Tryptophan Metabolism Contributes to Radiation

Published OnlineFirst April 24, 2018; DOI: 10.1158/1078-0432.CCR-18-0041

Translational Cancer Mechanisms and Therapy Clinical Cancer Research Tryptophan Metabolism Contributes to Radiation- Induced Immune Checkpoint Reactivation in Glioblastoma Pravin Kesarwani1, Antony Prabhu1, Shiva Kant1, Praveen Kumar2, Stewart F. Graham2, Katie L. Buelow1, George D. Wilson1, C. Ryan Miller3, and Prakash Chinnaiyan1,4

Abstract

Purpose: Immune checkpoint inhibitors designed to revert aberrant tryptophan metabolism as a metabolic node spe- tumor-induced immunosuppression have emerged as potent cific to the mesenchymal and classical subtypes of glioblas- anticancer therapies. Tryptophan metabolism represents an toma. GDC-0919 demonstrated potent inhibition of this immune checkpoint, and targeting this pathway's rate-limiting node and effectively crossed the blood–brain barrier. enzyme IDO1 is actively being investigated clinically. Here, we Although GDC-0919 as a single agent did not demonstrate studied the intermediary metabolism of tryptophan metabo- antitumor activity, it had a strong potential for enhancing lism in glioblastoma and evaluated the activity of the IDO1 RT response in glioblastoma, which was further augmented inhibitor GDC-0919, both alone and in combination with with a hypofractionated regimen. RT response in glioblas- radiation (RT). toma involves immune stimulation, reflected by increases in Experimental Design: LC/GC-MS and expression profiling activated and cytotoxic T cells, which was balanced by was performed for metabolomic and genomic analyses of immune checkpoint reactivation, reflected by an increase patient-derived glioma. Immunocompetent mice were in IDO1 expression and regulatory T cells (Treg). GDC-0919 injected orthotopically with genetically engineered murine mitigated RT-induced Tregs and enhanced T-cell activation. glioma cells and treated with GDC-0919 alone or combined Conclusions: Tryptophan metabolism represents a meta- with RT. Flow cytometry was performed on isolated tumors to bolic node in glioblastoma, and combining RT with IDO1 determine immune consequences of individual treatments. inhibition enhances therapeutic response by mitigating RT- Results: Integrated cross-platform analyses coupling glob- induced immunosuppression. Clin Cancer Res; 24(15); 3632–43. al metabolomic and gene expression profiling identified 2018 AACR.

Introduction failed. Therefore, developing innovative treatment strategies to improve outcome in glioblastoma is of clear importance (2). Glioblastoma is the most common adult primary brain tumor Recent clinical advancements using immune checkpoint inhi- (1). Despite continued advances in surgery and combined che- bitors designed to target tumor-mediated immune tolerance have moradiotherapy, clinical outcomes remain poor. An overwhelm- revolutionized our approach to cancer therapy and offers strong ing majority of tumors recur within a year of definitive therapy, promise in glioblastoma. Immune tolerance is important in and a very small percentage of patients survive beyond 3 years of normal physiology to prevent overreactivity of stimulated diagnosis. Unfortunately, concerted efforts to improve clinical immune responses to external pathogens and other stimuli. outcomes, including integration of molecularly targeted agents, Dysfunction of these immune response "brakes" may lead to a angiogenesis inhibitors, and vaccines to standard therapy, have all variety of disorders, including autoimmune diseases, type I dia- betes, inflammatory bowel disease, asthma, and allergies (3). Tumors have coopted multiple mechanisms to activate these 1Department of Radiation Oncology, Beaumont Health, Royal Oak, Michigan. "brakes" by inducing immunosuppressive signaling pathways, 2Metabolomics and Obstetrics/Gynecology, Beaumont Research Institute, creating a tolerogenic microenvironment allowing evasion of the Beaumont Health, Royal Oak, Michigan. 3Department of Pathology & Laboratory host immune response despite the presence of recognizable Medicine, Neurology, & Pharmacology, Lineberger Comprehensive Cancer Cen- antigens. Cytotoxic T lymphocyte–associated protein 4 (CTLA- ter and Neurosciences Center, University of North Carolina School of Medicine, 4) and programmed death-1 (PD-1), which negatively regulate 4 Chapel Hill, North Carolina. Oakland University William Beaumont School of T-cell activation, represent two specific immune checkpoints that Medicine, Royal Oak, Michigan. have received recent attention, with inhibitors targeting these Note: Supplementary data for this article are available at Clinical Cancer immune pathways demonstrating unprecedented clinical activity Research Online (http://clincancerres.aacrjournals.org/). in a variety of solid tumors (4, 5). Corresponding Author: Prakash Chinnaiyan, Beaumont Health System, 3811 The central nervous system (CNS) has been previously consid- West Thirteen Mile Road, Royal Oak, MI 48073. Phone: 248-551-4918; Fax: 248- ered an immune privileged site; however, growing data support a 551-1002; E-mail: [email protected] dynamic interaction between the CNS and the systemic immune doi: 10.1158/1078-0432.CCR-18-0041 system. Although the CNS lacks a traditional lymphatic system, 2018 American Association for Cancer Research. recent findings identified a rich lymphatic network in the dura

3632 Clin Cancer Res; 24(15) August 1, 2018

Radiation and Tryptophan Metabolism in Glioblastoma

AHRs are also responsible for providing a tolerogenic pheno- Translational Relevance type in dendritic cells, resulting in increased production of Through integrative, cross-platform analyses coupling glob- Tregs and reduced type 1 Th1 cells (17, 18). Activation of AHRs al metabolomic profiling with gene expression arrays in over also results in the induction of various cytokines, such as TGFb, þ 100 patient-derived tumors, we identified aberrant tryptophan IL6, and IL1b, that are responsible for converting CD4 T cells metabolism as an important metabolic node and immune to inducible Tregs (iTreg) and maintaining the suppressive checkpoint in glioblastoma. We discovered that intermediar- ability of MDSCs (14, 19, 20). ies associated with tryptophan metabolism demonstrated Interestingly, a variety of tumors, including glioblastoma, have >90% accuracy in discriminating between low- and high-grade evolved mechanisms to coopt this potent mode of immune glioma, offering novel diagnostic implications. We then tolerance to evade the host immune system (21). This relationship extended these findings preclinically using a novel, immuno- between cancer and elevated tryptophan catabolism was first competent, genetically engineered mouse model of adult recognized in the 1950s with the identification of its metabolic astrocytoma with a potent, clinically relevant inhibitor of intermediaries in the urine of cancer patients (22, 23). The IDO1 tryptophan metabolism/IDO1. Through these studies, we pathway was later proposed to play a more direct role in tumor uncovered a potent synergy when combining an IDO1 inhib- immune evasion, demonstrating a more robust T-cell response itor with hypofractionated radiation and a novel mechanism and delayed growth in vivo following pathway inhibition (24). linking tryptophan metabolism with radiation-induced Several recent studies have demonstrated that tryptophan catab- immunosuppression involving immune checkpoint reactiva- olism, through IDO1 and/or TDO upregulation, is particularly tion. These encouraging findings support future clinical efforts relevant in glioblastoma. TDO-mediated pathway activation in a designed to combine IDO1 inhibition with hypofractionated panel of glioma cell lines resulted in inhibition of T-cell prolif- radiation in glioblastoma, offering the promise of harnessing a eration, and modulating this pathway influenced tumor growth patient's immune system to attack these otherwise recalcitrant (14). Wainwright and colleagues demonstrated significant IDO1 tumors. expression in glioblastoma that promoted an immunosuppressive environment through recruitment of Tregs (15) and went on to show that the therapeutic efficacy of IDO1 inhibition can be significantly enhanced when combined with other immune mater that is able to absorb and transport CSF into deep cervical checkpoint inhibitors (25). lymph nodes where CNS antigens have been reported (6, 7). The majority of studies evaluating the immune consequences Furthermore, glioblastomas produce an array of chemokines, of aberrant tryptophan metabolism have largely focused on such as IL8, CCL2, CXCL12, that are able to recruit immunosup- expression of its rate-limiting enzymes rather than the individ- pressive tumor-associated macrophages (TAM) and myeloid- ual metabolites involved in immunosuppression. In this study, fi derived suppressive cells (MDSC), furthering the tolerogenic using global metabolomic pro ling as a framework, we explore tumor microenvironment (8, 9). This paradigm shift has contrib- this immunomodulatory pathway from a metabolic perspec- uted toward a growing interest in the evaluation of immunother- tive, further validating its relevance in glioblastoma, and eval- apeutic approaches in glioblastoma, although early studies have uate the antitumor potential of the potent IDO1 inhibitor demonstrated limited clinical activity when used alone or in GDC-0919 in a novel, immunocompetent preclinical model combination with bevacizumab (10). of adult glioblastoma. Emerging studies have identified multifaceted strategies by which alterations in tumor metabolism may also contribute to Materials and Methods a potent tolerogenic immune environment, thereby representing Study approval a line of next-generation immune checkpoints (11, 12). One of All mice were housed in the AAALAC-accredited Animal facility the most clinically advanced with particular relevance to glioblas- located at William Beaumont Research Institute (Royal Oak, MI). toma is tryptophan metabolism, whose most notable physiologic All animal studies were carried out under protocols approved by role has been attributed to peripheral immune tolerance and fetal the Institutional Animal Care and Use Committee at William protection from maternal immune rejection in the placenta (13). Beaumont Research Institute. Patient-derived tumors were Tryptophan is metabolized to kynurenine by the rate-limiting obtained from the H. Lee Moffitt Cancer Center (Tampa, FL), enzymes indoleamine 2,3-dioxygenase 1 (IDO1) and tryptophan following Institutional Review Board and Human Subjects 2,3-dioxygenase (TDO; refs. 14, 15) contributing toward an approval received from the H. Lee Moffitt Cancer Center Tissue immune tolerant environment at many levels. Increased metab- Core Facility as described previously (26). olism of tryptophan results in depletion of this critical metabolite in the microenvironment, resulting in cell-cycle arrest and/or Human tumor samples and metabolomic profiling anergy in effector T cells, while simultaneously fostering matu- All tumors used in this analysis were newly diagnosed. Tumors ration and activation of regulatory T cells (Treg; ref. 16). In were fresh-frozen and their integrity and histology confirmed by a addition to this passive consequence of tryptophan metabolism, staff pathologist before aliquoting samples (26). Metabolomic the resulting kynurenine is exported from the cell into the micro- studies were conducted at Metabolon Inc using methods environment, where it has the capacity to balance regulatory and described previously (26). For box plots, raw area count was effector responses of the immune system. This includes binding normalized and rescaled to set the median equal to one, followed and activation of aryl hydrocarbon receptors (AHR), a cyto- by inputting minimum to missing values. Rescaled data were used plasmic transcription factor, resulting in reduced proliferation to calculate fold change by comparing GBM with low-grade and infiltration of effector T cells (14). Kynurenine-activated glioma (LGG).

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3633

Kesarwani et al.

Microarray and database analysis Results For details, see Supplementary Methods. Metabolomic profiling identifies aberrant tryptophan metabolism in glioblastoma Cell culture and reagents Metabolomic profiling was performed in glioma using a robust Human glioblastoma cell lines U87, T98G, (obtained from library of standards enriched with intermediates of tryptophan ATCC), and U251 (obtained from CLS Cell Lines Service metabolism, comparing patient-derived glioblastoma (n ¼ 80) to GmbH) were cultured in Minimum Essential Media (MEM). LGG (n ¼ 28). Tryptophan levels were modestly elevated in Human glioblastoma neural stem cells (GNS) MES326 and glioblastoma (1.44-fold, P 0.011; Fig. 1A). Tryptophan has PN19 were generated and cultured in DMEM-F12 media with several metabolic fates independent of the IDO1/TDO2 pathway, GlutaMAX, heparin, and B27 supplement as described previ- including serving as a mediator for the formation of serotonin and ously (27). The genetically engineered murine GBM cell line indoles (Supplementary Fig. S1A). Global metabolomic profiling TRP was cultured in MEM (28–30). Primary cell culture of identified several indoles, which may serve as both catabolic mouse splenocyte (T cells) and murine GBM tumor lines byproducts and anabolic precursors of tryptophan, elevated in GL261 (obtained from NCI DTP, DCTD tumor repository) glioblastoma (Supplementary Fig. S1B). Of the identified meta- was performed in RPMI1640. All cell culture reagents were bolites associated with tryptophan metabolism, kynurenine dem- purchased from GIBCO/Thermo Fisher ScientificandCorning onstrated the highest accumulation in glioblastoma when com- Life Sciences. Tryptophan, kynurenine, p-dimethylaminoben- pared with LGG (6.07-fold, P 0.0002; Fig. 1A). Interestingly, zaldehyde, glacial acetic acid, TCA, and IDO1 inhibitors 1-L- kynurenate, a metabolite immediately downstream of kynure- methyl tryptophan (1-L-MT) were purchased from Sigma nine, was significantly lower in glioblastoma when compared Aldrich. GDC-0919 was provided by Genentech. with LGG (0.23-fold, P 0.0004), and the resulting kynurenine/ kynurenate ratio demonstrated the highest degree of sensitivity Western blot analysis, flow cytometry, magnetic sorting, and (91.1%) and specificity (96.5%) in discriminating between glio- ELISA blastoma and LGG in comparison with using kynurenine alone For details, see Supplementary Methods. (sensitivity 60.8%; specificity 91.6%; P < 0.00001; Fig. 1A; Sup- plementary Fig. S1C). Using the TCGA database, we then sought Metabolite extraction and targeted liquid chromatography to determine whether aberrant expression of enzymes involved fi mass spectrometry with tryptophan metabolism recapitulated these metabolic nd- For details, see Supplementary Methods. ings in glioma. Both IDO1 and TDO2 were elevated in glioblastoma when compared with LGG and consistent with the observed fi Immunocompetent tumor models accumulation of kynurenate in LGG, we identi ed increased C57BL/6 (H-2b, CD45.2) mice were exposed to 4 Gy total body expression of its biosynthetic enzyme kynurenine aminotransfer- irradiation one day prior to implantation to help tumor engraft- ase I (KYAT1) in these tumors when compared with glioblastoma ment. Mice were injected with 2 105 TRP (murine GBM) tumor (Fig. 1B). cells in the brain for intracranial tumor model or 2 106 TRP tumor cells in the right flank for subcutaneous tumor model. MRI Tryptophan metabolism is linked to molecular subtypes of was performed on 6 to 8 days after tumor implant. After verifi- glioblastoma cation of tumors, mice were randomized into described treatment We next sought to determine whether aberrant tryptophan arms. Detailed procedure is provided in Supplementary Materials metabolism was unique to a specific glioblastoma molecular and Methods. subtype. Transcriptional profiling has revealed considerable intertumoral heterogeneity in glioblastoma (31). These include the T-cell suppression assay proneural and mesenchymal subtypes, which have been identi- For details, see Supplementary Methods. fied most consistently in glioblastoma. Their transcriptional profiles are mutually exclusive and can be applied to approximately Kynurenine estimation one half of all tumors. The proneural subtype is characterized by For details, see Supplementary Methods. mutations in isocitrate dehydrogenase 1 (IDH1), frequent alterations in expression of p53 and platelet-derived growth factor Statistical analysis receptor, alpha polypeptide (PDGFR-a), and a transcriptional Comparisons across two different groups were performed signature typically present in LGG. In contrast, mesenchymal on original data using two-tailed Student t test. A log-rank test glioblastoma is characterized by mesenchymal gene expression, was used for survival analyses. Comparisons between tumor including CD44 and chitinase 3-like 1 (YKL40) and attributed to volumes in the subcutaneous tumor model were performed more aggressive disease. Other subtypes include classical and using two-way ANOVA in combination with post hoc opera- neural, which are characterized by EGFR and the expression of tions to generate P values between RT and RT þ GDC-0919. neuronal markers, respectively. Of the 80 glioblastoma specimens ROC curves were constructed using metabolomics data for metabolomically profiled, 56 had paired tissue available for kynurenine and kynurenine/kynurenine acid ratios. Highest transcriptional profiling, allowing for cross-platform analysis. concentrations of kynurenine and kynurenine/kynurenine Following molecular subtyping using methods described previ- acid ratios in LGG were used as thresholds for predicting ously (31) we demonstrated that tryptophan and kynurenine glioblastoma from LGG. All statistical analyses were per- accumulation was specific to classical and mesenchymal subtypes, formed using Origin Pro 2016 software (Origin Lab whereas kynurenate accumulation, the metabolite elevated in Corporation). LGG, was only evident in the proneural subtype. The

3634 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Radiation and Tryptophan Metabolism in Glioblastoma

Figure 1. Aberrant tryptophan metabolism represents a metabolic node in glioblastoma. A, Metabolomic profiling performed on patient-derived LGG (n ¼ 28) and glioblastoma (n ¼ 80) demonstrates differential accumulation of tryptophan (TRP), kynurenine (KYN), kynurenate (KYNA), and KYN/KYNA ratio. B, Expression analysis of enzymes involved in tryptophan metabolism in LGG and glioblastoma was performed using The Cancer Genome Atlas (TCGA). C and D, Cross-platform analysis was performed coupling metabolomic profiles with gene expression arrays in glioblastoma (n ¼ 56). The relative accumulation of intermediaries of tryptophan metabolism (C) and enzymes driving tryptophan metabolism (D) were evaluated in the context of the tumors' molecular subtype, classified as classical (CL), mesenchymal (M), neural (N), or proneural (PN). E, Expression of enzymes involved in tryptophan metabolism in the context of their anatomic structure derived from the Ivy GAP database. Regions were classified as periphery (leading edge and infiltrating tumor), cellular tumor, and necrotic core (perinecrotic and pseudopalisading necrosis). , P < 0.00005; , P < 0.0005; , P < 0.005; , P < 0.05.

kynurenine/kynurenate ratio was able to discriminate between as a metabolic phenotype in glioblastoma, we extended investi- individual molecular subtypes, with highest ratios present in the gations to preclinical models. IDO1, which is typically undetect- classical and mesenchymal subtypes when compared with neural able in cells at baseline, is rapidly induced by proinflammatory and proneural tumors (Fig. 1C). The expression levels of enzymes stimuli. Because of the multiple IFN-stimulated response ele- involved in tryptophan metabolism corroborated these metabolic ments in the IDO1 promoter region, in vitro treatment of cultured findings, with higher levels of IDO1 and TDO2 in classical and cells with the T effector cells and proinflammatory cytokine IFNg mesenchymal subtypes, whereas neural and proneural subtypes leads to robust expression of this enzyme (33). Accordingly, demonstrated elevated expression of KYAT1 (Fig. 1D). Using a baseline expression of IDO1 was low/undetectable in established unique data repository generated by the Ivy GAP, which used laser human glioblastoma cell lines and robustly expressed following capture microdissection to isolate RNA from different structural treatment with IFNg (Fig. 2A). Similar findings were observed regions of individual tumors that was subsequently molecularly when extended to novel, molecular subtype-specific GNSs, with profiled using RNA-seq, we recently uncovered a relationship more robust expression in the mesenchymal GNSs when com- between subtype heterogeneity in glioblastoma and its unique pared with proneural, which is consistent with our findings from tumor microenvironment. In this study, we demonstrated that patient-derived tumors. To determine whether IFNg-induced cells obtained from the infiltrating, leading edge of a tumor nearly expression of IDO1 corresponded to increased tryptophan metab- exclusively harbored a proneural signature, whereas the mesen- olism, kynurenine levels were evaluated. Consistent with chymal subtype was observed in perinecrotic regions within an increased IDO1 expression, IFNg induced accumulation of kynur- individual tumor (32). On the basis of these findings, we sought enine in the media of glioblastoma cell lines grown in culture, to determine whether aberrant tryptophan metabolism was confirming pathway activation (Fig. 2B). TDO2, on the other unique in specific regions within an individual tumor. Consistent hand, was inconsistently expressed in cell lines and interestingly with our previous findings, increased IDO1/TDO2 expression was appeared to be downregulated by IFNg in several of the lines present in the necrotic and central core of tumor when compared tested (Supplementary Fig. S2A). with the periphery, which corresponds to the mesenchymal/ To begin to explore the immune consequence of aberrant classical and proneural molecular subtypes, respectively (Fig. 1E). tryptophan metabolism in glioblastoma, we extended investigations to an adult astrocytic, genetically engineered mouse (GEM) The IDO1 pathway is activated in glioblastoma cell lines cell line to allow for in vivo studies using an immunocompetent As cross-platform analysis coupling metabolomics with tran- model. Specifically, primary astrocyte lines were generated from a scriptional profiling identified aberrant tryptophan metabolism series of conditional GEM in which core glioblastoma pathways

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3635

Kesarwani et al.

Figure 2. The IDO1 pathway is activated in glioblastoma cell lines. A, Human glioblastoma cell lines (T98G, U87, U251, MES326, and PN19) were cultured in the presence or absence of 50 ng/mL of human IFNg for 3 days and evaluated by Western blot analysis. B, Human cell lines were cultured in 250 mmol/L tryptophan and presence or absence of 50 ng/mL of IFNg. Media supernatant was collected after 3 days for kynurenine estimation using Ehrlich method. Data, mean SD. C, Orthotopic TRP tumors grown in C57BL/6 mice were analyzed for kynurenine using LC/MS and compared with normal murine brain. D, IDO1 expression was analyzed in TRP tumors isolated from C57BL/6 mice using Western blot analysis. Results are representative of at least three independent experiments. , P < 0.05.

were genetically targeted (30). Briefly, after Cre-mediated recom- direct small-molecule IDO1 inhibitor GDC-0919 (navoximod), bination in vitro, lines express (i) a truncation mutant of SV40 which is actively being investigated in clinical trials (34). As an large T antigen (T) from the human Gfap promoter that inactivates initial investigation, we evaluated levels of kynurenine following all 3 Rb family proteins; (ii) a constitutively active KrasG12D graded concentrations of GDC-0919 in our human glioblastoma mutant (R); and/or (iii) a homozygous Pten deletion (P). As an cell lines in vitro. GDC-0919 demonstrated potent inhibition of initial investigation, we performed metabolomic profiling of TRP kynurenine production, with median effective dose (ED50) con- cells grown orthotopically, which harbor all 3 of these mutations centrations ranging from 7 nmol/L to 1.77 mmol/L and near and display the most aggressive phenotype (30), which confirmed complete inhibition at 5 mmol/L (Fig. 3A). As we were unable pathway activation with high levels of kynurenine when com- to induce IDO1 expression in vitro in our mouse glioblastoma pared with normal brain (Fig. 2C). Although we were unable to models, we implemented an ex vivo approach to evaluate the induce IDO1 expression and kynurenine production in the TRP potential of GDC-0919 to inhibit kynurenine production in the line and the mouse glioblastoma cell line GL261 in vitro using a TRP line. In these studies, TRP tumors grown subcutaneously in variety of methods, including both human and mouse IFNg C57BL/6 mice were excised and disaggregated in cell culture (Supplementary Fig. S2B), expression was clearly evident in media using DNase I and collagenase IV. Dissociated tumors tumors grown in vivo (Fig. 2D). were then passed through a 70-mm cell strainer to obtain a uniform cell suspension. The cell suspension was treated with GDC-0919 demonstrates potent inhibition of tryptophan IFNg, and supernatant was collected after 24 hours for kynurenine metabolism in glioblastoma and achieves biologically relevant estimation. Using this approach, we demonstrated that TRP cells concentrations in the brain endogenously produced kynurenine ex vivo, which was signifi- As we established aberrant tryptophan metabolism in patient- cantly increased with IFNg. Kynurenine production was derived glioblastoma, which was recapitulated in preclinical completely inhibited in TRP ex vivo by GDC-0919 (5 mmol/L). models, we next sought to determine the potential for molecularly Interestingly, in a parallel study where mice were treated with targeting this immunomodulatory pathway using the potent and GDC-0919 (200 mg/kg, twice a day for 3 days) in vivo prior to

3636 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Radiation and Tryptophan Metabolism in Glioblastoma

Figure 3. GDC-0919 demonstrates potent inhibition of tryptophan metabolism in glioblastoma and achieves biologically relevant concentrations in the brain. A, Human GBM cell lines (T98G, U87, U251, and MES326) were cultured with human IFNg (50 ng/mL) and of tryptophan (250 mmol/L) and graded concentrations of the IDO1 inhibitor GDC-0919. The supernatant was collected after 3 days and evaluated for kynurenine. Data were plotted and used to calculate ED50 values for individual cell lines. B, Murine TRP tumors (in vitro and ex vivo tumors) isolated from C57BL/6 mice and C57BL/6 mice treated with GDC-0919 (200 mg/kg twice a day for 3 days given orally) were cultured in the presence and absence of mouse IFNg (100 ng/mL) and tryptophan (250 mmol/L). The supernatant was collected after 3 days and analyzed for kynurenine. C and D, C57BL/6 mice with subcutaneous or intracranial TRP tumors were treated with GDC-0919 (200 mg/kg twice a day for 3 days given orally). Tissue and plasma were isolated after 3 days of treatment and analyzed for GDC-0919 (C) and tryptophan and kynurenine (in intracranial tumors) using LC/MS (D). Results obtained in mg/kg were converted to mmol/L by assuming mg/mL ratio to be 1. Data, mean SD. , P < 0.05. A minimum of 4 tissue samples were used for each experiment.

tumor excision, TRP cells demonstrated diminished production both intracranial tumors and plasma from mice treated with the of both baseline and IFNg-induced kynurenine (Fig. 3B). above-described schedule, confirming drug activity with increased Next, we performed pharmacokinetic/pharmacodynamic stud- levels of tryptophan and decreased levels of kynurenine following ies to determine the potential of GDC-0919 to cross the blood– GDC-0919 treatment (Fig. 3D). brain barrier (BBB) and achieve biologically relevant concentrations in the brain. Using LC-MS, we quantified GDC-0919 levels GDC-0919 enhances radiation response in glioblastoma in normal brain (with no tumor implanted), intracranial tumors, Next, we determined the antitumor potential of IDO1 inhibi- subcutaneous tumors, and plasma of mice following treatment tion using GDC-0919 in glioblastoma. As an initial investigation, with GDC-0919 (200 mg/kg twice a day) for 3 days, with tissue we evaluated for growth delay in TRP cells grown in a subcuta- being extracted 2 hours after the last dose. Importantly, GDC- neous xenograft model. As a single agent, GDC-0919 did not 0919 achieved biologically relevant concentrations in intracranial demonstrate antitumor activity (Supplementary Fig. S3A). These tumors (15 mmol/L), which was comparable with levels findings were recapitulated using a first-generation IDO1 inhib- achieved in subcutaneous tumors. Furthermore, GDC-0919 itor 1-L-MT (Supplementary Fig. S3B). As radiotherapy (RT) is a achieved biologically relevant concentrations in the normal brain standard treatment in glioblastoma and emerging data have (5 mmol/L), which is particularly important in glioblastoma, identified a potential for synergy when combining RT with based on the infiltrative nature of this malignancy (Fig. 3C). In immune checkpoint blockade, we went on to evaluate the poten- addition, tryptophan and kynurenine levels were quantified in tial of GDC-0919 to enhance RT response. Continuing with the

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3637

Kesarwani et al.

Figure 4. GDC-0919 enhances radiation response in glioblastoma. TRP tumors were grown subcutaneously (A) or intracranially (IC; B and C) in immunocompetent C57BL/6 mice. MRI was obtained to conﬁrm the establishment of intracranial tumors prior to the initiation of therapy. Following establishment of tumors, mice were randomized to treatment as described in schema. GDC-0919 was given orally at a concentration of 200 mg/kg twice a day (6 days a week) for 2 to 3 weeks. Each treatment arm consisted of a minimum of 6 mice. , P < 0.005.

subcutaneous TRP model, GDC-0919 alone again demonstrated IDO1 inhibition demonstrated potent antitumor activity when no antitumor activity, however, when combined with RT (6 Gy combined with fractionated RT, with approximately one third of 1), demonstrated a significant growth delay when compared with mice demonstrating complete regression of tumors confirmed by RT alone (P < 0.005; Fig. 4A). Next, to allow for translating these MRI (Supplementary Fig. S4C) and increased survival when findings clinically, we extended studies to an orthotopic, intra- compared with fractionated RT alone (P < 0.005). cranial TRP tumor model. MRI was performed using our small- animal imaging platform prior to initiating therapy to confirm Tryptophan metabolism contributes to radiation-induced both presence of intracranial tumors and equal volumes between immune checkpoint reactivation in glioblastoma treatment arms (Supplementary Fig. S4A and S4B). Similar to We went on to evaluate immunologic factors underlying the results in the subcutaneous model, IDO1 inhibition using GDC- interface between tryptophan metabolism and radiation 0919 alone did not demonstrate antitumor activity; however, a response. As an initial investigation, correlative studies were trend toward improved survival when combined with RT (6 Gy performed using the experimental design detailed in Fig. 4B, 1; P ¼ 0.06; Fig. 4B) was evident. As fractionation has been consisting of mice with intracranial tumors treated with vehicle described to further potentiate the synergy between RT and control, GDC-0919, single-fraction RT, or the combination. Mice immune checkpoint agents, we extended studies using a hypo- were sacrificed 5 days following initiation of therapy, and flow fractionated regimen (6 Gy 3). As demonstrated in Fig. 4C, cytometry was performed on macrodissected tumors to evaluate

3638 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Radiation and Tryptophan Metabolism in Glioblastoma

Figure 5. Tryptophan metabolism contributes to radiation-induced immunosuppression in glioblastoma. A, Intracranial tumors were macrodissected from mice 5 days following the GDC-0919 and/or RT treatment schedule described in Fig. 4B and analyzed for immune correlates by flow cytometry, including presence of Tregs (CD4þFoxP3þCD25þ), MDSCs (CD45þCD11bþGr1þ), and tumor-associated macrophages (CD45þCD11bþF4/80þ). Scatter plot represents data from 5 to 6 tumors from each group. B, To analyze the suppressive ability of Tregs, splenocytes from C57BL/6 mice were used for isolating CD8þ T cells and Tregs (CD4þCD25þ) using magnetic sorting. CFSE-labeled CD8þ T cells were activated using plate-bound anti-CD3/CD28 antibody for 3 days in the presence or absence of Tregs (CD8:Tregs ¼ 1:3) and kynurenine (50 mmol/L) and represented as flow histogram plots as CFSE dilution (proliferation) or no proliferation (suppression). The bar graph represents cumulative data from three experiments. C, Intracranial tumors were macrodissected from mice 7 days following the GDC-0919 and/or fractionated RT treatment schedule described in Fig. 4C and analyzed for immune correlates by flow cytometry, including presence of Tregs (CD4þFoxP3þCD25þ). Scatter plot represents data as mean SEM from 4 tumors in each group. , P < 0.05.

for a panel of immunologic markers. Similar to above efficacy was also observed. When combined with RT, GDC-0919 appeared studies, GDC-0919 alone did not appear to modulate immune to mitigate this immunosuppressive response, normalizing Tregs, response in orthotopic TRP tumors, demonstrating no changes MDSCs, and TAMs back to baseline levels (Fig. 5A). Next, we þ þ þ in percentage of Tregs (CD4 FoxP3 CD25 ), MDSCs extended this line of investigation to studies using a fractionated þ þ þ þ (CD11b Gr1 ), or TAMs (CD11b F4/80 ; Fig. 5A). As GDC- RT regimen, a combination that led to a significant survival 0919 alone did not appear to influence the number of Tregs advantage in our model (Fig. 4C). Similar to a single fraction quantitatively in our model, we extended our studies to determine schedule, fractionated RT also led to an increase in Tregs, which whether IDO1/kynurenine upregulation could contribute toward was attenuated by GDC-0919 (Fig. 5C). In addition to immuno- þ Treg function using a CD8 T-cell proliferation/suppression suppression, we sought to determine whether GDC-0919, RT, or þ þ þ assay. Briefly, na€ve CD8 T cells and Tregs (CD4 CD25 T cells) the combination influenced immune activation. No changes in þ þ were isolated from splenocytes, and CD8 T cells were then the percentage of T effector cells (CD8 T cells) or helper T cells þ stained with cell proliferation dye (CFSE). These cells were then (CD4 T cells) were observed in either treatment arm (Supple- activated (using anti-CD3 and anti-CD28 antibodies) and cocul- mentary Fig. S5A and S5B). As radiation induces formation of tured in the presence or absence of Tregs, kynurenine, or the various neoantigens, thereby enhancing activation of cytotoxic T combination, for 3 days. Interestingly, neither kynurenine nor cells (35), we extended our studies to determine whether changes þ Tregs alone appeared to suppress the proliferation of CD8 T cells. in activated or cytotoxic T cells could be observed with the RT and However, kynurenine appeared to impart suppressiveness to GDC-0919 combination. Interestingly, fractionated RT alone led þ þ Tregs, as the combination led to diminished proliferation of to robust increases in both activated T cells (CD8 CD69 ) and þ þ þ CD8 T cells (Fig. 5B). cytotoxic T cells (CD8 GzmB ) in the tumor (Fig. 6A). This RT- Interestingly, we found that RT alone appeared to contribute induced immune activation was also reflected by increased IFNg toward an immunosuppressive environment in glioblastoma, levels in the plasma of mice receiving fractionated RT in compar- including over a 50% increase in Tregs (from 14.7% to 23.6% ison with controls (Fig. 6B). When combined with RT, GDC-0919 þ þ of CD45 CD4 cells). A trend in increase of MDSCs and TAMs led to further increases in both activated and cytotoxic T cells,

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3639

Kesarwani et al.

Figure 6. Radiation induces antitumor immune reactivation. A, Intracranial TRP tumors described in Fig. 4C were macrodissected from mice 7 days following the GDC-0919 and/or fractionated RT treatment schedule and analyzed for activated (CD8þCD69þ) and cytotoxic (GzmBþCD8þ) T cells. Scatter plot represents data as mean SEM from 4 tumors in each group. B, Plasma from mice harboring intracranial TRP tumors described in Fig. 4C were analyzed for mouse IFNg using ELISA. Scatter plot represents data from 4 mice from each group. C, IDO1 expression was analyzed in orthotopic TRP tumors isolated from C57BL/6 control mice and mice treated with f-RT and used to perform Western blot analysis for IDO1. Tubulin was used as loading control. D, Human PBMCs (HLA-A2þ) were cocultured þ with U251 (HLA-A2 ) and 250 mmol/L of tryptophan for 3 days. Cell culture supernatant was analyzed for kynurenine and cells were used for Western blotting for IDO1 (E). Results are representative of at least three independent experiments. , P < 0.005; , P < 0.05.

which supports the observed enhanced antitumor response by peripheral blood mononuclear cells (PBMC). As expected, neither this combination (Fig. 6A). We went on to demonstrate increased control nor irradiated U251 cells demonstrated IDO1 pathway expression of PD-1 in resected tumors following RT, suggesting an activation (Fig. 6D). Interestingly, coculturing U251 cells with increase in tumor-specific tumor infiltrating lymphocytes (Sup- na€ve PBMCs appeared to stimulate T-cell activation, leading to plementary Fig. S5C; ref. 36). IDO1 pathway activation (Fig. 6E) determined by increased We went on to explore potential mechanism(s) underlying the kynurenine production. Pathway activation was significantly potentiation of RT response with IDO1 inhibition. As both the potentiated when PBMCs were cocultured with irradiated U251 antitumor and immune response following IDO1 inhibition cells, providing further support linking RT response with trypto- appeared RT-specific in our glioblastoma model, as an initial phan metabolism in glioblastoma. investigation, we sought to determine whether RT could reactivate or further activate this immune checkpoint. As demonstrated in Fig. 6C, Western blot analysis performed on intracranial TRP Discussion tumors following fractionated RT demonstrated increased IDO1 There have been several recent reports implicating aberrant expression when compared with controls, confirming RT-induced tryptophan metabolism with glioblastoma immune tolerance pathway activation. As we have shown that RT stimulates both T- (14, 23, 37). A majority of these studies have indirectly evaluated cell activation and cytotoxicity, we next sought to link this this pathway through expression of its rate-limiting enzymes observed T-cell activation with increased IDO1 expression by IDO1 and TDO. Results from this line of investigation were determining whether activated T cells could in turn induce tryp- somewhat inconsistent, reporting IDO1 expression in 8% to tophan metabolism, thereby contributing toward RT-induced 61.5% of glioblastoma using IHC staining in 60 and 39 tumors, immunosuppression. To test this possibility, we cocultured irra- respectively (38, 39). Furthermore, the determination of increased þ diated and control U251 cells with HLA-A2 na€ve human IDO1 expression in glioblastoma is also mixed, ranging from

3640 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Radiation and Tryptophan Metabolism in Glioblastoma

medium to strong enzyme expression observed in 61.5% of these measured by inhibition of kynurenine production, and impor- tumors when using IHC to 29% when measured by mRNA levels. tantly, pharmacokinetic/pharmacodynamic studies demonstrat- Our study is among the first to delineate this pathway at the level of ed the strong potential for this agent to cross the BBB and achieve the individual metabolites contributing to immunomodulation. biologically relevant concentrations in the brain, supporting In addition to further validating the relevance of this pathway in further preclinical investigations with this compound in glio- gliomagenesis, evaluating levels of individual metabolites provide blastoma. To determine the potential of GDC-0919 to serve as a a far more striking distinction between glioblastoma and LGG novel immune checkpoint inhibitor in glioblastoma, we extend- than when determined through enzyme expression. For example, ed studies in vivo using an adult nongermline GEM astrocyte cell when quantifying relative values of kynurenine and using the line that recapitulated both the genotype and aggressive pheno- highest level of kynurenine detected in LGG as a threshold to type of glioblastoma (29, 30) and displayed aberrant tryptophan define pathway activation, approximately 60% of glioblastoma metabolism. Using this model, IDO1 inhibition with GDC-0919 demonstrated aberrant tryptophan metabolism. Furthermore, did not demonstrate antitumor activity as a single agent in through global metabolomic profiling, we uncovered a previously subcutaneous and orthotopic tumors or modulate the tumors' undescribed, inverse relationship between kynurenine levels and immune profile at baseline (41, 42). Although molecular knock- its downstream metabolic intermediate kynurenate in glioma. A down of IDO1 expression has been shown to influence glio- ratio generated by coupling these two metabolites led to a highly blastoma growth (43), our results are consistent with other significant distinction between LGG and glioblastoma, with over recent publications demonstrating minimal activity in glioblas- 90% of glioblastoma having a kynurenine/kynurenate ratio above toma when inhibiting IDO1 as a single targeted agent (44). the defined threshold (Supplementary Fig. S1C). These findings Although we did not observe any changes in the absolute have unique clinical applications. For example, the differential number of Tregs following IDO1 inhibition, we did uncover an accumulation of tryptophan-related metabolites in LGG and important role by which kynurenine modulates their activity, glioblastoma can potentially be imaged through MR spectroscopy serving as a critical metabolite required for Treg-mediated sup- þ and therefore has diagnostic implications. In addition, such imag- pression of CD8 T-cell proliferation. Further work is needed to ing correlates may be extended into clinical trials, allowing for the precisely define the role kynurenine and its downstream meta- enrichment in patients with tumors likely to respond to IDO1 bolic intermediates play in modulating Treg function and deter- pathway inhibition or by supporting phase 0 investigations mine its mechanistic underpinnings. designed to confirm target engagement of a specific compound Although GDC-0919 did not demonstrate antitumor activity as designed to inhibit this pathway. a single agent in our glioblastoma model, we identified a unique In addition to evaluating the IDO1 pathway from a metabolic synergy when combined with radiation, which was most notable perspective, integrated analyses coupling these global metabolo- when using a hypofractionated regimen. As recent strategies mic signatures with gene expression profiles performed on designed to incorporate this agent in cancer treatment have been matched tumors provided further insight into the biologic impli- largely negative (45), these findings, which are consistent with cations of this pathway. These cross-platform studies demonstrat- recent studies that have generated considerable interest in the ed that tryptophan metabolism is specific to the mesenchymal potential for immune checkpoint agents to be used in concert with and classical molecular subtypes of glioblastoma, which is con- radiation (46, 47), may provide a novel direction for continued sistent with recent work identifying immune heterogeneity clinical trial development. Our results provide a novel insight into between these individual subtypes (40). Our group has recently the complex mechanisms that may be underlying these interac- reported that rather than representing intertumoral heterogeneity, tions, suggesting radiation may perturb cancer immunoediting at molecular subtypes in glioblastoma were reflective of regional several levels. For example, in our working model, RT-induced intratumoral heterogeneity, with the proneural subtype compris- tumor cell death results in the release of tumor-associated antigen/ ing the infiltrative edge of an individual tumor, whereas the peptides, ensued by antigen processing and presentation to T cells mesenchymal subtype was specific to the perinecrotic core (35, 48). This, in turn, helps stimulate the immune system and (31). Consistent with these findings, evaluating regional expres- clonal expansion as demonstrated by a robust increase in activated sion of IDO1 using the Ivy GAP Atlas demonstrated highest and cytotoxic T cells, contributing toward an immunologic shift expression of this enzyme in the perinecrotic core and lowest in from a state of "equilibrium" or "escape" back to an immunologic the infiltrative edge of individual tumors. As increased expression state of "elimination" of tumor cells (49). However, this potential of IDO1 has been recently described as a prognostic factor in for an immunologic shift that would result in enhanced antitumor glioblastoma (23), the observed differential expression of this activity appears to be dampened by parallel increases in RT- enzyme may contribute toward the therapeutic resistance typi- induced immunosuppression, most notably, a consistent increase cally attributed to this region. In addition, these findings may in Tregs. Our data, along with another recent publication (50), imply that aberrant tryptophan metabolism is a late event in suggests that the observed increase in activated and cytotoxic T cells gliomagenesis and rational combinatorial strategies designed to following RT may in turn stimulate tryptophan metabolism, also utilize agents specifically targeting the unique immune and/ contributing to the observed increase in Tregs. Inhibiting trypto- or signaling associated with tumor cell populations expanding in phan metabolism mitigates the observed accumulation of RT- the infiltrative edge may lead to more durable control in these induced Tregs and attenuates their suppressiveness. This results in otherwise resistant tumors. a further increase in T-cell activation and cytotoxicity, thereby Aberrant tryptophan metabolism identified in patient-derived enhancing antitumor activity by tipping the immunologic balance glioblastoma was recapitulated in our preclinical models, with toward a state of "elimination" (37), thereby providing a mech- robust pathway activation observed in all glioblastoma cell lines anistic link between IDO1 inhibition and enhanced radiation tested. The IDO1 inhibitor GDC-0919 demonstrated potent response. Focused studies involving depletion of Tregs, and other inhibition of tryptophan metabolism in these models, which was potential targets of IDO1 inhibition, including MDSCs, are still

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3641

Kesarwani et al.

required to more definitively establish mechanisms governing Acquisition of data (provided animals, acquired and managed patients, potentiation of RT response following IDO1 inhibition. provided facilities, etc.): P. Kesarwani, A. Prabhu, S. Kant, P. Kumar, In summary, aberrant tryptophan metabolism represents an S.F. Graham, K.L. Buelow, C.R. Miller, P. Chinnaiyan fi Analysis and interpretation of data (e.g., statistical analysis, biostatistics, important metabolic node in glioblastoma. Although its speci c computational analysis): P. Kesarwani, A. Prabhu, S. Kant, P. Chinnaiyan role in gliomagenesis remains unclear, this pathway appears to Writing, review, and/or revision of the manuscript: P. Kesarwani, P. Kumar, play a significant role in RT-induced immune checkpoint reacti- S.F. Graham, K.L. Buelow, G.D. Wilson, C.R. Miller, P. Chinnaiyan vation. GDC-0919 is a potent IDO1 inhibitor with the capacity of Administrative, technical, or material support (i.e., reporting or organizing crossing the BBB and enhancing RT response by mitigating RT- data, constructing databases): P. Kesarwani, S. Kant induced immunosuppression, thereby shifting the immunologic Study supervision: P. Chinnaiyan balance toward a state of "elimination," which was particularly Acknowledgments striking when using a hypofractionated approach. These encour- This work was supported by the NIH/NINDS (R21NS090087), ACS (RSG- fi aging ndings support clinical efforts designed to combine IDO1 11-029-01), Bankhead-Coley Cancer Research Program, and Cancer Research inhibition with hypofractionated RT in glioblastoma, offering the Seed Grant Awards from Beaumont Heath (P. Chinnaiyan). C.R. Miller is promise of effectively harnessing a patient's own immune system supported by NIH/NCI (R01CA204136). to attack these otherwise recalcitrant tumors. The costs of publication of this article were defrayed in part by the Disclosure of Potential Conflicts of Interest payment of page charges. This article must therefore be hereby marked No potential conflicts of interest were disclosed. advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Authors' Contributions Conception and design: P. Kesarwani, P. Chinnaiyan Received January 5, 2018; revised March 6, 2018; accepted April 20, 2018; Development of methodology: P. Kesarwani, P. Chinnaiyan published first April 24, 2018.

References 1. Ostrom QT, Gittleman H, Liao P, Vecchione-Koval T, Wolinsky Y, Kruchko 15. Wainwright DA, Balyasnikova IV, Chang AL, Ahmed AU, Moon KS, C, et al. CBTRUS Statistical Report: primary brain and other central nervous Auffinger B, et al. IDO expression in brain tumors increases the recruitment system tumors diagnosed in the United States in 2010–2014. Neuro Oncol of regulatory T cells and negatively impacts survival. Clin Cancer Res 2017;19(suppl_5):v1–v88. 2012;18:6110–21. 2. Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008; 16. Munn DH, Sharma MD, Hou D, Baban B, Lee JR, Antonia SJ, et al. 359:492–507. Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic 3. Kamradt T, Mitchison NA. Tolerance and autoimmunity. N Engl J Med cells in tumor-draining lymph nodes. J Clin Invest 2004;114:280–90. 2001;344:655–4. 17. Mezrich JD, Fechner JH, Zhang X, Johnson BP, Burlingham WJ, Bradfield 4. Larkin J, Chiarion-Sileni V, Gonzalez R, Grob JJ, Cowey CL, Lao CD, et al. CA. An interaction between kynurenine and the aryl hydrocarbon receptor Combined nivolumab and ipilimumab or monotherapy in untreated can generate regulatory T cells. J Immunol 2010;185:3190–8. melanoma. N Engl J Med 2015;373:23–34. 18. Nguyen NT, Kimura A, Nakahama T, Chinen I, Masuda K, Nohara K, et al. 5. Pardoll DM. The blockade of immune checkpoints in cancer immuno- Aryl hydrocarbon receptor negatively regulates dendritic cell immunoge- therapy. Nat Rev Cancer 2012;12:252–64. nicity via a kynurenine-dependent mechanism. Proc Natl Acad Sci U S A 6. Louveau A, Harris TH, Kipnis J. Revisiting the mechanisms of CNS immune 2010;107:19961–6. privilege. Trends Immunol 2015;36:569–77. 19. Hollingshead BD, Beischlag TV, Dinatale BC, Ramadoss P, Perdew GH. 7. Aspelund A, Antila S, Proulx ST, Karlsen TV, Karaman S, Detmar M, et al. A Inflammatory signaling and aryl hydrocarbon receptor mediate synergistic dural lymphatic vascular system that drains brain interstitial fluid and induction of interleukin 6 in MCF-7 cells. Cancer Res 2008;68:3609–17. macromolecules. J Exp Med 2015;212:991–9. 20. Quintana FJ, Basso AS, Iglesias AH, Korn T, Farez MF, Bettelli E, et al. 8. Ludwig A, Schulte A, Schnack C, Hundhausen C, Reiss K, Brodway N, et al. Control of T(reg) and T(H)17 cell differentiation by the aryl hydrocarbon Enhanced expression and shedding of the transmembrane chemokine receptor. Nature 2008;453:65–71. CXCL16 by reactive astrocytes and glioma cells. J Neurochem 2005; 21. van Baren N, Van den Eynde BJ. Tryptophan-degrading enzymes in tumoral 93:1293–303. immune resistance. Front Immunol 2015;6:34. 9. Rempel SA, Dudas S, Ge S, Gutierrez JA. Identification and localization of 22. Boyland E, Williams DC. The metabolism of tryptophan. 2. The metab- the cytokine SDF1 and its receptor, CXC chemokine receptor 4, to regions of olism of tryptophan in patients suffering from cancer of the bladder. necrosis and angiogenesis in human glioblastoma. Clin Cancer Res 2000; Biochem J 1956;64:578–82. 6:102–11. 23. Zhai L, Lauing KL, Chang AL, Dey M, Qian J, Cheng Y, et al. The role of IDO 10. Reardon DA, Omuro A, Brandes AA, Rieger J, Wick A, Sepulveda J, et al. in brain tumor immunotherapy. J Neurooncol 2015;123:395–403. OS10.3 randomized phase 3 study evaluating the efficacy and safety of 24. Friberg M, Jennings R, Alsarraj M, Dessureault S, Cantor A, Extermann M, nivolumab vs. bevacizumab in patients with recurrent glioblastoma: et al. Indoleamine 2,3-dioxygenase contributes to tumor cell evasion of T CheckMate 143. Neuro Oncol 2017;19(suppl_3):iii21–iii. cell-mediated rejection. Int J Cancer 2002;101:151–5. 11. Boussiotis VA. Molecular and biochemical aspects of the PD-1 checkpoint 25. Zhai L, Ladomersky E, Dostal CR, Lauing KL, Swoap K, Billingham LK, et al. pathway. N Engl J Med 2016;375:1767–78. Non-tumor cell IDO1 predominantly contributes to enzyme activity and 12. Kesarwani P, Kant S, Prabhu A, Chinnaiyan P. The interplay between response to CTLA-4/PD-L1 inhibition in mouse glioblastoma. Brain Behav metabolic remodeling and immune regulation in glioblastoma. Neuro Immun 2017;62:24–9. Oncol 2017;19:1308–15. 26. Chinnaiyan P, Kensicki E, Bloom G, Prabhu A, Sarcar B, Kahali S, et al. The 13. Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B, et al. metabolomic signature of malignant glioma reflects accelerated anabolic Prevention of allogeneic fetal rejection by tryptophan catabolism. Science metabolism. Cancer Res 2012;72:5878–88. 1998;281:1191–3. 27. Mao P, Joshi K, Li J, Kim SH, Li P, Santana-Santos L, et al. Mesenchymal 14. Opitz CA, Litzenburger UM, Sahm F, Ott M, Tritschler I, Trump S, et al. An glioma stem cells are maintained by activated glycolytic metabolism endogenous tumour-promoting ligand of the human aryl hydrocarbon involving aldehyde dehydrogenase 1A3. Proc Natl Acad Sci U S A 2013; receptor. Nature 2011;478:197–203. 110:8644–9.

3642 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Radiation and Tryptophan Metabolism in Glioblastoma

28. Prabhu A, Sarcar B, Miller CR, Kim SH, Nakano I, Forsyth P, et al. Ras- protein in normal and tumoral human tissues. Cancer Immunol Res mediated modulation of pyruvate dehydrogenase activity regulates mito- 2015;3:161–72. chondrial reserve capacity and contributes to glioblastoma tumorigenesis. 40. Doucette T, Rao G, Rao A, Shen L, Aldape K, Wei J, et al. Immune Neuro Oncol 2015;17:1220–30. heterogeneity of glioblastoma subtypes: extrapolation from the cancer 29. Song Y, Zhang Q, Kutlu B, Difilippantonio S, Bash R, Gilbert D, et al. genome atlas. Cancer Immunol Res 2013;1:112–22. Evolutionary etiology of high-grade astrocytomas. Proc Natl Acad Sci U S A 41. Demaria S, Kawashima N, Yang AM, Devitt ML, Babb JS, Allison JP, et al. 2013;110:17933–8. Immune-mediated inhibition of metastases after treatment with local 30. Vitucci M, Karpinich NO, Bash RE, Werneke AM, Schmid RS, White KK, radiation and CTLA-4 blockade in a mouse model of breast cancer. Clin et al. Cooperativity between MAPK and PI3K signaling activation is Cancer Res 2005;11(2 Pt 1):728–34. required for glioblastoma pathogenesis. Neuro Oncol 2013;15:1317–29. 42. Wei H, Zhao L, Li W, Fan K, Qian W, Hou S, et al. Combinatorial PD-1 31. Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, et al. blockade and CD137 activation has therapeutic efficacy in murine cancer Integrated genomic analysis identifies clinically relevant subtypes of glio- models and synergizes with cisplatin. PLoS One 2013;8:e84927. blastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and 43. Wainwright DA, Chang AL, Dey M, Balyasnikova IV, Kim CK, Tobias A, et al. NF1. Cancer Cell 2010;17:98–110. Durable therapeutic efficacy utilizing combinatorial blockade against IDO, 32. Prabhu A, Kesarwani P, Kant S, Graham SF, Chinnaiyan P. Histologically CTLA-4, and PD-L1 in mice with brain tumors. Clin Cancer Res 2014; defined intratumoral sequencing uncovers evolutionary cues into con- 20:5290–301. served molecular events driving gliomagenesis. Neuro Oncol 2017;19: 44. Li M, Bolduc AR, Hoda MN, Gamble DN, Dolisca SB, Bolduc AK, et al. The 1599–606. indoleamine 2,3-dioxygenase pathway controls complement-dependent 33. Taylor MW, Feng GS. Relationship between interferon-gamma, indo- enhancement of chemo-radiation therapy against murine glioblastoma. J leamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J 1991;5: Immunother Cancer 2014;2:21. 2516–22. 45. Burris HA, Gordon MS, Hellmann MD, LoRusso P, Emens LA, Hodi FS, et al. 34. Zhai L, Spranger S, Binder DC, Gritsina G, Lauing KL, Giles FJ, et al. A phase Ib dose escalation study of combined inhibition of IDO1 (GDC- Molecular pathways: targeting IDO1 and other tryptophan dioxygenases 0919) and PD-L1 (atezolizumab) in patients (pts) with locally advanced or for cancer immunotherapy. Clin Cancer Res 2015;21:5427–33. metastatic solid tumors. J Clin Oncol 2017;35(15_suppl):105–. 35. Garnett CT, Palena C, ChakrabortyM,TsangKY,SchlomJ,HodgeJW. 46. Formenti SC. Silvia Formenti on the promise of combining radiotherapy Sublethal irradiation of human tumor cells modulates phenotype and immunotherapy to treat cancer. Oncology 2016;30:289, 292. resulting in enhanced killing by cytotoxic T lymphocytes. Cancer Res 47. Seyedin SN, Schoenhals JE, Lee DA, Cortez MA, Wang X, Niknam S, et al. 2004;64:7985–94. Strategies for combining immunotherapy with radiation for anticancer 36. Gros A, Robbins PF, Yao X, Li YF, Turcotte S, Tran E, et al. PD-1 identifies the therapy. Immunotherapy 2015;7:967–80. patient-specific CD8(þ) tumor-reactive repertoire infiltrating human 48. Reits EA, Hodge JW, Herberts CA, Groothuis TA, Chakraborty M, Wansley tumors. J Clin Invest 2014;124:2246–59. EK, et al. Radiation modulates the peptide repertoire, enhances MHC class I 37. Kalbasi A, June CH, Haas N, Vapiwala N. Radiation and immunotherapy: a expression, and induces successful antitumor immunotherapy. J Exp Med synergistic combination. J Clin Invest 2013;123:2756–63. 2006;203:1259–71. 38. Mitsuka K, Kawataki T, Satoh E, Asahara T, Horikoshi T, Kinouchi H. 49. Dunn GP, Old LJ, Schreiber RD. The three Es of cancer immunoediting. Expression of indoleamine 2,3-dioxygenase and correlation with patho- Annu Rev Immunol 2004;22:329–60. logical malignancy in gliomas. Neurosurgery 2013;72:1031–8. 50. Zhai L, Ladomersky E, Lauing KL, Wu M, Genet M, Gritsina G, et al. 39. Theate I, van Baren N, Pilotte L, Moulin P, Larrieu P, Renauld JC, et al. Infiltrating T cells increase IDO1 expression in glioblastoma and contribute Extensive profiling of the expression of the indoleamine 2,3-dioxygenase 1 to decreased patient survival. Clin Cancer Res 2017;23:6650–60.

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3643

Tryptophan Metabolism Contributes to Radiation-Induced Immune Checkpoint Reactivation in Glioblastoma

Pravin Kesarwani, Antony Prabhu, Shiva Kant, et al.

Clin Cancer Res 2018;24:3632-3643. Published OnlineFirst April 24, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-0041

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2018/04/24/1078-0432.CCR-18-0041.DC1

Cited articles This article cites 50 articles, 19 of which you can access for free at: http://clincancerres.aacrjournals.org/content/24/15/3632.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/24/15/3632.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/24/15/3632. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.