Surgical
Treatment of Squamous Cell Carcinoma of the Oral Tongue Factors
Influencing Survival
John O. Whitehurst, MD, Constantine A. Droulias, MD
\s=b\ One
hundred fifty patients with squacell carcinoma of the oral tongue (anterior two thirds) treated initially by surgery alone were analyzed and factors influencing survival were evaluated. Thirteen patients who died without evidence of disease in less than five years were excluded, leaving 137 determinate cases for review. The overall survival was 67.9%. The presence or absence of cervical node metastases appeared to be the most important determining factor in this series. The size of the primary lesion in itself did not play a substantial role in local control. However, overall survival was affected because the larger the primary lesion the higher the incidence of cervical metastases, and therefore, the lower the survival. It is concluded that partial glossectomy is effective treatment for control of localized disease (92.7%). Since cervical node metastases lowered the survival figure to 31%, it appears that more aggressive prophylactic treatment in the form of neck dissection or radiotherapy is indicated for lesions larger than 2 cm. mous
(Arch Otolaryngol 103:212-215, 1977)
Accepted
for publication Dec 3, 1976. From the Department of Surgery, Robert Winship Memorial Clinic, Emory University Clinic, Emory University School of Medicine, Atlanta. Dr Whitehurst is now in private practice in Columbus, Ga. Reprint requests to St Francis Medical Park, 2300 Manchester Rd, Columbus GA 31904 (Dr
Whitehurst).
literature on cancer of the is voluminous, often confusing, and still controversial. Surgeons and radiotherapists claim equal successes for early lesions, but are less than satisfied with their results in more advanced disease. The proponents of surgery insist that surgery, if performed initially, is more effective, less painful, and less expensive in time and cost.1 Opinions differ even among surgeons as to the extent of excision of the primary lesion and the proper place of radical neck dissection in the so-called early cancer. However, there is wide and general agreement that cervical node metastases are best treated by surgery or combined treat-
Thetongue
ment.2-4
Experienced radiotherapists believe that with modern equipment and refined techniques one can eradicate early cancer of the tongue in a substantial proportion of cases with minimal stigmata and with preservation of functional integrity exceeding that following a surgical procedure.5-7 Others take the view that both modalities are equally effective in early lesions (less than 3 cm) and the choice of one over the other depends on the facilities at hand and the experience and skill of the clinician.1.8 However, as larger and more advanced primary
Downloaded From: http://archotol.jamanetwork.com/ by a University of Iowa User on 06/20/2015
lesions are encountered and the incidence of metastases to the regional nodes increases, the chance of ultimate control diminishes in spite of more radical surgery or improved radiotherapeutic techniques.1.9 In the latter cases, some authors believe that a combination of preoperative radiotherapy followed by radical surgery offers the best chance for cure.1.3.4 We analyze the effectiveness of surgery as the initial modality in the treatment of cancer of the oral tongue and evaluate factors influencing prognosis. PATIENTS AND METHODS
through 1970, 275 previously untreated patients with squamous cell carcinoma of the tongue were seen and treated at Emory University Clinic. Seventy patients with lesions located in the posterior one third of the tongue were excluded from the study, since such patients represent an entirely different From 1940
problem.
Of the 205 patients with lesions of the anterior two thirds, 49 treated initially by radiotherapy and six by combination of radiotherapy and surgery were excluded. The number of different techniques and dosages used (especially in the presupervoltage era) and the fact that more patients with advanced disease were treated by radiotherapy precluded a reliable evaluation of this modality.
Of the remaining 150 patients with squacell carcinoma of the anterior two thirds treated initially by surgery alone, 13 who died without evidence of disease in less than five years following treatment were not included in the survival figures. Thus, 137 determinate patients with 100% follow-up (either by examination in the office or information from the local physi¬ cian) form the basis of this report. All classifications are retrospective and according to the TNM system of the Amer¬ ican Joint Committee for Cancer Staging and End Results Reporting"' (Table 1). However, in order to gain practical infor¬ mation, these patients were subsequently grouped in two categories; (1) those whose disease (of any size) on the first examina¬ tion was limited to the tongue, and (2) those who, in addition to the tongue lesion, had palpable cervical nodes.
Table 1.—Definition of TNM
mous
subsequent treatment (Table 4). This last figure is comparable to the results from Memorial Hospital," although, because of the small number of patients in the T.,N„ group, the
incidence of local control appears higher than might be expected. Eighty-five percent of the local fail-
Tumors About the Oral
T—Primary Tumor
TIS—Carcinoma in situ
T,— Tumor 2 cm or less in greatest diameter T.,—Tumor greater than 2 cm, but no greater than 4 cm in greatest diameter. T—Tumor greater than 4 cm in greatest diameter N—Regional Lymph Nodes N„—No clinically palpable cervical lymph node(s) or palpable node(s); metastasis not suspected N,—Clinically palpable homolateral cervical lymph node(s) that are not fixed; metastasis suspected N,—Clinically palpable contralateral or bilateral cervical lymph node(s) that are not fixed; metastasis suspected N—Clinically palpable lymph node(s) that are fixed; metastasis suspected M—Distant Metastasis M„—No distant metastasis M,—Clinical and/or radiographie evidence of metastasis other than to cervical nodes
lymph
'From Clinical Staging System for Carcinoma of the Oral Cavity.'"
Primary Treatment (Determinate Surgery Cases)
Table 2—Results of
RESULTS
The extent of surgical treatment in the first group with localized disease (124 patients) consisted of partial glossectomy alone in all but seven patients. These seven had prophy¬ lactic neck dissections as well. The second group with regional disease (13 patients) were treated by partial glos¬ sectomy and radical neck dissection. The five-year survival free of disease after primary treatment alone for the 137 determinate cases was 53% (Table 2). This figure and those for T,N„, T..N,,, and regional (67.0%, 43.0%, and 38.5% respectively) were compa¬ rable to results in previous reports." The conflicting results for T.N„ lesions (25%) can not be logically explained. Using retrospective de¬ scriptions, however, the figure is correct for our patient series. The failure rate following the initial surgery is shown in Table 3. Surgery alone proved to be adequate treat¬ ment for localized disease in about 80% of patients, although a significant number later developed neck node métastases (35%). However, the ulti¬ mate local control was 92.7% since some local failures were salvaged by
System of Malignant Cavity.*
Stage
No. Cured
(%) 55(67.1) 7(25.0) 6(42.9)
No. Treated 82 28 14
T,N„ T.,N„ T.,N„ Regional (any with nodes)
5(38.5) 73(53.2)
13 137
Total
Table 3.—Failure Rate
Following
Result
Surgical Treatment
Initial
T,N
T.N
8/82
4/28
Regional 0/14
3/13
Total 15/137
(10.9%)
Local failures
(9.8%) (14.3%)
(0%)
(23.1%)
Development of neck disease In previously untreated neck Local failure with development of disease in previously untreated
13/82
6/14
3*/13
34/137
(23.1%)
(24.8%)
12/28
(15.9%) (42.9%) (42.9%) 4/28
7/82
2/14
(8.5%) (14.3%) (14.3%)
neck
2*/13
15/137
(15.9%)
(10.9%)
-Contralateral.
Table 4.—Control of
Primary
Lesion
Result Total no. of patients Determinate patients Local recurrence
Local
Surgery
T,N„
T.N,
90 82
29 28
TN, 15
15
Salvaged by subsequent
treatment
Local control in determinate
Table
by
75/82 91.4%
patients
5.—Salvage
Rate with
26/28 92.8%
Secondary
Result
Downloaded From: http://archotol.jamanetwork.com/ by a University of Iowa User on 06/20/2015
100%
19 10 115/124 92.7%
Treatment
No. Treated
Local failures
Development of neck disease in previously untreated neck Local failure with development of neck disease in previously untreated neck (synchronous or metachronous)
14/14
Total 134 124
34
15
No.
Salvaged (%) 8(53.3) 12(35.3) 0(0)
Table 6.—Overall Results After
Primary
Stage
with
Secondary
Treatment
No. Treated 82
T,N„ T..N„ T,N„
Regional (any
and
No. Cured
(%) 64(78.0) 12(42.9) 10(71.4) 7(53.9) 93(67.9)
28 14 13
nodes)
Total
137
ures, cervical node métastases, or both, occurred within one year, 95% within two years, and all (100%) within three years from the initial treatment. There were no failures in the neck in those patients who had radical neck
dissections
as
part of their initial
management. However, there
Table 7.—Incidence of Cervical Node Metastasis After Local
Stage
No. of Patients in Whom Métastases Developed (%)
No. of Patients 82 28 14 124 s
T,N„ T,N„
T,N„ Total
"The 13 patients
initially
N+
are
Type of Dissection Prophylactic neck dissection Therapeutic neck dissection at of initial therapy Therapeutic neck dissection at later date when nodes
Neck dissection with negative nodes Neck dissection with positive nodes
20(24.4) 16(57.1) 8(57.1) 44(35.9)
not included.
8.—Analysis
Table
Surgery
of 51 Radical Neck Dissections No. Treated
No. Cured (%) 5 (71.4)
time 13
5
(38.5)
31
11
(35.5)
5
(71.4)
16
(36.4)
a
appeared
pathologically pathologically
Table 9.—Factors Factor
Influencing Survival No. Treated
No.
Surviving (%)
Age,
yr < 49 50-69
> 70 Sex Female Male Decade ol Treatment 1940s 1950s
1960s
Size of Primary Lesion < 2 cm 2-4 cm
30
23
36
49 21
57
40
80
53
34 58 45
37 35
82 28
> 4 cm
Presence of Cervical Métastases Absent*
Presenti
80
57
Alcohol and/or Tobacco
"During the entire tEither initially or later.
course
21
(76.6) (69) (58.3) (70.2) (66.3) (61.8) (63.8) (77.8)
(78.0) (42.9) 10 (71.4)
64 12
75 18
(92.7) (31)
No reliable information of their disease.
Downloaded From: http://archotol.jamanetwork.com/ by a University of Iowa User on 06/20/2015
were
five who developed contralateral metastastic nodes. Fifty-three percent of patients with local failures and 35% of those who later developed neck metastases were salvaged with subsequent treatment, but the combination of local failures with cervical node me¬ tastasis, synchronous or metachronous, was uniformly fatal (Table 5). The overall survival for initially localized disease was 69%; for region¬ al, 54%; and total, 69% (Table 6). The incidence of cervical node métastases in this study is shown in Table 7. An analysis of 51 radical neck dissections showed that 71.4% of those patients who had prophylactic dissec¬ tions survived five years or more free of disease, in contrast to 36.4% for those who had therapeutic dissections. Of the therapeutic dissections, how¬ ever, there was no statistically signif¬ icant difference between those done initially (38.5%) and those done later, when nodes developed (35.5%) (Table 8). The survival for patients with pathologically negative nodes was also twice as high as those with positive nodes. It appears that prophylactic neck dissection in all patients in this series might have salvaged eight addi¬ tional patients. This advantage would be at the expense of 62 prophylactic neck dissections in patients with negative nodes, and therefore, per¬ haps, needlessly done. FACTORS INFLUENCING SURVIVAL As previously reported,"'-' the sur¬ vival rate appears to be somewhat higher in the younger age group, with no difference in sex (Table 9). It would appear from this study that there has been no significant improvement in surgical results for cancer of the tongue during the last three decades {P < .02). The presence or absence of cervi¬ cal node métastases was an impor-
tant
factor
determining survival
reliable data on the effects of the use of tobacco or alcohol. Although the tumor size alone did not play an important role in local control, it certainly affected the ultimate survival.
(
Treatment of Squamous Cell Carcinoma of the Oral Tongue Factors
Influencing Survival
John O. Whitehurst, MD, Constantine A. Droulias, MD
\s=b\ One
hundred fifty patients with squacell carcinoma of the oral tongue (anterior two thirds) treated initially by surgery alone were analyzed and factors influencing survival were evaluated. Thirteen patients who died without evidence of disease in less than five years were excluded, leaving 137 determinate cases for review. The overall survival was 67.9%. The presence or absence of cervical node metastases appeared to be the most important determining factor in this series. The size of the primary lesion in itself did not play a substantial role in local control. However, overall survival was affected because the larger the primary lesion the higher the incidence of cervical metastases, and therefore, the lower the survival. It is concluded that partial glossectomy is effective treatment for control of localized disease (92.7%). Since cervical node metastases lowered the survival figure to 31%, it appears that more aggressive prophylactic treatment in the form of neck dissection or radiotherapy is indicated for lesions larger than 2 cm. mous
(Arch Otolaryngol 103:212-215, 1977)
Accepted
for publication Dec 3, 1976. From the Department of Surgery, Robert Winship Memorial Clinic, Emory University Clinic, Emory University School of Medicine, Atlanta. Dr Whitehurst is now in private practice in Columbus, Ga. Reprint requests to St Francis Medical Park, 2300 Manchester Rd, Columbus GA 31904 (Dr
Whitehurst).
literature on cancer of the is voluminous, often confusing, and still controversial. Surgeons and radiotherapists claim equal successes for early lesions, but are less than satisfied with their results in more advanced disease. The proponents of surgery insist that surgery, if performed initially, is more effective, less painful, and less expensive in time and cost.1 Opinions differ even among surgeons as to the extent of excision of the primary lesion and the proper place of radical neck dissection in the so-called early cancer. However, there is wide and general agreement that cervical node metastases are best treated by surgery or combined treat-
Thetongue
ment.2-4
Experienced radiotherapists believe that with modern equipment and refined techniques one can eradicate early cancer of the tongue in a substantial proportion of cases with minimal stigmata and with preservation of functional integrity exceeding that following a surgical procedure.5-7 Others take the view that both modalities are equally effective in early lesions (less than 3 cm) and the choice of one over the other depends on the facilities at hand and the experience and skill of the clinician.1.8 However, as larger and more advanced primary
Downloaded From: http://archotol.jamanetwork.com/ by a University of Iowa User on 06/20/2015
lesions are encountered and the incidence of metastases to the regional nodes increases, the chance of ultimate control diminishes in spite of more radical surgery or improved radiotherapeutic techniques.1.9 In the latter cases, some authors believe that a combination of preoperative radiotherapy followed by radical surgery offers the best chance for cure.1.3.4 We analyze the effectiveness of surgery as the initial modality in the treatment of cancer of the oral tongue and evaluate factors influencing prognosis. PATIENTS AND METHODS
through 1970, 275 previously untreated patients with squamous cell carcinoma of the tongue were seen and treated at Emory University Clinic. Seventy patients with lesions located in the posterior one third of the tongue were excluded from the study, since such patients represent an entirely different From 1940
problem.
Of the 205 patients with lesions of the anterior two thirds, 49 treated initially by radiotherapy and six by combination of radiotherapy and surgery were excluded. The number of different techniques and dosages used (especially in the presupervoltage era) and the fact that more patients with advanced disease were treated by radiotherapy precluded a reliable evaluation of this modality.
Of the remaining 150 patients with squacell carcinoma of the anterior two thirds treated initially by surgery alone, 13 who died without evidence of disease in less than five years following treatment were not included in the survival figures. Thus, 137 determinate patients with 100% follow-up (either by examination in the office or information from the local physi¬ cian) form the basis of this report. All classifications are retrospective and according to the TNM system of the Amer¬ ican Joint Committee for Cancer Staging and End Results Reporting"' (Table 1). However, in order to gain practical infor¬ mation, these patients were subsequently grouped in two categories; (1) those whose disease (of any size) on the first examina¬ tion was limited to the tongue, and (2) those who, in addition to the tongue lesion, had palpable cervical nodes.
Table 1.—Definition of TNM
mous
subsequent treatment (Table 4). This last figure is comparable to the results from Memorial Hospital," although, because of the small number of patients in the T.,N„ group, the
incidence of local control appears higher than might be expected. Eighty-five percent of the local fail-
Tumors About the Oral
T—Primary Tumor
TIS—Carcinoma in situ
T,— Tumor 2 cm or less in greatest diameter T.,—Tumor greater than 2 cm, but no greater than 4 cm in greatest diameter. T—Tumor greater than 4 cm in greatest diameter N—Regional Lymph Nodes N„—No clinically palpable cervical lymph node(s) or palpable node(s); metastasis not suspected N,—Clinically palpable homolateral cervical lymph node(s) that are not fixed; metastasis suspected N,—Clinically palpable contralateral or bilateral cervical lymph node(s) that are not fixed; metastasis suspected N—Clinically palpable lymph node(s) that are fixed; metastasis suspected M—Distant Metastasis M„—No distant metastasis M,—Clinical and/or radiographie evidence of metastasis other than to cervical nodes
lymph
'From Clinical Staging System for Carcinoma of the Oral Cavity.'"
Primary Treatment (Determinate Surgery Cases)
Table 2—Results of
RESULTS
The extent of surgical treatment in the first group with localized disease (124 patients) consisted of partial glossectomy alone in all but seven patients. These seven had prophy¬ lactic neck dissections as well. The second group with regional disease (13 patients) were treated by partial glos¬ sectomy and radical neck dissection. The five-year survival free of disease after primary treatment alone for the 137 determinate cases was 53% (Table 2). This figure and those for T,N„, T..N,,, and regional (67.0%, 43.0%, and 38.5% respectively) were compa¬ rable to results in previous reports." The conflicting results for T.N„ lesions (25%) can not be logically explained. Using retrospective de¬ scriptions, however, the figure is correct for our patient series. The failure rate following the initial surgery is shown in Table 3. Surgery alone proved to be adequate treat¬ ment for localized disease in about 80% of patients, although a significant number later developed neck node métastases (35%). However, the ulti¬ mate local control was 92.7% since some local failures were salvaged by
System of Malignant Cavity.*
Stage
No. Cured
(%) 55(67.1) 7(25.0) 6(42.9)
No. Treated 82 28 14
T,N„ T.,N„ T.,N„ Regional (any with nodes)
5(38.5) 73(53.2)
13 137
Total
Table 3.—Failure Rate
Following
Result
Surgical Treatment
Initial
T,N
T.N
8/82
4/28
Regional 0/14
3/13
Total 15/137
(10.9%)
Local failures
(9.8%) (14.3%)
(0%)
(23.1%)
Development of neck disease In previously untreated neck Local failure with development of disease in previously untreated
13/82
6/14
3*/13
34/137
(23.1%)
(24.8%)
12/28
(15.9%) (42.9%) (42.9%) 4/28
7/82
2/14
(8.5%) (14.3%) (14.3%)
neck
2*/13
15/137
(15.9%)
(10.9%)
-Contralateral.
Table 4.—Control of
Primary
Lesion
Result Total no. of patients Determinate patients Local recurrence
Local
Surgery
T,N„
T.N,
90 82
29 28
TN, 15
15
Salvaged by subsequent
treatment
Local control in determinate
Table
by
75/82 91.4%
patients
5.—Salvage
Rate with
26/28 92.8%
Secondary
Result
Downloaded From: http://archotol.jamanetwork.com/ by a University of Iowa User on 06/20/2015
100%
19 10 115/124 92.7%
Treatment
No. Treated
Local failures
Development of neck disease in previously untreated neck Local failure with development of neck disease in previously untreated neck (synchronous or metachronous)
14/14
Total 134 124
34
15
No.
Salvaged (%) 8(53.3) 12(35.3) 0(0)
Table 6.—Overall Results After
Primary
Stage
with
Secondary
Treatment
No. Treated 82
T,N„ T..N„ T,N„
Regional (any
and
No. Cured
(%) 64(78.0) 12(42.9) 10(71.4) 7(53.9) 93(67.9)
28 14 13
nodes)
Total
137
ures, cervical node métastases, or both, occurred within one year, 95% within two years, and all (100%) within three years from the initial treatment. There were no failures in the neck in those patients who had radical neck
dissections
as
part of their initial
management. However, there
Table 7.—Incidence of Cervical Node Metastasis After Local
Stage
No. of Patients in Whom Métastases Developed (%)
No. of Patients 82 28 14 124 s
T,N„ T,N„
T,N„ Total
"The 13 patients
initially
N+
are
Type of Dissection Prophylactic neck dissection Therapeutic neck dissection at of initial therapy Therapeutic neck dissection at later date when nodes
Neck dissection with negative nodes Neck dissection with positive nodes
20(24.4) 16(57.1) 8(57.1) 44(35.9)
not included.
8.—Analysis
Table
Surgery
of 51 Radical Neck Dissections No. Treated
No. Cured (%) 5 (71.4)
time 13
5
(38.5)
31
11
(35.5)
5
(71.4)
16
(36.4)
a
appeared
pathologically pathologically
Table 9.—Factors Factor
Influencing Survival No. Treated
No.
Surviving (%)
Age,
yr < 49 50-69
> 70 Sex Female Male Decade ol Treatment 1940s 1950s
1960s
Size of Primary Lesion < 2 cm 2-4 cm
30
23
36
49 21
57
40
80
53
34 58 45
37 35
82 28
> 4 cm
Presence of Cervical Métastases Absent*
Presenti
80
57
Alcohol and/or Tobacco
"During the entire tEither initially or later.
course
21
(76.6) (69) (58.3) (70.2) (66.3) (61.8) (63.8) (77.8)
(78.0) (42.9) 10 (71.4)
64 12
75 18
(92.7) (31)
No reliable information of their disease.
Downloaded From: http://archotol.jamanetwork.com/ by a University of Iowa User on 06/20/2015
were
five who developed contralateral metastastic nodes. Fifty-three percent of patients with local failures and 35% of those who later developed neck metastases were salvaged with subsequent treatment, but the combination of local failures with cervical node me¬ tastasis, synchronous or metachronous, was uniformly fatal (Table 5). The overall survival for initially localized disease was 69%; for region¬ al, 54%; and total, 69% (Table 6). The incidence of cervical node métastases in this study is shown in Table 7. An analysis of 51 radical neck dissections showed that 71.4% of those patients who had prophylactic dissec¬ tions survived five years or more free of disease, in contrast to 36.4% for those who had therapeutic dissections. Of the therapeutic dissections, how¬ ever, there was no statistically signif¬ icant difference between those done initially (38.5%) and those done later, when nodes developed (35.5%) (Table 8). The survival for patients with pathologically negative nodes was also twice as high as those with positive nodes. It appears that prophylactic neck dissection in all patients in this series might have salvaged eight addi¬ tional patients. This advantage would be at the expense of 62 prophylactic neck dissections in patients with negative nodes, and therefore, per¬ haps, needlessly done. FACTORS INFLUENCING SURVIVAL As previously reported,"'-' the sur¬ vival rate appears to be somewhat higher in the younger age group, with no difference in sex (Table 9). It would appear from this study that there has been no significant improvement in surgical results for cancer of the tongue during the last three decades {P < .02). The presence or absence of cervi¬ cal node métastases was an impor-
tant
factor
determining survival
reliable data on the effects of the use of tobacco or alcohol. Although the tumor size alone did not play an important role in local control, it certainly affected the ultimate survival.
(
