H ead and neck cancer is a major type of cancer, affecting 50 000 new patients in the United States every year (1). Patients with early head and neck cancers can undergo radiation therapy or surgery. Important causes of an unfavorable outcome in patients with advanced head and neck cancer can be delayed diagnosis and tardy salvage treatment at disease recurrence (1). Imaging goals for head and neck tumors include differential diagnosis of malignant tumors from benign lesions, establishment of tumor extent and size, and assessment of nodal disease. For these purposes, MRI is the preferred modality to evaluate head and neck areas because of its better contrast resolution and feasibility. Functional MRI modalities, such as diffusion-weighted (DW) imaging and dynamic contrast material-enhanced MRI provide anatomic and physiologic information, including tumor cellularity and vascularity for tumor characterization (2). However, there are several limitations to these techniques. DW images are affected by patient motion distortion and susceptibility difference in head and neck areas. Dynamic contrast-enhanced MRI is limited by the nonspecific nature of vessel leakage, which can lead to many false-negative and false-positive results (2). Thus, different imaging modalities have been applied to head and neck tumors to acquire more physiologic and molecular information.In this issue of Radiology, Law and colleagues (3) used amide proton transfer (APT) MRI to characterize head and neck tumors. They also investigated whether APT imaging has a complimentary role to DW imaging for this purpose. They enrolled 117 patients, 70 with nasopharyngeal undifferentiated carcinoma, 26 with squamous cell carcinoma, eight with non-Hodgkin lymphoma, and 13 with benign salivary gland tumors. Normal tissues were examined in 25 patients. They found that mean APT of malignant tumors (2.40% 6 0.97 [standard deviation]) was higher than that of brain (1.13% 6 0.43), muscle (0.23% 6 0.73), and benign (1.32% 6 1.20) tumors (P , .001), but no differences were observed between malignant groups. In addition, by adding APT parameters to the apparent diffusion coefficient (ADC) parameters, the area under the receiver operating characteristic curve could increase from a range of 0.84-0.87 to a range of 0.89-0.96 (P = .13 to P = .30).APT MRI is as an emerging molecular MRI technique based on the chemical exchange processes between free water and mobile amide protons of protein and peptide backbones. APT MRI depicts small variations in amide proton concentration associated with the tumor protein and peptide backbones. It is a potential tool of tumor detection and characterization, especially in the brain (4). In this study, Law et al asked, "Can APT MRI be feasible and useful in the evaluation of head and neck tumors?" The previous study by this author group (4) showed the feasibility and repeatability of performing APT MRI in head and neck areas at a clinical field strength of 3 T in healthy volunteers. In addition, they found that head and neck tumors had ...