Primary salivary gland-type carcinomas of the nasopharynx: Prognostic factors and outcome

Abstract

Primary salivary gland-type carcinomas of the nasopharynx (SNPC) are a rare malignancy with diverse clinical behaviour and different prognoses. Previous studies have reported on limited patient populations, and few long-term studies have outlined outcomes and prognostic factors. Controversy exists regarding the treatment policy for SNPC. The aim of this study was to define management approaches, therapeutic outcomes, and prognostic factors for SNPC. The medical records of 54 patients with SNPC at one institution between 1963 and 2006 were reviewed. Patient records were analysed for management approaches, outcomes, and prognostic factors. After a median follow-up of 61.3 (1.8–245.2) months, the 2-, and 5-year overall survival rates (OS), loco-regional failure free survival rates (LRFFS) and distant failure free survival rates (DFFS) were 84.6% and 61.3%, 74.4% and 55.4%, and 92.0% and 70.0%, respectively. Multivariate analyses indicated that lymph node metastases, date of treatment, and surgical treatment were independent factors for DFFS, whereas histological subtypes and distant metastases were independent factors affecting OS. The optimal treatment policy for patients with SNPC might be surgery plus radiotherapy.

The most frequent histological type of nasopharyngeal carcinoma is nonkeratinizing squamous cell carcinoma. Primary salivary gland-type carcinomas of the nasopharynx (SNPC) are a rare malignancy that represent about 0.48% of all nasopharyngeal cancers. The clinical characteristics, management approaches, outcomes, and prognostic factors of SNPC are poorly understood. Previous studies have shown that SNPC is a rare malignancy with some special biological features, compared with nonkeratinizing squamous cell carcinoma of the nasopharynx. Previous studies and case reports have reported on limited patient populations because of the rare incidence of SNPC. Few long-term studies have outlined outcomes and prognostic factors. Controversy exists as to whether radiotherapy alone, surgical treatment alone, or combined surgery with radiotherapy is the most appropriate treatment policy. The aim of this study was to define management approaches, therapeutic outcomes, and prognostic factors for SNPC.

Materials and methods

The Case Recording System was approved to identify patients diagnosed with SNPC in the authors’ institution from January 1963 to December 2006. All were considered valuable if they contained data about patient demographics, pathological diagnosis, tumour details, therapeutic outcomes, and follow-up in the authors’ centre. Disease in all patients was restaged according to the American Joint Committee on Cancer (AJCC) classification system (2002 edition). During the years 1963 through 2006, 54 patients were diagnosed with primary SNPC. Each of the 54 patients had a confirmed pathological diagnosis of SNPC, including adenoid cystic carcinoma (ACC) in 36 patients (66.7%), mucoepidermoid carcinoma (MEC) in 11 patients (20.4%), and adenocarcinoma not otherwise specified (AC) in 7 patients (12.9%).

Tables 1 and 2 show the common presenting symptoms and the clinicopathological characteristics of the 54 patients (26 males, 28 females). Their median age was 44.5 years (range 23–74 years). According to the AJCC 2002 criteria, 18 patients had stage I and stage II disease (early stage), 36 patients had stage III and stage IV disease (advanced stage). Ten patients had neck nodal metastases at admission. In this study, 18 patients were histologically grouped into 10 with low-grade tumours and 8 with high-grade tumours in MEC and AC; in the other 36 patients, tumour differentiation was not specified.

Table 1
The patients’ symptoms.
Symptom No. of patients (%)
Epistaxis 38 (70.4)
Nasal obstruction 41 (75.9)
Tinnitus 29 (53.7)
Headache 27 (50.0)
Numb face 18 (33.3)
Diplopia 10 (18.5)
Hearing loss 31 (57.4)
Neck mass 7 (13.0)

Table 2
Clinical features and treatment.
Factor No. of patients (%) p value
All cases ACC MEC AC
Sex
Male 26 (48.1) 18 (50.0) 4 (36.4) 4 (57.1) 0.641
Female 28 (51.9) 18 (50.0) 7 (63.6) 3 (42.9)
Age
<45 y 27 (50.0) 18 (50.0) 5 (45.5) 4 (57.1) 0.890
≥45 y 27 (50.0) 18 (50.0) 6 (54.5) 3 (42.9)
Skull base invasion
No 20 (37.0) 8 (22.2) 9 (81.8) 3 (42.9) 0.002
Yes 34 (63.0) 28 (77.8) 2 (18.2) 4 (57.1)
Cranial nerve invasion
No 30 (55.6) 15 (41.7) 11 (100.0) 4 (57.1) 0.003
Yes 24 (44.4) 21 (58.3) 0 (0.0) 3 (42.9)
T stage
T1,2 19 (35.2) 8 (22.2) 8 (62.7) 3 (42.9) 0.008
T3,4 35 (64.8) 28 (77.8) 3 (27.3) 4 (57.1)
Histologic grade
Low 10 (55.6) 5 (45.5) 5 (71.4) 0.280
High 8 (44.4) 6 (54.5) 2 (28.6)
Lymph node metastases
No 44 (81.5) 31 (86.1) 10 (90.9) 3 (42.9) 0.018
Yes 10 (18.5) 5 (13.9) 1 (9.1) 4 (57.1)
Distant metastases
No 41 (75.9) 25 (69.4) 11 (100.0) 5 (71.4) 0.111
Yes 13 (24.1) 11 (30.6) 0 (0.0) 2 (28.6)
Date of treatment
Before December 1990 21 (38.9) 16 (44.4) 1 (9.1) 4 (57.1) 0.062
After December 1990 33 (61.1) 20 (55.6) 10 (90.9) 3 (42.9)
RT
No 2 (3.7) 0 (0.0) 1 (9.1) 1 (14.3) 0.107
Yes 52 (96.3) 36 (100.0) 10 (90.9) 6 (85.7)
Surgical treatment
No 30 (55.6) 22 (61.1) 2 (18.2) 6 (85.7) 0.010
Yes 24 (44.4) 14 (38.9) 9 (81.8) 1 (14.3)
Chemotherapy
No 46 (85.2) 32 (88.9) 9 (81.8) 5 (71.4) 0.463
Yes 8 (14.8) 4 (11.1) 2 (18.2) 2 (28.6)

ACC, adenoid cystic carcinoma; MEC, mucoepidermoid carcinoma; AC, adenocarcinoma; RT, radiotherapy.

In this study, 2 patient received surgery alone, 30 patient received radiotherapy alone and 22 patients had combined surgery and radiotherapy. Of them, eight patients received chemotherapy, including postoperative adjuvant chemotherapy of 1 cycle cisplatin monotherapy in one, palliative chemotherapy of 3 cycles of paclitaxel or cisplatin based chemotherapy in six patients and concurrent chemotherapy with weekly paclitaxel 60 mg in one patient.

Of the 30 patients receiving radiotherapy, 27 were irradiated using conventional techniques by lateral-opposed facial-cervical fields or preauricular fields at a daily dose of 2 Gy, and 3 recent cases were treated using intensity modulated radiation therapy at a daily dose range of 2.12–2.24 Gy to the gross tumour volume. Twenty-six patients who were treated with radiotherapy alone received a median dose of 80 (70–100) Gy. Four patients who were treated with radiotherapy alone received a total dose of 50–60 Gy because they refused further radiotherapy or developed distant metastasis. Of the 22 patients who received combined modality treatment, 7 had radiation followed by surgery (RT + S), while the other 15 received surgery prior to radiation (S + RT). In the former group, a preoperative radiation dose of 50–60 Gy was given to five of them and a high dose was delivered for two patients who were initially misdiagnosed as having nasopharyngeal squamous cell carcinoma. One was given 70 Gy, and the other was given 76 Gy by conventional external beam radiation followed by an additional boost of 15 Gy by brachytherapy and 15 Gy by stereotactic radiotherapy, but had no response. All the patients in the latter group received a total dose of 60–80 Gy according to the margin status. Forty-four patients were given radiation to the neck of 40–80 Gy regardless of the status of nodal metastasis and 1 patient was given radiation to the neck of 30 Gy because of distant metastasis.

Of the 24 patients undergoing surgery, 17 patients received primary surgery, while 7 patients received surgery for residual tumour in the nasopharynx after radiotherapy. The surgical margin was negative in 11 patients, and positive in 13 patients. Nasopharyngectomy was performed mainly via the transoropalatal or transmaxillary swing approach apart from three patients who underwent nasal endoscopic surgery. Neck dissection was performed for two patients with clinical radiographically suspected enlarged lymph nodes.

Follow-up was carried out by the oncologic outpatient clinic and by contact with the patients or their relatives using letter and/or phone call. The date of last follow-up was May 2011.

Statistical analyses

The Statistical Package for Social Sciences, version 15.0 (SPSS, Chicago, IL, USA), software was used for statistical analysis. The Kaplan–Meier method was used to calculate the overall survival (OS), loco-regional failure-free survival (LRFFS), and distant failure-free survival (DFFS) rates. OS was measured from the date of the first treatment to the date of death or last follow-up examination. LRFFS and DFFS were measured from Day 1 of first treatment to the date of the event. A Cox proportional hazards model was used to determine any significant predictors of OS, LRFFS and DFFS. Comparisons of demographic, clinical, and pathological variables between cases were performed using Pearson’s χ 2 test. Differences were considered significant for p values <0.05 and all p values were 2-tailed.

Results

The median living follow-up time for all the patients was 61.3 (1.8–245.2) months. Thirty-six (66.7%) patients were dead at the end of follow-up, 32 (59.3%) of them died of tumour disease. Twenty-two patients had loco-regional failure only; 9 patients had distant metastases only; 4 had both distant and loco-regional failure.

Loco-regional control

Twenty-six (48.1%) patients had loco-regional failure (1 in situ and cervical lymph node, 23 in situ and 2 in cervical lymph node). The actuarial LRFFS rates were 74.4% at 2 years and 55.4% at 5 years. The 2-, and 5-year LRFRS for patients with ACC, MEC, or AC were 82.5% and 63.4%, 72.7% and 45.5%, and 21.4% and 21.4% respectively, but no significant differences were found in LRFRS between the three groups ( p = 0.058). The median loco-regional failure time was 53.2 (1.8–186.7) months.

Distant metastasis

Distant metastasis developed in 13 of the 54 patients (24.1%) and 4 patients developed multiple organ metastasis. The lung was the most common site of metastasis ( n = 10), followed by liver ( n = 3), bone ( n = 2), brain ( n = 1), and subcutaneous tissue ( n = 1). The DFFS rates were 92.0% at 2 years and 71.3% at 5 years. The 2-, and 5-year DFFS for patients with ACC, MEC, or AC were 93.8% and 65.0%, 100.0% and 100%, and 68.6%and 68.6%, respectively ( p = 0.082). The median distant failure time was 54.1 (1.4–245.2) months. Patients with cervical lymph node involvement at diagnosis had a DFFS rate of 78.0% at 5 years, compared to 42.2% for those with N0 necks ( p = 0.003).

Overall survival

At the end of the follow-up, 36 of 55 patients were dead, 32 of them due to tumour disease (59.3%), 1 of profuse epistaxis, and 3 of unknown causes. The OS rates were 84.6% at 2 years and 61.3% at 5 years. The 2-, and 5-year OS for patients with ACC, MEC, or AC were 91.2% and 70.2%, 81.8% and 63.6%, and 57.1%and 14.3%, respectively ( p = 0.000).

Prognostic factors

The value of various potential prognostic factors include age, gender, T stage, histologic subtype, skull base invasion, cranial nerve invasion, histologic grade, lymph node metastases, use of radiotherapy, use of surgical treatment, date of treatment and distant metastases on predicting LRFFS, DFFS, and OS ( Figs. 1–3 ) were evaluated. The outcomes are shown in Tables 3 and 4 . In multivariate analysis, lymph node metastases, date of treatment and surgical treatment were independent factors for DFFS, no particular factors were significant predictors of loco-regional failure, whereas histologic subtype and distant metastases were observed to be independent prognostic factors for overall survival.

Fig. 1
Kaplan–Meier curve showing loco-regional control of the patients.

Fig. 2
Kaplan–Meier curve showing DFFS of the patients.

Fig. 3
Kaplan–Meier curve showing OS of the patients.

Table 3
Univariate analyses results of factors affecting survival.
Factor All cases LRFFS DFFS OS
2-Year, % 5-Year, % p value 2-Year, % 5-Year, % p value 2-Year, % 5-Year, % p value
Sex
Male 26 (48.1) 71.0 49.7 0.447 91.8 66.6 0.671 84.0 60.0 0.536
Female 28 (51.9) 89.1 60.9 92.0 74.8 85.2 62.5
Age
<45 y 27 (50.0) 80.3 55.4 0.871 96.3 72.6 0.958 84.6 65.4 0.630
≥45 y 27 (50.0) 68.5 55.1 88.1 68.3 84.6 57.7
Histologic subtype
ACC 36 (66.7) 82.5 63.4 0.058 93.8 65.0 0.082 91.2 70.2 0.000
MEC 11 (20.4) 72.7 45.5 100.0 100.0 81.8 63.6
AC 7 (12.9) 21.4 21.4 68.6 68.6 57.1 14.3
Skull base invasion
No 20 (37.0) 72.7 50.3 0.981 94.7 78.9 0.199 84.2 63.2 0.496
Yes 34 (63.0) 75.5 58.7 90.4 67.3 84.8 60.1
Cranial nerve invasion
No 30 (55.6) 75.3 56.1 0.905 93.0 77.8 0.360 86.2 65.5 0.355
Yes 24 (44.4) 73.7 54.6 90.8 62.9 82.6 56.2
T stage
Early 19 (35.2) 71.1 47.4 0.781 94.4 77.3 0.255 83.3 61.1 0.454
Advanced 35 (64.8) 76.2 60.0 90.7 67.7 85.3 61.3
Histologic grade
Low 10 (55.6) 45.0 22.5 0.067 90.0 90.0 0.529 80.0 40.0 0.220
High 8 (44.4) 72.9 58.3 87.5 87.5 62.5 50.0
Lymph node metastases
No 44 (81.5) 76.3 53.5 0.996 100.0 78.0 0.003 88.1 64.1 0.152
Yes 10 (18.5) 64.3 64.3 56.3 42.2 70.0 50.0
Distant metastases
No 41 (75.9) 66.8 54.0 0.328 84.6 61.3 0.157
Yes 13 (24.1) 100.0 57.1 84.6 61.5
Date of treatment
Before December 1990 21 (38.9) 66.6 54.5 0.304 83.7 57.4 0.223 73.7 52.6 0.101
After December 1990 33 (61.1) 78.8 56.1 96.8 77.9 90.9 64.4
RT
No 2 (3.7) 100.0 50.0 0.956 100.0 100.0 0.400 100.0 100.0 0.264
Yes 52 (96.3) 73.4 55.6 91.6 69.8 84.0 59.8
Surgical treatment
No 30 (55.6) 62.8 54.9 0.420 84.7 72.5 0.925 75.0 57.1 0.097
Yes 24 (44.4) 87.5 55.8 100.0 71.4 95.8 66.4
Only gold members can continue reading. Log In or Register to continue

Stay updated, free dental videos. Join our Telegram channel

Jan 26, 2018 | Posted by in Oral and Maxillofacial Surgery | Comments Off on Primary salivary gland-type carcinomas of the nasopharynx: Prognostic factors and outcome

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos