Utility of magnetic resonance imaging in determining treatment response and local recurrence in nasopharyngeal carcinoma treated curatively

BackgroundTo determine the optimal timing of the first Magnetic Resonance Imaging (MRI) scan after curative-intent radiotherapy (RT) for nasopharyngeal carcinoma (NPC), and evaluate the role of MRI in surveillance for locoregional recurrence (LRR).MethodsPatients with non-metastatic NPC treated radically who had at least one post-treatment MRI (ptMRI) done were included for analysis. ptMRI reports were retrospectively reviewed and categorised as complete response (CR), partial response/residual disease (PR) or indeterminate (ID). Patients with LRR were assessed to determine if initial detection was by MRI or clinical means. Univariable and multivariable Cox proportional hazard regression analysis were performed to identify independent factors associated with CR on ptMRIs.ResultsBetween 2013 and 2017, 262 eligible patients were analysed, all treated with Intensity Modulated Radiotherapy (IMRT). Median time from end of RT to the first ptMRI was 93days (range 32-346). Of the first ptMRIs, 88 (33.2%) were CR, 133 (50.2%) ID, and 44 (16.6%) PR. A second ptMRI was done for 104 (78.2%) of 133 patients with ID status. In this group, 77 (57.9%) of the subsequent MRI were determined to be CR, 21(15.8%) remained ID and 6 (4.5%) PR. T1 tumour stage and AJCC stage I were associated with increased CR rates on first ptMRI on multivariable analysis. ID status was more likely at 75-105days (3months +/-15days) vs 106-135days (4months +/-15days) post RT (OR 2.13, 95% CI 1.16-4.12, p =0.024). LRR developed in 27 (10.1%) patients; 20 (74.1%) were initially detected through MRI, 3 (11.1%) by nasoendoscopy and 2 (7.4%) by PET-CT.ConclusionMRI is useful for detecting local recurrence or persistent disease after curative-intent treatment. Most patients will need more than one ptMRI to arrive at a definitive status. The rate of ID ptMRI may be reduced by delaying the first scan to around 4months post RT.