Jpn J Clin Oncol 2008;38(6)414418 doi:10.

1093/jjco/hyn045

Mucoepidermoid Carcinoma of the Head and Neck: Clinical Analysis of 43 Patients

Hiroyuki Ozawa1,2, Toshiki Tomita1, Koji Sakamoto1, Takamasa Tagawa1, Ryoichi Fujii1, Sho Kanzaki1, Kaoru Ogawa1, Kaori Kameyama3 and Masato Fujii4
Department of Otolaryngology, School of Medicine Keio University, Tokyo, 2Department of Otolaryngology, Shizuoka Red Cross Hospital, Shizuoka, 3Department of Pathology, School of Medicine Keio University and 4 Department of Otolaryngology, National Tokyo Medical Center, Tokyo, Japan
Received March 23, 2008; accepted May 15, 2008
1

Objective: It is well known that mucoepidermoid carcinoma (MEC) displays a variety of biological behaviors. While the high-grade type is a highly aggressive tumor, its low-grade counterpart usually demonstrates a more benign nature and several systems have, therefore, been proposed to grade this neoplasm. Methods: This report analyzes 43 patients suffering from head and neck MEC, who were treated in our department during the period from 1989 to 2005. The relationship between clinical and pathologic characteristics and survival rate was investigated. Results: The 5-year overall and disease-free survival rate was 62.3 and 57.2%. Multivariate analysis demonstrated that the parameters that signicantly affected survival were the patients age (P 0.040) and treatment method (P 0.011). Conclusions: The patients age and treatment method is the prognostic parameter in this study. Although complete surgical resection is the standard treatment for MEC, we should aggressively consider adjunctive radiotherapy in those cases that have a high risk of recurrence and poor prognosis. Key words: mucoepidermoid carcinoma head and neck cancer salivary gland prognosis

Downloaded from http://jjco.oxfordjournals.org/ by guest on April 17, 2012

INTRODUCTION
Mucoepidermoid carcinoma (MEC) is believed to arise from the reserve cells of excretory ducts, and the tumor consists of three cell types: epidermoid cells, mucous cells and poorly differentiated intermediate cells. It is well known that MEC displays a variety of biological behaviors, and that while the high-grade MEC is a highly aggressive tumor, its low-grade counterpart usually demonstrates a more benign nature. Several systems have been proposed to grade this neoplasm, but none has been universally accepted. A recent grading schema (Goodes grading) proposed by Auclair et al. (1) and Goode et al. (2) has been shown to be reproducible and to be predictive of the patients outcome by dening low, intermediate and high-grade tumors using ve histopathologic features. However, some patients with low-grade MECs according to Goodes grading at its early stage have occasionally developed distant metastases.
For reprints and all correspondence: Hiroyuki Ozawa, Department of Otolaryngology, Shizuoka Red Cross Hospital, 8-2 Outemachi, Aoi-ku, Shizuoka City 420-0853, Japan. E-mail: ozakky@cb.mbn.or.jp

Consequently, different investigators have proposed a variety of sub-classications and histopathologic grading criteria in order to predict clinical prognosis of MECs more accurately. We reviewed our experience with 43 cases of MEC occurring in the major and minor salivary glands, analyzed the clinical and histopathologic features of this type of tumor and attempted to correlate them with the biological behavior of the tumor.

PATIENTS AND METHOD

A total of 45 patients with MEC of the major and minor salivary glands diagnosed between 1989 and 2005 were included in the study. Of these 45 patients, 43 received treatment at the Department of Otolaryngology, Head and Neck Surgery, Keio University School of Medicine, Tokyo, Japan. The two biopsy alone performed cases were excluded from this study. All patients underwent primary treatment except two, who had previously been treated with other facilities and had had recurrence. The medical and surgical records of

# The Author (2008). Published by Oxford University Press. All rights reserved.

Jpn J Clin Oncol 2008;38(6)

415

these 43 cases were reviewed for their clinical and histopathologic features after obtaining approval from the institutional review board. Three patients who refused surgical treatment and two inoperable patients whose tumor invaded either the skull base extensively or the internal carotid artery received radical radiotherapy totaling 60 Gy. Two patients who were operated on received preoperative radiation totaling 40 Gy because tumor grew in the middle of radiotherapy, while ve patients with close margin or who showed multiple lymph node metastases based on a histologic study of their surgical specimens received postoperative radiation totaling 50 Gy. The clinicopathologic characteristics of the 43 patients are shown in Table 1. There were 26 male and 17 female patients, age ranging from 22 to 86 years, with an average age of 55.2. The median follow-up period was 874 days, with a minimum of 75 and a maximum of 4814 days. Thirty-one MECs originated from the major salivary glands, with the majority located in the parotid gland (28 cases). Twelve MECs developed in the minor salivary glands. Major salivary gland tumors were staged according to the TNM classication of malignant tumors (3). Minor salivary gland tumors were staged according to their site of origin, in a similar fashion to squamous cell carcinomas. The clinical disease stage was found to be T1 in four, T2 in 16, T3 in nine, T4 in 14, N0 in 27, N1 in two and N2 in 14 patients. No patients had N3 or M1 disease. Stage grouping according to TNM classication was: stage I, 13; stage II, 8; stage III, 3; stage IV, 19. Eighteen of the tumors were low grade, eight were intermediate and 17 were high grade according to Goodes histologic grading criteria. Thirty-one patients were treated with surgery alone, seven with both surgery and radiotherapy and ve were treated with radiotherapy alone. Operative method included total parotidectomy that were selected in the cases that MEC extend to the major part of parotid gland (12 cases), partial parotidectomy that were selected in the limited extent cases (14), total submandibulectomy (two cases), partial glossectomy containing removal of the oor of the mouth (ve cases), partial pharyngectomy (three cases), total maxillectomy (one case) and total glossectomy with total laryngectomy (one case). Neck lymph node dissection was performed on 14 cases, while six underwent some form of reconstructive surgery. Cisplatin-based chemotherapy was administered in ve cases. The efcacy of such chemotherapy was no response for all patients. All cases who received chemotherapy underwent some form of surgical treatment at some stage. The Kaplan Meier method was used to estimate the overall survival rate of all 43 cases treated, and disease-free survival rate of 40 cases excluding the three recurrent cases. The prognostic effects of sex, age (under 55 years of age that is average age of this study versus over 56), primary site (major versus minor salivary gland), TNM classication (T1, T2 and T3 versus T4; N0 versus N1 and N2), Goodes histologic grading (low versus intermediate and high grade) and treatment method (surgery alone versus surgery with

Table 1. Patient characteristics Male/female Age Index/recurred Primary site Parotid gland Submandibular gland Oral Oropharynx Larynx Epipharynx Maxillary sinus T 1 2 3 4 N 0 1 2 3 Stage I II III IV Histologic grade Low Intermediate High Treatment S S RTx RTx 31 7 5 17 10 16 13 8 3 19 27 2 14 0 4 28 3 5 4 1 1 1 26/17 2286 (average, 55.2) 40/3

Downloaded from http://jjco.oxfordjournals.org/ by guest on April 17, 2012

16 (rT2 2) 9 14 (rT4 1)

S, surgery; RTx, radiation; T and N, TNM Atlas 5th edition (2004).

radiation versus radiation alone) were tested using the log-rank test. Multivariate analysis of prognostic factors was also carried out using Cox regression model. These analyses were implemented with Stat View (version 5.0 for Windows; SAS Institute Inc., USA).

RESULTS
The 5-year overall survival rate was 62.3%, whereas the 5-year relapse-free survival rate was 57.2% (Fig. 1). In the recurrent cases, primary relapse developed in six cases,

416

Head and neck mucoepidermoid carcinoma

Figure 1. Kaplan Meier overall survival and relapse-free survival curve of 43 cases of mucoepidermoid carcinoma.

while regional lymph node recurrence without distant metastasis occurred in four patients. Lung metastasis was observed in ve cases, liver metastasis in two and brain metastasis was observed in one. Out of the 10 patients who died from their disease, ve were due to local recurrence, one due to lymph node recurrence and four due to distant metastasis. The results of the log-rank test are shown in Table 2. The group of cases aged over 56 showed poor prognosis. The

Table 2. Results of univariate analysis Subgroup Age (years) ,55 .56 Sex Male Female Primary site Major salivary gland Minor salivary gland T 123 4 N 0 12 Histologic grade Low Intermediate high Treatment S S RTx RTx 82.9 55.6 0 ,0.001 72.9 41.7 0 ,0.001 95.1 54.2 0.025 87.5 42.2 0.019 87.3 37.3 ,0.001 76.0 38.9 0.001 96.3 31.4 ,0.001 80.6 15.0 0.001 77.1 57.1 0.267 63.0 58.3 0.520 43.9 100 0.017 47.2 79.4 0.075 85.6 53.7 0.040 77.8 42.5 0.082 Overall survival (%) P value Relapse-free survival (%) P value

survival rate in this group was 53.7%, whereas those under 55 had a rate of 85.6% (P 0.040). Women had a 5-year overall survival rate of 100%, whereas for men it was 43.9% (P 0.017). With regard to tumor location, patients with a tumor in the major salivary gland had a 5-year overall survival rate of 77.1%, whereas those whose tumor was located in the minor salivary gland had a 5-year overall survival rate of 57.1% (P 0.267). The 5-year overall survival rate was 96.3% in the groups at stages T1, T2 and T3, but 31.4% in T4 (P , 0.001). The 5-year overall survival rate was 87.3% in the N0 group, but 37.3% in the N1 and N2 groups (P , 0.001). Histologic subtype was shown to affect the survival. The survival rate of the low-grade group was 95.1%, while that of the intermediate and high-grade group was 54.2% (P 0.025). The 5-year overall survival rate was 82.9% for the surgery group and 55.6% for the group that underwent surgery with radiotherapy. All patients treated with radiotherapy alone died due to their disease: there were statistically signicant differences in the survival rates between these three groups (P 0.025). Treatment modality, however, had bias because radiotherapy was selected for more extended cases whose tumor invaded either the skull base extensively or the internal carotid artery. Statistical multivariate analysis demonstrated that the parameters most signicantly affecting survival were the patients age [P 0.040, relative risk (RR) 0.071] and treatment method (P 0.011, RR 0.007 for surgery alone versus radiation; RR 0.004 for surgery with radiation versus radiation) (Table 3). Patients age (P 0.039, RR 0.216) and treatment method (P 0.003, RR 0.008 for surgery versus radiation; RR 0.003 for surgery with radiation versus radiation) also demonstrated a statistically signicant parameter with regard to relapse-free survival (Table 4).

Downloaded from http://jjco.oxfordjournals.org/ by guest on April 17, 2012

DISCUSSION
Mucoepidermoid carcinoma is the most common malignant neoplasm observed in the major and minor salivary glands among children and adults (4) and shows a wide, nearly uniform age distribution, with diminution in pediatric and geriatric life. There is a 3:2 female predilection. Approximately half of the tumors occur in the major salivary glands, with about 45% of MECs occurring in the parotid gland (4). In the present study, the 5-year overall survival of MECs occurring in the head and neck was 62.3%, while the 5-year disease-free survival was 57.2%. These data are in line with the study by Guzzo et al. (5), but slightly worse than those reported previously (6,7) as 19 of the 43 patients (44%) were at stage IV in this study. Multivariate analysis revealed that patients over 56 years of age were signicantly associated with decreased survival in our study. Goode et al. (2) reported that the average

Jpn J Clin Oncol 2008;38(6)

417

Table 3. Multivariate analysis associated with overall survival Factor Sex Age Primary site T N Histologic grade Treatment S S RTx 0.007 0.004 0.041 0.056 0.212 0.202 Subgroup Female ,55 MISG T1 2 3 N1 2 IH Relative risk 0.118 0.071 9.553 0.008 4.432 0.125 95% CI lower limit 0.005 0.006 0.566 0.010 0.276 0.023 95% CI upper limit 3.044 0.887 161.148 15.726 71.208 125.383 P value 0.198 0.040 0.117 0.213 0.293 0.555 0.011 0.004 0.006

MISG, minor salivary gland; IH, intermediate high; CI, condence interval.

Table 4. Multivariete analysis associated with relapse-free survival Factor Sex Age Primary site T N Histologic grade Treatment S S RTx 0.008 0.003 0.085 0.056 0.122 0.109 Subgroup Female ,55 MISG T1 2 3 N0 IH Relative risk 0.564 0.216 1.960 0.208 0.330 1.449 95% CI lower limit 0.095 0.05 0.386 0.038 0.048 0.209 95% CI upper limit 3.359 0.927 9.955 1.139 2.275 10.063 P value 0.529 0.039 0.417 0.070 0.260 0.709 0.003 0.006 0.002

age among the group of patients who died of their tumor was higher than among the other groups, and higher patient age correlated with a worse biologic outcome when patients with MEC of the major salivary glands were separated into one of four groups based on clinical outcome. Hicks et al. (8) reported that the mean age was lowest among patients with a histologic low grade, while those with a high grade were highest. In the present study, 42% of patients over 56 years had high-grade MEC, meanwhile 32% of patients under 55 years had high grade although not statistically signicant using the x2 test (P 0.27). Moreover, several reports have shown that the patients age is one of the prognostic factors in salivary carcinoma including MEC (9 11). Clinical stage is an important prognostic factor for MEC (5,6,12). In the present study, the results of univariate analysis showed that, according to TNM classication, T and N are signicant prognostic factors. T factor, strongly related to surgical margin and facial palsy, has been reported as a signicant factor of parotid gland cancer (11). Lymph node

metastasis is considered to be related to histologic malignancy. A high rate of metastasis was observed in MEC patients with high histologic grade as reported previously (8). In the present study, regional lymph node metastasis was observed in 24% of the patients with low histologic grade, 30% in those with intermediate grade and 56% in those with high grade. Although not statistically signicant using the x2 test (P 0.13), in the present study, high-grade MEC tended to metastasize to the lymph node. Distant metastasis was observed in the lung, liver and brain. The size of the primary tumor or histologic malignancy is considered to affect distant metastasis (12). Metastatic lesions from low-grade MEC may grow more slowly than in the case of high-grade MEC (13). Goodes grading system reects MEC progression and frequency of metastasis very well, though divergent opinions exist with regard to dening histologic classication. Brandwein et al. (12) reported that their new histologic grading system, which analyzed 89 cases of MECs, may have a better predictive value than Goodes system, and Goodes criteria tended to downgrade MEC. In the present study, MECs were classied according to Goodes grading system, and the group of patients with intermediate and highgrade MECs displayed a poor survival rate, which was shown by univariate analysis to be statistically signicant. However, we experienced two patients with low-grade MEC who were diagnosed as stage IV and had multiple neck metastases at rst medical examination and died as a result of distant metastasis and we therefore need to remember that there are cases among histologic low-grade MECs whose course is clinically very bad. The relationship between the primary site of the MEC and its prognosis is controversial. It has been reported that the differences in the primary site does not affect the survival of MEC (5,12), although submandibular MEC has been reported to have poor prognosis (14,15). In the present study, the primary site of the MEC did not affect the overall survival and did not cause a deection of histologic malignancy. Further studies with a large number of cases are needed to elucidate the relationship between primary site and prognosis. Standard treatment for MEC is surgical resection. The group of patients treated with surgery in this study showed good prognosis, and treatment modality was one of the signicant prognostic factors using multivariate analysis predictably. It is difcult to remove tumors with an adequate margin in some cases, especially T4 tumors that are too large and are localized near important organs. On the other hand, it is difcult to diagnose and grade MEC precisely from ne needle aspiration cytology (16) or preoperative biopsy (12). Therefore, MEC is often misdiagnosed as benign tumors preoperatively and is removed without appropriate margin. Such cases result in close or positive margin and show poor prognosis (5,11,17). The relationship between the study of the margin and prognosis was not assessed in this study because of the small number cases involved. It is

Downloaded from http://jjco.oxfordjournals.org/ by guest on April 17, 2012

418

Head and neck mucoepidermoid carcinoma

believed that cases of close or positive margin need some kind of different or additional treatment. Mucoepidermoid carcinoma has been considered a radioresistant tumor, though postoperative radiation is thought to be effective. Postoperative radiotherapy for MEC patients with positive surgical margin has been reported to decrease local failure (13,18). Radical surgery followed by postoperative radiotherapy for salivary gland malignancies has improved local control (19), but it is difcult to control parotid gland cancer by radiotherapy alone (19). In the present study, patients who received radiotherapy alone all died as a result of their disease, so radiotherapy alone was not effective for MEC. The group who underwent surgery with radiotherapy had a worse survival rate than those with surgery alone because postoperative radiation was given in unfavorable cases such as advanced locoregional disease, multiple lymph node metastasis and tumors with positive surgical margins. This bias, which has already been pointed out by another retrospective study (5), has led to an underestimation of postoperative radiation. We should aggressively consider adjunctive radiotherapy in those cases that have a high risk of recurrence and bad prognosis. Currently, there is no prognostically useful regimen of chemotherapy (20). Although cisplatin-based systemic chemotherapy was administered in ve cases in the present study, the effect of such chemotherapy was no change. Four of the ve cases had additional surgery, while one case underwent chemotherapy as palliative therapy against postoperative recurrence. However, histologic high-grade MEC needs chemotherapy as adjunctive treatment to prevent local recurrence or distant metastasis, so the possibility of a new regimen of chemotherapy containing molecular target agents should be considered.

References
1. Auclair PL, Goode RK, Ellis GL. Mucoepidermoid carcinoma of intraoral salivary glands. Evaluation and application of grading criteria in 143 cases. Cancer 1992;69:2021 30. 2. Goode RK, Auclair PL, Ellis GL. Mucoepidermoid carcinoma of the major salivary glands: clinical and histopathologic analysis of 234 cases with evaluation of grading criteria. Cancer 1998;82:1217 24. 3. Wittenkind C, Greene FL, Hutter RVP, Klimpnger M, Sobin LH. TNM Atlas. 5th edn. Berlin Heidelberg New York: Springer 2004,565. 4. Goode RK, El-Naggar AK. Mucoepidermoid Carcinoma. WHO Organization Classication of Tumours. Pathology and Genetics of Head and Neck Tumours. Lyon: IARC Press 2005,21920. 5. Guzzo M, Andreola S, Sirizzotti G, Cantu G. Mucoepidermoid carcinoma of the salivary glands: clinicopathologic review of 108 patients treated at the National Cancer Institute of Milan. Ann Surg Oncol. 2002;9:688 95. 6. Plambeck K, Friedrich RE, Hellner D, Donath K, Schmelzle R. Mucoepidermoid carcinoma of the salivary glands, Clinical data and follow-up of 52 cases. J Cancer Res Clin Oncol 1996;122:177 80. 7. Boahene DK, Olsen KD, Lewis JE, Pinheiro AD, Pankratz VS, Bagniewski SM. Mucoepidermoid carcinoma of the parotid gland: the Mayo clinic experience. Arch Otolaryngol Head Neck Surg 2004;130:849 56. 8. Hicks MJ, el-Naggar AK, Flaitz CM, Luna MA, Batsakis JG. Histocytologic grading of mucoepidermoid carcinoma of major salivary glands in prognosis and survival: a clinicopathologic and ow cytometric investigation. Head Neck 1995;17:89 95. 9. Vander Poorten VL, Balm AJ, Hilgers FJ, Tan IB, Loftus-Coll BM, Keus RB, et al. The development of a prognostic score for patients with parotid carcinoma. Cancer 1999;85:2057 67. 10. Luukkaa H, Klemi P, Leivo I, Koivunen P, Laranne J, Makitie A, et al. Salivary gland cancer in Finland 1991-96: an evaluation of 237 cases. Acta Otolaryngol 2005;125:207 14. 11. Carrillo JF, Vazquez R, Ramirez-Ortega MC, Cano A, Ochoa-Carrillo FJ, Onate-Ocana LF. Multivariate prediction of the probability of recurrence in patients with carcinoma of the parotid gland. Cancer 2007;109:204351. 12. Brandwein MS, Ivanov K, Wallace DI, Hille JJ, Wang B, Fahmy A, et al. Mucoepidermoid carcinoma: a clinicopathologic study of 80 patients with special reference to histological grading. Am J Surg Pathol 2001;25:835 45. 13. Rapidis AD, Givalos N, Gakiopoulou H, Stavrianos SD, Faratzis G, Lagogiannis GA, et al. Mucoepidermoid carcinoma of the salivary glands. Review of the literature and clinicopathological analysis of 18 patients. Oral Oncol 2007;43:1306. 14. Spiro RH, Huvos AG, Berk R, Strong EW. Mucoepidermoid carcinoma of salivary gland origin. A clinicopathologic study of 367 cases. Am J Surg 1978;136:461 8. 15. Evans HL. Mucoepidermoid carcinoma of salivary glands: a study of 69 cases with special attention to Histologic grading. Am J Clin Pasthol 1984;81:696701. 16. Kumar N, Kapila K, Verma K. Fine needle aspiration cytology of mucoepidermoid carcinoma. A diagnostic problem. Acta Cytol 1991;35:357 9. 17. Healey WV, Przin KH, Smith L. Mucoepidermoid carcinoma of salivary gland origin. Classication, clinical-pathologic correlation, and results of treatment. Cancer 1970;26:368 88. 18. Hosokawa Y, Shirato H, Kagei K, Hashimoto S, Nishioka T, Tei K, et al. Role of radiotherapy for mucoepidermoid carcinoma of salivary gland. Oral Oncol 1999;35:10511. 19. Mendenhall WM, Morris CG, Amdur RJ, Werning JW, Villaret DB. Radiotherapy alone or combined with surgery for salivary gland carcinoma. Cancer 2005;103:2544 50. 20. Laurie SA, Licitra L. Systemic therapy in the palliative management of advanced salivary gland cancers. J Clin Oncol 2006;24:2673 8.

Downloaded from http://jjco.oxfordjournals.org/ by guest on April 17, 2012

CONCLUSION
In conclusion, MECs display a variety of biological behaviors and a variable natural history. Many investigators have tried to dene histologic or clinical features that have prognostic signicance. The results from the present study suggest that the patients age is the parameter that is of the greatest prognostic signicance. Standard treatment for MEC is surgical resection, if possible, with an adequate margin. We believe that cases that have a high risk of recurrence due to a close margin or histologic high or intermediate grade need to undergo postoperative radiotherapy.

Conict of interest statement The authors have no involvements that might raise the question of bias in the work reported or in the conclusions, implications, or opinion stated.