A Combined Model Based on Bone Mineral Density for Noninvasive Prediction of Prognosis in Non‐Small Cell Lung Cancer Patients Receiving Immune Checkpoint Inhibitors: A Multicenter Retrospective Study
The prognostic value of baseline bone mineral density (BMD) and posttreatment BMD decrease (BMDD) in non-small cell lung cancer (NSCLC) patients receiving immune checkpoint inhibitor (ICI) treatment remains unclear. We assembled data of 2096 patients with advanced NSCLC from five institutions to develop a combined model incorporating BMD/BMDD and clinical characteristics for noninvasive prognosis prediction. BMD was automatically assessed using a deep learning-based method. Compared with the physiological BMD group and the non-severe BMDD group, the pathological BMD group and the severe BMDD group had shorter progression-free survival (PFS) (hazard ratio [HR]: 1.19, <i>p</i> = 0.003; and HR: 1.19, <i>p</i> = 0.002, respectively) and overall survival (OS) (HR: 1.31, <i>p</i> < 0.001; and HR: 1.30, <i>p</i> < 0.001). Compared with the single BMD/BMDD model, the combined model had higher Harrell's concordance indexes (<i>c</i>-indexes) (PFS: 0.580 and OS: 0.654). Transcriptomic analysis of 130 patients from the NSCLC radiogenomic cohort revealed upregulation of epithelial-mesenchymal transition, inflammatory, and hypoxia pathways, and increased macrophage infiltration in tumors of patients with pathological BMD. This study showed that lower baseline BMD and more severe BMDD are associated with poorer prognosis. BMD in combination with clinical characteristics can help to improve risk stratification and prognosis prediction.
Authors
- Zilong Wu (ORCID: https://orcid.org/0000-0003-3556-1943)
- Dongmei Wei (ORCID: https://orcid.org/0000-0001-7010-468X)
- Peng Mo
- Tao Xiong
- Hongxiang Zeng (ORCID: https://orcid.org/0000-0002-3028-9007)
- Zhaokai Wang (ORCID: https://orcid.org/0009-0007-4357-7802)
- Yi Li (ORCID: https://orcid.org/0000-0001-7994-5362)
- Xi Zhang (ORCID: https://orcid.org/0000-0002-0422-8980)
- Hui Wang (ORCID: https://orcid.org/0000-0001-6789-5185)
- Jie Lou (ORCID: https://orcid.org/0009-0006-6064-1450)
- Yiqun Chen (ORCID: https://orcid.org/0000-0003-4218-084X)
- Xiaofei Zhang (ORCID: https://orcid.org/0000-0002-2759-4254)
- Bingxin Gong
- Qi Wan (ORCID: https://orcid.org/0000-0002-6978-4777)
- Lian Yang (ORCID: https://orcid.org/0000-0003-4664-2474)
- Yusheng Guo (ORCID: https://orcid.org/0000-0003-2977-9356)
- Xiaowen Liu (ORCID: https://orcid.org/0000-0002-3172-8905)
Institutions
- Fujian Medical University (CN)
- Nantong University (CN)
- Southern University of Science and Technology (CN)
- Affiliated Hospital of Nantong University (CN)
- First Affiliated Hospital of Guangzhou Medical University (CN)
- Precision for Medicine (United States) (US)
- Wuhan Union Hospital (CN)
- Neusoft (China) (CN)
- Southern Medical University Shenzhen Hospital (CN)
- Union Hospital (CN)
- Shenzhen Second People's Hospital (CN)
- Huazhong University of Science and Technology (CN)
- Hubei University (CN)
- Guangzhou Medical University (CN)
Publication Details
- Journal
- MedComm
- Published
- 2025-09-21
- DOI
- https://doi.org/10.1002/mco2.70398
- Citations
- 1
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 1.36
Funders
- National Natural Science Foundation of China
- Natural Science Foundation of Hubei Province
- Fundamental Research Funds for the Central Universities