KM Research Hub
← 返回论文库
待处理 入排:待复核 图源:待检查 入排置信度 0.45

Zone-specific lymph node metastasis prediction model in clinical stage I-II non-small cell lung cancer: a supplementary analysis of JCOG1413.

Mitsuhiro Isaka, Tomoyuki Hishida, Hisashi Saji, Masashi Wakabayashi, Yasuhisa Ohde, Yasuhiro Tsutani, Makoto Endo, Takahiro Mimae, Tomohiro Miyoshi, Tetsuya Isaka, Aritoshi Hattori, Hiroshige Yoshioka, Ryu Nakajima, Masaya Yotsukura, Tomohiro Maniwa, Kazuo Nakagawa, Noriko Mitome, Yuta Sekino, Haruhiko Fukuda, Keiju Aokage, Shun-Ichi Watanabe, Morihito Okada, Japan Clinical Oncology Group (JCOG)

Eur J Surg Oncol 53(1) 112363

PMID43013968
PMCID-
DOI10.1016/j.ejso.2026.112363
发表日期2026-11-11
入排依据:未可靠识别 OS/PFS/DFS/EFS/RFS/TTP、Kaplan-Meier、HR 或其他 time-to-event 结局,需复核。
摘要

摘要

BACKGROUND: The optimal extent of lymph node dissection in early-stage non-small cell lung cancer remains unclear. We aimed to develop zone-specific lymph node metastasis prediction models for clinical stage I-II non-small cell lung cancer using data from the Japan Clinical Oncology Group (JCOG) 1413, a randomized phase III trial comparing lobe-specific and systematic lymph node dissection. METHODS: The analysis included 1699 eligible cases out of 1711 patients from 44 institutions in JCOG1413. Leave-one-out cross-validation was used to evaluate the prediction algorithm, and multivariable logistic regression identified independent predictors of metastasis for constructing nomograms. Model performance was assessed using receiver operating characteristic curves with concordance statistics and calibration curves with calibration slope. RESULTS: We developed models for four zones: upper (including the aortopulmonary zone) (n = 1416), subcarinal (n = 1170), hilar (n = 1662), and peripheral (n = 1662). Common independent predictors included histology, maximum standardized uptake on positron emission tomography, and, particularly for mediastinal zones, primary tumor lobe location. The subcarinal zone showed the best performance, with a concordance statistic of 0.7902 and a calibration slope of 1.0762. Across all zones, calibration slopes remained close to 1.0, and negative predictive values ranged from 96.6% to 99.4%. CONCLUSIONS: Zone-specific prediction models demonstrated moderate discrimination and well-calibrated estimates of lymph-node metastasis risk, with the highest performance observed in the subcarinal zone. These models could support risk stratification and facilitate individualized intraoperative nodal assessment to inform clinical decision-making.

KM 提取结果

0 个 Figure · 0 个 KM 面板

尚未提取到 KM 面板。