POS0971 THE RELATIONSHIP OF AXIAL [18F]FLUORIDE UPTAKE ON PET-CT TO ANATOMICAL CHANGES ON MRI AND X-RAY IN CLINICALLY ACTIVE ANKYLOSING SPONDYLITIS PATIENTS

Abstract

BackgroundBone formation is a histopathologic feature of ankylosing spondylitis (AS), causing structural changes through the formation of syndesmophytes and spinal ankyloses (1, 2) Recent studies with [18F]Fluoride Positron emission tomography (PET) demonstrated that disease activity of AS is best reflected by bone formation rather than inflammation and is sensitive to change during therapy (2, 3). [18F]Fluoride PET(/CT) identifies more lesions in active AS than bone marrow edema on MRI, so may provide complementary information about the disease activity (3). A direct comparison between [18F]Fluoride PET(/CT) and MRI and X-rays will allow better understanding of PET-CT positive lesions and will support further positioning of the technique for diagnostics and monitoring of AS.ObjectivesTo relate uptake of [18F]Fluoride on PET-CT to abnormalities on MRI and conventional radiography in patients with clinically active AS.MethodsTen patients (female 6/10, age 38±11 years) with AS based on modified New York criteria, a BASDAI of ≥ 4 and naïve to biologicals were included. All patients received a [18F]Fluoride PET(/CT) scan, MRI scan and x-rays of the spine and SI-joints. Experienced readers blinded to clinical data and findings on other modalities scored the respective scans. PET scans were dichotomously scored for PET-positivity. MRI scans were scored in the spine for fatty lesions, erosions, ankylosis and bone marrow edema (BME) and erosions, ankylosis and BME in the SI-joints. X-rays were scored for erosions, syndesmophytes and ankylosis. Lesions on all modalities were compared with Cohen’s kappa, linear regression, and univariate and multivariate analysis with a generalized mixed model.ResultsIn total, 69 lesions with enhanced [18F]Fluoride uptake compared to local background, 257 MRI lesions and 88 x-ray lesions were observed in the spine (Table 1). Most lesions were found in the thoracic spine. PET lesions were frequently located outside the field of view (FOV) of the MRI and X-ray in respectively costovertebral and facet joints (31/69; 45%). Univariate analysis PET positivity had the strongest association with the presence of MRI BME and x-ray bridges (OR 6.6, 95% C.I. 2.0 – 21.1, p = 0.002 and OR 34, 95% C.I. 6.0 – 183, p <0.001 respectively). Multivariate analysis confirmed these findings (OR 39, 95% C.I. 7 – 223, p < 0.01 and OR 6.4, 95% 1.7 – 23.5, p < 0.01 respectively) (Figure 1). However, of all PET-positive lesions only 7/24; 29% within the FOV of the MRI showed BME and 13/23; 56% of the PET lesions within the FOV of the X-ray showed bridges on X-ray. In the SI-joints, a total of 15 lesions (15/20; 75%) with PET enhancement were found. Erosions, ankylosis and BME lesions on MRI were observed in respectively 13/15; 87%, 4/15; 27% and 13/15; 87% of the lesions with PET enhancement.Table 1.Distribution of lesions throughout the spine on different modalities.Number of lesionsCervical spineThoracic spineLumbar spineTotalPET4422369PET, within FOV MRI1111224PET, within FOV x-ray1101223MRI – fatty lesions15301560MRI – erosions0011MRI – ankylosis212127150MRI – BME1351046X-ray erosions1425039x-ray syndesmophytes1012X-ray bridges4311247FOV: field of viewFigure 1.Example of Fluoride uptake on PET (A), corresponding BME on MRI (B) and bridges on X-ray (C).ConclusionLesions with [18F]Fluoride on PET showed significant correlation to MRI BME and x-ray bridge lesions. It showed partial agreement with BME of MRI and bridges on X-ray. This suggests that [18F]Fluoride PET visualizes different mechanisms central to AS disease activity compared to MRI and X-ray, thereby providing novel and additional information.References[1]Lee SG et al. Clin Exp Rheumatol. 2015;33(1):90-7.[2]Bruijnen STG et al. Rheumatology (Oxford). 2018;57(4):770.[3]Bruijnen ST et al. Arthritis Res Ther. 2012;14(2):R71.Disclosure of InterestsJerney de Jongh: None declared, Nicki Pouw-Verweij: None declared, Gerben C.J. Zwezerijnen: None declared, J.C. van Denderen: None declared, Irene van der Horst-Bruinsma Speakers bureau: BMS, AbbVie, Pfizer, MSD, Consultant of: AbbVie, UCB, MSD, Novartis, Eli Lilly, Grant/research support from: Unrestricted Grants received for investigator initiated studies from MSD, Pfizer, AbbVie, UCB, Joost Bot: None declared, B.J.H. Boden: None declared, Robert Hemke: None declared, Frank F Smithuis: None declared, Alexandre Voskuyl: None declared, Maarten Boers Consultant of: Novartis, Maqsood Yaqub: None declared, Conny J. van der Laken: None declared