Cost-Effectiveness Analysis of Routine Magnetic Resonance Imaging in the Follow-Up of Patients With Nasopharyngeal Carcinoma After Intensity Modulated Radiation Therapy

Purpose: To estimate the clinical benefit and cost-effectiveness of routine head and neck magnetic resonance imaging (MRI) in the follow-up of patients with nasopharyngeal carcinoma after definitive intensity modulated radiation therapy. Patients and Methods: Two Markov models were developed to compare the cost and effectiveness of 3 strategies: routine clinical surveillance without serial imaging and routine annual and biannual MRI surveillance in the first 5 years. Two hypothetical cohorts of patients with primary stage T1-2 or T3-4 nasopharyngeal carcinoma who achieved complete remission after radical treatment and remained asymptomatic were analyzed. Baseline probabilities, transition probabilities, utilities, and costs were derived from published studies. Markov models were used to calculate life-time costs, qualityadjusted life years (QALYs), and incremental cost-effectiveness ratios (ICERs). Model robustness was addressed via univariable and probabilistic sensitivity analyses. Results: In T1-2 patients, surveillance strategies utilizing routine MRI provided few QALYs compared with non-MRI clinical follow-up (annual MRI, 0.022 QALYs; biannual MRI, 0.035 QALYs), whereas the costs associated with MRI surveillance were considerable. Compared with clinical follow-up, the ICERs for annual and biannual MRI strategies were $ 328,389 and $ 403,857 per QALY. In T3-4 patients, annual and biannual MRI surveillance provided 0.052 and 0.088 incremental QALYs, with ICERs of $ 156,204 and $ 169,772 per QALY, respectively. Model conclusions were robust and remained stable in 1-way and probabilistic sensitivity analyses. Conclusions: Routine MRI surveillance was not cost-effective owing to the high cost of MRI coupled with low rates of failure in T1-2 patients, whereas annual MRI surveillance was the dominant and possibly a cost-effective strategy for T3-4 patients, depending on the social willingness to pay. (C) 2018 Elsevier Inc. All rights reserved.