Abstract
Background:
The role of elective neck dissection in the management of major salivary gland adenoid cystic carcinoma is unclear.
Methods:
Data were retrospectively extracted from the National Cancer Center Database. The study cohort included 1504 patients with adenoid cystic carcinoma of major salivary glands with clinical N0 necks who were treated with surgery between 2004 and 2014. The cohort was divided into four groups based on number of lymph nodes (LNs) examined on pathology: 0, 1-8, 9-17, and ≥18 LNs.
Results:
The rate of occult nodal metastasis was 9.0%. Number of LNs removed was not associated with survival (Reference, 0 LNs; HR = 0.98, 95% CI 0.73-1.32 for 1-8 LNs; HR = 1.22, 95% CI 0.80-1.88 for 9-17 LNs; HR = 0.94, 95% CI 0.61-1.46 for ≥18 LNs) after adjusting for important covariates.
Conclusions:
LN sampling is not associated with survival in cN0 major salivary gland ACC.
Keywords: adenoid cystic carcinoma, elective neck dissection, lymph node sampling, major salivary gland, neck dissection
1 ∣. INTRODUCTION
Adenoid cystic carcinoma (ACC) is an aggressive malignancy associated with poor survival, with estimates for the rate of 15-year recurrence-free survival from 30% to 37%.1,2 ACC is most common in major salivary glands and accounts for 22%-25% of all salivary gland malignancies.3,4 It behaves differently from other salivary gland cancers as it progresses slowly but has a high propensity for local recurrence, perineural invasion, and distant metastases.
ACC is typically believed to have infrequent metastasis to cervical lymph nodes (LNs). However, a recent study demonstrated a 17% rate of nodal metastases for ACC of major salivary glands, and a significant association between nodal metastasis and worse survival.5 In addition, a retrospective review of patients with clinical T1-2/N0 ACC of major salivary glands who received elective neck dissection (ND) found positive nodes in 7 out of 60 patients and that positive nodes were associated with distant metastases.6 Although these findings suggest that elective ND may be beneficial for these patients, a retrospective multicenter study of 270 patients did not show that elective ND had a survival advantage compared with those who did not.7 Ultimately, more data are needed to assess the impact of elective ND on survival. Therefore, the objective of the current study was to examine the impact of LN sampling on survival in patients with major salivary gland ACC.
2 ∣. PATIENTS AND METHODS
This study was granted an exemption by the Stanford University institutional review board. Data were extracted by from the National Cancer Database (NCDB), a hospital-based registry that draws from more than 1500 Commission on Cancer-accredited cancer programs that provides care to over 70% of all patients in the United States with newly diagnosed cancers. The NCDB is a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The study cohort included patients diagnosed with cN0 major salivary gland ACC (International Classification of Diseases for Oncology [ICD-O] histology code 8200, behavioral code 3) between 2004 and 2014. The following ICD-O codes were included: C07.0 (parotid gland), C08.0 (submandibular gland), C08.1 (sublingual gland), and C08.9 (major salivary gland, not other-wise specified). Only patients who underwent surgical resection of the primary tumor were included. Exclusion criteria included distant metastases, multiple primary tumors, and unknown treatment status.
The NCDB does not provide information on whether a ND was performed. However, it does provide information on the number of LNs examined in the pathology specimen. Patients were divided into four groups based on number of LNs examined on pathology: 0, 1-8, 9-17, and ≥18 LNs. It has been previously proposed that NDs with ≥18 LNs examined on pathology should be a quality metric for NDs in squamous cell carcinoma,8 which is the rationale behind the current grouping.
Information on clinical and pathologic stage was obtained from the NCDB. Disease stage was recorded in accordance with the American Joint Committee on Cancer Staging Manual, 7th edition.9 Race was recorded as “White”; “Black”; Other”; or “Unknown.” Charlson/Deyo score, which is a measure of comorbidity, was obtained from the NCDB (sum of scores assigned from 15 secondary ICD-9 diagnosis codes,10 with scores of 0, 1, 2, or ≥3).
The statistical analysis was performed using Stata 14 (StataCorp, 2015. Stata Statistical Software: Release 14. College Station, Texas: StataCorp LP). Survival analysis was performed using Kaplan–Meier analysis. The primary outcome variable was overall cumulative survival (OS). Multivariable analysis was performed using Cox proportional hazard regression analysis. Variables to be included for Cox regression were determined by individual univariable survival analyses of the following variables: T classification, age, sex, race, site, margin status, radiotherapy, Charlson–Deyo score, and year of diagnosis. Variables associated with survival at P ≤ .05 were entered into the model. Missing variables were handled by excluding cases from analysis. An estimate was considered statistically significant at α = 0.05.
3 ∣. RESULTS
We identified 1504 subjects with major salivary gland ACC who were cN0. The patient characteristics are displayed in Table 1. Seventy-seven percent of patients (n = 1504) had 0-8 LNs sampled. Of these, 314 had no LNs sampled. The number of nodes sampled was associated with clinical T classification (R2 = 0.04, F[1, 1502] = 49.84, P < .01) with increased number of nodes sampled in patients with higher clinical T classification (β = 0.19, 95% CI 0.13-0.23, P < .01). The overall rate of occult nodal metastasis for patients who had ≥1 LN examined was 9.0% (n = 867). The rate of occult nodal metastasis was 5.0% for patients with 1-8 LNs examined (n = 1163), 11.9% for patients with 9-17 LNs examined (n = 143), and 14.9% for patients with ≥18 LNs examined (n = 175). Radiotherapy was used in 74.5% of patients. The percentage of patients who received radiotherapy did not differ by number of LN examined (OR = 1.09, 95% CI 0.92-1.30, P = .32).
TABLE 1.
Patient characteristics. Subgroups with less than 10 patients were combined in order to adhere to data usage guidelines of the NCDB
| Patients, n (%) | |||||
|---|---|---|---|---|---|
| Characteristic | Overall (n = 1504) |
0 nodes examined (n = 562) |
1-8 nodes examined (n = 601) |
9-17 nodes examined (n = 157) |
≥18 nodes examined (n = 184) |
| Age, ya | 56.4 ± 16.5 | 57.7 ± 16.4 | 55.3 ± 17.1 | 56.4 ± 16.7 | 56.2 ± 14.4 |
| Female | 904 (60.1%) | 342 (60.9%) | 364 (60.6%) | 96 (61.2%) | 102 (55.4%) |
| Race | |||||
| White | 1207 (80.2%) | 456 (81.1%) | 482 (80.2%) | 121 (77.1%) | 148 (80.4%) |
| Black | 159 (10.6%) | 52 (9.3%) | 60 (10.0%) | 25 (15.9%) | 22 (12.0%) |
| Oth & Unk | 138 (9.2%) | 54 (9.6%) | 59 (9.8%) | 11 (7.0%) | 14 (7.6%) |
| Site | |||||
| Parotid | 781 (51.9%) | 231 (41.1%) | 361 (60.1%) | 95 (60.5%) | 94 (51.1%) |
| SM & SL | 723 (48.1%) | 331 (58.9%) | 240 (39.9%) | 62 (39.5%) | 90 (48.9%) |
| Charlson–Deyo | |||||
| 0 | 1308 (87%) | 532 (88.5%) | 532 (88.5%) | 128 (81.5%) | 158 (85.9%) |
| ≥1 | 196 (13%) | 69 (11.5%) | 69 (11.5%) | 29 (18.5%) | 26 (14.1%) |
| Radiation | 1121 (74.5%) | 417 (74.2%) | 441 (73.3%) | 123 (78.6%) | 140 (76%) |
| Positive margins | 668 (44.4%) | 275 (41.2%) | 258 (38.6%) | 61 (9.1%) | 74 (11.1%) |
| Positive nodes (path) | 104 (6.9%) | - | 58 (5.0%) | 20 (11.8%) | 26 (15.1%) |
| Clinical T classification | |||||
| T1 | 582 (38.7%) | 255 (45.4%) | 233 (38.7%) | 51 (32.7%) | 43 (23.7%) |
| T2 | 489 (32.5%) | 183 (32.6%) | 194 (32.3%) | 53 (34.0%) | 58 (31.5%) |
| T3 | 264 (17.5%) | 92 (16.3%) | 99 (16.5%) | 32 (20.1%) | 41 (22.4%) |
| T4 | 169 (11.3%) | 32 (5.7%) | 75 (12.5%) | 21 (13.2%) | 41 (22.4%) |
Abbreviations: Oth, Other; Unk, Unknown; SM, Submandibular; SL, Sublingual.
Mean ± SD.
Kaplan–Meier analysis showed no association between number of LNs examined and survival (5-year OS = 86%, 95% CI 82-89% for 0 LNs; 5-year OS = 83%, 95% CI 80-87% for 1-8 LNs; 5-year OS = 80%, 95% CI 71-87% for 9-17 LNs; 5-year OS = 83%, 95% CI 75-88% for ≥18 LNs; Figure 1). Positive occult nodes was associated with poorer survival in patients who had ≥9 nodes examined (5-year OS = 29%, 95% CI 5-59% for positive occult nodes; 5-year OS = 70%, 95% CI 58-79% for negative occult nodes; Figure 2).
FIGURE 1.
Overall survival based on number of lymph nodes removed. P = .53
FIGURE 2.
Overall survival based on presence of occult lymph nodes in patients who received lymph node sampling (≥9 nodes examined on pathology). P < .01
To identify variables to be included in the multivariable analysis, individual univariable survival analyses were conducted for T classification (P < .01), age (P < .01), sex (P = .04), race (P = .55), site (P = .52), margin status (P = .03), radiotherapy (P = .16), Charlson–Deyo score (P = .01), and year of diagnosis (P = 51). As such, the following variables were included in multivariable analysis: T classification, age, sex, margin status, and Charlson–Deyo score. The results of the multivariable analysis are shown in Table 2. In the overall cohort, the number of nodes removed was not associated with improved survival after adjusting for relevant covariates. The results of stratified multivariable analysis (by T classification) are shown in Table 3. The number of nodes removed was not associated with improved survival in either T1-2 or T3-4 groups.
TABLE 2.
Impact of number of lymph nodes examined on overall survival (multivariable analysis). Asterisks indicate significant (P < .05)
| Variable | Hazard ratio (95% CI) | P value |
|---|---|---|
| Lymph nodes examined | ||
| 0 | 1 | Reference |
| 1-8 | 0.98 (0.73-1.32) | .90 |
| 9-17 | 1.22 (0.80-1.88) | .36 |
| ≥18 | 0.94 (0.61-1.46) | .78 |
| Clinical T classification | ||
| T1 | 1.00 | Reference |
| T2 | 1.72 (1.18-2.52) | .01* |
| T3 | 2.25 (1.53-3.32) | <.01* |
| T4 | 5 (3.38-7.4) | <.01* |
| Age | 1.05 (1.04-1.06) | <.01* |
| Positive margins | 1.11 (0.86-1.44) | .42 |
| Female | 0.8 (0.62-1.03) | .08 |
| Charlson-Deyo score | 1.14 (0.81-1.59) | .45 |
TABLE 3.
Impact of umber of lymph nodes examined on overall survival by T classification (stratified multivariable analysis). Asterisks indicate significant (P < .05)
| T1-2 | T3-4 | |||
|---|---|---|---|---|
| Variable | Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value |
| Lymph nodes examined | Lymph nodes examined | |||
| 0 | 1 | Reference | 1 | Reference |
| 1-8 | 1.01 (0.65-1.56) | .97 | 1.25 (0.84-1.88) | .27 |
| 9-17 | 2.25 (1.24-4.09) | .07 | 0.87 (0.47-1.62) | .67 |
| ≥18 | 1.1 (0.49-2.46) | .82 | 1.13 (0.66-1.92) | .66 |
| Age | 1.06 (1.04-1.07) | <.01* | 1.04 (1.03-1.06) | <.01* |
| Female | 0.88 (0.59-1.29) | .51 | 0.74 (0.53-1.04) | .08 |
| Positive margins | 1.2 (0.81-1.78) | .36 | 1.03 (0.73-1.46) | .87 |
| Charlson–Deyo score | 1.13 (0.65-1.96) | .67 | 1.25 (0.82-1.91) | .30 |
4 ∣. DISCUSSION
The results of our study showed substantial occult nodal metastasis in patients with cN0 ACC of the major salivary glands undergoing LN sampling, with rates of occult nodal metastasis increasing with the number of LNs examined. Patients with positive occult nodes had worse survival than those negative for occult nodes. However, the number of LNs removed was not associated with survival, even after accounting for T classification and clinically important prognostic factors. This finding suggests that elective ND may not provide a survival benefit in this patient population. However, it may serve as a prognostic tool to identify patients with occult nodal metastasis, which is associated with worse survival.
The role for elective ND in major salivary gland ACC is unclear. Several authors recommend elective ND in all-cause salivary gland cancers if the probability of occult cervical metastasis is higher than 15-20%.11,12 Considering that ACC of major salivary glands has a 17-19% rate of nodal metastasis,5,13 this raises the question of whether elective ND is indicated in the management of ACC. Our current study showed 5-15% rate of occult nodal metastasis in our population of patients with ACC, with rates of occult nodal metastasis increasing with the number of LNs examined. Lee et al. examined the effect of elective ND on survival in 26 of 57 patients with cN0 ACC of major and minor salivary glands and found no impact on the rate of survival.14 Amit et al. also found no survival benefit for patients treated with elective ND in a cohort of 457 patients with head and neck ACC from multiple primary sites (250 from oral cavity, 133 from major salivary gland, 68 from sinonasal cavity, and 6 from larynx).7 Subgroup analysis by primary site and by early (T1-2) vs advanced (T3-4) disease also showed that elective ND was not associated with survival. These results are consistent with the findings of the current study.
To our knowledge, the present study comprising 1504 subjects is the largest to date. The study population is distinct from previous investigations in that it focuses exclusively on patients with major salivary gland ACC without clinical regional or distant metastases. Our study showed that the number of LNs sampled in these patients was not associated with survival regardless of T classification. This finding is important because previous studies show increased risk of nodal metastases in larger tumors. Megwalu et al. showed that T3 and T4 tumors were associated with 4- to 9-fold increase risk of nodal metastasis, compared to T1 and T2 tumors.5 Nevertheless, we found that LN sampling was not associated with improved survival. Interestingly, we found that the rates of occult nodal metastasis increased with the number of LNs examined. This may be due to increased likelihood of detecting occult metastasis with more extensive LN sampling. On the other hand, this may be explained by the findings that patients with advanced T classification, who are at higher risk of nodal metastasis, were more likely to undergo more extensive LN sampling.
The role of ND in cN0 necks has been debated extensively. Even in squamous cell carcinoma (SCC), there is an unpredictable propensity of occult neck metastases.15 In a prospective, randomized controlled trial, D'Cruz et al. showed that patients with early-stage oral SCC who received elective ND had higher rates of overall and disease-free survival than patients who were treated by watchful waiting followed by therapeutic ND for nodal relapse.16 Although elective ND is preferred by most authors for treatment of the cN0 neck in SCC,15-17 patients with ACC tend to receive radiation due to the high rate of perineural invasion. Postoperative radiation has been associated with a survival benefit in ACC with an advanced T classification and positive margins.18 Herman et al. found that elective ND in patients with salivary gland carcinomas of all entities did not benefit those who underwent postoperative radiation.19 In the present study, the majority of patients received postoperative radiation. Consequently, the lack of survival benefit for LN sampling in our cohort may be confounded by the fact that most patients received radiotherapy to the neck.
Although LN sampling was not associated with improved survival in our study, elective ND may be useful for prognostication. Our study found that the presence of occult metastases was associated with poorer survival. Therefore, elective ND may be useful if the desired outcome is to detect occult metastases for prognostic purposes.
The strength of our study lies in the large sample size and diverse patient characteristics. Utilizing the NCDB allows us to analyze a large and diverse population with outstanding quality control. The Commission on Cancer-accredited programs that serve at the catchment for the NCDB maintains a high-quality cancer reporting system. This study is primarily limited by the retrospective nature of the analysis. The NCDB does not definitively identify patients who underwent ND and does not identify the location of the sampled LNs (within the parotid gland vs in the neck). Consequently, we examined the impact of the number of LNs sampled on survival. Although large numbers of regional LNs examined by pathology may be interpreted as a proxy for ND, this cannot be concluded with certainty. This limitation must be considered if the present findings are considered in the management of cN0 ACC of major salivary glands.
As per the data use agreement with the NCDB, all groups containing less than 10 individuals must combined with larger groups. As such, patients with sublingual primary tumors were combined with submandibular group, although sublingual tumors have a higher rate of occult cervical metastases.20 The NCDB does not include information on disease recurrence, or development of nodal or distant metastasis beyond the initial treatment period, therefore, these outcomes could not be examined. The NCDB also does not provide information on perineural invasion, which has long been considered an important prognostic factor. However, several studies suggest that perineural invasion is not associated with survival.21,22 Finally, the NCDB does not provide information on disease-specific mortality. However, the impact of comorbidity on survival was accounted for using Charlson/Deyo score.
In conclusion, the presence of occult metastases in cN0 ACC of the major salivary glands is associated with poorer survival, but the number of LNs removed is not associated with improved survival. This suggests that elective ND may not be beneficial to survival in this situation, but may be useful as a prognostic tool to identify patients at risk for poor outcomes. Further studies are needed to validate our findings and clarify the role of ND in the management of major salivary gland ACC. Randomized control trials would be ideal, however, these may be hampered by small sample size. In the absence of randomized control trials, large, well-conducted comparative effectiveness analyses of observational data may provide much needed information on how to most effectively treat these patients.
ACKNOWLEDGMENTS
The NCDB is a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The data used in the study are derived from a de-identified NCDB file. The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology employed, or the conclusions drawn from these data by the investigators.
REFERENCES
- 1.Khan AJ, DiGiovanna MP, Ross DA, et al. Adenoid cystic carcinoma: a retrospective clinical review. Int J Cancer. 2001;96(3):149–158. http://www.ncbi.nlm.nih.gov/pubmed/11410883. [DOI] [PubMed] [Google Scholar]
- 2.Kokemueller H, Eckardt A, Brachvogel P, Hausamen J-E. Adenoid cystic carcinoma of the head and neck—a 20 years experience. Int J Oral Maxillofac Surg. 2004;33(1):25–31. 10.1054/ijom.2003.0448. [DOI] [PubMed] [Google Scholar]
- 3.Spiro RH. Salivary neoplasms: overview of a 35-year experience with 2,807 patients. Head Neck Surg. 1986;8(3):177–184. http://www.ncbi.nlm.nih.gov/pubmed/3744850. [DOI] [PubMed] [Google Scholar]
- 4.Laurie SA, Ho AL, Fury MG, Sherman E, Pfister DG. Systemic therapy in the management of metastatic or locally recurrent adenoid cystic carcinoma of the salivary glands: a systematic review. Lancet Oncol. 2011;12(8):815–824. 10.1016/S1470-2045(10)70245-X. [DOI] [PubMed] [Google Scholar]
- 5.Megwalu UC, Sirjani D. Risk of nodal metastasis in major salivary gland adenoid cystic carcinoma. Otolaryngol Head Neck Surg. 2017;156(4):660–664. 10.1177/0194599817690138. [DOI] [PubMed] [Google Scholar]
- 6.Bhayani MK, Yener M, El-Naggar A, et al. Prognosis and risk factors for early-stage adenoid cystic carcinoma of the major salivary glands. Cancer. 2012;118(11):2872–2878. 10.1002/cncr.26549. [DOI] [PubMed] [Google Scholar]
- 7.Amit M, Na'ara S, Sharma K, et al. Elective neck dissection in patients with head and neck adenoid cystic carcinoma: an international collaborative study. Ann Surg Oncol. 2015;22(4):1353–1359. 10.1245/s10434-014-4106-7. [DOI] [PubMed] [Google Scholar]
- 8.Divi V, Chen MM, Nussenbaum B, et al. Lymph node count from neck dissection predicts mortality in head and neck cancer. J Clin Oncol. 2016; 34(32):3892–3897. 10.1200/JCO.2016.67.3863. [DOI] [PubMed] [Google Scholar]
- 9.Edge SB. American Joint Committee on Cancer. AJCC Cancer Staging Manual. New York, NY, USA: Springer; 2010. https://www.springer.com/us/book/9780387884424. [DOI] [PubMed] [Google Scholar]
- 10.Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6): 613–619. http://www.ncbi.nlm.nih.gov/pubmed/1607900. [DOI] [PubMed] [Google Scholar]
- 11.Stennert E, Kisner D, Jungehuelsing M, et al. High incidence of lymph node metastasis in major salivary gland cancer. Arch Otolaryngol Neck Surg. 2003;129(7):720–723. [DOI] [PubMed] [Google Scholar]
- 12.Pitman KT. Rationale for elective neck dissection. Am J Otolaryngol. 2000; 21(1):31–37. http://www.ncbi.nlm.nih.gov/pubmed/10668674. [DOI] [PubMed] [Google Scholar]
- 13.Amit M, Binenbaum Y, Sharma K, et al. Incidence of cervical lymph node metastasis and its association with outcomes in patients with adenoid cystic carcinoma. An international collaborative study. Head Neck. 2015;37(7): 1032–1037. 10.1002/hed.23711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lee S-Y, Kim BH, Choi EC. Nineteen-year oncologic outcomes and the benefit of elective neck dissection in salivary gland adenoid cystic carcinoma. Head Neck. 2014;36(12):1796–1801. [DOI] [PubMed] [Google Scholar]
- 15.Cheng A, Schmidt BL. Management of the N0 neck in oral squamous cell carcinoma. Oral Maxillofac Surg Clin North Am. 2008;20(3):477–497. 10.1016/j.coms.2008.02.002. [DOI] [PubMed] [Google Scholar]
- 16.D'Cruz AK, Vaish R, Kapre N, et al. Elective versus therapeutic neck dissection in node-negative Oral cancer. N Engl J Med. 2015;373(6):521–529. 10.1056/NEJMoa1506007. [DOI] [PubMed] [Google Scholar]
- 17.O'Brien CJ, Traynor SJ, McNeil E, McMahon JD, Chaplin JM. The use of clinical criteria alone in the management of the clinically negative neck among patients with squamous cell carcinoma of the oral cavity and oropharynx. Arch Otolaryngol Head Neck Surg. 2000;126(3):360–365. http://www.ncbi.nlm.nih.gov/pubmed/10722009. [DOI] [PubMed] [Google Scholar]
- 18.Silverman DA, Carlson TP, Khuntia D, Bergstrom RT, Saxton J, Esclamado RM. Role for postoperative radiation therapy in adenoid cystic carcinoma of the head and neck. Laryngoscope. 2004;114(7):1194–1199. 10.1097/00005537-200407000-00012. [DOI] [PubMed] [Google Scholar]
- 19.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(3):205–208. 10.1016/j.amjoto.2012.11.012. [DOI] [PubMed] [Google Scholar]
- 20.Andreasen S, Bjørndal K, Agander TK, Wessel I, Homøe P. Tumors of the sublingual gland: a national clinicopathologic study of 29 cases. Eur Arch Oto-Rhino-Laryngology. 2016;273(11):3847–3856. 10.1007/s00405-016-4000-y. [DOI] [PubMed] [Google Scholar]
- 21.Andreasen S, Tan Q, Agander TK, et al. MicroRNA dysregulation in adenoid cystic carcinoma of the salivary gland in relation to prognosis and gene fusion status: a cohort study. Virchows Arch. 2018;473(3):329–340. 10.1007/s00428-018-2423-0. [DOI] [PubMed] [Google Scholar]
- 22.Amit M, Binenbaum Y, Trejo-Leider L, et al. International collaborative validation of intraneural invasion as a prognostic marker in adenoid cystic carcinoma of the head and neck. Head Neck. 2014;55(7):691–696. 10.1002/hed.23710. [DOI] [PubMed] [Google Scholar]