The safety and accuracy of intratumoral catheter placement to infuse viral immunotherapies in children with malignant brain tumors: a multi-institutional study

OBJECTIVE

Relatively little is known about the safety and accuracy of catheter placement for oncolytic viral therapy in children with malignant brain tumors. Accordingly, this study combines data from two phase I clinical trials that employed viral immunotherapy across two institutions to describe the adverse event profile, safety, and accuracy associated with the stereotactic placement and subsequent removal of intratumoral catheters.

METHODS

Children with progressive/recurrent supratentorial malignant tumors were enrolled in two clinical trials (NCT03043391 and NCT02457845) and treated with either the recombinant polio:rhinovirus (lerapolturev) or the genetically modified oncolytic herpesvirus (G207). Age, sex, race, tumor diagnosis, and tumor location were analyzed. Events related to the catheter placement or removal were categorized. A catheter that was either pulled back or could not be used was defined as "misplaced." Neuronavigation software was used to analyze the accuracy of catheter placement for NCT03043391. Descriptive statistics were performed.

RESULTS

Nineteen patients were treated across the two completed trials with a total of 49 catheters. The mean ± SD (range) age was 14.1 ± 3.6 (7-19) years. All tumors were grade 3 or 4 gliomas. Nonlobar catheter tip placement included the corpus callosum, thalamus, insula, and cingulate gyrus. Six of 19 patients (31.6%) had minor hemorrhage noted on CT; however, no patients were symptomatic and/or required intervention related to these findings. One of 19 patients had a delayed CSF leak after catheter removal that required oversewing of the surgical site. No patients developed infection or a neurological deficit. In 7 patients with accuracy data, the mean ± SD distance of the planned trajectory (PT) to the catheter tip was 1.57 ± 1.6 mm, the mean angle of the PT to the catheter was 2.43° ± 2.1°, and the greatest distance of PT to the catheter in the parallel plane was 1.54 ± 1.5 mm. Three of 49 (6.1%) catheters were considered misplaced.

CONCLUSIONS

Although instances of minor hemorrhage were encountered, they were clinically asymptomatic. One of 49 catheters required intervention for a CSF leak. Congruent with previous studies in the literature, the stereotactic placement of catheters in these pediatric tumor patients was accurate with approximately 95% of catheters having been adequately placed.

CTIM-15. RESULTS OF A PHASE IB TRIAL OF RECOMBINANT POLIO:RHINOVIRUS IMMUNOTHERAPY FOR RECURRENT PEDIATRIC HIGH GRADE GLIOMA

BACKGROUND

Outcomes of recurrent pediatric high grade glioma (pHHG) are poor with a median overall survival (OS) of < 6 months. Viral immunotherapy such as the polio:rhinovirus chimera, PVSRIPO, is a novel treatment approach for recurrent pHHG. PVSRIPO is genetically engineered to prevent neurovirulence. In adults with recurrent glioblastoma treated with PVSRIPO, 21% survived > 36 months. The poliovirus receptor, CD155, is ubiquitously expressed in malignant pediatric brain tumors including pHHG.

METHODS

The primary objective of this Phase 1b clinical trial was to evaluate the safety and feasibility of PVSRIPO for recurrent pHGG. PVSRIPO was given at a single dose, 5x107 50% tissue-culture infectious dose (TCID50) administered by convection enhanced delivery (CED) to children with biopsy-confirmed recurrent pHHG between ≥ 1 and ≤ 5.5 in diameter. 3 mL of PVSRIPO was delivered at 0.5 mL/hr via a single catheter.

RESULTS

Eight patients were treated with PVSRIPO including 5 males and 3 females with a median age of 16.5 (range 9-19). Six patients had recurrent glioblastoma, 2 had recurrent anaplastic astrocytoma. The median number of previous recurrences prior to enrollment was 3.5 (range 1-5). Four patients received bevacizumab on-study for treatment-related peritumoral inflammation/edema. Six of 8 patients experienced 26 treatment related adverse events (AEs) possibly, probably, or definitely related to protocol treatment. There were no Grade 4 or 5 AEs. There were 3 Grade 3 AEs: 2 headaches and 1 seizure. There were no AEs related to biopsy or CED catheter insertion/removal. Median OS was 4.13 months (range 1.23-NA). One patient is currently alive at > 21 months. Monocyte and T cell inflammatory phenotypes and total CD4+ T cells were increased in peripheral blood after treatment.

CONCLUSIONS

CED of PVSPRIO is both safe and feasible for the treatment of recurrent pHHG. Histologic correlative results will also be presented.

IMMU-43. CXCR3 SIGNALING ENGAGES GLIOMA INFILTRATING T CELLS AND IS REQUIRED FOR THE ANTITUMOR EFFICACY OF INNATE STIMULATING IMMUNOTHERAPY

BACKGROUND

T cell dysfunction is well documented in patients with gliomas, yet mechanisms explaining T cell dysfunction in gliomas are lacking. Recent evidence links CXCR3 signaling and functional immune surveillance in both preclinical models and human data sets.

METHODS

Bulk and single cell RNA-seq data sets were used to compare CXCR3 ligand expression in gliomas to that of other solid tumor types. The role of CXCR3 signaling in mediating the recruitment and phenotype of tumor infiltrating lymphocytes (TILs) was tested using CXCR3 blockade and genetic deletion of CXCR3 in murine tumor models. The capacity of relevant innate stimulating immunotherapies to induce CXCR3 ligands in human glioma tissue was assessed using an ex vivo glioma slice culture assay, and the contribution of CXCR3 signaling to polio virotherapy (PVSRIPO) and STING agonist therapy was tested in murine models.

RESULTS

Relative to other solid tumors, gliomas express significantly lower levels of CXCR3 ligands (CXCL9, 10, and 11), which are largely expressed by tumor associated myeloid cells. Blocking CXCR3 signaling profoundly restricted TIL recruitment in murine gliomas, and resulted in decreased TIL stemness features coinciding with exacerbated exhaustion phenotypes. Polio virotherapy and STING agonism efficiently induced CXCR3 ligand production in fresh human glioma tissue in a manner dependent upon myeloid cells, and at levels similar to that induced by the same agents in melanoma and breast cancer tissue. CXCR3+ TILs expressed higher levels of memory and activation markers, proliferated more, and were required for the antitumor efficacy of both polio virotherapy and a STING agonist in mice.

CONCLUSION

This work indicates that CXCR3 signaling, a key contributor to cancer immune surveillance, is disrupted in human gliomas relative to other solid tumor types. Intratumor virotherapy and STING agonists hold potential to rectify diminished CXCR3 ligand expression in gliomas.

EXTH-84. A NOVEL DUAL-SPECIFIC CHIMERIC ANTIGEN RECEPTOR T CELL WITH HIGH SPECIFICITY FOR EGFR AND EGFRVIII IMPROVES SURVIVAL IN EGFR EXPRESSING MEDULLOBLASTOMA

BACKGROUND

Medulloblastoma is the most common malignant brain tumor in children 0-19 years of age and current treatment requires surgical resection, followed by high dose chemotherapy and radiotherapy. Therapy carries high morbidity, with late effects including neurocognitive decline, endocrine dysfunction, and subsequent malignancies. Chimeric antigen receptor (CAR) T cell therapy presents the potential for less toxic and more effective treatment for medulloblastoma. Epidermal growth factor (EGFR) is expressed in medulloblastoma derived cell lines and appears to serve an important role in the metastatic potential of this tumor type. D2C7 is a recombinant monoclonal antibody short chain variable fragment (scFv) with dual specificity, binding to wild type EGFR (EGFRwt) and its mutant EGFR variant III (EGFRvIII). We previously developed a novel, third-generation chimeric antigen receptor T-cell construct utilizing the D2C7 scFv and performed analysis on both glioblastoma and medulloblastoma tumor models.

METHODS

U87 and U87vIII glioblastoma and DAOY medulloblastoma cell lines were characterized by flow cytometry to evaluate for EGFRwt and EGFRvIII expression. Cytotoxicity assays were performed utilizing flow cytometry on serially diluted effector: target ratios of U87, U87vII, and DAOY mixed with D2C7 CAR. DAOY was orthotopically implanted into the frontal lobe of NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice. After tumor implantation, D2C7 CAR was introduced within the tumor bed.

RESULTS

EGFR was detected in U87, U87vIII, and DAOY on flow analysis. EGFRvIII was expressed only on U87vIII. D2C7 CAR demonstrated effective in vitro cytotoxicity against all three cell lines with >90% cytotoxicity achieved at an effector: target ratio of 2.5:1. In vivo, D2C7 CAR treatment significantly prolonged survival compared to treatment with mock CAR.

CONCLUSION

D2C7 CAR demonstrates efficacy in vitro and in mouse models of medulloblastoma and glioblastoma. Future steps in our work will set out to assess the D2C7 CAR in NSG mice with DAOY in the posterior fossa.

CTIM-23. DOSE ESCALATION TRIAL OF FC-ENGINEERED ANTI-CD40 MONOCLONAL ANTIBODY (2141-V11) ADMINISTERED INTRATUMORALLY WITH D2C7-IT VIA CONVECTION-ENHANCED DELIVERY (CED) FOR RECURRENT MALIGNANT GLIOMAS (RMGS)

BACKGROUND

D2C7-IT, a novel immunotoxin-based cytotoxic therapy, targets epidermal growth factor receptor (EGFR) and mutant EGFR variant III. In preclinical studies, D2C7-IT kills tumor cells and prolongs survival, but is unable to generate cures in all animals. We hypothesized that immunosuppression in glioblastoma limits D2C7-IT efficacy. Eliminating glioblastoma immunosuppression via CD40 co-stimulation is anticipated to enhance D2C7-IT-induced antitumor responses. In murine glioma models, CED of D2C7-IT+αCD40 generated cures and long-term tumor-specific adaptive immunity. Hence, we are conducting a phase 1 trial of D2C7-IT+2141-V11 (αhuman-CD40) administered via CED in rMG patients.

METHODS

Eligibility includes adult patients with solitary supratentorial rMG (WHO grade 3/4); ≥ 4 weeks after chemotherapy, bevacizumab, or study drug; adequate organ function; and KPS ≥ 70%. Cohorts of 3 patients are treated with increasing doses of 2141-V11 to determine the maximum tolerated dose when administered sequentially following D2C7-IT (166,075 ng) via CED at 0.5 mL/hr. Five dose levels (DLs) are evaluated (2141-V11 at: #1: 0.70 mg; #2: 2.0 mg; #2**: 3.0 mg #2*: 4.0 mg; #3: 7.0 mg).

RESULTS

As of May 29, 2022, 13 patients were treated (3 patients on DL1, DL2, and DL2**; 2 patients on DL3 and DL2*). No dose-limiting toxicities were observed; however, lower DLs were added due to higher frequency of adverse events (AEs) expected with D2C7-IT+2141-V11 with DL3 and DL2* (fever, neurologic symptoms). All patients remain alive 0.7-10.6 months after therapy. Grade ≥ 2 AEs due to D2C7-IT+2141-V11 include: headache (grade 3, n = 1; grade 2, n = 4); pyramidal tract disorder (grade 3, n = 1; grade 2, n = 2); paresthesia (grade 3, n = 1); dysphasia (grade 3, n = 1); seizures (grade 2; n = 2); fever (grade 2; n = 2); and one each of grade 2 depressed level of consciousness, fatigue, and concentration impairment. Enrollment is ongoing.

CONCLUSIONS

Intratumoral administration of D2C7-IT+2141-V11 via CED is safe, encouraging efficacy results are observed.

SYST-12 D2C7 CAR: A NOVEL CAR T CELL THAT SIMULTANEOUSLY TARGETS WILDTYPE EGFR AND ITS MUTANT ISOFORM EGFRVIII FOR TREATMENT OF GLIOMA AND BRAIN METASTASES

INTRODUCTION

Chimeric antigen receptor (CAR) T-cells represent a revolutionary class of immunotherapy, achieving considerable success in hematological cancers but generally failing to control solid tumors, including gliomas, partly due to the lack of a ubiquitously-expressed target antigen. In this study, we engineered a novel CAR T-cell consisting of the D2C7scfv targeting moiety that binds a shared epitope between EGFR and EGFRvIII. EGFR is the most homogeneous antigen on glial brain tumors, and the mutant EGFR variant, EGFRvIII, is present on a considerable subset of high grade gliomas. CAR T-cells targeting EGFRvIII alone fail to treat tumors possessing as few as 5-10% EGFRvIII-negative cell. Thus, D2C7 CAR is expected to be superior to the EGFRvIII CAR.

METHODS

We retrovirally transduced T-cells with a vector encoding the D2C7scFv in tandem with intracellular signaling domains of CD28, 4-1BB, and CD3ζ to generate D2C7 CAR. We co-cultured D2C7 CAR or control CAR with fluorescently-tagged tumor cells expressing either EGFRwt or EGFRvIII to validate efficacy and specificity by flow cytometry. To determine in vivo efficacy, EGFRwt or EGFRvIII-expressing tumors were implanted intracranially in immunodeficient NSG mice. 48 hours later, D2C7 CAR, VIII CAR, or Mock CAR were administered intracranially and mice were monitored for survival.

RESULTS

D2C7 CAR specifically killed tumor cells that expressed either EGFRwt or EGFRvIII, but not cells that lacked EGFR. Intracranial D2C7 CAR administration resulted in significantly prolonged survival of mice bearing EGFRwt or EGFRvIII tumors compared to Mock CAR controls. Importantly, D2C7 CAR significantly benefitted mice bearing a heterogeneous mix of EGFRwt and EGFRvIII tumor cells, a model of tumor heterogeneity.

CONCLUSIONS

D2C7 CAR is efficacious against EGFRwt/EGFRvIII heterogeneous tumors. Intracranial administration of D2C7 CAR is predicted to safely and effectively treat a large cohort of patients due to the relatively high prevalence of EGFR and/or EGFRvIII-expressing brain tumors.

Reproducibility of outcomes in sequential trials using CMV-targeted dendritic cell vaccination for glioblastoma

2005

Background: Vaccination with dendritic cells (DCs) fares poorly in primary and recurrent glioblastoma (GBM). Moreover, GBM vaccine trials are often underpowered due to limited sample size. Methods: To address these limitations, we conducted three sequential clinical trials utilizing Cytomegalovirus (CMV)-specific DC vaccines in patients with newly diagnosed GBM eligible to receive standard of care resection and adjuvant radiation therapy and temozolomide chemotherapy. Autologous DCs were generated and electroporated with mRNA encoding for the CMV protein pp65. Serial vaccination was given throughout adjuvant temozolomide cycles, and 111Indium radiolabeling was implemented to assess migration efficiency of DC vaccines. Patients were followed for median overall survival (mOS) and OS. Results: Our initial study was the phase II ATTAC study (NCT00639639; total n = 12) with 6 patients randomized to vaccine site preconditioning with tetanus-diphtheria (Td) toxoid. This led to an expanded cohort trial (ATTAC-GM; NCT00639639) of 11 patients receiving CMV DC vaccines containing granulocyte-macrophage colony-stimulating factor (GM-CSF). Follow-up data from ATTAC and ATTAC-GM revealed 5-year OS rates of 33.3% (mOS 38.3 months; CI95 17.5-undefined) and 36.4% (mOS 37.7 months; CI95 18.2-109.1), respectively. ATTAC additionally revealed a significant increase in DC migration to draining lymph nodes following Td preconditioning ( P = 0.049). Increased DC migration was associated with OS (Cox proportional hazards model, HR = 0.820, P = 0.023). Td-mediated increased migration has been recapitulated in our larger confirmatory trial ELEVATE (NCT02366728) of 43 patients randomized to preconditioning (Wilcoxon rank sum, Td n = 24, unpulsed DC n = 19; 24h, P = 0.031 and 48h, P = 0.0195). In ELEVATE, median follow-up of 42.2 months revealed significantly longer OS in patients randomized to Td ( P = 0.026). The 3-year OS for Td-treated patients in ELEVATE was 34% (CI95 19-63%) compared to 6% given unpulsed DCs (CI95 1-42%). Conclusions: We report reproducibility of our findings across three sequential clinical trials using CMV pp65 DCs. Despite their small numbers, these successive trials demonstrate consistent survival outcomes, thus supporting the efficacy of CMV DC vaccine therapy in GBM. Clinical trial information: NCT00639639, NCT02366728.

D2C7 CAR: A novel CAR T cell that simultaneously targets wildtype EGFR and its mutant isoform EGFRvIII for treatment of glioma

Brain tumors are the leading cause of cancer death in children and a significant cause of morbidity and mortality in adults. Conventional treatments are suboptimal, thus signifying the need for novel therapeutic strategies, such as immunotherapy. Chimeric antigen receptor (CAR) T cells represent a revolutionary class of immunotherapy, achieving considerable success in eliminating hematological cancers but generally failing to control solid tumors in part due to the lack of a suitably-expressed target antigen. Epidermal growth factor receptor (EGFR) is the most ubiquitous and homogeneous antigen on glial brain tumors, and EGFR-directed therapies have been hotly pursued. Moreover, the mutant EGFR variant, EGFRvIII, is present on a subset of pediatric and adult high grade gliomas, representing a targetable, tumor-specific antigen. Unfortunately, CAR T cells targeting EGFRvIII fail to treat tumors possessing as few as 5–10% EGFRvIII-negative cells due to antigen escape. Thus, a CAR T cell that can target both EGFR and EGFRvIII is expected to be superior to a CAR that targets EGFRvIII alone. In this study, we developed a novel third generation CAR T cell consisting of the D2C7scfv targeting moiety that binds a shared epitope between EGFR and EGFRvIII. This D2C7 CAR was able to specifically and potently kill tumor cells expressing wildtype EGFR or EGFRvIII. Importantly, D2C7 CAR significantly prolonged survival of mice bearing EGFR or EGFRvIII-expressing gliomas of both adult (U87) and pediatric (DAOY) origin. Toxicity experiments involving EGFR-expressing human skin grafts provided evidence that D2C7 CAR is safe and effective when administered intracranially to mice bearing intracranial tumors.

EXTH-59. GENERATION OF A THIRD GENERATION CAR T CELL THAT SIMULTANEOUSLY TARGETS WILDTYPE EGFR AND ITS MUTANT ISOFORM EGFRVIII FOR TREATMENT OF GLIOBLASTOMA

Glioblastoma (GBM) is the most aggressive primary brain cancer with a median survival of less than 16 months. This dire prognosis signifies the urgent need for improved treatment options, such as immunotherapy. Chimeric antigen receptor (CAR) T cells have helped revolutionize immunotherapy, achieving considerable success in eliminating hematological cancers but generally failing to control solid tumors. One major hindrance to CAR T cell success in solid tumors is tumor heterogeneity. Tumor-associated or tumor-specific antigens (TAA or TSA, respectively) are rarely expressed by all malignant cells within a tumor. As a specific example in GBM, the most prevalent TSA, EGFRvIII, is present in just 30% of tumors, and then on only 30-50% of cells. Our pre-clinical and clinical experiences with CAR T cells reveal that tumors possessing as few as 5-10% EGFRvIII-negative cells will easily escape EGFRvIII-targeted CARs. Tumor cells that lack EGFRvIII expression often overexpress the wildtype isoform of EGFR (EGFRwt). Notably, EGFR is absent on normal brain. Therefore, a superior approach would be to simultaneously target EGFRvIII and EGFRwt, an approach that would bypass EGFR heterogeneity in EGFRwt/EGFRvIII-expressing tumors. Here, we generated a third generation CAR using the D2C7 single-chain variable fragment (scFv) targeting moiety that recognizes an epitope present in EGFRwt and EGFRvIII. Initial in vitro characterization of D2C7 CAR validated the specificity and function of D2C7 CAR, as it potently killed murine cell lines engineered to express either EGFRwt or EGFRvIII, but not a cell line expressing neither. Concomitant IFN-γ release supported these conclusions. Additionally, D2C7 CAR killed the human-derived GBM cell line U87 and vIII-transfected U87, U87vIII. Importantly, intracranially-administered D2C7 CAR significantly prolonged survival of mice bearing orthotopic U87vIII or U87/U87vIII heterogeneous tumors compared to mock-treated controls. Altogether, these data provide evidence that D2C7 CAR T cells represent a viable therapeutic option for EGFRwt/EGFRvIII heterogeneous tumors.

CTIM-10. REPRODUCIBILITY OF CLINICAL TRIALS USING CMV-TARGETED DENDRITIC CELL VACCINES IN PATIENTS WITH GLIOBLASTOMA

INTRODUCTION

Vaccination with dendritic cells (DCs) fares poorly in primary and recurrent glioblastoma (GBM). Moreover, GBM vaccine trials are often underpowered due to limited sample size.

METHODS

To address these limitations, we conducted three sequential clinical trials utilizing Cytomegalovirus (CMV)-specific DC vaccines in patients with primary GBM. Autologous DCs were generated and electroporated with mRNA encoding for the CMV protein pp65. Serial vaccination was given throughout adjuvant temozolomide cycles, and 111Indium radiolabeling was implemented to assess migration efficiency of DC vaccines. Patients were followed for median overall survival (mOS) and OS.

RESULTS

Our initial study was the phase II ATTAC study (NCT00639639; total n=12) with 6 patients randomized to vaccine site preconditioning with tetanus-diphtheria (Td) toxoid. This led to an expanded cohort trial (ATTAC-GM; NCT00639639) of 11 patients receiving CMV DC vaccines containing granulocyte-macrophage colony-stimulating factor (GM-CSF). Follow-up data from ATTAC and ATTAC-GM revealed 5-year OS rates of 33.3% (mOS 38.3 months; CI95 17.5-undefined) and 36.4% (mOS 37.7 months; CI95 18.2-109.1), respectively. ATTAC additionally revealed a significant increase in DC migration to draining lymph nodes following Td preconditioning (P=0.049). Increased DC migration was associated with OS (Cox proportional hazards model, HR=0.820, P=0.023). Td-mediated increased migration has been recapitulated in our larger confirmatory trial ELEVATE (NCT02366728) of 43 patients randomized to preconditioning (Wilcoxon rank sum, Td n=24, unpulsed DC n=19; 24h, P=0.031 and 48h, P=0.0195). In ELEVATE, median follow-up of 42.2 months revealed significantly longer OS in patients randomized to Td (P=0.026). The 3-year OS for Td-treated patients in ELEVATE was 34% (CI95 19-63%) compared to 6% given unpulsed DCs (CI95 1-42%).

CONCLUSION

We report reproducibility of our findings across three sequential clinical trials using CMV pp65 DCs. Despite their small numbers, these successive trials demonstrate consistent survival outcomes, thus supporting the efficacy of CMV DC vaccine therapy in GBM.

BIOM-20. TUMOR-INTRINSIC AND PERIPHERAL FEATURES ASSOCIATE WITH SURVIVAL AFTER POLIO VIROTHERAPY IN RECURRENT GBM

BACKGROUND

PVSRIPO is a live-attenuated recombinant rhino:poliovirus that mediates antitumor efficacy by engaging antitumor immunity. A subset (~20%) of patients with recurrent GBM (rGBM) survive >24 months after therapy. We previously reported that low tumor mutation burden (TMB) is associated with longer survival after PVSRIPO and immune checkpoint blockade therapy in rGBM, and that low TMB associates with higher inflammatory gene expression signatures in rGBM tumors.

METHODS

Clinical features were tested for association with survival after PVSRIPO therapy. Whole exome sequencing and RNA-sequencing of tumors were used to correlate mutational landscape, tumor mutation burden (TMB), and gene expression signatures of patient tumors with survival. An in vitro assay that measures inflammatory responses of patient PBMCs to PVSRIPO was performed. An independent cohort of paired primary and recurrent GBM tumors was used to assess longitudinal changes in TMB and gene expression signatures after standard of care treatment.

RESULTS

A short time to recurrence and low TMB associated with longer survival after PVSRIPO therapy; these features were not prognostic for longer survival in immunotherapy naïve rGBM cohorts. Unexpectedly, higher pre-treatment polio neutralizing antibody titers were also associated with longer survival after PVSRIPO therapy in two independent clinical cohorts. PBMCs from patients surviving longer after PVSRIPO therapy mounted higher TNF, but lower IFN-a responses after in vitro challenge with PVSRIPO. In analysis of paired primary vs recurrent GBM tumors, we discovered that patients with low TMB upon recurrence were more likely to experience increased tumor inflammation and suppression of overall TMB. Low TMB in rGBM tumors was also associated with neoantigen depletion. Collectively, these observations imply that patients with low TMB and/or shorter duration of standard of care therapy may have intact immune surveillance, and that pre-treatment immunological status may dictate survival response to polio virotherapy.

739 Intratumor childhood vaccine-specific CD4+ T cell recall helps antitumor CD8 T cells

Background

PVSRIPO, a recombinant poliovirus derived from the live-attenuated Sabin oral polio vaccine strain, is being tested in multi-institutional phase II clinical trials for recurrent glioblastoma (NCT04479241) and unresectable, PD-1 refractory melanoma (NCT04577807) in combination with PD1 blockade. PVSRIPO capsid is identical to the Sabin vaccine strain and >99% identical to the inactivated Polio vaccine (IPOL, Salk), against which public health mandated childhood vaccination is near universal. In non-vaccinated mice, PVSRIPO mediates antitumor efficacy in a replication-dependent manner via engaging innate inflammation and antitumor T cells. Accordingly, it is anticipated that pre-existing immunity to PVSRIPO impedes antitumor therapy. However, recent evidence indicates that immunological 'recall', or reactivation of memory T cells, may mediate anti-tumor effects.

Methods

The impact of prior polio vs control (KLH) vaccination on intratumor viral replication, tumor inflammation, and overall tumor growth after intratumor PVSRIPO therapy was assessed in murine tumor models. The role of polio capsid and tetanus recall antigens in mediating intratumor inflammation and antitumor efficacy was similarly studied in mice non-permissive to PVSRIPO infection. To mechanistically define antitumor effects of polio recall, B cell and CD8 T cell knockout mice were used, in addition to adoptive transfer of CD4+ T cells from vaccinated mice. Intratumor polio or tetanus recall antigen therapy was performed after OT-I transfer (OVA-specific T cells) in the B16-OVA melanoma model to gauge antitumor T cell activity. Lastly, the inflammatory effects of polio and tetanus antigens was tested in human peripheral blood mononuclear cells (PBMCs).

Results

Despite curtailing intratumor viral replication, prior polio vaccination in mice potentiated subsequent antitumor efficacy of PVSRIPO. Intratumor recall responses induced by polio and tetanus antigens also delayed tumor growth. Recall antigen therapy was associated with marked intratumor influx of eosinophils, conventional CD4+ T cells, and increased expression of IFN-g, TNF, and Granzyme B in tumor infiltrating T cells. The antitumor efficacy of polio recall antigen was mediated by CD4+ T cells, partially depended upon CD8+ T cells, and was impaired by B cells. Both polio and tetanus recall antigen therapy bolstered the antitumor function of tumor-specific OT-I CD8+ T cells. Polio and tetanus antigens induced CXCL10 and type I/II/III IFNs in PBMCs in vitro.

Conclusions

Childhood vaccine-specific CD4+ T cells hold cancer immunotherapy potential. In the context of PVSRIPO therapy, antitumor and inflammatory effects of polio vaccine-specific CD4+ T cell recall supersedes inhibitory effects of attenuated intratumor viral replication, and represents a novel mechanism of action.

Ethics Approval

The animal work described in this study was approved by the Duke University IACUC.

BIOL-10. DISTRIBUTION AND VULNERABILITY OF TRANSCRIPTIONAL OUTPUTS ACROSS THE GENOME IN MYC-AMPLIFIED MEDULLOBLASTOMA CELLS

Myc plays a central role in tumorigenesis by orchestrating the expression of genes essential to numerous cellular processes. While it is well established that Myc functions by binding to its target genes to regulate their transcription, the distribution of the transcriptional output across human genome in Myc-amplified cancer cells, and the susceptibility of such transcriptional outputs to therapeutic interferences remain to be fully elucidated. Here, we analyze the distribution of transcriptional outputs in Myc-amplified medulloblastoma (MB) cells by profiling nascent total RNAs within a temporal context. This profiling reveals a major portion of transcriptional action in these cells was directed at the genes fundamental to cellular infrastructures, including rRNAs and particularly those in the mitochondrial genome (mtDNA). Notably, even when Myc protein was depleted by as much as 80%, the impact on transcriptional outputs across the genome was limited, with notable reduction mostly in genes of involved in ribosomal biosynthesis, genes residing in mtDNA or encoding mitochondria-localized proteins, and those encoding histones. In contrast to the limited direct impact of Myc depletion, we found that the global transcriptional outputs were highly dependent on the activity of Inosine Monophosphate Dehydrogenases (IMPDHs), rate limiting enzymes for de novo guanine nucleotide synthesis and whose expression in tumor cells was positively correlated with Myc’s expression. Blockage of IMPDHs attenuated the global transcriptional outputs with a particularly strong inhibitory effect on the aforementioned infrastructure genes, which was accompanied by the abrogation of MB cell’s proliferation in vitro and in vivo. Together, our findings reveal a real time action of Myc as a transcriptional factor in tumor cells, gain new insight into the pathogenic mechanism underlying Myc-driven tumorigenesis, and support IMPDHs as a therapeutic vulnerability in MB cells empowered by a high level of Myc oncoprotein.

THER-05. GENETICALLY STABLE POLIOVIRUS VECTOR CARRYING H3.3K27M ANTIGEN FOR TREATMENT OF DIFFUSE MIDLINE GLIOMA BY INTRAMUSCULAR INJECTION

BACKGROUND

H3 K27M-mutant diffuse midline glioma (DMG) is invariably lethal. Viruses naturally engage innate immunity, induce antigen presentation, and mediate CD8 T cell priming against foreign antigens. Polioviruses, in particular, are uniquely tropic for dendritic cells (DC) and potently activate DC, inducing Th1-dominant cytokine profiles, CD8 T cell immunity, and enhanced epitope presentation. Thus, poliovirus is ideally suited for vectored delivery of signature tumor neoantigens, e.g. the H3 K27M feature of DMG. However, poliovirus vector design is inherently limited by genetic instability and the underlying neuropathogenicity of poliovirus.

METHODS

We created a genetically stable, polio:rhinovirus chimera vector devoid of neuropathogenicity and modified for stable expression of the HLA-A2 restricted H3.3 K27M antigen (RIPO (H3.3)).

RESULTS

RIPO(H3.3) infects, activates, and induces H3.3K27M antigen presentation in DCs in vitro. Given intramuscularly in vivo, RIPO(H3.3) recruits and activates DCs with Th1-dominant cytokine profiles, efficiently primes H3.3K27M-specific CD8 T cells, induces antigen-specific CD8 T cell migration to the tumor site, delays tumor growth, and enhances survival in murine tumor models.

CONCLUSION

This novel approach leverages the unique ability of polioviruses to activate DCs while simultaneously introducing the H3.3 K27M antigen. In this way, DCs are activated optimally in situ, while being simultaneously infected to express/present tumor antigen. RIPO(H3.3), given by intramuscular injection, will be evaluated in a clinical trial for children with H3 K27M-mutant diffuse midline glioma.

TMOD-17. ONCOLYTIC POLIOVIRUS AS A PROBE FOR MECHANISMS OF IMMUNE RESISTANCE IN GLIOBLASTOMA

Glioblastoma is an immunologically cold tumor that employs multiple resistance mechanisms to escape activation of both innate and adaptive immunity. Transient immune activation and responses induced by immunotherapies in glioblastoma are almost always suppressed by potent intrinsic and extrinsic resistance mechanisms, limiting therapeutic efficacy. PVSRIPO is a recombinant poliovirus/rhinovirus chimera that elicits robust innate immune activation through infecting and lysing glioblastoma cells expressing the human poliovirus receptor (CD155), and through sublethal infection of tumor associated macrophages, setting the stage for lingering viral replication. We have constructed a pathological model using murine CT2ACD155 intracranial tumors stereotactically infused with a single dose of PVSRIPO (5×107 pfu) or mock control to study the response of glioblastoma in an immune competent in vivo tumor microenvironment. Tumor bearing brains were harvested before therapy and on days 2, 4, 6, 9, 12 post therapy for histology. On day 4 post therapy, histopathology showed > 90% shrinkage of intracranial tumor diameter (n= 6/6). Residual tumor is surrounded by a necrotic rim on histology with profound infiltration of F4/80+ macrophages and substantially reduced expression of Ki-67 on IHC compared to controls. Other markers including CD3, CD8 and Cleaved-Caspase 3 showed no difference in tumors of both groups. Interestingly, most residual tumor progressed on day 6 post therapy and caught up with the size of control tumor by day 9. Long-term remission of tumor occurred in only 10% of mice. This model is valuable in studying the mechanisms of intrinsic resistance and the timeline of resistance to develop in order to obtain more durable immunotherapy responses in glioblastoma. Future steps are to track inflammation induced by viral infection by RNAseq, and to find potential therapeutic interventions (including blocking PD1:PD-L1 axis) to mitigate therapy resistance in this preclinical model.

CTIM-23. A PHASE 1 TRIAL OF D2C7-IT IN COMBINATION WITH ATEZOLIZUMAB IN RECURRENT WHO GRADE IV MALIGNANT GLIOMA (MG)

BACKGROUND

D2C7 immunotoxin (D2C7-IT) is a dual-specific recombinant immunotoxin comprising an EGFR-wt and mutant-specific (EGFRvIII) monoclonal antibody fragment and a genetically engineered form of the pseudomonas exotoxin. When injected directly into the tumor mass by convection enhanced delivery (CED), in addition to direct tumor cell killing, immunotoxins induce secondary immune responses by the activation of CD4+ and CD8+ T-cells. We completed a phase 1 dose escalation study of D2C7-IT injected by CED into recurrent WHO grade III-IV MG and identified the phase 2 dose (6,920 ng/mL). Three patients remain in partial response more than 58, 38, and 32 months after a single D2C7-IT intratumoral infusion. As optimal induction of anti-tumor immune responses by immunotoxins is impeded by potent MG-mediated immunosuppression, we are assessing the toxicity of the combination of D2C7-IT with atezolizumab in patients with recurrent WHO grade IV MG.

METHODS

Eligibility includes adult patients with recurrence of a solitary supratentorial WHO grade IV MG; ≥4 weeks after chemotherapy, bevacizumab or study drug; adequate organ function; and KPS >70%. Patient receives an intratumoral infusion of D2C7-IT and initiates two weeks later atezolizumab at 1200mg, followed by atezolizumab every 3 weeks for up to 2 years. Two cohorts of 3 patients are initially accrued to assess the toxicity of the combination. Assuming accrual continues after the initial two cohorts of 3 patients, an additional 12 patients will be accrued to the study.

RESULT

The first enrolled patient experienced a grade 3 DLT (grade III ALT elevation) after the first infusion of atezolizumab, but showed a more extensive immunotherapeutic effect by imaging than observed with patients on the D2C7-IT monotherapy trial. Enrollment is ongoing.

CONCLUSION

D2C7-IT monotherapy has shown prolonged survival and disease control in some patients. We are now evaluating the combination of D2C7-IT with checkpoint inhibition.

Pan-cancer analysis of whole genomes

Cancer is driven by genetic change, and the advent of massively parallel sequencing has enabled systematic documentation of this variation at the whole-genome scale1-3. Here we report the integrative analysis of 2,658 whole-cancer genomes and their matching normal tissues across 38 tumour types from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA). We describe the generation of the PCAWG resource, facilitated by international data sharing using compute clouds. On average, cancer genomes contained 4-5 driver mutations when combining coding and non-coding genomic elements; however, in around 5% of cases no drivers were identified, suggesting that cancer driver discovery is not yet complete. Chromothripsis, in which many clustered structural variants arise in a single catastrophic event, is frequently an early event in tumour evolution; in acral melanoma, for example, these events precede most somatic point mutations and affect several cancer-associated genes simultaneously. Cancers with abnormal telomere maintenance often originate from tissues with low replicative activity and show several mechanisms of preventing telomere attrition to critical levels. Common and rare germline variants affect patterns of somatic mutation, including point mutations, structural variants and somatic retrotransposition. A collection of papers from the PCAWG Consortium describes non-coding mutations that drive cancer beyond those in the TERT promoter4; identifies new signatures of mutational processes that cause base substitutions, small insertions and deletions and structural variation5,6; analyses timings and patterns of tumour evolution7; describes the diverse transcriptional consequences of somatic mutation on splicing, expression levels, fusion genes and promoter activity8,9; and evaluates a range of more-specialized features of cancer genomes8,10-18.

ATIM-24. DOSE FINDING AND DOSE EXPANSION TRIAL OF D2C7 IMMUNOTOXIN (D2C7-IT) ADMINISTERED INTRATUMORALLY VIA CONVECTION-ENHANCED DELIVERY (CED) FOR RECURRENT MALIGNANT GLIOMA (MG)

BACKGROUND

D2C7-IT is a recombinant immunotoxin comprised of a dual-specific antibody fragment targeting EGFRwt and EGFRvIII and a genetically engineered form of the Pseudomonas exotoxin, PE38-KDEL. We report results of a phase 1 trial, with dose expansion at the selected phase 2 dose, evaluating D2C7-IT delivered intratumorally by CED.

METHODS

Eligible patients are adults with recurrent supratentorial WHO grade III or IV MG; solitary tumor; ≥4 weeks after chemotherapy, bevacizumab or study drug; adequate organ function; and KPS >70%. Two patients per dose level (DL) were to be enrolled in the dose escalation portion (dose range: 40ng/mL to 23,354ng/mL), followed by dose expansion at the selected phase 2 dose (DL13).

RESULTS

As of 6/07/2019, 51 patients have been treated; 10 patients on the phase 2 dose. Observed dose limiting toxicities include: grade 4 seizure (n=1) on DL3, grade 3 confusion and pyramidal tract syndrome (n=1) on DL13, and grade 4 cerebral edema (n=1) and grade 3 dysphasia (n=1) on DL17. Grade 3 or higher adverse events possibly related to D2C7-IT include: seizure (grade 4, n=2, grade 3, n=3), cerebral edema (grade 4, n=1), hydrocephalus (grade 3, n=5), headache (grade 3, n=4), hemiparesis (grade 3, n=4), dysphasia (grade 3, n=4), lymphopenia (grade 3, n=3), thromboembolic event (grade 3, n=3); and one each of grade 3 elevated ALT, urinary tract infection, fall, wound complication, generalized muscle weakness, confusion, encephalopathy, and somnolence. Fourteen patients are alive. Three patients have partial radiographic response and remain alive without additional therapy more than 46, 27 and 21 months after D2C7-IT infusion.

CONCLUSION

Dose level 13 was selected as the optimal phase 2 dose and patient accrual is ongoing on the dose expansion arm. Encouraging efficacy results have been observed. A trial of D2C7-IT with checkpoint inhibitor is planned to start in the near future.

EXTH-51. GENETICALLY STABLE POLIOVIRUS VECTOR PLATFORM FOR DIPG IMMUNOTHERAPY

Options for the immunotherapy of diffuse intrinsic pontine glioma (DIPG), due to its anatomical location and inherent therapy resistance, are limited. The histone 3.3(K27M) mutation in ~80% of such tumors offers a unique opportunity for immunotherapy intervention, as it defines a high affinity, HLA-A2-restricted tumor neoantigen that spontaneously elicits CD8+ T cell responses in DIPG patients. Immunizing against the H3.3(K27M) signature in the clinic has been challenging, as conventional approaches (i.e. peptide-conjugates administered with adjuvants) lack the costimulatory signals known to drive CD8+ effector T cell responses. Therefore, we built on a viral vector approach for engaging innate immune responses to virus infection specifically in antigen presenting cells. Viruses naturally engage innate immunity, induce antigen presentation, and mediate CD8 T cell priming against foreign antigens. Polioviruses can provide a context optimal for generating antigen-specific CD8 T cells, as they have natural tropism for dendritic cells, preeminent inducers of CD8 T cell immunity; elicit Th1-promoting inflammation; and lack interference with innate or adaptive immunity. However, notorious genetic instability and underlying neuropathogenicity has hampered poliovirus-based vector applications. We devised a strategy based on the polio:rhinovirus chimera PVSRIPO, devoid of viral neuropathogenicity after intracerebral inoculation in human subjects, for stable expression of exogenous antigens. PVSRIPO vectors infect, activate, and induce epitope presentation in DCs in vitro; recruit and activate DCs with Th1-dominant cytokine profiles at the injection site in vivo. They efficiently prime tumor antigen-specific CD8 T cells in vivo, induce CD8 T cell migration to the tumor site, delay tumor growth and enhance survival in syngeneic rodent tumor models. We are preparing a prototype PVSRIPO-derived vector delivering the H3.3(K27M) signature for clinical investigation.

Poliovirus Receptor (CD155) Expression in Pediatric Brain Tumors Mediates Oncolysis of Medulloblastoma and Pleomorphic Xanthoastrocytoma

Poliovirus oncolytic immunotherapy is a putatively novel approach to treat pediatric brain tumors. This work sought to determine expression of the poliovirus receptor (PVR), CD155, in low-grade and malignant pediatric brain tumors and its ability to infect, propagate, and inhibit cell proliferation. CD155 expression in pleomorphic xanthoastrocytoma (PXA), medulloblastoma, atypical teratoid rhabdoid tumor, primitive neuroectodermal tumor, and anaplastic ependymoma specimens was assessed. The ability of the polio: rhinovirus recombinant, PVSRIPO, to infect PXA (645 [BRAF V600E mutation], 2363) and medulloblastoma (D283, D341) cells were determined by viral propagation measurement and cell proliferation. PVR mRNA expression was evaluated in 763 medulloblastoma and 1231 normal brain samples. CD155 was expressed in all 12 patient specimens and in PXA and medulloblastoma cell lines. One-step growth curves at a multiplicity of infection of 10 demonstrated productive infection and peak plaque formation units at 5-10 hours. PVSRIPO infection significantly decreased cellular proliferation in 2363, 645, and D341 cell lines at 48 hours (p < 0.05) and resulted in cell death. PVR expression was highest in medulloblastoma subtypes Group 3γ, WNTα, and WNTβ (p < 0.001). This proof-of-concept in vitro study demonstrates that PVSRIPO is capable of infecting, propagating, prohibiting cell proliferation, and killing PXA and Group 3 medulloblastoma.

Abstract A47: Inhibition of histone methyltransferase with novel epipolythiodioxopiperazines (ETP) alkaloids in pediatric high-grade glioma

Pediatric high-grade glioma (HGG), which ranks among the most lethal of cancers among children, contain two recurrent mutations within the histone H3 gene H3F3A (K27M and G34R/V). It is not known whether these mutations directly alter specific epigenetic regulatory mechanisms that may be involved in tumor proliferation and growth, or contribute to their therapeutic resistance and rapid recurrence. Preliminary findings from our lab have identified that Chaetocin (CH), a pan histone methyltransferase (HMT) inhibitor, and two newly synthesized epipolythiodioxopiperazines (ETP) alkaloids that are structurally and functionally distinct HMT inhibitors can effectively kill pediatric HGG (N=7), including three that contain the H3 K27M mutation. In addition, CH and ETPs can also inhibit tumor sphere formation and cell migration at low concentrations (&lt;0.5uM) through in vitro studies. In H3 K27M tumors, these HMTi were shown to reverse hypomethylation of K27me3 transiently (&lt;48 hours). We also found these HMTis to decrease tumor stem cell expression (Sox2, Oct4), and increase tumor-suppressor activity (PTEN, p53) for three pediatric HGG up to 72 hours. These results suggest ETP effectiveness that acts through direct HMT inhibition may also influence other tumorigenic pathways, which may in part contribute to their aggressive nature. Results from this investigation have advanced our understanding of epigenetic regulation of solid brain tumors, and identified drugs that may specifically target pediatric H3 K27M tumor subtypes and could ultimately improve overall survival in pediatric patients with high-grade glioma.

Citation Format: Stephen Dombrowski, Thamara Abou-Antoun, Anthony Mansour, Rewa Arar, Mohammad Movassaghi, Erin Murphy, Shuvojit Banerjee, Xiao-Nan Li, Peter Houghton, Darell Bigner, Angel Montero Carcaboso, Jeremy Rich, Violette Recinos. Inhibition of histone methyltransferase with novel epipolythiodioxopiperazines (ETP) alkaloids in pediatric high-grade glioma [abstract]. In: Proceedings of the AACR Special Conference: Pediatric Cancer Research: From Basic Science to the Clinic; 2017 Dec 3-6; Atlanta, Georgia. Philadelphia (PA): AACR; Cancer Res 2018;78(19 Suppl):Abstract nr A47.

Abstract LB-302: Oncolytic poliovirus-mediated inflammation and immunity in breast cancer

Breast cancer is the most common malignancy among women in the United States and the second leading cause of cancer deaths in women. Disease prognosis, local recurrence rates and response to therapy are all related to receptor status of the individual cancer. Triple negative breast cancer (TNBC; ER-, PR-, and HER2-negative) is a subset of breast cancer with poor prognosis and high rate of local recurrence and metastasis with limited therapeutic options.

The oncolytic poliovirus, PVSRIPO, is a recombinant polio:rhinovirus chimera that has been engineered to eliminate neuropathogenicity and has demonstrated promising responses in patients with recurrent glioblastoma. PVSRIPO is oncolytic in tumor cells and enters cells via its receptor CD155, which is overexpressed in solid malignancies, including TNBC. Treatment of primary TNBC tissue with PVSRIPO induces IFN-I and pro-inflammatory cytokines. In a murine immunotherapy model, intratumoral injection of PVSRIPO induces IFN-I, IL12 and IFN-γ and rapid recruitment of neutrophils into the tumor followed by infiltration of DCs, T cells and B cells. Treatment of orthotopic breast tumors with PVSRIPO results in delayed tumor growth and increased survival in an immunocompetent murine immunotherapy model.

We have designed a pilot study to examine the effects of intratumoral injection of PVSRIPO into six women with TNBC with at least 1 cm of residual tumor after chemotherapy. The primary exploratory objective is to describe the changes in the amount of tumor infiltrating immune cells in tumor tissue pre- and post-injection of PVSRIPO. We will also collect blood at different time points during PVSRIPO treatment to evaluate inflammatory and immune signatures. We hypothesize that intratumoral administration of PVSRIPO will cause inflammation thus stimulating innate and adaptive immune activation.

This study is funded by the Department of Defense Breast Cancer Research Program award W81XWH-16-1-0354 (PI, Smita Nair).

Citation Format: Karenia Landa, Eda Holl, Michael Brown, James Herndon, Lars Grimm, Jeremy Force, Darell Bigner, Shelley Hwang, Matthias Gromeier, Smita Nair. Oncolytic poliovirus-mediated inflammation and immunity in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-302.

AJAP1 expression modulates glioma cell motility and correlates with tumor growth and survival

Glioblastoma multiforme (GBM) is one of the most common primary malignant brain tumors. Unraveling the molecular and genetic complexity that determines GBM's pronounced migratory property could provide new options for therapeutic targeting that may significantly complement current surgical and chemoradiation therapy and alter the current poor outcome. In this study, we establish stable AJAP1 overexpressing glioma cells in order to examine in vivo tumor growth. We examine AJAP1 localization by confocal microscopy and AJAP1's functional effect on migration and invasion across surfaces coated with laminin. Finally, analysis of AJAP1 expression in murine xenografts and GBM primary tumors revealed its association with tumor growth and survival. Stable overexpression of AJAP1 promotes adherence, decreases invasion of glioma cells through an extracellular-like matrix, and slows migration in the presence of laminin. These observations are reversed by gene knockdown using multiple siRNAs. Additionally, overexpression of AJAP1 decreases colony formation in glioma cells, and leads to smaller tumor growth with increased survival in glioma xenograft mice. Loss of AJAP1 protein expression predicts worse survival in GBM patients. AJAP1 overexpression decreases cell motility in the presence of laminin and decreases tumor growth in xenografts. Its loss of expression predicts worse survival in patients. This study extends our prior observations and implicates AJAP1 as a potential prognostic marker and a viable target for therapeutic intervention in GBM.

Patient-derived DIPG cells preserve stem-like characteristics and generate orthotopic tumors

Diffuse intrinsic pontine glioma (DIPG) is a devastating brain tumor, with a median survival of less than one year. Due to enormous difficulties in the acquisition of DIPG specimens and the sophisticated technique required to perform brainstem orthotopic injection, only a handful of DIPG pre-clinical models are available. In this study, we successfully established eight patient-derived DIPG cell lines, mostly derived from treatment-naïve surgery or biopsy specimens. These patient-derived cell lines can be stably passaged in serum-free neural stem cell media and displayed distinct morphologies, growth rates and chromosome abnormalities. In addition, these cells retained genomic hallmarks identical to original human DIPG tumors. Notably, expression of several neural stem cell lineage markers was observed in DIPG cell lines. Moreover, three out of eight cell lines can form orthotopic tumors in mouse brainstem by stereotactic injection and these tumors faithfully represented the characteristics of human DIPG by magnetic resonance imaging (MRI) and histopathological staining. Taken together, we established DIPG pre-clinical models resembling human DIPG and they provided a valuable resource for future biological and therapeutic studies.

Long-term Survival in Glioblastoma with Cytomegalovirus pp65-Targeted Vaccination

Purpose: Patients with glioblastoma have less than 15-month median survival despite surgical resection, high-dose radiation, and chemotherapy with temozolomide. We previously demonstrated that targeting cytomegalovirus pp65 using dendritic cells (DC) can extend survival and, in a separate study, that dose-intensified temozolomide (DI-TMZ) and adjuvant granulocyte macrophage colony-stimulating factor (GM-CSF) potentiate tumor-specific immune responses in patients with glioblastoma. Here, we evaluated pp65-specific cellular responses following DI-TMZ with pp65-DCs and determined the effects on long-term progression-free survival (PFS) and overall survival (OS).Experimental Design: Following standard-of-care, 11 patients with newly diagnosed glioblastoma received DI-TMZ (100 mg/m2/d × 21 days per cycle) with at least three vaccines of pp65 lysosome-associated membrane glycoprotein mRNA-pulsed DCs admixed with GM-CSF on day 23 ± 1 of each cycle. Thereafter, monthly DI-TMZ cycles and pp65-DCs were continued if patients had not progressed.Results: Following DI-TMZ cycle 1 and three doses of pp65-DCs, pp65 cellular responses significantly increased. After DI-TMZ, both the proportion and proliferation of regulatory T cells (Tregs) increased and remained elevated with serial DI-TMZ cycles. Median PFS and OS were 25.3 months [95% confidence interval (CI), 11.0-∞] and 41.1 months (95% CI, 21.6-∞), exceeding survival using recursive partitioning analysis and matched historical controls. Four patients remained progression-free at 59 to 64 months from diagnosis. No known prognostic factors [age, Karnofsky performance status (KPS), IDH-1/2 mutation, and MGMT promoter methylation] predicted more favorable outcomes for the patients in this cohort.Conclusions: Despite increased Treg proportions following DI-TMZ, patients receiving pp65-DCs showed long-term PFS and OS, confirming prior studies targeting cytomegalovirus in glioblastoma. Clin Cancer Res; 23(8); 1898-909. ©2017 AACR.

Exomic sequencing of four rare central nervous system tumor types

A heterogeneous population of uncommon neoplasms of the central nervous system (CNS) cause significant morbidity and mortality. To explore their genetic origins, we sequenced the exomes of 12 pleomorphic xanthoastrocytomas (PXA), 17 non-brainstem pediatric glioblastomas (PGBM), 8 intracranial ependymomas (IEP) and 8 spinal cord ependymomas (SCEP). Analysis of the mutational spectra revealed that the predominant single base pair substitution was a C:G>T:A transition in each of the four tumor types. Our data confirm the critical roles of several known driver genes within CNS neoplasms, including TP53 and ATRX in PGBM, and NF2 in SCEPs. Additionally, we show that activating BRAF mutations play a central role in both low and high grade glial tumors. Furthermore, alterations in genes coding for members of the mammalian target of rapamycin (mTOR) pathway were observed in 33% of PXA. Our study supports the hypothesis that pathologically similar tumors arising in different age groups and from different compartments may represent distinct disease processes with varied genetic composition.

Recurrence patterns across medulloblastoma subgroups: an integrated clinical and molecular analysis

BACKGROUND

Recurrent medulloblastoma is a therapeutic challenge because it is almost always fatal. Studies have confirmed that medulloblastoma consists of at least four distinct subgroups. We sought to delineate subgroup-specific differences in medulloblastoma recurrence patterns.

METHODS

We retrospectively identified a discovery cohort of all recurrent medulloblastomas at the Hospital for Sick Children (Toronto, ON, Canada) from 1994 to 2012 (cohort 1), and established molecular subgroups using a nanoString-based assay on formalin-fixed paraffin-embedded tissues or frozen tissue. The anatomical site of recurrence (local tumour bed or leptomeningeal metastasis), time to recurrence, and survival after recurrence were assessed in a subgroup-specific manner. Two independent, non-overlapping cohorts (cohort 2: samples from patients with recurrent medulloblastomas from 13 centres worldwide, obtained between 1991 and 2012; cohort 3: samples from patients with recurrent medulloblastoma obtained at the NN Burdenko Neurosurgical Institute [Moscow, Russia] between 1994 and 2011) were analysed to confirm and validate observations. When possible, molecular subgrouping was done on tissue obtained from both the initial surgery and at recurrence.

RESULTS

Cohort 1 consisted of 30 patients with recurrent medulloblastomas; nine with local recurrences, and 21 with metastatic recurrences. Cohort 2 consisted of 77 patients and cohort 3 of 96 patients with recurrent medulloblastoma. Subgroup affiliation remained stable at recurrence in all 34 cases with available matched primary and recurrent pairs (five pairs from cohort 1 and 29 pairs from cohort 2 [15 SHH, five group 3, 14 group 4]). This finding was validated in 17 pairs from cohort 3. When analysed in a subgroup-specific manner, local recurrences in cohort 1 were more frequent in SHH tumours (eight of nine [89%]) and metastatic recurrences were more common in group 3 and group 4 tumours (17 of 20 [85%] with one WNT, p=0·0014, local vs metastatic recurrence, SHH vs group 3 vs group 4). The subgroup-specific location of recurrence was confirmed in cohort 2 (p=0·0013 for local vs metastatic recurrence, SHH vs group 3 vs group 4,), and cohort 3 (p<0·0001). Treatment with craniospinal irradiation at diagnosis was not significantly associated with the anatomical pattern of recurrence. Survival after recurrence was significantly longer in patients with group 4 tumours in cohort 1 (p=0·013) than with other subgroups, which was confirmed in cohort 2 (p=0·0075), but not cohort 3 (p=0·70).

INTERPRETATION

Medulloblastoma does not change subgroup at the time of recurrence, reinforcing the stability of the four main medulloblastoma subgroups. Significant differences in the location and timing of recurrence across medulloblastoma subgroups have potential treatment ramifications. Specifically, intensified local (posterior fossa) therapy should be tested in the initial treatment of patients with SHH tumours. Refinement of therapy for patients with group 3 or group 4 tumours should focus on metastases.