Nasopharyngeal carcinoma (NPC) occurs sporadically in Western countries but is endemic in East Asia.1 The disease has a distinct geographical distribution in Southern China, which accounts for more than 80% of the global NPC burden. The Chinese gross mortality of NPC is 1.34 per 100 000, which comprises 1.61% of tumor-caused deaths and ranks ninth in China.2
Radiation therapy remains the mainstay of treatment strategies of NPC because it is very radiosensitive. Treatment results for NPC have improved 7%-25% in the local context in recent decades. This advancement has been attributed to earlier detection, improved staging accuracy with computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography and computed tomography (PET-CT), and modern radiation therapy techniques.3–5 However, approximately 19%—56% patients develop recurrent disease at 5 years after their primary treatment.6,7 The result of salvage treatments for these patients still remains poor.
Recognition of post-radiation changes and early detection of relapse are important in improving the outcome of NPC. Compared with CT scan, which has been more extensively applied in the clinic, MRI and PET can depict residual or recurrence more accurately. There have been extensive reports about imaging features applied to facilitate the stage classification of NPC prior to radiation therapy,8–19 whereas post-radiation changes have not been fully described in the follow-up.
In this study, we retrospectively reviewed the treatment results in a group of patients with NPC, and investigated the relationship between post-radiation changes and recurrent diseases related to NPC.
Between January 1990 and June 2007, CT scans of 510 pre- and post-therapy patients with pathologically proven NPC, who had intact clinical and imaging files, were retrospectively reviewed. Those patients included 391 men (76.7%) and 119 women (23.3%) with an age range from 11 to 77 years (mean 47.1 years, median 48 years).
The pathological diagnosis: poorly differentiated squamous cell carcinoma (SCC) in 479 cases (93.9%), undifferentiated carcinoma in 17 (3.3%), moderately differentiated SCC in 7 cases (1.4%), keratinizing SCC in 5 cases (1.0%), poorly differentiated small round cell cancer, and spindle cell cancer in 1 case each (0.2%).
All patients were treated with the conventional radiation therapy for primary carcinomas of the nasopharynx. External-beam radiation was administered with or without brachytherapy at a mean dose of 70 Gy (range, 65–80 Gy). Five fractions of 1.8 to 2.0 Gy per fraction were given to the patients at a weekly interval.
CT scans were performed with the GE 9800, the PQ6000 and the GE Lightspeed Ultra 8 helical scanners. Of the cases, 405 (79.4%) were studied with the GE 9800, 10 (2.0%) were with the PQ 6000, and 95 (18.6%) were with the GE Lightspeed Ultra 8 helical scanners. Scans were performed from a level of about 3 to 5 cm above the base of the skull extending down to the upper neck in 510 cases (100%), the middle neck in 356 (69.8%), and the lower neck in 154 (30.2%).
The common scan was obtained with a 5-mm slice thickness and a 5-mm slice gap, while the helical scan was with a 5-mm-thickness, a 5-mm-interval, a pitch of 0.875, 120 kV, and 180 mA. Axial and coronal scans were performed with GE 9800 and PQ 6000, while axial scans were with an helical scanner; plus or reconstructed coronal and multiplanner reformatted sagittal images. Direct enhanced scans were performed in 437 cases; these scans were obtained immediately after a bolus intravenous injection of a contrast medium (60%, 100 ml) using a power injector at 3.0 ml/s with a delay of 30 seconds.
All images were retrospectively reviewed by two experienced radiologists. Post-radiation changes and recurrent manifestations were carefully analyzed, and the relationship between tumor classification, staging, and prognosis were evaluated. All data were statistically processed with a SPSS 13.0 software. The chi square test was used to show significant differences between the groups. P values of <0.05 were considered statistically significant.
In more than 2 years, 1829 scans had been obtained for the follow ups of the 510 cases. All cases had CT scans before irradiation, one scan or more during or after irradiation with two to twelve CT scans per case. The duration of the follow up varied from 24 to 184 months (mean 51.7 months, median 41 months).
The diagnostic criteria for post-radiation changes of NPC on the CT scan: (1) normal: the mass disappeared completely. (2) slightly thickening (Figures 1 and 2): the mass nearly disappeared and the nasopharyngeal mucosa were slightly thickening (smaller than 3 mm). (3) moderately thickening (Figures 3 and 4): the mass was shrunk by more than two thirds.
The diagnostic criteria for recurrence after irradiation: (1) A lesion was detected on the CT scan beyond six months after radiation therapy, and lesions which did not regress completely in 6 months were deemed to be persistent or progressive. (2) confirmation with nasopharyngeal biopsy (57 cases) or the nasopharyngeal mass, which grew larger than that on a previous CT scan, reduced after re-irradiation (8 cases). (3) lymph node metastasis proved by biopsy or operation (22 cases); or lymph nodes growing larger than before, according to the clinic and CT imaging (3 cases).
Of the 510 cases, a locally recurrent disease after irradiation were found in 75 cases (14.7%), including recurrence at primary sites in 65 cases (Figures 2–4) with relapse time from 6 to 94 months (mean 32.6 months, median 29 months), cervical lymph nodes metastasis in 25 cases with a relapse time of 6 to 128 months (mean 29.5 months, median 24.5 months), and both in 15 cases (Figures 5 and 6).
According to the 2002 AJCC malignant tumor TNM classification and staging system, 157 cases were in the T1 stage (30.8%), 173 cases were in T2 (33.9%), 115 cases were in T3 (22.5%) and 65 cases were in T4 (12.7%).
Local nasopharyngeal recurrence after radiation occurred in 9 cases in the T1 stage (5.7%), 23 cases in T2 (13.3%), 18 cases in T3 (15.7%), 15 cases in T4 (23.1%). There was a distinct difference for the local recurrence between cases in the T1 stage and the T4 stage (χ2 =14.103 and P=0.003).
Lymph node metastasis after radiation occurred in 8 cases in the T1 stage (5.1%), 7 cases in T2 (4.0%), 5 cases in T3 (4.3%), 5 cases in T4 stage (7.7%). There was no significant difference for lymph node metastasis between cases in the T1 stage and the T4 stage (χ2 =1.446 and P=0.695).
The relationship between post-radiation changes and recurrence at primary sites
As revealed on CT scans taken within 3 months, during the 4th to the 6th month, and beyond 7 months after radiation therapy, the proportions of cases in which the nasopharyngeal cavity recovered to normal appearance or with only a slight wall-thickening were 24.6% and 68.9%, 47.5% and 47.5%, 56.2% and 28.6%, respectively. The detailed post-radiation changes were demonstrated in Table 1.
The relapse rates in cases in which the nasopharyngeal walls were displayed on CT scans as normal, with a slight or moderate thickening, or with obvious residual masses within the initial 3 months were 7.1%, 11.7%, 23.5% and 23.1% respectively. The degree of tumor regression had no significant relevance to the risk of recurrence during this term (P=0.094). The relapse rates in cases in which the nasopharyngeal walls were displayed on CT scans as normal, with a slight or moderate thickening, stable, or with larger masses during the 4th to the 6th month were 6.6%, 19.7%, 40%, 100% and 100%, respectively. The relapse rates in the cases displaying normal, slight or more than a moderate thickening during the 7th to the 12th months were 9.3%, 20%, 88.9%, while 4.0%, 12.7%, and 100% were the statistics for cases beyond 13 months after radiation. The degree of tumor regression beyond 3 months after radiation therapy had a considerable reverse relevance to the risk of recurrence (P=0.000), tabulated in Tables 2–6.
The relationship between post-radiation changes and cervical lymphatic metastasis
The diagnostic criterion for post-radiation changes of the cervical metastatic lymph nodes on CT scans: (1) normal: the lymph nodes disappeared completely, or were less than 5 mm with edge shrinkage and obvious enhancement (Figure 5); (2) obvious lessening by more than two thirds of the original shown by the pre-irradiation CT scan (Figure 6); (3) slight lessening by less than two thirds compared to the original shown by the pre-irradiation CT scan.
Within six months after radiation therapy, as shown by the CT scans, the metastatic lymph nodes disappeared, markedly decreased, slightly decreased or enlarged 37.4%, 51.8%, 4.7%, and 0.4% respectively. During the 6 to 12 months after radiation therapy, the proportions were 78.5%, 19.2%, 0.6% and 1.7% correspondingly. Beyond 12 months, the proportions were 83.7%, 7.9%, 0%, and 8.4% with the details depicted in Table 7.
Of the 510 cases followed up for more than 2 years, cervical lymph node metastasis was present in 25 cases. In the groups with disappearing, markedly decreased, slightly decreased or enlarged malignant nodes on CT scan within six months after radiation therapy, the relapse rates were 2.9%, 4.5%, 12.5% and 100% respectively. In the groups with disappearing and markedly decreased malignant nodes shown on the CT scan during the 6th to the 12th month after radiation therapy, the relapse rates were 2.3% and 6.1% respectively.
All of the 4 cases had recurrence during this term, including 3 cases with enlarged nodes and 1 case with slightly decreased malignant nodes. The recurrent rate was 0.5% for cases with disappearing or markedly decreased malignant nodes shown on CT scans taken beyond 12 months. In contrast, all of the 17 cases with enlarged nodes relapsed within the same term. The regression degree of the malignant nodes after radiation therapy had a remarkable reverse relevance to the risk of recurrence in lymph nodes (P=0.000). The relationship between post-radiation changes and recurrence at cervical nodes during different periods after radiation therapy is shown in Tables 8–10.
NPC is sensitive to radiotherapy. Within 3 months after irradiation, the mass gradually decreases. From the 6th month, the primary site presents tumor regression with simultaneous fibrosis, and reactive edema may remain until the 12th month. About 30%-40% of NPC sites recover to a normal appearance which manifests as nasopharyngeal symmetry with clear, homogeneous walls and normal parapharyngeal space. Because of mucositis or granulation, 50% may show thickening nasopharyngeal walls, obliterated pharyngeal recess, and small parapharyngeal space. If such signs show gradual improvement or remain stable on a dynamic imaging, post-radiation changes are suggested.20–25
Our data showed that most cases presented normal or with a slightly-thickening nasopharyngeal wall on the CT scan as the time elapsed after radiation therapy. As shown on CT scans obtained within 3 months, during the 4th to the 6th month, and beyond 7 months after radiation therapy, the proportions of cases, in which the nasopharyngeal cavity recovered to normal appearance or with a slight wall-thickening, were 24.6% and 68.9%, 47.5% and 47.5%, 56.2% and 28.6%, respectively.
The degree of the tumor regression shown on the CT scan within the initial 3 months after irradiation had no significant relevance to the risk of recurrence, which maybe attributed to incomplete reactions to irradiation. The relapse rates in the cases in which the nasopharyngeal walls displayed on CT scans as normal, with a slight or moderate thickening, or with obvious residual masses during this term were 7.1%, 11.7%, 23.5% and 23.1% respectively. However, the degree of the tumor regression beyond 3 months after radiation therapy showed a reverse correlation to the risk of recurrence (P=0.000). The relapse rates were 13.2% and 10.2% respectively in cases with normal and slightly thickening nasopharyngeal walls during the 4th to the 6th month and the 13th to the 24th month after radiation, in contrast to 62.5% and 100% of cases with moderately or more aggressively thickening walls during the same periods.
The relapse rates decreased as the tumor regressed further. For those with a complete tumor disappearance shown on the CT scan, the relapse rates within 3 months, during the 4th to the 6th month, and during the 7th to the 12th month were 7.1%, 6.6% and 9.3%, respectively. Beyond 13 months, the relapse rate for cases with a moderately thickening wall shown on the CT scan was rather high at 100% in this study. Therefore, regular follow-ups were essential for all patients no matter how the lesions regressed, especially for cases with moderately or more aggressively thickening nasopharyngeal walls.
Generally speaking, cases with a normal nasopharyngeal cavity should be followed up every half year, and cases which did not recover to completely normal should be followed up every three months. For those with a persistent tumor or suspected recurrence, MRI or repeated biopsies were recommended.17,20–22
Metastatic lymph nodes were shown to be sensitive to irradiation as well. Most lymphadenopathies disappeared or shrunk obviously. The proportions of cases with nodes disappearing and obviously reducing as shown on the CT scan within 6 months, during the 6th to the 12th month, and beyond 13 months were 94.4%, 97.7% and 91.6%, respectively. The regression degree of the malignant nodes after radiation therapy had a remarkable reverse relevance to the risk of recurrence in lymph nodes (P=0.000). In groups with vanishing, markedly decreased, slightly decreased, or enlarged malignant nodes as shown on the CT scan within 6 months, the relapse rates were 2.9%, 4.5%, 12.5% and 100% respectively. Those casesincluding 1 case with a slightly decreased node, 3 cases with enlarged malignant nodes during the 6th to the 12th month, and 17 cases with enlarged nodes beyond 12 months after radiation- all relapsed. In the 25 cases with definite lymphadenopathy beyond six months after irradiation (minimal diameter is more than 1 cm), 23 cases presented with nodal recurrence. The cases with lymphadenopathy beyond 6 months after radiation should be closely followed-up. Surgerical removal was necessary for enlarged nodes.
The baseline images within 3 months after radiotherapy and dynamic CT studies are very important to the discovering of recurrent NPC in its earlier stage. We recommend that cases exhibiting a completely normal appearance should be followed up every half year, whereas those with an abnormality shown on a CT scan should be closely followed up every 3 months, especially for patients with moderately thickening walls. Recurrence is suspected for increscent lesions, which require early treatments once confirmed by nasopharyngoscopy and biopsy. Cases with lymphadenopathy beyond six months after irradiation should also be followed up closely. Metastasis is possible for the enlarged nodes as shown on a CT scan.
1. Marks JE, Phillips JL, Menck HR. The National Cancer Data Base report on the relationship of race and national origin to the histology of nasopharyngeal carcinoma. Cancer 1998; 83: 582–588.
2. Li L, Zhang S, Lu F. Research on characteristics of mortality spectrum and type composition of malignant tumors in China. Chin J Oncol (Chin) 1997; 19: 323–328.
3. Altun M, Fandi A, Dupuis O, Cvitkovic E, Krajina Z, Eschwege F. Undifferentiated nasopharyngeal cancer (UCNT): current diagnostic and therapeutic aspects. Int J Radiat Oncol Biol Phys 1995; 32: 859–877.
4. Yao Y, Wu GH, Xia SA, Zhou RH, Han ZW, Sun H. Influence of CT/MRI fusion image on target volume and 3-D conformal radiotherapy in nasopharyngeal carcinoma. Chin Oncol (Chin) 2006; 16: 498–502.
5. Wang XL, Yin JL, Li H, Li XD, Quan JT. PET/CT imaging of delayed radiation encephalopathy following radiotherapy for nasopharyngeal carcinoma. Chin Med J 2007; 120: 474–478.
6. Sanguineti G, Geara FB, Garden AS, Tucker SL, Ang KK, Morrison WH, et al. Carcinoma of the nasopharynx treated by radiotherapy alone: determinants of local and regional control. Int J Radiat Oncol Biol Phys1997; 37: 985–996.
7. Sutton JB, Green JP, Meyer JL, Louie D, Heltzel M, Karp AH. Nasopharyngeal carcinoma. A study examining Asian patients treated in the United States. Am J Clin Oncol 1995; 18: 337–342.
8. Ching AS, Chong VF, Khoo JB. CT evaluation of nasolacrimal relapse of nasopharyngeal carcinoma. Clin Radiol 2003; 58: 642–647.
9. Kim YI, Han MH, Cha SH, Sung MW, Kim KH, Chang KH. Nasopharyngeal carcinoma: posttreatment changes of imaging findings. Am J Otolaryngol 2003; 24: 224–230.
10. Chong VF, Fan YF, Khoo JB. Nasopharyngeal carcinoma with intracranial spread: CT and MR characteristics. J Comput Assist Tomogr 1996; 20: 563–569.
11. Chong VF, Fan YF. Skull base erosion in nasopharyngeal carcinoma: detection by CT and MRI. Clin Radiol 1996; 51: 625–631.
12. Chua DT, Sham JS, Kwong DL, Au GK, Chov DT. Retropharyngea lymphadenopathy in patients with nasopharyngeal carcinoma: a computed tomography-based study. Cancer 1997; 79: 869–877.
13. Yabuuchi H, Fukuya T, Murayama S, Sakai S, Okamura J, Fukuda T, et al. CT and MR features of nasopharyngeal carcinoma in children and young adults. Clin Radiol 2002; 57: 205–210.
14. Cai AQ, Zhou SH, Lin WB, Zhou XG. CT evaluation of 506 cases with nasopharyngeal carcinoma before radiotherapy. J Shantou U Med Coll (Chin) 2000; 13: 38–43.
15. Wei Y, Xiao JH, Zhou XP, Luo M. A study of pterygopalatine fossa involvement in nasopharyngeal carcinoma on CT. Chin J Radiol (Chin) 2001; 35: 675–678.
16. Yang GF, Wang ZZ, Liu JH, Shi DP, Wang LJ, Yu BL, et al. MRI in diagnosis and staging of nasopharyngeal carcinoma. Chin J Radiol (Chin) 1995; 29: 556–558.
17. Xie CM, Liang BL, Wu PH, Zheng L, Ruan CM, Li L. Spiral computed tomography (CT) and magnetic resonance imaging (MRI) in assessment of the skull base encroachment in nasopharyngeal carcinoma. Chin J Cancer (Chin) 2003; 22: 729–733.
18. Long WS, Luo XM, Tang JY, Liang JP, Chen MQ, Liang JQ. MRI signs of nasopharyngeal carcinoma with the involvement of the posterior cranial fossa. Chin J Radiol (Chin) 2000; 19: 681–683.
19. Long WS, Tang JY, Luo XM, Li ZJ, Huang Y, Chen MQ, et al. MRI of nasopharyngeal carcinoma invading the anterior and middle cranial fossae. Chin J Radiol (Chin) 1999; 33: 122–124.
20. Chong VF, Fan YF. Detection of recurrent nasopharyngeal carcinoma: MR imaging versus CT. Radiology 1997; 202: 453–470.
21. King A, Chan P, Leung S, Tse M, Kam K, Ahuja A, et al. MR mimics of recurrent nasopharyngeal carcinoma. Eur J Radiol Extra 2005; 55: 23–27.
22. Semiz OA, Ayanoglu E, Kodalli N, Oysu C, Uneri C, Erzen C. Dynamic contrast-enhanced MRI in the differentiation of posttreatment fibrosis from recurrent carcinoma of the head and neck. Clin Imag 2005; 29: 307–312.
23. Li HG, Li QQ, Wang YJ, Liu SR, Liang MZ. Early CT diagnosis of local recurrence of nasopharyngeal carcinoma after radiotherapy. Nat Sci Med Edit (Chin) 1995; 16: 66–69.
24. Li HG, Zhuang QH, Li QQ. Initial study of correlation of tumor types based on CT with radiation dose and prognosis in nasopharyngeal tumor. Chin J Radiol (Chin) 1999; 33: 558–560.
25. Ren J, Feng B, Xu MQ, Yan HY. CT findings of the recurrence of nasopharyngeal carcinoma after radiotherapy: a report of 73 cases. J Clin Radiol 2000; 19: 687–689.