Abstract
Objective. Regional recurrence is common following surgery for T1/T2 oral tongue squamous cell carcinoma (SCC). Tumor depth >4.0 mm is commonly assigned as an indication for prophylactic neck dissection to improve regional control. Prophylactic neck dissection may detect extracapsular extension, a poor prognostic sign where adjuvant chemotherapy is indicated. The hypothesis in this study is that regional recurrence is a significant problem in 2.1- to 4.0-mm-depth tumors, and detection of extracapsular extension may be important in this group.
Study Design. Retrospective chart review.
Setting. Australian tertiary referral center.
Subjects and Methods. Review of all patients with T1/T2 oral tongue SCC treated surgically between January 1991 and January 2009 (n = 81).
Results. Twenty-nine prophylactic and 5 therapeutic neck dissections followed for a median 34 months (range, 4-132 months). Tumor depths were 0 to 2.0 mm (n = 15), 2.1 to 4.0 mm (n = 18), 4.1 to 7.0 mm (n = 26), and >7.0 mm (n = 22). Tumors 2.1 to 4.0 mm depth had similar rates of occult nodes as 4.1 to 7.0 mm depth (25% vs 20%). Regional recurrence occurred in 31% overall, 44% in tumors 2.1 to 4.0 mm, and 27% in tumors 4.1 to 7.0 mm depth. Prophylactic neck dissection reduced regional recurrence (17% vs 43%, P = .02). Patients with pathologically negative necks had lower rates of regional recurrence than those with occult nodes (9% vs 50%, P < .01). Extracapsular extension increased regional recurrence (43% vs 7%, P = .02), including 25% of dissected necks with tumor depth 2.1 to 4.0 mm.
Conclusions. Regional recurrence is a significant problem in 2.1- to 4.0-mm-depth T1/T2 tongue tumors. Prophylactic neck dissection may improve regional control in patients with adequate primary resection margins and determine need for adjuvant therapies in 2.1- to 4.0-mm-depth tumors.
Locoregional recurrences occur in approximately 1 in 4 patients with T1/T2 oral tongue squamous cell carcinoma. 1 Immediate prophylactic neck dissection has been shown to reduce regional recurrence 2 and identifies 2 important prognostic factors: occult nodal disease and extracapsular extension in approximately 25% of oral tumors. 3 Identifying extracapsular extension allows selective use of adjuvant chemotherapy with postoperative radiotherapy, a toxic treatment proven to improve locoregional control and survival. 4 Tumor depth is currently the best available predictor of occult nodal disease 5 and is used to determine when the benefits of prophylactic neck dissection outweigh its morbidity. 6 Earlier authors have recommended lower depth cutoffs, 7 but most authors currently recommend immediate prophylactic neck dissection for tumors >4.0 mm in depth. 8
Given the high rate of regional failure and improvement in adjuvant therapies, neck dissection may be underused using this strict 4-mm cutoff. The a priori hypothesis of this study is that routine prophylactic neck dissection in 2.1- to 4.0-mm-depth tumors may improve regional control and guide adjuvant therapy.
Methods
Institutional review board approval was obtained from the Melbourne Health Office for Research. All previously untreated T1/T2 oral tongue squamous cell carcinoma managed at the Royal Melbourne Hospital between January 1991 and January 2009 were identified (n = 81) from the prospectively maintained Head and Neck Oncology Database and medical records (International Classification of Diseases Coding 9 and 10). Retrospective staging was based on the sixth edition of the American Joint Committee on Cancer staging manual. Retrospective chart review identified 68 T1 tumors and 13 T2 tumors with a median follow-up of 34 months (range, 4-132 months). Five patients preoperatively had clinically or radiologically positive necks and underwent therapeutic neck dissection. These “therapeutic neck dissections” were not counted as “prophylactic neck dissections” and were not counted in the tables or calculations as such. No patients had distal metastases at presentation. Cases were discussed at weekly multidisciplinary head and neck multidisciplinary team meetings where the decision to perform surgery and administer adjuvant therapy was based on head and neck surgeon, head and neck oncologist, and patient preference; patient age; comorbidities; and presence of other poor prognostic factors. Neck dissections were performed by trained head and neck consultant surgeons (assisted by surgical trainees) who were members of the audited multidisciplinary head and neck team.
Tumor depth was taken from pathology reports, and where doubt existed, slides were reevaluated by a senior head and neck pathologist (MB) and placed into categories of 0 to 2.0 mm (n = 15), 2.1 to 4.0 mm (n = 18), 4.1 to 7.0 mm (n = 26), and >7.0 mm (n = 22). Other prognostic variables were recorded, including tumor differentiation, maximum diameter, lymphovascular invasion, perineural invasion, pathological cervical nodes, and extracapsular extension. Margins were described as clear if >5 mm (n = 24), close if 1 to 5 mm (n = 46), and involved if <1 mm (n = 11).
Prophylactic neck dissection was defined a priori as affecting adjuvant therapy if there was the following:
Extracapsular extension without involved margins on the primary tumor (an indication for adjuvant chemotherapy) or
Two or more involved cervical nodes without extracapsular extension or involved margins on the primary tumor (an indication for radiotherapy)
Statistical analysis was performed using SPSS software (SPSS version 13.0; SPSS, Inc, an IBM Company, Chicago, Illinois). Survival was estimated using the Kaplan-Meier method from the date of surgery to the date of recurrence or censored at last follow-up.
Results
Patient Characteristics
The median age at diagnosis was 63 years (range, 21-90 years). Younger patients were more likely to have tumors >7.0 mm depth (P = .03;
Characteristics of Patients With T1/T2 Oral Tongue Squamous Cell Carcinoma (n = 81)
Margins were described as clear if >5 mm, close if 1 to 5 mm, and involved if <1 mm (n = 11).
Deeper tumors had a progressively higher rate of cervical lymph node metastases (P = .03;
Incidence of neck metastases by tumor depth for T1/T2 oral tongue squamous cell carcinoma (n = 81).
Prophylactic neck dissections were performed in 29 patients (36%). Of these, 18 (62%) were supraomohyoid neck dissections, 8 (28%) were modified radical neck dissections, and 3 (10%) were selective level I to IV neck dissections. The 5 patients with clinically positive necks were treated with therapeutic modified radical neck dissections (levels I-V). Deeper tumors were more likely to receive a neck dissection (P = .02). Four patients (22%) with tumor depth 2.1 to 4.0 mm and 15 patients (58%) with tumor depth 4.1 to 7.0 mm had a prophylactic neck dissection.
Seven of the 29 patients (24%) having prophylactic neck dissection had occult neck metastases (
Pathological Neck Status on Prophylactic Neck Dissection for Patients Clinically Staged N0 With T1/T2 Oral Tongue Squamous Cell Carcinoma (n = 29)
Postoperative radiotherapy was used in 31 patients (38%) with doses of 50 to 66 Gy delivered to the primary site and involved cervical nodes. Fifteen of 31 patients (48%) receiving radiotherapy had undergone neck dissections. Concurrent single-agent chemotherapy with cisplatin was used in 6 patients. Two of 6 patients receiving chemoradiotherapy (33%) had neck dissections. Use of radiotherapy (P < .01) and chemotherapy (P = .01) increased with primary tumor depth.
Recurrence and Survival
Local recurrence occurred in 23 patients (28%) at a median of 12 months (
Pathological Findings and Regional Recurrence in Patients Having Neck Dissections for T1/T2 Oral Tongue Squamous Cell Carcinoma (n = 34)
Extracapsular extension was present in 7 dissected necks, including 1 patient with tumor depth 2.1 to 4.0 mm (
Prophylactic neck dissection was associated with reduced regional recurrence compared with observation and delayed therapeutic neck dissection (17% vs 43%, P = .02;
Regional Recurrence in T1/T2 Oral Tongue Squamous Cell Carcinoma (n = 25) Following Prophylactic Neck Dissection Compared With Observation of Neck
Pathologic findings following prophylactic neck dissection were strongly predictive of risk of nodal recurrence; a pathologically negative neck had a significantly lower rate of regional recurrence than patients with involved nodes (9% vs 50%, P < .01;
Disease-free survival was generally associated with greater depth of tumor but similar between 2.1 to 4.0 mm and 4.1 to 7.0 mm depth (
Kaplan-Meier curve of survival by tumor depth for patients with T1/T2 oral tongue squamous cell carcinoma (n = 81).
Impact on Adjuvant Therapy
Prophylactic neck dissection affected adjuvant therapy by indicating the following:
Concurrent chemotherapy and radiotherapy in 7 clinically N0 patients (21%) who had extracapsular extension without involved margins on the primary tumor. This included 1 patient (17%) with tumor depth 2.1 to 4 mm.
Radiotherapy alone in 3 clinically N0 patients (9%) who had 2 or more involved cervical nodes without extracapsular extension or involved margins at the primary site.
Discussion
Currently, prophylactic neck dissection is commonly performed for oral tongue squamous cell carcinoma with greater than 4 mm depth of invasion. 8 Regional failure is a significant problem in tumors 2.1 to 4.0 mm deep, occurring in 44% in this series. Prophylactic neck dissection significantly reduced regional recurrence overall (17% vs 43%, P = .02), and this trend was also observed in the 2.1- to 4.0-mm deep subgroup (25% vs 50%, P = .38). Furthermore, prophylactic neck dissection helped stratify the need for adjuvant therapy by identifying extracapsular extension (21%) and multiple involved nodes (9%) in a proportion. These findings support routine prophylactic neck dissection for 2.1- to 4.0-mm-depth oral tongue squamous cell carcinoma.
Early oral tongue squamous cell carcinoma is a biologically aggressive disease with occult metastases not infrequent. 9 Occult cervical metastasis occurred in 24% of T1/T2 tumors in this series. Haddadin et al 10 found the incidence of reported occult cervical metastasis in early oral tongue T1/T2 tumors to range from 20% to 48%. Like many authors, 5 our study demonstrated an increasing incidence of cervical lymph node metastasis as tumor depth increased (P = .03).
The critical tumor depth at which authors have recommended dissection of the neck varies, but depth greater than 4.0 mm is commonly used. 8 Indeed, at our institution, prophylactic neck dissection was used in only 22% of 2.1- to 4.0-mm-depth tumors compared with 58% of 4.1- to 7.0-mm-depth tumors. Nevertheless, tumors 2.0 to 4.0 mm depth had a high regional recurrence rate (44%). We strongly favor the advice of Spiro et al, 7 who in 1986 recommended that prophylactic neck dissection be used in most patients with tumors more than 2.0 mm depth. Since then, improved data supporting the selective use of adjuvant therapy encourage identification of patients with occult nodal disease, especially with extracapsular extension. 4
Importantly, our series found occult metastasis in 25% of patients with tumors 2.1 to 4.0-mm deep. A decision analysis by Weiss et al 6 found that an occult nodal metastatic rate of more than 20% would mean that the benefit in local control of prophylactic neck dissection outweighs the morbidity of the procedure. Unfortunately, there is difficulty attributing statistical risk to this because of the small number of patients in our critical discussion group. However, although the sample size in this subgroup was limited, our data support prophylactic neck dissection in 2.1- to 4.0-mm-depth tumors.
In concordance with the rates of occult metastases, our study found regional failure to occur in 31% of patients. This is consistent with a recently published prospective randomized control trial by Yuen et al, 11 which found a 37% nodal recurrence rate for T1/T2 clinically N0 oral tongue squamous cell carcinoma. Of note, we found the rate of regional failure in the 2.1- to 4.0-mm depth group to be higher than that in the 4.1- to 7.0-mm-group (44% vs 27%), calling into question the rationale for infrequent use of prophylactic neck dissection in the 2.1- to 4.0-mm-depth tumor group at our institution (22% vs 58%).
Prophylactic neck dissection significantly reduced regional recurrence in the entire cohort (17% vs 43%, P = .02), with the same trend in the 2.1- to 4.0-mm-depth group (25% vs 50%, P = .38). Yuen et al 11 also found a significant reduction in regional recurrence with prophylactic neck dissection (6% vs 37%, P = .001). Similarly, a large retrospective study of 359 patients with T1/T2 clinically N0 of the oral cavity and oropharynx showed reduced regional recurrence following prophylactic neck dissection (8% vs 27%, P < .0001). 12 It should be noted that our study lacked a consistently defined protocol over the 2 decades to determine which patients should receive neck dissections. This limits statistical interpretation of the data.
It should be noted that although only 13% of primary tumors had involved <1-mm margins, a large proportion of patients had primary tumors that were excised with close 1- to 5-mm margins (57%). The high proportion of tumors excised with close margins was a trend in all groups, including thinner tumors. Failure of local control in cases of close or positive margins is likely to have contributed to regional spread, and this does cloud outcomes of treatment. It is unclear what role a prophylactic neck dissection has in the presence of local recurrence. This compromises interpretation of the results but does emphasize the need for close attention to achieving clear margins of the primary tumor. Indeed, more radical margins may reduce the need for neck dissection as well as radiotherapy treatment at the initial time of therapy.
The primary arguments against prophylactic dissection are that morbidity of neck dissection can be avoided by close observation and surgical salvage. Although Yuen et al 11 also found prophylactic dissection to reduce regional recurrence, they found little survival improvement compared with close observation (87% vs 89%). Interestingly, Yuen et al have found salvage of isolated nodal recurrence to be highly effective (100% of the 15 regional recurrences were salvaged). However, in experienced hands, routine prophylactic selective neck dissection has been shown to be a low-morbidity surgical option with equal efficacy as traditional modified radical neck dissection in the clinically negative neck. 13 Other authors have a much less favorable experience with salvage for neck recurrence in the observed neck. Kowalski 14 reviewed 513 patients with oral squamous cell carcinoma and found 16% had isolated neck recurrences, of which only 11% were successfully salvaged. If observation is chosen, our study highlights the need for closest monitoring in the first 2 years, with local and regional recurrence occurring at a median of 12 and 10 months, respectively.
Prophylactic neck dissection allows more accurate use of adjuvant therapies; patients with occult nodal disease had a greater need for adjuvant therapy to prevent regional recurrence (9% vs 50%, P < .01). Regardless of the controversy over observation with salvage for regional recurrence, the survival benefit of adjuvant chemotherapy for extracapsular extension is well proven and must be given at the time of adjuvant radiotherapy. 4 Importantly, 25% of prophylactically neck dissected patients in the 2.1- to 4.0-mm group had extracapsular extension, similar to the rate found in other series. 3 Consequently, our study demonstrates the impact of prophylactic neck dissection on adjuvant therapy, indicating chemoradiotherapy in 21% and radiotherapy alone in 9% of patients.
The role of routine prophylactic neck dissection in early oral tongue squamous cell carcinoma 2.1 to 4.0 mm depth remains controversial. However, the high incidence of occult cervical metastases closely correlated with the significant rate of regional failure, the variable results of salvage surgery, and the ability of prophylactic neck dissection to reduce regional recurrence provide strong arguments in favor of prophylactic dissection of the neck in this group. More recently, robust data highlight the need for prophylactic neck dissection to detect extracapsular extension in occult cervical lymph nodes and guide appropriate use of toxic chemotherapy to improve locoregional control and survival in this subgroup.
Conclusion
Regional recurrence is a significant problem following surgical treatment of T1/T2 oral tongue squamous cell carcinoma of 2.1 to 4.0 mm depth. Prophylactic neck dissection may improve regional control in patients with adequate primary resection margins and better guide the need for adjuvant therapies in 2.1- to 4.0-mm tumors of the oral tongue.
Author Contributions
Disclosures
Footnotes
No sponsorships or competing interests have been disclosed for this article.
Awarded the Australian Society of Otolaryngology Head and Neck Surgery Diamond Jubilee Poster Prize at the Australian Society of Otolaryngology Head and Neck Surgery Annual Scientific Meeting, Sydney, New South Wales, Australia, March 27-31, 2010