PET/CT vs PET/MRI in Lung Cancer: A Narrative Review of Diagnostic Performance, Practical Trade-offs, and Clinical Impact.

Lung cancer remains the leading cause of cancer death worldwide. Accurate initial staging is central to treatment selection and prognosis. Hybrid imaging with 18F fludeoxyglucose (FDG) positron emission tomography-computed tomography (PET-CT) is widely adopted as the pre-radical staging test, while positron emission tomography-magnetic resonance imaging (PET-MRI) has emerged as an alternative that replaces the CT component with multiparametric MRI, potentially improving soft-tissue assessment and reducing ionising dose. This narrative review evaluates PET-CT and PET-MRI for lung cancer staging across Tumour Node Metastasis (TNM) domains, drawing on English-language studies (2008-2025) with emphasis on head-to-head comparisons in non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC), complemented by guidelines, quantitative studies and operational literature. In NSCLC, PET-MRI achieves overall staging performance comparable to PET-CT and improves confidence for pleural, chest-wall and mediastinal invasion, particularly when diffusion-weighted imaging (DWI) and cine/free-breathing radial T1-weighted MRI at 3 T (a higher-field MRI strength than 1.5 T). PET-CT retains an advantage for sub-centimetre and subsolid pulmonary nodules; MRI-led pathways should therefore mandate a contemporaneous thin-slice diagnostic chest CT when such nodules could change intent. Nodal staging is broadly similar between hybrids; where results would alter management, selective invasive confirmation remains appropriate, with combined endobronchial and endoscopic ultrasound (EBUS-EUS) improving the sampling efficiency. For distant disease, PET-CT is an efficient whole-body screen at baseline, while PET-MRI adds value for brain and liver problem-solving. In SCLC, a prospective cohort reported higher T-category and overall TNM accuracy with whole-body MRI and coregistered PET-MRI than with PET-CT. Quantitatively, standardised uptake values (SUVs) correlate between platforms but are not interchangeable; centres should harmonise using European Association of Nuclear Medicine Research Ltd (EARL) and Quantitative Imaging Biomarkers Alliance (QIBA) guidance and document MRI attenuation-correction choices (e.g., Dixon vs ultrashort/zero echo time (UTE/ZTE)). Operationally, PET-MRI reduces anatomic radiation but typically requires longer slots and specialised workflow; PET-CT is faster, widely available, and linked to fewer futile thoracotomies and favourable cost-effectiveness. Overall, PET-CT should remain the default hybrid, with PET-MRI deployed selectively, particularly for complex local invasion, integrated brain-to-body staging, or dose reduction, while safeguarding small-nodule detection. Priorities include multicentre trials with harmonised protocols, management-impact endpoints, and explicit rules for integrating chest CT into MRI-led pathways.