pith. machine review for the scientific record. sign in

arxiv: 2604.05478 · v3 · submitted 2026-04-07 · 🧬 q-bio.GN · cs.LG

Recognition: 2 theorem links

· Lean Theorem

Transcriptomic Models for Immunotherapy Response Prediction Show Limited Cross-cohort Generalisability

Ahmadreza Argha, Amin Beheshti, Hamid Alinejad-Rokny, Lu Chen, Lucy Chhuo, Mehdi Hosseinzadeh, Nona Farbehi, Roohallah Alizadehsani, Thantrira Porntaveetusm, Youqiong Ye, Yuheng Liang

Authors on Pith no claims yet

Pith reviewed 2026-05-10 19:13 UTC · model grok-4.3

classification 🧬 q-bio.GN cs.LG
keywords transcriptomicsimmune checkpoint inhibitorsresponse predictioncross-cohort generalisabilityRNA-seqsingle-cell RNA-seqbiomarker consistencycancer immunotherapy
0
0 comments X

The pith

Transcriptomic models for predicting immune checkpoint inhibitor response show limited generalisability across independent patient cohorts.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper evaluates nine transcriptomic models that aim to forecast which cancer patients will respond to immune checkpoint inhibitor therapy. It applies these models to publicly available datasets that were not involved in their original training. Bulk RNA sequencing approaches perform at or near chance levels in most test groups, while single-cell RNA sequencing versions deliver only marginal gains. The biological pathways flagged by different models show little consistent overlap. This pattern suggests current transcriptomic predictors lack the stability required to support reliable treatment decisions in varied clinical settings.

Core claim

Benchmarking of five bulk RNA-seq models and four scRNA-seq models on unseen cohorts demonstrates modest predictive performance overall, with bulk models near chance level and scRNA-seq models showing only slight improvement, accompanied by sparse and non-reproducible immune-related pathway signals across models and datasets.

What carries the argument

Cross-cohort benchmarking of transcriptomic ICI response predictors on independent public datasets, exposing gaps in performance and biomarker consistency.

If this is right

  • Bulk RNA-seq models need substantial refinement to exceed near-chance prediction on new groups.
  • Single-cell models require targeted adjustments to convert marginal gains into reliable performance.
  • Greater emphasis on consistent immune-related biological signals would improve model reproducibility.
  • Standardised preprocessing and domain adaptation methods are required to enhance transfer across cohorts.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Transcriptomic signals alone may need supplementation with additional data types to overcome cohort-specific noise.
  • Unmeasured technical variations between studies could explain much of the observed inconsistency in biomarkers.
  • Focusing future models on core immune mechanisms shared across cohorts could raise transferability without larger datasets.

Load-bearing premise

The independent test datasets represent real-world clinical variation without batch effects, selection biases, or preprocessing differences that distort measured performance.

What would settle it

Observing that any of the nine models achieves high accuracy consistently across several additional unseen cohorts after minimal adaptation would challenge the limited generalisability conclusion.

Figures

Figures reproduced from arXiv: 2604.05478 by Ahmadreza Argha, Amin Beheshti, Hamid Alinejad-Rokny, Lu Chen, Lucy Chhuo, Mehdi Hosseinzadeh, Nona Farbehi, Roohallah Alizadehsani, Thantrira Porntaveetusm, Youqiong Ye, Yuheng Liang.

Figure 1
Figure 1. Figure 1: Conceptual overview of immune checkpoint inhibition and transcriptomics [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Architectural categories of transcriptomics [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Cross-dataset performance comparison of bulk and single-cell RNA-seq models. Macro F1, accuracy, and AUC for all evaluated bulk RNA-seq models within bulk and single cell RNA-seq dataset (Cho et al., Ribas et al., Poddubskaya et al., Gondal et al., Franken et al., Luoma et al., Reinstain et al.). As shown in [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Overlaps and interactions among significant biological pathways identified by [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Enrichment Map visualisation of pathway-level biological concepts across unseen cohorts. EnrichmentMap networks depict biological concepts identified by pathway￾level biomarkers for each dataset. Nodes represent enriched pathways, edges indicate gene-set overlap, and clusters denote groups of functionally related biological concepts. A. Cho et al.[45] (bulk RNA-seq), with node colours corresponding to bulk… view at source ↗
read the original abstract

Immune checkpoint inhibitors (ICIs) have transformed cancer therapy; yet substantial proportion of patients exhibit intrinsic or acquired resistance, making accurate pre-treatment response prediction a critical unmet need. Transcriptomics-based biomarkers derived from bulk and single-cell RNA sequencing (scRNA-seq) offer a promising avenue for capturing tumour-immune interactions, yet the cross-cohort generalisability of existing prediction models remains unclear.We systematically benchmark nine state-of-the-art transcriptomic ICI response predictors, five bulk RNA-seq-based models (COMPASS, IRNet, NetBio, IKCScore, and TNBC-ICI) and four scRNA-seq-based models (PRECISE, DeepGeneX, Tres and scCURE), using publicly available independent datasets unseen during model development. Overall, predictive performance was modest: bulk RNA-seq models performed at or near chance level across most cohorts, while scRNA-seq models showed only marginal improvements. Pathway-level analyses revealed sparse and inconsistent biomarker signals across models. Although scRNA-seq-based predictors converged on immune-related programs such as allograft rejection, bulk RNA-seq-based models exhibited little reproducible overlap. PRECISE and NetBio identified the most coherent immune-related themes, whereas IRNet predominantly captured metabolic pathways weakly aligned with ICI biology. Together, these findings demonstrate the limited cross-cohort robustness and biological consistency of current transcriptomic ICI prediction models, underscoring the need for improved domain adaptation, standardised preprocessing, and biologically grounded model design.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript systematically benchmarks nine transcriptomic ICI response predictors (five bulk RNA-seq models: COMPASS, IRNet, NetBio, IKCScore, TNBC-ICI; four scRNA-seq models: PRECISE, DeepGeneX, Tres, scCURE) on publicly available independent datasets. It reports modest overall performance, with bulk models at or near chance level across most cohorts and scRNA-seq models showing only marginal gains, alongside sparse and inconsistent pathway-level biomarker signals (e.g., immune-related themes in PRECISE and NetBio but metabolic focus in IRNet), concluding limited cross-cohort generalisability and calling for better domain adaptation and standardization.

Significance. If the empirical results hold, the work is significant as a comprehensive reference highlighting the robustness challenges facing current transcriptomic predictors in immuno-oncology. The multi-model, multi-cohort design and pathway consistency analysis provide concrete evidence that motivates improved methods; the reliance on public data is a reproducibility strength.

major comments (1)
  1. [Methods (independent datasets)] Methods section on independent datasets: the central claim that all evaluation cohorts are 'unseen during model development' rests on an assertion without a supporting cross-reference table or appendix that systematically compares the original training cohorts from the nine source papers against the test cohorts (e.g., potential overlaps with GSE78220 or IMvigor210). This verification is load-bearing for interpreting the 'near chance' performance as evidence of limited generalisability rather than an artifact of data leakage.
minor comments (2)
  1. [Abstract] Abstract: the description of 'modest' performance and 'sparse' signals would be clearer with explicit mention of the number of evaluation cohorts and the primary performance metric (e.g., AUC) used.
  2. [Results (pathway analyses)] Results (pathway analyses): the criteria for declaring 'inconsistent' or 'coherent' biomarker themes across models are not stated explicitly, making it hard to assess reproducibility of the biological findings.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback and positive evaluation of the study's significance. We address the single major comment below and will incorporate the requested clarification in the revised manuscript.

read point-by-point responses

  1. Referee: Methods section on independent datasets: the central claim that all evaluation cohorts are 'unseen during model development' rests on an assertion without a supporting cross-reference table or appendix that systematically compares the original training cohorts from the nine source papers against the test cohorts (e.g., potential overlaps with GSE78220 or IMvigor210). This verification is load-bearing for interpreting the 'near chance' performance as evidence of limited generalisability rather than an artifact of data leakage.

    Authors: We agree that an explicit cross-reference table would improve transparency and strengthen the manuscript. Although our evaluation cohorts were chosen after reviewing the training cohort descriptions in each of the nine original publications (COMPASS, IRNet, NetBio, IKCScore, TNBC-ICI, PRECISE, DeepGeneX, Tres, scCURE) and confirming no overlap with the test sets (including GSE78220 and IMvigor210), this verification was not documented in a dedicated table. In the revised version we will add Supplementary Table S1, which will list (i) the exact training cohorts and accession numbers reported in each source paper and (ii) the independent evaluation cohorts used here, with a clear statement of non-overlap. This addition will make the independence claim fully verifiable and support the interpretation of the observed performance as evidence of limited cross-cohort generalisability rather than data leakage. revision: yes

Circularity Check

0 steps flagged

No circularity: pure empirical benchmarking with no derivations or self-referential equations

full rationale

This is a systematic benchmarking study that evaluates nine pre-existing transcriptomic ICI response predictors on publicly available independent datasets. The abstract and structure contain no equations, fitted parameters, ansatzes, or derivation chains. Performance metrics (AUC, etc.) are computed directly from external test cohorts rather than being defined in terms of the models' own training data or prior outputs. Self-citations to the original model papers are standard and non-load-bearing for the central claim of limited generalisability. The skeptic concern about possible undetected cohort overlap is a validity issue for the evaluation design, not a circularity in any derivation step.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

This is an empirical benchmarking study with no new mathematical derivations or postulated entities. The central claim depends on standard assumptions in machine-learning evaluation regarding dataset independence and the validity of performance metrics.

axioms (1)
  • domain assumption The selected independent test cohorts are free from batch effects and representative of the target population for ICI response prediction.
    The evaluation of cross-cohort generalisability assumes that the chosen public datasets accurately reflect real-world variability without confounding factors.

pith-pipeline@v0.9.0 · 5617 in / 1442 out tokens · 75762 ms · 2026-05-10T19:13:44.418232+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

Reference graph

Works this paper leans on

79 extracted references · 79 canonical work pages

  1. [1]

    Network-based machine learning approach to predict immunotherapy response in cancer patients

    Kong J, Ha D, Lee J, Kim I, Park M, Im S-H, et al. Network-based machine learning approach to predict immunotherapy response in cancer patients. Nat Commun 2022;13:3703. https://doi.org/10.1038/s41467-022-31535-6

    work page doi:10.1038/s41467-022-31535-6 2022

  2. [2]

    IRnet: Immunotherapy response prediction using pathway knowledge-informed graph neural network

    Jiang Y, Immadi MS, Wang D, Zeng S, On Chan Y, Zhou J, et al. IRnet: Immunotherapy response prediction using pathway knowledge-informed graph neural network. J Adv Res 2024;72:319–31. https://doi.org/10.1016/j.jare.2024.07.036

    work page doi:10.1016/j.jare.2024.07.036 2024

  3. [3]

    Generalizable AI predicts immunotherapy outcomes across cancers and treatments

    Shen W, Nguyen TH, Li MM, Huang Y, Moon I, Nair N, et al. Generalizable AI predicts immunotherapy outcomes across cancers and treatments. medRxiv 2025:2025.05.01.25326820. https://doi.org/10.1101/2025.05.01.25326820

    work page doi:10.1101/2025.05.01.25326820 2025

  4. [4]

    Uncovering gene and cellular signatures of immune checkpoint response via machine learning and single-cell RNA-seq

    Pinhasi A, Yizhak K. Uncovering gene and cellular signatures of immune checkpoint response via machine learning and single-cell RNA-seq. Npj Precis Oncol 2025;9:95. https://doi.org/10.1038/s41698-025-00883-z

    work page doi:10.1038/s41698-025-00883-z 2025

  5. [5]

    Deep neural network modeling identifies biomarkers of response to immune-checkpoint therapy

    Kang Y, Vijay S, Gujral TS. Deep neural network modeling identifies biomarkers of response to immune-checkpoint therapy. iScience 2022;25. https://doi.org/10.1016/j.isci.2022.104228

    work page doi:10.1016/j.isci.2022.104228 2022

  6. [6]

    A T cell resilience model associated with response to immunotherapy in multiple tumor types

    Zhang Y, Vu T, Palmer DC, Kishton RJ, Gong L, Huang J, et al. A T cell resilience model associated with response to immunotherapy in multiple tumor types. Nat Med 2022;28:1421–31. https://doi.org/10.1038/s41591-022-01799-y

    work page doi:10.1038/s41591-022-01799-y 2022

  7. [7]

    Immune checkpoint signaling and cancer immunotherapy

    He X, Xu C. Immune checkpoint signaling and cancer immunotherapy. Cell Res 2020;30:660–9. https://doi.org/10.1038/s41422-020-0343-4

    work page doi:10.1038/s41422-020-0343-4 2020

  8. [8]

    PD-1 and PD-L1 in cancer immunotherapy: clinical implications and future considerations

    Jiang Y, Chen M, Nie H, Yuan Y. PD-1 and PD-L1 in cancer immunotherapy: clinical implications and future considerations. Hum Vaccines Immunother 2019;15:1111–22. https://doi.org/10.1080/21645515.2019.1571892

    work page doi:10.1080/21645515.2019.1571892 2019

  9. [9]

    Role of Immunotherapy in the Treatment of Cancer: A Systematic Review

    Ling SP, Ming LC, Dhaliwal JS, Gupta M, Ardianto C, Goh KW, et al. Role of Immunotherapy in the Treatment of Cancer: A Systematic Review. Cancers 2022;14:5205. https://doi.org/10.3390/cancers14215205

    work page doi:10.3390/cancers14215205 2022

  10. [10]

    Adaptive immune resistance: How cancer protects from immune attack

    Ribas A. Adaptive immune resistance: How cancer protects from immune attack. Cancer Discov 2015;5:915–9. https://doi.org/10.1158/2159-8290.CD-15-0563

    work page doi:10.1158/2159-8290.cd-15-0563 2015

  11. [11]

    Immunotherapy in Melanoma: Recent Advances and Future Directions

    Knight A, Karapetyan L, Kirkwood JM. Immunotherapy in Melanoma: Recent Advances and Future Directions. Cancers 2023;15:1106. https://doi.org/10.3390/cancers15041106

    work page doi:10.3390/cancers15041106 2023

  12. [12]

    PD-1/PD-L1 immune checkpoint blockade in breast cancer: research insights and sensitization strategies

    Jin M, Fang J, Peng J, Wang X, Xing P, Jia K, et al. PD-1/PD-L1 immune checkpoint blockade in breast cancer: research insights and sensitization strategies. Mol Cancer 2024;23:266. https://doi.org/10.1186/s12943-024-02176-8

    work page doi:10.1186/s12943-024-02176-8 2024

  13. [13]

    Primary, Adaptive and Acquired Resistance to Cancer Immunotherapy

    Sharma P, Hu-Lieskovan S, Wargo JA, Ribas A. Primary, Adaptive and Acquired Resistance to Cancer Immunotherapy. Cell 2017;168:707. https://doi.org/10.1016/j.cell.2017.01.017

    work page doi:10.1016/j.cell.2017.01.017 2017

  14. [14]

    Cell-Intrinsic Barriers of T Cell-Based Immunotherapy

    Ghoneim HE, Zamora AE, Thomas PG, Youngblood BA. Cell-Intrinsic Barriers of T Cell-Based Immunotherapy. Trends Mol Med 2016;22:1000–11. https://doi.org/10.1016/j.molmed.2016.10.002

    work page doi:10.1016/j.molmed.2016.10.002 2016

  15. [15]

    Immune evasion in cancer: mechanisms and cutting-edge therapeutic approaches

    Tufail M, Jiang C-H, Li N. Immune evasion in cancer: mechanisms and cutting-edge therapeutic approaches. Signal Transduct Target Ther 2025;10:227. https://doi.org/10.1038/s41392-025-02280-1. 38

    work page doi:10.1038/s41392-025-02280-1 2025

  16. [16]

    Effect of chemotherapy alone or combined with immunotherapy for locally advanced or metastatic genitourinary small cell carcinoma: a real-world retrospective study

    Huang R, Chen M, Li H, An X, Xue C, Hu A, et al. Effect of chemotherapy alone or combined with immunotherapy for locally advanced or metastatic genitourinary small cell carcinoma: a real-world retrospective study. BMC Cancer 2023;23:1002. https://doi.org/10.1186/s12885-023-11473-2

    work page doi:10.1186/s12885-023-11473-2 2023

  17. [17]

    The interaction of innate immune and adaptive immune system

    Wang R, Lan C, Benlagha K, Camara NOS, Miller H, Kubo M, et al. The interaction of innate immune and adaptive immune system. MedComm 2024;5:e714. https://doi.org/10.1002/mco2.714

    work page doi:10.1002/mco2.714 2024

  18. [18]

    MS, et al

    Niemeijer AN, Leung D, Huisman MC, Bahce I, Hoekstra OS, van Dongen G a. MS, et al. Whole body PD-1 and PD-L1 positron emission tomography in patients with non- small-cell lung cancer. Nat Commun 2018;9:4664. https://doi.org/10.1038/s41467-018- 07131-y

    work page doi:10.1038/s41467-018- 2018

  19. [19]

    High tumor mutation burden fails to predict immune checkpoint blockade response across all cancer types

    McGrail DJ, Pilié PG, Rashid NU, Voorwerk L, Slagter M, Kok M, et al. High tumor mutation burden fails to predict immune checkpoint blockade response across all cancer types. Ann Oncol 2021;32:661–72. https://doi.org/10.1016/j.annonc.2021.02.006

    work page doi:10.1016/j.annonc.2021.02.006 2021

  20. [20]

    Microsatellite Instability and Immune Response: From Microenvironment Features to Therapeutic Actionability—Lessons from Colorectal Cancer

    Greco L, Rubbino F, Dal Buono A, Laghi L. Microsatellite Instability and Immune Response: From Microenvironment Features to Therapeutic Actionability—Lessons from Colorectal Cancer. Genes 2023;14:1169. https://doi.org/10.3390/genes14061169

    work page doi:10.3390/genes14061169 2023

  21. [21]

    Neoantigens: promising targets for cancer therapy

    Xie N, Shen G, Gao W, Huang Z, Huang C, Fu L. Neoantigens: promising targets for cancer therapy. Signal Transduct Target Ther 2023;8:9. https://doi.org/10.1038/s41392- 022-01270-x

    work page doi:10.1038/s41392- 2023

  22. [22]

    Immune correlates and mechanisms of TIL therapy efficacy: current insights and knowledge gaps

    Navarro Rodrigo B, Ortiz Miranda Y, Corria-Osorio J, Coukos G, Harari A. Immune correlates and mechanisms of TIL therapy efficacy: current insights and knowledge gaps. Trends Cancer 2025;11:993–1004. https://doi.org/10.1016/j.trecan.2025.08.002

    work page doi:10.1016/j.trecan.2025.08.002 2025

  23. [23]

    Advances and prospects of biomarkers for immune checkpoint inhibitors

    Yamaguchi H, Hsu J-M, Sun L, Wang S-C, Hung M-C. Advances and prospects of biomarkers for immune checkpoint inhibitors. Cell Rep Med 2024;5:101621. https://doi.org/10.1016/j.xcrm.2024.101621

    work page doi:10.1016/j.xcrm.2024.101621 2024

  24. [24]

    Microsatellite- Stable Tumors with High Mutational Burden Benefit from Immunotherapy

    Goodman AM, Sokol ES, Frampton GM, Lippman SM, Kurzrock R. Microsatellite- Stable Tumors with High Mutational Burden Benefit from Immunotherapy. Cancer Immunol Res 2019;7:1570–3. https://doi.org/10.1158/2326-6066.CIR-19-0149

    work page doi:10.1158/2326-6066.cir-19-0149 2019

  25. [25]

    TMB as a predictive biomarker for ICI response in TNBC: current evidence and future directions for augmented anti-tumor responses

    Das R, Deb S, Suresh PK. TMB as a predictive biomarker for ICI response in TNBC: current evidence and future directions for augmented anti-tumor responses. Clin Exp Med 2025;26:25. https://doi.org/10.1007/s10238-025-01892-9

    work page doi:10.1007/s10238-025-01892-9 2025

  26. [26]

    Transcriptomics technologies

    Lowe R, Shirley N, Bleackley M, Dolan S, Shafee T. Transcriptomics technologies. PLoS Comput Biol 2017;13:e1005457. https://doi.org/10.1371/journal.pcbi.1005457

    work page doi:10.1371/journal.pcbi.1005457 2017

  27. [27]

    Conserved interferon-γ signaling drives clinical response to immune checkpoint blockade therapy in melanoma

    Grasso CS, Tsoi J, Onyshchenko M, Abril-Rodriguez G, Ross-Macdonald P, Wind- Rotolo M, et al. Conserved interferon-γ signaling drives clinical response to immune checkpoint blockade therapy in melanoma. Cancer Cell 2020;38:500-515.e3. https://doi.org/10.1016/j.ccell.2020.08.005

    work page doi:10.1016/j.ccell.2020.08.005 2020

  28. [28]

    Bulk RNA sequencing reveals the comprehensive genetic characteristics of human cord blood-derived natural killer cells

    Morimoto T, Nakazawa T, Maeoka R, Matsuda R, Nakamura M, Nishimura F, et al. Bulk RNA sequencing reveals the comprehensive genetic characteristics of human cord blood-derived natural killer cells. Regen Ther 2024;25:367–76. https://doi.org/10.1016/j.reth.2024.02.002

    work page doi:10.1016/j.reth.2024.02.002 2024

  29. [29]

    Integrated analysis of single- cell and bulk RNA sequencing data reveals a myeloid cell-related regulon predicting neoadjuvant immunotherapy response across cancers

    Liu H, Sima X, Xiao B, Gulizeba H, Zhao S, Zhou T, et al. Integrated analysis of single- cell and bulk RNA sequencing data reveals a myeloid cell-related regulon predicting neoadjuvant immunotherapy response across cancers. J Transl Med 2024;22:486. https://doi.org/10.1186/s12967-024-05123-9

    work page doi:10.1186/s12967-024-05123-9 2024

  30. [30]

    Ott PA, Bang Y-J, Piha-Paul SA, Razak ARA, Bennouna J, Soria J-C, et al. T-Cell– Inflamed Gene-Expression Profile, Programmed Death Ligand 1 Expression, and Tumor Mutational Burden Predict Efficacy in Patients Treated With Pembrolizumab Across 20 39 Cancers: KEYNOTE-028. J Clin Oncol 2019;37:318–27. https://doi.org/10.1200/JCO.2018.78.2276

    work page doi:10.1200/jco.2018.78.2276 2019

  31. [31]

    Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response

    Jiang P, Gu S, Pan D, Fu J, Sahu A, Hu X, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nat Med 2018;24:1550–8. https://doi.org/10.1038/s41591-018-0136-1

    work page doi:10.1038/s41591-018-0136-1 2018

  32. [32]

    Robust prediction of response to immune checkpoint blockade therapy in metastatic melanoma

    Auslander N, Zhang G, Lee JS, Frederick DT, Miao B, Moll T, et al. Robust prediction of response to immune checkpoint blockade therapy in metastatic melanoma. Nat Med 2018;24:1545–9. https://doi.org/10.1038/s41591-018-0157-9

    work page doi:10.1038/s41591-018-0157-9 2018

  33. [33]

    Single-cell RNA sequencing to explore immune cell heterogeneity

    Papalexi E, Satija R. Single-cell RNA sequencing to explore immune cell heterogeneity. Nat Rev Immunol 2018;18:35–45. https://doi.org/10.1038/nri.2017.76

    work page doi:10.1038/nri.2017.76 2018

  34. [34]

    Addressing fairness in artificial intelligence for medical imaging.Nature Communications, 13:4581, 2022

    Tietscher S, Wagner J, Anzeneder T, Langwieder C, Rees M, Sobottka B, et al. A comprehensive single-cell map of T cell exhaustion-associated immune environments in human breast cancer. Nat Commun 2023;14:98. https://doi.org/10.1038/s41467-022- 35238-w

    work page doi:10.1038/s41467-022- 2023

  35. [35]

    Huang L, Lou N, Xie T, Tang L, Han X, Shi Y. Identification of an antigen-presenting cells/T/NK cells-related gene signature to predict prognosis and CTSL to predict immunotherapeutic response for lung adenocarcinoma: an integrated analysis of bulk and single-cell RNA sequencing. Cancer Immunol Immunother CII 2023;72:3259–77. https://doi.org/10.1007/s0026...

    work page doi:10.1007/s00262-023-03485-5 2023

  36. [36]

    Tumor microenvironment assessment-based signatures for predicting response to immunotherapy in non-small cell lung cancer

    Wu J, Wang Y, Huang Z, Wu J, Sun H, Zhou R, et al. Tumor microenvironment assessment-based signatures for predicting response to immunotherapy in non-small cell lung cancer. iScience 2024;27:111340. https://doi.org/10.1016/j.isci.2024.111340

    work page doi:10.1016/j.isci.2024.111340 2024

  37. [37]

    Clinically oriented prediction of patient response to targeted and immunotherapies from the tumor transcriptome

    Dinstag G, Shulman ED, Elis E, Ben-Zvi DS, Tirosh O, Maimon E, et al. Clinically oriented prediction of patient response to targeted and immunotherapies from the tumor transcriptome. Med 2023;4:15-30.e8. https://doi.org/10.1016/j.medj.2022.11.001

    work page doi:10.1016/j.medj.2022.11.001 2023

  38. [38]

    Construction and validation of a gene expression classifier to predict immunotherapy response in primary triple-negative breast cancer

    Ensenyat-Mendez M, Orozco JIJ, Llinàs-Arias P, Íñiguez-Muñoz S, Baker JL, Salomon MP, et al. Construction and validation of a gene expression classifier to predict immunotherapy response in primary triple-negative breast cancer. Commun Med 2023;3:93. https://doi.org/10.1038/s43856-023-00311-y

    work page doi:10.1038/s43856-023-00311-y 2023

  39. [39]

    scCURE identifies cell types responding to immunotherapy and enables outcome prediction

    Zou X, Liu Y, Wang M, Zou J, Shi Y, Su X, et al. scCURE identifies cell types responding to immunotherapy and enables outcome prediction. Cell Rep Methods 2023;3:100643. https://doi.org/10.1016/j.crmeth.2023.100643

    work page doi:10.1016/j.crmeth.2023.100643 2023

  40. [40]

    The Cancer Genome Atlas Pan-Cancer analysis project

    Weinstein JN, Collisson EA, Mills GB, Shaw KRM, Ozenberger BA, Ellrott K, et al. The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet 2013;45:1113–20. https://doi.org/10.1038/ng.2764

    work page doi:10.1038/ng.2764 2013

  41. [41]

    A Cancer Cell Program Promotes T Cell Exclusion and Resistance to Checkpoint Blockade

    Jerby-Arnon L, Shah P, Cuoco MS, Rodman C, Su M-J, Melms JC, et al. A Cancer Cell Program Promotes T Cell Exclusion and Resistance to Checkpoint Blockade. Cell 2018;175:984-997.e24. https://doi.org/10.1016/j.cell.2018.09.006

    work page doi:10.1016/j.cell.2018.09.006 2018

  42. [42]

    A TCF4-dependent gene regulatory network confers resistance to immunotherapy in melanoma

    Pozniak J, Pedri D, Landeloos E, Herck YV, Antoranz A, Vanwynsberghe L, et al. A TCF4-dependent gene regulatory network confers resistance to immunotherapy in melanoma. Cell 2024;187:166-183.e25. https://doi.org/10.1016/j.cell.2023.11.037

    work page doi:10.1016/j.cell.2023.11.037 2024

  43. [43]

    Defining T cell states associated with response to checkpoint immunotherapy in melanoma

    Sade-Feldman M, Yizhak K, Bjorgaard SL, Ray JP, de Boer CG, Jenkins RW, et al. Defining T cell states associated with response to checkpoint immunotherapy in melanoma. Cell 2018;175:998-1013.e20. https://doi.org/10.1016/j.cell.2018.10.038

    work page doi:10.1016/j.cell.2018.10.038 2018

  44. [44]

    Genomic and Transcriptomic Predictors of Response to Immune Checkpoint Inhibitors in Melanoma Patients: A Machine Learning Approach

    Ahmed YB, Al-Bzour AN, Ababneh OE, Abushukair HM, Saeed A. Genomic and Transcriptomic Predictors of Response to Immune Checkpoint Inhibitors in Melanoma Patients: A Machine Learning Approach. Cancers 2022;14:5605. https://doi.org/10.3390/cancers14225605

    work page doi:10.3390/cancers14225605 2022

  45. [45]

    Genome-wide identification of differentially methylated promoters and enhancers associated with 40 response to anti-PD-1 therapy in non-small cell lung cancer

    Cho J-W, Hong MH, Ha S-J, Kim Y-J, Cho BC, Lee I, et al. Genome-wide identification of differentially methylated promoters and enhancers associated with 40 response to anti-PD-1 therapy in non-small cell lung cancer. Exp Mol Med 2020;52:1550–63. https://doi.org/10.1038/s12276-020-00493-8

    work page doi:10.1038/s12276-020-00493-8 2020

  46. [46]

    PD-L1 blockade in combination with inhibition of MAPK oncogenic signaling in patients with advanced melanoma

    Ribas A, Algazi A, Ascierto PA, Butler MO, Chandra S, Gordon M, et al. PD-L1 blockade in combination with inhibition of MAPK oncogenic signaling in patients with advanced melanoma. Nat Commun 2020;11:6262. https://doi.org/10.1038/s41467-020- 19810-w

    work page doi:10.1038/s41467-020- 2020

  47. [47]

    Poddubskaya E, Suntsova M, Lyadova M, Luppov D, Guryanova A, Lyadov V, et al. Biomarkers of success of anti-PD-(L)1 immunotherapy for non-small cell lung cancer derived from RNA- and whole-exome sequencing: results of a prospective observational study on a cohort of 85 patients. Front Immunol 2024;15. https://doi.org/10.3389/fimmu.2024.1493877

    work page doi:10.3389/fimmu.2024.1493877 2024

  48. [48]

    Integrated cancer cell-specific single-cell RNA-seq datasets of immune checkpoint blockade-treated patients

    Gondal MN, Cieslik M, Chinnaiyan AM. Integrated cancer cell-specific single-cell RNA-seq datasets of immune checkpoint blockade-treated patients. Sci Data 2025;12:139. https://doi.org/10.1038/s41597-025-04381-6

    work page doi:10.1038/s41597-025-04381-6 2025

  49. [49]

    CD4+ T cell activation distinguishes response to anti-PD-L1+anti-CTLA4 therapy from anti-PD-L1 monotherapy

    Franken A, Bila M, Mechels A, Kint S, Dessel JV, Pomella V, et al. CD4+ T cell activation distinguishes response to anti-PD-L1+anti-CTLA4 therapy from anti-PD-L1 monotherapy. Immunity 2024;57:541-558.e7. https://doi.org/10.1016/j.immuni.2024.02.007

    work page doi:10.1016/j.immuni.2024.02.007 2024

  50. [50]

    Tissue-resident memory and circulating T cells are early responders to pre-surgical cancer immunotherapy

    Luoma AM, Suo S, Wang Y, Gunasti L, Porter CBM, Nabilsi N, et al. Tissue-resident memory and circulating T cells are early responders to pre-surgical cancer immunotherapy. Cell 2022;185:2918-2935.e29. https://doi.org/10.1016/j.cell.2022.06.018

    work page doi:10.1016/j.cell.2022.06.018 2022

  51. [51]

    Pre-existing skin-resident CD8 and γδ T cell circuits mediate immune response in Merkel cell carcinoma and predict immunotherapy efficacy

    Reinstein ZZ, Zhang Y, Ospina OE, Nichols MD, Chu VA, de Mingo Pulido A, et al. Pre-existing skin-resident CD8 and γδ T cell circuits mediate immune response in Merkel cell carcinoma and predict immunotherapy efficacy. Cancer Discov 2024;14:1631–52. https://doi.org/10.1158/2159-8290.CD-23-0798

    work page doi:10.1158/2159-8290.cd-23-0798 2024

  52. [52]

    et al.: BEHRT: Transformer for electronic health records

    Hołyńska-Iwan I, Szewczyk-Golec K. Analysis of changes in sodium and chloride ion transport in the skin. Sci Rep 2020;10:18094. https://doi.org/10.1038/s41598-020- 75275-3

    work page doi:10.1038/s41598-020- 2020

  53. [53]

    Regionally specific levels and patterns of keratin 8 expression in the mouse embryo visceral endoderm emerge upon anterior-posterior axis determination

    Despin-Guitard E, Quenec’Hdu R, Nahaboo W, Schwarz N, Leube RE, Chazaud C, et al. Regionally specific levels and patterns of keratin 8 expression in the mouse embryo visceral endoderm emerge upon anterior-posterior axis determination. Front Cell Dev Biol 2022;10. https://doi.org/10.3389/fcell.2022.1037041

    work page doi:10.3389/fcell.2022.1037041 2022

  54. [54]

    NCBI RefSeq: reference sequence standards through 25 years of curation and annotation

    Goldfarb T, Kodali VK, Pujar S, Brover V, Robbertse B, Farrell CM, et al. NCBI RefSeq: reference sequence standards through 25 years of curation and annotation. Nucleic Acids Res 2024;53:D243. https://doi.org/10.1093/nar/gkae1038

    work page doi:10.1093/nar/gkae1038 2024

  55. [55]

    Granzyme family acts as a predict biomarker in cutaneous melanoma and indicates more benefit from anti-PD-1 immunotherapy

    Wu X, Wang X, Zhao Y, Li K, Yu B, Zhang J. Granzyme family acts as a predict biomarker in cutaneous melanoma and indicates more benefit from anti-PD-1 immunotherapy. Int J Med Sci 2021;18:1657–69. https://doi.org/10.7150/ijms.54747

    work page doi:10.7150/ijms.54747 2021

  56. [56]

    Nonsense-mediated mRNA decay inhibits TRAF6-dependent anti-tumor immunity in colorectal cancer

    Wang Y, Wang Z, Wang C, Wu Y, Li J, Feng L, et al. Nonsense-mediated mRNA decay inhibits TRAF6-dependent anti-tumor immunity in colorectal cancer. Cell Rep Med 2025;6:102463. https://doi.org/10.1016/j.xcrm.2025.102463

    work page doi:10.1016/j.xcrm.2025.102463 2025

  57. [57]

    Tertiary lymphoid structures and B cells: An intratumoral immunity cycle

    Fridman WH, Meylan M, Pupier G, Calvez A, Hernandez I, Sautès-Fridman C. Tertiary lymphoid structures and B cells: An intratumoral immunity cycle. Immunity 2023;56:2254–69. https://doi.org/10.1016/j.immuni.2023.08.009

    work page doi:10.1016/j.immuni.2023.08.009 2023

  58. [58]

    Immunotherapy for EGFR-mutant advanced non-small- cell lung cancer: Current status, possible mechanisms and application prospects

    Shi C, Wang Y, Xue J, Zhou X. Immunotherapy for EGFR-mutant advanced non-small- cell lung cancer: Current status, possible mechanisms and application prospects. Front Immunol 2022;13:940288. https://doi.org/10.3389/fimmu.2022.940288. 41

    work page doi:10.3389/fimmu.2022.940288 2022

  59. [59]

    Therapeutic Strategies Targeting Aerobic Glycolysis in Cancer and Dynamic Monitoring of Associated Metabolites

    Hu M, Zheng K, Zhang L, Kan Y, Zhao J, Chen D. Therapeutic Strategies Targeting Aerobic Glycolysis in Cancer and Dynamic Monitoring of Associated Metabolites. Cells 2025;14:1288. https://doi.org/10.3390/cells14161288

    work page doi:10.3390/cells14161288 2025

  60. [60]

    Targeting collagen to optimize cancer immunotherapy

    Wang Y, Zhang F, Qian Z, Jiang Y, Wu D, Liu L, et al. Targeting collagen to optimize cancer immunotherapy. Exp Hematol Oncol 2025;14:101. https://doi.org/10.1186/s40164-025-00691-y

    work page doi:10.1186/s40164-025-00691-y 2025

  61. [61]

    Cancer Immunity: Lessons From Infectious Diseases

    Trinchieri G. Cancer Immunity: Lessons From Infectious Diseases. J Infect Dis 2015;212:S67–73. https://doi.org/10.1093/infdis/jiv070

    work page doi:10.1093/infdis/jiv070 2015

  62. [62]

    Metabolic mechanisms of immunotherapy resistance

    Cabezón-Gutiérrez L, Palka-Kotlowska M, Custodio-Cabello S, Chacón-Ovejero B, Pacheco-Barcia V. Metabolic mechanisms of immunotherapy resistance. Explor Target Anti-Tumor Ther 2025;6:1002297. https://doi.org/10.37349/etat.2025.1002297

    work page doi:10.37349/etat.2025.1002297 2025

  63. [63]

    RhoGTPases and inflammasomes: Guardians of effector-triggered immunity

    Dufies O, Boyer L. RhoGTPases and inflammasomes: Guardians of effector-triggered immunity. PLoS Pathog 2021;17:e1009504. https://doi.org/10.1371/journal.ppat.1009504

    work page doi:10.1371/journal.ppat.1009504 2021

  64. [64]

    RNA Binding Proteins (RBPs) and their role in DNA Damage and radiation response in cancer

    Mehta M, Raguraman R, Ramesh R, Munshi A. RNA Binding Proteins (RBPs) and their role in DNA Damage and radiation response in cancer. Adv Drug Deliv Rev 2022;191:114569. https://doi.org/10.1016/j.addr.2022.114569

    work page doi:10.1016/j.addr.2022.114569 2022

  65. [65]

    Early signaling pathways in virus-infected cells

    Bonhomme D, Poirier EZ. Early signaling pathways in virus-infected cells. Curr Opin Virol 2024;66:101411. https://doi.org/10.1016/j.coviro.2024.101411

    work page doi:10.1016/j.coviro.2024.101411 2024

  66. [66]

    Immunometabolism – The Role of Branched-Chain Amino Acids

    Yahsi B, Gunaydin G. Immunometabolism – The Role of Branched-Chain Amino Acids. Front Immunol 2022;13:886822. https://doi.org/10.3389/fimmu.2022.886822

    work page doi:10.3389/fimmu.2022.886822 2022

  67. [67]

    Chemokines in the Landscape of Cancer Immunotherapy: How They and Their Receptors Can Be Used to Turn Cold Tumors into Hot Ones? Cancers 2021;13:6317

    Karin N. Chemokines in the Landscape of Cancer Immunotherapy: How They and Their Receptors Can Be Used to Turn Cold Tumors into Hot Ones? Cancers 2021;13:6317. https://doi.org/10.3390/cancers13246317

    work page doi:10.3390/cancers13246317 2021

  68. [68]

    Deciphering the deterministic role of TCR signaling in T cell fate determination

    Qin Z, Xu T. Deciphering the deterministic role of TCR signaling in T cell fate determination. Front Immunol 2025;16:1562248. https://doi.org/10.3389/fimmu.2025.1562248

    work page doi:10.3389/fimmu.2025.1562248 2025

  69. [69]

    Antigen presentation and tumor immunogenicity in cancer immunotherapy response prediction

    Wang S, He Z, Wang X, Li H, Liu X-S. Antigen presentation and tumor immunogenicity in cancer immunotherapy response prediction. eLife n.d.;8:e49020. https://doi.org/10.7554/eLife.49020

    work page doi:10.7554/elife.49020

  70. [70]

    NCBI GEO: archive for functional genomics data sets—update

    Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets—update. Nucleic Acids Res 2013;41:D991–5. https://doi.org/10.1093/nar/gks1193

    work page doi:10.1093/nar/gks1193 2013

  71. [71]

    Ajay Nadig, Joseph M Replogle, Angela N Pogson, et al

    Mudge JM, Carbonell-Sala S, Diekhans M, Martinez JG, Hunt T, Jungreis I, et al. GENCODE 2025: reference gene annotation for human and mouse. Nucleic Acids Res 2025;53:D966–75. https://doi.org/10.1093/nar/gkae1078

    work page doi:10.1093/nar/gkae1078 2025

  72. [72]

    New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1)

    Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–47. https://doi.org/10.1016/j.ejca.2008.10.026

    work page doi:10.1016/j.ejca.2008.10.026 2009

  73. [73]

    Neoadjuvant immune checkpoint blockade in high-risk resectable melanoma

    Amaria RN, Reddy SM, Tawbi HA, Davies MA, Ross MI, Glitza IC, et al. Neoadjuvant immune checkpoint blockade in high-risk resectable melanoma. Nat Med 2018;24:1649–54. https://doi.org/10.1038/s41591-018-0197-1

    work page doi:10.1038/s41591-018-0197-1 2018

  74. [74]

    Enrichr: a comprehensive gene set enrichment analysis web server 2016 update

    Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 2016;44:W90–7. https://doi.org/10.1093/nar/gkw377

    work page doi:10.1093/nar/gkw377 2016

  75. [75]

    WikiPathways 2024: next generation pathway database

    Agrawal A, Balcı H, Hanspers K, Coort SL, Martens M, Slenter DN, et al. WikiPathways 2024: next generation pathway database. Nucleic Acids Res 2024;52:D679–89. https://doi.org/10.1093/nar/gkad960

    work page doi:10.1093/nar/gkad960 2024

  76. [76]

    KEGG: Kyoto Encyclopedia of Genes and Genomes

    Kanehisa M, Goto S. KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res 2000;28:27–30. https://doi.org/10.1093/nar/28.1.27. 42

    work page doi:10.1093/nar/28.1.27 2000

  77. [77]

    The Reactome Pathway Knowledgebase 2024

    Milacic M, Beavers D, Conley P, Gong C, Gillespie M, Griss J, et al. The Reactome Pathway Knowledgebase 2024. Nucleic Acids Res 2024;52:D672–8. https://doi.org/10.1093/nar/gkad1025

    work page doi:10.1093/nar/gkad1025 2024

  78. [78]

    The Molecular Signatures Database (MSigDB) hallmark gene set collection

    Liberzon A, Birger C, Thorvaldsdóttir H, Ghandi M, Mesirov JP, Tamayo P. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 2015;1:417–25. https://doi.org/10.1016/j.cels.2015.12.004

    work page doi:10.1016/j.cels.2015.12.004 2015

  79. [79]

    Enrichment Map: A Network- Based Method for Gene-Set Enrichment Visualization and Interpretation

    Merico D, Isserlin R, Stueker O, Emili A, Bader GD. Enrichment Map: A Network- Based Method for Gene-Set Enrichment Visualization and Interpretation. PLOS ONE 2010;5:e13984. https://doi.org/10.1371/journal.pone.0013984. 43 Supplementary Materials Supplementary Table S1. Model inclusion-exclusion criteria. Models were included if they were published betwee...

    work page doi:10.1371/journal.pone.0013984 2010