LYMPH NODE METASTASIS IN SQUAMOUS CELL CARCINOMA OF THE ORAL CAVITY: CORRELATION BETWEEN HISTOLOGIC FEATURES AND THE PREVALENCE OF METASTASIS Masahiro Umeda, DDS, PhD, Satoshi Yokoo, DDS, Yoshiaki Take, DDS, Akiteru Omori, DDS, Koichi Nakanishi, DDS, PhD, and Keikichi Shimada, DDS, PhD
A retrospective study was made of the correlation between preoperative clinical or histologic findings and the prevalence of lymph node metastasis in 60 patients with squamous cell carcinoma of the oral cavity who had histologically confirmed neck metastasis. Of these 60 patients, 39 with clinically N+ necks underwent immediate therapeutic neck dissection, and 21 whose necks were initially NO but progressed to N+ during observation underwent subsequent therapeutic neck dissection. The primary site, TNM staging, histologic grade of malignancy of biopsy specimen, and location and number of histologically positive lymph nodes were reviewed in each case. The results were as follows: (1) The prevalence of neck metastasis was not significantly correlated with primary site and T stage; however, there was an apparent correlation between histologic grade of malignancy and the prevalence of neck metastasis. Patients with grade I- II histologic malignancy showed limited metastases that involved lymph nodes in levels 1-11. On the other hand, patients showing grade Ill-IV histologic malignancy often had metastases that extended beyond level 111, regardless of T stage. These results suggest that histologic grade of malignancy, as well as clinical features, must be taken into considerationwhen deciding whether supraomohyoid neck dissection is indicated. (2) The group that underwent subsequent neck dissection ex-
From the Department of Maxillofacial Surgery, Kobe University School of Medicine, Kobe, Japan Address reprint requests to Dr. Umeda at the Department of Maxillofacia1 Surgery, Kobe University School of Medicine, 650 7-5-1 KusunokiCho. Chuo-Ku, Kobe, Japan. Accepted for publication January 12, 1992. CCC 0148-6403/92/040263-010$04.00 0 1992 John Wiley & Sons, Inc.
Neck Metastasis of Oral Squamous Cell Carcinoma
hibited less advanced neck metastasis and a better prognosis than the group which underwent immediate neck dissection. These findings show that if they are closely followed up, it is possible to delay neck dissection in NO patients until a neck metastasis is detected. 0 1992 John Wiley & Sons, Inc. HEAD & NECK 1992;14:263-272
Because squamous cell carcinoma of the oral cavity often metastasizes to the regional lymph nodes, improved prognosis depends upon control of the metastatic as well as the primary tumor. Since its first description by Crile' in 1906, classical radical neck dissection (RND) has become the routine procedure for surgical treatment of lymph node metastases in head and neck cancers. RND consists of the complete removal of lymph nodes at all neck levels, but it produces cosmetic and functional disadvantages. Therefore, an increasing number of modifications have recently been made to RND.2-5 Some have recommended supraomohyoid neck dissection in patients with squamous cell carcinoma of the oral cavity with NO or N1 necks, but its rationale and indications remain controversial. In this article, we describe the correlation between clinicopathologic features and patterns of lymph node metastasis in oral squamous cell carcinoma, and, further, we discuss indications for supraomohyoid neck dissection.
HEAD & NECK
July/August 1992
263
MATERIALS AND METHODS
Of 247 patients with primary squamous cell carcinoma of the oral mucosa treated at the Department of Maxillofacial Surgery, Kobe University School of Medicine between 1980 and 1990, 101 underwent radical neck dissections. A retrospective review of 60 of these patients was undertaken to ascertain preoperative clinical appearance, histologic features of biopsy specimens, and distribution of lymph node metastases at the neck level. The remaining 41 patients were excluded from this study because of histologically negative findings in the neck (20 patients), preoperative radiotherapy of more than 30 Gy in the neck region (14), or neck metastasis from a recurrent primary tumor (7). Of 60 operations, 39 were immediate therapeutic dissections in clinically N + necks (immediate neck dissection group), and 21 were subsequent therapeutic dissections in patients whose necks were initially NO but progressed to N+ during observation (subsequent neck dissection group). Elective neck dissections were not performed in our department. The primary site, TNM staging, histologic grade of malignancy, and anatomic location of histologically positive lymph nodes were reviewed in each case. TNM staging was performed retrospectively according to the UICC classification.6 Lymph node levels used to describe the clinical and histologic findings were: level I, submaxillary; level 11, upper jugular; level 111, midjugular; level IV, lower jugular; and level V, posterior cervical (Figure 1).When multiple levels were involved, the most inferior level was recorded. Histologic evaluations of biopsy specimens were performed without any knowledge of the clinical course. The histologic parameters studied were degree of differentiation and mode of invasion. Degree of differentiation was classified into three types (well, moderately, and poorly differentiated) according to WHO histologic grading7 (Fig-
Table 1. Histoloaic evaluation. ~~
_______
Tumor cell differentiation
Mode of invasion
Well
Moderately
Poorly
Type w Type M Type D
Grade I Grade II Grade 111
Grade I1 Grade 111 Grade 111
Grade 111 Grade 111 Grade IV
264
Neck Metastasis of Oral Squamous Cell Carcinoma
LEVEL IV
FIGURE 1. Cervical lymph node levels used for description.
ure 2). Mode of invasion was also classified into three types (W, M, and D) according to the pattern of tumor cell invasion at the tumor-host interface. Type W had a generally well-defined borderline, type M exhibited groups of tumor cells and had no distinct borderline, whereas type D spread in small aggregates with finger-like projections or invaded diffusely without forming nests of tumor cells (Figure 3). We furthermore defined the histologic grade of malignancy as grade I-IV, with regard to both differentiation and mode of invasion (Table 1).Grade I indicates the lowest grade of malignancy; tumors in this category consisted of welldifferentiated cells with a well-defined borderline. Grade IV indicates the highest grade of malignancy; these lesions consisted of poorly differentiated cells with a diffusely invading borderline. RESULTS
The subjects consisted of 35 men and 25 whose ages ranged from 31 - 83 years (average
Table 2. Distribution of patients by primary site. Immediate neck dissection group
Subsequent neck dissection group
Total
Tongue Lower gingiva Floor of mouth Upper gingiva Lip Buccal mucosa Oropharyngeal isthmus
12
11
23
12
3
15 12
Total
39
Primary site
a
4 2 1
HEAD & NECK
4 2
6 2 1
1
1
21
60
July/August 1992
Table 4. Distribution of patients by histologic grade of malignancy.
Table 3. Distribution of patients by T stage. T stage T1 T2 T3 T4 Total
Immediate neck dissection group
Subsequent neck dissection group
Total
6
11
17
14 5 14
9 1
23 5
39
21
60
15
60 years). The distribution of primary sites is shown in Table 2. The oral tongue, gingiva, and floor of the mouth accounted for 93% of the primary sites. The T stage for each clinical group is shown in Table 3. Patients in the subsequent neck dissection group exhibited primary tumors in a less advanced stage than those in the immediate neck dissection group. Table 4 shows the distribution of histologic grades of malignancy. Of the 52 patients whose histologic grade was observable, 14 (27%) were
Histologic grade of malignancy
Immediate neck dissection group
Subsequent neck dissection group
Total
Grade I Grade II Grade 111 Grade IV Unknown
9 4 12 8 6
5 7 7 2
14 11 19 8 8
Total
39
21
60
classified as grade I, 11 (21%) as grade 11, 19 (37%) as grade 111, and eight (15%)as grade IV. The remaining eight patients could not be classified because of inadequate biopsy specimens (eg, the margin of the tumor was not located in the specimen). All patients classified as grade IV were in the immediate neck dissection group (although the distribution of grades I, 11, and I11 did not differ in the immediate and subsequent neck dissection groups). This finding indicates that patients exhibiting a high histologic grade of malignancy are more likely to show neck metastasis at the first visit. The relationship of primary sites and T stage to histologic grade of malignancy was also examined, but there was no correlation between these parameters, indicating that histologic grade is a parameter that is independent of primary sites and T stage. Histologically confirmed metastases were found at level I in 28 patients, at level I1 in 24, at level I11 in four, at level IV in two, and at level V in two (Table 5). Level 111-V metastases
FIGURE 2. Degree of differentiation. (A) well differentiated; (B) moderately differentiated; and (C) poorly differentiated.
Neck Metastasis of Oral Squamous Cell Carcinoma
HEAD & NECK
July/August 1992
265
Neck Level
Table 5. Prevalence of neck metastasis. Metastasis level Level I Level II Level 111 Level IV Level V Total
Immediate neck dissection group
Subsequent neck dissection group
20
8 13
11 4 2 2 39
level V
I 0
level lU
I
level I
?.
X
OOOOA
OOAA
A X X ?
A X X
A A X X
OOOOA
XX?
X?
OOA . .x
X
0 0 x 0
.??
A A X X X
X ? ?
I T1
were relatively rare, and these were all found in the immediate neck dissection group. The relation of primary site to metastasis level is shown in Table 6. Carcinoma of the lower gingiva and floor of the mouth often metastasized to level I, and carcinoma of the tongue to level 11. However, no correlation was found between primary site and neck metastasis beyond level 111. Level 111-V involvement was found in three patients with carcinoma of the tongue, two
0 x.
level N
level U
21
0
T2
T3
T4 T- Stage
0 : g r a d e I . A : g r a d e I I . X:gradem, 0 : gradem. ? : unknown grade
FIGURE 4. The correlation between histologic grade of malignancy and metastasis level.
with carcinoma of the lower gingiva, two with carcinoma of the floor of the mouth, and one with carcinoma of the upper gingiva. Table 7 shows T stage and metastasis level. There was a tendency for advances in neck disease to correspond with progression of the primary disease. However, level 111- V involvement was found in four T1-2 and four T3-4 patients. These findings show that primary site and T stage are not factors that will predict the prevalence of metastasis, especially in a level 111-V neck. Figure 4 shows the histologic grade of malignancy and metastasis level. A distinct correlation was found between these parameters. In grade 1-11 patients, nodal involvement was limited to level 1-11, regardless of T stage. On the other hand, grade 111-IV patients often exhibited neck metastasis in level 111-V regions, even if their primary lesions had no advanced locally.
FIGURE 3. Mode of invasion. (A) type W; (B) type M; and (C) type D.
266
Neck Metastasis of Oral Squamous Cell Carcinoma
HEAD & NECK
JulyiAugust 1992
hibited a histologic level that was more advanced by two to four levels than the clinical level. In other words, a tumor associated with histologically high malignancy was likely to show extensive neck metastases even if clinical examination revealed only localized metastasis. The average number of metastatic lymph nodes was 2.2 in the immediate neck dissection group and 1.7 in the subsequent neck dissection group (Table 9). This shows that the detection and treatment of secondary neck metastasis at our department were effective in the subsequent neck dissection group. The number of metastatic nodes was also related to the histologic grade of
In 45 of the 60 patients, the neck level of histologic metastasis coincided with levels detectable on preoperative clinical examination by palpation, computed tomography (CT), or magnetic resonance imaging (MRI). However, 15 patients exhibited more extensive neck disease than that which was diagnosed preoperatively . Table 8 shows the relationship between preoperative and postoperative findings and histologic grade of malignancy. In grade 1-11 patients, the histologic metastasis level was either equal to the clinical metastasis level or the former was more advanced than the latter by only one level. On the other hand, grade 111-IV patients often ex-
Table 6. Prevalence of metastasis by primary site. Metastasis level Level I
Level II
Level 111
Level IV
Level V
Total
Tongue Lower gingiva Floor of mouth Upper gingiva Lip Buccal mucosa Oropharyngeal isthmus
9 9 6 2 2 __
11 4 4 3
2 1 1 -
1 1 -
23 15 12 6 2
-
-
-
1 1
1 1 -
-
-
-
1
Total
28
24
4
2
2
Primary site
-
1 60
Table 7. Prevalence of metastasis by T stage. Metastasis level
T stage
Level I
Level II
Level 111
Level IV
Level V
T3 T4
7 13 1 7
9 7 2 6
1 2 -
-
1
2 -
1 1
Total 17 23 5 15
Total
28
24
4
2
2
60
~~
T1
T2
~
Table 8. Difference between clinically detectable node level and histologically confirmed node level. Difference between histologic and clinical level Histologic level equal to clinical level Histologic level more advanced than clinical level by 1 level 2 levels 3 levels 4 levels
Neck Metastasis of Oral Squarnous Cell Carcinoma
Histologic grade
of malignancy
Grade I
Grade II
Grade 111
Grade IV
Grade V
Total
10
10
15
3
7
45
4
1
-
-
2 1 1 -
1 2 1 1
1 -
8 4 2 1
-
HEAD & NECK
July/August 1992
267
Table 9. Number of metastatic lymph nodes. ~
~
Number of metastatic nodes
Subsequent neck dissection group (patients)
21 9 2 7
1 2 3
>4 Average number of metastatic nodes
subsequent
~
Immediate neck dissection group (patients)
13 5 0 3
40
1.7
2.2
2o
-
1
O '
malignancy. In grade 1-11 patients, an average of 1.4 nodes was found and in grade 111-IV patients, an average 2.9 nodes were counted (Table 10). The 5-year cumulative survival rate was 55% in the immediate neck dissection group and 73% in the subsequent neck dissection group (Figure 5). Neck failure without recurrence of the primary lesion was seen in three patients in the former and in two patients in the latter group (Table 11). DISCUSSION
The incidence of neck metastasis in oral squamous cell carcinoma, as cited in the literature, is relatively high, at 34% to 50%.8,9 As stated above, radical neck dissection (RND) is a routine surgical procedure performed on all N+ patients with oral cancer and sometimes performed in NO patients as a prophylactic therapy. However, because of its cosmetic and functional disadvantages, some authors have proposed various modifications of RND. Shah et a1.l' reviewed the prevalence of neck metastases in 501 patients with oral squamous cell carcinoma who underwent RNDs. These investigators found that levels I, 11, and I11 were at greatest risk of nodal metastasis, whereas level
Table 10. Average number of metastatic nodes by histologic grade of malignancy. Histologic grade of malignancy Grades I, II
Grades 111, IV
Total
T1 ,T2 T3,T4
1.3 1.6
2.7 3.1
1.9 2.6
Total
1.4
2.9
2.1
T stage
268
Neck Metastasis of Oral Squamous Cell Carcinoma
immediate RND group
1'
2
3
4
5 year
FIGURE 5. The cumulative survival rate.
IV was rarely involved in the NO patients but was more commonly involved in the N+ groups, and that level V was rarely involved in either NO or N+ groups. They recommended that supraomohyoid neck dissection for NO, and anterolateral neck dissection for N+ necks, would remove the majority of nodal metastases. However, the histologic findings of the primary lesions were not taken into consideration in their study, nor were the figures of neck failure and survival rate mentioned. Byers et al.4 reviewed 428 patients with head and neck cancers who underwent elective modified neck dissections; these investigators recommended supraomohyoid neck dissection for oral squamous cell carcinoma patients with an NO neck. Medina et al.5 also advocated supraomohyoid neck dissection in T2-T4NO and selected N1 squamous cell carcinomas of the oral cavity, but they did not supply any detailed data to support their conclusion. Thus, supraomohyoid neck dissection has been reported to be indicated mainly as an elective therapeutic modality for NO patients. The question of whether or not elective neck therapy is necessary for clinical NO patients remains controversial today. Many investigators have recommended elective neck dissection because of the high incidence of occult metastasis in patients with clinically NO necks and the poor survival rate of patients who undergo subsequent neck dissection^.^*^,^^ A review of the literature reveals that the incidence of secondary neck metastasis after initial therapy in NO patients with oral squamous cell carcinoma varies from 15% to 46%.11-13 However it has been reported by Friedman et al.14 that CT and MRI are
HEAD & NECK
July/August 1992
Table 11. Patients of neck failure Case
Primary site
TNM
Histologic grade of malignancy
Metastasis level
Number of metastatic nodes
Tongue Lower gingiva Upper gingiva
210 210 42b0
Grade I Grade IV Grade IV
Level I Level V Level V
1 6 7
Floor of mouth Upper gingiva
100 200
Grade 111 Grade II
Level II Level II
4
Immediate neck dissection group
Subsequent neck dissection group
associated with a higher sensitivity for detecting metastatic disease than physical exmaination, and that patients with clinically normal necks and negative CT or MRI had only a 12% probability of having occult metastatic disease. These investigators advocated that the concept of elective neck dissection should be re-examined in light of the improved ability to detect nodal disease. Vandenbrouck et al.15 reported the first randomized clinical trial for patients with T1-TSNO squamous cell carcinomas of the oral cavity. They divided their 75 patients into two groups: 39 who underwent elective neck dissection, and 36 who underwent therapeutic neck dissection when neck disease was discovered. In the second group, neck disease developed in 19 patients during follow-up, but the survival rate in the two groups did not differ. Fakih et a1.16 also conducted a randomized study of prophylactic vs therapeutic neck dissection in 35 patients with TlT2NO squamous cell carcinoma of the oral tongue; they concluded that the disease-free survival rate was higher in the prophylactic neck dissection group. The different results in these two prospective randomized trials seem to be due to an unsuccessful follow-up in the latter trial. Only 28% of the patients who developed neck disease during observation were salvaged in Fakih's study. This is thought t o be due to the difficulty of carrying out close follow-up in an indigent population in the vast rural expanse of India. In contrast, close follow-up is possible here in Japan. We make it a rule to observe oral cancer patients every week during the first 6 months after their initial treatment and every 2 weeks during the next 18 months, and we attempt to detect and treat neck diseases, as well as local recurrences, in their early stage. Furthermore,
Neck Metastasis of Oral Squamous Cell Carcinoma
2
adjuvant immunotherapy, with OK-432, is also given. 0K-432 is a biological response modifier consisting of a lyophilized powder made from cultures of a penicillin-treated, low-virulence Sustrain human Streptococcus pyogenes (Picibanil, Chugai Pharmaceutical Co. Ltd., Japan). This agent is injected intracutaneously once every 2 weeks in all oral cancer patients for at least 2 years. In this study, the average number of metastatic nodes was 1.7 in the subsequent RND group, fewer than the 2.2 in the immediate RND group. The cure rate for neck disease in our department was thus almost equal in the subsequent and immediate therapeutic neck dissection groups, and the survival rate was somewhat better in the former group. These findings seem to lead to the conclusion that, if there is a close follow-up, it is possible, without risk, to delay neck dissection until neck disease is detectable. There have been many reports indicating correlation between the incidence of neck metastasis and various clinical or histologic features. Several investigator^'^ have reported that the incidence of nodal involvement was correlated with the T stage, however, others's-20 found no correlation between these parameters. It has been reported that tumor thickness or depth of invasion is an important factor for predicting neck metastasis. Spiro et a1.20found that in 105 patients with oral and oropharyngeal carcinoma with NO necks, the risk of occult neck metastasis approached 40% when the tumor thickness was >2.0 mm. Mohit-Tabatabai et a1.l' documented a significant correlation between a tumor thickness greater than 1.5 mm and subsequent development of neck metastases in a series of 84 patients with stage I and I1 carcinomas of the floor of the mouth. These findings were confirmed by Frierson and Cooper2' in 186
HEAD & NECK
July/August 1992
269
patients with carcinoma of the lower lip, and by Rasgon et al.13 in 22 patients with oral and oropharyngeal carcinoma. Thus, it has been shown in some retrospective studies that primary tumor thickness is one of the parameters related to nodal metastasis. However, we believe tumor thickness is clinically not very useful as a preoperative parameter for predicting nodal involvement, since it is difficult to measure precisely by clinical examination or in a biopsy specimen. The degree of differentiation of the tumor has been found to be correlated with nodal metastasis. Willen et a1.,22McGavran et al.,23Mendelson et al.,24 and Frierson and Cooper2’ all demonstrated a higher incidence of neck involvement in poorly differentiated than in well-differentiated carcinomas. The pattern of invasion at the tumor-host interface has also been reported to be an important histologic parameter in neck metastasis and survival in patients with head and neck cancer. Willen et a1.22found, in their observations of 124 patients with squamous cell carcinoma of the gingiva, that those patients who showed diffuse invasion experienced a higher incidence of neck metastasis and poorer prognosis than those who showed well-defined borderlines. These findings were also confirmed by Yamamoto et al.,25Crissman et a1.,26and Frierson and Cooper.21 Several other histologic parameters have been reported to be related to neck metastasis. Close et al.,27Crissman et a1.,26and Poleksic and Kalwaic28 found that vascular invasion at the primary site correlated with nodal metastasis in head and neck cancer. On the other hand, McGavran et al.23and Frierson and Cooper21found no significant correlation between vascular invasion and neck metastasis. Although several investigators21.22have shown a strong correlation between perineural invasion at the primary site and neck nodal metastasis, other^^^,^' have not. Inflammatory infiltration is another histologic parameter reported to correlate with nodal metastasis. The incidence of neck metastasis was lower in those patients with a marked inflammatory response than in those with a minimal inflammatory response, according to studies by Mohit-Tabatabai et a1.,18 Frierson and Cooper,21 and Rasgon et al.I3 Hiratsuka et al.,30 using immunohistologic analysis, showed that the T cell infiltration was correlated with a decrease of neck metastasis in squamous cell carcinoma of the oral cavity. However, several other studies
270
Neck Metastasis of Oral Squamous Cell Carcinoma
have found no correlation between inflammatory response and neck m e t a s t a ~ i s . ~We ~ ’ ~used ~’~~ the extent of differentiation and mode of invasion as histologic parameters in this study because they are generally accepted to represent histologic malignancy and they are easily assessed in standard biopsy specimens. Many investigators have studied various histologic parameters in patients with squamous cell carcinoma of the oral cavity, and some of these parameters have been reported to be related to the incidence of nodal metastasis as mentioned above. However, correlations between histologic findings in the primary site and the neck level of nodal metastasis have not been documented up to now. The prediction of the extent of neck metastasis would be of great benefit to the surgeon. We believe that our retrospective study is unique in that it emphasizes the relationship of histologic findings both to the neck level of the metastasis and to the difference between clinically detectable and histologically confirmed neck level. Our study revealed a close relationship between histologic features at the primary site and the prevalence of metastasis, especially in level 111-V necks. Histologically low malignant tumors metastasized only to level 1-11, regardless of T classification, although histologically high malignant tumors often metastasized to level 111-V, even if they had not advanced locally. Our study also indicated that the histologically confirmed neck level of metastasis almost coincided with the preoperatively detectable level in patients showing low malignancy, but that the former was often much more advanced than the latter in patients showing high malignancy. Neck metastasis usually occurs when local lesions increase in size, but in patients with histologically high malignant tumors at the primary site, this metastasis may occur before local disease becomes aggressive, and extend to the whole neck during the early stage. The results of our study show the rationale of and indications for supraomohyoid neck dissection in patients with squamous cell carcinoma of the oral cavity. Some authors have reported that this procedure is indicated in NO (or N1) oral squamous cell carcinoma, as mentioned above; however, such reports were based on clinical features alone. If the histologic findings are also taken into consideration, indications will become more precise. Although it is difficult to draw any valid conclusions here, since a small population
HEAD & NECK
July/August 1992
of patients was studied, we would advocate the following: (1) Supraomohyoid neck dissection should be limited to N1 patients whose histologic grade of malignancy is grade I (well-differentiated squamous cell carcinoma with a well-defined borderline); (2) N1 patients of grade 11-IV and N2-3 patients should undergo RND; and (3) elective neck dissection is not necessary in NO patients if a close follow-up is conducted. A large multi-institutional prospective study is expected in the future. CONCLUSION
A retrospective study of patients with squamous cell carcinoma of the oral cavity was carried out, with special reference to the location of neck metastases. An apparent correlation was found between histologic grade of malignancy and the prevalence of neck metastases. Patients with grade I - I1 histologic malignancy exhibited limited neck metastases in the level 1-11 region. On the other hand, grade 111-IV patients often had neck metastases that extended beyond the level I11 region, regardless of their T stage. These results suggest that histologic grade of malignancy, as well as clinical features, should be considered when selective limited neck dissection is performed.
REFERENCES
1. Crile GW. Excision of cancer of the head and neck. JAMA 1906;47:1780-1786. 2. Byers RM. Modified neck dissection-a study of 967 cases from 1970 to 1980. A m J Surg 1985;150:414-421. 3. Bocca E, Pignataro 0, Oldine C, Cappa C. Functional neck dissection: an evaluation of review of 843 cases. Laryngoscope 1984;94:942- 945. 4. Byers RM, Wolf PF, Ballantyne AJ. Rationale for elective modified neck dissection. Head Neck Surg 1988;lO: 160-167. 5. Medina JE, Byers RM. Supraomohyoid neck dissection: rationale, indications, and surgical technique. Head Neck Surg 1989;11:111-122. 6. UICC. TNM Ktassifikation matigner Tumoren. Vierte Auflage. Berlin: Springer-Verlag, 1987. 7. WHO. Histological typing of oral and oropharyngeal tumors. International Histological Classification of Tumours no. 4. Geneva, 1971:17. 8. Lee JG, Krause CJ. Radical neck dissection: elective, t,herapeutic and secondary. Arch Otolayngol 1975;lOl: 656-659. 9. Martis C, Karabouta I, Lazaridis N. Incidence of lymph node metastasis in elective (prophylactic) neck dissection for oral carcinoma. J Maxillofuc Surg 1979;7:182-191. 10. Shah JP, Candela FC, Poddar AK. The patterns of cervical lymph node metastases from squamous carcinoma of the oral cavity. Cuncer 1990;66:109-113.
Neck Metastasis of Oral Squamous Cell Carcinoma
11. Cunningham MJ, Johnson JT, Myers EN, Schramm VL, Thearle PB. Cervical lymph node metastasis after local excision of early squamous cell carcinoma of the oral cavity. A m J Surg 1986;152:361-366. 12. Teichgraeber JF, Clairmont AA. The incidence of occult metastasis for cancer of the oral tongue and floor of the mouth: treatment rationale. Head Neck Surg 1984;7:1521. 13. Rasgon Bm, Cruz RM, Hilsinger RL, Sawicki JE. Relation of lymph-node metastasis to histopathologic appearance in oral cavity and oropharyngeal carcinoma: a case series and literature review. Laryngoscope 1989;99: 1103- 1110. 14. Friedman M, Mafee MF, Pacella BL Jr, Strorigl TL, Dew LL, Toriumi DM. Rationale for elective neck dissection in 1990. Laryngoscope 1990;100:54-59. 15. Vandenbrouck C, Sancho-Garnier H, Chassagne D, Saravne 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;46:386-390. 16. Fakih AR, Rao RS, Pate1 AR. Prophylactic neck dissection in squamous cell carcinoma of oral tongue: a prospective randomized study. Semin Surg Oncol 1989;5: 327-330. 17. Bradfield JS, Scruggs RP. Carcinoma of the mobile tongue: incidence of cervical metastases in early lesions related to method of primary treatment. Laryngoscope 1983;93:1332- 1336. 18. Lee JG, Litton WB. Symposium on malignancy, 11. Occult regional metastasis: carcinoma of the oral tongue. Laryngoscope 1972;82:1273- 1281. 19. Mohit-Tabatabai MA, Sobel HJ, Rush BF, Mashberg A. Relation of thickness of floor of mouth stage I and I1 cancers to regional metastasis. A m J Surg 1986;152:351353. 20. Spiro RH, Huvos AG, Wong GY, Spiro JD, Gnecco CA, Strong EW. Predictive value of tumor thickness in squamous cell carcinoma confined t o the tongue and floor of the mouth. A m J Surg 1986;152:345-350. 21. Frierson HF Jr, Cooper PH. Prognostic factors in squamous cell carcinoma of the lower lip. Hum Pathol 1986;17:346-354. 22. Willen R, Nathanson A, Moberger C, Anneroth G. Squamous cell carcinoma of the gingiva: histological classification and grading of malignancy. Acta Otolaryngo1 1975;79:146- 154. 31. McGavran MH, Bauer WC, Ogura JH. The incidence of cervical lymph node metastases from epidermoid carcinoma of the larynx and their relationship to certain characteristics of the primary tumor: a study based on the clinical and pathological findings for 96 patients treated by primary en bloc laryngectomy and radical neck dissection. Cancer 1961;14:55-65. 32. Mendelson BC, Woods JE, Beahrs OH. Neck dissection in the treatment of carcinoma of the anterior two-third of the tongue. Surg Gynecol Obstet 1976;143:75-80. 33. Yamamoto E, Miyakawa A, Kohama G. Mode of invasion and lymph node metastasis in squamous cell carcinoma of the oral cavity. Head Neck Surg 1984;6:938947. 34. Crissman JD, Liu WY, Gluckman JL, Cummings G. Prognostic value of histopathologic parameters in squamous cell carcinoma of the orpharynx. Cancer 1984;54:2995- 3001. 35. Close LG, Burns DK, Reisch J, Schaefer SD. Microvascular invasion in cancer of the oral cavity and oropharynx. Arch Otolaryngol Head Neck Surg 1987;113:1191- 1195. 36. Poleksic S, Kalwaic HJ. Prognostic value of vascular invasion in squamous cell carcinoma of the head and neck. Plast Reconstr Surg 1978;61:234-240. 37. So0 K-C, Carter RC, O'Brien CJ, Barr L, Bliss JM, Shaw
HEAD 8. NECK
July/August 1992
271
HJ. Prognostic implications of perineural spread in squamous cell carcinoma of the head and neck. Laryngoscope 1986;96:1145- 1148. 38. Hiratsuka H, Imamura M, Ishii Y, Kohama G, Kikuchi
272
Neck Metastasis of Oral Squarnous Cell Carcinoma
K. Immunohistologic detection of lymphocyte subpopulations infiltrating in human oral cancer with special reference to its clinical significance. Cancer 1984;53:24562466.
HEAD 8 NECK
JulyiAugust 1992
A retrospective study was made of the correlation between preoperative clinical or histologic findings and the prevalence of lymph node metastasis in 60 patients with squamous cell carcinoma of the oral cavity who had histologically confirmed neck metastasis. Of these 60 patients, 39 with clinically N+ necks underwent immediate therapeutic neck dissection, and 21 whose necks were initially NO but progressed to N+ during observation underwent subsequent therapeutic neck dissection. The primary site, TNM staging, histologic grade of malignancy of biopsy specimen, and location and number of histologically positive lymph nodes were reviewed in each case. The results were as follows: (1) The prevalence of neck metastasis was not significantly correlated with primary site and T stage; however, there was an apparent correlation between histologic grade of malignancy and the prevalence of neck metastasis. Patients with grade I- II histologic malignancy showed limited metastases that involved lymph nodes in levels 1-11. On the other hand, patients showing grade Ill-IV histologic malignancy often had metastases that extended beyond level 111, regardless of T stage. These results suggest that histologic grade of malignancy, as well as clinical features, must be taken into considerationwhen deciding whether supraomohyoid neck dissection is indicated. (2) The group that underwent subsequent neck dissection ex-
From the Department of Maxillofacial Surgery, Kobe University School of Medicine, Kobe, Japan Address reprint requests to Dr. Umeda at the Department of Maxillofacia1 Surgery, Kobe University School of Medicine, 650 7-5-1 KusunokiCho. Chuo-Ku, Kobe, Japan. Accepted for publication January 12, 1992. CCC 0148-6403/92/040263-010$04.00 0 1992 John Wiley & Sons, Inc.
Neck Metastasis of Oral Squamous Cell Carcinoma
hibited less advanced neck metastasis and a better prognosis than the group which underwent immediate neck dissection. These findings show that if they are closely followed up, it is possible to delay neck dissection in NO patients until a neck metastasis is detected. 0 1992 John Wiley & Sons, Inc. HEAD & NECK 1992;14:263-272
Because squamous cell carcinoma of the oral cavity often metastasizes to the regional lymph nodes, improved prognosis depends upon control of the metastatic as well as the primary tumor. Since its first description by Crile' in 1906, classical radical neck dissection (RND) has become the routine procedure for surgical treatment of lymph node metastases in head and neck cancers. RND consists of the complete removal of lymph nodes at all neck levels, but it produces cosmetic and functional disadvantages. Therefore, an increasing number of modifications have recently been made to RND.2-5 Some have recommended supraomohyoid neck dissection in patients with squamous cell carcinoma of the oral cavity with NO or N1 necks, but its rationale and indications remain controversial. In this article, we describe the correlation between clinicopathologic features and patterns of lymph node metastasis in oral squamous cell carcinoma, and, further, we discuss indications for supraomohyoid neck dissection.
HEAD & NECK
July/August 1992
263
MATERIALS AND METHODS
Of 247 patients with primary squamous cell carcinoma of the oral mucosa treated at the Department of Maxillofacial Surgery, Kobe University School of Medicine between 1980 and 1990, 101 underwent radical neck dissections. A retrospective review of 60 of these patients was undertaken to ascertain preoperative clinical appearance, histologic features of biopsy specimens, and distribution of lymph node metastases at the neck level. The remaining 41 patients were excluded from this study because of histologically negative findings in the neck (20 patients), preoperative radiotherapy of more than 30 Gy in the neck region (14), or neck metastasis from a recurrent primary tumor (7). Of 60 operations, 39 were immediate therapeutic dissections in clinically N + necks (immediate neck dissection group), and 21 were subsequent therapeutic dissections in patients whose necks were initially NO but progressed to N+ during observation (subsequent neck dissection group). Elective neck dissections were not performed in our department. The primary site, TNM staging, histologic grade of malignancy, and anatomic location of histologically positive lymph nodes were reviewed in each case. TNM staging was performed retrospectively according to the UICC classification.6 Lymph node levels used to describe the clinical and histologic findings were: level I, submaxillary; level 11, upper jugular; level 111, midjugular; level IV, lower jugular; and level V, posterior cervical (Figure 1).When multiple levels were involved, the most inferior level was recorded. Histologic evaluations of biopsy specimens were performed without any knowledge of the clinical course. The histologic parameters studied were degree of differentiation and mode of invasion. Degree of differentiation was classified into three types (well, moderately, and poorly differentiated) according to WHO histologic grading7 (Fig-
Table 1. Histoloaic evaluation. ~~
_______
Tumor cell differentiation
Mode of invasion
Well
Moderately
Poorly
Type w Type M Type D
Grade I Grade II Grade 111
Grade I1 Grade 111 Grade 111
Grade 111 Grade 111 Grade IV
264
Neck Metastasis of Oral Squamous Cell Carcinoma
LEVEL IV
FIGURE 1. Cervical lymph node levels used for description.
ure 2). Mode of invasion was also classified into three types (W, M, and D) according to the pattern of tumor cell invasion at the tumor-host interface. Type W had a generally well-defined borderline, type M exhibited groups of tumor cells and had no distinct borderline, whereas type D spread in small aggregates with finger-like projections or invaded diffusely without forming nests of tumor cells (Figure 3). We furthermore defined the histologic grade of malignancy as grade I-IV, with regard to both differentiation and mode of invasion (Table 1).Grade I indicates the lowest grade of malignancy; tumors in this category consisted of welldifferentiated cells with a well-defined borderline. Grade IV indicates the highest grade of malignancy; these lesions consisted of poorly differentiated cells with a diffusely invading borderline. RESULTS
The subjects consisted of 35 men and 25 whose ages ranged from 31 - 83 years (average
Table 2. Distribution of patients by primary site. Immediate neck dissection group
Subsequent neck dissection group
Total
Tongue Lower gingiva Floor of mouth Upper gingiva Lip Buccal mucosa Oropharyngeal isthmus
12
11
23
12
3
15 12
Total
39
Primary site
a
4 2 1
HEAD & NECK
4 2
6 2 1
1
1
21
60
July/August 1992
Table 4. Distribution of patients by histologic grade of malignancy.
Table 3. Distribution of patients by T stage. T stage T1 T2 T3 T4 Total
Immediate neck dissection group
Subsequent neck dissection group
Total
6
11
17
14 5 14
9 1
23 5
39
21
60
15
60 years). The distribution of primary sites is shown in Table 2. The oral tongue, gingiva, and floor of the mouth accounted for 93% of the primary sites. The T stage for each clinical group is shown in Table 3. Patients in the subsequent neck dissection group exhibited primary tumors in a less advanced stage than those in the immediate neck dissection group. Table 4 shows the distribution of histologic grades of malignancy. Of the 52 patients whose histologic grade was observable, 14 (27%) were
Histologic grade of malignancy
Immediate neck dissection group
Subsequent neck dissection group
Total
Grade I Grade II Grade 111 Grade IV Unknown
9 4 12 8 6
5 7 7 2
14 11 19 8 8
Total
39
21
60
classified as grade I, 11 (21%) as grade 11, 19 (37%) as grade 111, and eight (15%)as grade IV. The remaining eight patients could not be classified because of inadequate biopsy specimens (eg, the margin of the tumor was not located in the specimen). All patients classified as grade IV were in the immediate neck dissection group (although the distribution of grades I, 11, and I11 did not differ in the immediate and subsequent neck dissection groups). This finding indicates that patients exhibiting a high histologic grade of malignancy are more likely to show neck metastasis at the first visit. The relationship of primary sites and T stage to histologic grade of malignancy was also examined, but there was no correlation between these parameters, indicating that histologic grade is a parameter that is independent of primary sites and T stage. Histologically confirmed metastases were found at level I in 28 patients, at level I1 in 24, at level I11 in four, at level IV in two, and at level V in two (Table 5). Level 111-V metastases
FIGURE 2. Degree of differentiation. (A) well differentiated; (B) moderately differentiated; and (C) poorly differentiated.
Neck Metastasis of Oral Squamous Cell Carcinoma
HEAD & NECK
July/August 1992
265
Neck Level
Table 5. Prevalence of neck metastasis. Metastasis level Level I Level II Level 111 Level IV Level V Total
Immediate neck dissection group
Subsequent neck dissection group
20
8 13
11 4 2 2 39
level V
I 0
level lU
I
level I
?.
X
OOOOA
OOAA
A X X ?
A X X
A A X X
OOOOA
XX?
X?
OOA . .x
X
0 0 x 0
.??
A A X X X
X ? ?
I T1
were relatively rare, and these were all found in the immediate neck dissection group. The relation of primary site to metastasis level is shown in Table 6. Carcinoma of the lower gingiva and floor of the mouth often metastasized to level I, and carcinoma of the tongue to level 11. However, no correlation was found between primary site and neck metastasis beyond level 111. Level 111-V involvement was found in three patients with carcinoma of the tongue, two
0 x.
level N
level U
21
0
T2
T3
T4 T- Stage
0 : g r a d e I . A : g r a d e I I . X:gradem, 0 : gradem. ? : unknown grade
FIGURE 4. The correlation between histologic grade of malignancy and metastasis level.
with carcinoma of the lower gingiva, two with carcinoma of the floor of the mouth, and one with carcinoma of the upper gingiva. Table 7 shows T stage and metastasis level. There was a tendency for advances in neck disease to correspond with progression of the primary disease. However, level 111- V involvement was found in four T1-2 and four T3-4 patients. These findings show that primary site and T stage are not factors that will predict the prevalence of metastasis, especially in a level 111-V neck. Figure 4 shows the histologic grade of malignancy and metastasis level. A distinct correlation was found between these parameters. In grade 1-11 patients, nodal involvement was limited to level 1-11, regardless of T stage. On the other hand, grade 111-IV patients often exhibited neck metastasis in level 111-V regions, even if their primary lesions had no advanced locally.
FIGURE 3. Mode of invasion. (A) type W; (B) type M; and (C) type D.
266
Neck Metastasis of Oral Squamous Cell Carcinoma
HEAD & NECK
JulyiAugust 1992
hibited a histologic level that was more advanced by two to four levels than the clinical level. In other words, a tumor associated with histologically high malignancy was likely to show extensive neck metastases even if clinical examination revealed only localized metastasis. The average number of metastatic lymph nodes was 2.2 in the immediate neck dissection group and 1.7 in the subsequent neck dissection group (Table 9). This shows that the detection and treatment of secondary neck metastasis at our department were effective in the subsequent neck dissection group. The number of metastatic nodes was also related to the histologic grade of
In 45 of the 60 patients, the neck level of histologic metastasis coincided with levels detectable on preoperative clinical examination by palpation, computed tomography (CT), or magnetic resonance imaging (MRI). However, 15 patients exhibited more extensive neck disease than that which was diagnosed preoperatively . Table 8 shows the relationship between preoperative and postoperative findings and histologic grade of malignancy. In grade 1-11 patients, the histologic metastasis level was either equal to the clinical metastasis level or the former was more advanced than the latter by only one level. On the other hand, grade 111-IV patients often ex-
Table 6. Prevalence of metastasis by primary site. Metastasis level Level I
Level II
Level 111
Level IV
Level V
Total
Tongue Lower gingiva Floor of mouth Upper gingiva Lip Buccal mucosa Oropharyngeal isthmus
9 9 6 2 2 __
11 4 4 3
2 1 1 -
1 1 -
23 15 12 6 2
-
-
-
1 1
1 1 -
-
-
-
1
Total
28
24
4
2
2
Primary site
-
1 60
Table 7. Prevalence of metastasis by T stage. Metastasis level
T stage
Level I
Level II
Level 111
Level IV
Level V
T3 T4
7 13 1 7
9 7 2 6
1 2 -
-
1
2 -
1 1
Total 17 23 5 15
Total
28
24
4
2
2
60
~~
T1
T2
~
Table 8. Difference between clinically detectable node level and histologically confirmed node level. Difference between histologic and clinical level Histologic level equal to clinical level Histologic level more advanced than clinical level by 1 level 2 levels 3 levels 4 levels
Neck Metastasis of Oral Squarnous Cell Carcinoma
Histologic grade
of malignancy
Grade I
Grade II
Grade 111
Grade IV
Grade V
Total
10
10
15
3
7
45
4
1
-
-
2 1 1 -
1 2 1 1
1 -
8 4 2 1
-
HEAD & NECK
July/August 1992
267
Table 9. Number of metastatic lymph nodes. ~
~
Number of metastatic nodes
Subsequent neck dissection group (patients)
21 9 2 7
1 2 3
>4 Average number of metastatic nodes
subsequent
~
Immediate neck dissection group (patients)
13 5 0 3
40
1.7
2.2
2o
-
1
O '
malignancy. In grade 1-11 patients, an average of 1.4 nodes was found and in grade 111-IV patients, an average 2.9 nodes were counted (Table 10). The 5-year cumulative survival rate was 55% in the immediate neck dissection group and 73% in the subsequent neck dissection group (Figure 5). Neck failure without recurrence of the primary lesion was seen in three patients in the former and in two patients in the latter group (Table 11). DISCUSSION
The incidence of neck metastasis in oral squamous cell carcinoma, as cited in the literature, is relatively high, at 34% to 50%.8,9 As stated above, radical neck dissection (RND) is a routine surgical procedure performed on all N+ patients with oral cancer and sometimes performed in NO patients as a prophylactic therapy. However, because of its cosmetic and functional disadvantages, some authors have proposed various modifications of RND. Shah et a1.l' reviewed the prevalence of neck metastases in 501 patients with oral squamous cell carcinoma who underwent RNDs. These investigators found that levels I, 11, and I11 were at greatest risk of nodal metastasis, whereas level
Table 10. Average number of metastatic nodes by histologic grade of malignancy. Histologic grade of malignancy Grades I, II
Grades 111, IV
Total
T1 ,T2 T3,T4
1.3 1.6
2.7 3.1
1.9 2.6
Total
1.4
2.9
2.1
T stage
268
Neck Metastasis of Oral Squamous Cell Carcinoma
immediate RND group
1'
2
3
4
5 year
FIGURE 5. The cumulative survival rate.
IV was rarely involved in the NO patients but was more commonly involved in the N+ groups, and that level V was rarely involved in either NO or N+ groups. They recommended that supraomohyoid neck dissection for NO, and anterolateral neck dissection for N+ necks, would remove the majority of nodal metastases. However, the histologic findings of the primary lesions were not taken into consideration in their study, nor were the figures of neck failure and survival rate mentioned. Byers et al.4 reviewed 428 patients with head and neck cancers who underwent elective modified neck dissections; these investigators recommended supraomohyoid neck dissection for oral squamous cell carcinoma patients with an NO neck. Medina et al.5 also advocated supraomohyoid neck dissection in T2-T4NO and selected N1 squamous cell carcinomas of the oral cavity, but they did not supply any detailed data to support their conclusion. Thus, supraomohyoid neck dissection has been reported to be indicated mainly as an elective therapeutic modality for NO patients. The question of whether or not elective neck therapy is necessary for clinical NO patients remains controversial today. Many investigators have recommended elective neck dissection because of the high incidence of occult metastasis in patients with clinically NO necks and the poor survival rate of patients who undergo subsequent neck dissection^.^*^,^^ A review of the literature reveals that the incidence of secondary neck metastasis after initial therapy in NO patients with oral squamous cell carcinoma varies from 15% to 46%.11-13 However it has been reported by Friedman et al.14 that CT and MRI are
HEAD & NECK
July/August 1992
Table 11. Patients of neck failure Case
Primary site
TNM
Histologic grade of malignancy
Metastasis level
Number of metastatic nodes
Tongue Lower gingiva Upper gingiva
210 210 42b0
Grade I Grade IV Grade IV
Level I Level V Level V
1 6 7
Floor of mouth Upper gingiva
100 200
Grade 111 Grade II
Level II Level II
4
Immediate neck dissection group
Subsequent neck dissection group
associated with a higher sensitivity for detecting metastatic disease than physical exmaination, and that patients with clinically normal necks and negative CT or MRI had only a 12% probability of having occult metastatic disease. These investigators advocated that the concept of elective neck dissection should be re-examined in light of the improved ability to detect nodal disease. Vandenbrouck et al.15 reported the first randomized clinical trial for patients with T1-TSNO squamous cell carcinomas of the oral cavity. They divided their 75 patients into two groups: 39 who underwent elective neck dissection, and 36 who underwent therapeutic neck dissection when neck disease was discovered. In the second group, neck disease developed in 19 patients during follow-up, but the survival rate in the two groups did not differ. Fakih et a1.16 also conducted a randomized study of prophylactic vs therapeutic neck dissection in 35 patients with TlT2NO squamous cell carcinoma of the oral tongue; they concluded that the disease-free survival rate was higher in the prophylactic neck dissection group. The different results in these two prospective randomized trials seem to be due to an unsuccessful follow-up in the latter trial. Only 28% of the patients who developed neck disease during observation were salvaged in Fakih's study. This is thought t o be due to the difficulty of carrying out close follow-up in an indigent population in the vast rural expanse of India. In contrast, close follow-up is possible here in Japan. We make it a rule to observe oral cancer patients every week during the first 6 months after their initial treatment and every 2 weeks during the next 18 months, and we attempt to detect and treat neck diseases, as well as local recurrences, in their early stage. Furthermore,
Neck Metastasis of Oral Squamous Cell Carcinoma
2
adjuvant immunotherapy, with OK-432, is also given. 0K-432 is a biological response modifier consisting of a lyophilized powder made from cultures of a penicillin-treated, low-virulence Sustrain human Streptococcus pyogenes (Picibanil, Chugai Pharmaceutical Co. Ltd., Japan). This agent is injected intracutaneously once every 2 weeks in all oral cancer patients for at least 2 years. In this study, the average number of metastatic nodes was 1.7 in the subsequent RND group, fewer than the 2.2 in the immediate RND group. The cure rate for neck disease in our department was thus almost equal in the subsequent and immediate therapeutic neck dissection groups, and the survival rate was somewhat better in the former group. These findings seem to lead to the conclusion that, if there is a close follow-up, it is possible, without risk, to delay neck dissection until neck disease is detectable. There have been many reports indicating correlation between the incidence of neck metastasis and various clinical or histologic features. Several investigator^'^ have reported that the incidence of nodal involvement was correlated with the T stage, however, others's-20 found no correlation between these parameters. It has been reported that tumor thickness or depth of invasion is an important factor for predicting neck metastasis. Spiro et a1.20found that in 105 patients with oral and oropharyngeal carcinoma with NO necks, the risk of occult neck metastasis approached 40% when the tumor thickness was >2.0 mm. Mohit-Tabatabai et a1.l' documented a significant correlation between a tumor thickness greater than 1.5 mm and subsequent development of neck metastases in a series of 84 patients with stage I and I1 carcinomas of the floor of the mouth. These findings were confirmed by Frierson and Cooper2' in 186
HEAD & NECK
July/August 1992
269
patients with carcinoma of the lower lip, and by Rasgon et al.13 in 22 patients with oral and oropharyngeal carcinoma. Thus, it has been shown in some retrospective studies that primary tumor thickness is one of the parameters related to nodal metastasis. However, we believe tumor thickness is clinically not very useful as a preoperative parameter for predicting nodal involvement, since it is difficult to measure precisely by clinical examination or in a biopsy specimen. The degree of differentiation of the tumor has been found to be correlated with nodal metastasis. Willen et a1.,22McGavran et al.,23Mendelson et al.,24 and Frierson and Cooper2’ all demonstrated a higher incidence of neck involvement in poorly differentiated than in well-differentiated carcinomas. The pattern of invasion at the tumor-host interface has also been reported to be an important histologic parameter in neck metastasis and survival in patients with head and neck cancer. Willen et a1.22found, in their observations of 124 patients with squamous cell carcinoma of the gingiva, that those patients who showed diffuse invasion experienced a higher incidence of neck metastasis and poorer prognosis than those who showed well-defined borderlines. These findings were also confirmed by Yamamoto et al.,25Crissman et a1.,26and Frierson and Cooper.21 Several other histologic parameters have been reported to be related to neck metastasis. Close et al.,27Crissman et a1.,26and Poleksic and Kalwaic28 found that vascular invasion at the primary site correlated with nodal metastasis in head and neck cancer. On the other hand, McGavran et al.23and Frierson and Cooper21found no significant correlation between vascular invasion and neck metastasis. Although several investigators21.22have shown a strong correlation between perineural invasion at the primary site and neck nodal metastasis, other^^^,^' have not. Inflammatory infiltration is another histologic parameter reported to correlate with nodal metastasis. The incidence of neck metastasis was lower in those patients with a marked inflammatory response than in those with a minimal inflammatory response, according to studies by Mohit-Tabatabai et a1.,18 Frierson and Cooper,21 and Rasgon et al.I3 Hiratsuka et al.,30 using immunohistologic analysis, showed that the T cell infiltration was correlated with a decrease of neck metastasis in squamous cell carcinoma of the oral cavity. However, several other studies
270
Neck Metastasis of Oral Squamous Cell Carcinoma
have found no correlation between inflammatory response and neck m e t a s t a ~ i s . ~We ~ ’ ~used ~’~~ the extent of differentiation and mode of invasion as histologic parameters in this study because they are generally accepted to represent histologic malignancy and they are easily assessed in standard biopsy specimens. Many investigators have studied various histologic parameters in patients with squamous cell carcinoma of the oral cavity, and some of these parameters have been reported to be related to the incidence of nodal metastasis as mentioned above. However, correlations between histologic findings in the primary site and the neck level of nodal metastasis have not been documented up to now. The prediction of the extent of neck metastasis would be of great benefit to the surgeon. We believe that our retrospective study is unique in that it emphasizes the relationship of histologic findings both to the neck level of the metastasis and to the difference between clinically detectable and histologically confirmed neck level. Our study revealed a close relationship between histologic features at the primary site and the prevalence of metastasis, especially in level 111-V necks. Histologically low malignant tumors metastasized only to level 1-11, regardless of T classification, although histologically high malignant tumors often metastasized to level 111-V, even if they had not advanced locally. Our study also indicated that the histologically confirmed neck level of metastasis almost coincided with the preoperatively detectable level in patients showing low malignancy, but that the former was often much more advanced than the latter in patients showing high malignancy. Neck metastasis usually occurs when local lesions increase in size, but in patients with histologically high malignant tumors at the primary site, this metastasis may occur before local disease becomes aggressive, and extend to the whole neck during the early stage. The results of our study show the rationale of and indications for supraomohyoid neck dissection in patients with squamous cell carcinoma of the oral cavity. Some authors have reported that this procedure is indicated in NO (or N1) oral squamous cell carcinoma, as mentioned above; however, such reports were based on clinical features alone. If the histologic findings are also taken into consideration, indications will become more precise. Although it is difficult to draw any valid conclusions here, since a small population
HEAD & NECK
July/August 1992
of patients was studied, we would advocate the following: (1) Supraomohyoid neck dissection should be limited to N1 patients whose histologic grade of malignancy is grade I (well-differentiated squamous cell carcinoma with a well-defined borderline); (2) N1 patients of grade 11-IV and N2-3 patients should undergo RND; and (3) elective neck dissection is not necessary in NO patients if a close follow-up is conducted. A large multi-institutional prospective study is expected in the future. CONCLUSION
A retrospective study of patients with squamous cell carcinoma of the oral cavity was carried out, with special reference to the location of neck metastases. An apparent correlation was found between histologic grade of malignancy and the prevalence of neck metastases. Patients with grade I - I1 histologic malignancy exhibited limited neck metastases in the level 1-11 region. On the other hand, grade 111-IV patients often had neck metastases that extended beyond the level I11 region, regardless of their T stage. These results suggest that histologic grade of malignancy, as well as clinical features, should be considered when selective limited neck dissection is performed.
REFERENCES
1. Crile GW. Excision of cancer of the head and neck. JAMA 1906;47:1780-1786. 2. Byers RM. Modified neck dissection-a study of 967 cases from 1970 to 1980. A m J Surg 1985;150:414-421. 3. Bocca E, Pignataro 0, Oldine C, Cappa C. Functional neck dissection: an evaluation of review of 843 cases. Laryngoscope 1984;94:942- 945. 4. Byers RM, Wolf PF, Ballantyne AJ. Rationale for elective modified neck dissection. Head Neck Surg 1988;lO: 160-167. 5. Medina JE, Byers RM. Supraomohyoid neck dissection: rationale, indications, and surgical technique. Head Neck Surg 1989;11:111-122. 6. UICC. TNM Ktassifikation matigner Tumoren. Vierte Auflage. Berlin: Springer-Verlag, 1987. 7. WHO. Histological typing of oral and oropharyngeal tumors. International Histological Classification of Tumours no. 4. Geneva, 1971:17. 8. Lee JG, Krause CJ. Radical neck dissection: elective, t,herapeutic and secondary. Arch Otolayngol 1975;lOl: 656-659. 9. Martis C, Karabouta I, Lazaridis N. Incidence of lymph node metastasis in elective (prophylactic) neck dissection for oral carcinoma. J Maxillofuc Surg 1979;7:182-191. 10. Shah JP, Candela FC, Poddar AK. The patterns of cervical lymph node metastases from squamous carcinoma of the oral cavity. Cuncer 1990;66:109-113.
Neck Metastasis of Oral Squamous Cell Carcinoma
11. Cunningham MJ, Johnson JT, Myers EN, Schramm VL, Thearle PB. Cervical lymph node metastasis after local excision of early squamous cell carcinoma of the oral cavity. A m J Surg 1986;152:361-366. 12. Teichgraeber JF, Clairmont AA. The incidence of occult metastasis for cancer of the oral tongue and floor of the mouth: treatment rationale. Head Neck Surg 1984;7:1521. 13. Rasgon Bm, Cruz RM, Hilsinger RL, Sawicki JE. Relation of lymph-node metastasis to histopathologic appearance in oral cavity and oropharyngeal carcinoma: a case series and literature review. Laryngoscope 1989;99: 1103- 1110. 14. Friedman M, Mafee MF, Pacella BL Jr, Strorigl TL, Dew LL, Toriumi DM. Rationale for elective neck dissection in 1990. Laryngoscope 1990;100:54-59. 15. Vandenbrouck C, Sancho-Garnier H, Chassagne D, Saravne 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;46:386-390. 16. Fakih AR, Rao RS, Pate1 AR. Prophylactic neck dissection in squamous cell carcinoma of oral tongue: a prospective randomized study. Semin Surg Oncol 1989;5: 327-330. 17. Bradfield JS, Scruggs RP. Carcinoma of the mobile tongue: incidence of cervical metastases in early lesions related to method of primary treatment. Laryngoscope 1983;93:1332- 1336. 18. Lee JG, Litton WB. Symposium on malignancy, 11. Occult regional metastasis: carcinoma of the oral tongue. Laryngoscope 1972;82:1273- 1281. 19. Mohit-Tabatabai MA, Sobel HJ, Rush BF, Mashberg A. Relation of thickness of floor of mouth stage I and I1 cancers to regional metastasis. A m J Surg 1986;152:351353. 20. Spiro RH, Huvos AG, Wong GY, Spiro JD, Gnecco CA, Strong EW. Predictive value of tumor thickness in squamous cell carcinoma confined t o the tongue and floor of the mouth. A m J Surg 1986;152:345-350. 21. Frierson HF Jr, Cooper PH. Prognostic factors in squamous cell carcinoma of the lower lip. Hum Pathol 1986;17:346-354. 22. Willen R, Nathanson A, Moberger C, Anneroth G. Squamous cell carcinoma of the gingiva: histological classification and grading of malignancy. Acta Otolaryngo1 1975;79:146- 154. 31. McGavran MH, Bauer WC, Ogura JH. The incidence of cervical lymph node metastases from epidermoid carcinoma of the larynx and their relationship to certain characteristics of the primary tumor: a study based on the clinical and pathological findings for 96 patients treated by primary en bloc laryngectomy and radical neck dissection. Cancer 1961;14:55-65. 32. Mendelson BC, Woods JE, Beahrs OH. Neck dissection in the treatment of carcinoma of the anterior two-third of the tongue. Surg Gynecol Obstet 1976;143:75-80. 33. Yamamoto E, Miyakawa A, Kohama G. Mode of invasion and lymph node metastasis in squamous cell carcinoma of the oral cavity. Head Neck Surg 1984;6:938947. 34. Crissman JD, Liu WY, Gluckman JL, Cummings G. Prognostic value of histopathologic parameters in squamous cell carcinoma of the orpharynx. Cancer 1984;54:2995- 3001. 35. Close LG, Burns DK, Reisch J, Schaefer SD. Microvascular invasion in cancer of the oral cavity and oropharynx. Arch Otolaryngol Head Neck Surg 1987;113:1191- 1195. 36. Poleksic S, Kalwaic HJ. Prognostic value of vascular invasion in squamous cell carcinoma of the head and neck. Plast Reconstr Surg 1978;61:234-240. 37. So0 K-C, Carter RC, O'Brien CJ, Barr L, Bliss JM, Shaw
HEAD 8. NECK
July/August 1992
271
HJ. Prognostic implications of perineural spread in squamous cell carcinoma of the head and neck. Laryngoscope 1986;96:1145- 1148. 38. Hiratsuka H, Imamura M, Ishii Y, Kohama G, Kikuchi
272
Neck Metastasis of Oral Squarnous Cell Carcinoma
K. Immunohistologic detection of lymphocyte subpopulations infiltrating in human oral cancer with special reference to its clinical significance. Cancer 1984;53:24562466.
HEAD 8 NECK
JulyiAugust 1992
