1464
Extent of Extracapsular Spread
A Critical Prognosticator in Oral Tongue Cancer
Jayson S. Greenberg, M.D.1
Robert Fowler, M.D.2
Jose Gomez, M.D.3
Vivian Mo2
Dianna Roberts, Ph.D.2
Adel K. El Naggar, M.D., Ph.D.3
Jeffrey N. Myers, M.D., Ph.D.2
1
The Bobby R. Alford Department of Otorhinolar-
yngology and Communicative Sciences, Baylor
College of Medicine, Houston, Texas.
2
Department of Head and Neck Surgery, The Uni-
versity of Texas M. D. Anderson Cancer Center,
Houston, Texas.
3
Department of Pathology, The University of Texas
M. D. Anderson Cancer Center, Houston, Texas.
Address for reprints: Jeffrey N. Myers, M.D., Ph.D.,
Department of Head and Neck Surgery, The Uni-
versity of Texas M. D. Anderson Cancer Center,
1515 Holcombe Blvd., Box 441, Houston, TX
77030-4009; Fax: (713) 794-4662; E-mail:
jmyers@notes.mdacc.tmc.edu
Received August 6, 2002; revision received Octo-
ber 9, 2002; accepted October 23, 2002.
© 2003 American Cancer Society
BACKGROUND. Extracapsular spread (ECS) of metastatic squamous cell carcinoma
of the head and neck to regional lymph nodes is the most reliable predictor of poor
treatment outcomes. Recently, the authors have shown that ECS is significantly
associated with higher rates of locoregional recurrence, distant metastasis, and
decreased survival in patients with squamous cell carcinoma of the oral tongue
(SCCOT). The purpose of this review was to determine if the degree of ECS impacts
distant metastasis rates and survival.
METHODS. Two hundred sixty-six patients treated for SCCOT with surgery ⫹/⫺
adjuvant radiotherapy from 1980 –1995 were reviewed. The setting was a tertiary
referral center. The extent of ECS on histopathologic review of involved lymph
nodes was measured from the capsular margin to the farthest perinodal extension
in mm. Extent of ECS and the number of pathologic lymph nodes with or without
ECS were analyzed for disease-free interval, survival rates, and distant metastases.
RESULTS. No differences in the survival of patients with ECS of ⱕ 2 mm or ⬎ 2 mm
was found (P ⫽ 0.92). Patients with both ECS and multiple positive lymph nodes
had decreased overall survival (P ⫽ 0.0003), disease-specific survival (P ⫽ 0.0005),
and a shorter disease-free interval (P ⫽ 0.019) when compared with those with a
single positive lymph node with ECS. Those with multiple ECS⫹ lymph nodes had
the worst prognosis (P ⫽ 0.001).
CONCLUSIONS. Based on these findings, the authors recommended that all patients
with SCCOT with ECS or multiple positive lymph nodes with or without ECS on
pathologic review be considered for clinical trials that intensify regional and
systemic adjuvant therapy. Cancer 2003;97:1464 –70.
© 2003 American Cancer Society.
DOI 10.1002/cncr.11202
KEYWORDS: extracapsular spread, tongue cancer.
S quamous cell carcinoma of the oral tongue (SCCOT) poses a
clinical challenge because of its metastatic potential. Several stud-
ies have shown a high rate of occult nodal metastases (20 – 40%) in
SCCOT patients with no evidence of regional spread on clinical or
radiographic evaluation.1–7 Furthermore, pathologic evidence of re-
gional nodal metastases (pN⫹) has been associated with a marked
decrease in overall and disease-specific survival.8 The finding of ex-
tracapsular spread (ECS) further decreases survival rates and is asso-
ciated with increased rates of regional recurrence and distant metas-
tases.9 –16
In a recent study of 266 patients treated surgically with or without
adjuvant radiotherapy at our institution between the years 1980 –
1995, the 5-year disease-specific and overall survival rates were 88%
and 75% for pN0 patients, 65% and 50% for pN⫹/ECS⫺ patients, and
48% and 30% for pN⫹/ ECS⫹ patients. Thirteen percent of pN0
 Extracapsular Spread in Oral Tongue Cancer/Greenberg et al.
patients received postoperative radiotherapy in con-
trast to 70% of pN⫹ patients. The patterns of failure
for the pN0, pN⫹/ECS⫺, and pN⫹/ECS⫹ groups
showed overall recurrence rates of 19.8%, 34.2%, and
51.1%; regional failure rates of 11.5%, 19.2%, and
28.9%; and distant metastases rates of 3.3%, 8.2%, and
24.4%.16
Since a high percentage of patients in the ECS⫹
groups who received adjuvant radiotherapy (88%) still
had a ⬎ 33% regional recurrence rate, an intensified
adjuvant therapy including systemic chemotherapy
and/or biologic therapy along with radiation is pro-
posed. To distinguish patients who would most likely
benefit from this intensified therapy, we further sub-
divided patients with ECS into higher- and lower-risk
groups of treatment failure based on the degree of
ECS. In this study, we analyzed the outcomes of over-
all and disease-specific survival and distant metastasis
in ECS⫹ patients based on the extension at or beyond
2.0 mm into the perilymphoid soft tissue, the number
of pathologic lymph nodes in ECS⫹ patients, and the
number of ECS⫹ nodes per patient.
MATERIALS AND METHODS
A retrospective review of medical records of all pa-
tients treated at The University of Texas M. D. Ander-
son Cancer Center for SCCOT with resection of the
primary and neck dissection from January 1980
through December 1995 was performed; all patients
with synchronous or metachronous lesions were ex-
cluded. We identified 266 patients who had undergone
surgery that included a neck dissection. In general,
patients with N0 or limited N1 disease underwent a
supraomohyoid neck dissection. Those patients with
more advanced nodal disease underwent a more com-
prehensive modified radical neck dissection. One
hundred five of these patients (39.5%) received post-
operative radiotherapy as part of their treatment. Of
these patients, 45 were found to have ECS. The slides
of 33 of these 45 patients were available for pathologic
re-review and were examined independently by two
pathologists (A.E.N. and J.G.), who scored the speci-
mens for extent of ECS outside the lymph node.
The major variables evaluated were the extent of
ECS, either ⱕ 2 mm or ⬎ 2 mm, and number of lymph
nodes involved with tumor with or without ECS. The
endpoints evaluated included disease-free interval,
survival rates, and distant metastases; none of the
patients had distant metastasis on presentation.
The Kaplan–Meier method was used to create dis-
ease-specific survival curves and disease-free interval
curves. Differences between the actuarial curves were
tested by the log rank test. Follow-up time was calcu-
lated from the patient’s initial visit at M. D. Anderson
1465
Cancer Center for the primary tumor until the date of
last contact or death. Differences in the proportions of
patients who survived 3 and 5 years and who devel-
oped recurrences were tested by the Pearson chi-
square test. If there were 10 or fewer patients in a
group, the two-tailed Fisher exact test was used. Sur-
vival analysis and chi-square tests were performed
using the Statistica software application (StatSoft, Inc.,
Tulsa, OK). Crude odds ratios for death risk were as-
sessed by Cox regression analysis, calculated with the
assistance of SPSS for Windows (version 10.0, SPSS,
Inc., Chicago, IL) software.
RESULTS
Microscopic Distance of Extension beyond the Lymph
Node Capsule
All lymph node dissection hematoxylin and eosin-
stained slides were evaluated for the presence and the
extent of ECS by head and neck pathologists. Patients
were categorized as having ⱕ 2 mm or ⬎ 2 mm of ECS,
and were compared with disease-specific survival (Fig.
1A), overall survival (Fig. 1B), and incidence of distant
metastasis (Fig. 1C). No significant statistical differ-
ences in overall survival (P ⫽ 0.92), disease-specific
survival (P ⫽ 0.31), and rate of distant metastases (P
⫽ 0.48) were identified.
Multiple Lymph Nodes with ECS
The overall and disease-specific survival of patients
with more than one ECS⫹ node had a significantly
poorer prognosis with the majority of patients dying
within 1 year after treatment compared with those
patients with only one lymph node metastasis (Fig. 2).
Disease-specific (P ⫽ 0.001) (Fig. 2A) and overall sur-
vival (P ⫽ 0.001) (Fig. 2B) also were much worse in the
multiple ECS⫹ node group. The disease-free interval
of patients with multiple ECS⫹ nodes was found to be
extremely brief, with a median time to recurrence of 6
months (Fig. 2C).
Impact of Multiple Nodes
As shown in Figure 3, patients with two or more
pathologically positive lymph nodes had worse out-
comes than those with a single node irrespective of
ECS status for both disease-specific (P ⬍ 0.0001) (Fig.
3A) and overall survival (P ⬍ 0.00001) (Fig. 3B). The
disease-free interval also was markedly reduced in
those patients with multiple positive lymph nodes (P
⫽ 0.001). Furthermore, patients with multiple positive
nodes had more ECS (33/62, 53%) than patients with
only one positive node (12/58, 21%).
1466 CANCER March 15, 2003 / Volume 97 / Number 6
FIGURE 1. Impact of tumor spread beyond the lymph node capsule (ⱕ 2 mm
or ⬎ 2 mm of extracapsular spread [ECS]) on (A) disease-specific survival, (B)
overall survival, and (C) incidence of distant metastasis in patients with
squamous cell carcinoma of the oral tongue.
Impact of Multiple Nodes with or without ECS
Figure 4 shows a composite summary of the relative
impact of involvement of multiple nodes and ECS on
disease-specific survival (Fig. 4A), overall survival (Fig.
4B), and disease-free interval (Fig. 4C). Treatment out-
comes of all groups were clearly stratified into three
outcome types. Patients with pathologic nodal classi-
FIGURE 2. Impact of number of ECS⫹ lymph nodes on (A) disease-specific
survival, (B) overall survival, and (C) disease-free interval in patients with
squamous cell carcinoma of the oral tongue.
fications pN1ECS⫺ and pN1ECS⫹ had similar dis-
ease-specific survivals as pN0 patients (P ⫽ 0.12 and P
⫽ 0.26, respectively). Patients with pathologic classifi-
cations pN2ECS⫺ and pN2ECS⫹ in one node also
fared similarly to one another (P ⫽ 0.62) but much
worse than those who were pN0 or pN1 (P ⫽ 0.0002
and P ⫽ 0.01, respectively). Patients with multiple
 Extracapsular Spread in Oral Tongue Cancer/Greenberg et al. 1467

FIGURE 3. Impact of number of pathologically positive lymph nodes regard-

less of extracapsular spread status on (A) disease-specific survival, (B) overall
survival, and (C) disease-free interval in patients with squamous cell carcinoma
of the oral tongue. FIGURE 4. Impact of involvement of multiple pathologically positive nodes
and extracapsular spread on (A) disease-specific survival, (B) overall survival,
ECS⫹ nodes clearly had the worst prognosis with and (C) disease-free interval in patients with squamous cell carcinoma of the
higher odds ratio of death from disease (Fig. 5A), over- oral tongue.
all death (Fig. 5B), and increased recurrence (Fig. 5C).

Impact of Type of Treatment on Treatment Outcomes

Table 1 categorizes pathologic nodal stage versus
treatment including the incidence of adjuvant radia-
1468 CANCER March 15, 2003 / Volume 97 / Number 6
FIGURE 5. Impact of involvement of multiple pathologically positive nodes
and extracapsular spread on odds ratio of (A) disease-specific death, (B) overall
death, and (C) recurrence in patients with squamous cell carcinoma of the oral
tongue.
TABLE 1
Pathologic Nodal Stage Versus Treatment Type in Patients with
Squamous Cell Carcinoma of the Oral Tongue
Ratio
pN stage (no. of XRT/no. in group) Percent
NO 19/144 13.2
N1ECS⫺ 21/46 45.7
N1ECS⫹ 10/12 83.3
N2ECS⫺ 25/29 86.2
N2ECS⫹ 14/16 87.5
MultiECS⫹ 11/12 91.7
pN: Pathologic nodal stage; XRT: patients who received adjuvant radiation therapy; ECS: extracapsular
spread.
tion therapy. Although relatively few pN0 patients
(13.2%) received adjuvant radiotherapy, almost all of
those with N2ECS⫺ (86.2%), N2ECS⫹ (87.5%), and
multiple ECS⫹ nodes (91.7%) were treated with post-
operative radiotherapy. Within the N1 group, 45.7% of
ECS⫺ and 83.3% ECS⫹ patients received adjuvant
radiotherapy (P ⫽ 0.02).
Patterns of Failure within Different Groups
Table 2 compares nodal classification and ECS groups
(p0,pN1⫹/⫺ ECS, pN2⫹/⫺ ECS, multiple ECS) with
patterns of failure. Local failure rates ranged from
8.3–22.2% and were independent of nodal grouping,
whereas regional nodal failure increased with ad-
vanced nodal classification. Within nodal classifica-
tion groups, the presence of a single ECS⫹ lymph
node was not associated with increased regional fail-
ure, as seen in the N1 groups. The presence of multi-
ple ECS⫹ nodes, however, was associated with an
extremely high rate of regional failure (58.3%). Distant
metastases (DM) were rarely seen in those patients
with low pathologic nodal stage in contrast with a
33.3% DM rate for those with multiple ECS⫹ nodes.
DISCUSSION
We recently reported that tongue cancer patients with
ECS had decreased overall and disease-specific sur-
vival with higher rates of regional recurrence and dis-
tant metastases than patients with no positive nodes
or positive nodes without ECS.16 In the current study,
we have further investigated the prognostic signifi-
cance of the perinodal extent of ECS and number of
positive lymph nodes on the outcomes of patients
treated at a single institution for oral tongue cancer
with surgery including neck dissection and, in many
cases, adjuvant radiotherapy.
The extent of tumor outside the lymph node cap-
sule was not found to be significantly associated with
 Extracapsular Spread in Oral Tongue Cancer/Greenberg et al. 1469

TABLE 2
Pathologic Nodal Stage Versus Patterns of Failure in Patients with Squamous Cell Carcinoma of the Oral Tongue

Local Regional DM Any recurrence

Ratio Ratio Ratio Ratio

(no. with (no. with (no. with (no. with
recurrence/ recurrence/ recurrence/ recurrence/
pN stage no. in group) Percent no. in group) Percent no. in group) Percent no. group) Percent

N0 19/142 13.4 18/142 12.7 4/142 28 29/142 20.4

N1ECS⫺ 7/46 15.2 8/46 17.4 2/46 4.3 13/46 28.3
N1ECS⫹ 1/12 8.3 2/12 16.7 0/12 0.0 3/12 25.0
N2ECS⫺ 6/27 22.2 6/27 22.2 4/27 14.8 12/27 44.4
N2ECS⫹ 3/16 18.8 4/16 25.0 4/16 25.0 7/16 43.8
Multi ECS 2/12 16.7 7/12 58.3 4/12 33.3 10/12 83.3
Total 38/255 14.9 45/255 17.6 18/255 7.1 74/255 29.0

pN: Pathologic nodal stage; DM: distant metastases; ECS: extracapsular spread.

survival outcomes or the rate of distant metastases in
the current study. Previously, other investigators have
found that greater spread outside the lymph node
capsule is associated with worse outcomes in patients
treated for laryngeal cancer.13 One explanation for this
discrepancy is that different sites in the head and neck
have different metastatic patterns. A more likely ex-
planation is the small number of patients in our series
in whom extent of tumor outside the capsule could be
quantitated due to the retrospective nature of our
study. Further studies, particularly prospective analy-
ses, are needed to fully determine the issue of whether
extent of tumor outside the lymph node capsule is a
significant predictor of survival and distant metastasis.
A striking finding of this study is the significantly
poorer outcomes of patients with multiple ECS⫹
nodes. Although the concept of ECS as a poor prog-
nosticator is not a new one,9 –16 this study is the first to
examine the incidence of ECS in depth and stratify
actual numbers of ECS⫹ nodes, nodal classification,
and treatment outcomes at a specific head and neck
subsite. The finding of multiple ECS⫹ nodes was as-
sociated with median survival of ⬍ 1 year and a high
rate of distant metastasis. Similarly, multiple lymph
nodes with one or more nodes having ECS also was
significantly associated with early recurrence and poor
survival. However, the presence of a second ECS⫹
node carried the worst prognosis. These findings can-
not be explained by treatment differences because the
majority received identical therapy including surgical
resection of the primary, neck dissection, and postop-
erative radiation therapy.
Previous studies have reviewed the role of adju-
vant chemotherapy or concurrent chemoradiation for
patients with extracapsular spread. Johnson et al.17
studied adjuvant chemotherapy in patients with extra-
capsular spread and found 58% had 30-month survival
versus 40% for those treated with only surgery and
postoperative radiation therapy. Although this was a
prospective, single arm, nonrandomized study, these
results were statistically significant. Bachaud et al.18
studied the effects of concurrent chemoradiation ther-
apy for patients with advanced squamous cell carci-
noma from all head and neck sites and histologic
evidence of ECS. They also showed improved locore-
gional control and survival compared with postoper-
ative radiotherapy alone.
The poor survival and early recurrence rates of
patients with ECS and multiple nodes with or without
ECS support the long-held contention that all patients
with one or more of these findings is at high risk for
treatment failure and suggest that intensification of
therapy is warranted. These findings also demonstrate
that further studies into the mechanisms of regional
metastases in patients with tongue cancer are needed
to provide more rational approaches to the treatment
of these high-risk patients because the majority of
patients who received surgery and radiotherapy de-
velop recurrence after treatment.
REFERENCES
1. Ho CM, Lam KH, Wei WI, Lau SK, Lam LK. Occult lymph
node metastasis in small oral tongue cancers. Head Neck.
1992;14:359 –363.
2. Cunningham MJ, Johnson JT, Myers EN, Schramm VL,
Thearle PB. Cervical lymph node metastasis after local ex-
cision of early squamous cell carcinoma of the oral cavity.
Am J Surg. 1986;152:361–366.
3. Fakih AR, Rao RS, Patel AR. Prophylactic neck dissection in
squamous cell carcinoma of oral tongue: a prospective ran-
domized study. Semin Surg Oncol. 1989;5:327–330.
4. Lydiatt DD, Robbins KT, Byers RM, Wolf PF. Treatment of
stage I and II oral tongue cancer. Head Neck. 1993;15:308 –
312.
1470 CANCER March 15, 2003 / Volume 97 / Number 6
5. Teichgraeber JF, Clairmont AA. The incidence of occult me-
tastases for cancer of the oral tongue and floor of the mouth:
treatment rationale. Head Neck. 1984;7:15–21.
6. Yuen AP, Lam KY, Chan AC, et al. Clinicopathological anal-
ysis of elective neck dissection for N0 neck of early oral
tongue carcinoma. Am J Surg. 1999;177:90 –92.
7. Yuen AP, Wei WI, Wong YM, Tang KC. Elective neck dissec-
tion versus observation in the treatment of early oral tongue
carcinoma. Head Neck. 1997;19:583–538.
8. Grandi, Alloisio E, Moglia D, et al. Prognostic significance of
lymphatic spread in head and neck carcinoma: therapeutic
implications. Head Neck. 1985;8:67–73.
9. Alvi A, Johnson JT. Extracapsular spread in the clinically
negative neck (N0): implications and outcome. Otolaryngol
Head Neck Surg. 1996;114:65–70.
10. Carter RL, Barr LC, O’Brien CJ, Soo KC, Shaw HJ. Transcap-
sular spread of metastatic squamous cell carcinoma from
cervical lymph nodes. Am J Surg. 1985;150:495– 499.
11. Johnson JT, Barnes EL, Myers EN, Schramm VL, Borochovitz
D, Sigler BA. The extracapsular spread of tumors in cervical
node metastasis. Arch Otolaryngol. 1981;107:725–729.
12. Johnson JT, Myers EN, Bedetti CD, Barnbes EL, Schramm
VL, Thearle PB. Cervical lymph node metastases. Incidence
and implications of extracapsular carcinoma. Arch Otolar-
yngol. 1985;111:534 –537.
13. Brasilino de Carvalho M. Quantitative analysis of the extent
of extracapsular invasion and its prognostic significance: a
prospective study of 170 cases of carcinoma of the larynx
and hypopharynx. Head Neck. 1998;20:16 –21.
14. Hirabayashi H, Koshii K, Uno K, et al. Extracapsular spread
of squamous cell carcinoma in neck lymph nodes: prognos-
tic factor of laryngeal cancer. Laryngoscope. 1991;101:502–
506.
15. Snyderman NL, Johnson JT, Schramm VL, Myers EN, Bedetti
CD, Thearle PB. Extracapsular spread of carcinoma in cer-
vical lymph nodes. Impact upon survival in patients with
carcinoma of the supraglottic larynx. Cancer. 1985;56:1597–
1599.
16. Myers JN, Greenberg JS, Mo V, Roberts D. Extracapsular
spread: a significant predictor of treatment failure in pa-
tients with squamous cell carcinoma of the oral tongue.
Cancer. 2001;92:3030 –3036.
17. Johnson JT, Wagner RL, Myers EN. A long-term assessment
of adjuvant chemotherapy on outcome of patients with
extracapsular spread of cervical metastases from squamous
carcinoma of the head and neck. Cancer. 1996;77:181–185.
18. Bachaud JM, Cohen-Jonathan E, Alzieu C, et al. Combined
postoperative radiotherapy and weekly cisplatin infusion for
locally advanced head and neck carcinoma: final report of a
randomized trial. Int J Radiat Oncol Biol Phys. 1996;36:999 –
1004.