¹⁸F-FDG-PET for evaluation of the response to concurrent chemoradiation therapy with intensity-modulated radiation technique for Stage T4 nasopharyngeal carcinoma

Purpose: This article evaluates [¹⁸F] fluorodeoxyglucose positron emission tomography (¹⁸F-FDG-PET) findings as a predictor for local responders (R) vs. nonresponders (NR) in nasopharyngeal carcinoma (NPC) patients with Stage T4 lesions, before and at 3 months after completion of concurrent chemotherapy and radiation therapy (CCRT).

Methods and Materials: From January 2002 to November 2003, 39 T4 NPC patients were enrolled. All had magnetic resonance imaging and ¹⁸F-FDG-PET, both before and 3 months after CCRT. Any residual/recurrent lesions were confirmed histopathologically.

Results: Of the 39 eligible patients, after a follow-up of 24.2 ± 9.5 months, 35 became disease-free and 4 had residual or recurrent disease. Marginal differences in standard uptake values (SUV) were observed (10.9 ± 5.3 vs. 15.6 ± 3.4, p = 0.058) between R and NR before treatment, and value changes of SUV before and after CCRT were not significantly different. However, highly significantly lower values of SUV were noted for R than for NR 3 months after completion of CCRT (2.1 ± 0.8 vs. 5.5 ± 3.2, p = 0.001). One hundred percent positive and negative predictive values were observed for SUV values of 4.0, set 3 months after completion of CCRT.

Conclusions: Neither the pretreatment SUV nor the changes of SUV between pretreatment and posttreatment were significant predictors for local response. SUV at 3 months after completion of CCRT was a significant determinator for local response. The cutoff of 4.0 for SUV at 3 months after completion of CCRT was useful to be offered as a diagnostic reference for recurrent or residual tumor for NPC treatment.

Introduction

Nasopharyngeal carcinoma (NPC) is endemic in southern China and parts of southeast Asia (1, 2). Despite the high overall 5-year survival, the generally poor outcomes for recurrence and considerable toxicity and side effects during and after retreatment should be considered before starting treatment (3, 4, 5, 6). Recently, salvage treatment with radiosurgery or brachytherapy for residual tumor was shown to give excellent tumor control in residual disease before the onset of recurrence (7, 8, 9, 10, 11). To improve the final outcomes, a reliable prognostic test is required to predict treatment response before its onset, to allow a more aggressive treatment strategy. The test should also permit assessment of the residual tumor status so that tumor control can be improved by recognizing and managing very early recurrence. However, different tumors have characteristically different regression rates, so experience from other cancers may not be applicable to NPC. Thus, it is essential to be able to predict and monitor the therapeutic response of NPC patients, especially if there is a high risk of local residual tumor or recurrences.

Current criteria for assessment of treatment response to cancer therapy have been recommended by the World Health Organization. The most common criterion is the change in tumor size (12, 13). Computed tomography and magnetic resonance imaging (MRI) are current modalities of choice for the measurement of volume change after therapy, but such conventional imaging has at least two problems. After successful cytotoxic therapy of the tumor, consequent volume reduction in these images is often delayed. Also, such imaging is often not sensitive enough to discriminate tumor tissue from surrounding normal tissue, fibrous tissue, and necrotic tissue in the posttherapy period (14, 15, 16, 17, 18). [¹⁸F] Fluorodeoxyglucose positron emission tomography (¹⁸F-FDG-PET) has been used to evaluate treatment response in patients with various neoplasms, including esophagus, breast, lung, colorectal, and cervical cancers, before, after, and during treatment (19, 20, 21, 22). To the best of our knowledge, there is no such report for NPC.

Previously we found that NPC patients with initial Stage T4 had a significantly higher incidence for tumor local residuum/recurrence than those with less advanced disease (5, 23). Lin et al. showed that concurrent chemotherapy and radiation therapy (CCRT) had improved overall and progression-free survival compared with radiotherapy alone in advanced NPC (24). In our experience, CCRT may cause local inflammation in the nasopharyngeal roof. Besides, ¹⁸F-FDG uptake is not tumor-specific and some inflammatory processes, with leukocyte and phagocyte accumulation, may possibly influence ¹⁸F-FDG-PET uptake, during, or for short periods after, CCRT or radiation therapy. Then it is sometimes difficult to differentiate between increased ¹⁸F-FDG uptake due to residual, viable tumor cells and uptake in normal tissues caused by radiotoxicity or cytotoxicity (25). Based on these theoretical considerations, we proposed this prospective study to evaluate the standardized uptake value (SUV) immediately before and 3 months after the completion of CCRT. We also calculated the changes of the glucose metabolism in the main tumor, based on the above two SUV determinations. This prospective study had two aims. The first was to understand if SUV was a significant predictor for local response, either before or 3 months after CCRT. The second was to determine if the changes of SUV, between the two measurements, were a reliable predictor for local response vs. nonresponse to CCRT.