Development and validation of prognostic models for bone metastasis in Non-Small cell lung cancer based on Machine learning algorithms

IF 3.5 2区医学 Q2 Medicine

Journal of Bone Oncology Pub Date : 2025-10-04 DOI:10.1016/j.jbo.2025.100716

Jiabin Fang , Xiaojie Yang , Lingfeng Chen , Liuying Hong , Yingqiu He , Ji Huang , Jie Lin , Nengluan Xu , Hongru Li

{"title":"Development and validation of prognostic models for bone metastasis in Non-Small cell lung cancer based on Machine learning algorithms","authors":"Jiabin Fang , Xiaojie Yang , Lingfeng Chen , Liuying Hong , Yingqiu He , Ji Huang , Jie Lin , Nengluan Xu , Hongru Li","doi":"10.1016/j.jbo.2025.100716","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Bone is a common site of metastasis in non-small cell lung cancer (NSCLC), yet no validated prognostic model is currently available for patients presenting with bone metastases at diagnosis.</div></div><div><h3>Methods</h3><div>We retrospectively reviewed 1,299 NSCLC patients who underwent high-throughput sequencing between 2016 and 2023. Of these, 195 were diagnosed with bone metastases at presentation. Three machine learning algorithms were applied to identify prognostic variables. A nomogram constructed with Cox regression was used to predict overall survival (OS) and was internally validated with 1,000 bootstrap resamples.</div></div><div><h3>Results</h3><div>Four independent prognostic factors were identified, including age, serum calcium, monocyte-to-albumin ratio, and prognostic nutritional index. The nomogram demonstrated strong predictive performance, with areas under the curve (AUCs) of 86.53%, 78.32%, and 77.85% for 6-month, 1-year, and 2-year OS, respectively. Calibration plots showed excellent agreement between predicted and observed survival outcomes.</div></div><div><h3>Conclusion</h3><div>This validated nomogram provides a practical and individualized tool for predicting survival in NSCLC patients with bone metastases at diagnosis, supporting risk stratification and clinical practice.</div></div>","PeriodicalId":48806,"journal":{"name":"Journal of Bone Oncology","volume":"55 ","pages":"Article 100716"},"PeriodicalIF":3.5000,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Bone Oncology","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2212137425000570","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Medicine","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Bone is a common site of metastasis in non-small cell lung cancer (NSCLC), yet no validated prognostic model is currently available for patients presenting with bone metastases at diagnosis.

Methods

We retrospectively reviewed 1,299 NSCLC patients who underwent high-throughput sequencing between 2016 and 2023. Of these, 195 were diagnosed with bone metastases at presentation. Three machine learning algorithms were applied to identify prognostic variables. A nomogram constructed with Cox regression was used to predict overall survival (OS) and was internally validated with 1,000 bootstrap resamples.

Results

Four independent prognostic factors were identified, including age, serum calcium, monocyte-to-albumin ratio, and prognostic nutritional index. The nomogram demonstrated strong predictive performance, with areas under the curve (AUCs) of 86.53%, 78.32%, and 77.85% for 6-month, 1-year, and 2-year OS, respectively. Calibration plots showed excellent agreement between predicted and observed survival outcomes.

Conclusion

This validated nomogram provides a practical and individualized tool for predicting survival in NSCLC patients with bone metastases at diagnosis, supporting risk stratification and clinical practice.

Abstract Image

查看原文本刊更多论文

基于机器学习算法的非小细胞肺癌骨转移预后模型的建立和验证

骨是非小细胞肺癌（NSCLC）中常见的转移部位，但目前尚无有效的预后模型用于诊断时出现骨转移的患者。方法回顾性分析了2016年至2023年间接受高通量测序的1299例非小细胞肺癌患者。其中，195人在就诊时被诊断为骨转移。应用三种机器学习算法来识别预测变量。采用Cox回归构建的nomogram来预测总生存期（OS），并在1000个bootstrap样本中进行内部验证。结果确定了4个独立的预后因素，包括年龄、血钙、单核细胞/白蛋白比和预后营养指数。nomogram具有较强的预测能力，6个月、1年和2年OS的曲线下面积（auc）分别为86.53%、78.32%和77.85%。校正图显示预测和观察到的生存结果非常一致。结论：经验证的nomogram骨转移图为预测骨转移NSCLC患者的生存提供了一种实用且个性化的工具，支持风险分层和临床实践。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Bone Oncology ONCOLOGY-

CiteScore

7.20

自引率

2.90%

发文量

审稿时长

34 days

期刊介绍： The Journal of Bone Oncology is a peer-reviewed international journal aimed at presenting basic, translational and clinical high-quality research related to bone and cancer. As the first journal dedicated to cancer induced bone diseases, JBO welcomes original research articles, review articles, editorials and opinion pieces. Case reports will only be considered in exceptional circumstances and only when accompanied by a comprehensive review of the subject. The areas covered by the journal include: Bone metastases (pathophysiology, epidemiology, diagnostics, clinical features, prevention, treatment) Preclinical models of metastasis Bone microenvironment in cancer (stem cell, bone cell and cancer interactions) Bone targeted therapy (pharmacology, therapeutic targets, drug development, clinical trials, side-effects, outcome research, health economics) Cancer treatment induced bone loss (epidemiology, pathophysiology, prevention and management) Bone imaging (clinical and animal, skeletal interventional radiology) Bone biomarkers (clinical and translational applications) Radiotherapy and radio-isotopes Skeletal complications Bone pain (mechanisms and management) Orthopaedic cancer surgery Primary bone tumours Clinical guidelines Multidisciplinary care Keywords: bisphosphonate, bone, breast cancer, cancer, CTIBL, denosumab, metastasis, myeloma, osteoblast, osteoclast, osteooncology, osteo-oncology, prostate cancer, skeleton, tumour.