Background and purposeThe apparent diffusion coefficient (ADC), a potential imaging biomarker for radiotherapy response, needs to be reproducible before translation into clinical use. The aim of... Show moreBackground and purposeThe apparent diffusion coefficient (ADC), a potential imaging biomarker for radiotherapy response, needs to be reproducible before translation into clinical use. The aim of this study was to evaluate the multi-centre delineation- and calculation-related ADC variation and give recommendations to minimize it.Materials and methodsNine centres received identical diffusion-weighted and anatomical magnetic resonance images of different cancerous tumours (adrenal gland, pelvic oligo metastasis, pancreas, and prostate). All centres delineated the gross tumour volume (GTV), clinical target volume (CTV), and viable tumour volume (VTV), and calculated ADCs using both their local calculation methods and each of the following calculation conditions: b-values 0–500 vs. 150–500 s/mm2, region-of-interest (ROI)-based vs. voxel-based calculation, and mean vs. median. ADC variation was assessed using the mean coefficient of variation across delineations (CVD) and calculation methods (CVC). Absolute ADC differences between calculation conditions were evaluated using Friedman’s test. Recommendations for ADC calculation were formulated based on observations and discussions within the Elekta MRI-linac consortium image analysis working group.ResultsThe median (range) CVD and CVC were 0.06 (0.02–0.32) and 0.17 (0.08–0.26), respectively. The ADC estimates differed 18% between b-value sets and 4% between ROI/voxel-based calculation (p-values < 0.01). No significant difference was observed between mean and median (p = 0.64). Aligning calculation conditions between centres reduced CVC to 0.04 (0.01–0.16). CVD was comparable between ROI types.ConclusionOverall, calculation methods had a larger impact on ADC reproducibility compared to delineation. Based on the results, significant sources of variation were identified, which should be considered when initiating new studies, in particular multi-centre investigations. Show less
Background and purpose: A wide variation of MRI systems is a challenge in multicenter imaging biomarker studies as it adds variation in quantitative MRI values. The aim of this study was to design... Show moreBackground and purpose: A wide variation of MRI systems is a challenge in multicenter imaging biomarker studies as it adds variation in quantitative MRI values. The aim of this study was to design and test a quality assurance (QA) framework based on phantom measurements, for the quantitative MRI protocols of a multicenter imaging biomarker trial of locally advanced cervical cancer.Materials and methods: Fifteen institutes participated (five 1.5 T and ten 3 T scanners). Each institute optimized protocols for T2, diffusion-weighted imaging, T1, and dynamic contrast-enhanced (DCE-)MRI according to system possibilities, institutional preferences and study-specific constraints. Calibration phantoms with known values were used for validation. Benchmark protocols, similar on all systems, were used to investigate whether differences resulted from variations in institutional protocols or from system variations. Bias, repeatability (%RC), and reproducibility (%RDC) were determined. Ratios were used for T2 and T1 values.Results: The institutional protocols showed a range in bias of 0.88-0.98 for T2 (median %RC = 1%; % RDC = 12%), -0.007 to 0.029 x 10(-3) mm(2)/s for the apparent diffusion coefficient (median %RC = 3%; % RDC = 18%), and 0.39-1.29 for T1 (median %RC = 1%; %RDC = 33%). For DCE a nonlinear vendor-specific relation was observed between measured and true concentrations with magnitude data, whereas the relation was linear when phase data was used.Conclusion: We designed a QA framework for quantitative MRI protocols and demonstrated for a multi-center trial for cervical cancer that measurement of consistent T2 and apparent diffusion coefficient values is feasible despite protocol differences. For DCE-MRI and T1 mapping with the variable flip angle method, this was more challenging. (C) 2020 The Authors. Published by Elsevier B.V. Show less
