Session Item

Purpose/Objective: Acquiring functional magnetic resonance imaging in the form of diffusion weighted imaging (DWI) on a 1.5T MR integrated linear accelerator (MR-Linac) platform can be beneficial for longitudinal response assessments. In pelvic cancers such as bladder, cervix and rectal cancer this could inform on patients that may require treatment adaptation to improve response. Technical validation of DWI and apparent diffusion coefficient (ADC) measurements acquired on MR-Linac is required before use in future studies. We present results of a repeatability analysis performed in bladder, cervix and rectal cancers on a 1.5T MR-Linac.

Material/Methods: 30 patients with bladder, cervix or rectal cancer treated on MR-Linac had DWI acquired, as per guidelines, at radiotherapy planning scans (Day 0) and prior to treatment at first fraction (#1) to perform test-retest analysis. Gross tumour volume (GTV) was manually contoured on DWI and rigidly propagated onto ADC maps, which were created using an in-house tool. ADC mean was calculated for GTV on both Day 0 and #1 scans. Repeatability coefficient (RC), %RC and Interclass Correlation of Coefficient (ICC) were calculated.

Results: 54/60 datasets were analysed in total. Mean time difference between Day 0 and #1 scans was 19 days (range: 7-30 days). RC/%RC are 0.23/25% for test-retest analysis with bias and limit of agreement of 0.013± 0.17. ICC for all patients is 0.77.

Conclusion: The repeatability of ADC in cervix, bladder and rectal cancer on 1.5T MR-Linac appears acceptable. Future projects involving DWI on MR-Linac can be planned using above measurements as guidelines.