| evidence-b2-time-cart-1 | KF1: Spatial localization of lymphocytes provides information not captured by simple immune-cell abundance. | The review discusses immune-excluded and inflamed TIME patterns and spatial proximity from published studies, without constructing or validating a new spatial test. | Spatial localization of lymphocytes provides information not captured by simple immune-cell abundance. Immune patterns: Immune-excluded, immune-inflamed and TLS-containing contexts; Spatial dimensions: Stroma, tumor core and invasive margin; Evidence format: Narrative examples from cited observational research | The review discusses immune-excluded and inflamed TIME patterns and spatial proximity from published studies, without constructing or validating a new spatial test. | The review discusses immune-excluded and inflamed TIME patterns and spatial proximity from published studies, without constructing or validating a new spatial test. | PMC11201788: Classification of the TIME (Sec3), Spatial architecture of the TIME (Sec4), Distribution of immune cells and immune checkpoints (Sec5), Distance between immune cells and tumor/immune cells (Sec6); Figures 1–2. | KF1 |
|---|
| evidence-b2-time-cart-2 | KF2: Multiple host- and tumor-derived molecular layers can be used to characterize immune heterogeneity. | Genomics, transcriptomics, proteomics and microbiome observations are complementary literature domains; integration does not by itself establish intervention efficacy. | Multiple host- and tumor-derived molecular layers can be used to characterize immune heterogeneity. Molecular sources: Host genetics, tumor genetics/epigenetics, RNA profiles, proteomics, microbiome; Spatial dimension: Histology, cellular distribution and intercellular context; Reporting nature: Literature synthesis; no newly generated study cohort | Genomics, transcriptomics, proteomics and microbiome observations are complementary literature domains; integration does not by itself establish intervention efficacy. | Genomics, transcriptomics, proteomics and microbiome observations are complementary literature domains; integration does not by itself establish intervention efficacy. | PMC11201788: Multi-omics characteristics shape TIME profiles (Sec7); Hostogenetics (Sec8); Genetics and epigenetics (Sec9); Transcriptome (Sec10); Proteomics (Sec11); Microbiome (Sec12); Figure 3. | KF2 |
|---|
| evidence-b2-time-cart-3 | KF3: The review groups potential CAR-T improvements into antigen selection and cellular-state or trafficking strategies. | These are research design considerations illustrated by prior reports and are not head-to-head efficacy estimates from the review. | The review groups potential CAR-T improvements into antigen selection and cellular-state or trafficking strategies. Design axis: Target-antigen selection including neoantigen candidates; Cell-intrinsic axis: Metabolic maintenance and reduced exhaustion; Microenvironment axis: Resistance to suppressive cells/cytokines and improved infiltration | These are research design considerations illustrated by prior reports and are not head-to-head efficacy estimates from the review. | These are research design considerations illustrated by prior reports and are not head-to-head efficacy estimates from the review. | PMC11201788: Advancing the efficacy of CAR-T immunotherapy (Sec13); Optimize target antigen choice (Sec14); Maintain T cell metabolic activity (Sec15); Resist immunosuppression (Sec16–18); Enhance infiltration (Sec19); Prevent exhaustion (Sec20); Figure 4. | KF3 |
|---|
| evidence-b2-time-cart-4 | KF4: Machine-learning approaches are discussed as tools to analyze spatial TIME structure and predict candidate antigens or therapy response. | The article reviews computational research, not a newly trained and clinically validated classifier on an original patient sample. | Machine-learning approaches are discussed as tools to analyze spatial TIME structure and predict candidate antigens or therapy response. Applications: Pathology image spatial analysis; antigen/MHC binding prioritization; treatment response modeling; Input caveat: MHC affinity alone is not evidence that a peptide is immunogenic; Evidence boundary: Proposed or cited models need prospective validation in defined clinical settings | The article reviews computational research, not a newly trained and clinically validated classifier on an original patient sample. | The article reviews computational research, not a newly trained and clinically validated classifier on an original patient sample. | PMC11201788: ML and TIME (Sec22); CAR construction (Sec24); Neoantigen (Sec25); ML and prediction of immunotherapy response (Sec26). | KF4 |
|---|
| evidence-b2-time-cart-5 | KF5: The review identifies barriers to carrying omics-informed CAR-T strategies into clinical solid-tumor use. | Scientific possibilities remain conditional on tumor type, product design, toxicity, patient selection and independent evidence. | The review identifies barriers to carrying omics-informed CAR-T strategies into clinical solid-tumor use. Unresolved themes: Heterogeneous TIME, immune suppression, trafficking and cellular exhaustion; Clinical questions: Safety, reproducibility, patient benefit and scalable integrated analysis; Article time scope: Review perspective published 2024 | Scientific possibilities remain conditional on tumor type, product design, toxicity, patient selection and independent evidence. | Scientific possibilities remain conditional on tumor type, product design, toxicity, patient selection and independent evidence. | PMC11201788: Perspective and conclusion (Sec27); Figure 4 and cited studies. | KF5 |
|---|