beams. AZD1390 (20 mg/kg PO) was given just prior to each radiation dose (2 Gy x 5 fractions). AZD1390 monotherapy was mostly ineffective, IR alone was reasonably efficacious with an average 1.8 ± 0.1-fold-increase in survival relative to sham radiation (survival ratio) across all 10 models. IR/AZD1390 treatment resulted in significant survival extension relative to IR alone in 6 of 10 models. Analysis of the survival benefit of combination therapy compared to IR alone across the entire cohort of PDXs was statistically marginal (average survival ratio 2.3± 0.3 vs. 1.8 ± 0.1 with IR, p=0.08). However, when stratified by TP53 status, combination therapy was significantly more effective than IR (mean survival ratio 2.3 ± 0.1 vs. 1.6 ± 0.2 with IR alone, p=0.02) in TP53-mutant PDXs, where all 5 models benefited. In contrast, TP53-wt group had no benefit (mean survival ratio 2.2 ± 0.5 vs. 2.0 ± 0.2, p=0.61), GBM39 was only TP53-wt PDX that benefited from the combination. In conclusion, AZD1390 is an effective radio-sensitizer that causes disruption in HR and potentially other DNA repair pathways. Interestingly, in vivo radiosensitizing effects are mostly restricted to TP53-mutant GBM PDXs. Understandingmechanism of resistance in the context of different TP53 backgrounds remains an important future direction. Citation Format: Shiv K. Gupta, Jiajia Chen, Daniel J. Laverty, Surabhi Talele, Brett L. Carlson, Ann C. Mladek Tuma, Danielle Burgenske, Gaspar J. Kitange, Petra Hamerlik, Zachary D. Nagel, Stephen T. Durant, William F. Elmquist, Jann N. Sarkaria. AZD1390 radio-sensitizes p53-mutant GBM via disrupting homology directed DNA repair [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2598.

Abstract Background Despite aggressive upfront treatment in glioblastoma (GBM), recurrence remains inevitable for most patients. Accumulating evidence has identified hypermutation induced by temozolomide (TMZ) as an emerging subtype of recurrent GBM. However, its biological and therapeutic significance has yet to be described. Methods We combined GBM patient and derive GBM stem cells (GSCs) from tumors following TMZ to explore response of hypermutant and non-hypermutant emergent phenotypes and explore the immune relevance of hypermutant and non-hypermutant states in vivo. Results Hypermutation emerges as one of two possible mutational subtypes following TMZ treatment in vivo and demonstrates distinct phenotypic features compared to non-hypermutant recurrent GBM. Hypermutant tumors elicited robust immune rejection in subcutaneous contexts which was accompanied by increased immune cell infiltration. In contrast, immune rejection of hypermutant tumors were stunted in orthotopic settings where we observe limited immune infiltration. Use of anti-PD-1 immunotherapy showed that immunosuppression in orthotopic contexts was independent from the PD-1/PD-L1 axis. Finally, we demonstrate that mutational burden can be estimated from DNA contained in extracellular vesicles (EVs). Conclusion Hypermutation post-TMZ are phenotypically distinct from non-hypermutant GBM and requires personalization for appropriate treatment. The brain microenvironment may be immunosuppressive and exploration of the mechanisms behind this may be key to improving immunotherapy response in this subtype of recurrent GBM.

Abstract Background RBBP4 activates transcription by histone acetylation, but the partner histone acetyltransferases are unknown. Thus, we investigated the hypothesis that RBBP4 interacts with p300 in a complex in glioblastoma (GBM). Methods shRNA silencing of RBBP4 or p300 and RNAseq was used to identify genes co-regulated by RBBP4 and p300 in GBM43 patient-derived xenograft (PDX). RBBP4/p300 complex was demonstrated using proximity ligation assay (PLA) and ChIPseq delineated histone H3 acetylation and RBBP4/p300 complex binding in promoters/enhancers. Temozolomide (TMZ)-induced DNA double strand breaks (DSBs) were evaluated by γ-H2AX and proliferation by CyQuant and live cell monitoring assays. In vivo efficacy was based on survival of mice with orthotopic tumors. Results shRBBP4 and shp300 downregulated 4768 genes among which 1485 (31%) were commonly downregulated by both shRNAs, while upregulated genes were 2484, including 863 (35%) common genes. The pro-survival genes were the top-ranked among the downregulated genes, including C-MYC. RBBP4/p300 complex was demonstrated in the nucleus, and shRBBP4 or shp300 significantly sensitized GBM cells to TMZ compared to the control shNT in vitro (P < .05). Moreover, TMZ significantly prolonged the survival of mice bearing GBM22-shRBBP4 orthotopic tumors compared with control shNT tumors (median shNT survival 52 days vs. median shRBBP4 319 days; P = .001). CREB-binding protein (CBP)/p300 inhibitor CPI-1612 suppressed H3K27Ac and RBBP4/p300 complex target proteins, including C-MYC, and synergistically sensitized TMZ in vitro. Pharmacodynamic evaluation confirmed brain penetration by CPI-1612 supporting further investigation to evaluate efficacy to sensitize TMZ. Conclusions RBBP4/p300 complex is present in GBM cells and is a potential therapeutic target.

Abstract Isocitrate dehydrogenase 1/2 (IDH1/2) mutations are common in astrocytic glioma and are frequently coupled with TP53 and ATRX mutations. Collectively, these alterations cause genomic instability leading to high basal DNA double strand breaks (DSBs). Understanding how IDH/TP53/ATRX mutant cells process endogenous DSBs may help exploit inhibitors of DNA damage response (DDR) for the treatment of patients with IDH mutant gliomas. Through systematic effort to uncover the mechanisms involved in repair of endogenous DSBs in IDH1/2 mutant GBMs, we have discovered that high basal phosphorylated DNA-PK (p-DNA-PK) was characteristic of an IDH1/TP53/ATRX mutant GBM164 patient derived xenograft (PDX) but not in another IDH1 mutant GBM196 PDX. Immunofluorescence (IF) studies in patient specimen from which GBM164 was derived showed that p-DNA-PK co-localized with g-H2AX, 53BP1 or H4K20me2 (but not p-RPA) the known surrogates of DSBs. In contrast, p-DNA-PK was absent in the patient specimen from which GBM196 was derived, which otherwise had equally intense g-H2AX immunostaining colocalized with p-RPA. An independent IF study involving 11 IDH1 wild-type (WT) and 11 IDH1 mutant GBM patient samples, the p-DNA-PK was observed in 3 (27%) of 11 IDH1 mutant samples while IDH1 WT tumors were negative for p-DNA-PK. A telomere specific fluorescence in situ hybridization (Tel-FISH) confirmed elevated alternative lengthening of telomere (ALT) activity in GBM196 (but not in GBM164) indicative of HR proficiency. Consistently, HR related genes, including BRCA1 and MRE11A, were found upregulated in ALT-positive GBM196 as compared to those in GBM164. Interestingly, ALT+ GBM196 cells were highly vulnerable to inhibitors of ATM and ATR pathways. In conclusion, IDH1/TP53/ATRX mutant gliomas can be subdivided into HR-mediated ALT-positive group, which repairs the endogenous DSBs by HR (e.g. GBM196) and an ALT-negative/p-DNA-PK group, which repairs DSBs by c-NHEJ (e.g. GBM164) and this subdivision can be developed as a prescient biomarker of sensitivity to DDR inhibitors.

Abstract Background Glioblastoma (GBM) is an incurable disease with few approved therapeutic interventions. Radiation therapy (RT) and temozolomide (TMZ) remain the standards of care. The efficacy and optimal deployment schedule of the orally bioavailable small-molecule tumor checkpoint controller lisavanbulin alone, and in combination with, standards of care were assessed using a panel of IDH-wildtype GBM patient-derived xenografts. Methods Mice bearing intracranial tumors received lisavanbulin +/−RT +/−TMZ and followed for survival. Lisavanbulin concentrations in plasma and brain were determined by liquid chromatography with tandem mass spectrometry, while flow cytometry was used for cell cycle analysis. Results Lisavanbulin monotherapy showed significant benefit (P < .01) in 9 of 14 PDXs tested (median survival extension 9%-84%) and brain-to-plasma ratios of 1.3 and 1.6 at 2- and 6-hours postdose, respectively, validating previous data suggesting significant exposure in the brain. Prolonged lisavanbulin dosing from RT start until moribund was required for maximal benefit (GBM6: median survival lisavanbulin/RT 90 vs. RT alone 69 days, P = .0001; GBM150: lisavanbulin/RT 143 days vs. RT alone 73 days, P = .06). Similar observations were seen with RT/TMZ combinations (GBM39: RT/TMZ/lisavanbulin 502 days vs. RT/TMZ 249 days, P = .0001; GBM26: RT/TMZ/lisavanbulin 172 days vs. RT/TMZ 121 days, P = .04). Immunohistochemical analyses showed a significant increase in phospho-histone H3 with lisavanbulin treatment (P = .01). Conclusions Lisavanbulin demonstrated excellent brain penetration, significant extension of survival alone or in RT or RT/TMZ combinations, and was associated with mitotic arrest. These data provide a strong clinical rationale for testing lisavanbulin in combination with RT or RT/TMZ in GBM patients.

Abstract Background Appropriately designed preclinical patient-derived xenograft (PDX) experiments are important to accurately inform human clinical trials. There is little experimental design guidance regarding choosing the number of PDX lines to study, and the number of mice within each PDX line. Methods Retrospective data from IDH-wildtype glioblastoma preclinical experiments evaluating a uniform regimen of fractionated radiation (RT), temozolomide (TMZ) chemotherapy, and concurrent RT/TMZ across 27 PDX lines were used to evaluate experimental designs and empirically estimate statistical power for ANOVA and Cox regression. Results Increasing the number of PDX lines resulted in more precise and reproducible estimates of effect size. To achieve 80% statistical power using ANOVA, experiments using a single PDX line required subsampling of 6 mice per PDX for each treatment group to detect a difference in survival of 135 days, and 9 mice per PDX to detect a difference of 100 days. Alternatively, a design that used 10 PDX lines had greater than 80% power to detect a difference of 135 days with a single mouse per PDX per treatment group, a difference of 100 days with 2 mice per PDX per treatment, and 35 days with more than 10 mice per PDX per treatment. Power for Cox regression was slightly smaller than ANOVA for very small experiments regardless of effect size and slightly higher than ANOVA for detecting a smaller effect size of 35 days difference in survival for moderate-to-large experiments. Conclusions Experimental designs using few mice across many PDX lines can provide robust results and account for inter-tumor variability.

Abstract Background Antibody drug conjugates (ADCs) targeting the epidermal growth factor receptor (EGFR), such as depatuxizumab mafodotin (Depatux-M), is a promising therapeutic strategy for glioblastoma (GBM) but recent clinical trials did not demonstrate a survival benefit. Understanding the mechanisms of failure for this promising strategy is critically important. Methods PDX models were employed to study efficacy of systemic vs intracranial delivery of Depatux-M. Immunofluorescence and MALDI-MSI were performed to detect drug levels in the brain. EGFR levels and compensatory pathways were studied using quantitative flow cytometry, Western blots, RNAseq, FISH, and phosphoproteomics. Results Systemic delivery of Depatux-M was highly effective in nine of 10 EGFR-amplified heterotopic PDXs with survival extending beyond one year in eight PDXs. Acquired resistance in two PDXs (GBM12 and GBM46) was driven by suppression of EGFR expression or emergence of a novel short-variant of EGFR lacking the epitope for the Depatux-M antibody. In contrast to the profound benefit observed in heterotopic tumors, only two of seven intrinsically sensitive PDXs were responsive to Depatux-M as intracranial tumors. Poor efficacy in orthotopic PDXs was associated with limited and heterogeneous distribution of Depatux-M into tumor tissues, and artificial disruption of the BBB or bypass of the BBB by direct intracranial injection of Depatux-M into orthotopic tumors markedly enhanced the efficacy of drug treatment. Conclusions Despite profound intrinsic sensitivity to Depatux-M, limited drug delivery into brain tumor may have been a key contributor to lack of efficacy in recently failed clinical trials.

Abstract Radio-resistance mechanisms limit the benefit of radiation therapy (RT) for melanoma brain metastases. A key pathway for radiation-induced DNA double-strand break repair is non-homologous end joining, which is critically mediated by DNA-dependent protein kinase (DNA-PKcs). Here we evaluated radio-sensitizing effects of M3814, a selective oral inhibitor of DNA-PKcs, in patient-derived xenografts (PDXs) of melanoma brain metastases. In a clonogenic survival assay, M3841 augmented RT-induced killing of M12 cells at concentrations of ≥300 nM, and a minimum of 16 h exposure with ~300 nM M3814 was required for effective sensitization. M3814 inhibited RT-induced (5 Gy) auto-phosphorylation of serine-2056 of DNA-PKcs in primary cultures of M12, M15 and M27 PDX lines. Interestingly, inhibition of RT-induced DNA-PKcs by M3814 coincided with increased KAP1 phosphorylation, a DNA damage signaling regulated via ATM. Persistent γH2AX foci were observed in 28% M12 cells at 24 hours after co-treatment with M3814 and RT as compared to 12% cells following RT alone. In vivo pharmacokinetic analysis after single oral dose of 20 mg/kg M3814, showed reasonably short half-life (~2.44 hours) and poor brain distribution in wild-type FBV mice (Kpuu, 0.027). Consistent with an efflux liability, brain distribution of M3814 in triple knockout mice for BCRP/MDR1A/B was ~11 fold higher (Kpuu, 0.215). Compared to normal brain, much higher M3814 concentrations were detected in intracranially implanted M12 tumors (~23 fold and ~20 fold) 2 and 6 hours after a single oral dose of 50mg/kg respectively. The relative exclusion of M3814 from normal brain as compared to brain metastases suggests that this drug may have a favorable toxicity profile when combined with radiation for treatment of melanoma brain metastases, and this hypothesis is being tested in ongoing efficacy studies.