Immune and Cell Enrichment Is Associated with a Good Prognosis in

This information is current as Andrew M. Donson, Diane K. Birks, Valerie N. Barton, Qi of September 27, 2021. Wei, Bette K. Kleinschmidt-DeMasters, Michael H. Handler, Allen E. Waziri, Michael Wang and Nicholas K. Foreman J Immunol 2009; 183:7428-7440; Prepublished online 16 November 2009;

doi: 10.4049/jimmunol.0902811 Downloaded from http://www.jimmunol.org/content/183/11/7428

References This article cites 51 articles, 11 of which you can access for free at: http://www.jimmunol.org/content/183/11/7428.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Immune Gene and Cell Enrichment Is Associated with a Good Prognosis in Ependymoma1

Andrew M. Donson,2*§ Diane K. Birks,‡§ Valerie N. Barton,*§ Qi Wei,*§ Bette K. Kleinschmidt-DeMasters,† Michael H. Handler,‡§ Allen E. Waziri,‡ Michael Wang,*§ and Nicholas K. Foreman*‡§

Approximately 50% of children with ependymoma will suffer from tumor recurrences that will ultimately lead to death. Devel- opment of more effective therapies and patient stratification in ependymoma mandates better prognostication. In this study, tumor microarray profiles from pediatric ependymoma clinical samples were subject to ontological analyses to identify outcome-associated biological factors. was subsequently used to evaluate the results of ontological analyses. Ontology analyses revealed that associated with nonrecurrent ependymoma were predominantly immune function-related. Addition- ally, increased expression of immune-related genes was correlated with longer time to progression in recurrent ependymoma. Of Downloaded from those genes associated with both the nonrecurrent and that positively correlated with time to progression, 95% were associated with immune function. Histological analysis of a subset of these immune function genes revealed that their expression was restricted to a subpopulation of tumor-infiltrating cells. Analysis of tumor-infiltrating immune cells showed increased infil- tration of CD4؉ T cells in the nonrecurrent . No genomic sequences for SV40, BK, JC, or Merkel polyomaviruses were found in nonrecurrent ependymoma. This study reveals that up-regulation of immune function genes is the predominant http://www.jimmunol.org/ ontology associated with a good prognosis in ependymoma and it provides preliminary evidence of a beneficial host proinflam- matory and/or Ag-specific immune response. The Journal of Immunology, 2009, 183: 7428–7440.

pendymoma (EPN),3 the third most common recur from overly aggressive treatments. Identification of prognos- of children, is treated by surgical resection and radiation tic markers for EPN may have the added benefit of providing in- E therapy (1, 2). Complete resection, often requiring “sec- sight into the biological mechanisms of tumorigenesis, which ond-look” surgery, is critical for a favorable outcome (3, 4). Ra- could be exploited for the development of more effective therapies. diation therapy is also standard, and omission of this results in a To date, study of candidate prognostic markers for pediatric

higher number of tumor recurrences (4, 5). has so EPN have largely been confined to histological according by guest on September 27, 2021 far shown little or no benefit. Unfortunately, Ͼ50% of children to World Health Organization (WHO) tumor classification criteria treated with the standard regimen will suffer from tumor recur- (7–11), as well as to molecular markers such as Ki-67 (12, 13), rence, which will ultimately result in death (6). This high failure survivin (14, 15), human reverse transcriptase (16), and rate represents one of the most significant problems in pediatric (4). More recently, global molecular analyses such as neuro-oncology. Despite unfavorable outcome in more than half of array comparative genomic hybridization (17, 18) and gene ex- pediatric EPN patients, little progress has been made in the past 20 pression profiling (17, 19–21) have been employed to discover years either in treatment or identification of robust prognostic fac- prognostic chromosomal aberrations or gene expression signa- tors. The ability to identify up-front those EPN patients whose tures. These global studies have produced an even wider range of tumor will recur would allow clinicians to try more aggressive candidate prognostic markers, although none to date have identi- treatment regimens, better stratify patients on various treatment fied a biological mechanism of recurrence. Despite these numerous protocols, and spare those children whose tumors are unlikely to studies, there remains no predictor of tumor recurrence in EPN that is robustly reproducible from study to study. The driving hypoth- esis for this study is that gene expression patterns differ between *Department of Pediatrics, †Departments of and Neurology, and ‡Depart- ment of Neurosurgery, University of Colorado Denver, Aurora, CO 80045; and §The good and bad prognosis EPN, the details of which will allow for Children’s Hospital, Denver, CO 80045 better prognostication and provide insights into the biology of re- Received for publication August 25, 2009. Accepted for publication October 5, 2009. currence. To achieve this, tumor gene expression profiling com- The costs of publication of this article were defrayed in part by the payment of page bined with analysis was used as an unbiased ap- charges. This article must therefore be hereby marked advertisement in accordance proach to identify sets of functionally related genes that were with 18 U.S.C. Section 1734 solely to indicate this fact. associated with clinical outcome in EPN clinical samples. Using 1 This work was supported by the Tanner Seebaum Foundation. this approach, it was found that an up-regulation of immune func- 2 Address correspondence and reprint requests to Andrew M. Donson, Department of tion-related genes was the predominant ontology associated with a Pediatrics, University of Colorado Denver, Mail Stop 8302, P.O. Box 6511, Aurora, CO 80045. E-mail address: [email protected] complete response to therapy. 3 Abbreviations used in this paper: EPN, ependymoma; AIF-1, allograft inhibitory factor-1; DAVID; Database for Annotation, Visualization, and Integrated Discovery; FDR, false discovery rate; FFPE, formalin-fixed paraffin-embedded; gcRMA, Gene- Materials and Methods Chip robust multiarray average; GO, Gene Ontology; GSEA, Gene Set Enrichment Patient cohort Analysis; IHC, ; TIL, tumor-infiltrating lymphocyte; TTP, time to progression; GOTERM, Gene Ontology Project term. Surgical tumor samples were obtained from 19 patients who presented between 1997 and 2007 for treatment at The Children’s Hospital (Denver, Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 CO) who were diagnosed with EPN according to WHO guidelines (22). All www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902811 The Journal of Immunology 7429

identified by the user as significantly associated with a particular phenotype Table I. Patient cohort demographic and tumor detailsa or variable. Immunohistochemistry (IHC) Patient TTP Grade Age at ID Outcome (months) (WHO) Location Gender Dx (years) IHC was performed on 5-␮m FFPE tumor tissue sections. Slides were deparaffinized and then subjected to optimal Ag retrieval protocols. Sub- 80 Non — II IF M 2 110 Rec 31 II IF M 2 sequent steps were performed using the EnVision-HRP (Dako) on a 135 Rec 6 III ST F 14 Dako autostainer according to standard protocol. Incubation with primary 195 Rec 24 III IF M 2 Ab was performed for 2 h. The following dilutions of primary Ab were 242 Rec 1 III ST F 4 used, and applied to the sections for 1 h: 1/250 allograft inhibitory factor-1 246 Rec 35 II IF M 2 (AIF-1) (01-1974) from Waco Pure Chemicals; 1/50 HLA-DR (LN3) and 285 Non — III ST F 5.5 1/40 CD4 (IF6) from Novocastra; 1/100 CD8 (C8/144B), 1/200 CD20 306 Non — II IF M 13 ϩ 318 Non — III IF F 2 (L26), 1/50 CD45 (2B11 PD7/26), and 1/100 CD68 (PG-M1) from 319 Rec 51 II IF F 6 Dako. Each of these Abs stained a discrete subpopulation of cells that were 364 Rec 35 II IF M 13 distributed throughout the parenchyma of the tumor. Slides were analyzed 388 Non — III ST M 11 with the Olympus BX40 microscope, ϫ40 objective lens. Images were 392 Non — II IF F 1 captured using an Optronics MicroFire 1600 ϫ 1200 camera and Picture- 393 Rec 18 II IF F 6 Frame 2.3 imaging software (Optronics). Infiltrating cell abundancies were 416 Non — III ST M 5 measured as the mean number of positive staining cells per five fields of 419 Non — II IF M 3 459 Non — III IF F 0.5 view and differential expression between groups was determined using a 483 Rec 23 II IF M 7 Student’s t test with a p value cutoff of 0.05. 507 Rec 5 III IF M 5 Quantitative PCR for viral sequences Downloaded from a — denotes that tumor did not recur. Non, nonrecurrent; Rec, recurrent; WHO, World Health Organization tumor grade classification; IF, infratentorial; ST, supra- Quantitative PCR was performed for SV40, BK, JC, and Merkel polyoma- tentorial; Dx, diagnosis. viruses (PyV). DNA was extracted from surgical specimens using the Gen- trapure DNA extraction kit (Qiagen). All PCR analyses was performed using the ABI 7500 sequence dector (Applied Biosystems). TaqMan prim- ers and probes were synthesized by an Applied Biosystems facility. Probes Ј Ј patients included in the study were treated uniformly, undergoing complete were dual-labeled at the 5 end with FAM and the 3 end with TAMRA. A tumor resection followed by . Samples used in this study sequence search was performed to ensure the specificity of each http://www.jimmunol.org/ were obtained at the time of initial resection and before radiation therapy. primer/probe set. TaqMan PCR amplification data were analyzed with soft- Two tumor samples were collected for each patient: one sample was snap- ware provided by the manufacturer. All samples were tested in duplicate. frozen in liquid nitrogen, and one was formalin-fixed paraffin-embedded Results were expressed as cycle threshold (Ct), which was proportional to (FFPE) for routine light microscopy. Outcome data were available for all the starting copy numbers and was defined as the PCR cycle at which the patients in this study, which was conducted in compliance with Institu- fluorescence signal of the PCR kinetics exceeds the threshold value of the tional Review Board regulations (COMIRB 95-500 and 05-0149). Patient respective analysis. details are described in Table I. Results Gene expression microarray analysis Patient demographics, tumor grade, or location do not influence Five micrograms of RNA that had been extracted from tumor was ampli- risk of recurrence by guest on September 27, 2021 fied, biotin-labeled (Enzo Biochem), and hybridized to Affymetrix HG- In this study the median follow-up for nonrecurrent EPN patients U133 Plus 2 microarray chips. Analysis of gene expression microarray data was 5 years 3 mo. The median time to progression (TTP) for re- was performed using the Bioconductor R programming language (www. bioconductor.org). Microarray data were background corrected and nor- current EPN patients was 2 years. No statistically significant dif- malized using the guanine cytosine robust multiarray average (gcRMA) ference was seen between recurrent and nonrecurrent EPN patients algorithm (23), resulting in log2 expression values. The Affymetrix HG- with respect to tumor WHO grade, location, age at diagnosis, or U133 Plus 2 microarray contains 54,675 probe sets, including multiple gender. In those patients with recurrent EPN, anaplastic EPN probe sets for the same gene. To reduce errors associated with multiple testing, a filtered list containing a single probe set for each gene that (WHO grade III) had a significantly shorter TTP than did classic possessed the highest gcRMA expression level across all samples used EPN (WHO grade II) (9 mo vs 32 mo, respectively; p ϭ 0.012). A was created (18,624 genes). The microarray data discussed in this pub- shorter TTP was also seen in supratentorial vs infratentorial tumors lication are Minimum Information About a Microarray Experiment (3 mo vs 28 mo, respectively; p ϭ 0.044). No significant correla- (MIAME) compliant and have been deposited in National Center for tion was observed between TTP and either age at diagnosis or Biotechnology Information’s Gene Expression Omnibus (24) and are accessible through Gene Expression Omnibus series accession no. gender in recurrent patients. GSE16155 (www.ncbi.nlm.nih.gov/geo/query/). Genes associated with nonrecurrent EPN are predominantly Gene ontology analyses immune-related Two computer-based ontology analysis tools were used in this study: Gene expression microarray profiles generated from surgical spec- GSEA (Gene Set Enrichment Analysis: www.broad.mit.edu/gsea) (25) and imens of EPN at initial presentation were separated into 2 groups: DAVID (Database for Annotation, Visualization, and Integrated Discov- nonrecurrent (n ϭ 9) and recurrent (n ϭ 10). In the first gene ery: http://david.abcc.ncifcrf.gov) (26). Both analyses were used to assess gene lists for enrichment of genes annotated with specific Gene Ontology ontology analysis, GSEA was used to identify enriched biological Project terms (GOTERM; www.geneontology.org) (27). Enrichment is de- function in genes associated with either the nonrecurrent or the fined as more genes than would be expected by chance that are associated recurrent groups, respectively termed “the nonrecurrent pheno- with a specific phenotype or variable. type” and “the recurrent phenotype” (Table II). This revealed that Briefly, GSEA takes gene expression profiles that have been assigned a “adaptive immune response” was the most highly enriched specific phenotype (e.g., nonrecurrent or recurrent) or a continuous variable (e.g., time to progression) and creates a ranked list of genes based on the GOTERM in the nonrecurrent phenotype with a FDR of 0.059. In strength of the association with the phenotype or variable being interro- the recurrent phenotype the most enriched GOTERM was “gluta- gated. The output is an enrichment score with associated false discovery mate signaling pathway”, which did not reach statistical signifi- rate (FDR) adjusted q values and Student’s t test p values for each Gene cance by FDR (0.355). Ontology term. A Benjamini FDR cutoff of 0.25 was used as recommended by GSEA. DAVID was used as an additional measure of gene function DAVID is a web-based resource that provides Gene Ontology term en- enrichment. Two lists of genes that were associated either with richment scores for lists of genes that, unlike GSEA, have already been nonrecurrent or recurrent were generated before 7430 PROGNOSTIC IMMUNE FACTORS IN EPENDYMOMA

Table II. Ontologic analyses of genes associated with the nonrecurrent and recurrent phenotypes in EPNa

Enrichment

GOTERM Annotation GOTERM ID q Value p Value

GSEA ontology analysis Nonrecurrent phenotype Adaptive immune response 2250 0.059 0.00614 Adaptive immune response 2460 0.084 0.0103 Humoral immune response 6959 0.146 0.0294 Phagocytosis 6909 0.164 0.0120 Immune effector process 2252 0.224 0.00789 Recurrent phenotype Glutamate signaling pathway 7215 0.355 0.0435 Aromatic compound metabolic process 6725 0.450 0.0057 Regulation of G -coupled 8277 0.467 0.0468 protein signaling pathway Meiosis I 7127 0.468 0.0312 Regulation of muscle contraction 6937 0.476 0.0248 DAVID ontology analysis Nonrecurrent phenotype Immune response 6955 9.40 ϫ 10Ϫ9 4.91 ϫ 10Ϫ12 Downloaded from Immune system process 2376 9.84 ϫ 10Ϫ9 5.14 ϫ 10Ϫ12 Response to wounding 9611 5.79 ϫ 10Ϫ7 3.03 ϫ 10Ϫ10 Response to external stimulus 9605 1.49 ϫ 10Ϫ6 7.77 ϫ 10Ϫ10 Inflammatory response 6954 1.70 ϫ 10Ϫ5 8.87 ϫ 10Ϫ9 Recurrent phenotype Multicellular organismal process 32501 0.98 7.23 ϫ 10Ϫ4

Anatomical structure development 48856 0.95 0.00113 http://www.jimmunol.org/ Multicellular organismal development 7275 0.99 0.00244 Biological regulation 65007 1.00 0.00477 Developmental process 32502 0.99 0.00489

a The top five enriched ontologies for each phenotype ranked according to FDR (q value) are shown. Statistical significance is defined as FDR Ͻ 0.25 and Student’s t test p Ͻ 0.05.

DAVID analysis as input. One hundred twenty-seven of the 18,624 (HLA-DMA, HLA-DMB, HLA-DPB1, HLA-DRB5, and CD74). genes used in this analysis were overexpressed (Ͻ2-fold; p Ͻ MHC class II is predominantly expressed on APCs, the most pre- 0.05) in nonrecurrent EPN vs recurrent EPN groups. DAVID dem- dominant of which in the CNS are thought to be the microglia/ by guest on September 27, 2021 onstrated that the GOTERM “immune response” was the most macrophage population. A number of other genes that are associ- significantly enriched ontology (FDR ϭ 9.4 ϫ 10Ϫ9) in the non- ated with microglia or macrophages were found to be recurrent phenotype (Table II). In contrast, the most enriched overexpressed in the nonrecurrent EPN phenotype. Among these GOTERM in the recurrent EPN phenotype (47 genes) was “mul- was AIF1, which is a specific marker of activated microglia/mac- ticellular organismal process”, which was not statistically signifi- rophages (28). In the context of adaptive immune function, a num- cant by FDR (0.98). ber of genes specifically involved in T lymphocyte activity were Both GSEA and DAVID identified immune function-related associated with the nonrecurrent phenotype, including TCR ␣ con- genes as the most enriched ontology in the nonrecurrent EPN phe- stant (TRAC), CD37, FYN binding protein (FYB), hepatitis A notype. By contrast, there was no overlap in gene ontology en- virus cellular receptor 2 (HAVCR2), hematopoietic cell-specific richments identified by GSEA and DAVID in the recurrent EPN Lyn substrate-1 (HCLS1), and linker for activation of T cells fam- phenotype, nor did either approach identify any statistically sig- ily member 2 (LAT2). Other notable immune function-related nificant enrichment by FDR. genes associated with the nonrecurrent phenotype are Fc receptors A detailed analysis of the genes associated with nonrecurrent CD64A and B, STAT6, TNF (ligand) superfamily, member 10 EPN phenotype was performed to elaborate the results of the (TRAIL), and cytochrome b-245, ␣ and ␤ polypeptides (CYBA computer-based ontological analyses described above. All of and CYBB). the 127 genes that were overexpressed in nonrecurrent EPN (Ͼ2-fold; p Ͻ 0.05) were evaluated for their potential roles in In recurrent EPN, genes that positively correlate with longer any immune-related process as described in peer-reviewed pub- time to progression are predominantly immune-related lications. Fifty-four percent (68 out of127) of these genes had EPN generally recur within 3 years of initial presentation. Our documented immune-related functions. This approach identified recurrent EPN cohort had TTP ranging from 1 to 51 mo. To iden- a number of immune-related genes beyond those identified by tify genes associated with TTP in EPN that recurred (n ϭ 10), GSEA or DAVID; these genes had erroneously not been as- microarray gene expression data were correlated with TTP as a signed an annotation of immune function by GO. continuous variable using a modified version of the GSEA ap- Of the immune-related genes overexpressed in nonrecurrent proach described above. GSEA identified “humoral immune re- EPN, a number of genes that are involved in both innate and adap- sponse” as the highest enriched GOTERM in genes that positively tive immune responses were identified (Table III). Key initiating correlated with TTP (FDR ϭ 0.0694) (Table IV). In the reverse components of both the classical and lectin complement innate analysis, “biological process” was the highest enriched GOTERM response pathways (C1QC and MASP1, respectively) and down- in genes that negatively correlated with TTP (FDR ϭ 0.223), al- stream complement components C3, C3AR1, and ITGB2 (integrin though these were less statistically significant than was the im- ␤ 2) were identified. Multiple MHC class II alleles were identified mune gene correlation. The Journal of Immunology 7431

Table III. Immune function related genes overexpressed the nonrecurrent EPN phenotypea

Affymetrix Fold Gene Symbol Gene Name Probeset ID Increase P Value

Innate immune response APOBEC3G Apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like 3G 204205_at 4.89 0.00616 IRF7 IFN regulatory factor 7 208436_s_at 3.90 0.0134 OLR1 Oxidized low-density lipoprotein (lectin-like) receptor 1 210004_at 4.36 0.0417 RNASE6 Ribonuclease, RNase A family, k6 213566_at 2.76 0.0332 TREM2 Triggering receptor expressed on myeloid cells 2 219725_at 2.92 0.0363 TRIM22 Tripartite motif-containing 22 213293_s_at 3.79 0.0264 TRIM34 Tripartite motif-containing 34 221044_s_at 4.06 0.00371 TYROBP TYRO protein tyrosine kinase binding protein 204122_at 2.30 0.0183 Inflammation FRZB Frizzled-related protein 203697_at 6.89 0.00639

PLA2G4C Phospholipase A2, group IVC (cytosolic, calcium-independent) 209785_s_at 2.31 0.0464 PROS1 Protein S (␣) 207808_s_at 2.58 0.0287 Complement C1QC Complement component 1, q subcomponent, C chain 225353_s_at 2.46 0.00693 C3 Complement component 3 217767_at 4.43 0.0155 C3AR1 Complement component 3a receptor 1 209906_at 2.25 0.0301 ITGB2 Integrin, ␤ 2 (complement component 3 receptor 3 and 4 subunit) 202803_s_at 2.35 0.0445 Downloaded from MASP1 Mannan-binding lectin serine peptidase 1 (C4/C2-activating component 232224_at 2.55 0.0170 of Ra-reactive factor) SERPING1 Serpin peptidase inhibitor, clade G (C1 inhibitor), member 1 200986_at 2.77 0.0368 (angioedema, hereditary) Macrophage/microglia AIF1 Allograft inflammatory factor 1 215051_x_at 2.62 0.00730 ␤ ␤ ϫ Ϫ5 B3GALT4 UDP-Gal: GlcNAc 1,3-galactosyltransferase, polypeptide 4 210205_at 4.68 9.30 10 http://www.jimmunol.org/ CD74 CD74 Ag (invariant polypeptide of MHC, class II Ag-associated) 209619_at 2.13 0.0296 CLEC7A C-type lectin domain family 7 member A 221698_s_at 2.47 0.0312 CSF1R CSF1 receptor 203104_at 2.52 0.0204 HLA-DMA MHC, class II, DM␣ 217478_s_at 3.30 0.0319 HLA-DMB MHC, class II, DM␤ 203932_at 2.18 0.0268 HLA-DPB1 MHC, class II, DP␤ 1 201137_s_at 2.36 0.0251 HLA-DRB5 MHC, class II, DR␤ 5 208306_x_at 2.10 0.0449 IFIT3 IFN-induced protein with tetratricopeptide repeats 3 229450_at 3.85 0.0229 LILRB1 Leukocyte Ig-like receptor, subfamily b (with TM and ITIM domains), 229937_x_at 3.33 0.0389 member 1 NAPSB Napsin B aspartic peptidase pseudogene 228055_at 3.98 0.0478 by guest on September 27, 2021 SYK Spleen tyrosine kinase 226068_at 2.35 0.0307 Adaptive immune response T cell CD37 Leukocyte Ag CD37 204192_at 2.52 0.0218 FYB FYN-binding protein (FYB-120/130) 211795_s_at 2.96 0.0272 GPR65 G protein-coupled receptor 65 214467_at 2.54 0.0184 HAVCR2 Hepatitis A virus cellular receptor 2 235458_at 3.50 0.0391 HCLS1 Hematopoietic cell-specific Lyn substrate 1 202957_at 2.62 0.0385 LAT2 Linker for activation of T cells family member 2 221581_s_at 2.41 0.00564 PTPRJ Protein tyrosine phosphatase receptor J 227396_at 2.53 0.0388 TRAC TCR␣ constant 209670_at 3.29 0.00156 B cell BLNK B cell linker 207655_s_at 3.56 0.00115 GALNAC4S B cell RAG-associated protein 203066_at 3.06 0.0205 LPXN Leupaxin 216250_s_at 2.18 0.00254 MS4A6A CD20-like precursor 223280_x_at 2.82 0.0376 Antibody FCGR1A Fc fragment of IgG, high-affinity Ia (CD64A) 216950_s_at 6.27 0.00324 FCGR1B Fc fragment of IgG, high-affinity Ib (CD64B) 214511_x_at 7.07 0.00551 Cytokines, chemokines, and cytokine signaling CNTNAP1 Contactin-associated protein 1 219400_at 2.77 0.0345 IFIT1 IFN-induced protein with tetratricopeptide repeats 1 203153_at 2.71 0.0391 RARRES3 Retinoic acid receptor responder (tazarotene induced) 3 204070_at 2.38 0.0276 STAT6 Signal transducer and activator of transcription 6, IL-4 induced 201331_s_at 4.72 0.00153 TNFSF10 TNF (ligand) superfamily, member 10 202688_at 2.68 0.0470 XAF1 Xiap-associated factor-1 228617_at 3.97 0.00365 Oxidative burst ALOX5 Arachidonate 5-lipoxygenase 204446_s_at 4.25 0.0382 CYBA Cytochrome b-245, ␣ polypeptide 203028_s_at 2.67 0.0219 CYBB Cytochrome b-245, ␤ polypeptide 203923_s_at 4.08 0.0297 HCK Hemopoietic cell kinase 208018_s_at 3.66 0.0247 Tethering and rolling of immune cells APBB1IP Amyloid ␤ (a4) precursor protein-binding, family b, member 1 230925_at 2.67 0.0454 interacting protein (Table continues) 7432 PROGNOSTIC IMMUNE FACTORS IN EPENDYMOMA

Table III. (Continued)

Affymetrix Fold Gene Symbol Gene Name Probeset ID Increase P Value

ARHGAP4 Rho GTPase-activating protein 4 204425_at 5.45 0.0150 CORO1A Coronin, actin binding protein, 1A 209083_at 2.72 0.0187 DOCK2 Dedicator of cytokinesis protein 2 213160_at 2.51 0.0274 SELPLG Selectin P ligand 209879_at 3.15 0.0180 Hematopoietic cells ADAM28 ADAM metallopeptidase domain 28 205997_at 3.58 0.0373 CD300A CD300A Ag 209933_s_at 3.12 0.0300 ENTPD1 Ectonucleoside triphosphate diphosphohydrolase 1 209473_at 2.07 0.0491 PTPN6 Protein tyrosine phosphatase, nonreceptor type 6 206687_s_at 2.45 0.0140 Miscellaneous immune-related FBLN1 Fibulin 1 202994_s_at 2.41 0.0375 GIMAP2 GTPase IMAP family member 2 232024_at 2.33 0.0252 LY75 Lymphocyte Ag 75 205668_at 4.28 0.0158 PLAC8 Placenta-specific 8 219014_at 2.07 0.0330 SIGLEC10 Sialic acid binding Ig-like lectin 10 1552807_a_at 3.55 0.00521

a Genes overexpressed in the nonrecurrent EPN phenotype (Ͼ2-fold; p Ͻ 0.05) are categorized into specific immune categories according to peer-reviewed publications. The

level of overexpression is measured by fold increase and Student’s t test ( p value). Downloaded from

As an input for DAVID, a list of 395 genes that positively cor- Detailed analysis of genes that positively correlated with TTP related ( p Ͻ 0.05 estimated by two-sided Pearson correlation test) in recurrent EPN was performed to elaborate the results of the with TTP in recurrent EPN (n ϭ 10) was created using all 18,624 above computer-based ontological analyses. Twenty-eight per-

genes. Using the same approach, a list of 841 genes that were neg- cent (110 out of 395) of the genes positively correlated with http://www.jimmunol.org/ atively correlated with TTP was also created. Similar to the GSEA TTP in recurrent EPN with statistical significance ( p Ͻ 0.05) results, DAVID confirmed that immune function-related was the most were related to immune function. The results of this analysis, enriched ontology in genes that positively correlated with TTP (Table with genes listed and categorized into subgroups according to IV). Cell cycle-related ontologies were found to be enriched in genes their documented role in specific immune mechanisms, are pro- that were negatively correlated with TTP (FDR ϭ 2.11 ϫ 10Ϫ13), vided in Table V. As found in the previous analysis, a number with greater statistical significance than the positive TTP correlate- of genes beyond those identified by GSEA or DAVID were enriched ontologies (FDR ϭ 7.72 ϫ 10Ϫ7). found, due to their not having been assigned an annotation of by guest on September 27, 2021

Table IV. Gene ontology analyses of genes positively and negatively correlated with longer time to progression in EPNa

Enrichment

GOTERM Annotation GOTERM ID q Value Value

GSEA ontology analysis Positively correlated with TTP Humoral immune response 6959 0.0694 0.00 Extracellular structure organization and biogenesis 43062 0.308 0.0248 Synaptogenesis 7416 0.356 0.111 Synapse organization and biogenesis 50808 0.372 0.0462 Phagocytosis 6909 0.463 0.173 Negatively correlated with TTP Biological process 6270 0.223 0.0402 Spliceosome assembly 245 0.224 0.0532 Biological process 7093 0.227 0.0226 Ribonucleotide metabolic process 9259 0.228 0.0331 Sister chromatid segregation 819 0.239 0.00592 DAVID ontology analysis Positively correlated with TTP Defense response 6952 7.72 ϫ 10Ϫ7 4.03 ϫ 10Ϫ10 Immune response 6955 7.78 ϫ 10Ϫ6 4.07 ϫ 10Ϫ9 Immune system process 2376 2.66 ϫ 10Ϫ5 1.39 ϫ 10Ϫ8 Response to wounding 9611 2.55 ϫ 10Ϫ4 1.33 ϫ 10Ϫ7 Ag binding 3823 3.14 ϫ 10Ϫ4 1.75 ϫ 10Ϫ7 Negatively correlated with TTP Cell cycle 7049 2.11 ϫ 10Ϫ13 6.58 ϫ 10Ϫ17 DNA metabolic process 6259 4.67 ϫ 10Ϫ12 2.41 ϫ 10Ϫ15 Nucleobase, nucleoside, nucleotide, and 6139 6.58 ϫ 10Ϫ12 3.49 ϫ 10Ϫ15 nucleic acid metabolic process Mitotic cell cycle 278 1.68 ϫ 10Ϫ11 8.81 ϫ 10Ϫ15 Biopolymer metabolic process 43283 4.69 ϫ 10Ϫ11 2.45 ϫ 10Ϫ14

a The top five enriched ontologies for positive and negative correlates of longer TTP ranked according to FDR (q value) are shown. Statistical significance is defined as FDR Ͻ 0.25. The Journal of Immunology 7433

Table V. Immune-related genes positively correlated with TTPa

Affymetrix Gene Symbol Gene Name Probeset ID RpValue

Innate immune response CLEC1 Dendritic cell-associated lectin-1 1561899_at 0.64 0.0443 GATA6 GATA binding protein 6 210002_at 0.70 0.0233 RARA Retinoic acid receptor, ␣ 203750_s_at 0.74 0.0152 SIGLEC1 Sialic acid binding Ig-like lectin 1, 44673_at 0.79 0.00617 Viral response MX2 Myxovirus (influenza virus) resistance 2 (mouse) 204994_at 0.84 0.00228 OAS2 2Ј-5Ј-oligoadenylate synthetase 2, 69/71 kDa 204972_at 0.65 0.0408 TMC8 Transmembrane channel-like 8 227353_at 0.75 0.0130 Inflammation GPR84 G protein-coupled receptor 84 223767_at 0.69 0.0280 NFX1 Nuclear , X-box binding 1 1553103_at 0.73 0.0171 PSD Pleckstrin and Sec7 domain containing 208102_s_at 0.77 0.00982 TPSAB1 Tryptase ␣ 215382_x_at 0.66 0.0371 Complement C2 Complement component 2 203052_at 0.87 0.00109 C3 Complement component 3 217767_at 0.67 0.0351 C3AR1 Complement component 3a receptor 1 209906_at 0.69 0.0263 Downloaded from C6 Complement component 6 210168_at 0.71 0.0211 C7 Complement component 7 202992_at 0.68 0.0312 CD53 CD53 Ag 203416_at 0.69 0.0277 CD59 CD59 Ag, complement regulatory protein 200985_s_at 0.82 0.00361 CR1 Complement component (3b/4b) receptor 1 217552_x_at 0.69 0.0264 ITGB2 Integrin, ␤ 2 (complement component 3 receptor 3 and 4 subunit) 202803_s_at 0.73 0.0165

TLR signaling http://www.jimmunol.org/ BTK Bruton agammaglobulinemia tyrosine kinase 205504_at 0.75 0.0124 Macrophage/microglia AIF1 Allograft inflammatory factor 1 215051_x_at 0.63 0.0494 APOB48R Apolipoprotein B48 receptor 220023_at 0.86 0.00139 CD36 CD36 Ag (collagen type I receptor, thrombospondin receptor) 206488_s_at 0.76 0.0112 COLEC12 Collectin subfamily member 12 221019_s_at 0.77 0.00924 CSF2RB CSF2 receptor, ␤, low-affinity (granulocyte-macrophage) 205159_at 0.73 0.0162 FMNL1 Formin-like 1 204789_at 0.67 0.0356 HLA-DMB MHC, class II, DM␤ 203932_at 0.66 0.0389 LILRA2 Leukocyte Ig-like receptor, subfamily a (with TM domain), 207857_at 0.76 0.0109 member 2 by guest on September 27, 2021 LILRB1 Leukocyte Ig-like receptor, subfamily b (with TM and ITIM 229937_x_at 0.74 0.0135 domains), member 1 LILRB2 Leukocyte Ig-like receptor, subfamily b (with TM and ITIM 210146_x_at 0.68 0.0301 domains), member 2 LILRB4 Leukocyte Ig-like receptor, subfamily B (with TM and ITIM 210152_at 0.82 0.00338 domains), member 4 LY86 Lymphocyte Ag 86 205859_at 0.87 0.00122 PRDM1 PR domain containing 1, with ZNF domain 228964_at 0.64 0.0457 SIRPA Protein tyrosine phosphatase, nonreceptor type substrate 1 202897_at 0.66 0.0362 SLC15A1 Solute carrier family 15 (oligopeptide transporter), member 1 211349_at 0.75 0.0134 Adaptive immune response T cell APBB1IP Amyloid ␤ (a4) precursor protein-binding, family b, member 1 230925_at 0.75 0.0127 interacting protein FYB FYN binding protein (FYB-120/130) 211795_s_at 0.72 0.0201 GPR65 G protein-coupled receptor 65 214467_at 0.74 0.0136 HCLS1 Hematopoietic cell-specific Lyn substrate 1 202957_at 0.64 0.0472 LAT2 Linker for activation of T cells family member 2 221581_s_at 0.83 0.00299 TAGAP T cell activation GTPase-activating protein 229723_at 0.67 0.0325 TRDV2 TCR␦ variable 2 210972_x_at 0.65 0.0428 CD4ϩ T cell HP Haptoglobin 206697_s_at 0.70 0.0229 HPR Haptoglobin-related protein 208470_s_at 0.69 0.0260 SPDEF SAM pointed domain containing ETS transcription factor 214403_x_at 0.67 0.0357 MOG Myelin oligodendrocyte glycoprotein 205989_s_at 0.71 0.0222 B cell CD48 CD48 Ag (B cell membrane protein) 204118_at 0.72 0.0190 IGHA2 Ig heavy constant ␣ 2 (A2m marker) 214916_x_at 0.70 0.0232 IGHG3 Ig heavy constant ␥ 3 (G3m marker) 211868_x_at 0.65 0.0419 IGHM Ig heavy locus 209374_s_at 0.64 0.0455 IGJ Ig J polypeptide, linker protein for Ig␣ and Ig␮ polypeptides 212592_at 0.65 0.0416 IGKC Ig␬ constant 211644_x_at 0.64 0.0460 IGKV1D-13 Ig␬ variable 1D-13 216207_x_at 0.85 0.00199 (Table continues) 7434 PROGNOSTIC IMMUNE FACTORS IN EPENDYMOMA

Table V. (Continued)

Affymetrix Gene Symbol Gene Name Probeset ID RpValue

IGLC2 Ig␭ constant 1 (Mcg marker) 216984_x_at 0.64 0.0447 IGLJ3 Ig␭ joining 3 211798_x_at 0.72 0.0189 IGLL3 Similar to omega protein 215946_x_at 0.79 0.00695 IGSF6 Ig superfamily, member 6 206420_at 0.66 0.0377 RALY RNA binding protein, autoantigenic (hnRNP-associated with lethal 224096_at 0.69 0.0275 yellow homolog (mouse)) Antibody FCGR1A Fc fragment of IgG, high-affinity Ia 216950_s_at 0.64 0.0469 FCGR1B Fc␥ receptor I B2 214511_x_at 0.64 0.0443 FCGR2C Fc fragment of IgG, low-affinity IIc, receptor for (CD32) 211395_x_at 0.67 0.0356 Adhesion molecules ITGAL Integrin, ␣ L (Ag CD11A (p180), lymphocyte function-associated 213475_s_at 0.69 0.0260 Ag 1; ␣ polypeptide) SELE Selectin E (endothelial adhesion molecule 1) 206211_at 0.65 0.0422 Cytokines, chemokines, and cytokine signaling CCL7 Chemokine (C-C motif) ligand 7 208075_s_at 0.65 0.0411 CCL11 Chemokine (C-C motif) ligand 11 210133_at 0.65 0.0413 CCR5 Chemokine (C-C motif) receptor 5 206991_s_at 0.71 0.0220 Downloaded from CX3CR1 Chemokine (C-X3-C motif) receptor 1 205898_at 0.70 0.0234 CYSLTR1 Cysteinyl leukotriene receptor 1 230866_at 0.78 0.00813

EDA2R Ectodysplasin A2 receptor 221399_at 0.65 0.0435 IL1B IL-1, ␤ 39402_at 0.67 0.0322 IL4I1 IL-4-induced 1 214935_at 0.69 0.0259 IRF8 IFN regulatory factor 8 204057_at 0.82 0.00351

KLK7 Kallikrein 7 (chymotryptic, stratum corneum) 239381_at 0.67 0.0331 http://www.jimmunol.org/ MLCK3 MLCK3 protein 1568925_at 0.64 0.0442 SAA2 Serum amyloid A2 208607_s_at 0.67 0.0352 TRADD TNFRSF1A-associated via death domain 205641_s_at 0.87 0.00110 Oxidative burst CYBASC3 Cytochrome b, ascorbate dependent 3 224735_at 0.67 0.0346 CYBB Cytochrome b-245, ␤ polypeptide 203923_s_at 0.69 0.0266 GZMA Granzyme A (granzyme 1, cytotoxic T lymphocyte-associated serine 205488_at 0.64 0.0441 esterase 3) HCK Hemopoietic cell kinase 208018_s_at 0.63 0.0491 TPSB2 Tryptase ␤ 2 207134_x_at 0.82 0.00387 Tethering and rolling of lymphocytes by guest on September 27, 2021 ABCA1 ATP-binding cassette, subfamily A (ABC1), member 1 203505_at 0.73 0.0168 ARHGAP4 Rho GTPase-activating protein 4 204425_at 0.75 0.0118 ARHGAP9 Rho GTPase-activating protein 9 224451_x_at 0.67 0.0325 CORO1A Coronin, actin-binding protein, 1A 209083_at 0.70 0.0239 DOCK2 Dedicator of cytokinesis protein 2 213160_at 0.73 0.0156 FPRL2 Formyl peptide receptor-like 2 230422_at 0.66 0.0376 SELPLG Selectin P ligand 209879_at 0.82 0.00383 ST3GAL1 ST3 ␤-galactoside ␣-2,3-sialyltransferase 1 244074_at 0.72 0.0182 Expressed in hematopoietic cells BCL11B B cell CLL/lymphoma 11B (zinc finger protein) 222895_s_at 0.84 0.00234 CD109 CD109 Ag (GOV platelet alloantigens) 226545_at 0.73 0.0163 GNA15 Guanine nucleotide binding protein (G protein), ␣ 15 205349_at 0.77 0.00868 GIMAP4 GTPase IMAP family member 4 219243_at 0.66 0.0364 GMFG Glia maturation factor ␥ 204220_at 0.64 0.0468 LCP1 Lymphocyte cytosolic protein 1 (L-plastin) 208885_at 0.80 0.0054 MYLC2PL Myosin light chain 2, lymphocyte-specific 221660_at 0.65 0.0405 NCKAP1L NCK-associated protein 1-like 209734_at 0.76 0.00996 NT5E 5Ј-nucleotidase, ecto (CD73) 203939_at 0.68 0.0311 PCSK5 Proprotein convertase subtilisin/kexin type 5 213652_at 0.70 0.0254 PSCD4 Pleckstrin homology, Sec7 and coiled-coil domains 4 219183_s_at 0.83 0.00294 SPTB Spectrin, ␤, erythrocytic (Includes spherocytosis, clinical type I) 214145_s_at 0.85 0.00203 Miscellaneous immune-related GVIN1 GTPase, very large IFN inducible 1 220577_at 0.65 0.0399 IFI27 Interferon, ␣ -inducible protein 27 202411_at 0.72 0.0194 INPP5D Inositol polyphosphate-5-phosphatase, 145 kDa 203332_s_at 0.73 0.0166 LENG9 Leukocyte receptor cluster (LRC) member 9 1554589_at 0.66 0.0385 LGALS9 Lectin, galactoside-binding, soluble, 9 (galectin 9) 203236_s_at 0.68 0.0316 MALL Mal, T cell differentiation protein-like 209373_at 0.79 0.00642 SEMA3G Sema domain, Ig domain (Ig), short basic domain, secreted, 219689_at 0.65 0.0403 (semaphorin) 3G SERPINB9 Serpin peptidase inhibitor, clade B (ovalbumin), member 9 209723_at 0.68 0.0293 SLA SRC-like adapter 203761_at 0.75 0.0133

a Genes positively correlated with a longer time to progression ( p Ͻ 0.05) are categorized into specific immune categories according to peer-reviewed publications. The level by which listed genes are correlated with TTP is measured as a continuous variable by two-sided Pearson correlation test (R) and estimated p value. The Journal of Immunology 7435 immune function by GO. As seen in the nonrecurrent pheno- To validate the association of AIF1 and HLA-DR expression type, genes whose expression positively correlated with TTP with outcome, the frequency of positively immunostaining cells in included a number of molecules critically involved in both in- the parenchyma of nonrecurrent and recurrent EPN was measured. nate and adaptive immune responses. Some overlap in innate This analysis revealed that AIF1 positive staining cells were sig- and adaptive immune-related genes was observed between the nificantly more abundant in nonrecurrent EPN (1.91-fold; p ϭ nonrecurrent phenotype and TTP-positive correlates, analyzed 0.0082) (Fig. 1C). HLA-DR was on average 2-fold more abundant in more detail below. As seen in the nonrecurrent phenotype, in nonrecurrent EPN but was not significant (2.18-fold; p ϭ 0.082) multiple components of the complement system (C2, C3, (Fig. 1D). These data recapitulate the results of gene expression C3AR1, C6, C7, CD53, CD59, CR1, ITGB2) and genes asso- analysis that demonstrated overexpression of AIF1 and HLA- ciated with microglia/macrophages (AIF1, CD36, HLA-DMB, DR5B in nonrecurrent EPN compared with recurrent EPN. LILRA2, LILRB1, LILRB2, LILRB4) and T cells (FYB, HCLS1, LAT2, TAGAP2, TRDV2) were identified in positive Tumor-infiltrating immune cells are present in EPN and correlates of TTP in recurrent EPN. The main difference that associated with a good outcome distinguished TTP-positive correlates from the nonrecurrent In addition to the microglia/macrophage-associated markers ana- phenotype was the presence of a significant number of genes lyzed above, T and B cell-related transcripts were found to be commonly expressed by B cells. These included multiple Ig associated with outcome, suggesting a variety of infiltrating im- genes (IGHA2, IGHG3, IGHM, IGJ, IGKC, IGKV1D-13, mune cell subtypes in EPN. IHC was used to identify CD4ϩ T IGLC2, IGLJ3, IGLL3, and IGSF6). cells, CD8ϩ T cells, CD45ϩ leukocytes, microphage/microglia ϩ (CD68 ), and B cells (CD20) in FFPE tissue in nonrecurrent (n ϭ Downloaded from Overlapping genes between the nonrecurrent EPN phenotype 9) and recurrent (n ϭ 10) EPN. Representative staining of these and positive correlates of time to progression are almost immune cell subpopulations is depicted in Fig. 2. Microglia/mac- ϩ entirely immune function-related rophages and CD45 leukocytes were more abundant than T cells or B cells across all EPN analyzed. A number of genes were identified that were associated with both Frequency analysis of infiltrating cells showed increased num- the nonrecurrent EPN phenotype and that were also positively cor- ϩ ϩ

bers of CD4 T cells (16-fold; p ϭ 0.045), CD8 T cells (1.92- http://www.jimmunol.org/ related with TTP, emphasizing their involvement in EPN clinical ϩ fold; p ϭ 0.34), CD45 leukocytes (1.55-fold; p ϭ 0.16), and outcome as a whole. Ontological analysis revealed that 95% (19 microglia/macrophages (3.06-fold; p ϭ 0.18) in nonrecurrent EPN out of 20) of genes associated with both nonrecurrence and longer ϩ compared with recurrent EPN, although only CD4 T cells TTP have roles in innate and adaptive immune functions, the de- reached statistical significance (Fig. 3). Greater numbers of B cells tails of which are outlined in Table VI. These genes are involved were observed in recurrent EPN, although this difference was not in complement activity (C3, C3AR1, and ITGB2), macrophage statistically significant (3.92-fold; p ϭ 0.12). activity (AIF1), phagocytosis of Ab-coated cells (FCGR1A, FCGR1B, LILRB1, CYBB), Ag presentation (HLA-DMB), lym- Polyomavirus SV40, BK, JC, and Merkel are not present in

phocyte tethering and rolling (SELPLG, CORO1A, DOCK2, nonrecurrent EPN by guest on September 27, 2021 APBB1IP, ARHGAP4), and lymphocyte activation (HCLS1, A number of genes associated with the nonrecurrent EPN pheno- LAT2, FYB, GPR65, HCK). Only phosphorylase kinase, ␥-1 type are known to be involved in the immune response to viral (PHKG1) had no documented evidence of immune involvement, infection, in particular IFN regulatory factor-7 (IRF7), tripartite with its known function being as a key glycogenolytic enzyme. motif-containing-22 (TRIM22), and apolipoprotein B mRNA ed- However, the dependence of T-lymphocyte activity on glucose iting enzyme, catalytic polypeptide-like 3G (APOBEC3G) (30– metabolism suggests a potential role in immune function for this 32). Earlier research found SV40-like PyV sequences in ϳ50% of gene (29). In the reverse analysis of genes that overlapped between EPN but did not attempt to correlate viral positivity with clinical both bad prognosis groups, that is, the recurrent EPN phenotype outcome (33). Based on the fact that the percentage of EPN found and TTP negative correlates, only two genes were identified: pro- to contain viral sequences in this earlier study matched the per- grammed cell death-6 (PDCD6) and opsin-3 (OPN3), which are centage of patients who did not suffer from recurrence, it was known to have roles in TCR-induced apoptosis and photorecep- hypothesized that nonrecurrent EPN samples might contain virus, tion, respectively. triggering an increase in viral immune response gene expression. Presence of virus has been shown to predict a favorable outcome Immune-related genes associated with a good outcome in EPN in other tumor types, such as head and neck , which sup- are expressed by infiltrating cells within the tumor ported this hypothesis (34). Nonrecurrent (n ϭ 8) and recurrent It was predicted that up-regulated immune-related genes identified (n ϭ 9) EPN specimens were therefore screened for the presence by ontological analyses were expressed by tumor-infiltrating im- of SV40, BK, JC, and Merkel PyV DNA sequences using quan- mune cells within patient tumor samples. To provide some evi- titative PCR. No PyV sequences were found in any of the tumor dence for this, histology was used to identify individual cells ex- specimens tested apart from three of nine recurrent EPN that pressing AIF1 and HLA-DR. These immune-related genes showed weak positivity for SV40. Thus, no association between associated with the nonrecurrent phenotype are known to be ex- the nonrecurrent EPN phenotype and the presence of PyV DNA pressed by microglia/macrophages (28). IHC of AIF1 and sequences was observed. Despite these data, the possibility cannot HLA-DR protein expression was performed in FFPE tissue of non- be ruled out that some virus other than those tested is present in recurrent (n ϭ 9) and recurrent (n ϭ 10) EPN. Protein expression nonrecurrent EPN. of AIF1 (Fig. 1A) and HLA-DR (Fig. 1B) was restricted to a sub- population of cells in the parenchyma of the tumor with a cellular Discussion morphology that resembled microglia/macrophages. These data in- This study provides early circumstantial evidence that in ϳ50% dicate that at least a subset of immune function gene transcripts of EPN patients there is a host antitumor immune response identified by microarray analyses are derived from tumor-infiltrat- and/or proinflammatory microenvironment that, when com- ing immune cells. bined with standard therapy, results in complete eradication of 7436 PROGNOSTIC IMMUNE FACTORS IN EPENDYMOMA

Table VI. Overlapping genes that are associated with both the non-recurrent and long PFS phenotypes in ependymoma showing function, cellular distribution and key reference(s) pertaining to each of the 20 genes

Affymetrix Gene Symbol Gene Name ID Function Cellular Distribution

AIF1 Allograft inflammatory factor 1 215051_x_at Activation marker involved with membrane Macrophages/microglia ruffling APBB1IP Amyloid ␤ (a4) precursor protein- 230925_at Facilitates T cell receptor-mediated integrin T cells binding, family b, member 1 activation interacting protein ARHGAP4 Rho GTPase-activating protein 4 204425_at Cell movement Predominantly expressed in hematopoietic cells C3 Complement component 3 217767_at Plays a central role in the activation of Widely expressed, including both the classical and alternative macrophages complement system C3AR1 Complement component 3a 209906_at Stimulates chemotaxis, granule enzyme Widely expressed in differentiated receptor 1 release, and superoxide anion production hematopoietic cells CORO1A Coronin, actin-binding protein, 1A 209083_at Accumulates at the leading edge of Lymphocytes migrating neutrophils and at the nascent phagosome; cytoskeletal modification via actin Arp2/3 complex. CYBB Cytochrome b-245, ␤ polypeptide 203923_s_at Primary component of the microbicidal Monocytes, macrophages (chronic granulomatous oxidase system of phagocytes disease) DOCK2 Dedicator of cytokinesis protein 2 213160_at Hematopoietic cell-specific protein that is Specific to leukocytes indispensable for lymphocyte Downloaded from chemotaxis FCGR1A Fc fragment of IgG, high-affinity 216950_s_at CD64. Involved in phagocytosis of Ab- Predominantly expressed on Ia coated cells; high-affinity receptor to the monocytes and macrophages. Fc region of ␥ Igs FCGR1B Fc-gamma receptor I B2 214511_x_at Alternative splice form of CD64 Predominantly expressed on monocytes and macrophages FYB FYN binding protein 211795_s_at ADAP; adapter protein of the FYN and Expressed in hematopoietic (FYB-120/130) LCP2 signaling cascades in T cells tissues such as myeloid and T

cells, spleen, and thymus; not http://www.jimmunol.org/ expressed in B cells, nor in nonlymphoid tissues GPR65 G protein-coupled receptor 65 214467_at May have a role in activation-induced In organs and cells involved in differentiation or cell death of T cells hematopoiesis HCK Hemopoietic cell kinase 208018_s_at Part of a signaling pathway coupling the Fc Expressed predominantly in cells receptor to the activation of the of the myeloid and B respiratory burst; may also contribute to lymphoid lineages neutrophil migration and may regulate the degranulation process of neutrophils HCLS1 Hematopoietic cell-specific Lyn 202957_at Role in TCR signaling; substrate of the Ag Only on cells of hematopoietic substrate 1 receptor-coupled tyrosine kinase; plays a origin role in Ag receptor signaling for both

clonal expansion and deletion in by guest on September 27, 2021 lymphoid cells; cytoskeletal modification via actin Arp2/3 complex HLA-DMB MHC, class II, DM␤ 203932_at Ag processing and cross-presentation to APCs: macrophages, dendritics, CD4ϩ T cells B cells ITGB2 Integrin, ␤ 2 (complement 202803_s_at CD18, part of LFA1 and CR3, receptors Leukocytes component 3 receptor 3 and 4 for ICAMs and C3 (component iC3b), subunit) respectively. known to participate in cell adhesion as well as cell surface- mediated signaling LAT2 Linker for activation of T cells 221581_s_at Involved FCGR1 (CD64)-mediated Highly expressed in spleen, family member 2 signaling in myeloid cells; couples peripheral blood lymphocytes, activation of immune receptors and their and germinal centers of lymph associated kinases with distal nodes intracellular events LILRB1 Leukocyte Ig-like receptor, 229937_x_at CD85j; monocyte/macrophage Ig receptor; Expressed primarily by subfamily b (with TM and binds PTPN6 when phosphorylated; monocytes, macrophages, and ITIM domains), member 1 binds FCER1A and FCGR1B dendritic cells PHKG1 Phosphorylase kinase, ␥ 1 207312_at Crucial glycogenolytic regulatory enzyme Predominantly in muscle and (muscle) liver SELPLG Selectin P ligand 209879_at Critical role in tethering and rolling of Myeloid and T cells neutrophils and T lymphocytes on inflamed endothelial cells remaining residual tumor cells. Additionally, these data provide present study. Prospective validation of immune-related factors a novel perspective to the clinical problem of how to identify as an up-front prognostic marker in EPN is clearly warranted up-front those children whose EPN will recur by identifying a based on the results of this study. functional role for genes associated with prognosis, rather than The results of this study provide preliminary evidence for simply listing genes as in previous studies (17, 19–21). Similar involvement of both the innate and adaptive arms of the im- to the results of this study, correlation of lymphoma microarray mune response in host control of EPN. The innate immune sys- profiles with outcome demonstrated that immune gene expres- tem uses a diversity of pathways to recognize and respond to sion was the predominant feature that predicted survival (35). Ags, including potential cancer-specific Ags. The complement The presence of tumor reactive T and B cells and tumor-infil- system is the major humoral component of the innate immune trating lymphocytes (TIL) in clinical specimens has been cor- system, and multiple complement system genes were found to related with an improved outcome in a number of tumor types be associated with a good outcome in EPN (C1QC, C2, C3, C6, (36, 37). The presence of TIL is a prognostic marker in these C7, C3AR1, CR1, CD53, CD59, ITGB2, MASP1, SERPING1). tumors and provides a precedent for the correlation of immune Complement-dependent cytotoxicity is thought to be one of the cell infiltration with good clinical outcome in EPN seen in the most important mechanisms of action of therapeutic mAbs The Journal of Immunology 7437

FIGURE 1. Immunohistochemical staining of (A) AIF-1 and (B) HLA-DR in FFPE tumor sections of nonrecurrent EPN with hematoxylin counterstaining (ϫ400). Relative abundancy of (C) AIF-1 and (D) HLA-DR positive infiltrating cells in nonrecurrent (non-rec; n ϭ 9) and re- current (rec; n ϭ 10) EPN. Cells were

scored using the mean of the number Downloaded from of positive staining cells in five fields of view and statistical significance by Student’s t test was defined as p Ͻ 0.05. Horizontal bars represent the mean average of scores. http://www.jimmunol.org/ by guest on September 27, 2021 against cancer (38). In animal studies of rituximab-mediated tiating molecule of the classical, Ab-dependent complement tumor control, the presence of C1Q was found to be critical for pathway, was associated with the nonrecurrent EPN phenotype, effective complement-dependent cytotoxicity. C1QC, a key ini- but not with a long TTP in recurrent EPN.

FIGURE 2. Representative infiltration of (A) CD4ϩ and (B) CD8ϩ T cells in nonrecurrent EPN. C, CD45ϩ leukocytes and (D) CD68ϩ mi- croglia were observed in greater numbers than T cells across all samples. Immunostaining with hematoxylin counterstain (ϫ400). 7438 PROGNOSTIC IMMUNE FACTORS IN EPENDYMOMA

FIGURE 3. Tumor-infiltrating im- mune cells in nonrecurrent (non-rec; n ϭ 9) and recurrent (rec; n ϭ 10) EPN. A, CD45ϩ leukocytes, (B) CD4ϩ T cells, (C) CD8ϩ T cells, (D) CD68ϩ microglia/macrophages, and ϩ (E) CD20 B cells were identified in Downloaded from paraffin sections of tumor specimens using immunohistochemistry. Cells were scored using the mean of the number of positive staining cells in five fields of view and statistical sig- nificance by t test was defined as p Ͻ 0.05. Horizontal bars represent the http://www.jimmunol.org/ mean average of scores. by guest on September 27, 2021

A number of genes specifically associated with activity of mi- phages promote tumor activity in the brain and elsewhere (43, 44). croglia/macrophages, the key cellular component of the innate im- Note that most studies of tumor-infiltrating microglia/macrophages mune system, were found to correlate with good outcome in EPN. in the CNS have been performed in glioblastoma, which has a These included AIF1 (28), multiple MHC class II alleles (HLA- highly immunosuppressive tumor microenviroment and a uni- DMA, HLA-DMB, HLA-DPB1, HLA-DRB5 and CD74), and leu- formly dismal outcome. Direct comparison of infiltrating micro- kocyte Ig-like receptor, subfamily b1 (LILRB1). IHC analysis of glia/macrophages in good outcome EPN and glioblastoma may AIF1 and HLA-DR demonstrated that these molecules are re- shed light on this disparity. stricted to tumor-infiltrating cells. Based on the morphology of The up-regulation of numerous adaptive immune response re- AIF1 and HLA-DR staining, as compared with macrophage/mi- lated genes was observed in good prognosis EPN. In previous stud- croglia staining in matched samples, it appears that AIF1 and ies of CNS microglia, innate immune system activation was char- HLA-DR are being expressed by tumor-infiltrating microglia/mac- acterized by up-regulation of type-1 IFN and MHC class II rophages as expected. The increased expression of AIF1 in non- expression, resulting in cross-presentation of viral epitopes to recurrent EPN vs recurrent EPN was demonstrated by both mi- CD4ϩ T cells (45). Consistent with this, in nonrecurrent EPN we croarray analysis (2.62-fold; p ϭ 0.0073) and IHC (1.8-fold; p ϭ observed overexpression of IFN-induced genes (e.g., IFIT1, 0.0082). Consistent with our data, MHC class II expression posi- IFIT3), multiple MHC class II genes, genes specifically associated tively correlates with a favorable outcome in a variety of non-CNS with T cell activation (e.g., TRAC, CD37, FYB, HAVCR2, tumors such as diffuse large B cell lymphoma and hepatocellular HCLS1), and increased frequency of tumor-infiltrating CD4ϩ T carcinoma (39, 40). cells. A number of other examples of specific adaptive immune The association of microglia/macrophage-specific transcripts response activities are implied by EPN outcome-associated tran- with improved outcome in EPN is contrary to a number of reports scripts. These include the observation that B cell-associated tran- of compromised microglia/macrophage activity, including reduced scripts are correlated with delayed recurrence, but are not found in MHC class II expression, in other CNS tumors (41, 42). Further- the nonrecurrent phenotype. Although preliminary, this result more, there is growing evidence that tumor-infiltrating macro- suggest that an Ab response affords some resistance to tumor The Journal of Immunology 7439 recurrence, but a T cell-specific response is required for complete 7. Pollack, I. F., P. C. Gerszten, A. J. Martinez, K. H. Lo, B. Shultz, A. L. Albright, tumor elimination. The presence of a number of T cell function- J. Janosky, and M. Deutsch. 1995. Intracranial ependymomas of childhood: long- term outcome and prognostic factors. Neurosurgery 37: 655–666, discussion related transcripts elaborate specific T cell functions in good out- 666–657. come EPN. For example, polarization of nonrecurrent EPN infil- 8. Robertson, P. L., P. M. Zeltzer, J. M. Boyett, L. B. Rorke, J. C. Allen, J. R. Geyer, trating T cells to the Th1 phenotype is implied by the presence of P. Stanley, H. Li, A. L. Albright, P. McGuire-Cullen, et al. 1998. Survival and prognostic factors following radiation therapy and chemotherapy for ependymo- HAVCR2 (TIM3) (46). Taken together, these data provide pre- mas in children: a report of the Children’s Cancer Group. J. Neurosurg. 88: liminary evidence that, beyond the simple presence of an immune 695–703. 9. Figarella-Branger, D., M. Civatte, C. Bouvier-Labit, J. Gouvernet, D. Gambarelli, infiltrate, the phenotype and function of that infiltrate may influ- J. C. Gentet, G. Lena, M. Choux, and J. F. Pellissier. 2000. Prognostic factors in ence clinical outcome in EPN. This conclusion is consistent with intracranial ependymomas in children. J. Neurosurg. 93: 605–613. the report by Galon et al. demonstrating that the type (specifically 10. Korshunov, A., A. Golanov, and V. Timirgaz. 2002. Immunohistochemical mark- Th1), density, and location of immune cells within human colo- ers for prognosis of ependymal neoplasms. J. Neurooncol. 58: 255–270. 11. Tihan, T., T. Zhou, E. Holmes, P. C. Burger, S. Ozuysal, and E. J. Rushing. 2008. rectal tumors predict clinical outcome better than current staging The prognostic value of histological grading of posterior fossa ependymomas in criteria (47). children: a Children’s Oncology Group study and a review of prognostic factors. Finally, the interactive role between standard surgery, radiother- Mod. Pathol. 21: 165–177. 12. Bennetto, L., N. Foreman, B. Harding, R. Hayward, J. Ironside, S. Love, and apy, and chemotherapy and the host immune system cannot be D. Ellison. 1998. Ki-67 immunolabelling index is a prognostic indicator in child- overstated. There is increasing evidence that antitumor immune hood posterior fossa ependymomas. Neuropathol. Appl. Neurobiol. 24: 434–440. responses may contribute to the control of cancer after conven- 13. Preusser, M., H. Heinzl, E. Gelpi, R. Hoftberger, I. Fischer, I. Pipp, I. Milenkovic, A. Wohrer, F. Popovici, S. Wolfsberger, and J. A. Hainfellner. tional chemotherapy, by modulating the equilibrium between the 2008. Ki67 index in intracranial ependymoma: a promising histopathological tumor and the immune system (48, 49). This theory may apply to candidate biomarker. Histopathology 53: 39–47. Downloaded from our findings, whereby in those EPN that harbor a host immune 14. Altura, R. A., R. S. Olshefski, Y. Jiang, and D. R. Boue. 2003. Nuclear expres- sion of survivin in paediatric ependymomas and choroid plexus tumours corre- response, surgery and radiation therapy may shift the balance of lates with morphologic tumour grade. Br. J. Cancer 89: 1743–1749. the equilibrium in favor of the host by critically increasing the 15. Preusser, M., S. Wolfsberger, T. Czech, I. Slavc, H. Budka, and J. A. Hainfellner. immune/tumor cell ratio. This would then result in elimination of 2005. Survivin expression in intracranial ependymomas and its correlation with tumor cell proliferation and patient outcome. Am. J. Clin. Pathol. 124: 543–549. remaining residual tumor by the immune system, resulting in a 16. Tabori, U., J. Ma, M. Carter, M. Zielenska, J. Rutka, E. Bouffet, U. Bartels, favorable outcome in the patient. In those patients who do not D. Malkin, and C. Hawkins. 2006. Human reverse transcriptase expres- http://www.jimmunol.org/ receive a complete tumor resection or radiation therapy, the equi- sion predicts progression and survival in pediatric intracranial ependymoma. J. Clin. Oncol. 24: 1522–1528. librium remains in favor of the tumor, resulting in the poor out- 17. Modena, P., E. Lualdi, F. Facchinetti, J. Veltman, J. F. Reid, S. Minardi, come that is observed in such scenarios. In those patients that I. Janssen, F. Giangaspero, M. Forni, G. Finocchiaro, et al. 2006. Identification of receive standard therapy but lack an antitumor immune response, tumor-specific molecular signatures in intracranial ependymoma and association with clinical characteristics. J. Clin. Oncol. 24: 5223–5233. residual tumor continues to grow unhindered despite receiving 18. Pezzolo, A., V. Capra, A. Raso, F. Morandi, F. Parodi, C. Gambini, P. Nozza, standard therapy, resulting in tumor recurrence. F. Giangaspero, A. Cama, V. Pistoia, and M. L. Garre. 2008. Identification of Despite the promising results in animal studies of CNS cancer novel chromosomal abnormalities and prognostic cytogenetics markers in intra- cranial pediatric ependymoma. Cancer Lett. 261: 235–243. immunotherapy, clinical trials using immunotherapy in humans 19. Sowar, K., J. Straessle, A. M. Donson, M. Handler, and N. K. Foreman. 2006. have had limited success (50–53). This failure suggests that Predicting which children are at risk for ependymoma relapse. J. Neurooncol. 78: by guest on September 27, 2021 knowledge of the antitumor immune response in the human CNS 41–46. 20. Lukashova-v Zangen, I., S. Kneitz, C. M. Monoranu, S. Rutkowski, B. Hinkes, cannot be extrapolated from animal models as previously assumed. G. H. Vince, B. Huang, and W. Roggendorf. 2007. Ependymoma gene expression A more rational approach to successfully implementing immuno- profiles associated with histological subtype, proliferation, and patient survival. therapy would be to design strategies based on data taken from Acta Neuropathol. 113: 325–337. 21. de Bont, J. M., J. M. Kros, M. M. Passier, R. E. Reddingius, P. A. Sillevis Smitt, direct clinical studies of human host anti-CNS tumor immune re- T. M. Luider, M. L. den Boer, and R. Pieters. 2008. Differential expression and sponses. This report potentially illustrates just such a response, prognostic significance of SOX genes in pediatric and ependy- underscoring the value and potential impact of these findings. moma identified by microarray analysis. Neurol. Oncol. 10: 648–660. 22. McLendon, R. E., O. D. Wiestler, J. M. Kros, A. Korshunov, and H.-K. Ng. 2007. Ependymoma. In Pathology and of Tumours of the .P. Acknowledgments Kleihues and W. K. Cavenee, eds. IARC Press, Lyon, France, pp. 74–80. We thank Patsy Ruegg at IHCtech for assistance with immunohistochem- 23. Wu, Z., R. A. Irizarry, R. Gentleman, F. Martinez-Murillo, and F. Spencer. 2004. istry and Liza Litzenberger for photographic expertise. A model-based background adjustment for oligonucleotide expression arrays. J. Am. Stat. Assoc. 99: 909–917. 24. Edgar, R., M. Domrachev, and A. E. Lash. 2002. Gene Expression Omnibus: Disclosures NCBI gene expression and hybridization array data repository. Nucleic Acids Res. The authors have no financial conflicts of interest. 30: 207–210. 25. Mootha, V. K., C. M. Lindgren, K. F. Eriksson, A. Subramanian, S. Sihag, References J. Lehar, P. Puigserver, E. Carlsson, M. Ridderstrale, E. Laurila, et al. 2003. PGC-1␣-responsive genes involved in oxidative phosphorylation are coordinately 1. Rickert, C. H., and W. Paulus. 2001. Epidemiology of downregulated in human diabetes. Nat. Genet. 34: 267–273. tumors in childhood and adolescence based on the new WHO classification. 26. Dennis, G., Jr., B. T. Sherman, D. A. Hosack, J. Yang, W. Gao, H. C. Lane, and Childs Nerv. Syst. 17: 503–511. R. A. Lempicki. 2003. DAVID: Database for Annotation, Visualization, and In- 2. Central Brain Tumor Registry of the United States (CBTRUS). 2005. Statistical tegrated Discovery. Genome Biol. 4: P3. report: prmary brain tumors in the United States, 1998–2002. Central Brain Tu- mor Registry of the United States. Available at www.cbtrus.org/reports//2005– 27. Ashburner, M., C. A. Ball, J. A. Blake, D. Botstein, H. Butler, J. M. Cherry, 2006/2006report.pdf A. P. Davis, K. Dolinski, S. S. Dwight, J. T. Eppig, et al. 2000. Gene ontology: 3. Foreman, N. K., S. Love, S. S. Gill, and H. B. Coakham. 1997. Second-look tool for the unification of biology: The Gene Ontology Consortium. Nat. Genet. surgery for incompletely resected fourth ventricle ependymomas: technical case 25: 25–29. report. Neurosurgery 40: 856–860, discussion 860. 28. Yamada, M., K. Ohsawa, Y. Imai, S. Kohsaka, and S. Kamitori. 2006. X-ray 4. Ridley, L., R. Rahman, M. A. Brundler, D. Ellison, J. Lowe, K. Robson, structures of the microglia/macrophage-specific protein Iba1 from human and E. Prebble, I. Luckett, R. J. Gilbertson, S. Parkes, et al. 2008. Multifactorial mouse demonstrate novel molecular conformation change induced by calcium analysis of predictors of outcome in pediatric intracranial ependymoma. Neurol. binding. J. Mol. Biol. 364: 449–457. Oncol. 10: 675–689. 29. Cham, C. M., G. Driessens, J. P. O’Keefe, and T. F. Gajewski. 2008. Glucose 5. Rousseau, P., J. L. Habrand, D. Sarrazin, C. Kalifa, M. J. Terrier-Lacombe, deprivation inhibits multiple key gene expression events and effector functions in ϩ C. Rekacewicz, and A. Rey. 1994. Treatment of intracranial ependymomas of CD8 T cells. Eur. J. Immunol. 38: 2438–2450. children: review of a 15-year experience. Int. J. Radiat. Oncol. Biol. Phys. 28: 30. Song, Y. J., K. M. Izumi, N. P. Shinners, B. E. Gewurz, and E. Kieff. 2008. IRF7 381–386. activation by Epstein-Barr virus latent membrane protein 1 requires localization 6. Agaoglu, F. Y., I. Ayan, Y. Dizdar, R. Kebudi, O. Gorgun, and E. Darendeliler. at activation sites and TRAF6, but not TRAF2 or TRAF3. Proc. Natl. Acad. Sci. 2005. Ependymal tumors in childhood. Pediatr. Blood Cancer 45: 298–303. USA 105: 18448–18453. 7440 PROGNOSTIC IMMUNE FACTORS IN EPENDYMOMA

31. Tissot, C., and N. Mechti. 1995. Molecular cloning of a new interferon-induced 43. Markovic, D. S., R. Glass, M. Synowitz, N. Rooijen, and H. Kettenmann. 2005. factor that represses human immunodeficiency virus type 1 long terminal repeat Microglia stimulate the invasiveness of glioma cells by increasing the activity of expression. J. Biol. Chem. 270: 14891–14898. metalloprotease-2. J. Neuropathol. Exp. Neurol. 64: 754–762. 32. Sheehy, A. M., N. C. Gaddis, J. D. Choi, and M. H. Malim. 2002. Isolation of a 44. Biswas, S. K., L. Gangi, S. Paul, T. Schioppa, A. Saccani, M. Sironi, B. Bottazzi, human gene that inhibits HIV-1 infection and is suppressed by the viral Vif A. Doni, B. Vincenzo, F. Pasqualini, et al. 2006. A distinct and unique transcrip- protein. Nature 418: 646–650. tional program expressed by tumor-associated macrophages (defective NF-␬B 33. Bergsagel, D. J., M. J. Finegold, J. S. Butel, W. J. Kupsky, and R. L. Garcea. and enhanced IRF-3/STAT1 activation). Blood 107: 2112–2122. 1992. DNA sequences similar to those of simian virus 40 in ependymomas and 45. Olson, J. K., A. M. Girvin, and S. D. Miller. 2001. Direct activation of innate and choroid plexus tumors of childhood. N. Engl. J. Med. 326: 988–993. antigen-presenting functions of microglia following infection with Theiler’s vi- 34. Lindel, K., K. T. Beer, J. Laissue, R. H. Greiner, and D. M. Aebersold. 2001. rus. J. Virol. 75: 9780–9789. Human papillomavirus positive squamous cell carcinoma of the oropharynx: a 46. Monney, L., C. A. Sabatos, J. L. Gaglia, A. Ryu, H. Waldner, T. Chernova, radiosensitive subgroup of head and neck carcinoma. Cancer 92: 805–813. S. Manning, E. A. Greenfield, A. J. Coyle, R. A. Sobel, et al. 2002. Th1-specific 35. Dave, S. S., G. Wright, B. Tan, A. Rosenwald, R. D. Gascoyne, W. C. Chan, cell surface protein Tim-3 regulates macrophage activation and severity of an R. I. Fisher, R. M. Braziel, L. M. Rimsza, T. M. Grogan, et al. 2004. Prediction autoimmune disease. Nature 415: 536–541. of survival in follicular lymphoma based on molecular features of tumor-infil- 47. Galon, J., A. Costes, F. Sanchez-Cabo, A. Kirilovsky, B. Mlecnik, trating immune cells. N. Engl. J. Med. 351: 2159–2169. C. Lagorce-Pages, M. Tosolini, M. Camus, A. Berger, P. Wind, et al. 2006. Type, 36. Clemente, C. G., M. C. Mihm, Jr., R. Bufalino, S. Zurrida, P. Collini, and density, and location of immune cells within human colorectal tumors predict N. Cascinelli. 1996. Prognostic value of tumor infiltrating lymphocytes in the clinical outcome. Science 313: 1960–1964. vertical growth phase of primary cutaneous melanoma. Cancer 77: 1303–1310. 48. Danna, E. A., P. Sinha, M. Gilbert, V. K. Clements, B. A. Pulaski, and 37. Cho, Y., M. Miyamoto, K. Kato, A. Fukunaga, T. Shichinohe, Y. Kawarada, ϩ ϩ S. Ostrand-Rosenberg. 2004. Surgical removal of primary tumor reverses tumor- Y. Hida, T. Oshikiri, T. Kurokawa, M. Suzuoki, et al. 2003. CD4 and CD8 T induced immunosuppression despite the presence of metastatic disease. Cancer cells cooperate to improve prognosis of patients with esophageal squamous cell Res. 64: 2205–2211. carcinoma. Cancer Res. 63: 1555–1559. 38. Glennie, M. J., R. R. French, M. S. Cragg, and R. P. Taylor. 2007. Mechanisms 49. Zitvogel, L., L. Apetoh, F. Ghiringhelli, F. Andre, A. Tesniere, and G. Kroemer. of killing by anti-CD20 monoclonal antibodies. Mol. Immunol. 44: 3823–3837. 2008. The anticancer immune response: indispensable for therapeutic success? 39. Rimsza, L. M., R. A. Roberts, T. P. Miller, J. M. Unger, M. LeBlanc, J. Clin. Invest. 118: 1991–2001. Downloaded from R. M. Braziel, D. D. Weisenberger, W. C. Chan, H. K. Muller-Hermelink, 50. Farkkila, M., J. Jaaskelainen, M. Kallio, G. Blomstedt, R. Raininko, E. S. Jaffe, et al. 2004. Loss of MHC class II gene and protein expression in P. Virkkunen, A. Paetau, H. Sarelin, and M. Mantyla. 1994. Randomised, con- ␥ diffuse large B-cell lymphoma is related to decreased tumor immunosurveillance trolled study of intratumoral recombinant -interferon treatment in newly diag- and poor patient survival regardless of other prognostic factors: a follow-up study nosed glioblastoma. Br. J. Cancer 70: 138–141. from the Leukemia and Lymphoma Molecular Profiling Project. Blood 103: 51. Liau, L. M., R. M. Prins, S. M. Kiertscher, S. K. Odesa, T. J. Kremen, 4251–4258. A. J. Giovannone, J. W. Lin, D. J. Chute, P. S. Mischel, T. F. Cloughesy, and 40. Matoba, K., N. Iizuka, T. Gondo, T. Ishihara, H. Yamada-Okabe, T. Tamesa, M. D. Roth. 2005. Dendritic cell vaccination in glioblastoma patients induces systemic and intracranial T-cell responses modulated by the local central nervous N. Takemoto, K. Hashimoto, K. Sakamoto, T. Miyamoto, et al. 2005. Tumor http://www.jimmunol.org/ HLA-DR expression linked to early intrahepatic recurrence of hepatocellular car- system tumor microenvironment. Clin. Cancer Res. 11: 5515–5525. cinoma. Int. J. Cancer. 115: 231–240. 52. Plautz, G. E., G. H. Barnett, D. W. Miller, B. H. Cohen, R. A. Prayson, 41. Flugel, A., M. S. Labeur, E. M. Grasbon-Frodl, G. W. Kreutzberg, and J. C. Krauss, M. Luciano, D. B. Kangisser, and S. Shu. 1998. Systemic T cell M. B. Graeber. 1999. Microglia only weakly present glioma antigen to cytotoxic adoptive immunotherapy of malignant gliomas. J. Neurosurg. 89: 42–51. T cells. Int. J. Dev. Neurosci. 17: 547–556. 53. Riva, P., G. Franceschi, A. Arista, M. Frattarelli, N. Riva, A. M. Cremonini, 42. Hussain, S. F., D. Yang, D. Suki, E. Grimm, and A. B. Heimberger. 2006. Innate G. Giuliani, and M. Casi. 1997. Local application of radiolabeled monoclonal immune functions of microglia isolated from human glioma patients. J. Transl. antibodies in the treatment of high grade malignant gliomas: a six-year clinical Med. 4: 15. experience. Cancer 80: 2733–2742. by guest on September 27, 2021