The validation of low-dose CT scans from the [18F]-FDG PET-CT scan to assess skeletal muscle mass in comparison with diagnostic neck CT scans

Abstract

PurposeRadiologically defined sarcopenia, or a low skeletal muscle index (SMI), is an emerging biomarker for adverse clinical outcomes in head and neck cancer (HNC) patients. Recently, SMI measurements have been validated at the level of the third cervical vertebra (C3) on diagnostic neck CT scans but are not yet validated on low-dose (LD) neck CT scans from the [18F]-FDG PET-CT. This hampers SMI analysis in HNC patients without a diagnostic neck CT but with a [18F]-FDG PET-CT scan. Therefore, the aim was to study whether (low) SMI based on LD CT scan from [18F]-FDG PET-CT is comparable to those derived from diagnostic neck CT scans.MethodsHNC patients with both diagnostic CT and [18F]-FDG PET-CT of the neck were prospectively included into the OncoLifeS data-biobank. Skeletal muscle was retrospectively delineated at the level of the third cervical vertebra (C3), and (low) SMI (cm2/m2) was calculated for diagnostic and LD neck CTs. (Low) SMI from the diagnostic neck CT was considered the reference standard. Intra-class correlation coefficient (ICC), Bland–Altman plots, and Cohen’s Kappa analysis were performed.ResultsThe cohort (n = 233) mean age was 66.2 ± 12.8 years, and 74.2% of patients were male. Inter-rater reliability was excellent (ICC > 0.990, 95% confidence interval 0.975–0.996, p < 0.001). The agreement of SMI between both modalities was high according to the Bland–Altman plot (mean ΔSMI =  − 0.19 cm2/m2), and there was no substantial bias. Cohen’s Kappa analysis showed an almost perfect agreement of low SMI between the two modalities (κ = 0.911, p < 0.001). The position of arms didn't affect the high agreement of (low) SMI.ConclusionSkeletal muscle mass, as measured with (low) SMI, remains constant irrespective of CT acquisition parameters (diagnostic neck CT scans versus LD neck scans of the [18F]-FDG PET-CT scan), positioning of arms, and observers. These findings contribute to the construction of a clinically useful radiological biomarker for SMI and therefore identify patients at risk for adverse clinical outcomes.