Neck observation is appropriate in T1–2, cN0 oral squamous cell carcinoma without perineural invasion or lymphovascular invasion

Oral Oncology 50 (2014) 857–862

Contents lists available at ScienceDirect

Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology

Neck observation is appropriate in T1–2, cN0 oral squamous cell carcinoma without perineural invasion or lymphovascular invasion Chien-Fu Yeh a, Wing-Yin Li b, Muh-Hwa Yang c,d,e,f, Pen-Yuan Chu a, Yen-Ting Lu a, Yi-Fen Wang a, Peter Mu-Hsin Chang f, Shyh-Kuan Tai a,c,d,g,⇑ a

Departments of Otolaryngology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Road, Taipei 11217, Taiwan Departments of Pathology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Road, Taipei 11217, Taiwan Infection and Immunity Research Center, National Yang-Ming University, No. 155, Sec. 2, Linong Street, Taipei 11221, Taiwan d Immunology Center, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Road, Taipei 11217, Taiwan e Institute of Clinical Medicine, National Yang-Ming University, No. 155, Sec. 2, Linong Street, Taipei 11221, Taiwan f Division of Medical Oncology, Department of Medicine, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Road, Taipei 11217, Taiwan g Department of Otolaryngology, National Yang-Ming University, No. 155, Sec. 2, Linong Street, Taipei 11221, Taiwan b c

a r t i c l e

i n f o

Article history: Received 17 March 2014 Received in revised form 30 May 2014 Accepted 2 June 2014 Available online 4 July 2014 Keywords: Oral squamous cell carcinoma Perineural invasion Lymphovascular invasion Neck dissection Observation

s u m m a r y Objectives: Management of cN0 neck, elective neck dissection (END) or observation, remains controversial for T1–2 oral squamous cell carcinoma (OSCC). To allow for the safe observation of cN0 neck, it is mandatory to define predictors with high negative predictive value (NPV) for cervical lymph node (LN) status. Materials and Methods: Pathologic re-evaluation was performed in tumors of 253 consecutive patients with T1–2, cN0 OSCC. The predictive roles of pathologic parameters for cervical LN status in guiding neck management were investigated. Results: Cervical LN metastasis (LN+) occurred at a similar rate between observation and END groups (20.8% vs. 22.2%, p = 0.807), indicating poor discriminatory value for cervical LN status by clinical judgment. Compared with T classification, tumor thickness and differentiation, PNI/LVI (perineural invasion/lymphovascular invasion) demonstrated the highest NPV (85.5%). Hypothetically using PNI/LVI status to guide neck management, a dramatic reduction in overtreatment rate could be achieved (54.2% to 20.2%), with a minimal increase in undertreatment rate (6.3% to 9.9%). In patients without PNI or LVI (PNI/LVI ), the ultimate neck control rate (96.9% vs. 96.3%, p = 1.000) and 5-year disease-specific survival rate (91.1% vs. 92.8%, p = 0.863) were equivalent between observation and END. However, a significantly higher incidence of neck recurrence was found with observation (16.9% vs. 6.5%, p = 0.031), with 93.8% occurring within one year and 73.3% being successfully salvaged. Conclusion: Observation under close follow-up for the first year is appropriate in T1–2, cN0 OSCC without PNI or LVI, for the achievement of equivalent ultimate neck control and 5-year disease-specific survival rates compared with END. Ó 2014 Elsevier Ltd. All rights reserved.

Introduction Surgery is the mainstay of treatment for T1–2 oral squamous cell carcinoma (OSCC). Cervical lymph node (LN) metastasis is a major poor prognostic factor, and occult metastasis can exist in 15–40% of patients presenting as clinical N0 (cN0) [1,2]. For the management of cN0 neck, elective neck dissection (END) has been ⇑ Corresponding author at: Department of Otolaryngology, National Yang-Ming University, No. 155, Sec. 2, Linong Street, Taipei 11221, Taiwan. Tel.: +886 2 2875 7337; fax: +886 2 2875 7338. E-mail address: [email protected] (S.-K. Tai). http://dx.doi.org/10.1016/j.oraloncology.2014.06.002 1368-8375/Ó 2014 Elsevier Ltd. All rights reserved.

widely advocated to provide precise pathologic examination and better neck control [3–5]. However, END has not been shown to provide survival benefit in most studies [6–10]. Considering possible sequelae such as scarring and shoulder disability [11,12], END is an overtreatment for the majority of cN0 patients without occult metastasis. Sentinel LN biopsy (SLNB) has been reported as a promising alternative to END [13]. However, SLNB is technically demanding and still not considered a standard of care in routine practice [14,15]. Another treatment choice which has been shown to provide similar survival outcomes is observation [15,16]. Nevertheless, observation of cN0 neck with occult metastasis is obviously an undertreatment and carries the risk of advanced

858

C.-F. Yeh et al. / Oral Oncology 50 (2014) 857–862

disease progression. To date, no consensus regarding the optimal management has been made [17,18]. To safely observe cN0 neck, it is necessary to define parameters which effectively predict cervical LN status. The risk of cervical LN metastasis is commonly assessed by clinical judgment preoperatively, mainly based on T classification and thickness by physical examination. Both features can be further precisely evaluated by pathologic examination together [19] with additional parameters which has been known to correlate with the metastatic ability including differentiation [20,21], perineural invasion (PNI) [21–23] and lymphovascular invasion (LVI) [24,25]. PNI and LVI are established poor prognostic factors in human malignancies [26–28]. In head and neck cancers, PNI and LVI both correlate with advanced tumor stage and poor survival [22–25]. Our recent reports demonstrated that PNI and LVI both independently predict cervical LN metastasis in early T1–2 OSCC [29–31]. We therefore hypothesize that the absence of PNI or LVI may predict the lack of occult metastasis. In this study, we aimed to clarify the roles of PNI/LVI and other pathologic parameters in selecting T1–2, cN0 OSCC patients for neck observation. Materials and methods Patient population From June 2001 to August 2009, 272 consecutive newly diagnosed T1–2, cN0 OSCC patients underwent curative surgery at the Department of Otolaryngology, Taipei Veterans General Hospital. Patients with previous cancer history were excluded unless disease-free for >2 years. The cN0 status was determined by physical examination and imaging including computed tomography or magnetic resonance imaging. Nineteen patients who underwent neck observation were excluded as it was difficult to determine their initial cervical LN status because of: (1) arrangement of postoperative adjuvant therapy in 7; (2) development of local recurrence prior to neck recurrence in 1; and (3) short follow-up period <2 years in 11. The remaining 253 patients were included in this study. The mean age was 54.2 years (range, 22–85 years), and 219 (86.6%) patients were male. Medical records from the hospital database were retrospectively reviewed and updated under the approval of hospital’s institutional review board. Our institutional policy advocated END for cN0 neck, and the final decision was made by clinical judgment without strict criteria. END was performed in 176 (69.6%) patients (the END group), and observation in 77 (30.4%) patients (the observation group). Cervical LN metastasis (LN+) was defined as: (1) pathologic diagnosis of LN metastasis in the END group; (2) neck recurrence within 2 years without antecedent or synchronous local recurrence in the observation group. Postoperative adjuvant therapy was given to 31 patients in the END group for high-risk features including positive margin in 13, N2 disease in 14, or extracapsular spread in 7. All patients in the observation group underwent surgery alone.

2 months for the second year, and every 3 months thereafter. The median follow-up period was 61.9 months (range, 24–130 months) for those who underwent observation and remained free from neck recurrence. Good local and regional control was achieved in 221 (87.4%) and 218 (86.2%) patients, respectively, after long-term follow-up. The 5-year disease-specific survival (DSS) rate was 89.5% for the entire cohort, and 93.6% and 85.7% for T1 and T2 patients, respectively. Pathologic examination Archived hematoxylin and eosin-stained slides of the tumor resections from each patient were collected from the pathology department. A senior head and neck cancer pathologist blinded to clinical data re-reviewed all of the slides. Negative margin, defined as the absence of tumor cell involvement at the resection margins, was achieved in 238 (94.1%) patients. Tumor thickness was measured to the nearest 1 mm on serial sections vertically from the tumor surface or ulcer base to the deepest point of invasion. The median tumor thickness was 6.0 mm (range, 1–31 mm). PNI was defined as tumor cell infiltration in any layer of the nerve sheath or tumor in close proximity involving more than one third of the nerve circumference [26]. LVI was defined as tumor cell infiltration in endothelium-lined vessels or tumor nests within or attached to the endothelial cell lining of the lymphovascular space. PNI and LVI were combined in this study. Eighty-one (32.0%) patients whose tumor demonstrated PNI and/or LVI were grouped as PNI/LVI-positive (PNI/LVI+), and the other 172 (68.0%) patients without PNI or LVI were grouped as PNI/LVI-negative (PNI/LVI ). Statistical analysis Endpoints of the study were the roles of pathologic parameters in guiding cN0 management. Categorical data were compared using chi-square analysis or Fisher’s exact test. The negative predictive value (NPV), positive predictive value (PPV) and odds ratio (OR) of each parameter were calculated accordingly, with focus on NPV for guiding neck observation. Cervical LN status was used as the basis to define the appropriateness of neck management. Appropriate management indicated precise arrangement of END in LN+ or observation in LN patients. Inappropriate management included overtreatment (END in LN patients) and undertreatment (observation in LN+ patients). Follow-up time was defined as the interval between the date of surgery and events or last contact. Disease-specific death was defined as death from the index tumor or treatment-related events. The Kaplan–Meier method was used to construct survival curves, and the log-rank test was used to evaluate the differences. All statistical analyses were carried out using the Statistical Package for Social Sciences software (SPSS for Windows version 17.0, SPSS Inc., Chicago, IL). Results were considered significant at p < 0.05.

Results Tumor staging and oncologic results Efficacy of END was dependent on cervical LN status Tumor stage was defined postoperatively by the multidisciplinary head and neck cancer tumor board according to the 2002 sixth edition of the American Joint Commission on Cancer TNM staging system. Under our definition, 198 (78.3%) patients were negative for cervical LN metastasis (LN ) and 55 (21.7%) were positive (LN+), 39 by pathologic diagnosis after END and 16 by neck recurrence under observation. The LN+ rates were 18.0% and 25.6% for T1 and T2 patients, respectively. After treatment completion, patients were closely followed every month for the first year, every

Clinicopathologic features are listed in Table 1. Patients in the END group demonstrated more aggressive pathologic features including increased tumor thickness, T2 classification, less differentiation and the presence of PNI/LVI, compared with patients in the observation group (p < 0.001). However, LN+ rates in the observation and END groups were similar (20.8% vs. 22.2%, p = 0.807), demonstrating poor discriminatory value of clinical judgment for cervical LN status. Although neck recurrence rate was significantly

859

C.-F. Yeh et al. / Oral Oncology 50 (2014) 857–862 Table 1 Clinicopathological features by neck management. Characteristics Tumor thickness 0–6 mm >6 mm Pathologic T stage T1 T2 Differentiation Well Moderate/poor PNI/LVIa PNI/LVI PNI/LVI+ Cervical LN metastasisb LN LN+ Local recurrence No Yes Neck recurrence No Yes 5-year DSS (%) 5-year OS (%)

Obs no. (%)

END no. (%)

59 (76.6) 18 (23.4)

76 (43.2) 100 (56.8)

56 (72.7) 21 (27.3)

72 (40.9) 104 (59.1)

64 (83.1) 13 (16.9)

107 (60.8) 69 (39.2)

65 (84.4) 12 (15.6)

107 (60.8) 69 (39.2)

61 (79.2) 16 (20.8)

137 (77.8) 39 (22.2)

69 (89.6) 8 (10.4)

152 (86.4) 24 (13.6)

58 (75.3) 19 (24.7) 88.3 84.6

160 (90.9) 16 (9.1) 90.1 84.5

p <0.001

<0.001

<0.001

<0.001

0.807

0.475

0.001

0.858 0.438

Obs – observation, END – elective neck dissection, PNI – perineural invasion, LVI – lymphovascular invasion, LN – lymph node, DSS – disease-specific survival, OS – overall survival. a PNI/LVI+ indicates the presence of PNI and/or LVI; PNI/LVI indicates no PNI or LVI. b See definition of cervical LN metastasis in ‘‘Patients and Methods’’.

lower in the END group (9.1% vs. 24.7%, p = 0.001), END did not contribute to better 5-year DSS and overall survival. Patients were next stratified according to cervical LN status. In the LN group, there was no difference in the neck recurrence (5.8% vs. 4.9%, p = 1.000) and 5-year DSS rates (95.9% vs. 97.4%, p = 0.397, Fig. 1A) between patients undergoing END and observation. In contrast, in the LN+ group, END contributed to a significantly lower neck recurrence rate (20.5% vs. 100%, p < 0.001), and a trend toward a better 5-year DSS rate (70.9% vs 46.9%, p = 0.177, Fig. 1B). These results indicate that the efficacy of END is dependent on cervical LN status which is poorly predicted by clinical judgment. Predictive value of PNI/LVI for occult neck metastasis The relationships between pathologic parameters and LN+ are listed in Table 2. A non-significant trend of association was demonstrated for T classification, tumor thickness and differentiation. The predictive values of the three parameters were similar, with NPV and PPV around 80% and 25%, respectively, and odds ratios around 0.65. In contrast, PNI/LVI demonstrated a significant association, and the LN+ rate was significantly lower in the PNI/LVI group (14.5% vs. 37.0%, p < 0.001). PNI/LVI also exhibited the highest NPV of 85.5%, as well as the highest PPV of 37.0%. The odds ratio of being LN+ was 0.289 for PNI/LVI patients, indicating a > 70% reduction of the risk compared with PNI/LVI+ patients. Improved appropriate neck treatment rate by PNI/LVI status Using cervical LN status as the standard, 153 (60.5%) patients did not receive appropriate neck treatment by clinical judgment, including 137 (54.2%) cases of overtreatment and 16 (6.3%) cases of undertreatment (Fig. 2A). A hypothetical model using PNI/LVI status to guide neck management was proposed in Fig. 2B. In this model, a 50% decrease in the number of patients receiving END was achieved, with more than doubled the number of cases receiving

observation. Consequently, the appropriate neck management rate remarkably increased from 39.5% to 70.0%. Importantly, this improvement was achieved through a 2.5-fold decrease in the overtreatment rate (54.2% to 20.2%) while maintaining a low undertreatment rate (6.3% to 9.9%). Observation achieved equivalent survival in the PNI/LVI

group

To further clarify the appropriateness of observation in PNI/ LVI patients, the impact of neck management was evaluated by stratification according to PNI/LVI status. In PNI/LVI patients, the 5-year DSS rate was almost the same between those undergoing observation and those undergoing END (91.1% vs. 92.8%, p = 0.863, Fig. 1C). In contrast, in PNI/LVI+ patients, the 5-year DSS rate was 16% lower in those undergoing observation (70.0% vs. 85.9%, p = 0.265, Fig. 1D). Of note, in the PNI/LVI group, a significantly higher rate of neck recurrence was shown in those undergoing observation (16.9% vs. 6.5%, p = 0.031, Table 3). However, after salvage treatment, the ultimate neck control rates were equivalent between observation and END (96.9% vs. 96.3%, p = 1.000). Of the 16 patients who underwent observation and were determined to be LN+ by neck recurrence, the median time to neck recurrence was 5.6 months (range, 1–16 months), and 15 (93.8%) cases developed within 12 months. Excluding one patient who was lost to follow-up after neck recurrence, 11 (73.3%) were successfully salvaged, but 8 (72.7%) of them required adjuvant chemoradiotherapy in addition to salvage neck dissection. Discussion The decision regarding the management of cN0 neck in early T1–2 OSCC has been debated extensively [15,17,32,33]. Consistent with previous reports [34–36], clinical judgment did not discriminate cervical LN status, even though it effectively selected patients whose tumors exhibited more aggressive pathologic features in this study (Table 1). A simplified policy against modern concept of individualized cancer therapy is to perform END for all cN0 patients. This policy solves the problems of the uncertainty of preoperative cervical LN status and the risk of occult metastasis progression. However, most cN0 patients who are truly LN (78.3% in this study) will be overtreated. Because previous studies have failed to demonstrate the survival benefit of END despite improved neck control, high-quality randomized control trials have been suggested for clarification [8,15,21]. In the current study, END did not play a role in improving either neck control or DSS among the truly LN patients. This result, although not altogether surprising, should be taken into consideration in future study design. Without excluding the large number of LN patients who will do well with any treatment, it is very difficult to reach a definitive conclusion even for randomized control trials. This highlights the importance of robust stratification for the risk of occult metastasis. Defining efficient parameters should be regarded as the first essential step to develop targeted surgical management policies for cN0 neck in patients with T1–2 OSCC. Both PNI and LVI have been reported to be predictive for LN+ [20–24]. However, clinical contribution of LN+ predictors is limited with wide application of END. In contrast, predictors with high NPV will be most valuable for selecting appropriate patients for observation. In this study, PNI/LVI demonstrated the highest NPV, and was the only pathologic parameter significantly associated with occult metastasis (Table 2). When hypothetically tested in our patient cohort, PNI/LVI demonstrated the efficacy of reducing the overtreatment rate while maintaining a low undertreatment

860

C.-F. Yeh et al. / Oral Oncology 50 (2014) 857–862

Fig. 1. Disease-specific survival (DSS) according to neck management (A) DSS in patients without cervical LN metastasis (LN ), (B) DSS in patients with cervical LN metastasis (LN+), (C) DSS in PNI/LVI patients and (D) DSS in PNI/LVI+ patients. PNI/LVI indicates no PNI or LVI; PNI/LVI+ indicates the presence of PNI and/or LVI.

rate below 10% (Fig. 2). The fact that, in PNI/LVI patients, ultimate neck control and 5-year DSS rates were equivalent between observation and END groups (Table 3 and Fig. 1C) further supports the concept that observation is appropriate for T1–2, cN0 OSCC patients who are PNI/LVI . Because of the retrospective design, our results should be further validated in future prospective studies. More favorable clinical parameters could be observed in the observation group (Table 1). However, such selection bias did not interfere with LN status determination, or the poor survival outcome in LN+ or PNI/LVI+ patients who received neck observation (Fig. 1B and D). One major limitation of this study is that pathologic parameters are not available to aid in preoperative or intraoperative decision making, including thickness, differentiation and PNI/LVI. A second

operation under general anesthesia would be required when making the decision for END by PNI/LVI status. Whether or not to achieve a lower overtreatment rate with equivalent survival outcomes at the cost of such inconvenience is at the discretion of physicians’ decision. Because clinical judgment poorly predicts cervical LN status and molecular markers generally remain investigational [37], pathologic parameters are currently the most readily available indicators to assist in clinical decision making, even though they are not perfect. With further exploration of tumor biology between PNI/LVI and cervical LN metastasis, we believe that ideal predictors for individualized surgical planning can be developed in the future. Another limitation is the higher neck recurrence rate under the observation policy. This result, however, should not preclude the

861

C.-F. Yeh et al. / Oral Oncology 50 (2014) 857–862 Table 2 Histopathologic features predictive of cervical LN metastasis. Cervical LN metastasisa (no.)

Variables (no.)

NPV (%)

PPV (%)

OR

(95% CI)

p

LN

LN+

T classification T1 (128) T2 (125)

105 93

23 32

82.0

25.6

0.637 1

(0.341–1.165)

0.141

Thickness 0–6 mm (135) >6 mm (118)

110 88

25 30

81.5

25.4

0.667 1

(0.366–1.215)

0.184

Differentiation Well (171) Moderate/poor (82)

138 60

33 22

80.7

26.8

0.652 1

(0.351–1.211)

0.174

PNI/LVIb PNI/LVI (172) PNI/LVI+ (81)

147 51

25 30

85.5

37.0

0.289 1

(0.156–0.537)

<0.001

LN – lymph node, NPV – negative predictive value, PPV – positive predictive value, OR – odds ratio, CI – confidence interval, PNI – perineural invasion, LVI – lymphovascular invasion. a See definition of cervical LN metastasis in ‘‘Patients and Methods’’. b PNI/LVI+ indicates the presence of PNI and/or LVI; PNI/LVI indicates no PNI or LVI.

Fig. 2. Appropriateness of neck management. Cervical LN status, LN in 198 and LN+ in 55 patients, was used as the basis to determine the appropriateness. (A) Original clinical judgment. (B) Hypothetical PNI/LVI model assuming elective neck dissection (END) in 81 PNI/LVI+ patients and observation (Obs) in 172 PNI/LVI patients. PNI/LVI indicates no PNI or LVI; PNI/LVI+ indicates the presence of PNI and/or LVI.

Table 3 Impact of neck management on neck recurrence and ultimate neck control stratified by PNI/LVI status. Neck recurrence No. (%) PNI/LVI (n = 172) Obs END PNI/LVI+ (n = 81)a Obs END

Ultimate neck control P

No. (%)

0.031 11/65 (16.9) 7/107 (6.5)

1.000 63/65 (96.9) 103/107 (96.3)

<0.001 8/12 (66.7) 9/69 (13.0)

P

0.198 9/11 (81.8) 63/67 (94.0)

PNI – perineural invasion, LVI – lymphovascular invasion, PNI/LVI+ – PNI and/or LVI, PNI/LVI – no PNI or LVI, Obs – observation, END – elective neck dissection. a Three patients in PNI/LVI+ group were lost to follow-up after neck recurrence.

appropriateness of observation in PNI/LVI patients. The high salvage rate observed in this study has been considered critical

for the equivalent survival in patients undergoing observation [8,15]. However, the high percentage of patients requiring adjuvant chemoradiotherapy for salvage still indicates the need of improvement in early diagnosis of neck recurrence. In addition to close follow-up, Yuen et al. [8] performed neck ultrasonographic surveillance every 3 months for the first 3 years. However, they did not find nonpalpable neck recurrence and suggested a more frequent follow-up imaging schedule. Because 93.8% of neck recurrence in our LN+ patients who underwent observation occurred within one year, it is reasonable to suggest monthly follow-up and bimonthly neck ultrasonography for the first year. Aggressive work-up or salvage neck dissection for any suspicious finding should be performed to assure best survival outcomes. Further investigation is needed to clarify the cost-effectiveness of this follow-up protocol. In conclusion, using PNI/LVI status in guiding neck management can significantly reduce the overtreatment rate while maintaining a low undertreatment rate. Observation is thus appropriate for the management of cN0 neck in T1–2 OSCC without PNI or LVI, because ultimate neck control and 5-year DSS rates are equivalent compared with END. However, close monthly follow-up and bimonthly ultrasonogarphy are suggested for the first year.

Conflict of interest statement None declared. Acknowledgements The authors thank the Clinical Research Core Laboratory, Taipei Veterans General Hospital, for technical assistance. This work was supported in part by National Science Council of Taiwan (Grant Number 101-2314-B-010-021-MY3), and Taipei Veterans General Hospital (Grant Numbers V101C-057, V102C-087, V102E4-004, V103C-017). References [1] Beenken SW, Krontiras H, Maddox WA, Peters GE, Soong S, Urist MM. T1 and T2 squamous cell carcinoma of the oral tongue: prognostic factors and the role of elective lymph node dissection. Head Neck 1999;21(2):124–30. [2] Psychogios G, Mantsopoulos K, Bohr C, Koch M, Zenk J, Iro H. Incidence of occult cervical metastasis in head and neck carcinomas: development over time. J Surg Oncol 2013;107(4):384–7.

862

C.-F. Yeh et al. / Oral Oncology 50 (2014) 857–862

[3] Keski-Santti H, Atula T, Tornwall J, Koivunen P, Makitie A. Elective neck treatment versus observation in patients with T1/T2 N0 squamous cell carcinoma of oral tongue. Oral Oncol 2006;42(1):96–101. [4] Ferlito A, Silver CE, Rinaldo A. Elective management of the neck in oral cavity squamous carcinoma: current concepts supported by prospective studies. Br J Oral Maxillofac Surg 2009;47(1):5–9. [5] Fasunla AJ, Greene BH, Timmesfeld N, Wiegand S, Werner JA, Sesterhenn AM. A meta-analysis of the randomized controlled trials on elective neck dissection versus therapeutic neck dissection in oral cavity cancers with clinically nodenegative neck. Oral Oncol 2011;47(5):320–4. [6] Vandenbrouck C, Sancho-Garnier H, Chassagne D, Saravane D, Cachin Y, Micheau C. Elective versus therapeutic radical neck dissection in epidermoid carcinoma of the oral cavity: results of a randomized clinical trial. Cancer 1980;46(2):386–90. [7] Fakih AR, Rao RS, Borges AM, Patel AR. Elective versus therapeutic neck dissection in early carcinoma of the oral tongue. Am J Surg 1989;158(4): 309–13. [8] Yuen AP, Ho CM, Chow TL, Tang LC, Cheung WY, Ng RW, et al. Prospective randomized study of selective neck dissection versus observation for N0 neck of early tongue carcinoma. Head Neck 2009;31(6):765–72. [9] 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):360–5. [10] Khafif RA, Gelbfish GA, Tepper P, Attie JN. Elective radical neck dissection in epidermoid cancer of the head and neck. A retrospective analysis of 853 cases of mouth, pharynx, and larynx cancer. Cancer 1991;67(1):67–71. [11] Roh JL, Yoon YH, Kim SY, Park CI. Cervical sensory preservation during neck dissection. Oral Oncol 2007;43(5):491–8. [12] Speksnijder CM, van der Bilt A, Slappendel M, de Wijer A, Merkx MA, Koole R. Neck and shoulder function in patients treated for oral malignancies: a 1-year prospective cohort study. Head Neck 2013;35(9):1303–13. [13] Stoeckli SJ, Alkureishi LW, Ross GL. Sentinel node biopsy for early oral and oropharyngeal squamous cell carcinoma. Eur Arch Otorhinolaryngol 2009; 266(6):787–93. [14] Civantos FJ, Stoeckli SJ, Takes RP, Woolgar JA, de Bree R, Paleri V, et al. What is the role of sentinel lymph node biopsy in the management of oral cancer in 2010? Eur Arch Otorhinolaryngol 2010;267(6):839–44. [15] Rodrigo JP, Shah JP, Silver CE, Medina JE, Takes RP, Robbins KT, et al. Management of the clinically negative neck in early-stage head and neck cancers after transoral resection. Head Neck 2011;33(8):1210–9. [16] Rodjan F, de Bree R, Weijs J, Knol DL, Leemans CR, Castelijns JA. Refinement of selection criteria to perform ultrasound guided aspiration cytology during follow-up in patients with early staged oral cavity carcinoma and initially cN0 necks. Oral Oncol 2011;47(5):391–4. [17] Werning JW, Heard D, Pagano C, Khuder S. Elective management of the clinically negative neck by otolaryngologists in patients with oral tongue cancer. Arch Otolaryngol Head Neck Surg 2003;129(1):83–8. [18] Dunne AA, Folz BJ, Kuropkat C, Werner JA. Extent of surgical intervention in case of N0 neck in head and neck cancer patients: an analysis of data collection of 39 hospitals. Eur Arch Otorhinolaryngol 2004;261(6):295–303. [19] Mark Taylor S, Drover C, Maceachern R, Bullock M, Hart R, Psooy B, et al. Is preoperative ultrasonography accurate in measuring tumor thickness and predicting the incidence of cervical metastasis in oral cancer? Oral Oncol 2010;46(1):38–41.

[20] Borges AM, Shrikhande SS, Ganesh B. Surgical pathology of squamous carcinoma of the oral cavity: its impact on management. Semin Surg Oncol 1989;5(5):310–7. [21] D’Cruz AK, Siddachari RC, Walvekar RR, Pantvaidya GH, Chaukar DA, Deshpande MS, et al. Elective neck dissection for the management of the N0 neck in early cancer of the oral tongue: need for a randomized controlled trial. Head Neck 2009;31(5):618–24. [22] Fagan JJ, Collins B, Barnes L, D’Amico F, Myers EN, Johnson JT. Perineural invasion in squamous cell carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg 1998;124(6):637–40. [23] Rahima B, Shingaki S, Nagata M, Saito C. Prognostic significance of perineural invasion in oral and oropharyngeal carcinoma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004;97(4):423–31. [24] Michikawa C, Uzawa N, Kayamori K, Sonoda I, Ohyama Y, Okada N, et al. Clinical significance of lymphatic and blood vessel invasion in oral tongue squamous cell carcinomas. Oral Oncol 2012;48(4):320–4. [25] Jones HB, Sykes A, Bayman N, Sloan P, Swindell R, Patel M, et al. The impact of lymphovascular invasion on survival in oral carcinoma. Oral Oncol 2009; 45(1):10–5. [26] Liebig C, Ayala G, Wilks JA, Berger DH, Albo D. Perineural invasion in cancer: a review of the literature. Cancer 2009;115(15):3379–91. [27] Liebig C, Ayala G, Wilks J, et al. Perineural invasion is an independent predictor of outcome in colorectal cancer. J Clin Oncol 2009;27(31):5131–7. [28] Mohammed RA, Martin SG, Gill MS, Green AR, Paish EC, Ellis IO. Improved methods of detection of lymphovascular invasion demonstrate that it is the predominant method of vascular invasion in breast cancer and has important clinical consequences. Am J Surg Pathol 2007;31(12):1825–33. [29] Tai SK, Li WY, Chu PY, Chang SY, Tsai TL, Wang YF, et al. Risks and clinical implications of perineural invasion in T1-2 oral tongue squamous cell carcinoma. Head Neck 2012;34(7):994–1001. [30] Tai SK, Li WY, Yang MH, Chang SY, Chu PY, Tsai TL, et al. Treatment for T1-2 oral squamous cell carcinoma with or without perineural invasion: neck dissection and postoperative adjuvant therapy. Ann Surg Oncol 2012;19(6): 1995–2002. [31] Tai SK, Li WY, Yang MH, Chu PY, Wang YF. Perineural invasion in T1 oral squamous cell carcinoma indicates the need for aggressive elective neck dissection. Am J Surg Pathol 2013;37(8):1164–72. [32] Weiss MH, Harrison LB, Isaacs RS. Use of decision analysis in planning a management strategy for the stage N0 neck. Arch Otolaryngol Head Neck Surg 1994;120(7):699–702. [33] Okura M, Aikawa T, Sawai NY, Iida S, Kogo M. Decision analysis and treatment threshold in a management for the N0 neck of the oral cavity carcinoma. Oral Oncol 2009;45(10):908–11. [34] Kowalski LP, Sanabria A. Elective neck dissection in oral carcinoma: a critical review of the evidence. Acta Otorhinolaryngol Ital 2007;27(3):113–7. [35] Liu TR, Chen FJ, Yang AK, et al. Elective neck dissection in clinical stage I squamous cell carcinoma of the tongue: does it improve regional control or survival time? Oral Oncol 2011;47(2):136–41. [36] Sano D, Myers JN. Metastasis of squamous cell carcinoma of the oral tongue. Cancer Metast Rev 2007;26(3–4):645–62. [37] Kelner N, Rodrigues PC, Bufalino A, et al. Activin A immunoexpression as predictor of occult lymph node metastasis and overall survival in oral tongue squamous cell carcinoma. Head Neck 2014. http://dx.doi.org/10.1002/hed. 23627.