Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology · 2024
Prognostic models for progression-free survival in atypical meningioma: Comparison of machine learning-based approach and the COX model in an Asian multicenter study.
СтатьяМногоцентровое исследованиеСравнительное исследованиеПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод в очереди
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.
Kim Dowook, Kim Yeseul, Sung Wonmo, Kim In Ah, Cho Jaeho, Lee Joo Ho, Grassberger Clemens, Byun Hwa Kyung, Chang Won Ick, Ren Leihao, Gong Ye, Wee Chan Woo, Hua Lingyang, Yoon Hong In
Аннотация
BACKGROUND AND PURPOSE: Atypical meningiomas are prevalent intracranial tumors with varied prognoses and recurrence rates. The role of adjuvant radiotherapy (ART) in atypical meningiomas remains debated. This study aimed to develop and validate a prognostic model incorporating machine learning techniques and clinical factors to predict progression-free survival (PFS) in patients with atypical meningiomas and assess the impact of ART.
MATERIALS AND METHODS: A retrospective review of 669 patients from five institutions in Korea and China was conducted. Cox proportional hazards, gradient boosting machine, and random survival forest models were employed for comparative analysis, utilizing both internal and external validation sets. Model performance was assessed using Harrell's concordance index and permutation feature importance.
RESULTS: Of 581 eligible patients, age, post-operative platelet count, performance status, Simpson grade, and ART were identified as significant prognostic factors across all models. In the ART subgroup, age and tumor size were the top prognostic indicators. The Cox model outperformed other methods, achieving a training C-index of 0.73 (95 % CI: 0.72-0.73) and an external validation C-index of 0.74 (95 % CI: 0.73-0.74). The model effectively stratified patients into risk categories, revealing a differential impact of ART: low-risk patients in the active surveillance group showed a 5.6 % improvement in 5-year PFS with predicted ART addition, compared to a 15.9 % improvement in the high-risk group.
CONCLUSION: This multicenter study offers a validated prognostic model for atypical meningiomas, highlighting the need for tailored treatment plans. The model's ability to stratify patients into risk categories for PFS provides a valuable tool for clinical decision-making, potentially optimizing patient outcomes.
MATERIALS AND METHODS: A retrospective review of 669 patients from five institutions in Korea and China was conducted. Cox proportional hazards, gradient boosting machine, and random survival forest models were employed for comparative analysis, utilizing both internal and external validation sets. Model performance was assessed using Harrell's concordance index and permutation feature importance.
RESULTS: Of 581 eligible patients, age, post-operative platelet count, performance status, Simpson grade, and ART were identified as significant prognostic factors across all models. In the ART subgroup, age and tumor size were the top prognostic indicators. The Cox model outperformed other methods, achieving a training C-index of 0.73 (95 % CI: 0.72-0.73) and an external validation C-index of 0.74 (95 % CI: 0.73-0.74). The model effectively stratified patients into risk categories, revealing a differential impact of ART: low-risk patients in the active surveillance group showed a 5.6 % improvement in 5-year PFS with predicted ART addition, compared to a 15.9 % improvement in the high-risk group.
CONCLUSION: This multicenter study offers a validated prognostic model for atypical meningiomas, highlighting the need for tailored treatment plans. The model's ability to stratify patients into risk categories for PFS provides a valuable tool for clinical decision-making, potentially optimizing patient outcomes.