ORIGINAL ARTICLE
The Prognostic Significance of Facial Nerve Involvement in Carcinomas of the Parotid Gland Breanne E. Terakedis, MD,* Jason P. Hunt, MD,w Luke O. Buchmann, MD,w Vilija N. Avizonis, MD,z Christopher J. Anker, MD,y and Ying J. Hitchcock, MD*
Importance and Background: Facial nerve (FN) palsy and perineural invasion (PNI) are adverse features in carcinomas of the parotid gland. FN sacrifice at the time of surgery is associated with significant morbidity. The role of adjuvant radiotherapy in patients with high-risk features, including FN involvement, remains unclear. Objective: Analyze the disease-free survival (DFS) and overall survival (OS) and the impact of tumor characteristics, including FN involvement, for patients treated with surgical resection for carcinoma of the parotid gland. Design: This is a retrospective chart review. Setting: University of Utah and Intermountain Healthcare, Utah. Participants: A total of 129 patients who were treated with primary surgery for nonmetastatic primary malignancies of the parotid gland from 1988 to 2006. Interventions: Parotidectomy with or without adjuvant therapy. Main Outcome(s) and Measures: Kaplan-Meier analysis was used to obtain 5-year estimates of DFS and OS. Recurrence risk factors, particularly the impact of FN involvement, were analyzed. Results: Five-year DFS and OS rates were 79% and 78%, respectively. Thirty-two (28%) patients developed recurrent disease. Disease recurrence occurred in 64% of patients with both FN palsy and PNI, in 43% with FN palsy without PNI, in 27% with only PNI, and in 16% without either feature. Conclusions and Relevance: FN involvement, particularly FN palsy, is a predictor of increased risk of recurrence and death. Radiotherapy cannot substitute for FN sacrifice in high-risk patients. Key Words: parotid, carcinoma, malignancy, perineural invasion, facial nerve
(Am J Clin Oncol 2014;00:000–000)
T
umors of the salivary glands are relatively rare with an annual incidence of approximately 1 to 2 per 100,000 people.1 The incidence of salivary gland neoplasms is rising,2,3 and approximately 80% of cases involve the parotid gland. From the Departments of *Radiation Oncology; wOtolaryngology-Head and Neck Surgery, University of Utah Huntsman Cancer Hospital, Salt Lake City; zDepartment of Radiation Oncology, Intermountain Medical Center, Murray, UT; and yDepartment of Radiation Oncology, University of Vermont, Burlington, VT. Presented at the Multidisciplinary Head and Neck Cancer Symposium, January 2012, Phoenix, AZ. The authors declare no conflicts of interest. Reprints: Ying J. Hitchcock, MD, Department of Radiation Oncology, University of Utah Huntsman Cancer Hospital, 1950 Circle of Hope, Salt Lake City, UT 84112-5560. E-mail: [email protected]. Copyright r 2014 by Lippincott Williams & Wilkins ISSN: 0277-3732/14/000-000 DOI: 10.1097/COC.0000000000000157
American Journal of Clinical Oncology
Of these parotid gland tumors, only 20% to 30% are malignant. Prognostic factors for parotid gland carcinomas include tumor stage, histology, grading, facial nerve (FN) involvement, extraparotid tumor extension, and cervical lymph node involvement.2 Involvement of the FN by carcinomas of the parotid gland has several implications. Both clinical evidence of FN involvement4–6 and pathologic proof of perineural invasion (PNI)7,8 have been shown to be negative prognostic factors. Surgeons are challenged to preserve or reconstruct the FN while performing an effective oncologic surgery. These features of FN involvement are considered by multidisciplinary teams when deciding whether postoperative radiotherapy (RT) should be recommended. Elucidation of the risk factors of parotid carcinoma, as well as the role of RT are derived primarily from small single institution studies.9 Oftentimes, these reviews include all salivary gland locations, not just carcinomas of the parotid gland. In addition, complete information on clinical and pathologic FN involvement is often unavailable,10 and numerous series include patients treated with RT alone in addition to patients treated with surgery followed by RT.5,8,11 We performed this retrospective review of patients treated for parotid carcinoma with the goal of adding to the growing body of literature for this relatively rare disease. Herein we analyze the disease-free survival (DFS) and overall survival (OS) rates and the impact of tumor characteristics, including FN involvement, for patients treated with surgical resection with or without adjuvant therapy for carcinoma of the parotid gland.
MATERIALS AND METHODS Between 1988 and 2006, 129 consecutive patients with previously untreated, nonmetastatic primary parotid gland malignancies underwent therapeutic surgery at the University of Utah and at Intermountain Healthcare, Salt Lake City, Utah. This included patients with distant metastatic disease at the time of diagnosis, as well as those with parotid lymphoma. Surgery was the primary treatment modality for all patients included in this study. After Institutional Review Board approval, charts were retrospectively reviewed for data collection, and the following variables were recorded: patient age, sex, tumor histology, grade (low, intermediate, high), T stage (T1 to T4), N stage (N0 or N +), absence or presence (including partial or complete palsy) of FN palsy, surgical procedure (total vs. superficial parotidectomy with or without lymph node dissection), and pathologic details including lymph node status, presence or absence of PNI, and margin status. The presence or absence of FN palsy was determined by reviewing the clinical history and physical examination portions of the medical records. Use of adjuvant therapy including chemotherapy and/or RT was documented. The time and location of first recurrence, length of follow-up, and date of death were recorded.
Volume 00, Number 00, ’’ 2014
www.amjclinicaloncology.com |
1
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Terakedis et al
Surgery All patients in this series underwent surgery as the primary treatment modality. Parotid gland surgery was either a superficial or total parotidectomy. Extent of surgery was based on the size and location of the primary tumor, leaving extent of surgery to the discretion of the operating surgeon. The FN was spared when there was no evidence of gross nerve involvement. Obvious clinical invasion of the nerve resulted in resection of the FN to obtain negative margin at the gross tumor site and control the proximal and distal nerve margin with frozen pathologic section. Instances where there was not clear clinical evidence of nerve invasion, the operating surgeon would attempt to preserve the FN. Data regarding extent of surgery (total vs. superficial parotidectomy), preoperative FN function, PNI, and FN sacrifice were recorded for the purposes of this study.
Adjuvant Treatment and Follow-up RT or chemoRT was delivered at the discretion of the treating surgeon, radiation oncologist, and medical oncologist. Adjuvant RT alone was the most frequently utilized postoperative modality for patients with high-risk features, such as T3/T4 disease, extracapsular extension, intraoperative tumor spillage, PNI, close or microscopic positive margins, and intermediate-grade or high-grade tumor. Adjuvant radiation dose is standard 60 to 66 Gy at 1.8 to 2.0 Gy per fraction based on pathologic features. Timing and frequency of follow-up examinations were every 2 to 3 months for the first 2 years, 4 to 6 months for years 3 to 5, then annually. The first posttreatment imaging was performed approximately 3 months from the end of adjuvant RT or 2 to 3 months after surgery alone. Subsequent imaging study was at the discretion of the treating physicians.
Statistical Analysis Univariable Cox proportional hazard regression was performed to assess predictive factors for disease recurrence and survival. All variables considered significant in the univariate analysis were entered into a multivariate analysis. Hazard ratios with 95% confidence intervals were recorded. Statistical significance was defined as P < 0.05. Kaplan-Meier curves and estimates of DFS and OS were generated. All statistical analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC).
RESULTS Median follow-up was 72 months (range, 2.5 to 241 mo), and median age at diagnosis was 53 years (range, 9 to 94 y). The majority of patients had T1/T2 tumors (66%) and nodenegative disease (81%) (Table 1). Mucoepidermoid carcinoma was the most predominant histology (36%) followed by acinic cell carcinoma (25%). The number of patients undergoing superficial (n = 61) versus total parotidectomy (n = 68) were similar. Forty (35%) patients underwent a neck dissection. Mucoepidermoid carcinoma was the most likely histology to involve regional lymph nodes, accounting for 14 of the 28 cases of lymph node involvement at the time of diagnosis.
Adjuvant Therapy Of the 129 patients in this analysis, 86 (67%) received adjuvant RT. Median radiation dose was 60 Gy (range, 50 to 70.2 cGy). RT was utilized in the majority of patients (Table 1). Patients who underwent total parotidectomy and patients with microscopic positive margins or unknown margins were more likely to receive adjuvant RT than patients
2 | www.amjclinicaloncology.com
Volume 00, Number 00, ’’ 2014
TABLE 1. Total for all Patients Followed by Subgroup Analysis
n (%) All Adjuvant Patients Radiotherapy (n = 129) (n = 86, 67%) Sex Male Female T stage T1/T2 T3/T4 Unknown Grade High Intermediate Low Unknown Histology Mucoepidermoid Acinic cell Adenoid cystic Adenocarcinoma Other (salivary duct, carcinoma, ex pleomorphic adenoma, adenosquamous) Surgery Total parotidectomy Superficial parotidectomy Node positive Margin status Negative Microscopic positive Unknown Facial nerve palsy Perineural invasion Facial nerve sacrifice
No Adjuvant Radiotherapy (n = 43, 33%)
71 58
51 (72) 35 (60)
20 (28) 23 (40)
85 40 4
49 (58) 34 (85) 3 (75)
36 (42) 6 (15) 1 (25)
49 25 37 18
44 16 12 14
(90) (64) (32) (78)
5 9 25 4
(10) (36) (68) (22)
47 32 19 19 12
33 12 17 16 8
(70) (38) (89) (84) (67)
14 20 2 3 4
(30) (64) (11) (16) (33)
68
55 (81)
13 (19)
61
31 (51)
30 (49)
28
26 (93)
2 (7)
78 34
46 (59) 25 (74)
32 (41) 9 (26)
17 22 36 28
15 18 32 25
(88) (82) (89) (89)
2 4 4 3
(12) (18) (11) (11)
Percentage given is relative to the radiation grouping.
undergoing a superficial parotidectomy and those with negative surgical margins. Adjuvant chemotherapy use was documented in only 6 cases.
DFS and OS Five-year DFS was 79% for all patients. Disease recurrence occurred in 32 (25%) patients, including 28 in the adjuvant RT group and 4 patients in the group not receiving RT (Table 2). Distant recurrences were more common than locoregional recurrences with lung followed by bone being the most predominant sites of distant failure. The majority of distant recurrences were in patients with adenoid cystic (n = 7) and adenocarcinoma (n = 5) histology followed by mucoepidermoid (n = 3) and salivary duct carcinoma (n = 3). Univariate analysis revealed that tumor histology, advanced T stage, lymph node involvement, FN palsy, PNI, FN sacrifice, involved margins, and use of RT were predictors of disease recurrence (Table 3). On multivariate r
2014 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Volume 00, Number 00, ’’ 2014
TABLE 2. Site of First Recurrence (32 Recurrences)
Locoregional Distant Distant and locoregional Unknown site Total (n [%])
Adjuvant RT (n = 86 Patients)
No RT (n = 43 Patients)
9 16 2 1 28 (32.5)
4 0 0 0 4 (9.3)
RT indicates radiotherapy.
analysis, age, histology, and FN sacrifice were significant predictors of poor OS. Five-year and 10-year OS rates for the entire group were 78% and 67%, respectively. Male sex, intermediate-grade and high-grade tumors, histology other than acinic cell carcinoma and mucoepidermoid carcinoma, clinical FN dysfunction, PNI, FN sacrifice due to gross FN invasion, and use of RT were predictors of poor OS (Table 3). Histology, age 60 years and above, and FN sacrifice remained statistically significant predictors of worse OS.
FN Involvement Clinical FN dysfunction, PNI, and FN sacrifice for FN involvement were present in 22, 36, and 28 patients, respectively, and the majority of patients with any one of these features received adjuvant RT (Table 1). Of patients with PNI (n = 36), the most common histologies were mucoepidermoid (n = 13) and adenoid cystic (n = 8) carcinoma. Fourteen patients were found to have both PNI and FN palsy (Table 4), and of these, 11 received adjuvant RT and 9 (64%) developed recurrent disease. Eight patients had clinical FN palsy without pathologic-reported PNI, and 3 patients developed recurrent disease (43%). Twenty-two patients had pathologic PNI without clinical FN palsy, and the absolute
Carcinoma of the Parotid Gland
recurrence rate was 27%. In patients without FN palsy or PNI (n = 85), 16% developed recurrent disease. Of the 101 patients treated with FN preservation, 61 received adjuvant RT (Table 5). Of these, 13 (21.3%) developed recurrent or metastatic disease. Five patients recurred locally, 1 patient recurred regionally, and 7 patients experienced a distant-only recurrence. Among the 40 patients treated with FN preservation without RT, there were 3 (7.5%) recurrences including 1 local-only recurrence and 2 locoregional recurrences. Of the 25 patients treated with FN sacrifice and adjuvant RT, 15 (60%) developed recurrent disease, the majority of which were distant-only recurrences. There was 1 local recurrence in 3 patients treated with FN sacrifice without RT.
DISCUSSION Carcinoma of the parotid gland is a rare malignancy, and this single institution review aims to more clearly define the role of adjuvant therapy as well as the association between clinical FN involvement and PNI and how these factors relate to disease progression and survival. In our series, the site of first recurrence was distant in 50% of patients, and the most common site of failure was the lung. This pattern of recurrence is typical for patients who have undergone surgery with or without RT for localized parotid gland carcinoma. A review of 127 patients with parotid gland carcinoma treated at the Instituo Nacional de Cancerologia revealed that 41.7% of parotid cancer patients experienced disease recurrence, and 49% of these were distant metastases.4 The most common site of distant failure was lung (76.9%) followed by bone (19.23%) and liver (7.7%). Feinstein et al10 report that among 33 cases of recurrent disease in 74 patients treated with surgery and RT for salivary gland tumors, 19 (58%) experienced a distant recurrence without locoregional failure. Similar to other cancers of various disease sites, as the efficacy of locoregional tumor control improves, systemic therapies to prevent and treat
TABLE 3. Univariate Analysis for Risk of Recurrence and Death Male vs. female Age Z60 y Grade High Intermediate Histology Acinic cell Adenocarcinoma Adenoid cystic Mucoepidermoid Other T stage T1 T2 T3 T4 LN involvement Facial nerve palsy Perineural invasion Facial nerve sacrifice Involved margin Radiation use
Recurrence (HR [95% CI])
P
Death (HR [95% CI])
P
1.88 (0.89, 3.98) 0.8 (0.37, 1.74)
0.10 0.58 0.06
1.86 (1.01, 3.42) 2.79 (1.58, 4.94)
0.05 < 0.001 < 0.001
3.33 (1.24, 8.98) 2.24 (0.68, 7.33)
3.85 (1.68, 8.83) 2.07 (0.75, 5.71)
Reference 2.43 (0.54, 10.88) 6.84 (1.88, 24.89) 2.05 (0.55, 7.56) 9.67 (2.39, 39.18)
< 0.001
Reference 3.33 (1.09, 10.19) 4.66 (1.64, 13.25) 2.04 (0.73, 5.67) 13.37 (4.48, 39.92)
< 0.001
Reference 3.79 (1.14, 12.58) 7.82 (2.28, 26.84) 10.98 (3.52, 34.28) 2.85 (1.39, 5.84) 4.02 (1.96, 8.25) 3.0 (1.48, 6.08) 51.4 (2.55, 10.37) 2.74 (1.26, 5.94) 3.58 (1.25, 10.21)
< 0.001
Reference 3.4 (1.38, 8.35) 5.98 (2.16, 16.56) 9.27 (3.81, 22.52) 2.87 (1.59, 5.18) 3.06 (1.66, 5.63) 2.89 (1.62, 5.16) 4.92 (2.75, 8.79) 1.99 (1.07, 3.7) 2.05 (0.99, 4.23)
< 0.001
< 0.001 < 0.001 < 0.001 < 0.001 0.01 0.02
< 0.001 < 0.001 < 0.001 < 0.001 0.03 0.05
CI indicates confidence interval; HR, hazard ratio; LN, lymph node.
r
2014 Lippincott Williams & Wilkins
www.amjclinicaloncology.com |
3
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Terakedis et al
Volume 00, Number 00, ’’ 2014
TABLE 4. Relationship Between Clinical and Pathologic FN Involvement and Disease Recurrence
FN Palsy/PNI
n
No. Patients Receiving RT
+/+ + / / + /
14 8 22 85
11 7 21 47
First Recurrence Site
No. With Recurrent Disease (n [%]) 9 3 6 14
Local
Regional
LR
Distant
LR and Distant
Unknown
0 1 2 5
1 1 1 0
0 0 0 2
7 1 1 7
0 0 2 0
1 0 0 0
(64) (43) (27) (16)
FN indicates facial nerve; LR, locoregional; PNI, perineural invasion; RT, radiotherapy.
metastatic disease become increasingly important. Although use of systemic therapy has been shown to improve outcomes in patients with squamous cell carcinoma of the head and neck,12,13 use of systemic therapy for salivary gland malignancies has primarily been met with low to moderate response rates.14 In an effort to further define the role of chemotherapy, the Radiation Therapy Oncology Group is conducting a randomized, phase II study investigating the use of concurrent cisplatin with adjuvant RT for patients with resected salivary gland tumors. Eligible patients include those with intermediate-grade or high-grade tumors with a risk factor for recurrence, including T3 to T4 or N1 to N3 disease or T1 to 2N0 disease with a positive or close resection margin. Ideally this study will shed light on the role of systemic therapy in patients at high risk for distant failure.
Clinical Risk Factors Numerous retrospective, single institution reviews have identified poor prognostic factors in carcinomas of the parotid gland. Age, sex, histology, T stage, and LN involvement all influence survival in this patient population.4–6,9–11,15 The individual impact of FN palsy and PNI are less well documented. In our series, multivariate analysis revealed that age and histology, as well as FN sacrifice were significant predictors of poor OS. Patients undergoing FN sacrifice are likely those with a multitude of risk factors including advanced T stage, clinical FN palsy, and/or intraoperative gross PNI. Although FN palsy and PNI were not significant on multivariate analysis, patients with either feature had higher absolute recurrence rates than patients with neither feature, and patients with both features had the highest absolute recurrence rate.
Clinical FN Palsy In our study, clinical FN palsy was seen in 17% of patients which is consistent with published literature in which FN paralysis is documented in 14% to 30% of cases.4–6,8,11,13 Incomplete and complete FN paralysis have been identified as risk factors for recurrence and survival.5,6,11 Koul et al6
evaluated 184 patients with parotid gland malignancies and found that 26 (15%) patients presented with FN palsy. On univariate analysis, presence of FN palsy was associated with a lower 5-year disease-specific survival rate (67% vs. 32%, P = 0.0001). Godballe et al5 reported on 75 patients with carcinomas of the parotid gland treated with surgery and/or RT. FN dysfunction was present in 12 patients and was associated with lower rates of 5-year recurrence-free and disease-specific survival. A review of 324 patients with parotid carcinoma in the Dutch Head and Neck Oncology Cooperative Group revealed that FN function was partially and completely impaired in 14% and 7% of patients, respectively.9 The FN was sacrificed in 77% of patients, and postoperative RT was delivered in 77% of cases. On multivariate analysis, FN dysfunction was an independent predictor of DFS, and DFS rates correlated with the degree of FN impairment. In our analysis, clinical FN palsy was associated with a detriment in DFS and OS. Disease recurrence, the majority of which were distant recurrences, occurred in 9/14 patients (64%) patients with both FN palsy and PNI. FN palsy is a poor prognostic feature, and presence of FN dysfunction should be considered when discussing treatment options and prognosis with both patients and multidisciplinary teams. The propensity for disease failure, despite aggressive local therapies, again highlights the need for effective systemic therapy in this patient population.
PNI The prognostic significance of PNI is unclear as some series report worse outcomes in patients with PNI, whereas other investigators have shown that PNI does not impact disease recurrence or survival.7–9 Pohar and colleagues reviewed 163 cases of parotid carcinoma from 2 institutions and found that on univariate analysis, PNI was associated with a decrease in cause-specific survival (HR = 1.79, P = 0.02) and OS (HR = 1.69, P = 0.02) and a trend toward a decrease in local failure-free survival (HR = 1.55, P = 0.06).8 However, on multivariate analysis, PNI and facial weakness were no longer
TABLE 5. Site of First Recurrence for Patients With and Without FN Preservation and Adjuvant RT
FN FN FN FN
preservation, no RT preservation, with RT sacrifice, no RT sacrifice, with RT
First Recurrence Site
n
No. With Recurrent Disease
Local
Regional
LR
Distant
LR and Distant
Unknown
40 61 3 25
3 13 1 15
1 5 1 2
0 1 0 0
2 0 0 1
0 7 0 9
0 0 0 2
0 0 0 1
FN indicates facial nerve; LR, locoregional; RT, radiotherapy.
4 | www.amjclinicaloncology.com
r
2014 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Volume 00, Number 00, ’’ 2014
associated with these outcomes. In 85 cases of salivary gland tumors, 42 of which were carcinomas of the parotid gland, Bell et al9 found that PNI did not impact OS. Gomez et al7 reviewed cases of acinic cell carcinoma and found that, on univariate analysis, PNI was associated with a decrease in DFS (P = 0.005) but not OS, but that FN sacrifice was associated with a detriment in DFS (P = 0.046) and OS (P = 0.02). In our analysis of all parotid malignancies, including acinic cell carcinoma, PNI was associated with an increased risk of recurrence and death on univariate analysis but not a significant predictor of OS on multivariate analysis. Lack of significance on multivariate analysis could be attributable to a lack of power due to a small sample size. When comparing absolute risks of recurrence, there was a 27% recurrence rate in patients with PNI without FN palsy, of whom 95% received adjuvant RT. This is in contrast to a 16% recurrence rate in patients without FN palsy or PNI, of whom only 55% received adjuvant RT.
FN Sacrifice In our series, non-nerve sparing surgeries were associated with an increased risk of recurrence and death. It is unlikely that nerve sacrifice leads to increased mortality, but rather the patients in this treatment group have high-risk features which necessitate more aggressive surgery. FN sacrifice occurred in 17/71 (21.7%) patients, and this rate is consistent with the published literature.16 FN sacrifice is associated with significant morbidity and quality of life detriment. In treating patients with clinical FN involvement, we question whether postoperative RT can substitute for FN sacrifice in patients without FN palsy. In our series, 101 patients underwent nerve preservation surgery, and 16 patients developed recurrent disease. There were 7 (43.8%) cases of distant-only recurrence and 9 (56.2%) cases of either a local or regional recurrence. Six of these 9 locoregional recurrences occurred in patients who had received adjuvant RT. In patients who underwent FN sacrifice (n = 28), there were 16 cases of recurrent disease of which only 4 (25%) were local or regional only. Although this sample size precludes definitive conclusions, the data suggest that while RT plays an important role in obtaining local control, RT cannot substitute for FN sacrifice in highrisk patients.
RT In our series, the majority of patients were treated with adjuvant RT, and 28 patients in the adjuvant RT arm and 4 patients in the non-RT group developed recurrent disease. All 4 recurrences in the non-RT group were local, whereas in those patients receiving adjuvant RT, the recurrences were predominantly distant, suggesting that adjuvant RT may have benefited these 4 patients as well. Although RT cannot substitute for an aggressive oncologic surgery, adjuvant RT for parotid carcinoma has been shown to improve local control17–20 and is routinely used in patients with T3/T4 tumors, high-grade disease, and aggressive histologies, and for patients with positive margins postoperatively. Armstrong et al17 performed a matched-pair analysis of patients treated with combined surgery and RT versus patients treated with surgery alone and found that in patients with stage III/IV disease, combined modality treatment improved 5-year local control (51% vs. 17%) and survival (50% vs. 10%). The Dutch Head and Neck Cooperative Oncology Cooperative Group reported that postoperative RT improved 10-year local control when compared with surgery alone for patients with T3 to T4 tumors (84% vs. 81%), in patients with close (95% vs. 55%) and r
2014 Lippincott Williams & Wilkins
Carcinoma of the Parotid Gland
incomplete resection (82 vs. 44%), in bone invasion (86% vs. 54%) and PNI (88 vs. 60%).11 Limitations of this study include those inherent to a retrospective review of patient outcomes. In addition, due to a relatively small number of events, only a limited multivariate analysis was feasible. Despite these drawbacks, our patient series provides a better understanding of the relationship between FN palsy, PNI, and the role of nerve-sacrificing surgery and adjuvant therapy. This patient population is homogenous, only inclusive of patients with primary malignancies of the parotid gland treated with definitive surgical resection, and data regarding clinical FN dysfunction and PNI are available for all patients.
CONCLUSIONS FN involvement is a predictor of increased risk of recurrence and death from parotid cancer. FN palsy seems to be a stronger predictor than PNI; however, PNI without FN palsy is associated with an elevated absolute risk of recurrence despite the use of postoperative RT in most patients. Postoperative RT seems to provide a benefit in patients with high-risk features, and in patients treated with adjuvant RT, distant failure becomes an increasingly important therapeutic issue. Although RT plays an important role in obtaining locoregional control, adjuvant RT does not seem to substitute for FN sacrifice in high-risk patients. Adjuvant chemotherapy has been associated with a mixed response, and further studies are evaluating the benefit of systemic treatment in high-risk patients. REFERENCES 1. Guzzo M, Locati LD, Prott FJ, et al. Major and minor salivary gland tumors. Crit Rev Oncol Hematol. 2010;74:134–148. 2. Carvalho AL, Nichimoto IN, Califano JA, et al. Trends in incidence and prognosis for head and neck cancer in the United States: a site-specific analysis of the SEER database. Int J Cancer. 2005;114:806–816. 3. Davies L, Welch HG. Epidemiology of head and neck cancer in the United States. Otolaryngol Head Neck Surg. 2006;135: 451–457. 4. Carrillo JF, Va´zquez R, Ramı´rez-Ortega MC, et al. Multivariate prediction of the probability of recurrence in patients with carcinoma of the parotid gland. Cancer. 2007;109:2043–2051. 5. Godballe C, Schultz JH, Krogdahl A, et al. Parotid carcinoma: impact of clinical factors on prognosis in histologically revised series. Laryngoscope. 2003;113:1411–1417. 6. Koul R, Dubey A, Butler J, et al. Prognostic factors depicting disease-specific survival in parotid-gland tumors. Int J Radiat Oncol Biol Phys. 2007;68:714–718. 7. Gomez DR, Katabi N, Zhung J, et al. Clinical and pathologic prognostic features in acinic cell carcinoma of the parotid gland. Cancer. 2009;115:2128–2137. 8. Pohar S, Gay H, Rosenbaum P, et al. Malignant parotid tumors: presentation, clinical/pathologic prognostic factors, and treatment outcomes. Int J Radiat Oncol Biol Phys. 2005;61: 112–118. 9. Bell RB, Dierks EJ, Homer L, et al. Management and outcome of patients with malignant salivary gland tumors. J Oral Maxillofac Surg. 2005;63:917–928. 10. Feinstein TM, Lai S, Lenzner D, et al. Prognostic factors in patients with high-risk locally advanced salivary gland cancers treated with surgery and postoperative radiotherapy. Head Neck. 2011;33:318–323. 11. Terhaard C, Lubsen H, Tan B, et al. Facial nerve function in carcinoma of the parotid gland. Eur J Cancer. 2006;42: 2744–2750. 12. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med. 2004;350:1945–1952.
www.amjclinicaloncology.com |
5
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Terakedis et al
American Journal of Clinical Oncology
13. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med. 2004;350:1937–1944. 14. Lagha A, Chraiet N, Ayadi M, et al. Systemic therapy in the management of metastatic or advanced salivary gland cancers. Oral Oncol. 2012;48:948–957. 15. Mercante G, Marchese C, Giannarelli D, et al. Oncological outcome and prognostic factors in malignant parotid tumours. J Craniomaxillofac Surg. 2014;42:59–65. 16. Zba¨ren P, Schu¨pbach J, Nuyens M, et al. Carcinoma of the parotid gland. Am J Surg. 2003;186:57–62.
6 | www.amjclinicaloncology.com
Volume 00, Number 00, ’’ 2014
17. Armstrong JG, Harrison LB, Spiro RH, et al. Malignant tumors of major salivary gland origin. Arch Otolaryngol Head Neck Surg. 1990;116:290–293. 18. Garden AS, El-Naggar AK, Morrison WH, et al. Postoperative radiotherapy for malignant tumors of the parotid gland. Int J Radiat Oncol Biol Phys. 1997;37:79–85. 19. Mendenhall WM, Morris CG, Amdur RJ, et al. Radiotherapy alone or combined with surgery for salivary gland carcinoma. Cancer. 2005;103:2544–2550. 20. Terhaard CHJ, Lubsen H, Rasch CRN, et al. The role of radiotherapy in the treatment of malignant salivary gland tumors. Int J Radiat Oncol Biol Phys. 2005;61:103–111.
r
2014 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Prognostic Significance of Facial Nerve Involvement in Carcinomas of the Parotid Gland Breanne E. Terakedis, MD,* Jason P. Hunt, MD,w Luke O. Buchmann, MD,w Vilija N. Avizonis, MD,z Christopher J. Anker, MD,y and Ying J. Hitchcock, MD*
Importance and Background: Facial nerve (FN) palsy and perineural invasion (PNI) are adverse features in carcinomas of the parotid gland. FN sacrifice at the time of surgery is associated with significant morbidity. The role of adjuvant radiotherapy in patients with high-risk features, including FN involvement, remains unclear. Objective: Analyze the disease-free survival (DFS) and overall survival (OS) and the impact of tumor characteristics, including FN involvement, for patients treated with surgical resection for carcinoma of the parotid gland. Design: This is a retrospective chart review. Setting: University of Utah and Intermountain Healthcare, Utah. Participants: A total of 129 patients who were treated with primary surgery for nonmetastatic primary malignancies of the parotid gland from 1988 to 2006. Interventions: Parotidectomy with or without adjuvant therapy. Main Outcome(s) and Measures: Kaplan-Meier analysis was used to obtain 5-year estimates of DFS and OS. Recurrence risk factors, particularly the impact of FN involvement, were analyzed. Results: Five-year DFS and OS rates were 79% and 78%, respectively. Thirty-two (28%) patients developed recurrent disease. Disease recurrence occurred in 64% of patients with both FN palsy and PNI, in 43% with FN palsy without PNI, in 27% with only PNI, and in 16% without either feature. Conclusions and Relevance: FN involvement, particularly FN palsy, is a predictor of increased risk of recurrence and death. Radiotherapy cannot substitute for FN sacrifice in high-risk patients. Key Words: parotid, carcinoma, malignancy, perineural invasion, facial nerve
(Am J Clin Oncol 2014;00:000–000)
T
umors of the salivary glands are relatively rare with an annual incidence of approximately 1 to 2 per 100,000 people.1 The incidence of salivary gland neoplasms is rising,2,3 and approximately 80% of cases involve the parotid gland. From the Departments of *Radiation Oncology; wOtolaryngology-Head and Neck Surgery, University of Utah Huntsman Cancer Hospital, Salt Lake City; zDepartment of Radiation Oncology, Intermountain Medical Center, Murray, UT; and yDepartment of Radiation Oncology, University of Vermont, Burlington, VT. Presented at the Multidisciplinary Head and Neck Cancer Symposium, January 2012, Phoenix, AZ. The authors declare no conflicts of interest. Reprints: Ying J. Hitchcock, MD, Department of Radiation Oncology, University of Utah Huntsman Cancer Hospital, 1950 Circle of Hope, Salt Lake City, UT 84112-5560. E-mail: [email protected]. Copyright r 2014 by Lippincott Williams & Wilkins ISSN: 0277-3732/14/000-000 DOI: 10.1097/COC.0000000000000157
American Journal of Clinical Oncology
Of these parotid gland tumors, only 20% to 30% are malignant. Prognostic factors for parotid gland carcinomas include tumor stage, histology, grading, facial nerve (FN) involvement, extraparotid tumor extension, and cervical lymph node involvement.2 Involvement of the FN by carcinomas of the parotid gland has several implications. Both clinical evidence of FN involvement4–6 and pathologic proof of perineural invasion (PNI)7,8 have been shown to be negative prognostic factors. Surgeons are challenged to preserve or reconstruct the FN while performing an effective oncologic surgery. These features of FN involvement are considered by multidisciplinary teams when deciding whether postoperative radiotherapy (RT) should be recommended. Elucidation of the risk factors of parotid carcinoma, as well as the role of RT are derived primarily from small single institution studies.9 Oftentimes, these reviews include all salivary gland locations, not just carcinomas of the parotid gland. In addition, complete information on clinical and pathologic FN involvement is often unavailable,10 and numerous series include patients treated with RT alone in addition to patients treated with surgery followed by RT.5,8,11 We performed this retrospective review of patients treated for parotid carcinoma with the goal of adding to the growing body of literature for this relatively rare disease. Herein we analyze the disease-free survival (DFS) and overall survival (OS) rates and the impact of tumor characteristics, including FN involvement, for patients treated with surgical resection with or without adjuvant therapy for carcinoma of the parotid gland.
MATERIALS AND METHODS Between 1988 and 2006, 129 consecutive patients with previously untreated, nonmetastatic primary parotid gland malignancies underwent therapeutic surgery at the University of Utah and at Intermountain Healthcare, Salt Lake City, Utah. This included patients with distant metastatic disease at the time of diagnosis, as well as those with parotid lymphoma. Surgery was the primary treatment modality for all patients included in this study. After Institutional Review Board approval, charts were retrospectively reviewed for data collection, and the following variables were recorded: patient age, sex, tumor histology, grade (low, intermediate, high), T stage (T1 to T4), N stage (N0 or N +), absence or presence (including partial or complete palsy) of FN palsy, surgical procedure (total vs. superficial parotidectomy with or without lymph node dissection), and pathologic details including lymph node status, presence or absence of PNI, and margin status. The presence or absence of FN palsy was determined by reviewing the clinical history and physical examination portions of the medical records. Use of adjuvant therapy including chemotherapy and/or RT was documented. The time and location of first recurrence, length of follow-up, and date of death were recorded.
Volume 00, Number 00, ’’ 2014
www.amjclinicaloncology.com |
1
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Terakedis et al
Surgery All patients in this series underwent surgery as the primary treatment modality. Parotid gland surgery was either a superficial or total parotidectomy. Extent of surgery was based on the size and location of the primary tumor, leaving extent of surgery to the discretion of the operating surgeon. The FN was spared when there was no evidence of gross nerve involvement. Obvious clinical invasion of the nerve resulted in resection of the FN to obtain negative margin at the gross tumor site and control the proximal and distal nerve margin with frozen pathologic section. Instances where there was not clear clinical evidence of nerve invasion, the operating surgeon would attempt to preserve the FN. Data regarding extent of surgery (total vs. superficial parotidectomy), preoperative FN function, PNI, and FN sacrifice were recorded for the purposes of this study.
Adjuvant Treatment and Follow-up RT or chemoRT was delivered at the discretion of the treating surgeon, radiation oncologist, and medical oncologist. Adjuvant RT alone was the most frequently utilized postoperative modality for patients with high-risk features, such as T3/T4 disease, extracapsular extension, intraoperative tumor spillage, PNI, close or microscopic positive margins, and intermediate-grade or high-grade tumor. Adjuvant radiation dose is standard 60 to 66 Gy at 1.8 to 2.0 Gy per fraction based on pathologic features. Timing and frequency of follow-up examinations were every 2 to 3 months for the first 2 years, 4 to 6 months for years 3 to 5, then annually. The first posttreatment imaging was performed approximately 3 months from the end of adjuvant RT or 2 to 3 months after surgery alone. Subsequent imaging study was at the discretion of the treating physicians.
Statistical Analysis Univariable Cox proportional hazard regression was performed to assess predictive factors for disease recurrence and survival. All variables considered significant in the univariate analysis were entered into a multivariate analysis. Hazard ratios with 95% confidence intervals were recorded. Statistical significance was defined as P < 0.05. Kaplan-Meier curves and estimates of DFS and OS were generated. All statistical analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC).
RESULTS Median follow-up was 72 months (range, 2.5 to 241 mo), and median age at diagnosis was 53 years (range, 9 to 94 y). The majority of patients had T1/T2 tumors (66%) and nodenegative disease (81%) (Table 1). Mucoepidermoid carcinoma was the most predominant histology (36%) followed by acinic cell carcinoma (25%). The number of patients undergoing superficial (n = 61) versus total parotidectomy (n = 68) were similar. Forty (35%) patients underwent a neck dissection. Mucoepidermoid carcinoma was the most likely histology to involve regional lymph nodes, accounting for 14 of the 28 cases of lymph node involvement at the time of diagnosis.
Adjuvant Therapy Of the 129 patients in this analysis, 86 (67%) received adjuvant RT. Median radiation dose was 60 Gy (range, 50 to 70.2 cGy). RT was utilized in the majority of patients (Table 1). Patients who underwent total parotidectomy and patients with microscopic positive margins or unknown margins were more likely to receive adjuvant RT than patients
2 | www.amjclinicaloncology.com
Volume 00, Number 00, ’’ 2014
TABLE 1. Total for all Patients Followed by Subgroup Analysis
n (%) All Adjuvant Patients Radiotherapy (n = 129) (n = 86, 67%) Sex Male Female T stage T1/T2 T3/T4 Unknown Grade High Intermediate Low Unknown Histology Mucoepidermoid Acinic cell Adenoid cystic Adenocarcinoma Other (salivary duct, carcinoma, ex pleomorphic adenoma, adenosquamous) Surgery Total parotidectomy Superficial parotidectomy Node positive Margin status Negative Microscopic positive Unknown Facial nerve palsy Perineural invasion Facial nerve sacrifice
No Adjuvant Radiotherapy (n = 43, 33%)
71 58
51 (72) 35 (60)
20 (28) 23 (40)
85 40 4
49 (58) 34 (85) 3 (75)
36 (42) 6 (15) 1 (25)
49 25 37 18
44 16 12 14
(90) (64) (32) (78)
5 9 25 4
(10) (36) (68) (22)
47 32 19 19 12
33 12 17 16 8
(70) (38) (89) (84) (67)
14 20 2 3 4
(30) (64) (11) (16) (33)
68
55 (81)
13 (19)
61
31 (51)
30 (49)
28
26 (93)
2 (7)
78 34
46 (59) 25 (74)
32 (41) 9 (26)
17 22 36 28
15 18 32 25
(88) (82) (89) (89)
2 4 4 3
(12) (18) (11) (11)
Percentage given is relative to the radiation grouping.
undergoing a superficial parotidectomy and those with negative surgical margins. Adjuvant chemotherapy use was documented in only 6 cases.
DFS and OS Five-year DFS was 79% for all patients. Disease recurrence occurred in 32 (25%) patients, including 28 in the adjuvant RT group and 4 patients in the group not receiving RT (Table 2). Distant recurrences were more common than locoregional recurrences with lung followed by bone being the most predominant sites of distant failure. The majority of distant recurrences were in patients with adenoid cystic (n = 7) and adenocarcinoma (n = 5) histology followed by mucoepidermoid (n = 3) and salivary duct carcinoma (n = 3). Univariate analysis revealed that tumor histology, advanced T stage, lymph node involvement, FN palsy, PNI, FN sacrifice, involved margins, and use of RT were predictors of disease recurrence (Table 3). On multivariate r
2014 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Volume 00, Number 00, ’’ 2014
TABLE 2. Site of First Recurrence (32 Recurrences)
Locoregional Distant Distant and locoregional Unknown site Total (n [%])
Adjuvant RT (n = 86 Patients)
No RT (n = 43 Patients)
9 16 2 1 28 (32.5)
4 0 0 0 4 (9.3)
RT indicates radiotherapy.
analysis, age, histology, and FN sacrifice were significant predictors of poor OS. Five-year and 10-year OS rates for the entire group were 78% and 67%, respectively. Male sex, intermediate-grade and high-grade tumors, histology other than acinic cell carcinoma and mucoepidermoid carcinoma, clinical FN dysfunction, PNI, FN sacrifice due to gross FN invasion, and use of RT were predictors of poor OS (Table 3). Histology, age 60 years and above, and FN sacrifice remained statistically significant predictors of worse OS.
FN Involvement Clinical FN dysfunction, PNI, and FN sacrifice for FN involvement were present in 22, 36, and 28 patients, respectively, and the majority of patients with any one of these features received adjuvant RT (Table 1). Of patients with PNI (n = 36), the most common histologies were mucoepidermoid (n = 13) and adenoid cystic (n = 8) carcinoma. Fourteen patients were found to have both PNI and FN palsy (Table 4), and of these, 11 received adjuvant RT and 9 (64%) developed recurrent disease. Eight patients had clinical FN palsy without pathologic-reported PNI, and 3 patients developed recurrent disease (43%). Twenty-two patients had pathologic PNI without clinical FN palsy, and the absolute
Carcinoma of the Parotid Gland
recurrence rate was 27%. In patients without FN palsy or PNI (n = 85), 16% developed recurrent disease. Of the 101 patients treated with FN preservation, 61 received adjuvant RT (Table 5). Of these, 13 (21.3%) developed recurrent or metastatic disease. Five patients recurred locally, 1 patient recurred regionally, and 7 patients experienced a distant-only recurrence. Among the 40 patients treated with FN preservation without RT, there were 3 (7.5%) recurrences including 1 local-only recurrence and 2 locoregional recurrences. Of the 25 patients treated with FN sacrifice and adjuvant RT, 15 (60%) developed recurrent disease, the majority of which were distant-only recurrences. There was 1 local recurrence in 3 patients treated with FN sacrifice without RT.
DISCUSSION Carcinoma of the parotid gland is a rare malignancy, and this single institution review aims to more clearly define the role of adjuvant therapy as well as the association between clinical FN involvement and PNI and how these factors relate to disease progression and survival. In our series, the site of first recurrence was distant in 50% of patients, and the most common site of failure was the lung. This pattern of recurrence is typical for patients who have undergone surgery with or without RT for localized parotid gland carcinoma. A review of 127 patients with parotid gland carcinoma treated at the Instituo Nacional de Cancerologia revealed that 41.7% of parotid cancer patients experienced disease recurrence, and 49% of these were distant metastases.4 The most common site of distant failure was lung (76.9%) followed by bone (19.23%) and liver (7.7%). Feinstein et al10 report that among 33 cases of recurrent disease in 74 patients treated with surgery and RT for salivary gland tumors, 19 (58%) experienced a distant recurrence without locoregional failure. Similar to other cancers of various disease sites, as the efficacy of locoregional tumor control improves, systemic therapies to prevent and treat
TABLE 3. Univariate Analysis for Risk of Recurrence and Death Male vs. female Age Z60 y Grade High Intermediate Histology Acinic cell Adenocarcinoma Adenoid cystic Mucoepidermoid Other T stage T1 T2 T3 T4 LN involvement Facial nerve palsy Perineural invasion Facial nerve sacrifice Involved margin Radiation use
Recurrence (HR [95% CI])
P
Death (HR [95% CI])
P
1.88 (0.89, 3.98) 0.8 (0.37, 1.74)
0.10 0.58 0.06
1.86 (1.01, 3.42) 2.79 (1.58, 4.94)
0.05 < 0.001 < 0.001
3.33 (1.24, 8.98) 2.24 (0.68, 7.33)
3.85 (1.68, 8.83) 2.07 (0.75, 5.71)
Reference 2.43 (0.54, 10.88) 6.84 (1.88, 24.89) 2.05 (0.55, 7.56) 9.67 (2.39, 39.18)
< 0.001
Reference 3.33 (1.09, 10.19) 4.66 (1.64, 13.25) 2.04 (0.73, 5.67) 13.37 (4.48, 39.92)
< 0.001
Reference 3.79 (1.14, 12.58) 7.82 (2.28, 26.84) 10.98 (3.52, 34.28) 2.85 (1.39, 5.84) 4.02 (1.96, 8.25) 3.0 (1.48, 6.08) 51.4 (2.55, 10.37) 2.74 (1.26, 5.94) 3.58 (1.25, 10.21)
< 0.001
Reference 3.4 (1.38, 8.35) 5.98 (2.16, 16.56) 9.27 (3.81, 22.52) 2.87 (1.59, 5.18) 3.06 (1.66, 5.63) 2.89 (1.62, 5.16) 4.92 (2.75, 8.79) 1.99 (1.07, 3.7) 2.05 (0.99, 4.23)
< 0.001
< 0.001 < 0.001 < 0.001 < 0.001 0.01 0.02
< 0.001 < 0.001 < 0.001 < 0.001 0.03 0.05
CI indicates confidence interval; HR, hazard ratio; LN, lymph node.
r
2014 Lippincott Williams & Wilkins
www.amjclinicaloncology.com |
3
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Terakedis et al
Volume 00, Number 00, ’’ 2014
TABLE 4. Relationship Between Clinical and Pathologic FN Involvement and Disease Recurrence
FN Palsy/PNI
n
No. Patients Receiving RT
+/+ + / / + /
14 8 22 85
11 7 21 47
First Recurrence Site
No. With Recurrent Disease (n [%]) 9 3 6 14
Local
Regional
LR
Distant
LR and Distant
Unknown
0 1 2 5
1 1 1 0
0 0 0 2
7 1 1 7
0 0 2 0
1 0 0 0
(64) (43) (27) (16)
FN indicates facial nerve; LR, locoregional; PNI, perineural invasion; RT, radiotherapy.
metastatic disease become increasingly important. Although use of systemic therapy has been shown to improve outcomes in patients with squamous cell carcinoma of the head and neck,12,13 use of systemic therapy for salivary gland malignancies has primarily been met with low to moderate response rates.14 In an effort to further define the role of chemotherapy, the Radiation Therapy Oncology Group is conducting a randomized, phase II study investigating the use of concurrent cisplatin with adjuvant RT for patients with resected salivary gland tumors. Eligible patients include those with intermediate-grade or high-grade tumors with a risk factor for recurrence, including T3 to T4 or N1 to N3 disease or T1 to 2N0 disease with a positive or close resection margin. Ideally this study will shed light on the role of systemic therapy in patients at high risk for distant failure.
Clinical Risk Factors Numerous retrospective, single institution reviews have identified poor prognostic factors in carcinomas of the parotid gland. Age, sex, histology, T stage, and LN involvement all influence survival in this patient population.4–6,9–11,15 The individual impact of FN palsy and PNI are less well documented. In our series, multivariate analysis revealed that age and histology, as well as FN sacrifice were significant predictors of poor OS. Patients undergoing FN sacrifice are likely those with a multitude of risk factors including advanced T stage, clinical FN palsy, and/or intraoperative gross PNI. Although FN palsy and PNI were not significant on multivariate analysis, patients with either feature had higher absolute recurrence rates than patients with neither feature, and patients with both features had the highest absolute recurrence rate.
Clinical FN Palsy In our study, clinical FN palsy was seen in 17% of patients which is consistent with published literature in which FN paralysis is documented in 14% to 30% of cases.4–6,8,11,13 Incomplete and complete FN paralysis have been identified as risk factors for recurrence and survival.5,6,11 Koul et al6
evaluated 184 patients with parotid gland malignancies and found that 26 (15%) patients presented with FN palsy. On univariate analysis, presence of FN palsy was associated with a lower 5-year disease-specific survival rate (67% vs. 32%, P = 0.0001). Godballe et al5 reported on 75 patients with carcinomas of the parotid gland treated with surgery and/or RT. FN dysfunction was present in 12 patients and was associated with lower rates of 5-year recurrence-free and disease-specific survival. A review of 324 patients with parotid carcinoma in the Dutch Head and Neck Oncology Cooperative Group revealed that FN function was partially and completely impaired in 14% and 7% of patients, respectively.9 The FN was sacrificed in 77% of patients, and postoperative RT was delivered in 77% of cases. On multivariate analysis, FN dysfunction was an independent predictor of DFS, and DFS rates correlated with the degree of FN impairment. In our analysis, clinical FN palsy was associated with a detriment in DFS and OS. Disease recurrence, the majority of which were distant recurrences, occurred in 9/14 patients (64%) patients with both FN palsy and PNI. FN palsy is a poor prognostic feature, and presence of FN dysfunction should be considered when discussing treatment options and prognosis with both patients and multidisciplinary teams. The propensity for disease failure, despite aggressive local therapies, again highlights the need for effective systemic therapy in this patient population.
PNI The prognostic significance of PNI is unclear as some series report worse outcomes in patients with PNI, whereas other investigators have shown that PNI does not impact disease recurrence or survival.7–9 Pohar and colleagues reviewed 163 cases of parotid carcinoma from 2 institutions and found that on univariate analysis, PNI was associated with a decrease in cause-specific survival (HR = 1.79, P = 0.02) and OS (HR = 1.69, P = 0.02) and a trend toward a decrease in local failure-free survival (HR = 1.55, P = 0.06).8 However, on multivariate analysis, PNI and facial weakness were no longer
TABLE 5. Site of First Recurrence for Patients With and Without FN Preservation and Adjuvant RT
FN FN FN FN
preservation, no RT preservation, with RT sacrifice, no RT sacrifice, with RT
First Recurrence Site
n
No. With Recurrent Disease
Local
Regional
LR
Distant
LR and Distant
Unknown
40 61 3 25
3 13 1 15
1 5 1 2
0 1 0 0
2 0 0 1
0 7 0 9
0 0 0 2
0 0 0 1
FN indicates facial nerve; LR, locoregional; RT, radiotherapy.
4 | www.amjclinicaloncology.com
r
2014 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
American Journal of Clinical Oncology
Volume 00, Number 00, ’’ 2014
associated with these outcomes. In 85 cases of salivary gland tumors, 42 of which were carcinomas of the parotid gland, Bell et al9 found that PNI did not impact OS. Gomez et al7 reviewed cases of acinic cell carcinoma and found that, on univariate analysis, PNI was associated with a decrease in DFS (P = 0.005) but not OS, but that FN sacrifice was associated with a detriment in DFS (P = 0.046) and OS (P = 0.02). In our analysis of all parotid malignancies, including acinic cell carcinoma, PNI was associated with an increased risk of recurrence and death on univariate analysis but not a significant predictor of OS on multivariate analysis. Lack of significance on multivariate analysis could be attributable to a lack of power due to a small sample size. When comparing absolute risks of recurrence, there was a 27% recurrence rate in patients with PNI without FN palsy, of whom 95% received adjuvant RT. This is in contrast to a 16% recurrence rate in patients without FN palsy or PNI, of whom only 55% received adjuvant RT.
FN Sacrifice In our series, non-nerve sparing surgeries were associated with an increased risk of recurrence and death. It is unlikely that nerve sacrifice leads to increased mortality, but rather the patients in this treatment group have high-risk features which necessitate more aggressive surgery. FN sacrifice occurred in 17/71 (21.7%) patients, and this rate is consistent with the published literature.16 FN sacrifice is associated with significant morbidity and quality of life detriment. In treating patients with clinical FN involvement, we question whether postoperative RT can substitute for FN sacrifice in patients without FN palsy. In our series, 101 patients underwent nerve preservation surgery, and 16 patients developed recurrent disease. There were 7 (43.8%) cases of distant-only recurrence and 9 (56.2%) cases of either a local or regional recurrence. Six of these 9 locoregional recurrences occurred in patients who had received adjuvant RT. In patients who underwent FN sacrifice (n = 28), there were 16 cases of recurrent disease of which only 4 (25%) were local or regional only. Although this sample size precludes definitive conclusions, the data suggest that while RT plays an important role in obtaining local control, RT cannot substitute for FN sacrifice in highrisk patients.
RT In our series, the majority of patients were treated with adjuvant RT, and 28 patients in the adjuvant RT arm and 4 patients in the non-RT group developed recurrent disease. All 4 recurrences in the non-RT group were local, whereas in those patients receiving adjuvant RT, the recurrences were predominantly distant, suggesting that adjuvant RT may have benefited these 4 patients as well. Although RT cannot substitute for an aggressive oncologic surgery, adjuvant RT for parotid carcinoma has been shown to improve local control17–20 and is routinely used in patients with T3/T4 tumors, high-grade disease, and aggressive histologies, and for patients with positive margins postoperatively. Armstrong et al17 performed a matched-pair analysis of patients treated with combined surgery and RT versus patients treated with surgery alone and found that in patients with stage III/IV disease, combined modality treatment improved 5-year local control (51% vs. 17%) and survival (50% vs. 10%). The Dutch Head and Neck Cooperative Oncology Cooperative Group reported that postoperative RT improved 10-year local control when compared with surgery alone for patients with T3 to T4 tumors (84% vs. 81%), in patients with close (95% vs. 55%) and r
2014 Lippincott Williams & Wilkins
Carcinoma of the Parotid Gland
incomplete resection (82 vs. 44%), in bone invasion (86% vs. 54%) and PNI (88 vs. 60%).11 Limitations of this study include those inherent to a retrospective review of patient outcomes. In addition, due to a relatively small number of events, only a limited multivariate analysis was feasible. Despite these drawbacks, our patient series provides a better understanding of the relationship between FN palsy, PNI, and the role of nerve-sacrificing surgery and adjuvant therapy. This patient population is homogenous, only inclusive of patients with primary malignancies of the parotid gland treated with definitive surgical resection, and data regarding clinical FN dysfunction and PNI are available for all patients.
CONCLUSIONS FN involvement is a predictor of increased risk of recurrence and death from parotid cancer. FN palsy seems to be a stronger predictor than PNI; however, PNI without FN palsy is associated with an elevated absolute risk of recurrence despite the use of postoperative RT in most patients. Postoperative RT seems to provide a benefit in patients with high-risk features, and in patients treated with adjuvant RT, distant failure becomes an increasingly important therapeutic issue. Although RT plays an important role in obtaining locoregional control, adjuvant RT does not seem to substitute for FN sacrifice in high-risk patients. Adjuvant chemotherapy has been associated with a mixed response, and further studies are evaluating the benefit of systemic treatment in high-risk patients. REFERENCES 1. Guzzo M, Locati LD, Prott FJ, et al. Major and minor salivary gland tumors. Crit Rev Oncol Hematol. 2010;74:134–148. 2. Carvalho AL, Nichimoto IN, Califano JA, et al. Trends in incidence and prognosis for head and neck cancer in the United States: a site-specific analysis of the SEER database. Int J Cancer. 2005;114:806–816. 3. Davies L, Welch HG. Epidemiology of head and neck cancer in the United States. Otolaryngol Head Neck Surg. 2006;135: 451–457. 4. Carrillo JF, Va´zquez R, Ramı´rez-Ortega MC, et al. Multivariate prediction of the probability of recurrence in patients with carcinoma of the parotid gland. Cancer. 2007;109:2043–2051. 5. Godballe C, Schultz JH, Krogdahl A, et al. Parotid carcinoma: impact of clinical factors on prognosis in histologically revised series. Laryngoscope. 2003;113:1411–1417. 6. Koul R, Dubey A, Butler J, et al. Prognostic factors depicting disease-specific survival in parotid-gland tumors. Int J Radiat Oncol Biol Phys. 2007;68:714–718. 7. Gomez DR, Katabi N, Zhung J, et al. Clinical and pathologic prognostic features in acinic cell carcinoma of the parotid gland. Cancer. 2009;115:2128–2137. 8. Pohar S, Gay H, Rosenbaum P, et al. Malignant parotid tumors: presentation, clinical/pathologic prognostic factors, and treatment outcomes. Int J Radiat Oncol Biol Phys. 2005;61: 112–118. 9. Bell RB, Dierks EJ, Homer L, et al. Management and outcome of patients with malignant salivary gland tumors. J Oral Maxillofac Surg. 2005;63:917–928. 10. Feinstein TM, Lai S, Lenzner D, et al. Prognostic factors in patients with high-risk locally advanced salivary gland cancers treated with surgery and postoperative radiotherapy. Head Neck. 2011;33:318–323. 11. Terhaard C, Lubsen H, Tan B, et al. Facial nerve function in carcinoma of the parotid gland. Eur J Cancer. 2006;42: 2744–2750. 12. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med. 2004;350:1945–1952.
www.amjclinicaloncology.com |
5
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Terakedis et al
American Journal of Clinical Oncology
13. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med. 2004;350:1937–1944. 14. Lagha A, Chraiet N, Ayadi M, et al. Systemic therapy in the management of metastatic or advanced salivary gland cancers. Oral Oncol. 2012;48:948–957. 15. Mercante G, Marchese C, Giannarelli D, et al. Oncological outcome and prognostic factors in malignant parotid tumours. J Craniomaxillofac Surg. 2014;42:59–65. 16. Zba¨ren P, Schu¨pbach J, Nuyens M, et al. Carcinoma of the parotid gland. Am J Surg. 2003;186:57–62.
6 | www.amjclinicaloncology.com
Volume 00, Number 00, ’’ 2014
17. Armstrong JG, Harrison LB, Spiro RH, et al. Malignant tumors of major salivary gland origin. Arch Otolaryngol Head Neck Surg. 1990;116:290–293. 18. Garden AS, El-Naggar AK, Morrison WH, et al. Postoperative radiotherapy for malignant tumors of the parotid gland. Int J Radiat Oncol Biol Phys. 1997;37:79–85. 19. Mendenhall WM, Morris CG, Amdur RJ, et al. Radiotherapy alone or combined with surgery for salivary gland carcinoma. Cancer. 2005;103:2544–2550. 20. Terhaard CHJ, Lubsen H, Rasch CRN, et al. The role of radiotherapy in the treatment of malignant salivary gland tumors. Int J Radiat Oncol Biol Phys. 2005;61:103–111.
r
2014 Lippincott Williams & Wilkins
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
