Newly-diagnosed GBM Research Articles

Global post‑marketing safety surveillance of Tumor Treating Fields (TTFields) therapy in over 25,000 patients with CNS malignancies treated between 2011–2022

BackgroundTumor Treating Fields (TTFields) are alternating electric fields that disrupt cancer cell processes. TTFields therapy is approved for recurrent glioblastoma (rGBM), and newly-diagnosed (nd) GBM (with concomitant temozolomide for ndGBM; US), and for grade IV glioma (EU). We present an updated global, post-marketing surveillance safety analysis of patients with CNS malignancies treated with TTFields therapy.MethodsSafety data were collected from routine post-marketing activities for patients in North America, Europe, Israel, and Japan (October 2011–October 2022). Adverse events (AEs) were stratified by age, sex, and diagnosis.ResultsOverall, 25,898 patients were included (diagnoses: ndGBM [68%], rGBM [26%], anaplastic astrocytoma/oligodendroglioma [4%], other CNS malignancies [2%]). Median (range) age was 59 (3–103) years; 66% patients were male. Most (69%) patients were 18–65 years; 0.4% were < 18 years; 30% were > 65 years. All-cause and TTFields-related AEs occurred in 18,798 (73%) and 14,599 (56%) patients, respectively. Most common treatment-related AEs were beneath-array skin reactions (43%), electric sensation (tingling; 14%), and heat sensation (warmth; 12%). Treatment-related skin reactions were comparable in pediatric (39%), adult (42%), and elderly (45%) groups, and in males (41%) and females (46%); and similar across diagnostic subgroups (ndGBM, 46%; rGBM, 34%; anaplastic astrocytoma/oligodendroglioma, 42%; other, 40%). No TTFields-related systemic AEs were reported.ConclusionsThis long-term, real-world analysis of > 25,000 patients demonstrated good tolerability of TTFields in patients with CNS malignancies. Most therapy-related AEs were manageable localized, non-serious skin events. The TTFields therapy safety profile remained consistent across subgroups (age, sex, and diagnosis), indicative of its broad applicability.

Abstract CT044: A phase 0/1b study of AZD1390 plus radiotherapy in newly-diagnosed, MGMT-unmethylated glioblastoma patients

Abstract Background: ATM inhibition has been hypothesized to potentiate the effects of radiation by preventing acute phase DNA damage repair, and a Phase 1 trial of this combination in patients with GBM is ongoing (NCT03423628). To test this hypothesis, this Phase 0/1b study evaluated the pharmacokinetics (PK) and pharmacodynamics (PD) of AZD1390, a first-in-class ATM inhibitor, in combination with radiation in newly-diagnosed GBM patients (NCT05182905). Methods: Presumed newly-diagnosed GBM patients received 3 days of AZD1390 prior to planned resection at 4-6 or 23-25 hours following the final dose. Tumor tissue, cerebrospinal fluid (CSF), and plasma were collected. Total and unbound drug concentrations were measured using validated LC-MS/MS methods. A PK ‘trigger’ determined eligibility for the therapeutic expansion phase. ATM inhibition was assessed via ex vivo tissue radiation, quantifying pRAD50 levels compared to non-radiated control. Patients exceeding the PK threshold (2 nM unbound drug levels in Gd-non-enhancing tumor region) with unmethylated MGMT promoter were eligible for the expansion phase of study drug, combining AZD1390 with 6 weeks of radiotherapy. Results: Seventeen patients were enrolled into the study. All tested patients (n=14) exceeded the PK threshold and qualified for the expansion phase. Three patients were excluded from the PK analysis due to dosing deviation (n=1) and non-GBM diagnosis (n=2). The mean unbound concentration of AZD1390 in Gd-nonenhancing tumor was 12.4 and 4.4 pmol/g, respectively, for the 4-6 and 23-25-hours cohort. Following 5Gy ex vivo radiation of surgical specimen, pRAD50 expression was significantly suppressed in AZD1390 treated patients compared to untreated controls (median 0.2% for 4-6-hours cohorts and 1.9% for 23-25-hours cohort vs 32.3% for control cohort, average 99% and 95% reduction, respectively, p&amp;lt;0.01 for both cohorts vs control). As of 11/30/2023, 3 patients with unmethylated MGMT promoter are undergoing therapeutic dosing of AZD1390 plus radiotherapy, with median clinical follow-up of 2.3 months. Conclusion: AZD1390 is well-tolerated in newly-diagnosed GBM patients, achieving pharmacologically relevant concentrations in Gd-nonenhancing tumor tissue, and substantially suppressing radiation-induced increases in pRAD50 levels. These data are the first in GBM patients demonstrating AZD1390-mediated pharmacodynamic and pharmacokinetic response. Citation Format: Nader Sanai, Siddhi Desai, Tigran Margaryan, Jennifer Molloy, Costanza LoCascio, Mackenna Elliott, Jocelyn Harmon, Amy Hong, Ernesto Luna Melendez, John Wanebo, Kelly Braun, William Kennedy, Michael Garcia, Wonsuk Yoo, Artak Tovmasyan, An-Chi Tien, Shwetal Mehta. A phase 0/1b study of AZD1390 plus radiotherapy in newly-diagnosed, MGMT-unmethylated glioblastoma patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT044.

Efficacy and safety of chlorpromazine as an adjuvant therapy for glioblastoma in patients with unmethylated MGMT gene promoter: RACTAC, a phase II multicenter trial.

Drug repurposing is a promising strategy to develop new treatments for glioblastoma. In this phase II clinical trial, we evaluated the addition of chlorpromazine to temozolomide in the adjuvant phase of the standard first-line therapeutic protocol in patients with unmethylated MGMT gene promoter. This was a multicenter phase II single-arm clinical trial. The experimental procedure involved the combination of CPZ with standard treatment with TMZ in the adjuvant phase of the Stupp protocol in newly-diagnosed GBM patients carrying an unmethylated MGMT gene promoter. Progression-free survival was the primary endpoint. Secondary endpoints were overall survival and toxicity. Forty-one patients were evaluated. Twenty patients (48.7%) completed 6 cycles of treatment with TMZ+CPZ. At 6 months, 27 patients (65.8%) were without progression, achieving the primary endpoint. Median PFS was 8.0 months (95% CI: 7.0-9.0). Median OS was 15.0 months (95% CI: 13.1-16.9). Adverse events led to reduction or interruption of CPZ dosage in 4 patients (9.7%). The addition of CPZ to standard TMZ in the first-line treatment of GBM patients with unmethylated MGMT gene promoter was safe and led to a longer PFS than expected in this population of patients. These findings provide proof-of-concept for the potential of adding CPZ to standard TMZ treatment in GBM patients with unmethylated MGMT gene promoter. https://clinicaltrials.gov/study/NCT04224441, identifier NCT04224441.

CTIM-35. INB-400 PHASE 1B/2 DRUG RESISTANT IMMUNOTHERAPY WITH ACTIVATED, GENE MODIFIED ALLOGENEIC OR AUTOLOGOUS ΓΔ T CELLS IN COMBINATION WITH MAINTENANCE TEMOZOLOMIDE RECURRENT OR NEWLY DIAGNOSED GLIOBLASTOMA

Abstract Gamma delta (gd) T cells are innate immune cells that synergistically target cancer cells via NKG2D ligands (NKG2D-L) upregulated by chemotherapy. INB-200, a Phase 1 trial, demonstrated the safety and efficacy of autologously (auto) derived genetically modified drug resistant immunotherapy (DRI) gd T cells combined with maintenance temozolomide (TMZ) to treat newly diagnosed (ND) patients with GBM. Initiation of , a 3-arm phase 1b/2 study confirming the efficacy of the auto product in ND GBM and the safety and efficacy of an allogeneic (allo) product in both ND and relapsed GBM has begun. Arm A assesses the auto product in ND GBM patients with 1x107 cells delivered on Day 1 of cycles 1-6 along with TMZ maintenance therapy per the Stupp regimen. In the Phase 1b arm, the haploidentical allo product, will be assessed for safety in 6 relapsed GBM patients who will receive 1x107 cells in combination with 150mg/m2 of TMZ on Day 1 of every six 28-day cycles. Arm B will continue enrollment of relapsed patients following the Phase 1b and Arm C will test the safety of the allo derived product in ND, GBM patients. In Arm C, patients will receive 1x107 allo DeltEx DRI cells on Day 1 of cycles 1-6 in combination with TMZ maintenance. Cells are administered through a Rickham catheter in all arms. Primary endpoints for the Phase 1b define the RP2D and subject/ product characteristics to optimize manufacturing, while Phase 2 endpoints will determine the 12 mos OS rate for Arms A &amp; C and 9 mos OS rate for Arm B. Eligible patients are adults with IDH-wt ND or relapsed GBM with no concurrent TTF or immunotherapy, willing to place an indwelling Rickham catheter. Allo arm patients must have a haploidentical donor. Fifteen US centers will conduct study.

Eighteen years' experience with tumor treating fields in the treatment of newly diagnosed glioblastoma.

The prognosis of glioblastoma remains unfavorable. TTFields utilize low intensity electric fields (frequency 150-300 kHz) that disrupt cellular processes critical for cancer cell viability and tumor progression. TTFields are delivered via transducer arrays placed on the patients' scalp. Methods: Between the years 2004 and 2022, 55 patients (20 female), aged 21.9-77.8 years (mean age 47.3±11.8 years; median 47.6 years) were treated with TTFields for newly-diagnosed GBM, and compared to 54 control patients (20 females), aged 27.0-76.7 years (mean age 51.4±12.2 years; median 51.7 years) (p=0.08). All patients underwent gross total or partial resection of GBM. One patient had biopsy only. When available, MGMT promoter methylation status and IDH mutation was detected. Patients on TTFields therapy demonstrated improvements in PFS and OS relative to controls (hazard ratio: 0.64, p=0.031; and 0.61, p=0.028 respectively). TTFields average time on therapy was 74.8% (median 82%): median PFS of these patients was 19.75 months. Seven patients with TTFields usage ≤60% (23-60%, mean 46.3%, median 53%) had a median PFS of 7.95 months (p=0.0356). Control patients with no TTFields exposure had a median PFS of 12.45 months. Median OS of TTF patients was 31.67 months compared to 24.80 months for controls. This is the most extensive study on newly-diagnosed GBM patients treated with TTFields, covering a period of 18 years at a single center and presenting not only data from clinical trials but also a group of 36 patients treated with TTFields as a part of routine clinical practice.

Incidence, molecular characteristics, and imaging features of "clinically-defined pseudoprogression" in newly diagnosed glioblastoma treated with chemoradiation.

Pseudoprogression (PsP) remains an elusive and clinically important, yet ill-defined, phenomena that, generally, involves a period of early radiographic progression (enhancement) followed by a period of radiographic stability or regression. In the current study, we utilized data from the control arm of a phase III clinical trial in newly-diagnosed glioblastoma to explore imaging characteristics of "clinically-defined PsP", or early radiographic progression (PFS < 6months from chemoradiation) followed by a long post-progression residual overall survival (ROS > 12months). One hundred sixty-nine patients with newly-diagnosed GBM from the control arm of the AVAglio trial (NCT00943826) who presented with early radiographic progressive disease (PD) (< 6months) were included. Clinical characteristics, topographical patterns, and radiomic features were compared between newly-diagnosed GBM exhibiting early PD and early death (< 12-month ROS, "true PD") with those exhibiting early PD and a long residual survival (> 12-month ROS, "clinically-defined PsP"). "Clinically-defined PsP" occurred to 38.5% of patients with early PD, and was more associated with MGMT methylation (P = 0.02), younger age (P = 0.003), better neurological performance (P = 0.01), and lower contrast-enhancing tumor volume (P = 0.002) at baseline. GBM showing "true PD" occurred more frequently in the right internal capsule, thalamus, lentiform nucleus, and temporal lobe than those with "clinical PsP". Radiomic analysis predicted "clinical PsP" with > 70% accuracy on the validation dataset. Patients with early PD that eventually exhibit "clinically-defined PsP" have distinct clinical, molecular, and MRI characteristics. This information may be useful for treating clinicians to better understand the potential risks and outcome in patients exhibiting early radiographic changes following chemoradiation.

IMMU-50. GLIOBLASTOMA MANIPULATES HEMATOPOIETIC STEM CELL DIFFERENTIATION OUTCOMES AND DRIVES ALTERED IMMUNE RECONSTITUTION

Abstract INTRODUCTION Bone marrow-derived hematopoietic stem and progenitor cells (HSPCs) give rise to the cellular components of the immune system. Unfortunately, immune reconstitution from HSPCs are negatively impacted by solid cancers, including high-grade gliomas. For example, an expansion of myeloid progenitor cells has been previously described across several cancers that originate outside the CNS. A similar expansion of MDSCs coupled with diminished T cell function has also been described in the peripheral blood of patients with newly-diagnosed GBM. Alterations in both lymphoid and myeloid compartments due to CNS malignancy led us to determine how intracranial gliomas impact HSPCs in both their capacity to reconstitute the immune compartment and in their cell fate determination. This is important to better understand the impact of gliomas on immunity and how we can leverage these findings to better develop cellular immunotherapeutics. METHODS HSPCs were isolated from bone marrow of C57BL/6 mice with orthotopic KR158B glioma, or age-matched naïve mice. Experiments were conducted to compare relative changes in: gene expression (RNA-sequencing), precursor frequencies, cell fate determination, and cellular function of cells derived from HSPCs of glioma-bearing mice. RESULTS RNA-sequencing revealed 700+ genes whose expression was significantly up- or downregulated in HSPCs from glioma-bearing mice, particularly those involved with stemness and metabolic activity. Importantly, HSPCs from glioma-bearing mice expressed upregulation of genes involved in myelopoiesis relative to naïve mice. This was coupled with an expansion of granulocyte macrophage precursors (GMPs), the progenitors to gMDSCs. Next, differentiation assays revealed that HSPCs from glioma-bearing mice had higher propensity of differentiating into MDSC under homeostatic conditions relative to controls both in vitro and in vivo. Furthermore, mice bearing intracranial gliomas possess an expansion of MDSCs which are more suppressive on T cell proliferation and hinders T cell-mediated tumor cell killing relative to MDSCs derived from naïve control mice.

CTNI-35. UPDATE ON THE FEASIBILITY STUDY OF OKN-007 IN COMBINATION WITH TEMOZOLOMIDE AND RADIATION IN NEWLY DIAGNOSED GLIOBLASTOMA

Abstract BACKGROUND Temozolomide (TMZ) with concurrent radiation is the traditional standard of care for newly diagnosed glioblastoma. Unfortunately, this combination has limited efficacy and resistance can render TMZ ineffective. The novel anti-cancer agent OKN-007 plus TMZ increased survival in preclinical studies. Therefore, we initiated a phase Ib/feasibility clinical trial (NCT03587038) of OKN-007 in combination with TMZ and radiation therapy (RT). We report the safety and tolerability findings of this trial in-progress. METHODS Adults with newly-diagnosed GBM were eligible. OKN-007 was administered by IV at 60 mg/kg. There were three treatment phases: Concomitant, Pre-Maintenance, and Maintenance. In the Concomitant Phase, patients received OKN-007 three times per week (Cohort A) or five times per week (Cohort B); all patients receive TMZ at 75 mg/m2 daily and RT at 60 Gy over 30 fractions. In the 28-day Pre-Maintenance Phase, all patients receive OKN-007 thrice weekly. In the Maintenance Phase (MP), comprising up to eighteen 28-day cycles, TMZ was dosed at 150-200 mg/m2 on days 1-5 of each cycle for six cycles. OKN-007 was administered thrice weekly for six cycles, then twice weekly for three cycles, then once weekly for nine cycles. Each cohort was evaluated for safety with 3-6 patients followed by an expansion of cohorts if safety parameters were met. RESULTS Three patients completed Cohort A without dose-limiting toxicity (DLT). In Cohort B, two DLT’s (hematologic toxicities deemed to be related to TMZ) occurred, and this cohort was stopped. Currently, median PFS and OS have not been reached due to lack of events, but preliminary data indicate improved median PFS and OS compared to standard of care. CONCLUSIONS The treatment plan appears safe and well-tolerated at the Cohort A combination dosing level and may increase favorable treatment outcomes suggesting that the OKN-007 warrants further study.

PATH-16. EVALUATION OF SEX-BASED DIFFERENCES IN CLINICAL OUTCOMES AND TUMOR GENOMICS IN PATIENTS WITH NEWLY DIAGNOSED IDH-WILDTYPE GLIOBLASTOMA RECEIVING CHEMORADIATION

Abstract BACKGROUND We evaluated sex-based differences in clinical outcomes and tumor genomics in patients with newly-diagnosed GBM. METHODS We reviewed 665 IDH-wild type GBM patients with Karnofsky Performance Status (KPS) ≥60 treated at our institution from 2010-2019 including; 585 patients with targeted exome sequencing of 447 cancer associated genes (OncoPanel). Deleterious mutations were defined as homozygous deletions or loss of function mutations of known tumor suppressors (as reported in TCGA, ≥ 3 times in the COSMIC database, or predicted as “damaging” in SIFT and/or “probably damaging” in Polyphen 2) or known oncogenic mutations in proto-oncogenes (reported in TCGA or ≥ 3 times in COSMIC). RESULTS There were 384 (57.7%) males and 281 (42.3%) females. Median OS was 22.5 months for females and 19.3 months for males (hazard ratio [HR] 0.81, 95% CI 1.03-1.48, p = 0.02). On multivariable analysis adjusted for age, KPS ≥90, extent of resection, and MGMT methylation status, female sex (adjusted hazard ratio 0.78, 95% CI [0.64-0.95], p = 0.015) was associated with improved OS. Superior OS in females was observed in MGMT-unmethylated patients (HR 0.69, 95% CI [0.54-0.90], p = 0.005) but not MGMT-methylated patients. Thirteen genes were deleteriously altered in ≥5% of our cohort: CDK4 (12.1% male vs. 7.8% female), CDKN2A (46.5% vs. 45.7%), CDKN2B (41.8% vs. 43.3%), EGFR (34.7% vs. 40.0%, MTAP (18.2% vs. 18.8%), NF1 (11.5% vs. 9.4%), PTEN (28.2% vs. 29.8%), TP53 (28.2% vs. 30.2%), RB1 (5.6% vs. 6.5%), MDM4 (6.2% vs. 5.7%), ATM (5.9% vs. 3.7%), MDM2 (7.4% vs. 4.1%), PIK3R1 (6.2% vs. 4.1%). There were no differences in frequency of mutations in these individual genes between males and females (χ 2 [1, N=585] = 0.05-2.86, p = 0.09-0.86). CONCLUSIONS Female sex is associated with improved survival. We did not identify sex-based differences in deleterious genomic alterations amongst commonly altered genes in GBM.

NIMG-40. MRI-BASED ESTIMATION OF THE ABUNDANCE OF IMMUNOHISTOCHEMISTRY MARKERS IN GBM BRAIN USING DEEP LEARNING

Abstract Glioblastoma (GBM) is a devastating primary brain tumor known for its heterogeneity with a median survival of 15 months. Clinical imaging remains the primary modality to assess brain tumor response, but it is nearly impossible to distinguish between tumor growth and treatment response. Ki67 is a marker of active cell proliferation that shows inter- and intra-patient heterogeneity and should change under many therapies. In this work, we assessed the utility of a semi-supervised deep learning approach for regionally predicting high-vs-low Ki67 in GBM patients based on MRI. We used both labeled and unlabeled datasets to train the model. Labeled data included 114 MRI-localized biopsies from 43 unique GBM patients with available immunohistochemistry Ki67 labels. Unlabeled data included nine repeat routine pretreatment paired scans of newly-diagnosed GBM patients acquired within three days. Data augmentation techniques were utilized to enhance the size of our data and increase generalizability. Data was split between training, validation, and testing sets using 65-15-20 percent ratios. Model inputs were 16x16x3 patches around biopsies on T1Gd and T2 MRIs for labeled data, and around randomly selected patches inside the T2 abnormal region for unlabeled data. The network was a 4-conv layered VGG-inspired architecture. Training objective was accurate prediction of Ki67 in labeled patches and consistency in predictions across repeat unlabeled patches. We measured final model accuracy on held-out test samples. Our promising preliminary results suggest potential for deep learning in deconvolving the spatial heterogeneity of proliferative GBM subpopulations. If successful, this model can provide a non-invasive readout of cell proliferation and reveal the effectiveness of a given cytotoxic therapy dynamically during the patient's routine follow up. Further, the spatial resolution of our approach provides insights into the intra-tumoral heterogeneity of response which can be related to heterogeneity in localization of therapies (e.g. radiation therapy, drug dose heterogeneity).

The 70-year-old newly diagnosed glioblastoma patients are older than the 65-year-old? Outcome evaluation of the two categories in a matched case control study with propensity score balancing

BackgroundThe standard of care for elderly, newly-diagnosed glioblastoma patients consists, if feasible, of surgical resection followed by a short course of radiation therapy (RT) with concomitant and adjuvant temozolomide chemotherapy (TMZCHT). To date, the literature lacks of consistence in the definition of elderly, if older than 65 years, or 70 years. Aim of this study was to explore whether differences exist between these two cohorts, comparing outcomes using a propensity score matched analysis (PSM). Materials and methodsTwo hundred twenty-one elderly newly diagnosed glioblastoma patients were included. All patients received surgery followed by RT with concurrent and adjuvant TMZCHT. The RT dose prescribed was 60 Gy/30 fractions for patients 65–69-year-old or 40.5 Gy/15 fractions for ≥70-year-old. After 1:1 matching there were 86 patients in each group. Distribution of covariates was adequately balanced in the matched data set. ResultsAfter PSM median PFS time, 1,2,3-year PFS rates were 10 months, 33.3%, 13.1%, and 6.6% for the 65–69-year group, 9 months, 34.7%, 11% and 4.8% for the ≥70-year group (p = 0.530). Median OS time, and 1,2,3-year OS rates were 14 months, 54.1%, 23.4%, 13.9% for the 65–69-year old group, and 12 months, 49.3%, 21.5%, 10% for the ≥70-year group (p = 0.357). No differences were recorded in relation to different groups of age. ConclusionsThe PSM analyses showed a similar outcome in 65–69-year old patients compared to older ones notwithstanding a more burdensome RT schedule. Hypofractionated RT treatment has to be considered also in this group of younger elderly, newly-diagnosed GBM patients.

CTNI-16. FEASIBILITY PILOT STUDY OF OKN-007 IN COMBINATION WITH ADJUVANT TEMOZOLOMIDE CHEMORADIOTHERAPY IN PATIENTS WITH NEWLY DIAGNOSED GLIOBLASTOMA

Abstract BACKGROUND Temozolomide (TMZ) is the standard of care chemotherapy agent for glioblastoma. Unfortunately, intrinsic or acquired resistance eventually renders TMZ ineffective. The novel anti-cancer agent OKN-007 plus TMZ increased survival in glioma-bearing mice compared to TMZ alone. Furthermore, OKN-007 increased TMZ sensitivity in both TMZ-sensitive and TMZ-resistant cell lines. Therefore, we initiated a clinical trial (NCT03587038) of OKN-007 in combination with TMZ and radiation therapy (RT). Here we report the safety and tolerability findings of this trial in-progress. METHODS Adults with newly-diagnosed and resected GBM are eligible. OKN-007 is administered by IV at 60 mg/kg. There are three treatment phases: Concomitant, Pre-Maintenance, and Maintenance. In the Concomitant Phase, patients receive OKN-007 three times per week (Cohort 1) or five times per week (Cohort 2); all patients receive TMZ at 75 mg/m2 daily and RT at 60 Gy over 30 fractions. In the 28-day Pre-Maintenance Phase, all patients receive OKN-007 thrice weekly. In the Maintenance Phase (MP), comprising up to eighteen 28-day cycles, TMZ is dosed at 150–200 mg/m2 on days 1–5 of each cycle for six cycles. OKN-007 is administered thrice weekly for six cycles, then twice weekly for three cycles, then once weekly for nine cycles. An expansion cohort of up to 25 patients will use the highest tolerated dosing protocol. RESULTS To date, five patients have been enrolled. Three patients in Cohort 1 and one in Cohort 2 advanced to the MP, and no dose-limiting toxicities (DLTs) occurred. One patient in Cohort 2 was removed due to a DLT (grade 3 neutropenic fever). CONCLUSIONS The treatment plan in Cohort 1 appears safe and well-tolerated. Recruitment for Cohort 2 is ongoing and will be expanded to further evaluate safety. Once the outcome of Cohort 2 is known, the regimen for the expansion cohort will be determined.

CTIM-29. CRUCIAL APPLICATION OF SINGLE CELL GENE EXPRESSION FOR MONITORING CHANGES IN GLIOBLASTOMA PATIENTS WITH SUSTAINED RADIOGRAPHIC RESPONSE TO CMV pp65-LAMP RNA-PULSED DENDRITIC CELL-BASED VACCINES

Abstract BACKGROUND The application of single cell sequencing as a novel immune monitoring platform can be used to identify the molecular mechanisms of immune response to dendritic cell- based vaccines, trace the cell types and states involved, and uncover novel biomarkers for immunotherapy. We applied single-cell RNA Seq analysis of longitudinal peripheral blood mononuclear cells (PBMCs) in patients with newly-diagnosed GBM enrolled on the ATTAC II clinical trial (FDA IND BB-16530; Clinicaltrials.gov # NCT02465268) who experienced a sustained radiographic response to autologous CMV pp65-LAMP RNA-pulsed DC vaccines plus GM-CSF and tetanus-diphtheria booster administered during adjuvant cycles of dose-intensified temozolomide. METHODS We constructed 5’ gene expression libraries and T cell receptor enriched libraries for 10x Genomics single-cell 5’ and VDJ sequencing, generated from PBMCs collected prior to and during patient immunization using dendritic cells loaded with messenger ribonucleic acid encoding the human cytomegalovirus (CMV) matrix protein pp65 conjugated with the lysosomal associated membrane protein (LAMP) sequence. RESULTS Overall, we sequenced a total of 189,808 single-cell transcriptomes from 5 patients. We leveraged these transcriptome-wide features to distinguish 15 peripheral immune cell subtypes in tested PBMCs. Analysis revealed dynamic changes in immune cell subsets over the course of first three vaccines, including increases in cytotoxic CD8 T cells, CD4 T cells, and NK cell subsets. Increased markers of T cell activation were observed during vaccination. Surprisingly, we observed a very high-level frequency of natural killer T (NKT) cells in the patient with a complete durable response compared to other patients. After three DC vaccines, the level of NKT cells in PBMC of this patient increased up to 10%. CONCLUSIONS These results emphasize the importance of subset specific profiling to achieve higher resolution in monitoring immune responses compared with bulk expression profiling in patients receiving immunotherapeutic treatment.

Immunological responses through miRNA signatures in GBM plasma extracellular vesicles from patients receiving experimental immunotherapy

Abstract Glioblastoma (GBM) patients have a median survival of 15 months despite aggressive treatment. Treatment-related pseudo-progression confounds outcome assessment by MRI, particularly in patients receiving immunotherapy. Extracellular vesicles (EVs) contain tumor-specific miRNA that could serve as a liquid biopsy to distinguish true progression from treatment-related pseudo-progression. Small plasma EVs were isolated by serial density gradient ultracentrifugation from 20 newly-diagnosed GBM patients enrolled in a clinical trial of allogeneic tumor lysate-pulsed autologous dendritic cell (DC) vaccination. Short non-coding RNA sequencing was performed for each patient at three time points (TP): pre-vaccine (TP1), post-initial vaccine (TP2), and at end of treatment (TP3). Ingenuity Pathway Analysis from 31 differentially expressed miRNAs across time points (p-value &amp;lt;0.05, |logFC|&amp;gt;1) revealed that glioma-related signaling pathways previously seen in small EVs from glioma patients compared to normal donors were replicated when comparing patients at the beginning of the treatment (TP1-vs-TP2) and between early to late time points (TP1-vs-TP3). In addition, pathways corresponding to immunological responses typically seen in GBM patients undergoing treatment, such as IL-10, IL-6, and Toll-like Receptor (TLR) signaling, were exclusively found after the initial vaccine towards the end of treatment (TP2-vs-TP3). In conclusion, signaling pathways derived from differentially expressed miRNAs across time points suggest that as patients progress through treatment, consistent differences in plasma small EVs miRNA expression profile and immunological responses could be utilized as predictors of treatment response.

RARE-26. WHOLE GENOME SEQUENCING OF AN OSSEOUS METASTASIS DURING CHECKPOINT-CONTROLLED INTRACRANIAL GLIOBLASTOMA REVEALS NEW INSIGHTS INTO POTENTIAL MECHANISMS OF IMMUNE ESCAPE

Abstract Glioblastoma (GBM) has a devastating prognosis and recent advances in the treatment of a variety of cancer entities, e.g. through checkpoint inhibition, could so far not be translated into improved outcome in newly-diagnosed GBM. Characterizing rare cases of peripheral metastases which succeeded in overcoming immune control, can help to understand the mechanisms of immune escape. Here we describe the first reported case of a detailed genetic and immunological characterization of a peripheral bone metastasis from a GBM which was controlled intracranially by anti-PD1 checkpoint inhibition We performed whole genome sequencing (WGS) of the primary- and recurrent tumor, as well as the bone metastasis. Genomic data was analyzed for copy number variations and mutational profiles. In addition, immune monitoring with flow cytometric phenotyping and next-generation sequencing of the peripheral T-cell repertoire was used. A 70-year old patient developed multiple osseous metastases in the spine, while his IDHwt GBM recurrence was immunologically controlled with checkpoint inhibition. Biopsy confirmed peripheral GBM metastases. Immunophenotyping reflected the effective activation of the peripheral T-cell response, with, however, simultaneous upregulation of regulatory T-cells during disease progression. WGS sequencing demonstrated a distinct molecular profile of the GBM metastasis, with amplifications in chromosome 3 and 9, as well as genomic loss on chromosomes 4, 10 and 11. The peripheral metastasis was distinguished by mutations in mismatch repair genes, such as MSH4 and MLH1, associated with a hypermutated phenotype. Among the mutated genes we found potential candidates involved in immune escape of circulating tumor cells. This case represents a unique opportunity to analyze potential mechanisms of GBM-mediated immune escape during immune activation with anti-PD1 checkpoint therapy. It highlights the fact, that although an effective, disinhibited immune response can control the cranial GBM disease, hypermutated tumor clones can evade the tumor-specific T-cell response and disseminate to distant organs.

GENE-36. INTEGRATING TRANSCRIPTOMICS AND KINOMICS IDENTIFIES SYNERGISTIC DRUG COMBINATIONS FOR GLIOBLASTOMA TREATMENT

Abstract Glioblastoma (GBM) is the most common and malignant adult brain tumor. Despite years of research, few advancements have been made in its management. One promising avenue of research has been treatment with BRD4 inhibitors, which decrease oncogene expression in GBM cells. However, resistance to these inhibitors is rapidly acquired. Kinome reprogramming is thought to underlie this resistance, suggesting a need for combination therapy with kinase inhibitors. The goal of this study is to determine whether transcriptomic and kinomic profiling of GBM tumors will identify synergistic drug pairs for GBM treatment. We profiled the active kinome on a set of three newly-diagnosed GBM patient-derived xenograft (PDX) tumors and three recurrent tumors using quantitative SILAC mass spectrometry. Additionally, kinome reprogramming following BET inhibition was profiled in vitro using the BET inhibitor JQ1. Kinome activity was uploaded into our novel computational platform, SynergySeq, to assess the synergistic potential of kinase inhibitors with JQ1. Additionally, single cell RNA-sequencing of a GBM tumor was used to determine the cell populations affected by each inhibitor. To quantify synergy in response to combination therapy, cells were treated with a combination matrix of JQ1 and a kinase inhibitor in variable concentrations and cell death was quantified via ATP levels. Our results showed that newly-diagnosed tumors were predicted to be sensitive to combined BET inhibition with Bcr/Abl, EGFR, and FGFR kinase inhibitors. Recurrent tumors were sensitive to combined Bcr/Abl and FGFR inhibitors but were not sensitive to EGFR inhibitors. We screened inhibitors in vitro and found a synergistic effect for the combination of JQ1 and TAS120, a pan-FGFR inhibitor. These data suggest that clinically a brain penetrant BET inhibitor should be effective in combination with a brain penetrant FGFR inhibitor. Importantly, our computational platform is a novel informatics-based approach for targeted therapy in a patient-specific and disease-specific manner.

ATIM-34. SINGLE-CELL RNA SEQUENCING REVEALS DYNAMIC IMMUNE RESPONSE CHANGES IN GLIOBLASTOMA PATIENT WITH DURABLE COMPLETE RESPONSE TO CMV PP65-LAMP RNA-PULSED DENDRITIC VACCINES

Abstract BACKGROUND The application of single cell sequencing as a novel immune monitoring platform can enhance the ability to interrogate at unprecedented depth the single-cell level dynamic phenotypic and functional attributes of immune cells in patients undergoing cancer immunotherapy treatment. We applied single-cell sequencing analysis of PBMCs in a patient with newly-diagnosed GBM enrolled on the ATTAC II clinical trial (FDA IND BB-16530) who experienced a sustained complete radiographic response to autologous CMV pp65-LAMP RNA-pulsed DC vaccines plus GM-CSF and tetanus-diphtheria booster administered during adjuvant cycles of dose-intensified TMZ. METHODS We constructed 5’ gene expression libraries and T cell receptor enriched libraries for 10x Genomics single-cell sequencing, generated from PBMCs collected prior to and during patient immunization using dendritic cells loaded with RNA encoding the CMV matrix protein pp65 conjugated with the lysosomal associated membrane protein (LAMP) sequence. RESULTS RNA-Seq revealed dynamic changes in immune cell subsets over course of first three vaccines, including increases in cytotoxic T cells and memory T cell subsets. Increased markers of T cell activation were observed during vaccination including enhanced T cell signaling pathways, proliferative signals, and cytokine production. We found that proportion of cytotoxic T cells increased from 3.42% to 11.74% in whole PBMCs after immunization. Surprisingly, we observed very high level of frequency natural killer T (NKT) cells comprising 4.75 % of this patient’s PBMCs at baseline. After three DC vaccines, the level of NKT cells in PBMC increased up to 10%. CONCLUSIONS Single cell RNA sequencing of a patient with sustained complete response to DC vaccination during cycles of dose-intensified TMZ reveals dynamic immune activation and expansion within the cytotoxic T cells and NKT compartments. These results emphasize the importance of subset specific profiling to achieve higher resolution in monitoring immune responses compared with bulk expression profiling in patients receiving immunotherapeutic treatment.

NIMG-39. REVEALING THE TUMOR-IMMUNE LANDSCAPE THROUGH SPATIALLY-RESOLVED RADIOMICS: CASE STUDIES

Abstract BACKGROUND Conventional magnetic resonance imaging (MR) guides patient care in GBM. However, there is mounting awareness that MR enhancement is non-specific reflecting either tumor progression or non-tumoral inflammatory changes. Histological evaluation of GBM is held as the gold standard for disease assessment. However, the invasiveness of this methodology and the sample sparsity limit its usefulness. Methods to infer histological underpinnings of MRI are needed to improve clinical care. METHODS A transfer learning mixed effects model based on T1Gd and FLAIR MR voxel based image features trained and cross-validated to predict FPKM values of CD68, CASP3, CD8A and IL13RA2. Training data included RNAseq from 38 image-localized biopsies from 15 newly-diagnosed GBM patients. These models were then applied to two independent patients, chosen based on therapy and the quality of image registration, at three different time points, just prior to receiving IL13RA2 targeted CAR-T therapy for rGBM, after 2 cycles and after 4 cycles, resulting in a voxel-based prediction of the relative expression levels of these genes. RESULTS Cross-validation of the proposed machine learning models demonstrated a Pearson Correlation coefficient between predicted and observed FPKM values of 0.89 (CD68), 0.92 (CASP3), 0.93 (CD8A), and 0.88 (IL13RA2). When the models were applied to the two CAR-T patients, CD68-modeled expression levels were seen to increase throughout therapy for one patient, while remaining stable for the other. Survival from first CAR-T infusion was, respectively, 412 and 83 days suggesting that the increased CD68 was indicative of therapeutic effectivity. Further, the spatial activity predictions were consistent with expected therapeutic action as the correlation coefficient between CASP3 and the product of IL13R2 and CD8A expression was 0.81, suggesting the T-cells targeting IL13RA2 were inducing cell death. CONCLUSIONS Preliminary results with our model highlights the potential of spatially resolved radiomic maps to provide insight into regional therapeutic effectivity.

Abstract A021: Vaccination with autologous, nonattenuated, live glioblastoma cells induces potent peripheral and intratumoral anti-tumoral responses: A first-in-human study

Abstract Glioblastoma (GBM), which grows uninhibited in the brain, almost never metastasizes outside of it. The rare occurrence (&amp;lt;0.5%) of overt GBM extracranial metastasis is paradoxically associated with significantly enhanced patient survival. We previously reported that the highly malignant F98 Fischer-rat GBM model, which grows aggressively in the brain, spontaneously regresses when injected subcutaneously as live unattenuated cells. The peripheral growth and regression of live glioma cells markedly enhanced the survival of all rats, and cured (survival &amp;gt;9 months) approximately 50% of rats challenged with F98 cells intracranially. The results in the F98 model were corroborated in the CNS1 astrocytoma model in Lewis rats. The fraction of cured rats in both models were the highest reported in published literature. We found that the spontaneous rejection of live glioma cells inoculated in peripheral sites was immune-mediated. This location-biased immune response was termed “Split Immunity”—a tumor that thrives in an immune-privileged site (e.g., the brain) may be inhibited by injecting live unmodified tumor cells in a site that is not privileged, generating protective immunity that spreads back to the privileged site.To evaluate the applicability of the “Split Immunity” approach in humans, we ran a first-in-human (FIH) study on two recurrent GBM patients. The patients were vaccinated subcutaneously, initially with autologous irradiated tumor cells, and then with autologous, non-attenuated, live tumor cells. The treatment caused no serious adverse events. The injected live tumor cells did not grow at the inoculation site, nor did they spread metastatically, as evaluated by choline 18F PET-CT. Standard biochemical and hematologic blood tests showed no treatment-related adverse effects. Both patients’ quality of life (QoL) questionnaires demonstrated subjective improvements in self-evaluated “global health status.” The treatment had demonstrated several signs indicative of efficacy. MRI evaluation using “treatment response assessment maps” (TRAM) demonstrated strong increases in the “treatment-response” component following live-cell vaccination, compatible with “pseudo-progression-like” responses in both patients. Using elaborate 8-color flow cytometric panels, we detected potent tumor-cell specific polyfunctional activation in 2.5% and 5% of peripheral cytotoxic T-cell (CTL) and T helper cells (Th), respectively, 3 days following live-cell vaccination. Flow cytometric immune mapping of all intratumoral immune cell subsets showed major changes from baseline that had occurred shortly after live-cell vaccinations. We observed major increases (8 to 200 fold) in both patients of all monitored intratumoral dendritic cell (DC) subsets (CD1c+ and CD141+ myeloid DC, CD123+ plasmacytoid DC). Patient-1 showed the highest frequency of intratumoral CD56brt natural killer (NK) cells from all previously fully mapped brain tumor patients. Patient-2’s intratumoral CTL and Th were increase by 6- and 17-fold, respectively. Patient-2’s CD141+ and CD1c+ myeloid DC had exhibited upregulated CD86 maturation marker following treatment. Overall, the immune infiltrate following treatment in Patient-2 exceeded the immune infiltrates found in 42 previously fully-mapped brain tumor patients. The flow cytometric results were corroborated by the patients’ immunohistochemistry data. Patient-2, who had been treated by a tighter vaccination protocol than Patient-1, had survived 19 months after GBM recurrence, which is considerably longer than the median 6- to 8-month survival of recurrent GBM patients. Taken together, the accumulated FIH results are suggestive of safety and efficacy of the “Split Immunity” approach. Based on the these results, we have recently received an ethical approval for a phase IIa clinical trial on 8 recurrent- and 4 newly-diagnosed GBM patients. Citation Format: Ilan Volovitz, Nati Shapira, Rachel Grossman, Zvi Ram. Vaccination with autologous, nonattenuated, live glioblastoma cells induces potent peripheral and intratumoral anti-tumoral responses: A first-in-human study [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr A021.

Overview on current treatment standards in high-grade gliomas.

High-grade gliomas (HGGs) are the most common primary tumors of the central nervous system, which include anaplastic gliomas (grade III) and glioblastomas (GBM, grade IV). Surgery is the mainstay of treatment in HGGs in order to achieve a histological and molecular characterization, as well as relieve neurological symptoms and improve seizure control. Combinations of some molecular factors, such as IDH 1-2 mutations, 1p/19q codeletion and MGMT methylation status, allow to classify different subtypes of gliomas and identify patients with different outcome. The SOC in HGGs consists in a combination of radiotherapy and chemotherapy with alkylating agents. Despite this therapeutic approach, tumor recurrence occurs in HGGs, and new surgical debulking, reirradiation or second-line chemotherapy are needed. Considering the poor results in terms of survival, several clinical trials have explored the efficacy and tolerability of antiangiogenic agents, targeted therapies against epidermal growth factor receptor (EGFR) and different immunotherapeutic approaches in recurrent and newly-diagnosed GBM, including immune checkpoint inhibitors (ICIs), and cell- or peptide-based vaccination with unsatisfactory results in term of disease control. In this review we describe the major updates in molecular biology of HGGs according to 2016 WHO Classification, the current management in newly-diagnosed and recurrent GBM and grade III gliomas, and the results of the most relevant clinical trials on targeted agents and immunotherapy.