Development and biological characterization of a random survival forest prognostic model for colon adenocarcinoma using eight machine learning algorithms.
- 1Department of Gastroenterology, Minda Hospital of Hubei Minzu University, Enshi, China.
- 2Hubei Provincial Key Laboratory of Occurrence and Intervention of Rheumatic Diseases (Hubei Minzu University), Enshi, China.
- 3Hubei Provincial Clinical Medical Research Center for Nephropathy (Hubei Minzu University), Enshi, China.
Abstract
Prognostic models for colon adenocarcinoma (COAD) frequently rely on predefined biologic gene sets or a single analytic framework, and few studies simultaneously address algorithm selection, biologic interpretation, comparison with conventional staging, and benchmarking against published COAD signatures. This study aimed to develop and biologically characterize a random survival forest (RSF) prognostic model for COAD. RNA-sequencing (RNA-seq) and clinical data from 455 The Cancer Genome Atlas (TCGA)-COAD patients were analyzed in a training cohort (n=324) and an internal test cohort (n=131). A 12-gene signature was derived through differential-expression screening, log-rank testing, and univariate Cox analysis. Eight survival algorithms were compared, and the best-performing RSF model generated the final risk score. Additional analyses included Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment, variable-importance and SHapley Additive exPlanations (SHAP) interpretation, Kaplan-Meier and subgroup interaction analyses, Cox regression, comparison of RSF vs. tumor-node-metastasis (TNM) vs. RSF + TNM using time-dependent receiver operating characteristic (ROC), concordance index (C-index), net reclassification improvement (NRI), and integrated discrimination improvement (IDI), Immuno-Oncology Biological Research (IOBR)-based immune-signature profiling, and benchmarking against published COAD signatures. All 12 model genes mapped successfully for enrichment analysis, yielding 115 significant GO biological-process terms, 33 cellular-component terms, 25 molecular-function terms, and 4 KEGG pathways centered on telomere regulation, chromosome organization, and transcription-related programs. RSF showed near-perfect internal discrimination, with 1-, 3-, and 5-year areas under the ROC curves (AUCs) of 0.999, 0.999, and 0.998 in the training set, but external performance was more modest in the test set (0.631, 0.608, and 0.552). PPFIA4, GTF2H4, H4C5, ARL6IP4, LRP2, and TERT were consistently prioritized by variable importance (VIMP) and SHAP. Only pathologic M in training and lymphatic invasion in testing showed significant interaction with RSF score. In the test cohort, TNM and RSF + TNM outperformed RSF alone, and adding TNM to RSF improved NRI at 1 and 3 years. IOBR analyses identified 72 significant correlations in training and 57 in testing, dominated by exosome-, glycan-, metabolism-, and T-cell-related signatures. The 12-gene RSF score captures biologically meaningful transcriptomic and immune-related signals in COAD, but its most defensible translational role is as a complement to TNM rather than a replacement. Rigorous external RNA-seq validation remains necessary.
Related
- Spatial multi-omics uncovers an invasive-desmoplastic cellular community with prognostic and therapeutic implications in pancreatic ductal adenocarcinoma
- Prognostic and Clinicopathological Significance of CD74 Expression in Pancreatic Ductal Adenocarcinoma: A Systematic Review and Meta-Analysis
- Kynureninase Is a Prognostic Biomarker of Poor Survival and Gemcitabine Resistance in Pancreatic Ductal Adenocarcinoma
- LLPS-Driven Membraneless Organelle Activity Landscape Reveals Molecular Subtypes, Immune States, and Prognostic Stratification in Lung Adenocarcinoma
- Characterization of BRD4 during Mammalian Postmeiotic Sperm Development
BibTeX
@article{zhang2026development,
title = {{Development and biological characterization of a random survival forest prognostic model for colon adenocarcinoma using eight machine learning algorithms.}},
author = {Jiqiao Zhang and Jiawu Tan and Mingjun Lu and Dan Luo and Mouguan Yang and Feng Gao and Chan He and Maoyin He and Lifeng Qin},
journal = {Translational cancer research},
year = {2026},
month = sep,
doi = {10.21037/tcr-2026-0877},
url = {https://europepmc.org/article/MED/42997891},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Development and biological characterization of a random survival forest prognostic model for colon adenocarcinoma using eight machine learning algorithms.
AU - Jiqiao Zhang
AU - Jiawu Tan
AU - Mingjun Lu
AU - Dan Luo
AU - Mouguan Yang
AU - Feng Gao
AU - Chan He
AU - Maoyin He
AU - Lifeng Qin
PY - 2026
DA - 2026/09/28
JO - Translational cancer research
DO - 10.21037/tcr-2026-0877
UR - https://europepmc.org/article/MED/42997891
AB - Prognostic models for colon adenocarcinoma (COAD) frequently rely on predefined biologic gene sets or a single analytic framework, and few studies simultaneously address algorithm selection, biologic interpretation, comparison with conventional staging, and benchmarking against published COAD signatures. This study aimed to develop and biologically characterize a random survival forest (RSF) prognostic model for COAD. RNA-sequencing (RNA-seq) and clinical data from 455 The Cancer Genome Atlas (TCGA)-COAD patients were analyzed in a training cohort (n=324) and an internal test cohort (n=131). A 12-gene signature was derived through differential-expression screening, log-rank testing, and univariate Cox analysis. Eight survival algorithms were compared, and the best-performing RSF model generated the final risk score. Additional analyses included Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment, variable-importance and SHapley Additive exPlanations (SHAP) interpretation, Kaplan-Meier and subgroup interaction analyses, Cox regression, comparison of RSF vs. tumor-node-metastasis (TNM) vs. RSF + TNM using time-dependent receiver operating characteristic (ROC), concordance index (C-index), net reclassification improvement (NRI), and integrated discrimination improvement (IDI), Immuno-Oncology Biological Research (IOBR)-based immune-signature profiling, and benchmarking against published COAD signatures. All 12 model genes mapped successfully for enrichment analysis, yielding 115 significant GO biological-process terms, 33 cellular-component terms, 25 molecular-function terms, and 4 KEGG pathways centered on telomere regulation, chromosome organization, and transcription-related programs. RSF showed near-perfect internal discrimination, with 1-, 3-, and 5-year areas under the ROC curves (AUCs) of 0.999, 0.999, and 0.998 in the training set, but external performance was more modest in the test set (0.631, 0.608, and 0.552). PPFIA4, GTF2H4, H4C5, ARL6IP4, LRP2, and TERT were consistently prioritized by variable importance (VIMP) and SHAP. Only pathologic M in training and lymphatic invasion in testing showed significant interaction with RSF score. In the test cohort, TNM and RSF + TNM outperformed RSF alone, and adding TNM to RSF improved NRI at 1 and 3 years. IOBR analyses identified 72 significant correlations in training and 57 in testing, dominated by exosome-, glycan-, metabolism-, and T-cell-related signatures. The 12-gene RSF score captures biologically meaningful transcriptomic and immune-related signals in COAD, but its most defensible translational role is as a complement to TNM rather than a replacement. Rigorous external RNA-seq validation remains necessary.
SN - 2218-676X
SN - 2219-6803
ER -