Quantitative and clinical implications of the EARL2 versus EARL1 [18F]FDG PET-CT performance standards in head and neck squamous cell carcinoma

Abstract Background: The EANM Research Ltd. (EARL) guidelines give recommendations for harmonization of [18F]FDG PET-CT image acquisition and reconstruction, aiming to ensure reproducibility of quantitative data between PET scanners. Recent technological advancements in PET-CT imaging resulted in an updated version of the EARL guidelines (EARL2). The aim of this study is to compare quantitative [18F]FDG uptake metrics of the primary tumor and lymph nodes in patients with head and neck squamous cell carcinoma (HNSCC) on EARL2 versus EARL1 reconstructed images and to describe clinical implications for nodal staging and treatment.Methods: Forty-nine consecutive patients with HNSCC were included. For all, both EARL1 and EARL2 images were reconstructed from a singular [18F]FDG PET-CT scan. Primary tumors and non-necrotic lymph nodes ≥ 5mm were delineated on CT-scan. In the quantitative analysis, maximum standardized uptake values (SUVmax) and standardized uptake ratios (SURmax, i.e. SUVmax normalized to cervical spinal cord uptake) were calculated for all lesions on EARL1 and EARL2 reconstructions. Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were compared between EARL1 and EARL2 using different segmentation methods (adaptive threshold; SUV2.5/3.5/4.5; SUR2.5/3.5/4.5; MAX40%/50%). In the qualitative analysis, each lymph node was scored independently by two nuclear medicine physicians on both EARL1 and EARL2 images on different occasions using a 4-point scale.Results: There was a significant increase of SUVmax (19%) and SURmax (11%) of primary tumor and lymph nodes on EARL2 versus EARL1 imaging (p < 0.001). The proportional difference of both SUVmax and SURmax between EARL2 and EARL1 decreased with increasing tumor volume (p < 0.001). Absolute differences in MTVs between both reconstructions were small (< 1.0cm3), independent of the segmentation method. Relative differences in MTVs and TLGs were small using the adaptive threshold method and larger using static SUV or SUR thresholds. With visual scoring of lymph nodes 38% (11/29) of nodes with score 2 on EARL1 were upstaged to score 3 on EARL2, which resulted in an alteration of nodal stage in 18% (6/33) of the patients.Conclusions: Using the EARL2 method for PET image reconstruction resulted in higher SUVmax and SURmax compared to EARL1, with nodal upstaging in a significant number of patients.