Prospective comparison of 18F-FDG PET with conventional imaging modalities (MRI, CT, and 67Ga scintigraphy) in assessment of combined intraarterial chemotherapy and radiotherapy for head and neck carcinoma

Abstract

To preserve the oral organs and functions in patients with head and neck carcinoma, accurate determination of the appropriate treatment after neoadjuvant chemotherapy and radiotherapy is of critical importance. We evaluated the diagnostic accuracy of (18)F-FDG PET relative to that of other conventional imaging modalities in the assessment of therapeutic response after combined intraarterial chemotherapy and radiotherapy as an organ preservation protocol.

Methods: The study was prospectively performed on 23 consecutive patients with head and neck squamous cell carcinoma who completed the treatment regimen and underwent 2 (18)F-FDG PET studies before and after neoadjuvant chemoradiotherapy. (67)Ga scintigraphy (only before therapy) as well as MRI and CT (both before and after therapy) were also performed. All images were blindly and independently interpreted without knowledge of histologic findings. The level of confidence in image interpretation was graded by means of a 5-point rating system (0 = definitely no tumor to 4 = definite tumor).

Results: Before treatment, (18)F-FDG PET detected primary tumors in all 23 patients and was more sensitive (100%) than MRI (18/23; 78.3%), CT (15/22; 68.2%), and (67)Ga scintigraphy (8/20; 40%), with a confidence level of 3 or 4 as a positive tumor finding. After chemoradiotherapy, residual tumors were histologically confirmed in 4 patients (pathologic complete response rate, 19/23; 82.6%). Although posttreatment (18)F-FDG PET showed almost equal sensitivity (4/4; 100%) compared with MRI (3/3; 100%) or CT (3/4; 75%), its specificity (17/19; 89.5%) was superior to MRI (7/17, 41.2%) and to CT (10/17; 58.8%) for primary lesions. Regarding metastases to neck lymph nodes, only specificity for posttreatment images was calculated because no metastasis was confirmed in any patients after treatment. Six subjects had (18)F-FDG PET-positive lymph nodes, which had pathologically no tumor cells and suggested an inflammatory reactive change after therapy. Therefore, the specificity of posttreatment (18)F-FDG PET (17/23; 73.9%) was almost identical to that of MRI (17/20; 85%) and CT (16/21; 76.2%) for neck metastasis. With combined chemoradiotherapy monitored with (18)F-FDG PET, 8 patients avoided surgery and the remaining 15 patients underwent a reduced form of surgery.

Conclusion: (18)F-FDG PET facilitates differentiation of residual tumors from treatment-related changes after chemoradiotherapy, which may be occasionally difficult to characterize by anatomic images. (18)F-FDG PET has a clinical impact for the management of patients with head and neck cancers after neoadjuvant chemoradiotherapy by optimizing surgical treatment for each patient and contributes to the improvement of the patient's quality of life.