Hallmark pathway scores and prognosis in lower-grade glioma: validation in two cohorts against random gene sets

Background Hallmark pathway scores may capture prognostic biology in lower-grade glioma, but their added value beyond clinical and molecular factors remains uncertain. Methods We analyzed TCGA-LGG (352 patients, 87 deaths) and CGGA (561 patients, 329 deaths). Clinical Cox models were compared with models adding Hallmark pathway scores (eight selected pathways, all 50, or five components) through cross-validation and bidirectional external validation. Random expression-matched gene sets and whole-transcriptome components served as benchmarks. Results The clinical model achieved C indices of 0.85 in TCGA cross-validation, 0.75 in forward validation, and 0.82 in reverse validation. Pathway scores changed the C index by − 0.017 to + 0.015 across four primary evaluations, and no change remained significant after Holm adjustment. Five pathway components added information beyond clinical predictors in pooled analyses (χ² = 25.1, 5 df, P = 0.00013), while random sets and whole-transcriptome components provided comparable or greater information. Sixteen pathways were nominally more prognostic than matched random sets in both cohorts. Eight-pathway scores selected in one cohort were associated with survival in the other cohort without a consistent gain in C index. Conclusions Hallmark pathway scores did not consistently improve discrimination beyond clinical and molecular factors. Some pathways were more prognostic than matched random sets in both cohorts, and random-set benchmarks help separate pathway specific from generic transcriptomic information.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-19
DOI
https://doi.org/10.21203/rs.3.rs-11677226/v1
Primary Topic
Glioma Diagnosis and Treatment
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Hallmark pathway scores and prognosis in lower-grade glioma: validation in two cohorts against random gene sets

Moustafa Sherif, Aya Darwish, Ahmed Kassab
Research Square
Glioma Diagnosis and Treatment
preprint

Hallmark pathway scores and prognosis in lower-grade glioma: validation in two cohorts against random gene sets

Moustafa Sherif, Aya Darwish, Ahmed Kassab
preprint en

Abstract

Background Hallmark pathway scores may capture prognostic biology in lower-grade glioma, but their added value beyond clinical and molecular factors remains uncertain. Methods We analyzed TCGA-LGG (352 patients, 87 deaths) and CGGA (561 patients, 329 deaths). Clinical Cox models were compared with models adding Hallmark pathway scores (eight selected pathways, all 50, or five components) through cross-validation and bidirectional external validation. Random expression-matched gene sets and whole-transcriptome components served as benchmarks. Results The clinical model achieved C indices of 0.85 in TCGA cross-validation, 0.75 in forward validation, and 0.82 in reverse validation. Pathway scores changed the C index by − 0.017 to + 0.015 across four primary evaluations, and no change remained significant after Holm adjustment. Five pathway components added information beyond clinical predictors in pooled analyses (χ² = 25.1, 5 df, P = 0.00013), while random sets and whole-transcriptome components provided comparable or greater information. Sixteen pathways were nominally more prognostic than matched random sets in both cohorts. Eight-pathway scores selected in one cohort were associated with survival in the other cohort without a consistent gain in C index. Conclusions Hallmark pathway scores did not consistently improve discrimination beyond clinical and molecular factors. Some pathways were more prognostic than matched random sets in both cohorts, and random-set benchmarks help separate pathway specific from generic transcriptomic information.

Research Square
Pathfinder International (US), Benha University (EG)
Glioma Diagnosis and Treatment
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon