GRADING OF MRI-DETECTED SKULL-BASE INVASION IN NASOPHARYNGEAL CARCINOMA AND ITS PROGNOSTIC VALUE

Transcription

1 ORIGINAL ARTICLE GRADING OF MRI-DETECTED SKULL-BASE INVASION IN NASOPHARYNGEAL CARCINOMA AND ITS PROGNOSTIC VALUE Lei Chen, MD, 1 Li-Zhi Liu, MD, 2 Yan-Ping Mao, MD, 1 Ling-Long Tang, MD, 1 Ying Sun, MD, PhD, 1 Yong Chen, MD, 1 Ai-Hua Lin, MD, PhD, 3 Li Li, MD, PhD, 2 Jun Ma, MD 1 1 State Key Laboratory of Oncology in Southern China, Department of Radiation Oncology, Cancer Center, Sun Yat-sen University, Guangzhou, People s Republic of China. majun2@mail.sysu.edu.cn 2 State Key Laboratory of Oncology in Southern China, Imaging Diagnosis and Interventional Center, Cancer Center, Sun Yat-sen University, Guangzhou, People s Republic of China 3 Department of Medical Statistics and Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou, People s Republic of China Accepted 9 July 2010 Published online 10 November 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: /hed Abstract: Background. Our aim was to grade MRI detected skull-base invasion in nasopharyngeal carcinoma and evaluate the prognostic value of the grading. Methods. The MRI scans and medical records of 924 patients with histologically diagnosed nondisseminated nasopharyngeal carcinoma were reviewed retrospectively. Results. MRI-detected skull-base invasion was not found to be an independent prognostic factor for overall survival, distant metastasis-free survival, or local relapse-free survival (p >.05 for all). Grading of skull-base erosion according to the site of invasion was found to be an independent prognostic factor for both overall survival (p ¼.003 and p ¼.006, respectively) and distant metastasis-free survival (p ¼.001 for both) in the 512 patients with skull-base invasions and 315 patients with T3 disease. Conclusions. MRI-detected skull-base invasion is not an independent prognostic factor for nasopharyngeal carcinoma. However, grading according to the site of invasion as either low grade or severe has prognostic value.. VC 2010 Wiley Periodicals, Inc. Head Neck 33: , 2011 Keywords: nasopharyngeal carcinoma; skull-base invasion; prognostic value; magnetic resonance imaging; grading Skull-base invasion of nasopharyngeal carcinoma (NPC) is the result of posterosuperior extension of the tumor. The incidence of skull-base erosion, on the basis of CT, occurs in between 30% and 40% of cases. 1,2 This Correspondence to: J. Ma Lei Chen and Li-Zhi Liu contributed equally to this study. This work was supported by grants from the Science Foundation of Key Hospital Clinical Program of Ministry of Health P.R. China (No ), the Hi-Tech Research and Development Program of China (No. 2006AA02AA404) and the International Cooperation Foundation of Guangdong Science and Technology Department of China (No. 2008B ). VC 2010 Wiley Periodicals, Inc. erosion is considered to be 1 of the unfavorable prognostic factors 3,4 and has been classified as T3 according to the staging system of NPC in the seventh edition of the American Joint Committee on Cancer. 5 This classification was primarily based on retrospective studies of patients treated in the 1980s and 1990s, during which CT was the predominant diagnostic tool for evaluating the extent of tumors. However, it has now become common practice to use MRI for pretreatment tumor evaluation. MRI has proven to be a more valuable tool than CT in the diagnosis of NPC and, accordingly, is likely to have an influence on the NPC staging. 6,7 It is more sensitive in detecting the early infiltration of the tumor cell into bone marrow, thus making it easier to identify skull-base abnormalities; consequently, their incidence has increased to between 50% and 70%. 8,9 Moreover, with advances in both diagnostic and radiotherapeutic technologies and improvements in combined therapy with chemotherapy, the 5-year relative survival rate for NPC has increased from approximately 50% to 75% over the past 10 years. 10,11 Therefore it is not clear whether MRI-detected skull-base abnormalities remain a poor prognostic factor for NPC. In 2010, the National Comprehensive Cancer Network (NCCN) recommended concurrent chemoradiotherapy plus adjuvant or induction chemotherapy as the standard treatment for patients with skull-base erosions. However, recent research has indicated that some of the skull-base invasions on MRI had significantly better prognosis than the others. 12,13 Hence, further research on their prognostic value has been carried out with a large sample of MRI-detected skull-base abnormalities. In this study, we aimed to grade skull-base invasion on MRI and judge its prognosis precisely. The results of this study should aid in the individualized treatment of patients. Grading of Skull-Base Invasion HEAD & NECK DOI /hed September

2 MATERIALS AND METHODS Participants. Between January 2003 and December 2004, 943 patients with newly diagnosed, untreated, and nondisseminated NPC were included in this study. Of these, 19 patients were excluded, including 13 patients who discontinued undergoing radiotherapy, 4 patients who were unable to complete radiotherapy, 1 patient in whom metastasis was detected during treatment, and 1 patient who also had Hodgkin s disease. Thus, the data from a total of 924 patients were included in the analysis. Of these, 685 were male and 239 were female (a male-to-female ratio of 2.9:1). The median age was 45 years (range, years). Histologically, 99.0% (915/924) of the patients had World Health Organization type II or type III disease, 0.8% (7/924) had World Health Organization type I disease, and the remaining 0.2% (2/924) had adenocarcinoma. All patients underwent pretreatment evaluation including a complete history, physical examination, hematology and biochemistry profiles, MRI scanning of the nasopharynx and neck, chest radiography, and abdominal sonography. To confirm that the disease was nondisseminated, 37.7% (349/924) patients underwent emission CT, which is equivalent to bone scintigraphy, including all patients with N2-N3 disease, and 6.1% (56/924) underwent positron emission tomography CT. Medical records and imaging studies were retrospectively analyzed, and all patients were staged according to the staging system of the seventh edition of the American Joint Committee on Cancer. The stage distribution for the patients was as follows: 29.8% (275/924) had a T1 classification, 14.6% (135/924) had a T2 classification, 34.1% (315/924) had a T3 classification, and 21.5% (199/924) had a T4 classification; 14.9% (138/924) had a N0 classification, 55.4% (512/ 924) were N1, 20.6% (190/924) were N2, and 9.1% (84/ 924) were N3; 7.6% (70/924) had stage I disease, 24.4% (225/924) had stage II disease, 39.3% (363/924) had stage III disease, and 28.8% (266/924) had stages IVA- B disease. Imaging Protocol. All patients underwent MRI with a 1.5-Tesla system (Signa CV/I; General Electric Healthcare, Chalfont St. Giles, United Kingdom). The region from the suprasellar cistern to the inferior margin at the sternal end of the clavicle was examined with a head and neck combined coil. T 1 -weighted fast spin-echo images were obtained in the axial, coronal, and sagittal planes (repetition time of ms and echo time of ms, 2 excitations, a 22-cm field of view, and a frequency matrix); T 2 - weighted fast spin-echo images in the axial plane (repetition time, ms and echo time, ms, 1 excitation, a 22-cm field of view, and a frequency matrix) were obtained before injecting the contrast material. After the intravenous administration of gadopentetate dimeglumine (Gd-DTPA; Magnevist, Schering, Berlin, Germany) at a dose of 0.1 mmol/kg body weight, spin-echo T 1 -weighted axial and sagittal sequences, and spin-echo T 1 -weighted fat-suppressed coronal sequences were performed sequentially with the same parameters that were used before Gd-DTPA injection. A section thickness of 5 mm and a matrix size of were used. Image Assessment. Two radiologists, both with a clinical focus in head and neck cancers, certifications of professional diagnostic imaging in China, and more than 10 years of experience, evaluated the MRI studies separately. Any disagreements were resolved by consensus. Skull-base invasion on MRI was established by the presence of low-intensity tissue in highsignal bone marrow on the T 1 -weighted image and Gd-DTPA enhancement of the abnormal tissue 8,13 MRI findings of skull-base invasion were assessed at the following sites: pterygoid process, base of sphenoid bone, petrous apex, clivus, foramen lacerum, pterygoid canal, great wing of sphenoid bone, pterygopalantine fossa, foramen rotundum, foramen ovale, jugular foramen, hypoglossal canal, internal acoustic meatus, and facial canal. Treatment. All patients were treated by definitiveintent radiation therapy. The policy was to cover the nasopharynx and the retropharyngeal lymph nodes within the primary target in every radical attempt and to treat patients with gross lymphadenopathy with whole-neck irradiation. Most patients (773/924; 83.7%) were treated with conventional techniques, 12.7% (118/924) with intensity-modulated radiation therapy (IMRT), and 3.6% (33/924) with 3-dimensional conformal radiation therapy (3D-CRT). Details regarding the radiation therapy techniques at the Cancer Center of Sun Yat-Sen University have been reported previously Most patients (517/629; 82.2%) with stage III or stage IV disease (classified as T3-T4 or N2-N3) received neoadjuvant, concomitant, or adjuvant chemotherapy in conjunction with a platinum-based therapeutic clinical trial. The remaining patients (112/629; 17.8%) did not receive chemotherapy because of advanced age, heart disease, severe diabetes, or inadequate renal or hepatic function, or if they were enrolled in the control group of a clinical trial. When possible, salvage treatments (including afterloading, surgery, and chemotherapy) were provided in the event of documented relapse or when the disease persisted despite therapy. Statistical Analysis. The follow-up period was estimated from the first day of treatment until the day the patient died or the day of the last examination. Patients were evaluated at least every 3 months during the first 2 years; thereafter, follow-up was conducted every 5 months until death. The median 1310 Grading of Skull-Base Invasion HEAD & NECK DOI /hed September 2011

3 Table 1. Incidence of invasion of each site in 512 patients with skullbase invasions according to MRI. No. of patients Site of skull-base invasion (incidence) Pterygoid process 423 (82.6%) Base of sphenoid bone 402 (78.5%) Petrous apex 351 (68.6%) Clivus 350 (68.4%) Foramen lacerum 328 (64.1%) Pterygoid canal 250 (48.8%) Foramen ovale 214 (41.8%) Great wing of sphenoid bone 202 (39.5%) Pterygopalatin fossa 154 (30.1%) Hypoglossal canal 94 (18.4%) Foramen rotundum 83 (16.2%) Jugular foramen 47 (9.2%) Internal acoustic meatus 5 (1.0%) Facial canal 0 Abbreviation: MRI, magnetic resonance imaging. follow-up period for the whole group was 55 months (range, 3 73 months). All events were measured from the date of commencement of treatment. The following end points (time to the first defining event) were assessed: overall survival (OS), local relapse-free survival (LRFS), and distant metastasis-free survival (DMFS). Local recurrence was established by fiberoptic endoscopy and biopsy or MRI. Distant metastases were diagnosed on the basis of clinical symptoms, physical examination, and imaging methods, including chest radiography, bone scanning, CT, and abdominal sonography. All statistical analyses were conducted with the Statistical Package for the Social Sciences (SPSS) 15.0 software (SPSS Inc., Chicago, IL). The actuarial rates were calculated by the Kaplan-Meier method, and differences were compared by use of the log-rank test. Multivariate analyses with the Cox proportional hazards model were used to calculate the hazard ratio (HR) and test independent significance by backward elimination of insignificant explanatory variables. Host factors (age and sex) were included as the covariates in all tests. The criterion for statistical significance was set at a ¼ 0.05, and p values were based on 2-sided test results. RESULTS Incidence of Invasion of Each Skull-Base Site. Of the 924 patients, 512 (55.4%) had skull-base invasions according to MRI. Incidence of invasion of each site in the 512 patients is shown in Table 1. Prognostic Value of MRI-Detected Skull-Base Invasion. In all of the 924 patients, significant differences were observed with regard to OS (the 5-year OS rate: 72.1% vs 85.1%, p <.001), DMFS (79.4% vs 87.4%, p ¼.001), and LRFS (86.0% vs 92.7%, p <.001) between patients with and without MRIdetected skull-base abnormalities. The following parameters, which could possibly influence the prognosis, were included in the Cox proportional hazards model for multivariate analysis: age (50 years vs <50 years), sex, nasal cavity, oropharynx, parapharyngeal space, paranasal sinuses, cranial nerve palsy, intracranial extension, orbit, masticator space, hypopharynx, N classification, technique of radiotherapy, chemotherapy and skull-base invasion on MRI. MRI-detected skull-base invasion was not found to be an independent prognostic factor for OS and DMFS. However, a marginally significant difference on LRFS was observed. The results of the multivariate analysis are shown in Table 2. Grading of MRI-Detected Skull-Base Invasion. The skull-base invasions were classified into 2 grades on the basis of the incidence of each site (Table 1). Because of the high incidence of invasion, group 1 included the pterygoid process, base of sphenoid bone, petrous apex, clivus, and foramen lacerum. Patients were classified into group 1 if they had invasions of only 1 or more sites from the group. Because of the low incidence of invasion, the pterygoid canal, great wing of sphenoid bone, pterygopalantine fossa, foramen rotundum, foramen ovale, jugular foramen, hypoglossal canal, and internal acoustic meatus were classified into group 2. Patients were classified into group 2 if they had invasions of 1 or more sites from the group. Of the 512 patients, 136 patients were classified into group 1, and 376 patients were classified into group 2. The results of the univariate analysis showed that grading on the basis of the site of invasion had Table 2. Summary of multivariate analysis of prognostic factors in 924 patients. End point Variable HR OS DMFS LRFS 95% CI for HR p value Skull-base NS NS NS N classification Parapharyngeal space Paranasal sinuses Age Intracranial extension Skull-base NS NS NS N classification Parapharyngeal space Paranasal sinuses Skull-base Intracranial extension Sex Age Oropharynx Abbreviations: HR, hazard ratio; CI, confidence interval; OS, overall survival; NS, not significant; DMFS, distant metastasis-free survival; LRFS, local relapse-free survival. Grading of Skull-Base Invasion HEAD & NECK DOI /hed September

4 FIGURE 1. The overall survival Kaplan-Meier curves for T3a and T3b. statistically significant influence on OS and DMFS. The 5-year OS and DMFS rates of group 1 were 85.3% and 90.3%, respectively, and were 67.4% and 75.2%, respectively, in group 2 (p <.001 for both). However, the grading had no significant effect on LRFS (89.6% vs 84.7%, p ¼.124). The following parameters were included in the Cox proportional hazards model for multivariate analysis: age (50 years vs <50 years), sex, T classification, N classification, technique of radiotherapy, chemotherapy and grading of skull-base invasion. The results show that grading of MRI-detected skullbase invasion is an independent prognostic factor for OS (HR ¼ 2.231, p ¼.003) and DMFS (HR ¼ 2.714, p ¼.001), but not for LRFS. Prognostic Value of the Grading in Patients with T3 Disease. Patients with T3 disease (315 patients) were divided into 2 subgroups on the basis of the grading of skull-base invasion: patients in the T3a group (132 patients, who belonged to group 1) and those in the T3b group (183 patients, who belonged to group 2). Significant differences were observed between T3a and T3b with regard to OS (the 5-year OS rate: 84.9% vs 73.8%, p ¼.006) and DMFS (91.5% vs 77.1%, p ¼.001; Figure 1 and Figure 2). However, no significant difference was observed between the groups in LRFS (89.2% vs. 91.6%, p ¼.685). Furthermore, patients with T3 disease were divided into 2 subgroups on the basis of the technique of radiotherapy: 260 patients treated with conventional techniques and 55 with IMRT or 3-DCRT. The 2 subgroups were analyzed separately. Of the 260 patients treated with conventional techniques, 108 belonged to group 1, and 152 belonged to group 2 according to the grading of skull-base invasion. Significant differences were observed with regard to OS FIGURE 2. The distant metastasis-free survival Kaplan-Meier curves for T3a and T3b. and DMFS between groups 1 and 2, but no significant difference was found between the groups in LRFS (Table 3). Of the 55 patients treated with IMRT or 3- DCRT, 24 belonged to group 1 and 31 belonged to group 2. Marginally significant differences were observed between groups 1 and 2 with regard to OS and DMFS, but, again, no significant difference was found between the groups in LRFS (Table 3). However, because of the small subgroup sample size, the test power may be insufficient. The following parameters were included in the Cox proportional hazards model for multivariate analysis: age (50 years vs. <50 years), sex, N classification, technique of radiotherapy, chemotherapy, and grading of skull-base invasion. It was observed that grading of skull-base invasion was a significant predictive factor for OS and DMFS in patients with T3 disease (Table 4). Table 3. Subgroup analysis of 315 patients with T3 disease on the basis of radiotherapy technique used (5-year different survival rates). Group 1* Group 2* p value 260 patients with T3 disease treated with conventional techniques N ¼ 108 N ¼ 152 OS rate 83.5% 73.7%.026 DMFS rate 92.5% 79.0%.004 LRFS rate 88.8% 89.8% patients with T3 disease treated with IMRT or 3D-CRT N ¼ 24 N ¼ 31 OS rate 91.5% 74.2%.093 DMFS rate 87.5% 67.7%.077 LRFS rate 90.9% 100%.135 Abbreviations: OS, overall survival; DMFS, distant metastasis-free survival; LRFS, local relapse-free survival; IMRT, intensity-modulated radiation therapy; 3-DCRT, three-dimensional conformal radiation therapy. *Subgroups were classified into group 1 and group 2 on the basis of the grading of skull-base invasion. The p value was calculated with the use of a log-rank test Grading of Skull-Base Invasion HEAD & NECK DOI /hed September 2011

5 Table 4. Summary of multivariate analysis of prognostic factors in 315 T3 patients. End point Variable HR Further Prognostic Analysis of Patients with T3 Disease with Low-Grade Skull-Base Erosions. Patients with T3 disease with low-grade erosions (T3a, 132 patients) were compared with patients with T1 disease (275 patients) and T2 disease (135 patients), respectively, with regard to prognosis. No significant differences were observed in OS, LRFS, or DMFS between T3a and T1 disease (5-year OS rate: 84.9% vs 89.2%, p ¼.341; 89.2% vs 93.4%, p ¼.172; 91.5% vs 91.7%, p ¼.885, respectively). No significant differences were observed in OS or LRFS between T3a and T2 (5-year OS rate: 84.9% vs 76.5%, p ¼.082; 89.2% vs 90.8%, p ¼.388). However, a significant difference was observed in DMFS between the groups (91.5% vs 78.2%, p ¼.004). The following parameters were included in the Cox proportional hazards model for multivariate analysis: age (50 years vs <50 years), sex, N classification, technique of radiotherapy, chemotherapy, and T classification (T3a or T1). No significant differences were observed in OS, LRFS, or DMFS between T3a and T1 disease. Likewise, no significant differences were observed in OS or LRFS between T3a and T2 disease. However, a significant difference was observed in DMFS between the groups (HR ¼ 2.489, p ¼.010). DISCUSSION 95% CI for HR p value OS Grading N classification DMFS Grading N classification LRFS Grading NS NS NS Sex Abbreviations: HR, hazard ratio; CI, confidence interval; OS, overall survival; DMFS, distant metastasis-free survival; LRFS, local relapse-free survival; NS, not significant. Why Were Different Parts of the Skull Base Significantly Different in the Incidence of Invasion? In this study, the incidence of MRI-detected skull-base invasion (55.4%) was lower than that found in previous research (approximately 70%). 8,9 This may be due to smaller samples or larger percentages of patients with category T3 and T4 disease in other studies. The skull base is divided into several different parts, and NPC has a particular pattern of expansion, which results in different incidences of invasion, and different prognoses after invasion, in the different skull base sites. The fossa of Rosenmuller is the most common site for the initial development of NPC, which often extends posterosuperiorly to the skull-base and the intracranial region or laterally to penetrate the pharyngobasilar fascia to the parapharyngeal space, carotid sheath and masticator space. 17 In this study, sites in group 2 are further from the nasopharynx than group 1. Moreover, sites of group 1 are all located in the pharyngobasilar fascia, except the clivus, of which only part is in the fascia. In contrast, group 2 sites are all outside the pharyngobasilar fascia. It has been suggested that being further from the nasopharynx and having the barrier function of the pharyngobasilar fascia may make group 2 sites more difficult to invade. 17,18 Moreover, it is supposed that the prognosis is always poor once group 2 sites are invaded. Why Was MRI-Detected Skull Base Invasion Not an Independent Prognostic Factor for NPC? In this study, which is based on a large sample, skull-base invasion on MRI was not found to be a significant factor influencing OS, DMFS, or LRFS. This was probably related to the following 2 reasons. First, with the aid of MRI, CT simulation, IMRT, and 3-DCRT, the range of local lesions could be evaluated with greater accuracy and a more rational target and field could be designed. Of the 512 patients with skull-base invasions, 438 were treated with conventional techniques and 87 (87/438; 19.9%) received boost therapy. On the basis of the above measures, we had increased the expose dose of skull base accurately, and therefore the local control had been improved. Furthermore, 398 patients (398/512; 77.7%) were treated with chemotherapy, which could also affect local control. 19 Hence, the LRFS in patients with skull-base invasions had become higher than before and was probably not poorer than in patients without skull-base invasions. 11,20 Second, MRI showed early, subtle skull-base abnormalities, which CT scanning could not detect. Most patients with such abnormalities had involvement of the pterygoid process, base of sphenoid bone and/or clivus, including some which showed only single-site invasion. 7 According to Lu et al 12 and Nishioka et al, 20 these patients have a more favorable prognosis. Why Were Different Grades Of Skull Base Invasion Significantly Different In Prognosis? Grading of skull-base invasion in this study was shown to have some prognostic utility with regard to OS and DMFS. The pterygoid canal, pterygopalantine fossa, foramen rotundum, foramen ovale, jugular foramen, hypoglossal canal, and internal acoustic meatus all belong to group 2 and are neural foramina. According to our previous study, 21 these site invasions were frequently associated with MRI-detected cranial nerve involvement. Disease with MRI-detected cranial nerve involvement has a high distant metastasis rate and poor survival rate, independent of lesion localization and symptoms. 21 Several studies have demonstrated that perineural tumor spread in other head and neck cancers may result in a higher incidence of distant metastases. 22,23 Batsakis et al 24 reported that carcinoma proliferated along the nerves within the lymphatics of the epineurium and the perineural sheaths. Grading of Skull-Base Invasion HEAD & NECK DOI /hed September

6 It is presumed that tumor proliferation within the lymphatics may increase the risk of distant metastasis, which may be one of the reasons that group 2 sites frequently lead to distant metastasis once invaded. Because distant metastasis is one of the main causes of death, it is easy to understand why the OS and DMFS of group 1 were higher than those of group 2. With the development of imaging and therapy, local control has been markedly enhanced and distant metastasis has been the main pattern of treatment failure at present. 11 This may explain why the grading had no significant effect on LRFS. Do Patients with Different Grades of Skull-Base Invasion Need Different Treatment Regimens? In 2010, the NCCN recommended that the treatment regimen for patients with NPC with MRI-detected skullbase involvement should consist of concurrent chemoradiotherapy plus adjuvant or induction chemotherapy as the standard treatment. However, the results of this study found that there were no significant differences between patients with T3a disease and those with T1 disease with regard to prognosis. Thus further studies are required with regard to whether radiotherapy alone is sufficient treatment for patients with lowgrade erosions and no regional metastasis. In contrast, patients with severe-grade erosions had a high distant metastasis rate; therefore, more intensive chemotherapy regimens should be considered for these patients. These findings will have implications for individualized treatments. Limitations of This Study. First, NPC is treated primarily with radiotherapy, and no surgical or pathologic verification of bone involvement identified through imaging studies was available in this study. This is a common and difficult problem encountered in imaging studies of skull-base lesions. Secondly, it should be noted that only 12.8% of patients were treated with IMRT and only 3.5% with 3D-CRT because of resource limitations. Although excellent local control has been achieved with IMRT, patients unfortunately continue to have development of distant metastases. 25 When included as a covariate in this study, the radiotherapy technique was not found to be an independent prognostic factor in multivariate analyses. Therefore the degree of inaccuracy that resulted from the use of different radiotherapy techniques was minimal in this investigation. CONCLUSIONS MRI-detected skull-base invasion is not an independent prognostic factor for NPC. However, grading according to the site of invasion as either low grade or severe has prognostic value. REFERENCES 1. Musa Altun, Nuri Tenekeci, Esra Kaytan, et al. Locally advanced nasopharyngeal carcinoma: computed tomography findings, clinical evaluation, and treatment outcome. Int J Radiat Oncol Biol Phys 2000;47: JS Sham, YK Cheung, D Choy, et al. Nasopharyngeal carcinoma: CT evaluation of patterns of tumor spread. Am J Neuroradiol 1991;12: Teo P, Yu P, Lee WY, et al. Significant prognosticators after primary radiotherapy in 903 nondisseminated nasopharyngeal carcinoma evaluated by computer tomography. Int J Radiat Oncol Biol Phys 1996;36: Heng DMK, Wee J, Fong KW, et al. Prognostic factors in 677 patients in Singapore with nondisseminated nasopharyngeal carcinoma. Cancer 1999;86: Stephen B. Edge, David R. Byrd, Carolyn C. Compton, April G. Fritz, Frederick L. Greene, Andy Trotti. American Joint Committee on Cancer Staging Manual. Seventh Edition, New York: Springer; Chong VF, Mukherji SK, Ng SH, et al. Nasopharyngeal carcinoma: review of how imaging affects staging. Comput Assist Tomogr 1999;23: Xin-Biao Liao, Yan-Ping Mao, Li-Zhi Liu, et al. How does magnetic resonance imaging influence staging according to AJCC staging system for nasopharyngeal carcinoma compared with computed tomography? Int J Radiat Oncol Biol Phys 2008;72: Chong VF, Fan YF. Skull base erosion in nasopharyngeal carcinoma: detection by CT and MRI. Clin Radiol 1996;51: Ng SH, Chang TC, Ko SF, et al. Nasopharyngeal carcinoma: MR and CT assessment. Neuroradiology 1997;39: Ma J, Mai HQ, Hong MH, et al. Is the 1997 AJCC staging system for nasopharyngeal carcinoma prognostically useful for Chinese patient population? Int J Radiat Oncol Biol Physiol 2001;50: Lee AW, Sze WM, Au JS, et al. Treatment results for nasopharyngeal carcinoma in the modern era: the Hong Kong experience. Int J Radiat Oncol Biol Physiol 2005; 61: Jin-Cheng Lu, Bao-Qing Wei, Wen-Zhan Chen, et al. Staging of the nasopharyngeal carcinoma investigated by magnetic resonance imaging. Radiother Oncol 2006;79: Jin-Cheng Lu, Qing Wei, Yi-Qin Zhang, et al. Influence of MRI abnormality in skull-base bone on prognosis of nasopharyngeal carcinoma. Cancer/Radiothérapie 2004; 8: Jun Ma, Lizhi Liu, Linglong Tang, et al. Retropharyngeal lymphadenopathy in nasopharyngeal carcinoma: prognostic value and staging categories. Clinical Cancer Research 2007;13: Zhao Chong, Han Fei, Lu Li-Xia, et al. Intensity modulated radiotherapy for local-regional advanced nasopharyngeal carcinoma. Ai Zheng 2004;23: Luo Wei, Deng Xiao-Wu, Lu Tai-Xiang. Dosimetric evaluation for three dimensional radiotherapy plans for patients with early nasopharyngeal carcinoma. Ai Zheng 2004;23: Mukherji SK, Pillsbury HR, Castillo M. Imaging squamous cell carcinomas of the upper aerodigestive tract: what clinicians need to know. Radiology 1997;205: King AD, Teo P, Lam WW, Leung SF, Metreweli C. Paranasopharyngeal space involvement in nasopharyngeal cancer: detection by CT and MRI. Clin Oncol (R Coll Radiol) 2000;12: Baujat B, Audry H, Bourhis J, et al. Chemotherapy in locally advanced nasopharyngeal carcinoma: an individual patient data meta-analysis of eight randomized trials and 1753 patients. Int J Radiat Oncol Biol Phys 2006;64: Nishioka T, Shirato H, Kagei K, et al. Skull-base invasion of nasopharyngeal carcinoma: magnetic resonance imaging findings and therapeutic implications. Int J Radiat Oncol Biol Phys 2000;47: Lizhi Liu, Shaobo Liang, Li Li, et al. Prognostic impact of magnetic resonance imaging-detected cranial nerve involvement in nasopharyngeal carcinoma. Cancer 2009;119: Ballantyne AJ, McCarten AB, Ibanez ML. The extension of cancer of the head and neck through peripheral nerves. Am J Surg 1963;106: Byers RM, O Brien J, Waxler J. The therapeutic and prognostic implications of nerve invasion in cancer of the lower lip. Int J Radiat Oncol Biol Physiol 1978;4: Batsakis JG. Nerves and neurotropic carcinomas. Ann Otol Rhinol Laryngol 1985;94: Kam MK, Teo PM, Chau RM, et al. Treatment of nasopharyngeal carcinoma with intensity modulated radiotherapy: the Hong Kong experience. Int J Radiat Oncol Biol Physiol 2004;60: Grading of Skull-Base Invasion HEAD & NECK DOI /hed September 2011