Learned immune phenotypes associate with lung cancer survival
Source check: Source checked. Not independent scientific confirmation.
Finding: Unsupervised clustering identified two lung-adenocarcinoma immune-expression patterns, and a random forest assigned tumors in additional cohorts. The more immune-infiltrated pattern was associated with better overall survival, including in external gene-expression datasets. This observational classification does not establish treatment response or a causal protective effect of immune infiltration; immunotherapy outcomes were not available for validation.
Technology: Consensus unsupervised clustering; random forest
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Unsupervised clustering learns infiltration patterns and a random forest transfers those patterns to additional cohorts for clinical survival-association analysis.
- Test and result
- Expression-derived phenotypes in 514 TCGA tumors, with classifier assignment and survival comparisons in 1,063 GEO tumors and a 34-patient institutional cohort.
- Limitation / still unknown
- Prospective treatment-response utility and causal biology remain unestablished; immune-cell abundance is inferred from bulk expression and aggregate patient totals differ between the abstract and full text.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC11027644/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 21 Nov 2026. Review scope: Primary relevant model Methods, Results and limitations; exact primary identity and current primary PubMed correction/retraction relationships checked..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.