DRPPM-PATH-SURVEIOR: Plug-and-Play Survival Analysis of Pathway-level Signatures and Immune Components

doi:10.21203/rs.3.rs-2688545/v1

Download PDF

Method Article

DRPPM-PATH-SURVEIOR: Plug-and-Play Survival Analysis of Pathway-level Signatures and Immune Components

https://doi.org/10.21203/rs.3.rs-2688545/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Pathway-level survival analysis offers the opportunity to examine molecular pathways and immune signatures that influence patient outcomes. However, available survival analysis algorithms are limited in pathway-level function and lack a streamlined analytical process. Here we present a comprehensive pathway-level survival analysis suite, DRPPM-PATH-SURVEIOR, which includes a Shiny user interface with extensive features for systematic exploration of pathways and covariates in a Cox proportional-hazard model. Moreover, our framework offers an integrative strategy for performing Hazard Ratio ranked Gene Set Enrichment Analysis (GSEA) and pathway clustering. As an example, we applied our tool in a combined cohort of melanoma patients treated with checkpoint inhibition (ICI) and identified several immune populations and biomarkers predictive of ICI efficacy. We also analyzed gene expression data of pediatric acute myeloid leukemia (AML) and performed an inverse association of drug targets with the patient’s clinical endpoint. Our analysis derived several drug targets in high-risk KMT2A-fusion-positive patients, which were then validated in AML cell lines in the Genomics of Drug Sensitivity database. Altogether, the tool offers a comprehensive suite for pathway-level survival analysis and a user interface for exploring drug targets, molecular features, and immune populations at different resolutions.

Computational Biology

Pathway-level survival analysis

R Shiny app

hazard ratio GSEA

pathway clustering

Organizing biological knowledge into pathways facilitates the integrative analysis of gene expression data derived from RNA sequencing and proteomics profiling. Common pathway-level analysis tools, such as ENRICHR[1], GSEA [2], are able to perform pathway enrichment analysis based on gene set databases (e.g., KEGG [3], REACTOME [4], MSIGDB [5], LINCS1000 [6], and the Cell Marker database [7]). While these pathway analysis tools tend to focus on differentially expressed genes between two groups of samples, an alternative approach is to infer a pathway activity score in a single sample by transforming the expression of a set of genes into a single value using gene summary statistics, such as maxmean statistics [8], PLAGE [9], GSVA [10], and ssGSEA [2, 11]. Based on this approach, scores derived from custom gene sets or from network analyses [12–14] can then be used to dichotomize the patient population for survival analysis. For example, PERK-associated gene activity was found to be associated with a higher risk in melanoma patients [15], RAS dependency index was developed in pancreatic adenocarcinoma [16], LCK network activity was associated with T-cell acute lymphoblastic leukemia patient survival [17], and an epithelial-mesenchymal transition score was found to be associated with poorer disease-free survival in ovarian and colorectal patients [18]. Moreover, single scores derived from immune markers can be used as an estimate of immune components (e.g., Xcell [19], TIMER 2.0 [20], and geometric mean estimation of tumor infiltrative leukocytes [21]). These immune scores can then be applied in cancer patient classification [22], as a biomarker of check-point immunotherapy response [23], or as a prognosis marker that’s predictive of clinical outcome [24].

These integrative summary scores represent a useful approach in highlighting signaling pathways and immune populations that correlate with the clinical outcome. But existing survival analysis tools either lack a user-friendly interface or has limited functionality for systematic screening of large pathway databases. They are often restricted in available patient cohort or limited to a small subset of pathways [25–27] and are often incapable of accepting external user input data [25–28] or clinically relevant covariates [26, 29, 30]. Thus, to facilitate the public mining of retrospective clinical studies, we introduce DRPPM-PATH-SURVEIOR, a comprehensive plug-and-play suite for pathway-level survival analysis of signature databases. Our tool is presented with the following unique features:

A one-stop tool for expression-based survival analyses.

The ability to include multiple covariates inside the Cox-proportion hazard pathway model.

The ability to summarize prioritized gene signatures into relevant clusters and pathway modules.

The ability to perform hazard ratio ranked gene set enrichment analysis

Altogether, survival analysis is a critical branch of statistics for analyzing the time-to-event, and our tool facilitates a comprehensive survival analysis of pathway-level scores (an additional comparison of features is available in the Supplementary Result Section).

Overview of the entire pipeline. DRPPM-PATH-SURVEIOR is implemented in the R environment, and packages will be automatically installed during runtime. There are four major components of the DRPPM-PATH-SURVEIOR (Fig. 1), which include: 1) The Interactive (UI) Mode. This feature allows for a point-and-click pathway survival analysis. The interactive mode offers the ability to perform select immune deconvolution in real time and perform univariate or complex multivariate analyses of clinical features. 2) The Pipeline (Advanced) Mode. This feature performs a complete survival analysis of pathway databases and gene features. Covariates can be included as part of the systematic screening, and the P-values are corrected by Benjamini-Hochberg. 3) Pathway Connectivity. This feature allows the user to evaluate the similarity between pathways and group pathways that are predictive of the clinical outcome. 4) Hazard Ratio Ranked Gene Set Enrichment Analysis (GSEA). This user interface performs a GSEA analysis based on the gene-level hazard ratio ranking derived from the Pipeline Mode. This feature facilitates the identification of clinically relevant pathways and, in turn, identifies regulators that can drive the underlying gene expression. Additional installation and usage instructions is available in the Supplementary Method Section.

“On-the-fly” Mode: Shiny Interface

DRPPM-PATH-SURVEIOR is a comprehensive framework for analyzing and visualizing “on-the-fly” associations of immune signatures and pathways scores with a clinical endpoint. The application facilitates the partitioning of patients based on pathway scores, estimated immune cells, and gene expression level, followed by univariate Cox-regression survival analysis and multivariate Cox-regression analysis. DRPPM-PATH-SURVEIOR uses several R packages, including survival (v3.2-11), survminer (v0.4.9), GSVA (v1.40.1) [10], and immunedeconv [31](v2.1.0). Pathway score is calculated with the gsva() function based on ssGSEA, GSVA, plage, or zscore. Immune deconvolution is performed with the immunedeconv R package, which includes several deconvolution packages, such as CIBERSORT [32], ESTIMATE [33], and MCP counter [34]. For multivariate analysis, a covariate can be selected from the user-provided meta-information file. The multivariate survival analysis can be performed through additive and multiplicative interaction of two or more variables.

To evaluate the association between pathway and survival S over time t, S(t), is defined through hazard function, h(t), as

$${S}\left({t}\right)={e}{x}{p}(-\underset{0}{\overset{{t}}{\int }}{h}\left({x}\right){d}{x})$$

$${h}\left({t}\right)={{h}}_{0}\left({t}\right)\times {{e}}^{\left({{\beta }}_{1}\times {{X}}_{1}\right)}$$

where h(t) is a hazard rate at time t and h₀(t) is the baseline hazard rate, β₁ is the coefficient related to hazard with β₁ > 0 as high risk and β₁ < 0 as low risk for X₁, a dichotomized based on the gene or pathway score.

To evaluate the pathway association with survival after adjusting for patient meta information X₂ is defined as

$${h}\left({t}\right)={{h}}_{0}\left({t}\right)\times {{e}}^{\left({{\beta }}_{1}\times {{X}}_{1}+{{\beta }}_{2}\times {{X}}_{2}\right)}$$

To evaluate the associated interaction between the pathway and patient meta information is defined by β₃ as

$${h}\left({t}\right)={{h}}_{0}\left({t}\right)\times {{e}}^{\left({{\beta }}_{1}\times {{X}}_{1}+{{\beta }}_{2}\times {{X}}_{2}+{{\beta }}_{3}\times {{X}1\times {X}}_{2}\right)}$$

Pipeline Mode: Systematic Pathway-level Survival Analysis

To facilitate the identification of top high-risk pathways and genes, we have developed a pipeline to systematically assess pathways associated with hazard by a Cox proportional hazard function. The user can provide or select individual genes and pathway databases to perform a systematic screening. Each expression feature is stratified based on a median cutoff. The user also has the option of performing a systematic screening with the inclusion of a covariate as an additive or multiplicative interactive model. The p.value is calculated on the likelihood ratio, wald test. An adjusted p.values can be calculated based on Benjamini-Hochberg correction method. In the output table, genes and pathways are ranked by the likelihood ratio p.value.

Connectivity Mode: Pathway Gene-set Connectivity

The Connectivity Mode offers the user the ability to analyze the similarity between pathways associated with survival. The hazard ratio ranked pathways from the pipeline mode can be used as input to the DRPPM-Jaccard-Connectivity R Shiny app to generate hierarchical clustering based on gene-set similarity. A Jaccard function can calculate distance between pathways:

The Jaccard score function J for two pathways A and B is defined as

J(A, B) = | A ∩ B | / | A ∪ B |

where

A contains n set of genes, A = [a1, a2, …, an]

B contains m set of genes, B = [b1, b2, …, bm]

The Jaccard matrix can be visualized as a heatmap.

Next, the pathways can be clustered using the hclust function from R (v4.1.0) into k-groups (user-specified). Clusters can be visualized as a dendrogram. To overlap survival-associated gene expression, genes within the pathway can be displayed as a table with a flexible sorting feature and added annotation information.

GSEA Mode: Hazard Ratio Ranked Gene Set Enrichment Analysis

From the pipeline mode, we can derive a hazard ratio ranked gene list, which can then be applied as input to the DRPPM-HazardRatio-Ranked-GSEA R Shiny app. This application takes a two-column file of the gene symbols and hazards ratios, which is used as input to the GSEA function from clusterProfiler (v4.0.5). The application performs GSEA, and results can be visualized as a table with additional options for visualizing the GSEA plots through the gseaplot2 function from enrichplot (v1.12.3).

To demonstrate functionalities of the DRPPM-PATH-SURVEIOR framework, we have included use-case examples of biomarker discovery in a cohort of immunotherapy-treated melanoma patients. We have also provided an example use-case strategy for drug repurposing in pediatric acute myeloid leukemia patients.

Identifying Immune Pathways Associated with Effective Checkpoint Inhibition Treatment

To identify predictive biomarkers that facilitate patient selection of patients suitable for immune checkpoint inhibitor (ICI) treatment, we integrated 313 melanoma patients treated with ICI from Riaz et al. (n = 51) [35], Hugo et al. (n = 25) [36], Van Allen et al. (n = 25) [37](n = 42), Liu et al. (n = 122) [38], and Gide et al. (n = 73) [39]. First, we performed a systematic univariate Cox-hazard analysis of individual gene expression in the “Pipeline Mode” and identified 100 genes associated with the better prognosis (Supplementary Table S1). These include PRF1 and HLA-DPA1, which have been previously reported as predictive biomarkers for ICI therapy [40] (Fig. 2). “On-the-fly analysis mode” further demonstrate PRF1 and HLA-DPA1 were significantly higher in patients who respond to ICI treatment (Supplementary Figure S1). We then ranked the genes based on hazard ratio derived from the cox-proportion hazard model and performed a Hazard Ratio Ranked GSEA analysis of the Hallmark database (Fig. 3A). Interferon Gamma was found associated with Low-risk patients (Fig. 3B). Consistently, immune signatures associated with LCK and CSF1 were also associated with Low-risk patients (Fig. 3C, 3D). Through immune deconvolution, we derived an immune score from xCell [19] and an estimated M2-like macrophage population from Cibersort [32]. We found that high immune infiltration with low M2-like (immune suppressive) macrophages were associated with better outcome (Fig. 4A, 4B). Next, we used the “Pipeline Mode” to perform a systematic univariate Cox-hazard analysis of immune signatures and identified to identify 69 immune signatures associated with a better outcome (Supplementary Table S2). A systematic assessment of immune pathway followed by Pathway Connectivity analysis demonstrated 13 immune modules captured a favorable outcome in pretreated RNA samples, including CD8 cytotoxicity, antigen presentation, interferon and immune checkpoint marker signatures (Fig. 5, Supplementary Figure S2). Altogether, our tool provides suggests favorable outcome and is linked with immune activation.

Survival-Directed Therapeutic Discovery in Acute Myeloid Leukemia

To leverage our framework for therapeutic discovery, we obtained the gene expression data and clinical annotation of 220 patients with the KMT2A fusion event from the National Cancer Institute TARGET pediatric acute myeloid leukemia (AML) 1031 cohort (0–22 years of age). The translocation event of the gene KMT2A, also known as mixed lineage leukemia (MLL), is frequently identified in pediatric AML. Through its multiple fusion partners arises a diverse patient population with a need for advanced risk stratification [43]. Through the DRPPM-PATH-SURVEIOR suite of tools, we examined pathways and genes associated with poor outcome and identify therapeutic targets in high-risk patients. First, single samples gene set enrichment analysis (ssGSEA) was performed using the expression data in tandem with the Library of Integrated Network-based Cellular Signatures (LINCS; 31,028 gene sets) LINCS1000 gene sets to calculate the pathway scores (Supplementary Figure S3). Next, the patients were dichotomized through a median cut-point of each pathway score into an above-median or below median group, followed by a Cox proportional hazards regression using the patient’s overall survival (OS) and event free survival (EFS). A hazard ratio value greater than one and a P-value less than 0.05 was applied to identify significant pathways associated with high risk. To prioritize a putative therapeutic target that downregulates genes associated with high-risk AML in the KMT2A subgroup, we examined enriched connectivity map (cMAP) name and Mechanism of Action. We found 12 enriched Cmap names and 6 enriched drug categories grouped by their mechanism of action (Fig. 6A). Notably, genes downregulated by the HDAC inhibitor, Vorinostat, were associated with the worst prognosis based on OS and EFS (Fig. 6B). Furthermore, Vorinostat was highly sensitive in KMT2A fusion-positive AML cell lines based on the genomics of drug sensitivity in cancer (GDSC) database (Fig. 6C). Taken together, we presented an integrative strategy utilizing DRPPM-PATH-SURVEIOR to prioritize pathways based on patient risk and identified a known therapeutic target in high-risk KMT2A fusion-positive AML patients.

DRPPM-PATH-SURVEIOR is designed to visualize and perform systematic survival analysis based on gene pathway information. The application is designed for users with limited experience in programming as well as for advanced users to perform systematic high-throughput pathway screening. In the interactive mode, the Shiny application will ensure reproducibility and can be easily set up and applied in any cohort. In the pipeline mode, the user can apply univariate and multivariate analysis of pathway and patient covariates associated with patient outlook. Our current application can also perform GSEA based on hazard ratio ranking as well as pathway clustering to examine shared gene and pathway features associated with survival. As more RNA sequencing and proteomics data are being captured in large clinical trials, we anticipate DRPPM-PATH-SURVEIOR will enable a collaborative environment for exploring pathway-level and immune features that is predictive of treatment efficacy, especially for immunotherapy.

Acute Myeloid Leukemia (AML), Gene Set Enrichment Analysis (GSEA), Drug sensitivity in cancer (GDSC) database, Immune checkpoint inhibitor (ICI), Overall survival (OS), Event free survival (EFS), Connectivity Map (CMAP), Mechanism of Action (MOA), Therapeutically Applicable Research To Generate Effective Treatments (TARGET)

AVAILABILTY AND REQUIREMENTS

Project name: DRPPM-PATH-SURVEIOR

Project home page: https://github.com/shawlab-moffitt/DRPPM-PATH-SURVEIOR-Suite

Operating system: Platform independent.

Programming language: R version 4.1 or higher.

License: BSD License.

ETHICS DECLARATIONS

Ethics approval and consent to participate:

Not Applicable

Consent for publication:

Not Applicable

Availability of data and materials:

An example of the Shiny app with user upload function is available http://shawlab.science/shiny/DRPPM_PATH_SURVEIOR_UserInput/. An overview of the DRPPM-PATH-SURVEIOR Suite of tools can be found on our GitHub page (https://github.com/shawlab-moffitt/DRPPM-PATH-SURVEIOR-Suite), which includes source code, example data, and an installation guide. An example startup page is available to guide through the DRPPM-PATH-SURVEIOR-Suite with downloadable example files and example scripts. The processed RNA sequencing data were downloaded from iATLAS https://www.cri-iatlas.org/. The raw RNA-seq data for the TARGET data can be downloaded from the Database of Genotypes and Phenotypes (dbGaP) under the study ID phs000465.v21.p8. Subject to the NIH Genomic Data Sharing Policy, the raw data are freely available to all researchers via https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000465.v21.p8. Processed data are available at the National Cancer Institute’s Genomic Data Commons (https://portal.gdc.cancer.gov/) under the TARGET-AML project.ee Supplementary Method for additional details on installation and requirements on file inputs.

Competing interests:

TIS, AO, and DTC report a provisional patent application for the DRPPM-PATH-SURVEIOR software. SAEreports intellectual property (RSI) and stock in Cvergenx. AT reports grants from Bristol Myers Squib, grants from Genentech-Roche, grants from Regeneron, grants from Sanofi-Genzyme, grants from Nektar, grants from Clinigen, grants from Merck, grants from Acrotech, grants from Pfizer, grants from Checkmate, grants from OncoSec, personal fees from Bristol Myers Squibb, personal fees from Merck, personal fees from Easai, personal fees from Instil Bio, personal fees from Clinigin, personal fees from Regeneron, personal fees from Sanofi-Genzyme, personal fees from Novartis, personal fees from Partner Therapeutics, personal fees from Genentech/Roche, personal fees from BioNTech, outside the submitted work. DC, GN, MT, ACT, GDG, XW, PR, and SM declare no other conflict of interest.

Funding:

The study was funded by American Cancer Society IRG-21-145-25, Moffitt Cancer Center Quantitative Science/Team Science, Moffitt Cancer Center Department of Biostatistics and Bioinformatics Pilot Award. This work has been supported in part by the Biostatistics and Bioinformatics Shared Resource at the H. Lee Moffitt Cancer Center & Research Institute, an NCI designated Comprehensive Cancer Center (P30-CA076292).

Acknowledgement:

We are grateful to the American Cancer Society for supporting this work. We thank Sam Coleman for assisting with the downloading of the iATLAS melanoma patient ICI cohort. We extend our sincere thanks to the Biostatistics and Informatics Core of the H. Lee Moffitt Cancer Center and Research Institute.

Authors' contributions:

Contributions AO completed the software development. GN and DC assisted with the analysis and testing of the software. TIS and AO wrote the manuscript. MT, ACT, XW, SE, PR, DG, SM, AT gave suggestions and revised the manuscript. DTC and TIS supervised the whole project. All authors read and approved the final manuscript.

Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, Koplev S, Jenkins SL, Jagodnik KM, Lachmann A et al: Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 2016, 44(W1):W90-97.
Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES et al: Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005, 102(43):15545-15550.
Kanehisa M, Goto S: KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res 2000, 28(1):27-30.
Joshi-Tope G, Gillespie M, Vastrik I, D'Eustachio P, Schmidt E, de Bono B, Jassal B, Gopinath GR, Wu GR, Matthews L et al: Reactome: a knowledgebase of biological pathways. Nucleic Acids Res 2005, 33(Database issue):D428-432.
Liberzon A, Birger C, Thorvaldsdottir H, Ghandi M, Mesirov JP, Tamayo P: The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 2015, 1(6):417-425.
Subramanian A, Narayan R, Corsello SM, Peck DD, Natoli TE, Lu X, Gould J, Davis JF, Tubelli AA, Asiedu JK et al: A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles. Cell 2017, 171(6):1437-1452 e1417.
Zhang X, Lan Y, Xu J, Quan F, Zhao E, Deng C, Luo T, Xu L, Liao G, Yan M et al: CellMarker: a manually curated resource of cell markers in human and mouse. Nucleic Acids Res 2019, 47(D1):D721-D728.
Efron B, Tibshirani R: On testing the significance of sets of genes. The Annals of Applied Statistics 2007, 1(1):107-129, 123.
Tomfohr J, Lu J, Kepler TB: Pathway level analysis of gene expression using singular value decomposition. BMC Bioinformatics 2005, 6:225.
Hanzelmann S, Castelo R, Guinney J: GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013, 14:7.
Barbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF, Schinzel AC, Sandy P, Meylan E, Scholl C et al: Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009, 462(7269):108-112.
Carro MS, Lim WK, Alvarez MJ, Bollo RJ, Zhao X, Snyder EY, Sulman EP, Anne SL, Doetsch F, Colman H et al: The transcriptional network for mesenchymal transformation of brain tumours. Nature 2010, 463(7279):318-325.
Alvarez MJ, Shen Y, Giorgi FM, Lachmann A, Ding BB, Ye BH, Califano A: Functional characterization of somatic mutations in cancer using network-based inference of protein activity. Nat Genet 2016, 48(8):838-847.
Du X, Wen J, Wang Y, Karmaus PWF, Khatamian A, Tan H, Li Y, Guy C, Nguyen TM, Dhungana Y et al: Hippo/Mst signalling couples metabolic state and immune function of CD8alpha(+) dendritic cells. Nature 2018, 558(7708):141-145.
Mandula JK, Chang S, Mohamed E, Jimenez R, Sierra-Mondragon RA, Chang DC, Obermayer AN, Moran-Segura CM, Das S, Vazquez-Martinez JA et al: Ablation of the endoplasmic reticulum stress kinase PERK induces paraptosis and type I interferon to promote anti-tumor T cell responses. Cancer Cell 2022.
Yi M, Nissley DV, McCormick F, Stephens RM: ssGSEA score-based Ras dependency indexes derived from gene expression data reveal potential Ras addiction mechanisms with possible clinical implications. Sci Rep 2020, 10(1):10258.
Gocho Y, Liu J, Hu J, Yang W, Dharia NV, Zhang J, Shi H, Du G, John A, Lin TN et al: Network-based systems pharmacology reveals heterogeneity in LCK and BCL2 signaling and therapeutic sensitivity of T-cell acute lymphoblastic leukemia. Nat Cancer 2021, 2(3):284-299.
Tan TZ, Miow QH, Miki Y, Noda T, Mori S, Huang RY, Thiery JP: Epithelial-mesenchymal transition spectrum quantification and its efficacy in deciphering survival and drug responses of cancer patients. EMBO Mol Med 2014, 6(10):1279-1293.
Aran D, Hu Z, Butte AJ: xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol 2017, 18(1):220.
Li T, Fu J, Zeng Z, Cohen D, Li J, Chen Q, Li B, Liu XS: TIMER2.0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Res 2020, 48(W1):W509-W514.
Danaher P, Warren S, Dennis L, D'Amico L, White A, Disis ML, Geller MA, Odunsi K, Beechem J, Fling SP: Gene expression markers of Tumor Infiltrating Leukocytes. J Immunother Cancer 2017, 5:18.
Thorsson V, Gibbs DL, Brown SD, Wolf D, Bortone DS, Ou Yang TH, Porta-Pardo E, Gao GF, Plaisier CL, Eddy JA et al: The Immune Landscape of Cancer. Immunity 2018, 48(4):812-830 e814.
Coleman S, Xie M, Tarhini AA, Tan AC: Systematic evaluation of the predictive gene expression signatures of immune checkpoint inhibitors in metastatic melanoma. Mol Carcinog 2022.
Petitprez F, de Reynies A, Keung EZ, Chen TW, Sun CM, Calderaro J, Jeng YM, Hsiao LP, Lacroix L, Bougouin A et al: B cells are associated with survival and immunotherapy response in sarcoma. Nature 2020, 577(7791):556-560.
Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Ponce-Rodriguez I, Chakravarthi B, Varambally S: UALCAN: A Portal for Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia 2017, 19(8):649-658.
Lanczky A, Gyorffy B: Web-Based Survival Analysis Tool Tailored for Medical Research (KMplot): Development and Implementation. J Med Internet Res 2021, 23(7):e27633.
Dwivedi B, Mumme H, Satpathy S, Bhasin SS, Bhasin M: Survival Genie, a web platform for survival analysis across pediatric and adult cancers. Sci Rep 2022, 12(1):3069.
Borcherding N, Bormann NL, Voigt AP, Zhang W: TRGAted: A web tool for survival analysis using protein data in the Cancer Genome Atlas. F1000Res 2018, 7:1235.
Rupji M, Zhang X, Kowalski J: CASAS: Cancer Survival Analysis Suite, a web based application. F1000Res 2017, 6:919.
Pak K, Oh SO, Goh TS, Heo HJ, Han ME, Jeong DC, Lee CS, Sun H, Kang J, Choi S et al: A User-Friendly, Web-Based Integrative Tool (ESurv) for Survival Analysis: Development and Validation Study. J Med Internet Res 2020, 22(5):e16084.
Sturm G, Finotello F, List M: Immunedeconv: An R Package for Unified Access to Computational Methods for Estimating Immune Cell Fractions from Bulk RNA-Sequencing Data. Methods Mol Biol 2020, 2120:223-232.
Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, Hoang CD, Diehn M, Alizadeh AA: Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 2015, 12(5):453-457.
Yoshihara K, Shahmoradgoli M, Martinez E, Vegesna R, Kim H, Torres-Garcia W, Trevino V, Shen H, Laird PW, Levine DA et al: Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 2013, 4:2612.
Becht E, Giraldo NA, Lacroix L, Buttard B, Elarouci N, Petitprez F, Selves J, Laurent-Puig P, Sautes-Fridman C, Fridman WH et al: Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol 2016, 17(1):218.
Riaz N, Havel JJ, Makarov V, Desrichard A, Urba WJ, Sims JS, Hodi FS, Martin-Algarra S, Mandal R, Sharfman WH et al: Tumor and Microenvironment Evolution during Immunotherapy with Nivolumab. Cell 2017, 171(4):934-949 e916.
Hugo W, Zaretsky JM, Sun L, Song C, Moreno BH, Hu-Lieskovan S, Berent-Maoz B, Pang J, Chmielowski B, Cherry G et al: Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma. Cell 2016, 165(1):35-44.
Van Allen EM, Miao D, Schilling B, Shukla SA, Blank C, Zimmer L, Sucker A, Hillen U, Foppen MHG, Goldinger SM et al: Genomic correlates of response to CTLA-4 blockade in metastatic melanoma. Science 2015, 350(6257):207-211.
Liu D, Schilling B, Liu D, Sucker A, Livingstone E, Jerby-Arnon L, Zimmer L, Gutzmer R, Satzger I, Loquai C et al: Integrative molecular and clinical modeling of clinical outcomes to PD1 blockade in patients with metastatic melanoma. Nat Med 2019, 25(12):1916-1927.
Gide TN, Quek C, Menzies AM, Tasker AT, Shang P, Holst J, Madore J, Lim SY, Velickovic R, Wongchenko M et al: Distinct Immune Cell Populations Define Response to Anti-PD-1 Monotherapy and Anti-PD-1/Anti-CTLA-4 Combined Therapy. Cancer Cell 2019, 35(2):238-255 e236.
Gibney GT, Weiner LM, Atkins MB: Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Lancet Oncol 2016, 17(12):e542-e551.

The Supplementary Methods and Results, including Tables S1-S2, are not available with this version.

SupplementaryFigureS1.tiff
Figure S1. Associating PRF1 (A) and HLA-DPA1 (B) expression with ICI Treatment Response in melanoma patients. C) User interface can be accessed from the Data Exploration tab (1) then from the Feature Boxplot tab (2) with the selected feature: “Responder” (3).
SupplementaryFigureS2.tiff
Supplementary Figure S2. 12 Immune Hubs assocaited with a favorable outcome in skin cancer patients treated with ICI. Pathways were clustered by a Jaccard distance calculated based on overlapping genes. Major immune categories were individually circled.
SupplementaryFigureS3.tiff
Supplementary Figure S3. Schematic workflow of ssGSEA Pathway Analysis method. With the input data of an expression matrix, clinical data, and a gene set file, single sample gene set enrichment (ssGSEA) scores were calculated. These scores were further binned into categories of above and below the median ssGSEA score for each gene set. A cox regression analysis was performed on each gene set to generate a comprehensive table of gene sets to filter according to significant, high-risk patients (hazard ratio > 1, p.value < 0.05). The “High Risk” gene sets were further prioritized by Xsurv’s XGBoost machine learning algorithm to identify enriched drug targets in high-risk patients.

Download PDF

Version 1

posted

You are reading this latest preprint version

DRPPM-PATH-SURVEIOR: Plug-and-Play Survival Analysis of Pathway-level Signatures and Immune Components

Status:

Version 1

Abstract

Figures

1. Background

2. Implementation

3. Results

4. Conclusion

List Of Abbreviations

Declarations

References

Supplementary information

Supplementary Files

Status:

Version 1