LRIG1 engages ligand VISTA and impairs tumor-specific CD8+ T cell responses.

Immune checkpoint blockade is a promising approach to activate antitumor immunity and improve the survival of patients with cancer. V-domain immunoglobulin suppressor of T cell activation (VISTA) is an immune checkpoint target; however, the downstream signaling mechanisms are elusive. Here, we identify leucine-rich repeats and immunoglobulin-like domains 1 (LRIG1) as a VISTA binding partner, which acts as an inhibitory receptor by engaging VISTA and suppressing T cell receptor signaling pathways. Mice with T cell-specific LRIG1 deletion developed superior antitumor responses because of expansion of tumor-specific cytotoxic T lymphocytes (CTLs) with increased effector function and survival. Sustained tumor control was associated with a reduction of quiescent CTLs (TCF1+ CD62Lhi PD-1low) and a reciprocal increase in progenitor and memory-like CTLs (TCF1+ PD-1+). In patients with melanoma, elevated LRIG1 expression on tumor-infiltrating CD8+ CTLs correlated with resistance to immunotherapies. These results delineate the role of LRIG1 as an inhibitory immune checkpoint receptor and propose a rationale for targeting the VISTA/LRIG1 axis for cancer immunotherapy.

VISTA promotes the metabolism and differentiation of myeloid-derived suppressor cells by STAT3 and polyamine-dependent mechanisms.

Myeloid-derived suppressor cells (MDSCs) impair antitumor immune responses. Identifying regulatory circuits during MDSC development may bring new opportunities for therapeutic interventions. We report that the V-domain suppressor of T cell activation (VISTA) functions as a key enabler of MDSC differentiation. VISTA deficiency reduced STAT3 activation and STAT3-dependent production of polyamines, which causally impaired mitochondrial respiration and MDSC expansion. In both mixed bone marrow (BM) chimera mice and myeloid-specific VISTA conditional knockout mice, VISTA deficiency significantly reduced tumor-associated MDSCs but expanded monocyte-derived dendritic cells (DCs) and enhanced T cell-mediated tumor control. Correlated expression of VISTA and arginase-1 (ARG1), a key enzyme supporting polyamine biosynthesis, was observed in multiple human cancer types. In human endometrial cancer, co-expression of VISTA and ARG1 on tumor-associated myeloid cells is associated with poor survival. Taken together, these findings unveil the VISTA/polyamine axis as a central regulator of MDSC differentiation and warrant therapeutically targeting this axis for cancer immunotherapy.

Phase II Clinical Trial of Pembrolizumab and Chemotherapy Reveals Distinct Transcriptomic Profiles by Radiologic Response in Metastatic Triple-Negative Breast Cancer.

PURPOSE: A single arm, phase II trial of carboplatin, nab-paclitaxel, and pembrolizumab (CNP) in metastatic triple-negative breast cancer (mTNBC) was designed to evaluate overall response rate (ORR), progression-free survival (PFS), duration of response (DOR), safety/tolerability, overall survival (OS), and identify pathologic and transcriptomic correlates of response to therapy. PATIENTS AND METHODS: Patients with ≤2 prior therapies for metastatic disease were treated with CNP regardless of tumor programmed cell death-ligand 1 status. Core tissue biopsies were obtained prior to treatment initiation. ORR was assessed using a binomial distribution. Survival was analyzed via the Kaplan-Meier method. Bulk RNA sequencing was employed for correlative studies. RESULTS: Thirty patients were enrolled. The ORR was 48.0%: 2 (7%) complete responses (CR), 11 (41%) partial responses (PR), and 8 (30%) stable disease (SD). The median DOR for patients with CR or PR was 6.4 months [95% confidence interval (CI), 4-8.5 months]. For patients with CR, DOR was >24 months. Overall median PFS and OS were 5.8 (95% CI, 4.7-8.5 months) and 13.4 months (8.9-17.3 months), respectively. We identified unique transcriptomic landscapes associated with each RECIST category of radiographic treatment response. In CR and durable PR, IGHG1 expression was enriched. IGHG1high tumors were associated with improved OS (P = 0.045) and were concurrently enriched with B cells and follicular helper T cells, indicating IGHG1 as a promising marker for lymphocytic infiltration and robust response to chemo-immunotherapy. CONCLUSIONS: Pretreatment tissue sampling in mTNBC treated with CNP reveals transcriptomic signatures that may predict radiographic responses to chemo-immunotherapy.

The MIF promoter SNP rs755622 is associated with immune activation in glioblastoma.

Intratumoral heterogeneity is a defining hallmark of glioblastoma, driving drug resistance and ultimately recurrence. Many somatic drivers of microenvironmental change have been shown to affect this heterogeneity and, ultimately, the treatment response. However, little is known about how germline mutations affect the tumoral microenvironment. Here, we find that the single-nucleotide polymorphism (SNP) rs755622 in the promoter of the cytokine macrophage migration inhibitory factor (MIF) is associated with increased leukocyte infiltration in glioblastoma. Furthermore, we identified an association between rs755622 and lactotransferrin expression, which could also be used as a biomarker for immune-infiltrated tumors. These findings demonstrate that a germline SNP in the promoter region of MIF may affect the immune microenvironment and further reveal a link between lactotransferrin and immune activation.

Spatial immunosampling of MRI-defined glioblastoma regions reveals immunologic fingerprint of non-contrast enhancing, infiltrative tumor margins.

Glioblastoma (GBM) treatment includes maximal safe resection of the core and MRI contrast-enhancing (CE) tumor. Complete resection of the infiltrative non-contrast-enhancing (NCE) tumor rim is rarely achieved. We established a safe, semi-automated workflow for spatially-registered sampling of MRI-defined GBM regions in 19 patients with downstream analysis and biobanking, enabling studies of NCE, wherefrom recurrence/progression typically occurs. Immunophenotyping revealed underrepresentation of myeloid cell subsets and CD8+ T cells in the NCE. While NCE T cells phenotypically and functionally resembled those in matching CE tumor, subsets of activated (CD69hi) effector memory CD8+ T cells were overrepresented. Contrarily, CD25hi Tregs and other subsets were underrepresented. Overall, our study demonstrated that MRI-guided, spatially-registered, intraoperative immunosampling is feasible as part of routine GBM surgery. Further elucidation of the shared and spatially distinct microenvironmental biology of GBM will enable development of therapeutic approaches targeting the NCE infiltrative tumor to decrease GBM recurrence.

Unsupervised and supervised AI on molecular dynamics simulations reveals complex characteristics of HLA-A2-peptide immunogenicity.

Immunologic recognition of peptide antigens bound to class I major histocompatibility complex (MHC) molecules is essential to both novel immunotherapeutic development and human health at large. Current methods for predicting antigen peptide immunogenicity rely primarily on simple sequence representations, which allow for some understanding of immunogenic features but provide inadequate consideration of the full scale of molecular mechanisms tied to peptide recognition. We here characterize contributions that unsupervised and supervised artificial intelligence (AI) methods can make toward understanding and predicting MHC(HLA-A2)-peptide complex immunogenicity when applied to large ensembles of molecular dynamics simulations. We first show that an unsupervised AI method allows us to identify subtle features that drive immunogenicity differences between a cancer neoantigen and its wild-type peptide counterpart. Next, we demonstrate that a supervised AI method for class I MHC(HLA-A2)-peptide complex classification significantly outperforms a sequence model on small datasets corrected for trivial sequence correlations. Furthermore, we show that both unsupervised and supervised approaches reveal determinants of immunogenicity based on time-dependent molecular fluctuations and anchor position dynamics outside the MHC binding groove. We discuss implications of these structural and dynamic immunogenicity correlates for the induction of T cell responses and therapeutic T cell receptor design.

Functional landscapes of POLE and POLD1 mutations in checkpoint blockade-dependent antitumor immunity.

Defects in pathways governing genomic fidelity have been linked to improved response to immune checkpoint blockade therapy (ICB). Pathogenic POLE/POLD1 mutations can cause hypermutation, yet how diverse mutations in POLE/POLD1 influence antitumor immunity following ICB is unclear. Here, we comprehensively determined the effect of POLE/POLD1 mutations in ICB and elucidated the mechanistic impact of these mutations on tumor immunity. Murine syngeneic tumors harboring Pole/Pold1 functional mutations displayed enhanced antitumor immunity and were sensitive to ICB. Patients with POLE/POLD1 mutated tumors harboring telltale mutational signatures respond better to ICB than patients harboring wild-type or signature-negative tumors. A mutant POLE/D1 function-associated signature-based model outperformed several traditional approaches for identifying POLE/POLD1 mutated patients that benefit from ICB. Strikingly, the spectrum of mutational signatures correlates with the biochemical features of neoantigens. Alterations that cause POLE/POLD1 function-associated signatures generate T cell receptor (TCR)-contact residues with increased hydrophobicity, potentially facilitating T cell recognition. Altogether, the functional landscapes of POLE/POLD1 mutations shape immunotherapy efficacy.

Phenotypic and molecular states of IDH1 mutation-induced CD24-positive glioma stem-like cells.

Mutations in IDH1 and IDH2 drive the development of gliomas. These genetic alterations promote tumor cell renewal, disrupt differentiation states, and induce stem-like properties. Understanding how this phenotypic reprogramming occurs remains an area of high interest in glioma research. Previously, we showed that IDH mutation results in the development of a CD24-positive cell population in gliomas. Here, we demonstrate that this CD24-positive population possesses striking stem-like properties at the molecular and phenotypic levels. We found that CD24 expression is associated with stem-like features in IDH-mutant tumors, a patient-derived gliomasphere model, and a neural stem cell model of IDH1-mutant glioma. In orthotopic models, CD24-positive cells display enhanced tumor initiating potency compared to CD24-negative cells. Furthermore, CD24 knockdown results in changes in cell viability, proliferation rate, and gene expression that closely resemble a CD24-negative phenotype. Our data demonstrate that induction of a CD24-positive population is one mechanism by which IDH-mutant tumors acquire stem-like properties. These findings have significant implications for our understanding of the molecular underpinnings of IDH-mutant gliomas.

The Translocator Protein (TSPO) Genetic Polymorphism A147T Is Associated with Worse Survival in Male Glioblastoma Patients.

Glioblastoma (GBM) is the most common primary brain tumor in adults, with few available therapies and a five-year survival rate of 7.2%. Hence, strategies for improving GBM prognosis are urgently needed. The translocator protein 18kDa (TSPO) plays crucial roles in essential mitochondria-based physiological processes and is a validated biomarker of neuroinflammation, which is implicated in GBM progression. The TSPO gene has a germline single nucleotide polymorphism, rs6971, which is the most common SNP in the Caucasian population. High TSPO gene expression is associated with reduced survival in GBM patients; however, the relation between the most frequent TSPO genetic variant and GBM pathogenesis is not known. The present study retrospectively analyzed the correlation of the TSPO polymorphic variant rs6971 with overall and progression-free survival in GBM patients using three independent cohorts. TSPO rs6971 polymorphism was significantly associated with shorter overall survival and progression-free survival in male GBM patients but not in females in one large cohort of 441 patients. We observed similar trends in two other independent cohorts. These observations suggest that the TSPO rs6971 polymorphism could be a significant predictor of poor prognosis in GBM, with a potential for use as a prognosis biomarker in GBM patients. These results reveal for the first time a biological sex-specific relation between rs6971 TSPO polymorphism and GBM.

Bazedoxifene inhibits sustained STAT3 activation and increases survival in GBM.

An important factor correlated with poor survival in glioblastoma (GBM) is the aberrant and persistent activation of STAT3, a critical transcription factor that regulates multiple genes with key roles in cell survival, proliferation, resistance to chemotherapy, and stem cell maintenance. The Interleukin-6 (IL6)-STAT3 signaling axis has been studied extensively in inflammation and cancer. However, it is not completely understood how high levels of activated STAT3 are sustained in tumors. Previously, we identified a novel mechanism of biphasic activation of STAT3 in response to gp130-linked cytokines, including IL6, in which activation of STAT3 is prolonged by circumventing the negative regulatory mechanisms induced by its initial activationTo target prolonged STAT3 activation, we used the small molecule inhibitor bazedoxifene (BZA), which blocks formation of the IL6 receptor-gp130 complex. Glioma stem-like cells (GSCs) are more tumorigenic and more resistant to therapy. STAT3 is a key driver of the expression of stem cell transcription factors, making it a therapeutically important target in GBM. We show that treating GSCs with BZA decreases their self-renewal capacity and the expression of GSC markers in vitro. Additionally, BZA crosses the blood-brain barrier and confers a survival advantage in an orthotopic syngeneic mouse model of GBM. Although IL6-STAT3 signaling is important for GSC survival, a therapeutic agent that inhibits this pathway without toxicity has yet to be identified. Our findings reveal a mechanism of sustained STAT3 signaling in GBM and reveal its role in GSC maintenance, and we identify BZA as a novel candidate for treating GBM.

Increased Incidence of Venous Thromboembolism with Cancer Immunotherapy.

BACKGROUND: Cancer immunotherapy is associated with several immune-related adverse events, but the relationship between immunotherapy and venous thromboembolism has not been thoroughly studied. METHODS: We conducted a retrospective cohort study of 1,686 patients who received immunotherapy for a variety of malignancies to determine the incidence of venous thromboembolism and the impact of venous thromboembolism on survival. To examine the potential role of inflammation in venous thromboembolism, we also profiled immune cells and plasma cytokines in blood samples obtained prior to initiation of immunotherapy in a sub-cohort of patients treated on clinical trials who subsequently did (N = 15), or did not (N = 10) develop venous thromboembolism. FINDINGS: Venous thromboembolism occurred while on immunotherapy in 404/1686 patients (24%) and was associated with decreased overall survival [HR=1.22 (95% CI 1.06-1.41), p<0.008]. Patients that developed venous thromboembolism had significantly higher pretreatment levels of myeloid-derived suppressor cells (5.382 ± 0.873 vs. 3.341 ± 0.3402, mean ± SEM; p=0.0045), interleukin 8 (221.2 ± 37.53 vs. 111.6 ± 25.36, mean ± SEM; p=0.016), and soluble vascular cell adhesion protein 1 (1210 ± 120.6 vs. 895.5 ± 53.34, mean ± SEM; p=0.0385). CONCLUSIONS: These findings demonstrate that venous thromboembolism is an underappreciated and important immune-related adverse event associated with cancer immunotherapy, and may implicate an interleukin 8 and myeloid-derived suppressor cell-driven pathway in pathogenesis.

Preclinical Modeling of Surgery and Steroid Therapy for Glioblastoma Reveals Changes in Immunophenotype that are Associated with Tumor Growth and Outcome.

PURPOSE: Glioblastoma (GBM) immunotherapy clinical trials are generally initiated after standard-of-care treatment-including surgical resection, perioperative high-dose steroid therapy, chemotherapy, and radiation treatment-has either begun or failed. However, the impact of these interventions on the antitumoral immune response is not well studied. While discoveries regarding the impact of chemotherapy and radiation on immune response have been made and translated into clinical trial design, the impact of surgical resection and steroids on the antitumor immune response has yet to be determined. EXPERIMENTAL DESIGN: We developed a murine model integrating tumor resection and steroid treatment and used flow cytometry to analyze systemic and local immune changes. These mouse model findings were validated in a cohort of 95 patients with primary GBM. RESULTS: Using our murine resection model, we observed a systemic reduction in lymphocytes corresponding to increased tumor volume and decreased circulating lymphocytes that was masked by dexamethasone treatment. The reduction in circulating T cells was due to reduced CCR7 expression, resulting in T-cell sequestration in lymphoid organs and the bone marrow. We confirmed these findings in a cohort of patients with primary GBM and found that prior to steroid treatment, circulating lymphocytes inversely correlated with tumor volume. Finally, we demonstrated that peripheral lymphocyte content varies with progression-free survival and overall survival, independent of tumor volume, steroid use, or molecular profiles. CONCLUSIONS: These data reveal that prior to intervention, increased tumor volume corresponds with reduced systemic immune function and that peripheral lymphocyte counts are prognostic when steroid treatment is taken into account.

High-Dimensional Analysis of Circulating and Tissue-Derived Myeloid-Derived Suppressor Cells from Patients with Glioblastoma.

We will first describe analysis of MDSC subsets from patient tumors with multicolor flow cytometry. The key components of this methodology are to obtain viable single cell suspensions and eliminate red blood cell contamination.

Hepatobiliary malignancies have distinct peripheral myeloid-derived suppressor cell signatures and tumor myeloid cell profiles.

Myeloid-derived suppressor cells (MDSCs) are immunosuppressive cells that are increased in patients with numerous malignancies including viral-derived hepatocellular carcinoma (HCC). Here, we report an elevation of MDSCs in the peripheral blood of patients with other hepatobiliary malignancies including non-viral HCC, neuroendocrine tumors (NET), and colorectal carcinoma with liver metastases (CRLM), but not cholangiocarcinoma (CCA). The investigation of myeloid cell infiltration in HCC, NET and intrahepatic CCA tumors further established that the frequency of antigen-presenting cells was limited compared to benign lesions, suggesting that primary and metastatic hepatobiliary cancers have distinct peripheral and tumoral myeloid signatures. Bioinformatics analysis of The Cancer Genome Atlas dataset demonstrated that a high MDSC score in HCC patients is associated with poor disease outcome. Given our observation that MDSCs are increased in non-CCA malignant liver cancers, these cells may represent suitable targets for effective immunotherapy approaches.

Glioblastoma Myeloid-Derived Suppressor Cell Subsets Express Differential Macrophage Migration Inhibitory Factor Receptor Profiles That Can Be Targeted to Reduce Immune Suppression.

The application of tumor immunotherapy to glioblastoma (GBM) is limited by an unprecedented degree of immune suppression due to factors that include high numbers of immune suppressive myeloid cells, the blood brain barrier, and T cell sequestration to the bone marrow. We previously identified an increase in immune suppressive myeloid-derived suppressor cells (MDSCs) in GBM patients, which correlated with poor prognosis and was dependent on macrophage migration inhibitory factor (MIF). Here we examine the MIF signaling axis in detail in murine MDSC models, GBM-educated MDSCs and human GBM. We found that the monocytic subset of MDSCs (M-MDSCs) expressed high levels of the MIF cognate receptor CD74 and was localized in the tumor microenvironment. In contrast, granulocytic MDSCs (G-MDSCs) expressed high levels of the MIF non-cognate receptor CXCR2 and showed minimal accumulation in the tumor microenvironment. Furthermore, targeting M-MDSCs with Ibudilast, a brain penetrant MIF-CD74 interaction inhibitor, reduced MDSC function and enhanced CD8 T cell activity in the tumor microenvironment. These findings demonstrate the MDSC subsets differentially express MIF receptors and may be leveraged for specific MDSC targeting.

JAM-A functions as a female microglial tumor suppressor in glioblastoma.

BACKGROUND: Glioblastoma (GBM) is the most aggressive primary brain tumor and has a dismal prognosis. Previously, we identified that junctional adhesion molecule A (JAM-A), a cell adhesion molecule, is highly elevated in human GBM cancer stem cells and predicts poor patient prognosis. While JAM-A is also highly expressed in other cells in the tumor microenvironment, specifically microglia and macrophages, how JAM-A expression in these cells affects tumor growth has yet to be determined. The goal of this study was to understand the role of microenvironmental JAM-A in mediating GBM growth. METHODS: Male and female wild-type (WT) and JAM-A-deficient mice were transplanted intracranially with the syngeneic glioma cell lines GL261 and SB28 and were assessed for differences in survival and microglial activation in tumors and in vitro. RNA-sequencing was performed to identify differentially regulated genes among all genotypes, and differences were validated in vitro and in vivo. RESULTS: We found that JAM-A-deficient female mice succumbed to GBM more quickly compared with WT females and JAM-A-deficient and male WT mice. Analysis of microglia in the tumors revealed that female JAM-A-deficient microglia were more activated, and RNA-sequencing identified elevated expression of Fizz1 and Ifi202b specifically in JAM-A-deficient female microglia. CONCLUSIONS: Our findings suggest that JAM-A functions to suppress pathogenic microglial activation in the female tumor microenvironment, highlighting an emerging role for sex differences in the GBM microenvironment and suggesting that sex differences extend beyond previously reported tumor cell-intrinsic differences.

Myeloid-Derived Suppressor Cell Subsets Drive Glioblastoma Growth in a Sex-Specific Manner.

Myeloid-derived suppressor cells (MDSC) that block antitumor immunity are elevated in glioblastoma (GBM) patient blood and tumors. However, the distinct contributions of monocytic (mMDSC) versus granulocytic (gMDSC) subsets have yet to be determined. In mouse models of GBM, we observed that mMDSCs were enriched in the male tumors, whereas gMDSCs were elevated in the blood of females. Depletion of gMDSCs extended survival only in female mice. Using gene-expression signatures coupled with network medicine analysis, we demonstrated in preclinical models that mMDSCs could be targeted with antiproliferative agents in males, whereas gMDSC function could be inhibited by IL1ß blockade in females. Analysis of patient data confirmed that proliferating mMDSCs were predominant in male tumors and that a high gMDSC/IL1ß gene signature correlated with poor prognosis in female patients. These findings demonstrate that MDSC subsets differentially drive immune suppression in a sex-specific manner and can be leveraged for therapeutic intervention in GBM. SIGNIFICANCE: Sexual dimorphism at the level of MDSC subset prevalence, localization, and gene-expression profile constitutes a therapeutic opportunity. Our results indicate that chemotherapy can be used to target mMDSCs in males, whereas IL1 pathway inhibitors can provide benefit to females via inhibition of gMDSCs.See related commentary by Gabrilovich et al., p. 1100.This article is highlighted in the In This Issue feature, p. 1079.

Identifying conserved molecular targets required for cell migration of glioblastoma cancer stem cells.

Glioblastoma (GBM) is the most prevalent primary malignant brain tumor and is associated with extensive tumor cell infiltration into the adjacent brain parenchyma. However, there are limited targeted therapies that address this disease hallmark. While the invasive capacity of self-renewing cancer stem cells (CSCs) and their non-CSC progeny has been investigated, the mode(s) of migration used by CSCs during invasion is currently unknown. Here we used time-lapse microscopy to evaluate the migratory behavior of CSCs, with a focus on identifying key regulators of migration. A head-to-head migration assay demonstrated that CSCs are more invasive than non-CSCs. Time-lapse live cell imaging further revealed that GBM patient-derived CSC models either migrate in a collective manner or in a single cell fashion. To uncover conserved molecular regulators responsible for collective cell invasion, we utilized the genetically tractable Drosophila border cell collective migration model. Candidates for functional studies were generated using results from a targeted Drosophila genetic screen followed by gene expression analysis of the human homologs in GBM tumors and associated GBM patient prognosis. This strategy identified the highly conserved small GTPase, Rap1a, as a potential regulator of cell invasion. Alteration of Rap1a activity impaired the forward progress of Drosophila border cells during development. Rap1a expression was elevated in GBM and associated with higher tumor grade. Functionally, the levels of activated Rap1a impacted CSC migration speed out of spheres onto extracellular matrix. The data presented here demonstrate that CSCs are more invasive than non-CSCs, are capable of both collective and single cell migration, and express conserved genes that are required for migration and invasion. Using this integrated approach, we identified a new role for Rap1a in the migration of GBM CSCs.

Metronomic capecitabine as an immune modulator in glioblastoma patients reduces myeloid-derived suppressor cells.

BACKGROUNDMyeloid-derived suppressor cells (MDSCs) are elevated in the circulation of patients with glioblastoma (GBM), present in tumor tissue, and associated with poor prognosis. While low-dose chemotherapy reduces MDSCs in preclinical models, the use of this strategy to reduce MDSCs in GBM patients has yet to be evaluated.METHODSA phase 0/I dose-escalation clinical trial was conducted in patients with recurrent GBM treated 5-7 days before surgery with low-dose chemotherapy via capecitabine, followed by concomitant low-dose capecitabine and bevacizumab. Clinical outcomes, including progression-free and overall survival, were measured, along with safety and toxicity profiles. Over the treatment time course, circulating MDSC levels were measured by multiparameter flow cytometry, and tumor tissue immune profiles were assessed via time-of-flight mass cytometry.RESULTSEleven patients total were enrolled across escalating dose cohorts of 150, 300, and 450 mg bid. No serious adverse events related to the drug combination were observed. Compared with pretreatment baseline, circulating MDSCs were found to be higher after surgery in the 150-mg treatment arm and lower in the 300-mg and 450-mg treatment arms. Increased cytotoxic immune infiltration was observed after low-dose capecitabine compared with untreated GBM patients in the 300-mg and 450-mg treatment arms.CONCLUSIONSLow-dose, metronomic capecitabine in combination with bevacizumab was well tolerated in GBM patients and was associated with a reduction in circulating MDSC levels and an increase in cytotoxic immune infiltration into the tumor microenvironment.TRIAL REGISTRATIONClinicalTrials.gov NCT02669173.FUNDINGThis research was funded by the Cleveland Clinic, Case Comprehensive Cancer Center, the Musella Foundation, B*CURED, the NIH, the National Cancer Institute, the Sontag Foundation, Blast GBM, the James B. Pendleton Charitable Trust, and the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation. Capecitabine was provided in kind by Mylan Pharmaceuticals.