Exhibit Hall | Forum 2
Purpose: Diffusion-weighted imaging (DWI) on MR-linac systems can potentially enable adaptive radiotherapy based on treatment response in oropharyngeal cancers (OPC). We compare the repeatability, apparent diffusion coefficient (ADC) bias, and signal-to-noise ratio (SNR) of six total DWI sequences on an MR-linac and MR simulator (MR-sim).
Methods: Ten OPC patients and ten volunteers underwent DWI on a 1.5T MR-linac with EPI and two low-distortion sequences, SPLICE and TSE. Volunteers were also imaged on a 1.5T MR-sim with EPI and two low-distortion sequences, BLADE and RESOLVE. Participants underwent two scan sessions per device and two repeats of each sequence per session. Short-term and long-term within-subject coefficient of variation (wCV) of mean ADC were calculated for tumors and lymph nodes (patients) and parotid glands (volunteers) according to Quantitative Imaging Biomarker Alliance (QIBA) guidelines. Differences in measured ADC values between sequences were quantified using Bland-Altman analysis. Four sequential images were acquired of the QIBA diffusion phantom for each sequence to calculate ADC bias, repeatability metrics, and SNR, per QIBA guidelines.
Results: In vivo short-term/long-term wCV for parotids were 5.41%/6.72%, 3.83%/8.80%, 5.66%/10.03%, 3.44%/5.70%, 5.04%/5.66%, 4.23%/7.36% for EPI-MR-linac, SPLICE, TSE, EPI-MR-sim, BLADE, RESOLVE. Short-term/long-term wCV for EPI-MR-linac, SPLICE, TSE were 9.64%/10.28%, 7.84%/8.96%, 7.60%/11.68% for tumors and 7.80%/9.95%, 7.23%/8.48%, 10.82%/10.44% for nodes. Bland-Altman analysis revealed significant differences between all sequence pairs except BLADE/EPI-MR-linac and RESOLVE/SPLICE (largest difference (0.56 x10^-3 mm^2/s) for EPI-MR-sim/TSE). All sequences except TSE had phantom ADC biases within ±0.1x10^-3 mm2/s for most vials. MR-linac sequences had inconsistent ADC values between different vials with the same known ADC value, indicating spatial inhomogeneities. SNR of b=0 images was 87.3, 180.5, 161.3, 171.0, 171.9, 130.2 for EPI-MR-linac, SPLICE, TSE, EPI-MR-sim, BLADE, RESOLVE.
Conclusion: MR-linac DWI sequences, particularly SPLICE, demonstrate near-comparable performance to MR-sim sequences and are worthy of further investigation for treatment response assessment in OPC.
Funding Support, Disclosures, and Conflict of Interest: Funding: NIH/NIDCR grants R01DE028290 and 1F31DE029093. J. Christodouleas is an employee of Elekta and A. Dresner an employee of Philips, who have collaborated on this project under an academic-industrial partnership grant. C. Fuller and B. McDonald have received travel funding and speaking honoraria from Elekta AB unrelated to this project.