Enhanced detection of skull-base bone invasion in nasopharyngeal carcinoma: the supplementary diagnostic value of 18fluorine-sodium fluoride (18F-NaF) positron emission tomography/computed tomography (PET/CT) combined with magnetic resonance imaging (MRI)
Meina Liang,Xufeng Guo,Chengmao Guo,Jingxing Xiao
DOI: https://doi.org/10.21037/qims-24-265
2024-10-01
Quantitative Imaging in Medicine and Surgery
Abstract:Background: Nasopharyngeal carcinoma (NPC) originates in the nasopharyngeal mucosa, the lateral wall of the nasopharynx. A significant challenge in NPC management is skull-base bone invasion (SBBI), which affects prognosis and treatment planning. Magnetic resonance imaging (MRI) is the primary diagnostic tool for SBBI in NPC patients; however, the detection of SBBI can be challenging due to skull-base complexity and overlapping MRI signals. <sup>18</sup>fluorine-sodium fluoride (<sup>18</sup>F-NaF) positron emission tomography/computed tomography (PET/CT) is an emerging imaging technique that has shown promise in detecting osseous lesions. This cohort study aimed to assess the supplementary diagnostic value of <sup>18</sup>F-NaF PET/CT in detecting SBBI in NPC patients compared to that of MRI alone.Methods: Imaging data were retrospectively collected from <sup>18</sup>F-NaF PET/CT and head-and-neck MRI examinations conducted within a 7-day period. The sensitivity, specificity, and accuracy of <sup>18</sup>F-NaF PET/CT, MRI, and the combination of both modalities in detecting SBBI were individually assessed. Both lesion- and patient-based analyses were employed for the comparison. Cochran's Q test was used to compare the accuracy of these methods, while the Bonferroni-corrected McNemar test was used for the pairwise comparisons. The data analysis was performed using the R software package, and a significance level of P<i><</i>0.05 was considered statistically significant.Results: A total of 164 patients were enrolled in the study. Using <sup>18</sup>F-NaF PET/CT, MRI, and the combined modality of <sup>18</sup>F-NaF PET/CT with MRI, 97, 84, and 94 cases of SBBI were diagnosed, respectively. At the patient level, the diagnostic efficacy (sensitivity, specificity, and accuracy) was as follows: <sup>18</sup>F-NaF PET/CT had 100% sensitivity, 93.1% specificity, and 97.0% accuracy; MRI had 90.2% sensitivity, 98.6% specificity, and 93.9% accuracy; and the combination of <sup>18</sup>F-NaF PET/CT and MRI had 100% sensitivity, 97.2% specificity, and 98.8% accuracy. The accuracy rate of <sup>18</sup>F-NaF PET/CT combined with MRI were significantly higher than that of MRI alone (P<i>=</i>0.034). A total of 284, 243, and 276 SBBI lesions were diagnosed using <sup>18</sup>F-NaF PET/CT, MRI, and <sup>18</sup>F-NaF PET/CT combined with MRI, respectively. The diagnostic efficacy (sensitivity, specificity, and accuracy) at the lesion level was as follows: <sup>18</sup>F-NaF PET/CT had 99.6% sensitivity, 75.9% specificity, and 95.4% accuracy; MRI had 88.2% sensitivity, 93.1% specificity, and 89.1% accuracy; and the combination of <sup>18</sup>F-NaF PET/CT with MRI had 100% sensitivity, 91.4% specificity, and 98.5% accuracy. The combination of <sup>18</sup>F-NaF PET/CT with MRI significantly improved the accuracy rate compared to that of MRI alone, and the difference was statistically significant (P<0.001).Conclusions: The combined use of <sup>18</sup>F-NaF PET/CT and MRI significantly enhanced the diagnosis of SBBI in NPC patients, and the combined method had improved diagnostic sensitivity and accuracy than MRI alone.
radiology, nuclear medicine & medical imaging