Current gene therapy · 2026
Machine Learning-Driven PCDI Classifier for Invasive PitNETs.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Wang Guanyu, Yan Song, Zhang Luyang, Lin Lu, Liu Rentong, Han Yiling, Zhao Yan
Аннотация
INTRODUCTION: Aggressive Pituitary Neuroendocrine Tumors (PitNETs) pose significant therapeutic challenges due to their invasive behavior and resistance to conventional therapies. Current prognostic markers lack the ability to capture molecular heterogeneity, necessitating novel biomarkers. Dysregulated Programmed Cell Death (PCD) pathways are implicated in tumorigenesis, but their prognostic relevance in invasive PitNETs remains unexplored.
METHODS: GEO datasets (GSE51618, GSE169498, GSE260487) were analyzed to identify differential gene expression between noninvasive and invasive PitNETs. A curated panel of 1,548 PCDrelated genes was integrated. Machine learning (LASSO regression and SVM-RFE) was employed to construct a PCD-associated Index (PCDI). For validation, ROC analysis, immune infiltration assessment (CIBERSORT, TIMER, ssGSEA), and experimental validation via RT-qPCR were performed.
RESULTS: The PCDI, comprising 11 genes (e.g., FGFR3, MAPK11, SLC7A11), distinguished invasive from noninvasive PitNETs with high accuracy. High-PCDI tumors exhibited enriched metabolic pathways and immune activation. Consensus clustering stratified PitNETs into two molecular subtypes (C1/C2), with C2 (high-PCDI) showing elevated immune scores and pathway activity. Experimental validation confirmed the differential expression of key genes in invasive tumors (*p<0.05).
DISCUSSION: The PCDI outperforms traditional prognostic models by capturing PCD-immunemetabolic crosstalk. High-PCDI tumors demonstrate adaptive immune evasion despite an elevated checkpoint molecule expression, suggesting therapeutic potential for combined MAPK inhibitors and immunotherapy. Limitations include retrospective data and small validation cohorts.
CONCLUSION: The PCDI provides a robust molecular framework for risk stratification and personalized therapy in invasive PitNETs. Future studies should validate its clinical utility and explore pancancer relevance.
METHODS: GEO datasets (GSE51618, GSE169498, GSE260487) were analyzed to identify differential gene expression between noninvasive and invasive PitNETs. A curated panel of 1,548 PCDrelated genes was integrated. Machine learning (LASSO regression and SVM-RFE) was employed to construct a PCD-associated Index (PCDI). For validation, ROC analysis, immune infiltration assessment (CIBERSORT, TIMER, ssGSEA), and experimental validation via RT-qPCR were performed.
RESULTS: The PCDI, comprising 11 genes (e.g., FGFR3, MAPK11, SLC7A11), distinguished invasive from noninvasive PitNETs with high accuracy. High-PCDI tumors exhibited enriched metabolic pathways and immune activation. Consensus clustering stratified PitNETs into two molecular subtypes (C1/C2), with C2 (high-PCDI) showing elevated immune scores and pathway activity. Experimental validation confirmed the differential expression of key genes in invasive tumors (*p<0.05).
DISCUSSION: The PCDI outperforms traditional prognostic models by capturing PCD-immunemetabolic crosstalk. High-PCDI tumors demonstrate adaptive immune evasion despite an elevated checkpoint molecule expression, suggesting therapeutic potential for combined MAPK inhibitors and immunotherapy. Limitations include retrospective data and small validation cohorts.
CONCLUSION: The PCDI provides a robust molecular framework for risk stratification and personalized therapy in invasive PitNETs. Future studies should validate its clinical utility and explore pancancer relevance.