A Pan-Cancer Comparative Analysis of The Cancer Genome Atlas Transcriptomic TIL-Immune Signatures

A Pan-Cancer Comparative Analysis of The Cancer Genome Atlas Transcriptomic TIL-Immune Signatures | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A Pan-Cancer Comparative Analysis of The Cancer Genome Atlas Transcriptomic TIL-Immune Signatures Kyle Hitscherich, Darryl Noussome, Aaron Dinerman, Victoria Dulemba, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6441170/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 07 Aug, 2025 Read the published version in Cancer Immunology, Immunotherapy → Version 1 posted 11 You are reading this latest preprint version Abstract Efforts to understand the tumor microenvironment (TME) through basic science research and The Cancer Genome Atlas (TCGA) data analysis have led to the creation of unique immune transcriptomic signatures from tumor-infiltrating lymphocytes (TIL). However, no pan-cancer analysis has been conducted to compare the prognostic performance of these signatures using overall survival (OS) or progression-free interval (PFI) as endpoints. We compiled a library of 146 TIL-immune signatures and evaluated gene signature score correlation with OS and PFI for 9,961 available TCGA samples across 33 cancer types. Zhang CD8 TCS demonstrated higher accuracy in prognosticating both OS and PFI across the pan-cancer landscape, however, variability was seen across cancer types and germ cell origin. Cluster analysis compiled a group of six signatures (Oh.Cd8.MAIT, Grog.8KLRB1, Oh.TIL_CD4.GZMK, Grog.CD4.TCF7, Oh.CD8.RPL, Grog.CD4.RPL32) whose association with OS and PFI could potentially be conserved across multiple cancer types. Cancer Immunotherapy Single Cell Sequencing Tumor Microenvironment Tumor Infiltrating Lymphocytes Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction The Cancer Genome Atlas (TCGA) program curated multi-omic data, clinical characteristics, and outcomes of over 10,000 primary cancers from 33 cancer types over the past 17 years. 1 The goal of the project evolved from understanding the genetics of a select cancer histology, (glioblastoma multiforme) to now using combined tumor and microenvironment transcriptomics to better understand cancer biology as it relates to outcomes. 1 Immunotherapy has evolved over the past two decades as a promising arm of cancer therapy that can provide robust and durable responses for a wide variety of cancer types. 2 As success with checkpoint inhibitors (anti-PD-1, -PD-L1, -CTLA-4) and adoptive cell transfer (ACT, including chimeric antigen receptor (CAR) and tumor-infiltrating lymphocyte (TIL)) become more prevalent, the interest in the regulation of immune system in cancer microenvironment has grown. 3 , 4 , 5 , 6 , 7 Multiple immune gene signatures have been developed from transcriptomic data looking to better describe TIL populations and how they may impact the tumor microenvironment (TME). 8 , 9 , 10 , 11 Several of these in-depth evaluations of the tumor-immune microenvironment have sought to identify populations of T-cells that may be associated with disease progression through cytotoxic function, neoantigen recognition, or more “stem-like” phenotypic state in specific cancers. 7 In querying the TCGA, numerous research teams have sought to incorporate patient outcomes data into the development of such predictive gene signatures. 12 , 13 , 14 , 15 , 16 17 In many instances, these signatures include similar genes compared to those constructed from direct primary patient source data, however, coupling outcomes data from TCGA database has allowed for the investigation of novel genes, some specific to rare cancers and their subtypes. 18 , 19 Although there were over 150 TIL-TIL-immune signatures published, no study has compared the prognostic performance of these TIL-immune signatures to identify the top-ranked predictive TIL-immune signatures across all cancer, individual cancer types, and germ-cell origins. Previous work has demonstrated utility in such signatures in predicting response rates to immunotherapy such as checkpoint inhibitors. 13 , 14 Further understanding of the ideal immune cell population found within TIL could expand such therapeutic tools and aid clinicians in selecting patients who may benefit from such therapies. Methods Gene signature library construction and sample accrual A library of Tumor Infiltrating Lymphocyte (TIL) immune transcriptomic signatures was generated by PubMed literature review, specifically focusing on recent publications analyzing RNA-sequencing data on TIL derived from patients with metastatic cancers without restriction placed on histology. Emphasis was placed on publications identifying one or multiple T cell populations characterized by defined TIL-TIL-immune signatures (eg. “stem-like”, “terminally differentiated”, “effector memory”, “tissue resident memory” etc.). This library was broadened by searching “TCGA”, “transcriptomic” and “signature” to accumulate publications compiled from RNA-sequencing signatures specifically derived from TCGA database review, compared to those previously identified from direct, patient-sourced studies of immune transcriptomics and phenotypes. It should be noted that although some of these signatures were constructed for prognostic purposes, many were developed as a descriptive effort to define TIL populations within the TME. These TIL-TIL-immune signatures were then individually queried for all upregulated genes within their composition. Signature gene lists were reviewed to ensure consistent nomenclature across publications. Duplicated signatures were excluded from the final analysis. The TCGA recount3 project is an online data source containing the accumulated RNA-sequencing data contained within the TCGA database and across 8,679 studies of human samples. 20 RNA-sequencing data were downloaded for 33 cancer types and 9,961 samples. The recount3 project processed all RNA-seq samples via the Monorail system and provided gene-level counts using Gencode v26 (Table 1 ). The GSVA R/ Bioconductor package was used to calculate individual level gene set enrichment scores for each sample. Overall survival (OS) and progression-free interval (PFI) were chosen as primary endpoints similar to the previous publication by Liu et al. 1 Based on the nature of TCGA database comprised of non-metastatic primary tumor lesions, we determined PFI as a potentially insightful metric for our study. Table 1 TCGA tumor types, Nomenclature, and Available Samples for Analysis Ectoderm/Neural Crest n = 2646 (26.6%) Head and Neck Squamous Cell Carcinoma (HNSC) 522 (5.3%) Breast invasive Carcinoma (BRCA) 1,093 (11%) Pheochromocytoma and Paraganglioma (PCPG) 179 (1.8%) Skin Cutaneous melanoma (SKCM) 103 (1.0%) Uveal Melanoma (UVM) 80 (0.8%) Brain Lower Grade Glioma (LGG) 514 (5.2%) Glioblastoma Multiforme (GBM) 155 (1.6%) Mesoderm n = 3457 (34.7%) Mesothelioma (MESO) 87 (0.9%) Sarcoma (SARC) 259 (2.7%) Acute Myeloid Leukemia (LAML) 126 (1.3%) Adrenocortical Carcinoma (ACC) 79 (0.8%) Cervical Squamous Cell Carcinoma (CESC) 304 (3.2%) Kidney Chromophobe (KICH) 66 (0.7%) Kidney Renal Clear Cell Carcinoma (KIRC) 531 (5.3%) Kidney Renal Papillary Cell Carcinoma (KIRP) 290 (3.1%) Uterine Corpus Endometrial Carcinoma (UCEC) 541 (5.4%) Uterine Carcinosarcoma (UCS) 57 (0.6%) Testicular Germ Cell Tumor (TGCT) 150 (1.5%) Prostate Adenocarcinoma (PRAD) 497 (5.0%) Lymphoid Neoplasm Diffuse Large B-Cell Lymphoma (DLBC) 48 (0.5%) Ovarian Serous Cystadenocarcinoma (OV) 422 (4.2%) Endoderm n = 3858 (38.7%) Thymoma (THYM) 120 (1.2%) Bladder Urothelial Carcinoma (BLCA) 408 (74.5%) Cholangiocarcinoma (CHOL) 36 (0.4%) Colon Adenocarcinoma (COAD) 458 (4.6%) Esophageal Carcinoma (ESCA) 184 (1.9%) Liver Hepatocellular Carcinoma (LIHC) 371 (3.7%) Lung Adenocarcinoma (LUAD) 516 (5.2%) Lung Squamous Cell Carcinoma (LUSC) 501 (5.0%) Thyroid Carcinoma (THCA) 505 (5.1%) Stomach Adenocarcinoma (STAD) 415 (4.2%) Rectum Adenocarcinoma (READ) 166 (1.7%) Pancreatic Adenocarcinoma (PAAD) 178 (1.8%) Gene signature analysis and construction of a novel signature OS and PFI coefficients were calculated based on the applicability of each TIL-TIL-immune signature for individual samples based on grouped populations. Our analysis was conducted across all cancer specimens, distinction by tissue germ cell origins, individual cancer type, and descriptive immune clusters. 21 The association of each TIL-immune signature score and OS and PFI were compared. Analysis was performed for the above-mentioned populations to assess the performance of signatures and broad applicability. Conserved genes found across multiple high-performing TIL-TIL-immune signatures were compiled. The top frequently conserved genes were used to construct a novel signature (Novel_Sig) and compared against the performance of those originally identified within our constructed library. Cluster analysis and concordance evaluation Cluster analysis was performed on all signatures based on the genetic composition of each signature. The organization of clusters was determined to be 10 based on differentiation of positively and negatively related OS and PFI outcomes based on cluster association. Prognostication capability was evaluated for each cluster based on randomly sampled specimens via TCGA data. The remaining 10% of samples not included in prognostication were then used for Kaplan-Meyer analysis across constructed clusters. Results Construction of a gene signature library We identified 153 immune transcriptomic signatures from the literature review. Three were not included as they were not specific to T cells, and four were excluded as they only contained 1 gene. We thus examined 146 signatures across 45 publications (Fig. 1 A). One hundred and twenty (120) signatures were described in the setting of basic science research and derived from single-cell TIL sequencing sourced directly from patient samples. Twenty-six (26) were developed from the review of available TCGA database and/or in combination with other databases such as the Gene Expression Omnibus (GEO) database. These signatures comprised 3088 unique genes with nearly half (1432) shared across multiple signatures (Supplemental Table 1). The average number of sourced samples for the development of a molecular signature was 132.7 with a median of 10 patient samples used. Most sourcing was restricted to a single cancer type (77%, 112/146) with the dominant cancer types including non-small cell lung cancer (NSCLC, 23%, 32/146), bladder cancer (20%, 29/146), and melanoma (20%, 29/146). Examining the content of each signature, 1432 of the 3088 genes (46.37%) were shared between at least 2 signatures (Fig. 1 B). Of these genes shared among multiple signatures, they were found in an average of 4.97 signatures with 65.5% shared across more than 3 signatures. More than 3 quarters (76.71%) of signatures contained 75–100% of genes shared amongst at least 2 signatures from our library (Fig. 1 C). Only 2.74% of compiled signatures were comprised of entirely unique genes (Fig. 1 C). The most overlapped genes across signatures were ENTPD1 (n = 34), PDCD1 (n = 32), and HAVCR2 (n = 32) ( Fig. 1 D ) . The median number of genes per signature was 50 with a range of 2 to 1114 genes (Supplemental table 1 ). From querying TCGA database we accumulated bulk RNA transcriptomic data from 9,961 patient samples across 33 tumor types (Table 1 ) (Supplemental Table 2). Examining performance across pan-cancer To examine performance across pan-cancer we required grouping of samples to undergo gene signature score analysis due to the large number of available primary tumor samples. We opted to cluster samples by similarities in gene expression to forego any other grouping metric in an unbiased fashion. In doing so, we were able to evaluate gene signature scores as they related to OS and PFI coefficients of similar transcriptomic cancer samples (Supplemental Fig. 1A). Across our 146 gene signatures, the Zhang CD8 T-Cell associated gene signature for prognosis risk in lung adenocarcinoma (Zhang CD8 TCS) appeared to have the lowest OS and PFI coefficients, consistent with the strongest association with longer OS and PFI (Fig. 2 A, Supplemental Table 3). 16 Alternatively, the Liu hypoxia-associated gene score in bladder cancer (Liu_Hypoxia) appeared to have the highest OS and PFI coefficients, consistent with an association with shorter OS and PFI (Fig. 2 A). 1 Within our pan-cancer analysis, patient OS correlated with the American Joint Committee on Cancer (AJCC) staging, as expected (Fig. 2 B). Scoring samples based on Zhang CD8 TCS gene expression, patient samples were separated into quartiles based on concordance with molecular signature expression. When comparing OS from Q2 to Q1 we found a hazards ratio of 0.74 (P = 9.95e- 7 ) consistent with significantly longer OS with higher concordance of Zhang CD8 TCS gene signature. Similar significance was found comparing Q3 (HR = 0.67, P = 2.39e- 9 ) and Q4 (HR = 0.68, P = 1.72e- 4 ) (Fig. 2 C). Thus the higher correlation quartiles (Q2-Q4) were associated with prolonged OS and PFI compared to the lowest sample quartile (Q1) (Fig. 2 C and 2 D). Examining performance across germ cell origin An alternative grouping of samples was performed by germ-cell origin rather than overall transcriptomic similarities. This offered additional insight into the performance of our 146 signatures (Fig. 3 A-C). Although many signatures did not demonstrate a statistically significant correlation between OS or PFI coefficients and signature score, several signatures began to show a direct or inverse association with OS and/or PFI coefficients (Supplemental Table 4). When examining signature score by germ cell origin, the Oh CD8 central memory signature (Oh.CD8.CM) and Zhang.CD8.TCS had the highest correlations with OS and PFI coefficients for ectoderm and mesoderm-derived cancers, respectively (Fig. 3 A and 3 C). 9 , 16 , 22 In endoderm-derived malignancies, the top signatures correlating with OS and PFI coefficient were the Caushi Stem-like memory T-cell signature (Caushi.Stem-Like memory) and Oh CD4 central memory signature (Oh.CD4.CM) respectively (Fig. 3 B). 9 , 22 Thus, patients whose primary tumors contained more T-cells corresponding with these gene signatures tended towards longer OS and PFI following resection (Table 2 ). On closer examination, we identified 21.5% of composite genes within these signatures (n = 35/163) were shared across multiple signatures (Supplemental Table 5). The most conserved genes across these signatures included GPR183 , CCR7 , SELL , ARID5A , and others (Fig. 2 D). Table 2 Top TIL-immune signatures that correlated with overall survival (OS) and progression-free interval (PFI) by germ cell origin Germ Cell Origin Signature Title OS Coefficient OS Coefficient P Value PFI Coefficient PFI Coefficient P value Neural Crest Shi_TCS -1.29137 0.001457 -1.66787 9.21E-6 Mesoderm Yost_CD8_Memory -0.66138 0.000233 -0.75684 5.8E-6 Ectoderm Oh.CD8.CM -0.61975 0.012588 -0.47997 0.029186 Endoderm Zhang CD8 TCS -0.60349 1.76E-10 -0.66531 1.61E-13 Mesoderm Oh.CD4.CXCL13 -0.54667 0.017151 -0.98599 5.72E-6 Neural Crest Oh.CD8.PRO 0.81151 0.005707 1.13123 6.13E-5 We conversely identified that several signatures had an inverse correlation with OS and PFI coefficients. The signature titled Liu_Hypoxia had the poorest prognostication for Ectoderm and Endoderm derived malignancies (Figs. 3 A and 3 B). 23 The Oliveira proliferating T-cell gene signature (Oliveira_Prolif_T) and Caushi proliferating CD8 T-cell signature (Caushi.CD8.Proliferating) correlated with shorter PFI and OS respectively within the mesoderm-derived malignancies (Fig. 3 C). 9 , 10 Of these TIL-immune signatures, 10.6% (n = 18/170) of their genes were shared amongst two of these signatures, however, no gene was shared across all 3 signatures associated with poor OS and PFI. Examining performance across cancer types By distinguishing individual cancer types, rather than germ cell origin, we saw greater variability across OS and PFI coefficients as correlates to gene signature score (Supplemental Table 6). We determined that across our 33 cancer types, 22 had statistically significant variability across the signature score to correlate with OS coefficient and 28 regarding PFI coefficient (Tables 3 and 4 ). Table 3 Top TIL-immune signatures that correlated with overall survival (OS) by tumor type Tumor Histology Signature Title OS Coef OS Pvalue PF Coef PF Pvalue PCPG Zhang CD8 TCS -6.24199 0.022149674 -3.76982 0.002143 KICH Oh.CD8.HSP -4.30571 0.008573393 -2.44199 0.042444 ACC Zhang CD8 TCS -3.34056 4.82561E-05 -2.38948 0.000287 UVM Shi_TCS -2.05644 0.003787896 -2.57423 7.04E-05 HNSC Tang_Ferroptosis -2.03379 3.02851E-10 -1.5004 6.04E-06 KIRC Zhang CD8 TCS -1.96574 4.80286E-10 -1.63795 3.9E-07 SKCM Guo_Supp Treg -1.96074 0.01343793 -1.98457 0.004273 CESC Tang_Ferroptosis -1.57918 0.003553691 -1.22583 0.024684 UCEC Oh.CD4.CM -1.40552 0.000572013 -1.27701 0.000207 SARC Tang_Ferroptosis -1.39883 0.00220507 -0.73848 0.047949 BLCA Tang_Ferroptosis -1.24705 0.000695187 -1.31438 0.000466 LIHC Caushi.CD8.Stem-like memory -1.0595 0.001002037 -0.91496 0.000695 LUAD Zhang CD8 TCS -1.01043 7.38238E-05 -0.81351 0.000407 BRCA B16_PROG.EX_Miller -0.91433 0.000962026 -0.84596 0.001901 LGG Oh.CD8.MAIT -0.81017 0.009074414 -0.66166 0.00686 STAD Liu_Treg -0.73121 0.000587008 -0.73239 0.001334 COAD Krishna.ACT.CD8.Term.Diff -0.5895 0.020006639 -0.55552 0.012663 KIRP Oh.CD4.MITO -0.55617 0.04930259 -0.7407 0.003681 GBM Caushi.CD4-Th(3) 0.502069 0.046295118 0.735503 0.00316 PAAD Wang_Cytokine Exp 0.503322 0.014469427 0.556764 0.003478 MESO Exhaust_1_Feldman 1.38126 5.86199E-09 0.933397 0.000167 THYM Chatani_TP 1.629962 0.042234695 1.211128 0.021425 Table 4 Top Progression Free Interval (PFI) Correlation for Each Histology Tumor Histology Signature Title OS Coef OS Pvalue PF Coef PF Pvalue ACC Jansen_Stem like -2.89192 0.000666 -3.30626 6.13E-06 BLCA Tang_Ferroptosis -1.24705 0.000695 -1.31438 0.000466 BRCA Oh.CD8.CM -0.71512 0.040587 -0.99211 0.00404 CESC Krishna.ACT.CD8.Stem.Like -0.9932 0.012153 -1.50782 0.000245 CHOL LCMV_PROG.EX_Miller -1.82733 0.127838 -4.2128 0.002698 COAD TOX_Scott -0.66957 0.051828 -0.63124 0.038797 DLBC Oliveira_TM -1.13396 0.506353 -3.18516 0.029162 GBM Li_Pyroptosis 0.083296 0.649199 0.445846 0.019261 HNSC Tang_Ferroptosis -2.03379 3.03E-10 -1.5004 6.04E-06 KICH Tang_Ferroptosis -4.19384 0.021295 -5.02205 0.004432 KIRC Zhang CD8 TCS -1.96574 4.8E-10 -1.63795 3.9E-07 KIRP Zhang CD8 TCS -1.07239 0.087701 -1.65042 0.003526 LGG Oh.CD8.CM -0.68851 0.038533 -0.72399 0.008386 LIHC Caushi.Stem-like memory -1.03655 0.001353 -0.9826 0.000267 LUAD Zhang CD8 TCS -1.01043 7.38E-05 -0.81351 0.000407 LUSC Dai_Activation CD4 CD8 -0.10698 0.520351 -0.42427 0.03222 MESO Exhaust_1_Feldman 1.38126 5.86E-09 0.933397 0.000167 OV Wang_OS Benefit -0.12112 0.57786 -0.41078 0.042113 PAAD Wang_Cytokine Exp 0.503322 0.014469 0.556764 0.003478 PCPG Zhang CD8 TCS -6.24199 0.02215 -3.76982 0.002143 PRAD Zhang CD8 TCS -0.17488 0.892955 -1.18777 0.003254 READ Ahuluwalia_Prognostic Cell Death 0.165135 0.840801 -1.43142 0.041254 SARC Tang_Ferroptosis -1.39883 0.002205 -0.73848 0.047949 SKCM Oliveira_TM -0.96642 0.240344 -2.22317 0.003376 STAD Oliveira_Tumor Spec_Prolif T -0.41472 0.081555 -0.77828 0.002654 THCA Exhaust_1_Feldman -0.11051 0.823879 1.022291 0.001125 THYM Grog.Treg.1 0.805856 0.293964 1.044679 0.037388 UCEC Oh.CD4.CM -1.40552 0.000572 -1.27701 0.000207 UCS Dai_Activation CD4 CD8 -0.564 0.147362 -0.84535 0.035912 UVM Shi_TCS -2.05644 0.003788 -2.57423 7.04E-05 Examining the correlation between immune gene signature and OS showed variability in performance across different cancer types. Across the 22 cancers, 16 unique signatures were found to have the highest correlation to OS coefficient. Despite this variability, two signatures were found to correlate with OS coefficient across multiple cancer types, the Tang ferroptosis related gene signature in head and neck squamous cell carcinoma (Tang_Ferroptosis) and Zhang.CD8.TCS, each seen across 4 cancer types (BLCA, CESC, HNSC, SARC and ACC, KIRC, LUAD, PCPG) (Fig. 4 A, Table 3 ). 15 , 16 Regarding the PFI coefficient, 17 signatures were found to be correlated with longer PFI across our 28 cancer types. Similar to OS, Tang_Ferroptosis and Zhang.CD8.TCS signatures shared high correlation with PFI. The Zhang.CD8.TCS signature was the best correlate to PFI within five cancer types (KIRC, KIRP, LUAD, PCPG, PRAD) and Tang_Ferroptosis in four (BLCA, HNSC, KICH, SARC) (Fig. 4 B, Table 4 ). 15 , 16 The number of shared genes across top correlates for cancer type was 20.2% (n = 223/1106). The most common overlapped genes across these signatures were TCF7, IL7R, CCR7, GPR183, KLF3, PABPC1, SELL and LTB (Fig. 4 C). An inverse correlation with multiple TIL-immune signatures across different types of cancer was also identified. Signatures Caushi.CD8.Proliferating and the An immune cell prognosticating signature in cervical cancer (AnCervicalCA) were most frequently found to be inversely correlated to OS across multiple cancer types, both for 3 different cancer types (ACC, MESO, KIRP and KICH, LIHC, SARC) (Supplemental Table 6). 9 , 24 Looking at PFI, Caushi.CD8.Proliferating was most frequently the top inverse correlation found across 6 cancer types (ACC, LIHC MESO, SARC, KIRC, KIRP) (Supplemental Table 6). 9 When we examined the gene composition of these gene signatures (Caushi.CD8.Proliferating, AnCervical CA, Duhen_Tumor React CD8, Liu_Hypoxia, Chatani_TP, Tang_Ferroptosis, LCMV_PROG.EX_Miller, Qi_TREG, Li_Pyroptosis, Jansen_Term diff, Ahuluwalia_Prognostic Cell Death, Wu_OS Pancreatic CA, Yan_TCS, Exhaust_1_Feldman, Caushi.CD4-Th(3), Grog.8TRM.2, Hou_T Cell Prolif, Krishna.ACT.CD8.Term.Diff, Oh.CD4.PROLIF, Oh.CD8.MITO, Oliveira_AAT, Oliveira_Prolif_T, Oliveira_Tumor Spec_Prolif T, Yang_Cupropptosis), we found that 23.74% of their genes were shared across multiple signatures (n = 151/636). The most conserved genes across these signatures were TYMS, TOP2A , and UBE2C each found within 28% of signatures (n = 7/25) (Fig. 4 D). Examining performance across immune cell clusters We next attempted to analyze gene signature performance by grouping samples by immune cells clustered based on likely phenotypic state. These groups included such categories as wound healing, IFN-gamma dominant, Inflammatory, Lymphocyte-depleted, immunologically quiet and TGF-B dominant. 21 Our analysis, however, did not demonstrate a statistically significant correlation between gene signature score and OS or PFI coefficients. Cluster analysis of TIL-Immune Signatures Mapping our TIL-immune signature library by composite genes allowed us to cluster similar signatures into groups. We opted to cluster all signatures into a total of 10 groups as this was the minimal cluster number where the distinction for both correlation and inverse correlation for OS and PFI could be detected (Fig. 5 , Table 5 , Supplemental Fig. 1B). Cluster sizes ranged from 2 to 48 signatures per cluster with a median size of 9 signatures (Supplemental Table 7). By examining group prognostication of OS and PFI we were able to extrapolate that cluster 10 was associated with the longest mean OS and PFI across pan-cancer samples (Table 5 ). The mean OS and PFI coefficients were − 0.11207 and − 0.16673 respectively. Table 5 OS Coefficients for TIL-Immune Signatures Within Cluster #10 Signature Name OS Coefficient Grog.8KLRB1 -0.03943 Grog.CD4.RPL32 -0.44604 Grog.CD4.TCF7 -0.00901 Oh.CD8.MAIT -0.08677 Oh.CD8.RPL 0.012838 Oh.TIL_CD4.GZMK -0.10399 Constructing a novel gene signature By examining the top gene signatures correlating with both OS and PFI across our distinctions of germ cell origin and cancer tumor type, we found that there were 28 unique gene signatures in total that were the highest correlates to OS or PFI within a given category. 8 – 10 , 12 , 15 , 16 , 18 , 22 – 23 , 25 – 34 . We extracted the top 22 genes shared across these signatures including IL7R, TCF6 (seen in 8 signatures), CCR7 and GPR183 (Seen in 7 signatures), etc. (Table 6 , Supplemental Table 8). Table 6 Gene Composition of Novel Signature Gene Name Number of Signatures Present Within (n) IL7R 8 TCF7 8 CCR7 7 GPR183 7 KLF3 6 LTB 5 PABPC1 5 SELL 5 ASPM 4 CCR6 4 CD55 4 CD83 4 CRYBG1 4 DUSP2 4 EMP3 4 ENTPDq 4 FAM177A1 4 LMNA 4 P2RY8 4 PLAC8 4 S1PR1 4 SC5D 4 Examining our novel signature (Novel_Sig) showed concurrence with improved OS (OS coefficient − 0.072) and PFI (PFI coefficient (-0.126 ) compared to the mean across all signatures (OS coefficient: -0.016, PFI coefficient: -0.059) within pan-cancer (Supplemental Table 9). However, our Novel_Sig was not within the top 50 signatures correlated with either prolonged OS or PFI. When examining performance within germ cell origin, our Novel_Sig signature was associated with an OS coefficient less than the average for all signatures within Mesoderm (-0.445 vs. -0.349) and Endoderm (-0.125 vs 0.014) derived cancers (Supplemental Table 10). Regarding PFI, our Novel_Sig was only associated with a PFI coefficient less than the average of all signatures within Endoderm derived cancers (-0.146 vs. 0.039) (Supplemental Table 10). Within both OS and PFI, our Novel_Sig was not within the top correlates, or inverse correlates for OS or PFI across all germ cell origins (supplemental Table 10). When examining tumor types, our Novel_Sig signature only had a statistically significant association with OS or PFI coefficient in 6 cancer types (BRCA, CESC, CHOL, LGG, LIHC, PAAD). Considering OS, our Novel_Sig had an OS coefficient less than the mean across all other signatures within CESC (-0.697 vs. -0.520), CHOL (-0.946 vs. -0.919), LIHC (-0.532 vs. -0.375) and PAAD (0.573 vs. 0.934) (Supplemental Table 11). Only within PAAD did our Novel_Sig score amongst the top 5 signatures with the lowest OS coefficient, otherwise our signature was only modestly below the mean for OS coefficients (Supplemental Table 11). Regarding PFI our Novel_Sig had a PFI coefficient below the mean of all other signatures within BRCA (-0.512 vs. -0.492), CESC (-1.043 vs. -0.089), CHOL (-2.035 vs. -1.878), LIHC (-0.737 vs. -0.538) and PAAD (0.696 vs. 0.99), however in none of these cancer types was our signature within the top lowest PFI coefficients (Supplemental Table 11). Discussion Given the availability and accessibility of genome-wide high-throughput transcriptomic data in cancers, numerous TIL-immune signatures were created and used in bulk transcriptomic data to aid prognostication and predict treatment response to various immunotherapy regimens. In addition, these TIL-immune signatures can improve our understanding of the presence of different immune cells in the cancer microenvironment and the associations with clinical outcomes in the absence of single nuclei transcriptomics performed in large cohorts. However, there were systemic and direct comparisons to assess the differences (or similarities) between these TIL-immune signatures to understand their roles in prognostication across cancer types. To address this gap of knowledge, we curated and compared the components and the performance in prognostication of 146 published TIL-immune signatures in all cancers in TCGA database. Reviewing signature performance across pan-cancer samples, we showed that the Zhang CD8 TCS signature had the overall closest correlation with patient OS and PFI. 16 Survival curves for both OS and PFI demonstrated statistically significant prolonged OS and PFI in TCGA samples demonstrating higher concordance with this Zhang CD8 TCS score (Fig. 2 ). A more nuanced investigation into the performance of these signatures, however, realizes the disparity in signature performance across cancer types. Through our germ cell origin analysis, we see that the Zhang CD8 TCS signature was the most proficient signature correlate of OS and PFI in mesoderm-derived malignancies, however in ectoderm and endoderm-derived malignancies, other signatures (Oh.CD8.CM, Oh.CD4.CM, Caushi-Stem like memory) were associated with lower OS and PFI coefficients. The performance of each TIL-immune signature in predicting OS and PFI varied across different cancer types. Although Zhang CD8 TCS had the lowest OS coefficient across ACC, KIRC, LUAD, PCPG and lowest PFI coefficient in KIRC, KIRP, LUAD, PCPG, and PRAD there is much more variability in top-performing signatures such as the Miller progenitor exhausted T-cell signature (B16_prog.Ex_miller), the Guo suppressive CD4 T-regulatory cell signature (Guo_Supp Treg) and the Chatani CD8 tumor recognition signature (Chatani_TP) having the lowest OS coefficients in BRCA, SKCM, and THYM respectively. In many instances, one-off signatures not seen as top correlates to OS or PFI across pan-cancer or germ cell origin are top predictors of outcomes for individual cancer types. Reliance on the Zhang CD8 TCS signature for all cancers as a prognostic indicator for OS and PFI could provide valuable insight into future cancer behavior, however, our study demonstrates the importance of consideration of numerous published signatures tailored to tumor type and origin to better inform patient-centered decision-making. It should be noted that within our studies of interest, there was variability in the cancer types by which their signatures were derived. Most publications utilized samples from a single histology to generate descriptive signatures (77% of signatures, 112/146) (Supplemental Table 12). Only 6 publications with 34 resultant signatures (23%, 34/146) utilized multiple cancer types to generate TIL-immune signatures. 7 , 14 , 29 , 35 – 37 Of studies utilizing a single cancer histology, the most common included non-small cell lung cancer (NSCLC, 23%, 32/146), bladder cancer (20%, 29/146), and melanoma (20%, 29/146). (Supplemental Table 12). It is no coincidence that these cancer types represent the most “immune-friendly” solid tumors, with checkpoint blockade therapy FDA-approved for all three, including in the management of adjuvant therapy after primary tumor resection. 38 – 40 The emerging role of neoadjuvant checkpoint blockade in the management of these tumors creates a scientific dilemma as checkpoint blockade has an unclarified role in altering T-cell phenotype. Despite the small number of studies utilizing numerous cancer types, we did identify multiple TIL-immune signatures that were top prognostic indicators across multiple cancers (Tang_Ferroptosis and Zhang.CD8.TCS) (Fig. 4 A, Supplemental Table 6). Both were derived from a single histology, HNSC and LUAD respectively. Unsurprisingly, they were also top prognosticators of OS and PFI within their derivative cancer types. Regardless, these two signatures and two others, the Li T-cell pyroptosis signature in glioblastoma (Li-Pyroptosis) and the Oliveira T-cell memory gene signature (Oliveira_TM) outperformed gene signatures derived from the same histology they were prognosticating. Additionally, only 1 signature derived from multiple cancer types was found to correlate highly with PFI or OS. This signature, the Grog T regulatory CD4 signature (Grog.Treg.1), did not correlate with higher performance within any of the cancer types it was derived from (Fig. 4 A and 4 B, Supplemental Table 6). 29 It follows that the immune landscape amongst differing cancer types is likely more complex than anticipated. Signature composition across our library revealed that despite different goals in developing signatures, many genes were conserved across multiple signatures. Notably, many of these conserved genes (ENTPD1, PDCD1, HAVCR2, CXCL13) are those most frequently associated with T cell exhaustion. 7 , 11 , 31 At the onset of our investigation, we surmised that evidence of tumor recognition, as exemplified by increasing signatures of T-cell exhaustion, would drive immune response and lead to improved overall prognosis (Hanada 2022, Lowery 2022, Duhen 2022, Duhen 2018, Chatani 2023). 7 , 11 , 26 , 36 – 37 However, only two signatures describing neoantigen reactivity (Chatani_TP, Oliveira_Tumor Spec_Prolif T) were found to have correlations with OS or PFI within histology or germ cell origin analysis. 10 , 26 Likely, neoantigen recognition alone is not enough to prognosticate tumors as phenotypic state plays a role in describing the overall immune response to cancer. 31 Although the original hypothesis of tumor recognition did not seem apparent, our data did frequently identify signatures expressing genes associated with a “less exhausted” cell phenotype as top prognosticators of OS and PFI (Caushi.stem-like memory, Caushi.CD8.Stem-like memory, B16_Prog.Ex_Miller, Jansen_Stem like, Krishna.ACT.CD8.Stem.Like, LCMV_Prog.EX_Miller). 9 , 25 , 28 , 31 Many of these signatures feature genes associated with more “stem-like” phenotype ( IL7R, TCF7, SELL, CCR7 , etc) (Table 6 ). Despite this association, our novel signature, comprised of these conserved genes, had middling performance across pan-cancer, germ cell, and individual cancer types. Much like neoantigen recognition, “stem-like” phenotype alone is not indicative of better prognostication amongst primary cancers. Across our multiple iterations of signature library analysis, Zhang CD8 TCS was the most consistent prognosticator of OS and PFI. This signature was the top prognosticator across multiple cancer types, one germ cell origin category, and across our pan-cancer analysis as well. The signature was initially constructed through an analysis of available RNA sequencing data for LUAD from TCGA with the intention of constructing a signature of CD8 markers that could predict likely response to immune checkpoint therapy in LUAD patients. 16 The signature itself consists of multiple genes describing T-cell adhesion, early activation, cytokine receptors, and aquaporins. No particular indication of the T-cell phenotypic state was considered when constructing this signature, again lending credence to the idea that the TME and immune cell phenotypic environment is much more complex than anticipated. That said, we believe that use of the Zhang CD8 TCS score could be utilized in patient counseling following surgical resection, or possibly even in the decision algorithm for receipt of adjuvant therapy. Consideration for other well-performing signatures based on cancer cell origin and cancer type should also be considered if mRNA sequencing data is available to potentially assist in prognosticating cancer behavior and the immunoregulatory response following resection of primary tumors. Despite promising potential, these conclusions require additional investigation and confirmation. Although TCGA does include useful metrics such as patient outcomes measures (OS, PFI, etc.), their capture of treatment modalities remains a barrier to more in-depth analysis. As was prevalent amongst many of the signatures we investigated, association of immune score with immune checkpoint blockade would be invaluable in selecting appropriate patients for therapy. Our study did not assess the association between these TIL-immune signatures and patient outcomes by the use of immune checkpoint blockade because the transcriptomic data in many cancers in TCGA were completed before the use of the first FDA-approved immune checkpoint blockade in 2011 became widely adopted. Further investigation into the application of metastatic cancer lesions could prove useful in guiding patients and practitioners in determining complex patient care strategies to combat advanced-stage cancers with systemic immune response activation. In summary, the analysis across pan-cancer, germ cell origin, and individual histology revealed that the Zhang CD8 TCS signature demonstrated the best performance across the broadest scenarios in prognosticating OS and PFI for primary resected tumors. Numerous other signatures, however, perform well in OS and PFI performance when restricted to individual germ cell origin or within individual histology. Variability in prognostication could be due to numerous factors such as cancer behavior, histology, T-cell population, and phenotypic state. Further investigation is warranted to better understand the landscape of TIL populations and their potential in prognosticating and directing patient therapy. Declarations Author Contributions Kyle J. Hitscherich 1* , Darryl Noussome 2 , Aaron J. Dinerman 1 , Victoria Dulemba 1 , Naris Nilubol 3 Kyle J. Hitscherich contributed intellectual conceptualization of this project along with data accumulation, analysis, authorship and publication of figures and tables. Darryl Noussome contributed intellectual conceptualization of this project along with statistical analysis and publication of figures. Aaron J. Dinerman contributed intellectual conceptualization of this project along with data accumulation, analysis, authorship and publication of figures and tables. Victoria Dulemba contributed publication of tables, accumulation of data and authorship of this manuscript. Frank J. Lowery contributed intellectual conceptualization and authorship of this manuscript. Naris Nilubol contributed intellectual conceptualization and authorship of this manuscript. All above listed authors have read this manuscript in full and are in agreement with submission to OncoImmunology for publication. Database Use Disclosure Due to the retrospective and publicly available nature of The Cancer Genome Atlas (TCGA) database this manuscript was exempt from ethics review from the NIH Institutional Review Board (IRB). As the TCGA database is a publicly available dataset, having been previously published for peer review and public use, provided by the NCI specific permission was not required for use within this work. No biological samples from this dataset were used in the preparation of this manuscript. Data Availability Statement The data that support the findings of this study are available from the corresponding author, KJH, upon reasonable request. Data is otherwise publicly available within the TCGA database available in its initial publication at doi: 10.1016/j.cell.2018.03.022 or at their website https://www.cancer.gov/ccg/research/genome-sequencing/tcga. No additional data was generated in preparation of this manuscript. Materials and Correspondence Kyle J Hitscherich DO will serve as corresponding author for this work. Disclosures of Interest Although the authors have previously published TIL signatures, we do not believe this has effected our interpretation of the above data. The authors thus declare that there are no intellectual or financial conflicts of interest regarding this work. Declaration of Funding The research activity in this manuscript was supported by the Intramural Research Program of the NIH Grant #ZIABC011286. References Liu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD, Kovatich AJ, Benz CC, Levine DA, Lee AV, Omberg L, Wolf DM, Shriver CD, Thorsson V, The Cancer Genome Atlas Research Network, Hu H. An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics. Cell 2018. 173(2): 400-416. Zhang Y, Zhang Z. The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. Cell Mol Immunol. 2020 Aug;17(8):807-821. doi: 10.1038/s41423-020-0488-6. Epub 2020 Jul 1. PMID: 32612154; PMCID: PMC7395159. Brudno JN, Lam N, Vanasse D, Shen YW, Rose JJ, Rossi J, Xue A, Bot A, Scholler N, Mikkilineni L, Roschewski M, Dean R, Cachau R, Youkharibache P, Patel R, Hansen B, Stroncek DF, Rosenberg SA, Gress RE, Kochenderfer JN. Safety and feasibility of anti-CD19 CAR T cells with fully human binding domains in patients with B-cell lymphoma. Nat Med. 2020 Feb;26(2):270-280. doi: 10.1038/s41591-019-0737-3. Epub 2020 Jan 20. Erratum in: Nat Med. 2020 May;26(5):803. PMID: 31959992; PMCID: PMC7781235. Rosenberg SA, Yang JC, Sherry RM, Kammula US, Hughes MS, Phan GQ, Citrin DE, Restifo NP, Robbins PF, Wunderlich JR, Morton KE, Laurencot CM, Steinberg SM, White DE, Dudley ME. Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer immunotherapy. Clin Cancer Res. 2011 Jul 1;17(13):4550-7. doi: 10.1158/1078-0432.CCR-11-0116. Epub 2011 Apr 15. PMID: 21498393; PMCID: PMC3131487. Tran E, Robbins PF, Lu YC, Prickett TD, Gartner JJ, Jia L, Pasetto A, Zheng Z, Ray S, Groh EM, Kriley IR, Rosenberg SA. T-Cell Transfer Therapy Targeting Mutant KRAS in Cancer. N Engl J Med. 2016 Dec 8;375(23):2255-2262. doi: 10.1056/NEJMoa1609279. PMID: 27959684; PMCID: PMC5178827. Leidner R, Sanjuan Silva N, Huang H, Sprott D, Zheng C, Shih YP, Leung A, Payne R, Sutcliffe K, Cramer J, Rosenberg SA, Fox BA, Urba WJ, Tran E. Neoantigen T-Cell Receptor Gene Therapy in Pancreatic Cancer. N Engl J Med. 2022 Jun 2;386(22):2112-2119. doi: 10.1056/NEJMoa2119662. PMID: 35648703; PMCID: PMC9531755. Lowery FJ, Krishna S, Yossef R, Parikh NB, Chatani PD, Zacharakis N, Parkhurst MR, Levin N, Sindiri S, Sachs A, Hitscherich KJ, Yu Z, Vale NR, Lu YC, Zheng Z, Jia L, Gartner JJ, Hill VK, Copeland AR, Nah SK, Masi RV, Gasmi B, Kivitz S, Paria BC, Florentin M, Kim SP, Hanada KI, Li YF, Ngo LT, Ray S, Shindorf ML, Levi ST, Shepherd R, Toy C, Parikh AY, Prickett TD, Kelly MC, Beyer R, Goff SL, Yang JC, Robbins PF, Rosenberg SA. Molecular signatures of antitumor neoantigen-reactive T cells from metastatic human cancers. Science. 2022 Feb 25;375(6583):877-884. doi: 10.1126/science.abl5447. Epub 2022 Feb 3. PMID: 35113651; PMCID: PMC8996692. Sade-Feldman M, Yizhak K, Bjorgaard SL, Ray JP, de Boer CG, Jenkins RW, Lieb DJ, Chen JH, Frederick DT, Barzily-Rokni M, Freeman SS, Reuben A, Hoover PJ, Villani AC, Ivanova E, Portell A, Lizotte PH, Aref AR, Eliane JP, Hammond MR, Vitzthum H, Blackmon SM, Li B, Gopalakrishnan V, Reddy SM, Cooper ZA, Paweletz CP, Barbie DA, Stemmer-Rachamimov A, Flaherty KT, Wargo JA, Boland GM, Sullivan RJ, Getz G, Hacohen N. Defining T Cell States Associated with Response to Checkpoint Immunotherapy in Melanoma. Cell. 2018 Nov 1;175(4):998-1013.e20. doi: 10.1016/j.cell.2018.10.038. Erratum in: Cell. 2019 Jan 10;176(1-2):404. PMID: 30388456; PMCID: PMC6641984. Caushi JX, Zhang J, Ji Z, Vaghasia A, Zhang B, Hsiue EH, Mog BJ, Hou W, Justesen S, Blosser R, Tam A, Anagnostou V, Cottrell TR, Guo H, Chan HY, Singh D, Thapa S, Dykema AG, Burman P, Choudhury B, Aparicio L, Cheung LS, Lanis M, Belcaid Z, El Asmar M, Illei PB, Wang R, Meyers J, Schuebel K, Gupta A, Skaist A, Wheelan S, Naidoo J, Marrone KA, Brock M, Ha J, Bush EL, Park BJ, Bott M, Jones DR, Reuss JE, Velculescu VE, Chaft JE, Kinzler KW, Zhou S, Vogelstein B, Taube JM, Hellmann MD, Brahmer JR, Merghoub T, Forde PM, Yegnasubramanian S, Ji H, Pardoll DM, Smith KN. Transcriptional programs of neoantigen-specific TIL in anti-PD-1-treated lung cancers. Nature. 2021 Aug;596(7870):126-132. doi: 10.1038/s41586-021-03752-4. Epub 2021 Jul 21. Erratum in: Nature. 2021 Oct;598(7881):E1. PMID: 34290408; PMCID: PMC8338555. Oliveira G, Stromhaug K, Klaeger S, Kula T, Frederick DT, Le PM, Forman J, Huang T, Li S, Zhang W, Xu Q, Cieri N, Clauser KR, Shukla SA, Neuberg D, Justesen S, MacBeath G, Carr SA, Fritsch EF, Hacohen N, Sade-Feldman M, Livak KJ, Boland GM, Ott PA, Keskin DB, Wu CJ. Phenotype, specificity and avidity of antitumour CD8 + T cells in melanoma. Nature. 2021 Aug;596(7870):119-125. doi: 10.1038/s41586-021-03704-y. Epub 2021 Jul 21. PMID: 34290406; PMCID: PMC9187974. Hanada KI, Zhao C, Gil-Hoyos R, Gartner JJ, Chow-Parmer C, Lowery FJ, Krishna S, Prickett TD, Kivitz S, Parkhurst MR, Wong N, Rae Z, Kelly MC, Goff SL, Robbins PF, Rosenberg SA, Yang JC. A phenotypic signature that identifies neoantigen-reactive T cells in fresh human lung cancers. Cancer Cell. 2022 May 9;40(5):479-493.e6. doi: 10.1016/j.ccell.2022.03.012. Epub 2022 Apr 21. PMID: 35452604; PMCID: PMC9196205. Shi X, Dong A, Jia X, Zheng G, Wang N, Wang Y, Yang C, Lu J, Yang Y. Integrated analysis of single-cell and bulk RNA-sequencing identifies a signature based on T-cell marker genes to predict prognosis and therapeutic response in lung squamous cell carcinoma. Front Immunol. 2022 Oct 14;13:992990. doi: 10.3389/fimmu.2022.992990. PMID: 36311764; PMCID: PMC9614104. Yan K, Lu Y, Yan Z, Wang Y. 9-Gene Signature Correlated With CD8 + T Cell Infiltration Activated by IFN-γ: A Biomarker of Immune Checkpoint Therapy Response in Melanoma. Front Immunol. 2021 Jun 17;12:622563. doi: 10.3389/fimmu.2021.622563. PMID: 34220795; PMCID: PMC8248551. Yang Z, Deng Y, Cheng J, Wei S, Luo H, Liu L. Tumor-Infiltrating PD-1 hi CD8 + -T-Cell Signature as an Effective Biomarker for Immune Checkpoint Inhibitor Therapy Response Across Multiple Cancers. Front Oncol. 2021 Sep 15;11:695006. doi: 10.3389/fonc.2021.695006. PMID: 34604032; PMCID: PMC8479164. Tang Y, Li C, Zhang YJ, Wu ZH. Ferroptosis-Related Long Non-Coding RNA signature predicts the prognosis of Head and neck squamous cell carcinoma. Int J Biol Sci. 2021 Jan 31;17(3):702-711. doi: 10.7150/ijbs.55552. PMID: 33767582; PMCID: PMC7975700. Zhang M, Ma J, Guo Q, Ding S, Wang Y, Pu H. CD8 + T Cell-Associated Gene Signature Correlates With Prognosis Risk and Immunotherapy Response in Patients With Lung Adenocarcinoma. Front Immunol. 2022 Feb 22;13:806877. doi: 10.3389/fimmu.2022.806877. PMID: 35273597; PMCID: PMC8902308. Wu M, Li X, Zhang T, Liu Z, Zhao Y. Identification of a Nine-Gene Signature and Establishment of a Prognostic Nomogram Predicting Overall Survival of Pancreatic Cancer. Front Oncol. 2019 Sep 27;9:996. doi: 10.3389/fonc.2019.00996. PMID: 31612115; PMCID: PMC6776930. Li XY, Zhang LY, Li XY, Yang XT, Su LX. A Pyroptosis-Related Gene Signature for Predicting Survival in Glioblastoma. Front Oncol. 2021 Aug 17;11:697198. doi: 10.3389/fonc.2021.697198. PMID: 34485134; PMCID: PMC8416108. Li J, Cao J, Li P, Deng R, Yao Z, Ying L, Tian J. A Bioinformatic Analysis of Immune-Related Prognostic Genes in Clear Cell Renal Cell Carcinoma Based on TCGA and GEO Databases. Int J Gen Med. 2022 Jan 8;15:325-342. doi: 10.2147/IJGM.S341801. PMID: 35035230; PMCID: PMC8754506. Wilks C, Zheng SC, Chen FY, Charles R, Solomon B, Ling JP, Imada EL, Zhang D, Joseph L, Leek JT, Jaffe AE, Nellore A, Collado-Torres L, Hansen KD, Langmead B. recount3: Summaries and Queries for Large-Scale RNA-seq Expression and Splicing. Genome Biology 2021. 22:323. Thorsson V, Gibbs DL, Brown SD, Wolf D, Bortone DS, Ou Yang TH, Porta-Pardo E, Gao GF, Plaisier CL, Eddy JA, Ziv E, Culhane AC, Paull EO, Sivakumar IKA, Gentles AJ, Malhotra R, Farshidfar F, Colaprico A, Parker JS, Mose LE, Vo NS, Liu J, Liu Y, Rader J, Dhankani V, Reynolds SM, Bowlby R, Califano A, Cherniack AD, Anastassiou D, Bedognetti D, Mokrab Y, Newman AM, Rao A, Chen K, Krasnitz A, Hu H, Malta TM, Noushmehr H, Pedamallu CS, Bullman S, Ojesina AI, Lamb A, Zhou W, Shen H, Choueiri TK, Weinstein JN, Guinney J, Saltz J, Holt RA, Rabkin CS; Cancer Genome Atlas Research Network; Lazar AJ, Serody JS, Demicco EG, Disis ML, Vincent BG, Shmulevich I. The Immune Landscape of Cancer. Immunity. 2018 Apr 17;48(4):812-830.e14. doi: 10.1016/j.immuni.2018.03.023. Epub 2018 Apr 5. Erratum in: Immunity. 2019 Aug 20;51(2):411-412. PMID: 29628290; PMCID: PMC5982584. Oh DY, Kwek SS, Raju SS, Li T, McCarthy E, Chow E, Aran D, Ilano A, Pai CS, Rancan C, Allaire K, Burra A, Sun Y, Spitzer MH, Mangul S, Porten S, Meng MV, Friedlander TW, Ye CJ, Fong L. Intratumoral CD4+ T Cells Mediate Anti-tumor Cytotoxicity in Human Bladder Cancer. Cell. 2020 Jun 25;181(7):1612-1625.e13. doi: 10.1016/j.cell.2020.05.017. Epub 2020 Jun 3. PMID: 32497499; PMCID: PMC7321885. Liu J, Geng R, Yang S, Shao F, Zhong Z, Yang M, Ni S, Cai L, Bai J. Development and Clinical Validation of Novel 8-Gene Prognostic Signature Associated With the Proportion of Regulatory T Cells by Weighted Gene Co-Expression Network Analysis in Uterine Corpus Endometrial Carcinoma. Front Immunol. 2021 Dec 14;12:788431. doi: 10.3389/fimmu.2021.788431. PMID: 34970268; PMCID: PMC8712567. An R, Meng S, Qian H. Identification of Key Pathways and Establishment of a Seven-Gene Prognostic Signature in Cervical Cancer. J Oncol. 2022 Feb 4;2022:4748796. doi: 10.1155/2022/4748796. PMID: 35154316; PMCID: PMC8837458. Jansen CS, Prokhnevska N, Master VA, Sanda MG, Carlisle JW, Bilen MA, Cardenas M, Wilkinson S, Lake R, Sowalsky AG, Valanparambil RM, Hudson WH, McGuire D, Melnick K, Khan AI, Kim K, Chang YM, Kim A, Filson CP, Alemozaffar M, Osunkoya AO, Mullane P, Ellis C, Akondy R, Im SJ, Kamphorst AO, Reyes A, Liu Y, Kissick H. An intra-tumoral niche maintains and differentiates stem-like CD8 T cells. Nature. 2019 Dec;576(7787):465-470. doi: 10.1038/s41586-019-1836-5. Epub 2019 Dec 11. PMID: 31827286; PMCID: PMC7108171. Chatani PD, Lowery FJ, Parikh NB, Hitscherich KJ, Yossef R, Hill V, Gartner JJ, Paria B, Florentin M, Ray S, Bera A, Parkhust M, Robbins P, Krishna S, Rosenberg SA. Cell surface marker-based capture of neoantigen-reactive CD8+ T-cell receptors from metastatic tumor digests. J Immunother Cancer. 2023 May;11(5):e006264. doi: 10.1136/jitc-2022-006264. PMID: 37258038; PMCID: PMC10255266. Dai S, Liu T, Liu XQ, Li XY, Xu K, Ren T, Luo F. Identification of an Immune-Related Signature Predicting Survival Risk and Immune Microenvironment in Gastric Cancer. Front Cell Dev Biol. 2021 Nov 2;9:687473. doi: 10.3389/fcell.2021.687473. PMID: 34805135; PMCID: PMC8596572. Miller BC, Sen DR, Al Abosy R, Bi K, Virkud YV, LaFleur MW, Yates KB, Lako A, Felt K, Naik GS, Manos M, Gjini E, Kuchroo JR, Ishizuka JJ, Collier JL, Griffin GK, Maleri S, Comstock DE, Weiss SA, Brown FD, Panda A, Zimmer MD, Manguso RT, Hodi FS, Rodig SJ, Sharpe AH, Haining WN. Subsets of exhausted CD8+ T cells differentially mediate tumor control and respond to checkpoint blockade. Nat Immunol. 2019 Mar;20(3):326-336. doi: 10.1038/s41590-019-0312-6. Epub 2019 Feb 18. Erratum in: Nat Immunol. 2019 Nov;20(11):1556. PMID: 30778252; PMCID: PMC6673650. Wu TD, Madireddi S, de Almeida PE, Banchereau R, Chen YJ, Chitre AS, Chiang EY, Iftikhar H, O'Gorman WE, Au-Yeung A, Takahashi C, Goldstein LD, Poon C, Keerthivasan S, de Almeida Nagata DE, Du X, Lee HM, Banta KL, Mariathasan S, Das Thakur M, Huseni MA, Ballinger M, Estay I, Caplazi P, Modrusan Z, Delamarre L, Mellman I, Bourgon R, Grogan JL. Peripheral T cell expansion predicts tumour infiltration and clinical response. Nature. 2020 Mar;579(7798):274-278. doi: 10.1038/s41586-020-2056-8. Epub 2020 Feb 26. PMID: 32103181. Guo X, Zhang Y, Zheng L, Zheng C, Song J, Zhang Q, Kang B, Liu Z, Jin L, Xing R, Gao R, Zhang L, Dong M, Hu X, Ren X, Kirchhoff D, Roider HG, Yan T, Zhang Z. Global characterization of T cells in non-small-cell lung cancer by single-cell sequencing. Nat Med. 2018 Jul;24(7):978-985. doi: 10.1038/s41591-018-0045-3. Epub 2018 Jun 25. Erratum in: Nat Med. 2018 Oct;24(10):1628. PMID: 29942094. Krishna S, Lowery FJ, Copeland AR, Bahadiroglu E, Mukherjee R, Jia L, Anibal JT, Sachs A, Adebola SO, Gurusamy D, Yu Z, Hill V, Gartner JJ, Li YF, Parkhurst M, Paria B, Kvistborg P, Kelly MC, Goff SL, Altan-Bonnet G, Robbins PF, Rosenberg SA. Stem-like CD8 T cells mediate response of adoptive cell immunotherapy against human cancer. Science. 2020 Dec 11;370(6522):1328-1334. doi: 10.1126/science.abb9847. PMID: 33303615; PMCID: PMC8883579. Scott AC, Dündar F, Zumbo P, Chandran SS, Klebanoff CA, Shakiba M, Trivedi P, Menocal L, Appleby H, Camara S, Zamarin D, Walther T, Snyder A, Femia MR, Comen EA, Wen HY, Hellmann MD, Anandasabapathy N, Liu Y, Altorki NK, Lauer P, Levy O, Glickman MS, Kaye J, Betel D, Philip M, Schietinger A. TOX is a critical regulator of tumour-specific T cell differentiation. Nature. 2019 Jul;571(7764):270-274. doi: 10.1038/s41586-019-1324-y. Epub 2019 Jun 17. PMID: 31207604; PMCID: PMC7698992. Wang T, Shi J, Li L, Zhou X, Zhang H, Zhang X, Wang Y, Liu L, Sheng L. Single-Cell Transcriptome Analysis Reveals Inter-Tumor Heterogeneity in Bilateral Papillary Thyroid Carcinoma. Front Immunol. 2022 Apr 20;13:840811. doi: 10.3389/fimmu.2022.840811. PMID: 35515000; PMCID: PMC9065345. Wang Y, Chen L, Yu M, Fang Y, Qian K, Wang G, Ju L, Xiao Y, Wang X. Immune-related signature predicts the prognosis and immunotherapy benefit in bladder cancer. Cancer Med. 2020 Oct;9(20):7729-7741. doi: 10.1002/cam4.3400. Epub 2020 Aug 25. PMID: 32841548; PMCID: PMC7571842. Yost KE, Satpathy AT, Wells DK, Qi Y, Wang C, Kageyama R, McNamara KL, Granja JM, Sarin KY, Brown RA, Gupta RK, Curtis C, Bucktrout SL, Davis MM, Chang ALS, Chang HY. Clonal replacement of tumor-specific T cells following PD-1 blockade. Nat Med. 2019 Aug;25(8):1251-1259. doi: 10.1038/s41591-019-0522-3. Epub 2019 Jul 29. PMID: 31359002; PMCID: PMC6689255. Duhen T, Duhen R, Montler R, Moses J, Moudgil T, de Miranda NF, Goodall CP, Blair TC, Fox BA, McDermott JE, Chang SC, Grunkemeier G, Leidner R, Bell RB, Weinberg AD. Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors. Nat Commun. 2018 Jul 13;9(1):2724. doi: 10.1038/s41467-018-05072-0. PMID: 30006565; PMCID: PMC6045647. Duhen R, Fesneau O, Samson KA, Frye AK, Beymer M, Rajamanickam V, Ross D, Tran E, Bernard B, Weinberg AD, Duhen T. PD-1 and ICOS coexpression identifies tumor-reactive CD4+ T cells in human solid tumors. J Clin Invest. 2022 Jun 15;132(12):e156821. doi: 10.1172/JCI156821. PMID: 35439168; PMCID: PMC9197519. Felip E, Altorki N, Zhou C, Csőszi T, Vynnychenko I, Goloborodko O, Luft A, Akopov A, Martinez-Marti A, Kenmotsu H, Chen YM, Chella A, Sugawara S, Voong D, Wu F, Yi J, Deng Y, McCleland M, Bennett E, Gitlitz B, Wakelee H; IMpower010 Investigators. Adjuvant atezolizumab after adjuvant chemotherapy in resected stage IB-IIIA non-small-cell lung cancer (IMpower010): a randomised, multicentre, open-label, phase 3 trial. Lancet. 2021 Oct 9;398(10308):1344-1357. doi: 10.1016/S0140-6736(21)02098-5. Epub 2021 Sep 20. Erratum in: Lancet. 2021 Sep 23;: PMID: 34555333. Bajorin DF, Witjes JA, Gschwend JE, Schenker M, Valderrama BP, Tomita Y, Bamias A, Lebret T, Shariat SF, Park SH, Ye D, Agerbaek M, Enting D, McDermott R, Gajate P, Peer A, Milowsky MI, Nosov A, Neif Antonio J Jr, Tupikowski K, Toms L, Fischer BS, Qureshi A, Collette S, Unsal-Kacmaz K, Broughton E, Zardavas D, Koon HB, Galsky MD. Adjuvant Nivolumab versus Placebo in Muscle-Invasive Urothelial Carcinoma. N Engl J Med. 2021 Jun 3;384(22):2102-2114. doi: 10.1056/NEJMoa2034442. Erratum in: N Engl J Med. 2021 Aug 26;385(9):864. PMID: 34077643; PMCID: PMC8215888. Eggermont AMM, Blank CU, Mandala M, Long GV, Atkinson V, Dalle S, Haydon A, Lichinitser M, Khattak A, Carlino MS, Sandhu S, Larkin J, Puig S, Ascierto PA, Rutkowski P, Schadendorf D, Koornstra R, Hernandez-Aya L, Maio M, van den Eertwegh AJM, Grob JJ, Gutzmer R, Jamal R, Lorigan P, Ibrahim N, Marreaud S, van Akkooi ACJ, Suciu S, Robert C. Adjuvant Pembrolizumab versus Placebo in Resected Stage III Melanoma. N Engl J Med. 2018 May 10;378(19):1789-1801. doi: 10.1056/NEJMoa1802357. Epub 2018 Apr 15. PMID: 29658430. Additional Declarations No competing interests reported. Supplementary Files SupplementalFigure1.jpg ManuscriptSupplementalTables.xlsx Cite Share Download PDF Status: Published Journal Publication published 07 Aug, 2025 Read the published version in Cancer Immunology, Immunotherapy → Version 1 posted Editorial decision: Revision requested 28 Apr, 2025 Reviews received at journal 27 Apr, 2025 Reviews received at journal 25 Apr, 2025 Reviews received at journal 25 Apr, 2025 Reviewers agreed at journal 17 Apr, 2025 Reviewers agreed at journal 16 Apr, 2025 Reviewers agreed at journal 16 Apr, 2025 Reviewers invited by journal 15 Apr, 2025 Editor assigned by journal 15 Apr, 2025 Submission checks completed at journal 15 Apr, 2025 First submitted to journal 13 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6441170","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":445572721,"identity":"611e0df4-f274-469e-953c-6442a1055534","order_by":0,"name":"Kyle Hitscherich","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIie3PMQrCMBSA4YRAXIJ1TKl6hkDA0V7FUHAqKDiJg0pBFw8Q8BJOmRVBr1Cog9UTiCiKDqZdujWOgvmH5FHywSsANtsPRgGaIMDyCR71RapmAjPCcpxJgr8jICeYZl+MxJVieu73376zjHbDW9iuY4DSU1xCPCoiLhkT8rDrJg0V6MUw52EJaVIx8whjHUbDVuIqpAnBnoHMX5r4jPbuA1eNzUQvNkOawBUNMbyorZm4izTSi3Eh4y73oNoTjAz/QvfB5kreTd+RQXp5qpHvVKL0XEZArVPMiORn2fMsZ13M8GF6bbPZbH/ZB1mDQBBmsZfuAAAAAElFTkSuQmCC","orcid":"","institution":"National Cancer Institute","correspondingAuthor":true,"prefix":"","firstName":"Kyle","middleName":"","lastName":"Hitscherich","suffix":""},{"id":445572722,"identity":"4edf9609-8f91-4564-942c-6a63c38efdf8","order_by":1,"name":"Darryl Noussome","email":"","orcid":"","institution":"National Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Darryl","middleName":"","lastName":"Noussome","suffix":""},{"id":445572725,"identity":"ea2aceb6-9454-41d9-ad81-817cf8316402","order_by":2,"name":"Aaron Dinerman","email":"","orcid":"","institution":"National Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Aaron","middleName":"","lastName":"Dinerman","suffix":""},{"id":445572726,"identity":"b710a925-b628-4e35-b679-03ed04353c0a","order_by":3,"name":"Victoria Dulemba","email":"","orcid":"","institution":"National Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Victoria","middleName":"","lastName":"Dulemba","suffix":""},{"id":445572727,"identity":"3a7618e7-52f1-42ff-b28d-22bccec9cc44","order_by":4,"name":"Frank Lowery","email":"","orcid":"","institution":"National Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Lowery","suffix":""},{"id":445572728,"identity":"055bc4bc-b0f8-48a0-9838-2b49dd6a7af3","order_by":5,"name":"Naris Nilubol","email":"","orcid":"","institution":"National Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Naris","middleName":"","lastName":"Nilubol","suffix":""}],"badges":[],"createdAt":"2025-04-13 21:38:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6441170/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6441170/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00262-025-04102-3","type":"published","date":"2025-08-07T15:58:04+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":81382515,"identity":"3a1ebc07-c5d5-4e7e-9214-e15fd0d2ea4f","added_by":"auto","created_at":"2025-04-25 13:07:43","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":493897,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConstruction of RNA-sequencing TIL-Immune Signature Library Reveals High Conservation of Composite Genes Across Signatures:\u003c/strong\u003e Flowchart outlining the construction of our RNA-sequencing TIL-immune signature library with the inclusion of both primary samples sourced molecular signatures as well as TCGA-sourced samples (A). Comparing all 146 gene signatures amongst one another, we see a high level of similarity between signatures demonstrated via heatmap (B). Pie chart demonstrating the high level of shared genes comprising the majority of these compiled signatures (C). The most commonly shared genes include those commonly associated with more “exhausted” phenotypes (ENTPD1, PDCD1, HAVCR2, etc.) (D).\u003c/p\u003e","description":"","filename":"ManuscriptFigure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/393885c7edaf4613a9cd3358.jpg"},{"id":81383426,"identity":"4a2eeb19-e45b-4df9-b1b5-74e3a3e90855","added_by":"auto","created_at":"2025-04-25 13:15:47","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":215600,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTIL-Immune Signature Performance Across Pan-Cancer Reveals Prognostic Capabilities for OS and PFI:\u003c/strong\u003e Across our pan-cancer analysis, one signature (Zhang CD8 TCS) was associated with the longest OS and PFI across samples (A). When accounting for AJCC cancer staging, we were able to plot OS survival for patients demonstrated via the Kaplan-Meier curve (B). Stratifying samples instead by Zhang CD8 TCS score into quartiles, we see a distinction in OS and PFI correlating with higher signature scores (C and D).\u003c/p\u003e","description":"","filename":"ManuscriptFigure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/de1669a4e02eceb40ae6a875.jpg"},{"id":81382524,"identity":"bf60f356-fef7-4364-a530-e069ba4430a8","added_by":"auto","created_at":"2025-04-25 13:07:44","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":283540,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTIL-Immune Signature Performance Differentiates Across Germ Cell Origin Samples:\u003c/strong\u003e By grouping samples based on germ cell origin, we identify several signatures whose scores correlated with OS and PFI across Ectoderm (A), Endoderm (B) and Mesoderm (C) malignancies. There were several conserved genes found across several of these top-performing signatures (D).\u003c/p\u003e","description":"","filename":"ManuscriptFigure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/b642fd6446bf9f0ececa36fa.jpg"},{"id":81382529,"identity":"97e8e9bb-880c-4acf-a95d-3f52f501f4dc","added_by":"auto","created_at":"2025-04-25 13:07:44","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":363937,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTIL-Immune Signature Performance Differentiates Across Individual Cancer Histology:\u003c/strong\u003e By grouping samples based on cancer histology, we demonstrate several signatures whose score correlates with improved OS (A) and PFI (B). Across these top-performing signatures, numerous genes are shared (C). Similarly, for poorly performing signatures numerous genes are shared (D).\u003c/p\u003e","description":"","filename":"ManuscriptFigure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/b3f1981e49b3b69dca6ff019.jpg"},{"id":81383423,"identity":"5f5c6228-7089-4a07-9e43-2f8ca209b8c6","added_by":"auto","created_at":"2025-04-25 13:15:46","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":200548,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCluster Analysis of Signatures Identifies High Performing Similar Signatures:\u003c/strong\u003e By clustering signatures based on the similarity of their composite genes, we can analyze cluster performance against pan-cancer samples. The mean OS coefficient is plotted for each cluster.\u003c/p\u003e","description":"","filename":"ManuscriptFigure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/d4580be59a0fbc2bbad13968.jpg"},{"id":88814794,"identity":"19705b3f-cb64-4c62-951b-afb0a92f0be3","added_by":"auto","created_at":"2025-08-11 16:10:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3023373,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/5fb1a0a4-278e-4774-9ebe-287b2c05ec3b.pdf"},{"id":81382518,"identity":"350c0a40-5c43-48a0-8f15-dccd46c8bee2","added_by":"auto","created_at":"2025-04-25 13:07:44","extension":"jpg","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":298656,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementalFigure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/3ee74c83de2a04017ac95c18.jpg"},{"id":81383417,"identity":"e8c20cb9-e566-44e5-8196-da0513786d1f","added_by":"auto","created_at":"2025-04-25 13:15:45","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":405394,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptSupplementalTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6441170/v1/c131b492846d755885621739.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Pan-Cancer Comparative Analysis of The Cancer Genome Atlas Transcriptomic TIL-Immune Signatures","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe Cancer Genome Atlas (TCGA) program curated multi-omic data, clinical characteristics, and outcomes of over 10,000 primary cancers from 33 cancer types over the past 17 years.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e The goal of the project evolved from understanding the genetics of a select cancer histology, (glioblastoma multiforme) to now using combined tumor and microenvironment transcriptomics to better understand cancer biology as it relates to outcomes.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eImmunotherapy has evolved over the past two decades as a promising arm of cancer therapy that can provide robust and durable responses for a wide variety of cancer types.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e As success with checkpoint inhibitors (anti-PD-1, -PD-L1, -CTLA-4) and adoptive cell transfer (ACT, including chimeric antigen receptor (CAR) and tumor-infiltrating lymphocyte (TIL)) become more prevalent, the interest in the regulation of immune system in cancer microenvironment has grown. \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e Multiple immune gene signatures have been developed from transcriptomic data looking to better describe TIL populations and how they may impact the tumor microenvironment (TME).\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e Several of these in-depth evaluations of the tumor-immune microenvironment have sought to identify populations of T-cells that may be associated with disease progression through cytotoxic function, neoantigen recognition, or more \u0026ldquo;stem-like\u0026rdquo; phenotypic state in specific cancers.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eIn querying the TCGA, numerous research teams have sought to incorporate patient outcomes data into the development of such predictive gene signatures. \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e In many instances, these signatures include similar genes compared to those constructed from direct primary patient source data, however, coupling outcomes data from TCGA database has allowed for the investigation of novel genes, some specific to rare cancers and their subtypes.\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eAlthough there were over 150 TIL-TIL-immune signatures published, no study has compared the prognostic performance of these TIL-immune signatures to identify the top-ranked predictive TIL-immune signatures across all cancer, individual cancer types, and germ-cell origins. Previous work has demonstrated utility in such signatures in predicting response rates to immunotherapy such as checkpoint inhibitors.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e Further understanding of the ideal immune cell population found within TIL could expand such therapeutic tools and aid clinicians in selecting patients who may benefit from such therapies.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eGene signature library construction and sample accrual\u003c/h2\u003e \u003cp\u003eA library of Tumor Infiltrating Lymphocyte (TIL) immune transcriptomic signatures was generated by PubMed literature review, specifically focusing on recent publications analyzing RNA-sequencing data on TIL derived from patients with metastatic cancers without restriction placed on histology. Emphasis was placed on publications identifying one or multiple T cell populations characterized by defined TIL-TIL-immune signatures (eg. \u0026ldquo;stem-like\u0026rdquo;, \u0026ldquo;terminally differentiated\u0026rdquo;, \u0026ldquo;effector memory\u0026rdquo;, \u0026ldquo;tissue resident memory\u0026rdquo; etc.). This library was broadened by searching \u0026ldquo;TCGA\u0026rdquo;, \u0026ldquo;transcriptomic\u0026rdquo; and \u0026ldquo;signature\u0026rdquo; to accumulate publications compiled from RNA-sequencing signatures specifically derived from TCGA database review, compared to those previously identified from direct, patient-sourced studies of immune transcriptomics and phenotypes. It should be noted that although some of these signatures were constructed for prognostic purposes, many were developed as a descriptive effort to define TIL populations within the TME. These TIL-TIL-immune signatures were then individually queried for all upregulated genes within their composition. Signature gene lists were reviewed to ensure consistent nomenclature across publications. Duplicated signatures were excluded from the final analysis.\u003c/p\u003e \u003cp\u003eThe TCGA recount3 project is an online data source containing the accumulated RNA-sequencing data contained within the TCGA database and across 8,679 studies of human samples.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e RNA-sequencing data were downloaded for 33 cancer types and 9,961 samples. The recount3 project processed all RNA-seq samples via the Monorail system and provided gene-level counts using Gencode v26 (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The GSVA R/ Bioconductor package was used to calculate individual level gene set enrichment scores for each sample. Overall survival (OS) and progression-free interval (PFI) were chosen as primary endpoints similar to the previous publication by Liu et al.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e Based on the nature of TCGA database comprised of non-metastatic primary tumor lesions, we determined PFI as a potentially insightful metric for our study.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTCGA tumor types, Nomenclature, and Available Samples for Analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eEctoderm/Neural Crest\u003c/p\u003e \u003cp\u003en\u0026thinsp;=\u0026thinsp;2646 (26.6%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHead and Neck Squamous Cell Carcinoma (HNSC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e522\u003c/p\u003e \u003cp\u003e(5.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBreast invasive Carcinoma (BRCA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1,093\u003c/p\u003e \u003cp\u003e(11%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePheochromocytoma and Paraganglioma (PCPG)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e179\u003c/p\u003e \u003cp\u003e(1.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSkin Cutaneous melanoma (SKCM)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e103\u003c/p\u003e \u003cp\u003e(1.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUveal Melanoma (UVM)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80\u003c/p\u003e \u003cp\u003e(0.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBrain Lower Grade Glioma (LGG)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e514\u003c/p\u003e \u003cp\u003e(5.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlioblastoma Multiforme (GBM)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e155\u003c/p\u003e \u003cp\u003e(1.6%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMesoderm\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003en\u0026thinsp;=\u0026thinsp;3457 (34.7%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMesothelioma (MESO)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e87\u003c/p\u003e \u003cp\u003e(0.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSarcoma (SARC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e259\u003c/p\u003e \u003cp\u003e(2.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAcute Myeloid Leukemia (LAML)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e126\u003c/p\u003e \u003cp\u003e(1.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdrenocortical Carcinoma (ACC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e79\u003c/p\u003e \u003cp\u003e(0.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCervical Squamous Cell Carcinoma (CESC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e304\u003c/p\u003e \u003cp\u003e(3.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKidney Chromophobe (KICH)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e66\u003c/p\u003e \u003cp\u003e(0.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKidney Renal Clear Cell Carcinoma (KIRC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e531\u003c/p\u003e \u003cp\u003e(5.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKidney Renal Papillary Cell Carcinoma (KIRP)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e290\u003c/p\u003e \u003cp\u003e(3.1%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUterine Corpus Endometrial Carcinoma (UCEC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e541\u003c/p\u003e \u003cp\u003e(5.4%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUterine Carcinosarcoma (UCS)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e57\u003c/p\u003e \u003cp\u003e(0.6%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTesticular Germ Cell Tumor (TGCT)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003cp\u003e(1.5%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProstate Adenocarcinoma (PRAD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e497\u003c/p\u003e \u003cp\u003e(5.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphoid Neoplasm Diffuse Large B-Cell Lymphoma (DLBC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e48\u003c/p\u003e \u003cp\u003e(0.5%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOvarian Serous Cystadenocarcinoma (OV)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e422\u003c/p\u003e \u003cp\u003e(4.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEndoderm\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003en\u0026thinsp;=\u0026thinsp;3858 (38.7%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThymoma (THYM)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e120\u003c/p\u003e \u003cp\u003e(1.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBladder Urothelial Carcinoma (BLCA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e408\u003c/p\u003e \u003cp\u003e(74.5%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCholangiocarcinoma (CHOL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36\u003c/p\u003e \u003cp\u003e(0.4%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eColon Adenocarcinoma (COAD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e458\u003c/p\u003e \u003cp\u003e(4.6%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEsophageal Carcinoma (ESCA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e184\u003c/p\u003e \u003cp\u003e(1.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLiver Hepatocellular Carcinoma (LIHC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e371\u003c/p\u003e \u003cp\u003e(3.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLung Adenocarcinoma (LUAD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e516\u003c/p\u003e \u003cp\u003e(5.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLung Squamous Cell Carcinoma (LUSC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e501\u003c/p\u003e \u003cp\u003e(5.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThyroid Carcinoma (THCA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e505\u003c/p\u003e \u003cp\u003e(5.1%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStomach Adenocarcinoma (STAD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e415\u003c/p\u003e \u003cp\u003e(4.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRectum Adenocarcinoma (READ)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e166\u003c/p\u003e \u003cp\u003e(1.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePancreatic Adenocarcinoma (PAAD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e178\u003c/p\u003e \u003cp\u003e(1.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eGene signature analysis and construction of a novel signature\u003c/h3\u003e\n\u003cp\u003eOS and PFI coefficients were calculated based on the applicability of each TIL-TIL-immune signature for individual samples based on grouped populations. Our analysis was conducted across all cancer specimens, distinction by tissue germ cell origins, individual cancer type, and descriptive immune clusters.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e The association of each TIL-immune signature score and OS and PFI were compared. Analysis was performed for the above-mentioned populations to assess the performance of signatures and broad applicability.\u003c/p\u003e \u003cp\u003eConserved genes found across multiple high-performing TIL-TIL-immune signatures were compiled. The top frequently conserved genes were used to construct a novel signature (Novel_Sig) and compared against the performance of those originally identified within our constructed library.\u003c/p\u003e\n\u003ch3\u003eCluster analysis and concordance evaluation\u003c/h3\u003e\n\u003cp\u003eCluster analysis was performed on all signatures based on the genetic composition of each signature. The organization of clusters was determined to be 10 based on differentiation of positively and negatively related OS and PFI outcomes based on cluster association. Prognostication capability was evaluated for each cluster based on randomly sampled specimens via TCGA data. The remaining 10% of samples not included in prognostication were then used for Kaplan-Meyer analysis across constructed clusters.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of a gene signature library\u003c/h2\u003e \u003cp\u003eWe identified 153 immune transcriptomic signatures from the literature review. Three were not included as they were not specific to T cells, and four were excluded as they only contained 1 gene. We thus examined 146 signatures across 45 publications (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). One hundred and twenty (120) signatures were described in the setting of basic science research and derived from single-cell TIL sequencing sourced directly from patient samples. Twenty-six (26) were developed from the review of available TCGA database and/or in combination with other databases such as the Gene Expression Omnibus (GEO) database. These signatures comprised 3088 unique genes with nearly half (1432) shared across multiple signatures (Supplemental Table\u0026nbsp;1). The average number of sourced samples for the development of a molecular signature was 132.7 with a median of 10 patient samples used. Most sourcing was restricted to a single cancer type (77%, 112/146) with the dominant cancer types including non-small cell lung cancer (NSCLC, 23%, 32/146), bladder cancer (20%, 29/146), and melanoma (20%, 29/146).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eExamining the content of each signature, 1432 of the 3088 genes (46.37%) were shared between at least 2 signatures (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Of these genes shared among multiple signatures, they were found in an average of 4.97 signatures with 65.5% shared across more than 3 signatures. More than 3 quarters (76.71%) of signatures contained 75\u0026ndash;100% of genes shared amongst at least 2 signatures from our library (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). Only 2.74% of compiled signatures were comprised of entirely unique genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). The most overlapped genes across signatures were \u003cem\u003eENTPD1\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;34), \u003cem\u003ePDCD1\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;32), and \u003cem\u003eHAVCR2\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;32) \u003cem\u003e(\u003c/em\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD\u003cem\u003e)\u003c/em\u003e. The median number of genes per signature was 50 with a range of 2 to 1114 genes (Supplemental table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). From querying TCGA database we accumulated bulk RNA transcriptomic data from 9,961 patient samples across 33 tumor types (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) (Supplemental Table\u0026nbsp;2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eExamining performance across pan-cancer\u003c/h2\u003e \u003cp\u003eTo examine performance across pan-cancer we required grouping of samples to undergo gene signature score analysis due to the large number of available primary tumor samples. We opted to cluster samples by similarities in gene expression to forego any other grouping metric in an unbiased fashion. In doing so, we were able to evaluate gene signature scores as they related to OS and PFI coefficients of similar transcriptomic cancer samples (Supplemental Fig.\u0026nbsp;1A). Across our 146 gene signatures, the Zhang CD8 T-Cell associated gene signature for prognosis risk in lung adenocarcinoma (Zhang CD8 TCS) appeared to have the lowest OS and PFI coefficients, consistent with the strongest association with longer OS and PFI (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, Supplemental Table\u0026nbsp;3).\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e Alternatively, the Liu hypoxia-associated gene score in bladder cancer (Liu_Hypoxia) appeared to have the highest OS and PFI coefficients, consistent with an association with shorter OS and PFI (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA).\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWithin our pan-cancer analysis, patient OS correlated with the American Joint Committee on Cancer (AJCC) staging, as expected (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Scoring samples based on Zhang CD8 TCS gene expression, patient samples were separated into quartiles based on concordance with molecular signature expression. When comparing OS from Q2 to Q1 we found a hazards ratio of 0.74 (P\u0026thinsp;=\u0026thinsp;9.95e-\u003csup\u003e7\u003c/sup\u003e) consistent with significantly longer OS with higher concordance of Zhang CD8 TCS gene signature. Similar significance was found comparing Q3 (HR\u0026thinsp;=\u0026thinsp;0.67, P\u0026thinsp;=\u0026thinsp;2.39e-\u003csup\u003e9\u003c/sup\u003e) and Q4 (HR\u0026thinsp;=\u0026thinsp;0.68, P\u0026thinsp;=\u0026thinsp;1.72e-\u003csup\u003e4\u003c/sup\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Thus the higher correlation quartiles (Q2-Q4) were associated with prolonged OS and PFI compared to the lowest sample quartile (Q1) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExamining performance across germ cell origin\u003c/h3\u003e\n\u003cp\u003eAn alternative grouping of samples was performed by germ-cell origin rather than overall transcriptomic similarities. This offered additional insight into the performance of our 146 signatures (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-C). Although many signatures did not demonstrate a statistically significant correlation between OS or PFI coefficients and signature score, several signatures began to show a direct or inverse association with OS and/or PFI coefficients (Supplemental Table\u0026nbsp;4).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhen examining signature score by germ cell origin, the Oh CD8 central memory signature (Oh.CD8.CM) and Zhang.CD8.TCS had the highest correlations with OS and PFI coefficients for ectoderm and mesoderm-derived cancers, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e In endoderm-derived malignancies, the top signatures correlating with OS and PFI coefficient were the Caushi Stem-like memory T-cell signature (Caushi.Stem-Like memory) and Oh CD4 central memory signature (Oh.CD4.CM) respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e Thus, patients whose primary tumors contained more T-cells corresponding with these gene signatures tended towards longer OS and PFI following resection (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). On closer examination, we identified 21.5% of composite genes within these signatures (n\u0026thinsp;=\u0026thinsp;35/163) were shared across multiple signatures (Supplemental Table\u0026nbsp;5). The most conserved genes across these signatures included \u003cem\u003eGPR183\u003c/em\u003e, \u003cem\u003eCCR7\u003c/em\u003e, \u003cem\u003eSELL\u003c/em\u003e, \u003cem\u003eARID5A\u003c/em\u003e, and others (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop TIL-immune signatures that correlated with overall survival (OS) and progression-free interval (PFI) by germ cell origin\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGerm Cell Origin\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSignature Title\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOS Coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOS Coefficient P Value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePFI Coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePFI Coefficient P value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeural Crest\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShi_TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.29137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001457\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.66787\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9.21E-6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMesoderm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYost_CD8_Memory\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.66138\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000233\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.75684\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.8E-6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEctoderm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD8.CM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.61975\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.012588\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.47997\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.029186\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEndoderm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.60349\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.76E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.66531\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.61E-13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMesoderm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD4.CXCL13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.54667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.017151\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.98599\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.72E-6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeural Crest\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD8.PRO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.81151\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.005707\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.13123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.13E-5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eWe conversely identified that several signatures had an inverse correlation with OS and PFI coefficients. The signature titled Liu_Hypoxia had the poorest prognostication for Ectoderm and Endoderm derived malignancies (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e The Oliveira proliferating T-cell gene signature (Oliveira_Prolif_T) and Caushi proliferating CD8 T-cell signature (Caushi.CD8.Proliferating) correlated with shorter PFI and OS respectively within the mesoderm-derived malignancies (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Of these TIL-immune signatures, 10.6% (n\u0026thinsp;=\u0026thinsp;18/170) of their genes were shared amongst two of these signatures, however, no gene was shared across all 3 signatures associated with poor OS and PFI.\u003c/p\u003e\n\u003ch3\u003eExamining performance across cancer types\u003c/h3\u003e\n\u003cp\u003eBy distinguishing individual cancer types, rather than germ cell origin, we saw greater variability across OS and PFI coefficients as correlates to gene signature score (Supplemental Table\u0026nbsp;6). We determined that across our 33 cancer types, 22 had statistically significant variability across the signature score to correlate with OS coefficient and 28 regarding PFI coefficient (Tables\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop TIL-immune signatures that correlated with overall survival (OS) by tumor type\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTumor Histology\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSignature Title\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOS Coef\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOS Pvalue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePF Coef\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePF Pvalue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCPG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-6.24199\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.022149674\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-3.76982\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.002143\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKICH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD8.HSP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.30571\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.008573393\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-2.44199\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.042444\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eACC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-3.34056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.82561E-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-2.38948\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShi_TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-2.05644\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.003787896\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-2.57423\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.04E-05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHNSC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-2.03379\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.02851E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.5004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.04E-06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKIRC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.96574\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.80286E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.63795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.9E-07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSKCM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGuo_Supp Treg\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.96074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.01343793\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.98457\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.004273\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCESC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.57918\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.003553691\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.22583\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.024684\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUCEC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD4.CM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.40552\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000572013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.27701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000207\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSARC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.39883\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.00220507\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.73848\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.047949\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBLCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.24705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000695187\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.31438\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000466\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLIHC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCaushi.CD8.Stem-like memory\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.0595\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001002037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.91496\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000695\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLUAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.01043\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.38238E-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.81351\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000407\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBRCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB16_PROG.EX_Miller\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.91433\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000962026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.84596\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.001901\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLGG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD8.MAIT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.81017\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.009074414\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.66166\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.00686\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLiu_Treg\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.73121\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000587008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.73239\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.001334\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKrishna.ACT.CD8.Term.Diff\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.5895\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.020006639\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.55552\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.012663\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKIRP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD4.MITO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.55617\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.04930259\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.7407\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003681\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGBM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCaushi.CD4-Th(3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.502069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.046295118\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.735503\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.00316\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePAAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWang_Cytokine Exp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.503322\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.014469427\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.556764\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003478\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMESO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eExhaust_1_Feldman\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.38126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.86199E-09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.933397\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000167\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTHYM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eChatani_TP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.629962\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.042234695\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.211128\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.021425\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop Progression Free Interval (PFI) Correlation for Each Histology\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTumor Histology\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSignature Title\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOS Coef\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOS Pvalue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePF Coef\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePF Pvalue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eACC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJansen_Stem like\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-2.89192\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000666\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-3.30626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.13E-06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBLCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.24705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000695\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.31438\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000466\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBRCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD8.CM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.71512\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.040587\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.99211\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.00404\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCESC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKrishna.ACT.CD8.Stem.Like\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.9932\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.012153\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.50782\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000245\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCHOL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLCMV_PROG.EX_Miller\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.82733\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.127838\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-4.2128\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.002698\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTOX_Scott\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.66957\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.051828\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.63124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.038797\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDLBC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOliveira_TM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.13396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.506353\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-3.18516\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.029162\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGBM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLi_Pyroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.083296\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.649199\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.445846\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.019261\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHNSC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-2.03379\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.03E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.5004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.04E-06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKICH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.19384\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.021295\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-5.02205\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.004432\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKIRC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.96574\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.8E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.63795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.9E-07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKIRP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.07239\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.087701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.65042\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003526\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLGG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD8.CM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.68851\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.038533\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.72399\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.008386\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLIHC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCaushi.Stem-like memory\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.03655\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001353\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.9826\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000267\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLUAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.01043\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.38E-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.81351\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000407\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLUSC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDai_Activation CD4 CD8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.10698\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.520351\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.42427\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.03222\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMESO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eExhaust_1_Feldman\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.38126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.86E-09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.933397\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000167\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWang_OS Benefit\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.12112\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.57786\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.41078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.042113\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePAAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWang_Cytokine Exp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.503322\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.014469\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.556764\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003478\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCPG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-6.24199\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.02215\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-3.76982\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.002143\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePRAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZhang CD8 TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.17488\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.892955\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.18777\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003254\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eREAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAhuluwalia_Prognostic Cell Death\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.165135\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.840801\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.43142\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.041254\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSARC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTang_Ferroptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.39883\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.002205\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.73848\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.047949\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSKCM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOliveira_TM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.96642\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.240344\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-2.22317\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003376\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOliveira_Tumor Spec_Prolif T\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.41472\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.081555\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.77828\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.002654\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTHCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eExhaust_1_Feldman\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.11051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.823879\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.022291\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.001125\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTHYM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGrog.Treg.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.805856\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.293964\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.044679\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.037388\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUCEC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOh.CD4.CM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-1.40552\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000572\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-1.27701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000207\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDai_Activation CD4 CD8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.564\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.147362\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.84535\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.035912\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShi_TCS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-2.05644\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.003788\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-2.57423\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.04E-05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eExamining the correlation between immune gene signature and OS showed variability in performance across different cancer types. Across the 22 cancers, 16 unique signatures were found to have the highest correlation to OS coefficient. Despite this variability, two signatures were found to correlate with OS coefficient across multiple cancer types, the Tang ferroptosis related gene signature in head and neck squamous cell carcinoma (Tang_Ferroptosis) and Zhang.CD8.TCS, each seen across 4 cancer types (BLCA, CESC, HNSC, SARC and ACC, KIRC, LUAD, PCPG) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eRegarding the PFI coefficient, 17 signatures were found to be correlated with longer PFI across our 28 cancer types. Similar to OS, Tang_Ferroptosis and Zhang.CD8.TCS signatures shared high correlation with PFI. The Zhang.CD8.TCS signature was the best correlate to PFI within five cancer types (KIRC, KIRP, LUAD, PCPG, PRAD) and Tang_Ferroptosis in four (BLCA, HNSC, KICH, SARC) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB, Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e The number of shared genes across top correlates for cancer type was 20.2% (n\u0026thinsp;=\u0026thinsp;223/1106). The most common overlapped genes across these signatures were \u003cem\u003eTCF7, IL7R, CCR7, GPR183, KLF3, PABPC1, SELL and LTB\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003eAn inverse correlation with multiple TIL-immune signatures across different types of cancer was also identified. Signatures Caushi.CD8.Proliferating and the An immune cell prognosticating signature in cervical cancer (AnCervicalCA) were most frequently found to be inversely correlated to OS across multiple cancer types, both for 3 different cancer types (ACC, MESO, KIRP and KICH, LIHC, SARC) (Supplemental Table\u0026nbsp;6).\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e Looking at PFI, Caushi.CD8.Proliferating was most frequently the top inverse correlation found across 6 cancer types (ACC, LIHC MESO, SARC, KIRC, KIRP) (Supplemental Table\u0026nbsp;6).\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e When we examined the gene composition of these gene signatures (Caushi.CD8.Proliferating, AnCervical CA, Duhen_Tumor React CD8, Liu_Hypoxia, Chatani_TP, Tang_Ferroptosis, LCMV_PROG.EX_Miller, Qi_TREG, Li_Pyroptosis, Jansen_Term diff, Ahuluwalia_Prognostic Cell Death, Wu_OS Pancreatic CA, Yan_TCS, Exhaust_1_Feldman, Caushi.CD4-Th(3), Grog.8TRM.2, Hou_T Cell Prolif, Krishna.ACT.CD8.Term.Diff, Oh.CD4.PROLIF, Oh.CD8.MITO, Oliveira_AAT, Oliveira_Prolif_T, Oliveira_Tumor Spec_Prolif T, Yang_Cupropptosis), we found that 23.74% of their genes were shared across multiple signatures (n\u0026thinsp;=\u0026thinsp;151/636). The most conserved genes across these signatures were \u003cem\u003eTYMS, TOP2A\u003c/em\u003e, and \u003cem\u003eUBE2C\u003c/em\u003e each found within 28% of signatures (n\u0026thinsp;=\u0026thinsp;7/25) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eExamining performance across immune cell clusters\u003c/h2\u003e \u003cp\u003eWe next attempted to analyze gene signature performance by grouping samples by immune cells clustered based on likely phenotypic state. These groups included such categories as wound healing, IFN-gamma dominant, Inflammatory, Lymphocyte-depleted, immunologically quiet and TGF-B dominant.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e Our analysis, however, did not demonstrate a statistically significant correlation between gene signature score and OS or PFI coefficients.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCluster analysis of TIL-Immune Signatures\u003c/h2\u003e \u003cp\u003eMapping our TIL-immune signature library by composite genes allowed us to cluster similar signatures into groups. We opted to cluster all signatures into a total of 10 groups as this was the minimal cluster number where the distinction for both correlation and inverse correlation for OS and PFI could be detected (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Supplemental Fig.\u0026nbsp;1B). Cluster sizes ranged from 2 to 48 signatures per cluster with a median size of 9 signatures (Supplemental Table\u0026nbsp;7). By examining group prognostication of OS and PFI we were able to extrapolate that cluster 10 was associated with the longest mean OS and PFI across pan-cancer samples (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The mean OS and PFI coefficients were \u0026minus;\u0026thinsp;0.11207 and \u0026minus;\u0026thinsp;0.16673 respectively.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOS Coefficients for TIL-Immune Signatures Within Cluster #10\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSignature Name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOS Coefficient\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrog.8KLRB1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.03943\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrog.CD4.RPL32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.44604\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrog.CD4.TCF7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.00901\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOh.CD8.MAIT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.08677\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOh.CD8.RPL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.012838\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOh.TIL_CD4.GZMK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.10399\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eConstructing a novel gene signature\u003c/h2\u003e \u003cp\u003eBy examining the top gene signatures correlating with both OS and PFI across our distinctions of germ cell origin and cancer tumor type, we found that there were 28 unique gene signatures in total that were the highest correlates to OS or PFI within a given category.\u003csup\u003e\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan additionalcitationids=\"CR26 CR27 CR28 CR29 CR30 CR31 CR32 CR33\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. We extracted the top 22 genes shared across these signatures including IL7R, TCF6 (seen in 8 signatures), CCR7 and GPR183 (Seen in 7 signatures), etc. (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, Supplemental Table\u0026nbsp;8).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eGene Composition of Novel Signature\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGene Name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of Signatures Present Within (n)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eIL7R\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eTCF7\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCCR7\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eGPR183\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eKLF3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eLTB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePABPC1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eSELL\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eASPM\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCCR6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCD55\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCD83\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCRYBG1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eDUSP2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eEMP3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eENTPDq\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eFAM177A1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eLMNA\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eP2RY8\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePLAC8\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eS1PR1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eSC5D\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eExamining our novel signature (Novel_Sig) showed concurrence with improved OS (OS coefficient \u0026minus;\u0026thinsp;0.072) and PFI (PFI coefficient (-0.126 ) compared to the mean across all signatures (OS coefficient: -0.016, PFI coefficient: -0.059) within pan-cancer (Supplemental Table\u0026nbsp;9). However, our Novel_Sig was not within the top 50 signatures correlated with either prolonged OS or PFI. When examining performance within germ cell origin, our Novel_Sig signature was associated with an OS coefficient less than the average for all signatures within Mesoderm (-0.445 vs. -0.349) and Endoderm (-0.125 vs 0.014) derived cancers (Supplemental Table\u0026nbsp;10). Regarding PFI, our Novel_Sig was only associated with a PFI coefficient less than the average of all signatures within Endoderm derived cancers (-0.146 vs. 0.039) (Supplemental Table\u0026nbsp;10). Within both OS and PFI, our Novel_Sig was not within the top correlates, or inverse correlates for OS or PFI across all germ cell origins (supplemental Table\u0026nbsp;10).\u003c/p\u003e \u003cp\u003eWhen examining tumor types, our Novel_Sig signature only had a statistically significant association with OS or PFI coefficient in 6 cancer types (BRCA, CESC, CHOL, LGG, LIHC, PAAD). Considering OS, our Novel_Sig had an OS coefficient less than the mean across all other signatures within CESC (-0.697 vs. -0.520), CHOL (-0.946 vs. -0.919), LIHC (-0.532 vs. -0.375) and PAAD (0.573 vs. 0.934) (Supplemental Table\u0026nbsp;11). Only within PAAD did our Novel_Sig score amongst the top 5 signatures with the lowest OS coefficient, otherwise our signature was only modestly below the mean for OS coefficients (Supplemental Table\u0026nbsp;11). Regarding PFI our Novel_Sig had a PFI coefficient below the mean of all other signatures within BRCA (-0.512 vs. -0.492), CESC (-1.043 vs. -0.089), CHOL (-2.035 vs. -1.878), LIHC (-0.737 vs. -0.538) and PAAD (0.696 vs. 0.99), however in none of these cancer types was our signature within the top lowest PFI coefficients (Supplemental Table\u0026nbsp;11).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eGiven the availability and accessibility of genome-wide high-throughput transcriptomic data in cancers, numerous TIL-immune signatures were created and used in bulk transcriptomic data to aid prognostication and predict treatment response to various immunotherapy regimens. In addition, these TIL-immune signatures can improve our understanding of the presence of different immune cells in the cancer microenvironment and the associations with clinical outcomes in the absence of single nuclei transcriptomics performed in large cohorts. However, there were systemic and direct comparisons to assess the differences (or similarities) between these TIL-immune signatures to understand their roles in prognostication across cancer types. To address this gap of knowledge, we curated and compared the components and the performance in prognostication of 146 published TIL-immune signatures in all cancers in TCGA database. Reviewing signature performance across pan-cancer samples, we showed that the Zhang CD8 TCS signature had the overall closest correlation with patient OS and PFI.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e Survival curves for both OS and PFI demonstrated statistically significant prolonged OS and PFI in TCGA samples demonstrating higher concordance with this Zhang CD8 TCS score (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). A more nuanced investigation into the performance of these signatures, however, realizes the disparity in signature performance across cancer types. Through our germ cell origin analysis, we see that the Zhang CD8 TCS signature was the most proficient signature correlate of OS and PFI in mesoderm-derived malignancies, however in ectoderm and endoderm-derived malignancies, other signatures (Oh.CD8.CM, Oh.CD4.CM, Caushi-Stem like memory) were associated with lower OS and PFI coefficients. The performance of each TIL-immune signature in predicting OS and PFI varied across different cancer types. Although Zhang CD8 TCS had the lowest OS coefficient across ACC, KIRC, LUAD, PCPG and lowest PFI coefficient in KIRC, KIRP, LUAD, PCPG, and PRAD there is much more variability in top-performing signatures such as the Miller progenitor exhausted T-cell signature (B16_prog.Ex_miller), the Guo suppressive CD4 T-regulatory cell signature (Guo_Supp Treg) and the Chatani CD8 tumor recognition signature (Chatani_TP) having the lowest OS coefficients in BRCA, SKCM, and THYM respectively. In many instances, one-off signatures not seen as top correlates to OS or PFI across pan-cancer or germ cell origin are top predictors of outcomes for individual cancer types. Reliance on the Zhang CD8 TCS signature for all cancers as a prognostic indicator for OS and PFI could provide valuable insight into future cancer behavior, however, our study demonstrates the importance of consideration of numerous published signatures tailored to tumor type and origin to better inform patient-centered decision-making.\u003c/p\u003e \u003cp\u003eIt should be noted that within our studies of interest, there was variability in the cancer types by which their signatures were derived. Most publications utilized samples from a single histology to generate descriptive signatures (77% of signatures, 112/146) (Supplemental Table\u0026nbsp;12). Only 6 publications with 34 resultant signatures (23%, 34/146) utilized multiple cancer types to generate TIL-immune signatures.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e Of studies utilizing a single cancer histology, the most common included non-small cell lung cancer (NSCLC, 23%, 32/146), bladder cancer (20%, 29/146), and melanoma (20%, 29/146). (Supplemental Table\u0026nbsp;12). It is no coincidence that these cancer types represent the most \u0026ldquo;immune-friendly\u0026rdquo; solid tumors, with checkpoint blockade therapy FDA-approved for all three, including in the management of adjuvant therapy after primary tumor resection.\u003csup\u003e\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e The emerging role of neoadjuvant checkpoint blockade in the management of these tumors creates a scientific dilemma as checkpoint blockade has an unclarified role in altering T-cell phenotype. Despite the small number of studies utilizing numerous cancer types, we did identify multiple TIL-immune signatures that were top prognostic indicators across multiple cancers (Tang_Ferroptosis and Zhang.CD8.TCS) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, Supplemental Table\u0026nbsp;6). Both were derived from a single histology, HNSC and LUAD respectively. Unsurprisingly, they were also top prognosticators of OS and PFI within their derivative cancer types. Regardless, these two signatures and two others, the Li T-cell pyroptosis signature in glioblastoma (Li-Pyroptosis) and the Oliveira T-cell memory gene signature (Oliveira_TM) outperformed gene signatures derived from the same histology they were prognosticating. Additionally, only 1 signature derived from multiple cancer types was found to correlate highly with PFI or OS. This signature, the Grog T regulatory CD4 signature (Grog.Treg.1), did not correlate with higher performance within any of the cancer types it was derived from (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB, Supplemental Table\u0026nbsp;6).\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e It follows that the immune landscape amongst differing cancer types is likely more complex than anticipated.\u003c/p\u003e \u003cp\u003eSignature composition across our library revealed that despite different goals in developing signatures, many genes were conserved across multiple signatures. Notably, many of these conserved genes (ENTPD1, PDCD1, HAVCR2, CXCL13) are those most frequently associated with T cell exhaustion.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e At the onset of our investigation, we surmised that evidence of tumor recognition, as exemplified by increasing signatures of T-cell exhaustion, would drive immune response and lead to improved overall prognosis (Hanada 2022, Lowery 2022, Duhen 2022, Duhen 2018, Chatani 2023).\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e However, only two signatures describing neoantigen reactivity (Chatani_TP, Oliveira_Tumor Spec_Prolif T) were found to have correlations with OS or PFI within histology or germ cell origin analysis.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e Likely, neoantigen recognition alone is not enough to prognosticate tumors as phenotypic state plays a role in describing the overall immune response to cancer.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eAlthough the original hypothesis of tumor recognition did not seem apparent, our data did frequently identify signatures expressing genes associated with a \u0026ldquo;less exhausted\u0026rdquo; cell phenotype as top prognosticators of OS and PFI (Caushi.stem-like memory, Caushi.CD8.Stem-like memory, B16_Prog.Ex_Miller, Jansen_Stem like, Krishna.ACT.CD8.Stem.Like, LCMV_Prog.EX_Miller).\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e Many of these signatures feature genes associated with more \u0026ldquo;stem-like\u0026rdquo; phenotype (\u003cem\u003eIL7R, TCF7, SELL, CCR7\u003c/em\u003e, etc) (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Despite this association, our novel signature, comprised of these conserved genes, had middling performance across pan-cancer, germ cell, and individual cancer types. Much like neoantigen recognition, \u0026ldquo;stem-like\u0026rdquo; phenotype alone is not indicative of better prognostication amongst primary cancers.\u003c/p\u003e \u003cp\u003eAcross our multiple iterations of signature library analysis, Zhang CD8 TCS was the most consistent prognosticator of OS and PFI. This signature was the top prognosticator across multiple cancer types, one germ cell origin category, and across our pan-cancer analysis as well. The signature was initially constructed through an analysis of available RNA sequencing data for LUAD from TCGA with the intention of constructing a signature of CD8 markers that could predict likely response to immune checkpoint therapy in LUAD patients.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e The signature itself consists of multiple genes describing T-cell adhesion, early activation, cytokine receptors, and aquaporins. No particular indication of the T-cell phenotypic state was considered when constructing this signature, again lending credence to the idea that the TME and immune cell phenotypic environment is much more complex than anticipated. That said, we believe that use of the Zhang CD8 TCS score could be utilized in patient counseling following surgical resection, or possibly even in the decision algorithm for receipt of adjuvant therapy. Consideration for other well-performing signatures based on cancer cell origin and cancer type should also be considered if mRNA sequencing data is available to potentially assist in prognosticating cancer behavior and the immunoregulatory response following resection of primary tumors.\u003c/p\u003e \u003cp\u003eDespite promising potential, these conclusions require additional investigation and confirmation. Although TCGA does include useful metrics such as patient outcomes measures (OS, PFI, etc.), their capture of treatment modalities remains a barrier to more in-depth analysis. As was prevalent amongst many of the signatures we investigated, association of immune score with immune checkpoint blockade would be invaluable in selecting appropriate patients for therapy. Our study did not assess the association between these TIL-immune signatures and patient outcomes by the use of immune checkpoint blockade because the transcriptomic data in many cancers in TCGA were completed before the use of the first FDA-approved immune checkpoint blockade in 2011 became widely adopted. Further investigation into the application of metastatic cancer lesions could prove useful in guiding patients and practitioners in determining complex patient care strategies to combat advanced-stage cancers with systemic immune response activation.\u003c/p\u003e \u003cp\u003eIn summary, the analysis across pan-cancer, germ cell origin, and individual histology revealed that the Zhang CD8 TCS signature demonstrated the best performance across the broadest scenarios in prognosticating OS and PFI for primary resected tumors. Numerous other signatures, however, perform well in OS and PFI performance when restricted to individual germ cell origin or within individual histology. Variability in prognostication could be due to numerous factors such as cancer behavior, histology, T-cell population, and phenotypic state. Further investigation is warranted to better understand the landscape of TIL populations and their potential in prognosticating and directing patient therapy.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cu\u003eAuthor Contributions\u0026nbsp;\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eKyle J. Hitscherich\u003csup\u003e1*\u003c/sup\u003e, Darryl Noussome\u003csup\u003e2\u003c/sup\u003e, Aaron J. Dinerman\u003csup\u003e1\u003c/sup\u003e, Victoria Dulemba\u003csup\u003e1\u003c/sup\u003e, Naris Nilubol\u003csup\u003e3\u003c/sup\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eKyle J. Hitscherich contributed intellectual conceptualization of this project along with data accumulation, analysis, authorship and publication of figures and tables.\u003c/p\u003e\n\u003cp\u003eDarryl Noussome contributed intellectual conceptualization of this project along with statistical analysis and publication of figures.\u003c/p\u003e\n\u003cp\u003eAaron J. Dinerman contributed intellectual conceptualization of this project along with data accumulation, analysis, authorship and publication of figures and tables.\u003c/p\u003e\n\u003cp\u003eVictoria Dulemba contributed publication of tables, accumulation of data and authorship of this manuscript.\u003c/p\u003e\n\u003cp\u003eFrank J. Lowery contributed intellectual conceptualization and authorship of this manuscript.\u003c/p\u003e\n\u003cp\u003eNaris Nilubol contributed intellectual conceptualization and authorship of this manuscript.\u003c/p\u003e\n\u003cp\u003eAll above listed authors have read this manuscript in full and are in agreement with submission to OncoImmunology for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eDatabase Use Disclosure\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDue to the retrospective and publicly available nature of The Cancer Genome Atlas (TCGA) database this manuscript was exempt from ethics review from the NIH Institutional Review Board (IRB). As the TCGA database is a publicly available dataset, having been previously published for peer review and public use, provided by the NCI specific permission was not required for use within this work. No biological samples from this dataset were used in the preparation of this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eData Availability Statement\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are available from the corresponding author, KJH, upon reasonable request. Data is otherwise publicly available within the TCGA database available in its initial publication at doi: 10.1016/j.cell.2018.03.022 or at their website https://www.cancer.gov/ccg/research/genome-sequencing/tcga. No additional data was generated in preparation of this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eMaterials and Correspondence\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKyle J Hitscherich DO will serve as corresponding author for this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eDisclosures of Interest\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAlthough the authors have previously published TIL signatures, we do not believe this has effected our interpretation of the above data. The authors thus declare that there are no intellectual or financial conflicts of interest regarding this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eDeclaration of Funding\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research activity in this manuscript was supported by the Intramural Research Program of the NIH Grant #ZIABC011286.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLiu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD, Kovatich AJ, Benz CC, Levine DA, Lee AV, Omberg L, Wolf DM, Shriver CD, Thorsson V, The Cancer Genome Atlas Research Network, Hu H. An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics. Cell 2018. 173(2): 400-416.\u003c/li\u003e\n\u003cli\u003eZhang Y, Zhang Z. The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. Cell Mol Immunol. 2020 Aug;17(8):807-821. doi: 10.1038/s41423-020-0488-6. Epub 2020 Jul 1. PMID: 32612154; PMCID: PMC7395159.\u003c/li\u003e\n\u003cli\u003eBrudno JN, Lam N, Vanasse D, Shen YW, Rose JJ, Rossi J, Xue A, Bot A, Scholler N, Mikkilineni L, Roschewski M, Dean R, Cachau R, Youkharibache P, Patel R, Hansen B, Stroncek DF, Rosenberg SA, Gress RE, Kochenderfer JN. Safety and feasibility of anti-CD19 CAR T cells with fully human binding domains in patients with B-cell lymphoma. Nat Med. 2020 Feb;26(2):270-280. doi: 10.1038/s41591-019-0737-3. Epub 2020 Jan 20. Erratum in: Nat Med. 2020 May;26(5):803. PMID: 31959992; PMCID: PMC7781235.\u003c/li\u003e\n\u003cli\u003eRosenberg SA, Yang JC, Sherry RM, Kammula US, Hughes MS, Phan GQ, Citrin DE, Restifo NP, Robbins PF, Wunderlich JR, Morton KE, Laurencot CM, Steinberg SM, White DE, Dudley ME. Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer immunotherapy. Clin Cancer Res. 2011 Jul 1;17(13):4550-7. doi: 10.1158/1078-0432.CCR-11-0116. Epub 2011 Apr 15. PMID: 21498393; PMCID: PMC3131487.\u003c/li\u003e\n\u003cli\u003eTran E, Robbins PF, Lu YC, Prickett TD, Gartner JJ, Jia L, Pasetto A, Zheng Z, Ray S, Groh EM, Kriley IR, Rosenberg SA. T-Cell Transfer Therapy Targeting Mutant KRAS in Cancer. N Engl J Med. 2016 Dec 8;375(23):2255-2262. doi: 10.1056/NEJMoa1609279. PMID: 27959684; PMCID: PMC5178827.\u003c/li\u003e\n\u003cli\u003eLeidner R, Sanjuan Silva N, Huang H, Sprott D, Zheng C, Shih YP, Leung A, Payne R, Sutcliffe K, Cramer J, Rosenberg SA, Fox BA, Urba WJ, Tran E. Neoantigen T-Cell Receptor Gene Therapy in Pancreatic Cancer. N Engl J Med. 2022 Jun 2;386(22):2112-2119. doi: 10.1056/NEJMoa2119662. PMID: 35648703; PMCID: PMC9531755.\u003c/li\u003e\n\u003cli\u003eLowery FJ, Krishna S, Yossef R, Parikh NB, Chatani PD, Zacharakis N, Parkhurst MR, Levin N, Sindiri S, Sachs A, Hitscherich KJ, Yu Z, Vale NR, Lu YC, Zheng Z, Jia L, Gartner JJ, Hill VK, Copeland AR, Nah SK, Masi RV, Gasmi B, Kivitz S, Paria BC, Florentin M, Kim SP, Hanada KI, Li YF, Ngo LT, Ray S, Shindorf ML, Levi ST, Shepherd R, Toy C, Parikh AY, Prickett TD, Kelly MC, Beyer R, Goff SL, Yang JC, Robbins PF, Rosenberg SA. Molecular signatures of antitumor neoantigen-reactive T cells from metastatic human cancers. Science. 2022 Feb 25;375(6583):877-884. doi: 10.1126/science.abl5447. Epub 2022 Feb 3. PMID: 35113651; PMCID: PMC8996692.\u003c/li\u003e\n\u003cli\u003eSade-Feldman M, Yizhak K, Bjorgaard SL, Ray JP, de Boer CG, Jenkins RW, Lieb DJ, Chen JH, Frederick DT, Barzily-Rokni M, Freeman SS, Reuben A, Hoover PJ, Villani AC, Ivanova E, Portell A, Lizotte PH, Aref AR, Eliane JP, Hammond MR, Vitzthum H, Blackmon SM, Li B, Gopalakrishnan V, Reddy SM, Cooper ZA, Paweletz CP, Barbie DA, Stemmer-Rachamimov A, Flaherty KT, Wargo JA, Boland GM, Sullivan RJ, Getz G, Hacohen N. Defining T Cell States Associated with Response to Checkpoint Immunotherapy in Melanoma. Cell. 2018 Nov 1;175(4):998-1013.e20. doi: 10.1016/j.cell.2018.10.038. Erratum in: Cell. 2019 Jan 10;176(1-2):404. PMID: 30388456; PMCID: PMC6641984.\u003c/li\u003e\n\u003cli\u003eCaushi JX, Zhang J, Ji Z, Vaghasia A, Zhang B, Hsiue EH, Mog BJ, Hou W, Justesen S, Blosser R, Tam A, Anagnostou V, Cottrell TR, Guo H, Chan HY, Singh D, Thapa S, Dykema AG, Burman P, Choudhury B, Aparicio L, Cheung LS, Lanis M, Belcaid Z, El Asmar M, Illei PB, Wang R, Meyers J, Schuebel K, Gupta A, Skaist A, Wheelan S, Naidoo J, Marrone KA, Brock M, Ha J, Bush EL, Park BJ, Bott M, Jones DR, Reuss JE, Velculescu VE, Chaft JE, Kinzler KW, Zhou S, Vogelstein B, Taube JM, Hellmann MD, Brahmer JR, Merghoub T, Forde PM, Yegnasubramanian S, Ji H, Pardoll DM, Smith KN. Transcriptional programs of neoantigen-specific TIL in anti-PD-1-treated lung cancers. Nature. 2021 Aug;596(7870):126-132. doi: 10.1038/s41586-021-03752-4. Epub 2021 Jul 21. Erratum in: Nature. 2021 Oct;598(7881):E1. PMID: 34290408; PMCID: PMC8338555.\u003c/li\u003e\n\u003cli\u003eOliveira G, Stromhaug K, Klaeger S, Kula T, Frederick DT, Le PM, Forman J, Huang T, Li S, Zhang W, Xu Q, Cieri N, Clauser KR, Shukla SA, Neuberg D, Justesen S, MacBeath G, Carr SA, Fritsch EF, Hacohen N, Sade-Feldman M, Livak KJ, Boland GM, Ott PA, Keskin DB, Wu CJ. Phenotype, specificity and avidity of antitumour CD8\u003csup\u003e+\u003c/sup\u003e T cells in melanoma. Nature. 2021 Aug;596(7870):119-125. doi: 10.1038/s41586-021-03704-y. Epub 2021 Jul 21. PMID: 34290406; PMCID: PMC9187974.\u003c/li\u003e\n\u003cli\u003eHanada KI, Zhao C, Gil-Hoyos R, Gartner JJ, Chow-Parmer C, Lowery FJ, Krishna S, Prickett TD, Kivitz S, Parkhurst MR, Wong N, Rae Z, Kelly MC, Goff SL, Robbins PF, Rosenberg SA, Yang JC. A phenotypic signature that identifies neoantigen-reactive T cells in fresh human lung cancers. Cancer Cell. 2022 May 9;40(5):479-493.e6. doi: 10.1016/j.ccell.2022.03.012. Epub 2022 Apr 21. PMID: 35452604; PMCID: PMC9196205.\u003c/li\u003e\n\u003cli\u003eShi X, Dong A, Jia X, Zheng G, Wang N, Wang Y, Yang C, Lu J, Yang Y. Integrated analysis of single-cell and bulk RNA-sequencing identifies a signature based on T-cell marker genes to predict prognosis and therapeutic response in lung squamous cell carcinoma. Front Immunol. 2022 Oct 14;13:992990. doi: 10.3389/fimmu.2022.992990. PMID: 36311764; PMCID: PMC9614104.\u003c/li\u003e\n\u003cli\u003eYan K, Lu Y, Yan Z, Wang Y. 9-Gene Signature Correlated With CD8\u003csup\u003e+\u003c/sup\u003e T Cell Infiltration Activated by IFN-\u0026gamma;: A Biomarker of Immune Checkpoint Therapy Response in Melanoma. Front Immunol. 2021 Jun 17;12:622563. doi: 10.3389/fimmu.2021.622563. PMID: 34220795; PMCID: PMC8248551.\u003c/li\u003e\n\u003cli\u003eYang Z, Deng Y, Cheng J, Wei S, Luo H, Liu L. Tumor-Infiltrating PD-1\u003csup\u003ehi\u003c/sup\u003eCD8\u003csup\u003e+\u003c/sup\u003e-T-Cell Signature as an Effective Biomarker for Immune Checkpoint Inhibitor Therapy Response Across Multiple Cancers. Front Oncol. 2021 Sep 15;11:695006. doi: 10.3389/fonc.2021.695006. PMID: 34604032; PMCID: PMC8479164.\u003c/li\u003e\n\u003cli\u003eTang Y, Li C, Zhang YJ, Wu ZH. Ferroptosis-Related Long Non-Coding RNA signature predicts the prognosis of Head and neck squamous cell carcinoma. Int J Biol Sci. 2021 Jan 31;17(3):702-711. doi: 10.7150/ijbs.55552. PMID: 33767582; PMCID: PMC7975700.\u003c/li\u003e\n\u003cli\u003eZhang M, Ma J, Guo Q, Ding S, Wang Y, Pu H. CD8\u003csup\u003e+\u003c/sup\u003e T Cell-Associated Gene Signature Correlates With Prognosis Risk and Immunotherapy Response in Patients With Lung Adenocarcinoma. Front Immunol. 2022 Feb 22;13:806877. doi: 10.3389/fimmu.2022.806877. PMID: 35273597; PMCID: PMC8902308.\u003c/li\u003e\n\u003cli\u003eWu M, Li X, Zhang T, Liu Z, Zhao Y. Identification of a Nine-Gene Signature and Establishment of a Prognostic Nomogram Predicting Overall Survival of Pancreatic Cancer. Front Oncol. 2019 Sep 27;9:996. doi: 10.3389/fonc.2019.00996. PMID: 31612115; PMCID: PMC6776930.\u003c/li\u003e\n\u003cli\u003eLi XY, Zhang LY, Li XY, Yang XT, Su LX. A Pyroptosis-Related Gene Signature for Predicting Survival in Glioblastoma. Front Oncol. 2021 Aug 17;11:697198. doi: 10.3389/fonc.2021.697198. PMID: 34485134; PMCID: PMC8416108.\u003c/li\u003e\n\u003cli\u003eLi J, Cao J, Li P, Deng R, Yao Z, Ying L, Tian J. A Bioinformatic Analysis of Immune-Related Prognostic Genes in Clear Cell Renal Cell Carcinoma Based on TCGA and GEO Databases. Int J Gen Med. 2022 Jan 8;15:325-342. doi: 10.2147/IJGM.S341801. PMID: 35035230; PMCID: PMC8754506.\u003c/li\u003e\n\u003cli\u003eWilks C, Zheng SC, Chen FY, Charles R, Solomon B, Ling JP, Imada EL, Zhang D, Joseph L, Leek JT, Jaffe AE, Nellore A, Collado-Torres L, Hansen KD, Langmead B. recount3: Summaries and Queries for Large-Scale RNA-seq Expression and Splicing. Genome Biology 2021. 22:323.\u003c/li\u003e\n\u003cli\u003eThorsson V, Gibbs DL, Brown SD, Wolf D, Bortone DS, Ou Yang TH, Porta-Pardo E, Gao GF, Plaisier CL, Eddy JA, Ziv E, Culhane AC, Paull EO, Sivakumar IKA, Gentles AJ, Malhotra R, Farshidfar F, Colaprico A, Parker JS, Mose LE, Vo NS, Liu J, Liu Y, Rader J, Dhankani V, Reynolds SM, Bowlby R, Califano A, Cherniack AD, Anastassiou D, Bedognetti D, Mokrab Y, Newman AM, Rao A, Chen K, Krasnitz A, Hu H, Malta TM, Noushmehr H, Pedamallu CS, Bullman S, Ojesina AI, Lamb A, Zhou W, Shen H, Choueiri TK, Weinstein JN, Guinney J, Saltz J, Holt RA, Rabkin CS; Cancer Genome Atlas Research Network; Lazar AJ, Serody JS, Demicco EG, Disis ML, Vincent BG, Shmulevich I. The Immune Landscape of Cancer. Immunity. 2018 Apr 17;48(4):812-830.e14. doi: 10.1016/j.immuni.2018.03.023. Epub 2018 Apr 5. Erratum in: Immunity. 2019 Aug 20;51(2):411-412. PMID: 29628290; PMCID: PMC5982584.\u003c/li\u003e\n\u003cli\u003eOh DY, Kwek SS, Raju SS, Li T, McCarthy E, Chow E, Aran D, Ilano A, Pai CS, Rancan C, Allaire K, Burra A, Sun Y, Spitzer MH, Mangul S, Porten S, Meng MV, Friedlander TW, Ye CJ, Fong L. Intratumoral CD4+ T Cells Mediate Anti-tumor Cytotoxicity in Human Bladder Cancer. Cell. 2020 Jun 25;181(7):1612-1625.e13. doi: 10.1016/j.cell.2020.05.017. Epub 2020 Jun 3. PMID: 32497499; PMCID: PMC7321885.\u003c/li\u003e\n\u003cli\u003eLiu J, Geng R, Yang S, Shao F, Zhong Z, Yang M, Ni S, Cai L, Bai J. Development and Clinical Validation of Novel 8-Gene Prognostic Signature Associated With the Proportion of Regulatory T Cells by Weighted Gene Co-Expression Network Analysis in Uterine Corpus Endometrial Carcinoma. Front Immunol. 2021 Dec 14;12:788431. doi: 10.3389/fimmu.2021.788431. PMID: 34970268; PMCID: PMC8712567.\u003c/li\u003e\n\u003cli\u003eAn R, Meng S, Qian H. Identification of Key Pathways and Establishment of a Seven-Gene Prognostic Signature in Cervical Cancer. J Oncol. 2022 Feb 4;2022:4748796. doi: 10.1155/2022/4748796. PMID: 35154316; PMCID: PMC8837458.\u003c/li\u003e\n\u003cli\u003eJansen CS, Prokhnevska N, Master VA, Sanda MG, Carlisle JW, Bilen MA, Cardenas M, Wilkinson S, Lake R, Sowalsky AG, Valanparambil RM, Hudson WH, McGuire D, Melnick K, Khan AI, Kim K, Chang YM, Kim A, Filson CP, Alemozaffar M, Osunkoya AO, Mullane P, Ellis C, Akondy R, Im SJ, Kamphorst AO, Reyes A, Liu Y, Kissick H. An intra-tumoral niche maintains and differentiates stem-like CD8 T cells. Nature. 2019 Dec;576(7787):465-470. doi: 10.1038/s41586-019-1836-5. Epub 2019 Dec 11. PMID: 31827286; PMCID: PMC7108171.\u003c/li\u003e\n\u003cli\u003eChatani PD, Lowery FJ, Parikh NB, Hitscherich KJ, Yossef R, Hill V, Gartner JJ, Paria B, Florentin M, Ray S, Bera A, Parkhust M, Robbins P, Krishna S, Rosenberg SA. Cell surface marker-based capture of neoantigen-reactive CD8+ T-cell receptors from metastatic tumor digests. J Immunother Cancer. 2023 May;11(5):e006264. doi: 10.1136/jitc-2022-006264. PMID: 37258038; PMCID: PMC10255266.\u003c/li\u003e\n\u003cli\u003eDai S, Liu T, Liu XQ, Li XY, Xu K, Ren T, Luo F. Identification of an Immune-Related Signature Predicting Survival Risk and Immune Microenvironment in Gastric Cancer. Front Cell Dev Biol. 2021 Nov 2;9:687473. doi: 10.3389/fcell.2021.687473. PMID: 34805135; PMCID: PMC8596572.\u003c/li\u003e\n\u003cli\u003eMiller BC, Sen DR, Al Abosy R, Bi K, Virkud YV, LaFleur MW, Yates KB, Lako A, Felt K, Naik GS, Manos M, Gjini E, Kuchroo JR, Ishizuka JJ, Collier JL, Griffin GK, Maleri S, Comstock DE, Weiss SA, Brown FD, Panda A, Zimmer MD, Manguso RT, Hodi FS, Rodig SJ, Sharpe AH, Haining WN. Subsets of exhausted CD8+ T cells differentially mediate tumor control and respond to checkpoint blockade. Nat Immunol. 2019 Mar;20(3):326-336. doi: 10.1038/s41590-019-0312-6. Epub 2019 Feb 18. Erratum in: Nat Immunol. 2019 Nov;20(11):1556. PMID: 30778252; PMCID: PMC6673650.\u003c/li\u003e\n\u003cli\u003eWu TD, Madireddi S, de Almeida PE, Banchereau R, Chen YJ, Chitre AS, Chiang EY, Iftikhar H, O\u0026apos;Gorman WE, Au-Yeung A, Takahashi C, Goldstein LD, Poon C, Keerthivasan S, de Almeida Nagata DE, Du X, Lee HM, Banta KL, Mariathasan S, Das Thakur M, Huseni MA, Ballinger M, Estay I, Caplazi P, Modrusan Z, Delamarre L, Mellman I, Bourgon R, Grogan JL. Peripheral T cell expansion predicts tumour infiltration and clinical response. Nature. 2020 Mar;579(7798):274-278. doi: 10.1038/s41586-020-2056-8. Epub 2020 Feb 26. PMID: 32103181.\u003c/li\u003e\n\u003cli\u003eGuo X, Zhang Y, Zheng L, Zheng C, Song J, Zhang Q, Kang B, Liu Z, Jin L, Xing R, Gao R, Zhang L, Dong M, Hu X, Ren X, Kirchhoff D, Roider HG, Yan T, Zhang Z. Global characterization of T cells in non-small-cell lung cancer by single-cell sequencing. Nat Med. 2018 Jul;24(7):978-985. doi: 10.1038/s41591-018-0045-3. Epub 2018 Jun 25. Erratum in: Nat Med. 2018 Oct;24(10):1628. PMID: 29942094.\u003c/li\u003e\n\u003cli\u003eKrishna S, Lowery FJ, Copeland AR, Bahadiroglu E, Mukherjee R, Jia L, Anibal JT, Sachs A, Adebola SO, Gurusamy D, Yu Z, Hill V, Gartner JJ, Li YF, Parkhurst M, Paria B, Kvistborg P, Kelly MC, Goff SL, Altan-Bonnet G, Robbins PF, Rosenberg SA. Stem-like CD8 T cells mediate response of adoptive cell immunotherapy against human cancer. Science. 2020 Dec 11;370(6522):1328-1334. doi: 10.1126/science.abb9847. PMID: 33303615; PMCID: PMC8883579.\u003c/li\u003e\n\u003cli\u003eScott AC, D\u0026uuml;ndar F, Zumbo P, Chandran SS, Klebanoff CA, Shakiba M, Trivedi P, Menocal L, Appleby H, Camara S, Zamarin D, Walther T, Snyder A, Femia MR, Comen EA, Wen HY, Hellmann MD, Anandasabapathy N, Liu Y, Altorki NK, Lauer P, Levy O, Glickman MS, Kaye J, Betel D, Philip M, Schietinger A. TOX is a critical regulator of tumour-specific T cell differentiation. Nature. 2019 Jul;571(7764):270-274. doi: 10.1038/s41586-019-1324-y. Epub 2019 Jun 17. PMID: 31207604; PMCID: PMC7698992.\u003c/li\u003e\n\u003cli\u003eWang T, Shi J, Li L, Zhou X, Zhang H, Zhang X, Wang Y, Liu L, Sheng L. Single-Cell Transcriptome Analysis Reveals Inter-Tumor Heterogeneity in Bilateral Papillary Thyroid Carcinoma. Front Immunol. 2022 Apr 20;13:840811. doi: 10.3389/fimmu.2022.840811. PMID: 35515000; PMCID: PMC9065345.\u003c/li\u003e\n\u003cli\u003eWang Y, Chen L, Yu M, Fang Y, Qian K, Wang G, Ju L, Xiao Y, Wang X. Immune-related signature predicts the prognosis and immunotherapy benefit in bladder cancer. Cancer Med. 2020 Oct;9(20):7729-7741. doi: 10.1002/cam4.3400. Epub 2020 Aug 25. PMID: 32841548; PMCID: PMC7571842.\u003c/li\u003e\n\u003cli\u003eYost KE, Satpathy AT, Wells DK, Qi Y, Wang C, Kageyama R, McNamara KL, Granja JM, Sarin KY, Brown RA, Gupta RK, Curtis C, Bucktrout SL, Davis MM, Chang ALS, Chang HY. Clonal replacement of tumor-specific T cells following PD-1 blockade. Nat Med. 2019 Aug;25(8):1251-1259. doi: 10.1038/s41591-019-0522-3. Epub 2019 Jul 29. PMID: 31359002; PMCID: PMC6689255.\u003c/li\u003e\n\u003cli\u003eDuhen T, Duhen R, Montler R, Moses J, Moudgil T, de Miranda NF, Goodall CP, Blair TC, Fox BA, McDermott JE, Chang SC, Grunkemeier G, Leidner R, Bell RB, Weinberg AD. Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors. Nat Commun. 2018 Jul 13;9(1):2724. doi: 10.1038/s41467-018-05072-0. PMID: 30006565; PMCID: PMC6045647.\u003c/li\u003e\n\u003cli\u003eDuhen R, Fesneau O, Samson KA, Frye AK, Beymer M, Rajamanickam V, Ross D, Tran E, Bernard B, Weinberg AD, Duhen T. PD-1 and ICOS coexpression identifies tumor-reactive CD4+ T cells in human solid tumors. J Clin Invest. 2022 Jun 15;132(12):e156821. doi: 10.1172/JCI156821. PMID: 35439168; PMCID: PMC9197519.\u003c/li\u003e\n\u003cli\u003eFelip E, Altorki N, Zhou C, Csőszi T, Vynnychenko I, Goloborodko O, Luft A, Akopov A, Martinez-Marti A, Kenmotsu H, Chen YM, Chella A, Sugawara S, Voong D, Wu F, Yi J, Deng Y, McCleland M, Bennett E, Gitlitz B, Wakelee H; IMpower010 Investigators. Adjuvant atezolizumab after adjuvant chemotherapy in resected stage IB-IIIA non-small-cell lung cancer (IMpower010): a randomised, multicentre, open-label, phase 3 trial. Lancet. 2021 Oct 9;398(10308):1344-1357. doi: 10.1016/S0140-6736(21)02098-5. Epub 2021 Sep 20. Erratum in: Lancet. 2021 Sep 23;: PMID: 34555333.\u003c/li\u003e\n\u003cli\u003eBajorin DF, Witjes JA, Gschwend JE, Schenker M, Valderrama BP, Tomita Y, Bamias A, Lebret T, Shariat SF, Park SH, Ye D, Agerbaek M, Enting D, McDermott R, Gajate P, Peer A, Milowsky MI, Nosov A, Neif Antonio J Jr, Tupikowski K, Toms L, Fischer BS, Qureshi A, Collette S, Unsal-Kacmaz K, Broughton E, Zardavas D, Koon HB, Galsky MD. Adjuvant Nivolumab versus Placebo in Muscle-Invasive Urothelial Carcinoma. N Engl J Med. 2021 Jun 3;384(22):2102-2114. doi: 10.1056/NEJMoa2034442. Erratum in: N Engl J Med. 2021 Aug 26;385(9):864. PMID: 34077643; PMCID: PMC8215888.\u003c/li\u003e\n\u003cli\u003eEggermont AMM, Blank CU, Mandala M, Long GV, Atkinson V, Dalle S, Haydon A, Lichinitser M, Khattak A, Carlino MS, Sandhu S, Larkin J, Puig S, Ascierto PA, Rutkowski P, Schadendorf D, Koornstra R, Hernandez-Aya L, Maio M, van den Eertwegh AJM, Grob JJ, Gutzmer R, Jamal R, Lorigan P, Ibrahim N, Marreaud S, van Akkooi ACJ, Suciu S, Robert C. Adjuvant Pembrolizumab versus Placebo in Resected Stage III Melanoma. N Engl J Med. 2018 May 10;378(19):1789-1801. doi: 10.1056/NEJMoa1802357. Epub 2018 Apr 15. PMID: 29658430.\u003cstrong\u003e\u003cu\u003e\u003cstrong\u003e\u003cu\u003e\u003c/u\u003e\u003c/strong\u003e\u003c/u\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cancer-immunology-immunotherapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ciim","sideBox":"Learn more about [Cancer Immunology, Immunotherapy](http://link.springer.com/journal/262)","snPcode":"262","submissionUrl":"https://submission.nature.com/new-submission/262/3","title":"Cancer Immunology, Immunotherapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Cancer, Immunotherapy, Single Cell Sequencing, Tumor Microenvironment, Tumor Infiltrating Lymphocytes","lastPublishedDoi":"10.21203/rs.3.rs-6441170/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6441170/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEfforts to understand the tumor microenvironment (TME) through basic science research and The Cancer Genome Atlas (TCGA) data analysis have led to the creation of unique immune transcriptomic signatures from tumor-infiltrating lymphocytes (TIL). However, no pan-cancer analysis has been conducted to compare the prognostic performance of these signatures using overall survival (OS) or progression-free interval (PFI) as endpoints. We compiled a library of 146 TIL-immune signatures and evaluated gene signature score correlation with OS and PFI for 9,961 available TCGA samples across 33 cancer types. Zhang CD8 TCS demonstrated higher accuracy in prognosticating both OS and PFI across the pan-cancer landscape, however, variability was seen across cancer types and germ cell origin. Cluster analysis compiled a group of six signatures (Oh.Cd8.MAIT, Grog.8KLRB1, Oh.TIL_CD4.GZMK, Grog.CD4.TCF7, Oh.CD8.RPL, Grog.CD4.RPL32) whose association with OS and PFI could potentially be conserved across multiple cancer types.\u003c/p\u003e","manuscriptTitle":"A Pan-Cancer Comparative Analysis of The Cancer Genome Atlas Transcriptomic TIL-Immune Signatures","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-25 13:07:39","doi":"10.21203/rs.3.rs-6441170/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-04-28T05:55:12+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-27T14:39:11+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-25T18:57:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-25T08:30:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"52637950202753288895814563027723849383","date":"2025-04-17T16:35:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"271213072223278233320409216826265410264","date":"2025-04-16T11:32:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"11235916069249902991422021968878945645","date":"2025-04-16T09:51:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-16T01:02:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-04-15T09:28:10+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-15T09:26:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cancer Immunology, Immunotherapy","date":"2025-04-13T21:30:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cancer-immunology-immunotherapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ciim","sideBox":"Learn more about [Cancer Immunology, Immunotherapy](http://link.springer.com/journal/262)","snPcode":"262","submissionUrl":"https://submission.nature.com/new-submission/262/3","title":"Cancer Immunology, Immunotherapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"93ab3fda-f9e2-48ad-8109-1bd64d669432","owner":[],"postedDate":"April 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-08-11T16:08:33+00:00","versionOfRecord":{"articleIdentity":"rs-6441170","link":"https://doi.org/10.1007/s00262-025-04102-3","journal":{"identity":"cancer-immunology-immunotherapy","isVorOnly":false,"title":"Cancer Immunology, Immunotherapy"},"publishedOn":"2025-08-07 15:58:04","publishedOnDateReadable":"August 7th, 2025"},"versionCreatedAt":"2025-04-25 13:07:39","video":"","vorDoi":"10.1007/s00262-025-04102-3","vorDoiUrl":"https://doi.org/10.1007/s00262-025-04102-3","workflowStages":[]},"version":"v1","identity":"rs-6441170","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6441170","identity":"rs-6441170","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}