Ann Surg Oncol DOI 10.1245/s10434-014-3681-y

ORIGINAL ARTICLE – HEAD AND NECK ONCOLOGY

Treatment of the Neck in Carcinoma of the Parotid Gland Safina Ali, MD, Frank L. Palmer, BA, Monica DiLorenzo, BA, Jatin P. Shah, MD, Snehal G. Patel, MD, and Ian Ganly, MD, PhD Head and Neck Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY

ABSTRACT Purpose. To review our experience in the treatment of the neck in patients with carcinoma of the parotid gland. Methods. A total of 263 patients were stratified into 3 groups: no neck dissection (NoND), elective neck dissection (END), and therapeutic neck dissection (TND). Clinicopathological characteristics of END and TND versus NoND were compared by Chi square test. Pathological positivity of each neck level was quantified. Neck recurrence–free survival was determined by Kaplan–Meier statistics. Results. There were 232 cN0 and 31 cN? patients. Of the cN0 patients, 74 had END. All cN? patients had TND. Of the END group, occult neck metastases were detected in 26 (35 %) patients. The percentage of positivity was 6.7, 28.3, 21.3, 10.8, and 6.7 % for levels I to V, respectively. Compared to the NoND group, the END group was more likely to be over 60 years old, to have cT3T4 disease, and to have disease with more aggressive histology. Of the TND group, pathological positivity was found in 87 %. The percentage of positivity was 51.6, 77, 73, 53, and 40 % for levels I to V, respectively. Patients who had diseasepositive necks had a poorer neck recurrence-free survival of 84.8 %. Conclusions. In patients with cN0 disease, observation of the neck is safe in those who are under 60 years of age with clinical T1 or T2 tumors and who have low-grade histology. END should be carried out in patients with cT3T4 disease or high-grade histology and should involve levels II to IV at a minimum. Patients with cN? disease commonly have all neck levels involved and therefore should be managed with comprehensive neck dissection.

Ó Society of Surgical Oncology 2014 First Received: 3 January 2014 I. Ganly, MD, PhD e-mail: [email protected]

There is no current consensus about the treatment of the clinically negative neck in carcinoma of the parotid gland. The decision when to carry out an elective neck dissection (END) and what levels to dissect remain poorly defined. Before 1966, our practice at Memorial Sloan-Kettering Cancer Center (MSKCC) was to carry out radical neck dissection in clinically disease-positive necks and observe patients with clinically negative necks.1 From 1966 onward, our practice changed because we observed that some patients, notably patients with disease with high T stage and high-grade histologies, experienced disease recurrence in the neck. We reported our results of END in these select patients in 1992.2,3 Since then, there have been several conflicting studies advocating END in all patients regardless of T stage and histology, with some researchers recommending comprehensive neck dissection rather than selective neck dissection.4–6 As a result of the controversy surrounding this topic, we decided to review our more recent experience in neck treatment in patients with carcinoma of the parotid gland to better define the indications and type of neck dissection in these patients. MATERIALS AND METHODS After the study received institutional review board approval, patients with previously untreated malignant salivary gland tumors treated at our institution between the years 1985 and 2009 were identified. Patients who had previous surgery, previous nonsurgical treatment (radiotherapy or chemotherapy), nonsalivary malignancies (lymphoma, sarcoma), metastases to the parotid gland, benign pathology, treatment outside of MSKCC, and incomplete notes were excluded. This left data of 301 patients available for analysis. Of these, 266 had carcinomas of the parotid gland. Three patients had M1 disease at presentation and were removed from analysis, leaving 263 patients for the study. The most common pathology was mucoepidermoid carcinoma (33 %), followed by

S. Ali et al.

carcinoma ex-pleomorphic adenoma (21 %), acinic cell carcinoma (13 %), adenocarcinoma (10 %), and adenoid cystic carcinoma (7 %). Patient, tumor, and treatment characteristics were collected by retrospective review of patient charts. Patients were stratified by neck treatment into 3 groups: no neck dissection (NoND), END, and therapeutic neck dissection (TND). After neck dissection, the specimen was routinely divided into neck levels, enabling level-specific pathology to be determined. A total of 136 men and 127 women (median age 62 years) were included in the study. Of the 263 patients, 232 had cN0 and 31 had cN? disease. Of the cN0 patients, 74 were selected to undergo END. All cN? patients underwent TND. A flow chart summarizing the treatment of the neck is shown in Fig. 1. Clinicopathological characteristics of the END and TND groups compared to the NoND group were determined by the v2 test. A p value of less than 0.05 was taken as statistically significant. The pathological positivity of each neck level was quantified, and 5-year neck recurrence-free survival (NRFS) was determined by the Kaplan–Meier statistic. Clinical and pathological characteristics of patients who had recurrence of neck disease were then analyzed on a case-by-case basis. Statistical analysis was carried out by SPSS software (IBM, Armonk, NY).

RESULTS Treatment of the cN0 Neck Comparison of the END and NoND groups are shown in Table 1. Compared to the NoND group, patients in the END group were more likely to be over 60 years of age, to have clinical stage T3T4 disease, and to have more aggressive pathology, with a greater percentage of highgrade tumors, vascular invasion, perineural invasion, and positive margins, as well as higher pT stage. Most of the pathologic data were available after the fact—that is, after the neck dissection—and thus it did not enter into the decision of selecting the patient for END. Only the clinical parameters of T stage and the histology were available before the operation, and therefore the decision for END was based only on these factors. Of the END group, occult neck metastases were detected in 26 (35 %) patients. Figure 2 shows the proportion of occult metastases stratified by histology. As expected, occult metastases were more frequent in high-grade histologies such as adenocarcinoma, anaplastic and undifferentiated cancer, high-grade mucoepidermoid carcinoma, and salivary duct carcinoma. The proportion of patients who had each neck level dissected and the

Total Population 263

cN232

cN+ 31

END 74

pN48 (65%)

No PORT 15 (31%) NR=0

PORT 33 (69%) NR=1

Observation 158

pN+ 26 (35%)

No PORT 2 (8%) NR=0

PORT 24 (92%) NR=3

No PORT 104 (66%) NR=0

TND 31

PORT 54 (34%) NR=1

pN4 (13%)

No PORT 1 (25%) NR=0

FIG. 1 Flow chart showing treatment of the neck in patients with carcinoma of the parotid gland

PORT 3 (75%) NR=0

pN+ 27 (87%)

No PORT 2 (7%) NR=0

PORT 25 (93%) NR=0

Treatment of the Neck TABLE 1 Comparison of clinical characteristics of patients Variable

NoND

END

TND

(n = 158)

(n = 74)

(n = 31)

TABLE 1 continued

NoND vs. END p value

NoND vs. TND p value

0.030

0.109

Sex 86 (54 %) 29 (39 %) 12 (39 %)

Male

72 (46 %) 45 (61 %) 19 (61 %)

Age (years) \60

84 (53 %) 28 (38 %)

[60

74 (47 %) 46 (62 %) 23 (74 %)

8 (26 %)

0.029

0.005

cT1

39 (27 %)

cT2

74 (51 %) 29 (41 %) 11 (37 %)

cT3 cT4

27 (18 %) 23 (33 %) 7 (23 %) 6 (4 %) 12 (17 %) 10 (33 %)

6 (9 %)

2 (7 %)

1

39 (27 %)

2

74 (50 %) 29 (41 %)

0 (0 %)

3

28 (19 %) 23 (33 %)

8 (26 %)

4

6 (4 %)

6 (9 %)

0 (0 %)

\0.001 \0.001

12 (17 %) 23 (74 %)

Smoker Never

68 (46 %) 31 (46 %)

Ever

79 (54 %) 36 (54 %) 20 (69 %)

9 (31 %)

0.999

0.131

Drinker Never

42 (31 %) 24 (36 %)

Ever

92 (69 %) 42 (64 %) 22 (81 %)

5 (19 %)

0.478

0.181

Facial paralysis 153 (97 %) 59 (80 %) 18 (58 %) \0.001 \0.001

Yes 5 (3 %) Skin involvement

15 (20 %) 13 (42 %)

154 (98 %) 70 (95 %) 28 (90 %)

Yes

4 (2 %)

4 (5 %)

0.264

0.054

0.599

0.886

3 (10 %)

Comorbidities No

99 (63 %) 49 (66 %) 19 (61 %)

Yes

59 (37 %) 25 (34 %) 12 (39 %)

Disease grade Low

73 (62 %)

4 (8 %)

1 (4 %)

Intermediate

13 (11 %)

4 (8 %)

2 (8 %)

High

31 (27 %) 43 (84 %) 21 (88 %)

(n = 158)

(n = 74)

(n = 31)

NoND vs. END p value

NoND vs. TND p value

79 (89 %) 25 (48 %)

Yes

10 (11 %) 27 (52 %) 20 (80 %)

Neural invasion 76 (82 %) 12 (21 %)

4 (16 %) \0.001 \0.001

17 (18 %) 44 (79 %) 21 (84 %)

Negative

65 (47 %) 24 (34 %)

Close/ positive

72 (53 %) 47 (66 %) 20 (71 %)

8 (29 %)

75 (50 %) 13 (18 %)

2 (7 %)

pT2

53 (35 %) 18 (24 %)

4 (13 %)

pT3

11 (7 %)

2 (7 %)

pT4

12 (8 %)

5 (7 %)

\0.001 \0.001

38 (51 %) 22 (73 %)

pathological positivity of each were 6.7, 28.3, 21.3, 10.8, and 6.7 % for levels I to V, respectively. Most patients whose diseases was pN? in the END group (93 %) had postoperative radiotherapy (PORT). Of patients who had END, 4 patients had recurrence of neck disease (3 pN?, 1 pN-). Of the 158 patients who had no END, 54 (34 %) patients had PORT as a result of adverse features of the primary tumor. There was 1 recurrence in this group (Fig. 1). Treatment of the cN? Neck Comparison of the TND and NoND groups is shown in Table 1. As in the END patients, those in the TND group were more likely to be over 60 years of age, to have clinical stage T3T4 disease, and to have more aggressive pathology compared to the NoND group. Of the TND group, pathological positivity was found in 87 % patients. Figure 3 shows the proportion of pathological positive nodes stratified by histology. As in END, high-grade histologies accounted for pathological positivity of neck nodes. The proportion of patients who had each neck level dissected and the pathological positivity of each was 51.6, 77, 73, 53, and 40 % for levels I to V, respectively. The majority of patients (92 %) had PORT. No patient experienced recurrent neck disease (Fig. 1). Details of Patients with Recurrence of Neck Disease

5 (20 %) \0.001 \0.001

No

pT1

\0.001 \0.001

Vascular invasion

Yes Margins

TND

\0.001 \0.001

cStage

No

END

NoND no neck dissection, END elective neck dissection, TND therapeutic neck dissection

cT

No

NoND

pT

Female

No

Variable

0.059

0.067

Table 2 and Fig. 1 show the neck failure and NRFS rates. The NoND group had an excellent 5-year NRFS of 98.7 %. END patients who were pN? had a poorer 5-year NRFS compared to pN0 patients, 80.5 versus 97.3 % (p = 0.05). Three patients who had an END experienced disease recurrence; all patients had undergone comprehensive neck dissection (levels I to V) and PORT. The reason for recurrence in these patients was aggressive pathology (high-grade adenocarcinoma, carcinoma ex-pleomorphic adenoma, salivary duct carcinoma).

S. Ali et al.

No of patients (n=74)

Histology Acinic Cell Adenocarcinoma Adenoid Cystic Anaplastic Ca Ex-Pleomorphic MEC Poorly Differentiated Myoepithelial Salivary Duct Undifferentiated

9 13 3 2 16 17 3 3 6 2

pN+ (n=26) 3(33%) 7(54%) 1(50%) 6(37.5%) 4(23.5%) 1(33%) 3(50%) 1(50%)

I 6.7%

Proportion of occult metastases stratified by histology in patients having END Neck level dissected

III 21.6%

No of patients (% ) Pathological (n=74) positivity (% ) 53(72% ) 74(100% ) 74(100% ) 47(64% ) 43(58% )

1 2 3 4 5

II 28.3%

IV 10.8%

V 6.7%

6.7% 28.3% 21.6% 10.8% 6.7%

Neck levels dissected in patients having END

FIG. 2 Proportion of metastases stratified by histology and percentage positivity per neck level in patients undergoing END

Histology Adenocarcinoma Ca Ex-Pleomorphic MEC Poorly Differentiated Salivary Duct

No of patients (n=31)

pN+ (n=27)

4 5 9 6 7

4(100%) 5(100%) 6(67%) 5(83%) 7(100%)

Proportion of pathological positive nodes stratified by histology in patients having TND I 51.6% Neck level dissected 1 2 3 4 5

No of patients (%) Pathological (n=31) positivity (%) 26(86%) 31(100%) 31(100%) 29(94%) 27(87%)

51.6% 77.0% 73.0% 53.0% 40.0%

II 77.% III 73% IV 53%

V 40%

Neck levels dissected in patients having TND

FIG. 3 Proportion of metastases stratified by histology and percentage positivity per neck level in patients undergoing TND

DISCUSSION There is no current consensus on the treatment of the clinically negative neck in carcinoma of the parotid gland.

The decision regarding when to carry out an END and what levels to dissect remains poorly defined. This is in contrast to squamous cell cancer of the head and neck, where the role of END is now well established.7 According to the

Treatment of the Neck TABLE 2 Clinicopathological details of patients with recurrent neck disease Patient no.

Age, years

Sex

cN

Neck group

ND extent

Histology

pT

pN

Port

Recurrence

1

44

M

cN0

END, pN?

Left I to V

Adenocarcinoma

T1

N2

Yes

Right IV

2

67

F

cN0

END, pN?

Right I to V

Ca Ex pleomorphic adenoma

T2

N2

Yes

Right V; left I to VII

3

61

M

cN0

END, pN?

Right I to V

Salivary duct carcinoma

T2

N2

Yes

Right III to V

4

60

M

cN0

END, pN0

Left I to III

Mucoepidermoid

T4a

N0

Yes

Left III to IV

5

73

M

cN0

Observation

None

Adenocarcinoma

T2



Yes

Right V

ND neck dissection, END elective neck dissection

literature, occult metastases are detected in 12–48 % of patients.1,6 This range in occult metastases is a reflection of the varied histologies that can occur in parotid gland carcinoma, with tumors such as salivary duct carcinoma or adenocarcinoma having high rates of neck metastases compared to tumors such as adenoid cystic cancer or acinic cell cancer, which have low rates. Before 1966, our practice at MSKCC was to carry out radical neck dissection in clinically disease-positive necks and observe patients with clinically disease-negative necks.1 Since 1966, our practice changed because we observed that some patients, notably patients with high T stage and high-grade histologies, experienced neck recurrence. We reported our results of END in select patients with parotid cancer in 1992.2,3 The study by Armstrong in 1992 reported one of the largest series of 474 patients.2 END was recommended only in patients deemed to be at high risk, namely those with T3/ T4 tumors and disease with high-grade histologies. In addition, it was concluded that neck levels II to IV were mainly at risk and should be electively dissected. However, some studies have reported the risk of neck metastases to be related to other factors.8–12 Medina summarized the indications for END to be high-grade tumors, T3T4 tumors, tumors [3 cm in size, facial paralysis, age over 54 years, extraglandular extension, and perilymphatic invasion.12 Further, some authors advocate a routine END in all cases of primary parotid carcinoma. They justify this on 2 arguments. First, there is inaccuracy in the preoperative diagnosis of cytological grade; for example, Kawata et al. 4 reported the rate at which the histological grade was accurately diagnosed preoperatively was low, especially in those with low-grade malignancy. Second, the rate of occult metastases was high, with nodal disease detected in 61 % of patients with high-grade and 23 % with intermediate-grade disease. Therefore, given the inaccuracy of preoperative fine needle aspiration and the high rate of occult metastases reported, it could be justified to carry out an END in all patients. In the present study, of 232 patients with a clinically disease-negative neck, only 74 were selected to undergo END. These patients more than 60 years old, had T3/T4 tumors, and had high-grade histology. The findings on

pathology of vascular invasion, perineural invasion, and positive margins correlated well with the preoperative impression of aggressive histology. In these patients, the occult rate of metastases was 35 %. Our data therefore suggest that END should be considered for these highergrade histologies and for any tumor that is T3 or T4. After deciding on END, one then has to decide what levels to dissect. This remains an area of controversy. Klussmann et al. 5 found that levels II, III, and V were predominantly involved but that involvement of level I and/or level IV were important risk factors for locoregional recurrence. The authors therefore recommended that END should include dissection of all neck levels, I to V. Armstrong et al.2 reported occult lymph node involvement predominately in levels II to IV. In the current study, nodal positivity was also more likely in levels II to IV. As such, the current data are in agreement with that of Armstrong et al.2 This implies that at a minimum, one should consider dissecting levels II to IV for these high-risk patients. However, consideration should be given to dissecting levels I or V, depending on the size and location of the primary tumor. For example, for cancers located more anteriorly, metastases to level I are more likely, whereas a large tumor located in the parotid tail is at risk of spread to level V. It is possible that the addition of 18-FDG PET/CT imaging may help with the necessity and extent of neck dissection in patients with salivary gland carcinoma.12 Another controversial area is whether END has any proven efficacy. In our study, a comparison of efficacy between those having NoND (observation) and those having an END is not possible as a result of the noncomparability of the 2 groups according to clinicopathological variables, but also by the fact that 69 % of our NoND group underwent PORT. Others have proposed that END has a proven benefit over observation. For example, Zbaren et al.6 reported on 41 patients who had END and 42 patients who had no END. Both groups had comparable clinical and tumor characteristics. Both groups also had comparable rates of PORT. The regional recurrence rates were 12 % in the END group and 26 % in the observation groups. They also reported the disease-free survival to be superior in the END group (86 vs. 69 %). On the basis of

S. Ali et al.

these data, the authors recommended END for all patients with parotid cancer. However, noncomparability of the END and observation groups may still be present because of the retrospective nature of the data, nonrandomization of the decision to carry out END, and nonrandomization of the decision to use PORT. Making recommendations of END for all patients must therefore be interpreted with caution. If the neck proves to be positive for disease, PORT is generally advocated. This is based on evidence from reduced recurrence of neck disease in squamous cell cancer of the head and neck treated with adjuvant PORT after neck dissection.13 In addition, studies in parotid gland cancer have shown that surgery and PORT resulted in excellent outcomes with minimal side effects and preservation of good quality of life scores.14 Armstrong et al. 2 have also reported that the rate of recurrence of neck disease was 29 % in patients with pN? disease after END who did not receive PORT, compared to 0 in those having PORT. Therefore, PORT is generally advocated for the pN? neck after END. Of more controversy is whether elective neck radiation (ENI) for the cN0 neck has equivalent efficacy to END. A recent article assessing END (n = 41) versus ENI (n = 18) for high-grade salivary carcinoma reported a 10 % recurrence rate in the END versus 0 recurrence in the ENI group.15 However, again, these patients were not randomized to these treatment arms. Therefore, confounding factors, as well as selection bias, mean that these data have to be interpreted cautiously. At our institution, END is generally carried out if patients have clinical T3/T4 disease before surgery or if preoperative biopsy reveals a high-grade malignancy. When parotidectomy is carried out without neck dissection for a presumed benign tumor or low-grade cancer, one can later find the cancer to be high grade or have other worrisome features, such as perineural or vascular invasion. In these cases, PORT is provided for the primary surgical bed, and nodal basins at risk are also included in the radiation field. From Fig. 1, we can see that 69 % of patients who did not have an END and who had cN0 disease still received PORT as a result of adverse features of the primary cancer. The control rate of neck disease in these patients was excellent, with a 5-year NRFS of 97 %. In patients who present with clinically positive neck disease, there is no debate that TND is warranted.1,6 In the current study, pathological positivity was confirmed in 87 % patients who presented with a clinically diseasepositive neck. In addition to a high rate of pathological positive nodes in levels II to IV, disease positivity in the neck was also more likely in level I (51.6 %) and level V (40 %). Therefore, it would seem prudent to carry out a comprehensive neck dissection at levels I to V for all patients with a clinically disease-positive neck.

With regard to recurrence of neck disease, the literature suggests that the risk is higher in patients with cN? disease than N0 disease. For example, Rodriguez-Cuevas et al.16 reported recurrence to be 23.5 % in N? compared to 3.2 % in N0 disease. In the current study, we found that patients who had a pathologically disease-positive neck had a poorer NRFS of 84.8 %. Interestingly, the recurrence rate in patients having a TND was 0, whereas patients with pN? neck disease after END had a higher recurrence rate (Fig. 1). This could suggest incomplete treatment of the neck by END in these patients. We therefore analyzed the 3 END pN? patients in more detail (Table 2). In all 3 patients, all levels (I to V) of the neck were dissected and all received PORT. Therefore, treatment failure in these patients was a reflection of aggressive histology rather than incomplete treatment. Our study is not without its limitations. Most important is the retrospective nature of the study and the limitations associated with such data. In particular, one can never completely account for the selection bias that is present as a result of surgeon-, radiation oncology-, and patientrelated factors that are present in determining the extent of surgery and use of PORT. Making any recommendations on treatment, especially with relatively rare histologies present in salivary gland cancers, must always be interpreted with caution. Nevertheless, our study is from an institution with over 50 years’ experience in managing such patients, and as a consequence, we report a very large series of such patients treated by a multidisciplinary team. In conclusion, we can say that in patients with cN0 disease, observation of the neck is safe in patients who are under 60 years of age with clinical T1 or T2 tumors and with disease with low-grade histology. END should be carried out in patients with cT3T4 disease or high-grade histology. At a minimum, levels II to IV should be dissected, with dissection of levels I and V being done according to size and location of the primary tumor. In patients with cN? disease, all levels of the neck are at risk of metastatic disease, and therefore comprehensive neck dissection of levels I to V is advocated. In patients who do not have an END but in whom subsequent analysis shows the primary tumor to have adverse features on pathological analysis, PORT to the upper neck is recommended in addition to the primary surgical bed. REFERENCES 1. Spiro RH, Huvos AG, Strong EW. A clinicopathologic study of 288 primary cases. Cancer of the parotid gland. Am J Surg. 1975;130:452–9. 2. Armstrong JG, Harrison LB, Thaler HT, et al. The indications for elective treatment of the neck in cancer of the major salivary glands. Cancer. 1992;69:615–9.

Treatment of the Neck 3. Kelley DJ, Spiro RH. Management of the neck in parotid carcinoma. Am J Surg. 1996;172:695–7. 4. Kawata R, Koutetsu L, Yoshimura K, Nishikawa S, Takenaka H. Indication for elective neck dissection for N0 carcinoma of the parotid gland: a single institution’s 20-year experience. Acta Otolaryngol. 2010;130:286–92. 5. Klussmann JP, Ponert T, Mueller RP, Dienes HP, GuntinasLichius O. Patterns of lymph node spread and its influence on outcome in resectable parotid cancer. Eur J Surg Oncol. 2008;34:932–7. 6. Zbaren P, Schupbach J, Nuyens M, Stauffer E. Elective neck dissection versus observation in primary parotid carcinoma. Otolaryngol Head Neck Surg. 2005;132:387–91. 7. Byers RM, Wolf PF, Ballantyne AJ. Rationale for elective modified neck dissection. Head Neck. 1988;10:160–7. 8. Herman MP, Werning JW, Morris CG, Kirwan JM, Amdur RJ, Mendenhall WM. Elective neck management for high-grade salivary gland carcinoma. Am J Otolaryngol. 2013;34:205–8. 9. Spiro RH, Armstrong J, Harrison L, Geller NL, Lin SY, Strong EW. Carcinoma of major salivary glands. Recent trends. Arch Otolaryngol Head Neck Surg. 1989;115:316–21. 10. Lima RA, Tavares MR, Dias FL, et al. Clinical prognostic factors in malignant parotid gland tumors. Otolaryngol Head Neck Surg. 2005;133:702–8.

11. Medina JE. Neck dissection in the treatment of cancer of major salivary glands. Otolaryngol Clin North Am. 1998;31:815–22. 12. Kim MJ, Kim JS, Roh JL, et al. Utility of (18)F-FDG PET/CT for detecting neck metastasis in patients with salivary gland carcinomas: preoperative planning for necessity and extent of neck dissection. Ann Surg Oncol. 2013;20:899–905. 13. Ang KK, Trotti A, Brown BW, et al. Randomized trial addressing risk features and time factors of surgery plus radiotherapy in advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys. 2001;51:571–8. 14. Al-Mamgani A, van Rooij P, Verduijn GM, Meeuwis CA, Levendag PC. Long-term outcomes and quality of life of 186 patients with primary parotid carcinoma treated with surgery and radiotherapy at the Daniel den Hoed Cancer Center. Int J Radiat Oncol Biol Phys. 2012;84:189–95. 15. Chen AM, Garcia J, Lee NY, Bucci MK, Eisele DW. Patterns of nodal relapse after surgery and postoperative radiation therapy for carcinomas of the major and minor salivary glands: what is the role of elective neck irradiation? Int J Radiat Oncol Biol Phys. 2007;67:988–94. 16. Rodriguez-Cuevas S, Labastida S, Baena L, Gallegos F. Risk of nodal metastases from malignant salivary gland tumors related to tumor size and grade of malignancy. Eur Arch Otorhinolaryngol. 1995;252:139–42.

Treatment of the neck in carcinoma of the parotid gland.

To review our experience in the treatment of the neck in patients with carcinoma of the parotid gland...
411KB Sizes 3 Downloads 5 Views