PET radiomics for histologic subtype classification of non-small cell lung cancer: a systematic review and meta-analysis

• Published in: European journal of nuclear medicine and molecular imaging Vol. 52; no. 6; pp. 2212 - 2224
• Main Authors: Zhang, Jucheng, Zhang, Xiaohui, Zhong, Yan, Wang, Jing, Zhong, Chao, Xiao, Meiling, Chen, Yuhan, Zhang, Hong
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.05.2025; Springer Nature B.V
• Subjects: Adenocarcinoma; Carcinoma, Non-Small-Cell Lung - classification; Carcinoma, Non-Small-Cell Lung - diagnostic imaging; Carcinoma, Non-Small-Cell Lung - pathology; Cardiology; Classification; Confidence intervals; Heterogeneity; Humans; Image Processing, Computer-Assisted - methods; Imaging; Lung cancer; Lung Neoplasms - classification; Lung Neoplasms - diagnostic imaging; Lung Neoplasms - pathology; Medicine; Medicine & Public Health; Meta-analysis; Non-small cell lung carcinoma; Nuclear Medicine; Oncology; Orthopedics; Positron-Emission Tomography; Quality assessment; Quality control; Radiology; Radiomics; Review Article; Small cell lung carcinoma; Squamous cell carcinoma; Statistical analysis; Statistical models; Subgroups; Positron emission tomography (PET); NSCLC; Histologic subtype; Radiomics
• ISSN: 1619-7070; 1619-7089; 1619-7089
• DOI: 10.1007/s00259-025-07069-6

Purpose To systematically review the literature and perform a meta-analysis of PET radiomics for histologic subtype classification in non-small cell lung cancer (NSCLC). Methods PubMed, Embase, Scopus, and Web of Science databases were systematically searched in English on human subjects for studies on distinguishing adenocarcinoma (ADC) from squamous cell carcinoma (SCC) using PET radiomics published from inception until November 2024. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool and the Radiomics Quality Score (RQS) were utilized to assess the methodological quality of the included studies. The area under the receiver operating characteristic curves (AUC) was pooled to estimate predictive performance. An overall effect size was estimated using a random-effects model. Statistical heterogeneity was evaluated by the I 2 value. Subgroup analyses were conducted to explore sources of heterogeneity. Results Twelve studies were included in the analysis, yielding a pooled AUC of 0.92 (95% confidence interval [CI]: 0.89–0.94). Despite this promising result, the studies showed limitations in both study design and methodological quality, as evidenced by a median RQS of 11/36. A significant degree of heterogeneity was observed among the studies, with an I 2 of 92.20% (95% CI: 89.01–95.39) for sensitivity and 89.29% (95% CI: 84.48–94.10) for specificity. Conclusions This meta-analysis highlights the potential utility of PET radiomics in distinguishing ADC from SCC. However, the observed high heterogeneity indicates substantial methodological variability across the included studies. Future research should focus on standardization, transparency, and multicenter collaborations to improve the reliability and clinical applicability of PET radiomics for histologic subtype classification in NSCLC.