Jain, Sonia

Temozolomide (TMZ) remains the cornerstone of chemotherapeutic agent for glioblastoma (GBM), yet intrinsic and acquired resistance severely limits its clinical benefit. While O6-methylguanine-DNA methyltransferase (MGMT)–mediated repair of TMZ-induced O6-methylguanine (O6-meG) lesions has been extensively studied, the DNA mismatch repair (MMR) pathway is increasingly recognized as a key determinant of TMZ cytotoxicity. Canonical MMR, mediated by MutSα (MSH2–MSH6) and MutLα (MLH1–PMS2) complexes, recognizes O6-meG: thymine mispairs generated during replication and initiates futile repair cycles that culminate in replication stress, replication fork collapse, and apoptotic signaling; intact canonical MMR is, therefore, required for TMZ-induced cell death. Disruption of canonical MMR, frequently via acquired MSH6 mutations, confers TMZ tolerance and drives hypermutated recurrent GBM. Beyond mismatch correction, MMR proteins perform non-canonical functions in DNA damage signaling, replication stress responses, transcriptional regulation, chromatin dynamics, and immune modulation. These activities may shift the outcome from cytotoxic futile repair toward replication stress adaptation, Translesion synthesis (TLS)-mediated lesion tolerance, immune remodeling, and therapeutic resistance. Notably, partial attenuation or functional diversion of MMR may decouple lesion recognition from cytotoxic signaling, enabling TLS-mediated lesion tolerance without complete loss of MMR activity. This review integrates current insights into canonical and non-canonical MMR functions in GBM, defines their distinct contributions to TMZ sensitivity and resistance, and highlights therapeutic opportunities to exploit MMR-associated dependencies, including synthetic lethal strategies and immunotherapeutic vulnerabilities linked to MMR deficiency-driven hypermutation.

Abstract Glioblastoma (GBM) is an aggressive, high-grade glioma with a near-universally fatal prognosis. Therapeutic failure is often attributed to the highly selective blood brain barrier (BBB), the diffuse infiltrative nature of the tumor, and the marked intratumoral heterogeneity of GBM. Although antibody drug conjugates ADCs have shown promise for high grade gliomas such as GBM, efficacy is limited by ADC size. Aptamers—short, synthetic, single-stranded DNA or RNA molecules—can be ∼6-fold lower in molecular weight than IgG antibodies and have the potential to cross the intact BBB. Unlike other nucleic acid-based therapies, aptamer function arises from three-dimensional shape rather than genetic coding. Here we aim to replace the targeting component of the ADC paradigm with a DNA aptamer, thus creating an aptamer-drug conjugate (ApDC). We employed in vivo SELEX using an orthotopic patient-derived xenograft (PDX) GBM mouse model and a vast (∼100 trillion 80-mer sequences) ApDC library. We report the results from this first in vivo ApDC selection of its kind. We characterize target tissue binding ex vivo , cell association, biodistribution, and pharmacokinetics from this selection. This study exemplifies an unbiased approach to a problem that rational design has yet to overcome, offering a new direction for GBM therapeutic development.

Abstract Background Murine double minute 2 (MDM2) inhibitors reactivate wild-type p53 and are a promising therapy for glioblastoma, IDH-wildtype (GBM). Brigimadlin is a highly potent MDM2 inhibitor being tested in a phase 0/1 clinical trial in combination with radiation in GBM. Methods Brigimadlin pharmacokinetics, pharmacodynamics, and efficacy were evaluated in GBM patient-derived xenografts (PDXs). Results In vitro, brigimadlin impaired viability in TP53 wild-type GBM with an IC50 of 0.8-6.6 nmol/L, but sensitivity did not correlate with MDM2 amplification. In vivo, MDM2 amplification was highly correlated with efficacy. In subcutaneous PDXs, 1 or 2 mg/kg brigimadlin dosed weekly was highly effective in 2 MDM2-amplified PDXs. At 2 mg/kg, brigimadlin delayed tumor regrowth by >5-fold in the MDM2-amplified PDXs compared to 1.5-fold in a non-amplified PDX. In orthotopic PDXs, efficacy was more limited, but 2 mg/kg brigimadlin enhanced the response to fractionated radiation in MDM2-amplified PDXs. Consistent with blood-brain barrier efflux limiting drug distribution, 2 mg/kg brigimadlin extended survival by >5-fold in an MDM2-amplified orthotopic PDX established in Rag1−/−Abcb1a−/− Abcg2−/− mice. In pharmacodynamic studies, p53 target genes were upregulated at both subtherapeutic and therapeutic dose levels, and the extent of activation did not correlate with MDM2 status. Concentrations of brigimadlin in tumor tissue were approximately 10-fold higher in MDM2-amplified tumors, and intracellular drug levels directly correlated with drug-dependent MDM2 upregulation, suggesting target binding affects drug accumulation. Conclusions Brigimadlin is highly effective in MDM2-amplified GBM when adequate drug levels are achieved in tumor tissue. MDM2 amplification impacts both treatment efficacy and intratumoral drug accumulation.

Abstract Purpose: Navtemadlin is a potent small-molecule inhibitor of MDM2, which has completed a phase 0 window-of-opportunity study in glioblastoma (GBM). To optimally interpret the clinical data, a detailed analysis of navtemadlin pharmacokinetics (PK), pharmacodynamics, and efficacy was performed in GBM patient-derived xenografts (PDX). Experimental Design: Response to navtemadlin was characterized in vitro and in vivo in GBM PDXs with and without MDM2 amplification. Efficacy in vivo was integrated with measured plasma and intra-tumoral drug levels to develop a translational PK/efficacy model comparing exposure effective in PDX to exposure achieved in phase 0 patient samples. Results: In vitro, navtemadlin showed robust on-target activity in TP53–wild-type GBM. In vivo efficacy strongly correlated with MDM2 amplification status. In subcutaneous PDXs, navtemadlin significantly extended survival when dosed at 25 mg/kg in an MDM2-amplified PDX compared with 100 mg/kg in a non-amplified PDX. Central nervous system distribution was limited by blood–brain barrier efflux (Kp_brain = 0.009). In an MDM2-amplified orthotopic PDX model, navtemadlin was ineffective at 100 mg/kg; when established in mice with deficient blood–brain barrier efflux (Rag−/−Abcb1a−/−Abcg2−/−), 25 mg/kg doubled survival. A tumor PK/efficacy model was built to define target exposure for efficacy in GBM, using the effective 25 mg/kg dose. Modeled exposures exceeded this threshold in three (of 16) tumor samples from phase 0 study patients at the 240 mg dose level. Conclusions: Navtemadlin efficacy was highly dependent on adequate brain penetration. Our translational PK/efficacy model suggests that the minimum effective tumor exposures were achieved only in a minority of patients with GBM.

Temozolomide (TMZ) is the main chemotherapy for GBM, but resistance develops in most patients, highlighting the need for sensitizing strategies. PARP inhibitors (PARPi), which disrupt DNA repair, are being investigated as TMZ-sensitizing agents. Here we evaluated the efficacy and explored molecular mechanisms of pamiparib-mediated TMZ sensitization. In vitro, low concentrations of pamiparib (0.1-1.0 μM) synergized with TMZ and enhanced DNA damage, particularly during the S- and G2/M phases. In vivo, pamiparib monotherapy (0.75 or 3 mg/kg, twice daily) was ineffective, however, TMZ afforded survival benefit in orthotopic patient-derived xenograft (PDX) models (GBM12, GBM22, and GBM84) of MGMT-methylated primary GBM. Pamiparib combined with TMZ (25 or 50 mg/kg daily) prolonged survival in GBM12 [median survival 272 vs. 114 days with TMZ (P=0.02), 30 days with placebo (P<0.001)] and GBM22 (median survival 88 vs. 60 days with TMZ (P=0.05), 26 days with placebo, (P<0.001)] but not in GBM84 [median survival 355 vs. 404 days with TMZ (P=0.44), 37 days with placebo (P<0.001)]. These findings align with prior studies indicating that PARPi can potentiate TMZ efficacy in a subset of primary GBM models sensitive to TMZ. TMZ induces cytotoxic DNA adducts, N7MeG, N3MeA, and O6MeG. While N7MeG and N3MeA are repaired by base-excision repair (BER), MGMT removes O6MeG. Suppressed MGMT expression, due to promoter methylation, drives TMZ sensitivity in ∼40% of primary GBM cases. In MGMT-methylated tumors, O6MeG-thymine mispairs trigger futile mismatch repair, causing replication stress. Although BER is PARP-dependent, BER inhibition did not replicate the TMZ-sensitizing effects of pamiparib. Beyond DNA repair, PARP regulates replication stress by recruiting translocases like SMARCAL1 and controlling endonucleases such as MRE11 and EXO1, protecting stressed forks and nascent DNA. We targeted these components using siRNA and small-molecule inhibitors to elucidate their roles in TMZ sensitization. Unexpectedly, silencing SMARCAL1 or MRE11, or inhibiting MRE11 with mirin, did not affect pamiparib-mediated sensitization in U251 cells. While EXO1 silencing attenuated DNA damage signaling, it had minimal impact on TMZ sensitivity or pamiparib-mediated sensitization. These results suggest that fork protection mechanisms may be epistatic to PARP but redundant for TMZ sensitivity. Alternatively, replication stress may be mitigated by lesion bypass via translesion synthesis (TLS), where replicative polymerases are replaced with low-fidelity polymerases. Consistent with this, silencing RAD18, involved in TLS, enhanced sensitivity to both TMZ and pamiparib, with the combination being more effective than either drug alone. In conclusion, pamiparib enhances TMZ efficacy in a subset of primary GBM. While further in vivo validation is needed, TLS deficiency may be a key determinant of pamiparib-mediated sensitization. Citation Format: Shiv K. Gupta, Ann C. Mladek, Sonia Jain, Katrina K. Bakken, Zeng HU, Brett L. Carlson, Danielle M. Burgenske, Jann N. Sarkaria. Pamiparib-induced replication stress augments the efficacy of temozolomide in GBM PDX models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2910.

Abstract Background and Objective: Glioblastoma (GBM) is most common primary brain tumor with dire need of novel therapeutics. Studies with intracranial tumors are required for robust characterization and development of novel drugs for brain cancer. Tumor harvest and processing is an important aspect to study pharmacokinetics, pharmacodynamics, drug delivery, and tumor evolution. Typically, delineating and dissecting tumor from surrounding brain for varied downstream processes is dependent on the distinct morphology of the tumor and/or expression of fluorescent marker. However, genetic modification of the tumor cells for the expression of fluorescent protein is not always possible, especially for patient-derived xenograft (PDX) tumor models. Thus, we have developed a dye-based staining method to confidently visualize tumor in the brain and assessed usability of isolated tumors for drug and DNA analysis. Methodology: Whole brains were harvested from mice with intracranial Mayo GBM PDX tumors and stored at -80oC freezer. Few days prior to staining, brains were transferred to -20oC freezer. Brain was cut into 1 mm thick coronal sections and transferred onto glass slides. Slices were covered with toluidine blue (TB) solution (0.05 to 10 mg/mL) for 45 s. Excess stain solution was removed, and slices were observed under white light. Regions of tumor became more transparent, while normal brain remained opaque. This enabled dissection of tumor from normal brain for downstream analyses. Results: TB concentration (0.05 to 10 mg/mL) was optimized to stain 1 mm thick sections of fresh frozen brain, and the best contrast between tumor to normal brain was observed at 0.1-0.3 mg/mL. TB aided differential visualization of tumors in several PDX models - GBM12, GBM43, GBM39 and GBM108 - suggesting broad applicability. To evaluate the accuracy of TB staining, brain with PDX tumor transduced with pSINeGFP/fLuc2 was observed by both TB staining and eGFP fluorescence. This analysis demonstrated excellent concordance across 6 animals with three distinct PDX models. Genomic DNA was isolated from tumor dissected under TB guidance and analyzed by PCR for primers specific for murine vs human PTGER2. 50-70% of the DNA in tumor samples was detected to be of human origin while normal brain tissue had 90-95% of the mouse DNA highlighting the specificity of TB stain for tumor detection. Tumors from drug treated GBM108 PDX were isolated under TB guidance and comparable drug levels were detected by LC-MS/MS as analyzed in GFP-assisted tumor samples. Conclusion: Staining fresh frozen thick brain sections with toluidine blue enables visualization of tumor from normal brain regions that is compatible with varied downstream processes. Development of such staining assays can help faster and easier tumor processing and is not dependent on genetic modification of tumor cells prior to experimentation. Citation Format: Sonia Jain, Anders Herfindahl-Quint, Caroline D. Doherty, Wenjuan Zhang, Danielle M. Burgenske, Jann N. Sarkaria. Novel staining method for visualization, isolation and processing of intracranial brain tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2531.

Abstract Background The ataxia-telangiectasia mutated (ATM) protein coordinates the cellular response to therapeutic radiation, and ATM inhibitors can potentially enhance the efficacy of radiation in otherwise radiation-resistant tumors. Methods The small molecule ATM kinase inhibitor WSD0628 was specifically designed for enhanced distribution across the blood-brain barrier to more effectively treat glioblastoma (GBM) and brain metastases in combination with radiation therapy (RT). GBM and brain metastasis patient-derived xenograft models were used to understand target inhibition, radiosensitization, inhibition of the DNA damage response, and in vivo efficacy. Results Initial in vitro characterization of WSD0628 demonstrates a high level of selectivity across kinase families, limited aldehyde oxidase liability, and low risk of hERG interactions. Consistent with a central role for ATM in radiation response, WSD0628 blocked radiation-induced signaling and enhanced radiosensitivity in U251 glioma cells and brain tumor PDXs GBM120 and M12. In comparison to control or RT alone in orthotopic PDXs, the combination of WSD0628 with RT markedly prolonged median survival—GBM12 (19, 55, and 408 days, respectively); GBM43 (26, 44, and 143 days, respectively); GBM120 (51, 89, and 231 days, respectively); M12 (17, 39, and 190 days, respectively). Pharmacokinetic and pharmacodynamic testing after treatment in orthotopic GBM43 tumors showed inhibitory levels of WSD0628 and a reduction of γH2AX foci in the combination-treated tumors. Conclusions Collectively, these results suggest a promising role for WSD0628 in combination with RT in brain tumors and provide the rationale for an ongoing Phase 0/1A clinical trial testing this combination in recurrent GBM.

Abstract Glioblastoma (GBM) is the most common and aggressive primary malignant brain tumor of adults. Current therapeutic options yield dismal prognoses that have remained essentially unchanged over nearly two decades. Diffuse growth patterns, high intratumoral heterogeneity, and variable blood–brain barrier integrity limit treatment efficacy, creating challenges that rational small molecule design has not overcome. Antibody–drug conjugates have shown some promise, leading us to hypothesize that smaller folded DNA aptamers, developed in vivo via principles of natural selection, might eventually have advantages for drug delivery. Here, we document the first in vivo DNA aptamer selection involving an orthotopic patient-derived xenograft GBM mouse model to identify tumor-homing DNA aptamers. We demonstrate the preferential accumulation of these aptamers in the tumor relative to other tissues 4 h after intraperitoneal injection. The aptamers can be detected by quantitative polymerase chain reaction, fluorescent tumor staining, and stain GBM tumor section from untreated mice and the GBM tumor cells in culture. Two of three candidates are selective for the target cell line in vitro and do not bind other human tumor cells. In vivo selection of tumor-specific DNA aptamers demonstrates a novel approach for diagnostics or toxin delivery that might allow for the development of individualized therapies.

2012 Background: Navtemadlin (nvtm) is a potent, selective, orally available small molecule inhibitor of murine double minute 2 homologue (MDM2), which has completed a Ph 0 surgical window of opportunity study in glioblastoma, IDH-wildtype (GBM; Lee SNO 2022). To optimally interpret the clinical data, an extensive evaluation of nvtm PK, PD, and efficacy was performed in GBM patient derived xenografts (PDXs). Methods: Response to nvtm +/- radiotherapy (RT) was characterized in vitro and in vivo across a panel of GBM PDXs with various TP53 and MDM2 alterations. Efficacy, PK, and PD were evaluated in flank or orthotopic PDX models. p53 pathway activation was evaluated by qPCR, Western blot, and IHC. Nvtm brain penetration was evaluated in immunodeficient abcb1a/b-/- (P-glycoprotein; P-gp), abcg2-/- (BCRP), or double-knockout mice. Binding was assessed by rapid equilibrium dialysis, and nvtm concentrations by LC/MS-MS and MALDI-MSI. Results: Nvtm decreased in vitro viability of TP53WT GBM PDX explant cultures with IC50s <100 nM, but was ineffective in TP53-mutant GBM. In flank PDX models, nvtm (100 mg/kg daily for 4 weeks) delayed tumor regrowth by 1.9 to 4.8-fold in TP53 WT PDXs without MDM2amplification (n=9 models), and more markedly by 20.1 to 29.4-fold in MDM2amplified models (n=4; 2 PDXs did not recur). In a flank PDX dose-response study, nvtm delayed the growth of MDM2amplified GBM108 at 25 mg/kg vs. placebo (61 vs. 11 days; p=0.014; mean intra-tumoral nvtm = 590 nM). In contrast, growth of MDM2 non-amplified GBM14 was delayed only at 100 mg/kg (37 vs. 15 days; p=0.082; mean intra-tumoral nvtm = 2990 nM). In both flank models, p53 target gene upregulation was similar between 25 and 100 mg/kg and did not correlate with regrowth. Nvtm was highly bound (fraction unbound = 0.090 in culture media and 0.014 in GBM tissue homogenate). In mice after oral dosing, distribution to CNS was limited by P-gp, with a brain to plasma ratio of 0.009 and unbound ratio of 0.005, whereas BCRP deficiency did not affect CNS distribution. With orthotopic GBM108, nvtm efficacy was evaluated in nude vs. efflux-deficient mice. In nude mice, nvtm was ineffective at doses up to 100 mg/kg; however, in efflux deficient mice, the mean intra-tumoral nvtm level at 25 mg/kg was similar to flank tumors (690 nM) and survival was significantly extended (134 vs. 62 days, p=0.0002) with additional benefit at 100 mg/kg (>300 vs. 62 days, p=0.002). Conclusions: Nvtm showed significant efficacy in TP53 WT GBM flank PDX models but efficacy in orthotopic PDX was highly dependent on CNS penetration. Targeting nvtm levels based on the in vitroIC50 may significantly underestimate the in vivo effective concentration . In vivo studies showed lower intra-tumoral nvtm levels are required to significantly inhibit growth of MDM2amplified compared with non-amplified tumors.

Abstract Purpose: Antibody-drug conjugates (ADC) are targeted therapies with robust efficacy in solid cancers, and there is intense interest in using EGFR-specific ADCs to target EGFR-amplified glioblastoma (GBM). Given GBM’s molecular heterogeneity, the bystander activity of ADCs may be important for determining treatment efficacy. In this study, the activity and toxicity of two EGFR-targeted ADCs with similar auristatin toxins, Losatuxizumab vedotin (ABBV-221) and Depatuxizumab mafodotin (Depatux-M), were compared in GBM patient-derived xenografts (PDX) and normal murine brain following direct infusion by convection-enhanced delivery (CED). Experimental Design: EGFRviii-amplified and non-amplified GBM PDXs were used to determine in vitro cytotoxicity, in vivo efficacy, and bystander activities of ABBV-221 and Depatux-M. Nontumor-bearing mice were used to evaluate the pharmacokinetics (PK) and toxicity of ADCs using LC-MS/MS and immunohistochemistry. Results: CED improved intracranial efficacy of Depatux-M and ABBV-221 in three EGFRviii-amplified GBM PDX models (Median survival: 125 to &gt;300 days vs. 20–49 days with isotype control AB095). Both ADCs had comparable in vitro and in vivo efficacy. However, neuronal toxicity and CD68+ microglia/macrophage infiltration were significantly higher in brains infused with ABBV-221 with the cell-permeable monomethyl auristatin E (MMAE), compared with Depatux-M with the cell-impermeant monomethyl auristatin F. CED infusion of ABBV-221 into the brain or incubation of ABBV-221 with normal brain homogenate resulted in a significant release of MMAE, consistent with linker instability in the brain microenvironment. Conclusions: EGFR-targeting ADCs are promising therapeutic options for GBM when delivered intratumorally by CED. However, the linker and payload for the ADC must be carefully considered to maximize the therapeutic window.