ARTICLES

Colon Cancer Survival Rates With the New American

Joint Committee on Cancer Sixth Edition Staging
Jessica B. OConnell, Melinda A. Maggard, Clifford Y. Ko
Background: The recently revised American Joint Committee on Cancer (AJCC) sixth edition cancer staging system
increased the stratification within colon cancer stages II and
III defined by the AJCC fifth edition system. Using nationally representative Surveillance, Epidemiology, and End Results (SEER) data, we compared survival rates associated
with colon cancer stages defined according to both AJCC
systems. Methods: Using SEER data (from January 1, 1991,
through December 31, 2000), we identified 119 363 patients
with colon adenocarcinoma and included all patients in two
analyses by stages defined by AJCC fifth and sixth edition
systems. Tumors were stratified by SEERs extent of disease and number of positive [lymph] nodes coding
schemes. KaplanMeier analyses were used to compare
overall and stage-specific 5-year survival. All statistical tests
were two-sided. Results: Overall 5-year survival was 65.2%.
According to stages defined by the AJCC fifth edition system, 5-year stage-specific survivals were 93.2% for stage I,
82.5% for stage II, 59.5% for stage III, and 8.1% for stage
IV. According to stages defined by the AJCC sixth edition
system, 5-year stage-specific survivals were 93.2% for stage
I, 84.7% for stage IIa, 72.2% for stage IIb, 83.4% for stage
IIIa, 64.1% for stage IIIb, 44.3% for stage IIIc, and 8.1% for
stage IV. Under the sixth edition system, 5-year survival was
statistically significantly better for patients with stage IIIa
colon cancer (83.4%) than for patients with stage IIb disease
(72.2%) (P<.001). Conclusions: The AJCC sixth edition system for colon cancer stratifies survival more distinctly than
the fifth edition system by providing more substages. The
association of stage IIIa colon cancer with statistically significantly better survival than stage IIb in the new system
may reflect current clinical practice, in which stage III patients receive chemotherapy but stage II patients generally
do not. [J Natl Cancer Inst 2004;96:1420 5]
Since 1959, the American Joint Committee on Cancer
(AJCC) has been working to formulate systems of classification
for cancer staging. These systems are designed to enable physicians to stratify patients in terms of expected predicted survival,
to help select the most effective treatments, to determine prognoses, and to evaluate cancer control measures (1,2).
As of January 1, 2003, the newest edition of the AJCC Cancer
Staging Manual, sixth edition, is being used to stage all new
cancer cases. This edition contains updated classification
schemes for several cancers, including colon cancer. The prior
(fifth edition) staging system for colon cancer had four categories that were based on the tumornodemetastasis (TNM) classification (stages I, II, III, and IV). The new (sixth edition)
staging system for colon cancer stratifies stages II and III further
1420 ARTICLES

by use of the T stage (i.e., tumor depth of penetration) and the N

stage (i.e., number of positive lymph nodes), resulting in a total
of seven stages (stages I, IIa, IIb, IIIa, IIIb, IIIc, and IV; Table
1) (1).
In this article, we use data from the Surveillance, Epidemiology, and End Results (SEER1) program, a large national
cancer registry, to compare survival outcomes by stage as defined by both the older (fifth edition) and the newer (sixth
edition) AJCC staging systems for colon cancer. We also evaluate other factors, in addition to TNM stages, as potential
prognostic factors.

PATIENTS

AND

METHODS

We identified and evaluated all 119 363 patients diagnosed

with colon adenocarcinoma in the SEER national cancer registry
from January 1, 1991, through December 31, 2000. The SEER
program collects patient records from multiple sites across the
United States and is regarded as a model population-based tumor
registry. This national program includes 12 regional registries
that cover approximately 14% of the U.S. population. The database was designed to reflect the overall characteristics of the
U.S. population, including the diverse array of racial and/or
ethnic groups, geographic locations, and types of cities and
states (3).
We characterized tumors according to the location in the
colon: cecum, ascending colon, hepatic flexure, transverse colon, splenic flexure, descending colon, sigmoid colon, and large
intestine NOS (not otherwise specified). Appendiceal, rectosigmoid, and rectal locations were excluded, as were in situ tumors.
We further restricted the tumors included by specific histologic
type, as defined by the following individual International Classification of Diseases for Oncology, second edition, codes for
adenocarcinomas (8010, 8020 8022, 8140 8145, 8210 8211,
8220 8221, 8230 8231, 8260 8263), mucinous adenocarcinomas (8470, 8480, 8481), and signet ring cell carcinoma (9490).

Affiliations of authors: Department of Surgery, David Geffen School of

Medicine at University of California, Los Angeles; Department of Surgery, West
Los Angeles Veteran Affairs Hospital, Los Angeles, CA.
Correspondence to: Jessica Beth OConnell, MD, Department of Surgery,
David Geffen School of Medicine at University of California, Los Angeles,
10833 Le Conte Ave., Rm. 72-215 CHS, Los Angeles, CA 90095 (e-mail:
[email protected]).
See Notes following References.
DOI: 10.1093/jnci/djh275
Journal of the National Cancer Institute, Vol. 96, No. 19, Oxford University
Press 2004, all rights reserved.

Journal of the National Cancer Institute, Vol. 96, No. 19, October 6, 2004

Table 1. Stages as defined by the American Joint Committee on Cancer

(AJCC) fifth and sixth edition staging systems
Staging system

T stage

N stage

M stage

AJCC fifth edition

I
II
III
IV

T1 or
T3 or
Any
Any

T2
T4
T
T

N0
N0
N1
Any N

M0
M0
M0
M1

AJCC sixth edition

I
IIa
IIb
IIIa
IIIb
IIIc
IV

T1 or T2
T3
T4
T1 or T2
T3 or T4
Any T
Any T

N0
N0
N0
N1
N1
N2
Any N

M0
M0
M0
M0
M0
M0
M1

T1 tumor invades submucosa; T2 tumor invades muscularis propria; T3

tumor invades through the muscularis propria into the subserosa or into
nonperitonealized pericolic tissues; T4 tumor directly invades other organs or
structures and/or perforates visceral peritoneum; N0 no regional lymph node
metastasis; N1 metastasis to one to three regional lymph nodes; N2
metastasis to four or more regional lymph nodes; M0 no distant metastasis;
M1 distant metastasis.

Patient Demographics
Demographic information recorded for each patient included
age, sex, race, and/or ethnicity (white, black, Hispanic, Asian, or
other), marital status (single or married at diagnosis), and
SEER registry site (Alaska, Atlanta, Connecticut, Detroit, Hawaii, Iowa, Los Angeles [after 1992], New Mexico, San Francisco, San Jose [after 1992], Seattle, and Utah).
Tumor and Disease Characteristics
Cancer-specific data evaluated for each patient included stage
at presentation, tumor grade, specific histology, tumor location,
number of positive lymph nodes, and metastases. Each tumor
stage was coded as described by the AJCC fifth and sixth
editions according to the TNM stage organization for each
edition (T1 tumor invades submucosa; T2 tumor invades
muscularis propria; T3 tumor invades through the muscularis
propria into the subserosa or into nonperitonealized pericolic
tissues; T4 tumor directly invades other organs or structures
and/or perforates visceral peritoneum; N0 no regional lymph
node metastasis; N1 metastasis to one to three regional lymph
nodes; N2 metastasis to four or more regional lymph nodes;
M0 no distant metastasis; M1 distant metastasis). TNM
stage was determined by SEERs extent of disease (for T stage
and M stage) and number of positive [lymph] nodes (for N
stage) coding schemes. All patients were included in both analyses of survival for both staging systems. Tumor grade was
categorized as low grade (well or moderately differentiated) and
high grade (poorly differentiated, anaplastic, or undifferentiated). Tumor location was categorized as right (cecum, ascending colon, hepatic flexure), transverse, left (splenic flexure, descending colon), and sigmoid colon. The numbers of positive
lymph nodes were categorized and examined.
Survival and Statistical Analyses
KaplanMeier analyses were performed to determine and to
compare overall and stage-specific 5-year survivals for stages
defined by both the fifth and sixth edition staging systems. Colon

cancerspecific survival (as opposed to observed survival or

survival from any cause of death) was determined to obtain the
most specific survival associated with colon cancer. In additional
analyses, we used AJCC fifth edition staging as a base and
stratified data into additional stages by other potential prognostic
factors. For this analysis, we used the KaplanMeier method and
the fifth edition staging system and then stratified the stages by
tumor grade, specific histologic type, tumor location, and number of positive lymph nodes. We also evaluated a proposed
system of staging that separated the N stage into stages N1 (one
to three positive lymph nodes), N2 (four or five positive lymph
nodes), N3 (six to eight positive lymph nodes), and N4 (nine or
more positive lymph nodes) (Table 2). Because there is no
consensus in the literature regarding potential further separation
of N stage, we used data on the number of positive lymph nodes
for all patients evaluated and analyzed these data by use of a
histogram, and broke the data into reasonable groups based on
the number of patients with each number of lymph nodes and the
dropoffs noted on the histogram after three lymph nodes, five
lymph nodes, or eight lymph nodes.
All statistical analyses used SAS version 8.02 (SAS Institute,
Cary, NC) or Stata Intercooled version 7.0 (Stata, College Station, TX). P values for survival curves were determined from the
KaplanMeier survival curves by use of the log-rank test. Those
values less than .05 were considered statistically significant. All
statistical tests were two-sided.

RESULTS
Demographics
Data from a total of 119 363 patients with colon adenocarcinoma were evaluated. Mean age ( standard deviation) for the
cohort was 71.1 12.6 years. Females represented 51.6% of the
group, and 54.4% of patients were married at the time of
diagnosis. The overall racial and/or ethnic distribution was
77.0% whites, 9.7% blacks, 5.4% Hispanics, 5.9% Asians, and
2.0% other.
Survival by AJCC Fifth and Sixth Edition Staging
Overall 5-year colon cancerspecific survival for the entire
cohort was 65.2%. By use of stages defined by the AJCC fifth
Table 2. Proposed staging system
Stage
I
IIa
IIb
IIIa
IIIb
IIIc
IIId
IIIc
IV

T stage

N stage

M stage

T1 or T2
T3
T4
T1 or T2
T3 or T4
Any T
Any T
Any T
Any T

N0
N0
N0
N1
N1
N2
N3
N4
Any N

M0
M0
M0
M0
M0
M0
M0
M0
M1

Stages I, IIa, IIb, IIIa, and IIIb are the same as in the American Joint
Committee on Cancer sixth edition system. T1 tumor invades submucosa; T2
tumor invades muscularis propria; T3 tumor invades through the muscularis
propria into the subscrosa or into nonperitonealized pericolic tissues; T4
tumor directly invades other organs or structures and/or perforates visceral
peritoneum; N0 no positive lymph nodes; N1 one to three positive lymph
nodes; N2 four or five positive lymph nodes; N3 six to eight positive lymph
nodes; N4 nine or more positive lymph nodes; M0 no distant metastasis;
M1 distant metastasis.

Journal of the National Cancer Institute, Vol. 96, No. 19, October 6, 2004

ARTICLES 1421

edition system, 5-year colon cancer-specific survival by stage

was 93.2% for stage I, 82.5% for stage II, 59.5% for stage III,
and 8.1% for stage IV. By use of stages defined by the revised
AJCC sixth edition system, 5-year colon cancerspecific survival by stage was 93.2% for stage I, 84.7% for stage IIa, 72.2%
for stage IIb, 83.4% for stage IIIa, 64.1% for stage IIIb, 44.3%
for stage IIIc, and 8.1% for stage IV. The survival analysis for
the cohort using the fifth edition staging system is shown in Fig.
1, and that using the sixth edition staging system is in Fig. 2. It
is noteworthy that, as defined by the new sixth edition system,
stage IIIa colon cancer (83.4%) has a statistically significantly
better prognosis than stage IIb colon cancer (72.2%) (P.001).
Survival by Tumor Grade
We next used colon cancer stages as defined by the AJCC
fifth edition system and stratified data in each stage further by
other factors to assess their prognostic value. Among all patients
evaluated in the cohort, 67.8% (n 81 493) had low-grade
tumors, 19.4% (n 23 287) had high-grade tumors, and 12.8%
(n 15 343) had tumors whose grade was unknown. For those
patients whose tumor grade (high versus low) was known (n
104 780), tumor grade was statistically significantly associated
with survival in stages II, III, and IV, but not in stage I (Fig. 3).
Survival by Histologic Subtype
Among patients in the entire cohort, 87.4% had adenocarcinomas, 11.6% had mucinous adenocarcinomas, and 1.0% had
signet ring cell carcinomas. Among the entire cohort, a worse
5-year survival was statistically significantly associated with
signet ring cell carcinomas (36.0%) than with adenocarcinomas
(65.9%) (P.001) or with mucinous adenocarcinomas (61.8%)
(P.001). When we further stratified data in each stage (as
defined by the fifth edition system) by histologic subtype, we
observed similar survival distributions in stages II, III, and IV,
but not in stage I (Fig. 4). For example, in stage III, the 5-year
survival was 36.6% for signet ring cell carcinomas, 60.1% for
adenocarcinomas (P.001), and 58.7% for mucinous adenocar-

Fig. 1. Five-year survival by American Joint Committee on Cancer fifth edition

system stages IIV. P value determined with the log-rank test refers to the
corresponding stage and the stage in the row above. All statistical tests were
two-sided.

1422 ARTICLES

Fig. 2. Five-year survival by the American Joint Committee on Cancer sixth

edition system stages IIV. P value determined by the log-rank test refers to the
corresponding stage and the stage in the row above, unless otherwise indicated.
All statistical tests were two-sided. * IIIa versus IIb; IIa versus IIIa;
IIb versus IIIb; NS not statistically significant.

cinomas (P.001). For stage I, however, the 5-year survival was

100.0% for signet ring cell carcinomas, 93.3% for adenocarcinomas, and 92.0% for mucinous adenocarcinomas; these values
were not statistically significantly different from each other.
Survival by Tumor Location
Among patients in the entire cohort, 44.6% had tumors in the
right colon, 9.4% had tumors in the transverse colon, 10.4% had
tumors in the left colon, 31.6% had tumors in the sigmoid colon,
and 4.0% had tumors whose location was unknown. Among the
overall cohort, a better 5-year survival was statistically significantly associated with tumors located in the sigmoid colon
(69.8%) than with tumors located in the right colon (63.7%)
(P.001), in the transverse colon (65.0%) (P.001), and in the
left colon (65.1%) (P.001). When we further stratified each
stage (as defined by the fifth edition system) by these tumor
locations, we observed similar survival distributions in stages I,
III, and IV, but not in stage II (Fig. 5). For example, in stage III,
5-year survival was 64.3% for sigmoid lesions, 57.0% for rightcolon lesions (P.001), 57.9% for transverse (P.001), and
60.2% for left-colon lesions (P.001), whereas in stage II,

Fig. 3. Five-year survival for American Joint Committee on Cancer fifth edition
by grade. Solid bars, low-grade tumors; shaded bars, high-grade tumors. Star,
P.001, log-rank test. All statistical tests were two-sided.

Journal of the National Cancer Institute, Vol. 96, No. 19, October 6, 2004

Fig. 4. Five-year survival for American Joint Committee on Cancer fifth edition
by histologic subtype. Solid bars, adeno; open bars, mucinous; shaded bars,
signet cell. Star, P.001, log-rank test, compared with signet cell carcinoma.
All statistical tests were two-sided.

5-year survival was 83.6% and 83.7%, respectively, for rightand transverse-colon lesions, 81.5% for the left colon, and
80.7% for sigmoid lesions.
Number of Positive Lymph Nodes and Proposed
Staging System
Among patients in the entire cohort, 32.5% had positive
lymph nodes. When we used a histogram analysis of the number
of positive lymph nodes, we found that the N stage could be
stratified into the following four categories: N1 (one to three
positive lymph nodes), N2 (four or five positive lymph nodes),
N3 (six to eight positive lymph nodes), and N4 (nine or more
positive lymph nodes). We used the proposed N stages in combination with the AJCC sixth edition staging system as a new
staging system (Table 2). In this new system, stages I, IIa, IIb,
IIIa, and IIIb are the same as corresponding stages in the sixth
edition system, but the new stages IIIc, IIId, and IIIe are stratified by categories N2, N3, and N4, respectively, as defined
above. The 5-year survival by these proposed stages is 93.2% for
stage I, 84.7% for stage IIa, 72.2% for stage IIb, 83.4% for stage
IIIa, 64.1% for stage IIIb, 52.3% for stage IIIc, 43.0% for stage
IIId, 26.8% for stage IIIe, and 8.1% for stage IV. Corresponding
KaplanMeier survival curves for this system are shown in Fig. 6.

DISCUSSION
The TNM system for staging cancer was developed in the
1940s (4) and since then has become an important and dynamic
part of our cancer language. Many physicians also still continue
to use other staging systems with long-standing histories, the
Dukes staging system (5) and the AstlerColler (6) and Kirklin
(7) modifications of the Dukes staging system. For the evalua-

Fig. 5. Five-year survival for American Joint Committee on Cancer fifth edition
by tumor location. Open bars, left; diagonal-hatched bars, right; solid bars,
transverse; vertical-hatched bars, sigmoid. Star, P.001, log-rank test, compared with sigmoid colon. All statistical tests were two-sided.

Fig. 6. Survival by American Joint Committee on Cancer sixth edition staging

with proposed lymph node (N) stages. *, P values determined by the log-rank
test refers to the corresponding stage and the stage in the row above, unless
otherwise indicated. IIIa versus IIb; IIa versus IIIa; IIb versus IIIb;
NS not statistically significant. All statistical tests were two-sided.

tion of administrative data samples and/or data from cancer

registries, however, a unified language for colorectal cancer
staging is crucial. These large, population-based datasets are
valuable assets in that they provide very large data samples of
patients to study and are often population based so that they
more clearly reflect the nation as a whole. However, they are
useful for cancer research only if a clear and consistent staging
system is used (8).
We used national population-based cancer data from the
SEER cancer registry to determine the 5-year survival rates for
colon cancer with stages defined by both the AJCC fifth edition
and the revised AJCC sixth edition staging systems. The fifth
edition system separated patients with colon cancer into four
stages, with somewhat wide spacing between survival curves,
particularly around stages II and III. The sixth edition system for
colon cancer stratified stages II and III into two and three
subsets, respectively, thereby filling in some of the gaps in the
survival curves.
An interesting finding of this study is that, with the AJCC
sixth edition staging system, stage IIIa disease appears to be
associated with a statistically significantly improved survival
compared with that of stage IIb disease. This finding may reflect
the National Institutes of Health (NIH) 1990 Consensus Conference guidelines (9) for colorectal cancer that recommend adjuvant chemotherapy for stage III colon cancer patients but do not
recommend adjuvant chemotherapy for stage II patients. That is,
treatment practices that followed these guidelines may contribute to the finding that early stage III patients (i.e., stage IIIa) had
better survival rates than late stage II patients (i.e., stage IIb).
Unfortunately, we could not analyze this issue further with
SEER data because information about chemotherapy is not
available in the SEER database. Another possible explanation
for the finding is misclassification of T4N1 (stage IIIa) tumors as
T4N0 (stage IIb) tumors. If misclassification occurred, stage IIb

Journal of the National Cancer Institute, Vol. 96, No. 19, October 6, 2004

ARTICLES 1423

patients may also have had micrometastases, which, if untreated,

may have been associated with decreased survival. Alternatively, stage IIb tumors may simply be more aggressive tumors,
which have a poorer prognosis. Finally, stage IIb tumors may
not have received proper en bloc surgical resection, so that
residual disease was not excised or an incision went through the
tumor, both of which have been associated with a worse prognosis (10).
Since publication of the NIH guidelines (9), there has been
some controversy about whether stage II patients, or a certain
subset of them, should receive chemotherapy. Recent studies
have identified potential high-risk stage II patients who may
benefit from such treatment (11,12). Given our findings, patients
with stage IIb (i.e., T4N0) colon cancer appear to be one group
that may benefit from adjuvant chemotherapy.
Another aim of this study was to evaluate other factors that
could be incorporated into a new colon cancer staging system as
prognostic factors. We studied four variables (grade, histologic
subtypes, tumor location, and number of positive lymph nodes)
and applied them to data separated into stages as defined by the
AJCC fifth edition staging system. We found, similar to previous
results (13), statistically significant differences in survival between patients with high-grade tumors and patients with lowgrade tumors. These differences were observed for all stages as
defined by the AJCC fifth edition system, except stage I. The
International Union Against Cancer and the AJCC are constantly
reevaluating the effectiveness of the TNM system and of various
options to it. Grade may be a factor to investigate further for
colon cancer staging systems (14).
We also evaluated the association of histologic type of colon
cancer and tumor location with survival. We found that signet
ring cell carcinomas were statistically significantly associated
with worse survival than adenocarcinomas and mucinous adenocarcinomas. However, only 1% of the overall cohort has
signet ring cell carcinomas. Consequently, histologic subtype is
probably not a useful addition to staging systems. When we
investigated tumor location, we found that sigmoid colon tumors
were associated with statistically significantly better survival
than tumors of the right, transverse, and left colon. However, the
differences, although statistically significant, were small and are
not likely to be clinically significant.
The fourth factor we evaluated, the number of positive lymph
nodes, is already part of the TNM classification. However, our
proposed scheme would use four N stages rather than two. When
the survival curves generated by use of this proposed system
were examined, we found that the curves provided more detailed
(i.e., stratified) prognostic information than did current systems.
Furthermore, because the AJCC and the International Union
Against Cancer have recommended that a minimum of 12 lymph
nodes be examined for proper staging of stage III patients
(8,1517), it is not unreasonable to stratify the N stage into the
four stages that we propose, which stratify up to nine or more
positive lymph nodes. Although the increased stratification of a
staging system may not necessarily lead to different treatments
(e.g., stage IIIa versus IIIb), it does provide important prognostic
information that may help future clinical trials, improve the
evidence-based literature, and provide more information to
health care providers and patients. Several other potential staging stratifications have been created for stratifying stage III
tumors (8), and ours is another option.
1424 ARTICLES

Although this is a large population-based study evaluating

survival and stages for colon cancer defined by the AJCC fifth
edition staging system compared with stages defined by the new
sixth edition system, it has several potential limitations. First,
although the SEER registry maintains stringent quality control
measures to prevent coding errors, miscoding and inaccurate
data may be present. SEER, however, does uphold several
measures to ensure accuracy and maintains the highest level of
certification of data quality and completeness, as reported by the
North American Association of Central Cancer Registries (18).
Second, the SEER registry does not contain all clinically relevant treatment data, the most important for our study being the
receipt of chemotherapy. Data on treatment would be helpful in
analyzing and understanding some of our results.
Nevertheless, our use of the population-based SEER registry
also provided benefits to our study. First, we were able to
analyze data from 119 363 patients diagnosed with colon adenocarcinoma over a 10-year period. Second, SEER is a national
sample representing approximately 14% of the U.S. population
(3). As such, our finding should generalize to the overall population of the United States.
In summary, this work is one of the largest studies to date
evaluating the new AJCC sixth edition staging system for colon
cancer. We found that the new AJCC sixth edition staging
system for colon cancer results in better estimates of survival by
providing more substages than the fifth edition system. In addition, we demonstrate that other factors may be used to improve
colon cancer staging, such as tumor grade and additional stratification of the number of positive lymph nodes. It is unclear
why stage IIIa patients appear to have better survival than that of
stage IIb, but this finding brings up the important and debated
question of whether patients with stage II colon cancer should
receive adjuvant therapy. Further studies to determine whether
stage IIb patients would benefit from receiving chemotherapy
routinely may be indicated.

REFERENCES
(1) American Joint Committee on Cancer. Missions and objectives. Available
at: http://www.cancerstaging.org. [Last accessed: August 6, 2004.]
(2) AJCC cancer staging manual, 6th ed. New York (NY): Springer, 2002.
(3) Surveillance, Epidemiology, and End Results (SEER) Program Public-Use
Data (1973-2000), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2002, based on
the November 2001 submission. Available at: http://www.seer.cancer.gov.
[Last accessed: August 6, 2004.]
(4) Denoix PF, Schwartz D. [General rules for classification of cancers and
presentation of the therapeutic results.] Mem Acad Chir (Paris) 1959;85:
41524. French.
(5) Dukes CE, Bussey HJ. The spread of rectal cancer and its effect on
prognosis. Br J Cancer 1958;12:309 20.
(6) Astler VB, Coller FA. The prognostic significance of direct extension of
carcinoma of the colon and rectum. Ann Surg 1954;139:846 52.
(7) Kirklin JW, Dockerty MD, Waugh JW. The role of the peritoneal reflection
in the prognosis of carcinoma of the rectum and sigmoid colon. Surg
Gynecol Obstet 1949;88:326 31.
(8) Greene FL. TNM staging for malignancies of the digestive tract: 2003
changes and beyond. Semin Surg Oncol 2003;21:239.
(9) NIH consensus conference. Adjuvant therapy for patients with colon and
rectal cancer. JAMA 1990;264:1444 50.
(10) Nelson H, Petrelli N, Carlin A, Couture J, Fleshman J, Guillem J, et al.
Guidelines 2000 for colon and rectal cancer surgery. J Natl Cancer Inst
2001;93:58396.

Journal of the National Cancer Institute, Vol. 96, No. 19, October 6, 2004

(11) Mulcahy HE, Toner M, Patchett SE, Daly L, ODonoghue DP. Identifying
stage B colorectal cancer patients at high risk of tumor recurrence and
death. Dis Colon Rectum 1997;40:326 31.
(12) Merkel S, Wein A, Gunther K, Papadopoulos T, Hohenberger W, Hermanek P. High-risk groups of patients with stage II colon carcinoma.
Cancer 2001;92:1435 43.
(13) Compton CC, Fielding LP, Burgart LJ, Conley B, Cooper HS, Hamilton
SR, et al. Prognostic factors in colorectal cancer. College of American
Pathologists Consensus Statement 1999. Arch Pathol Lab Med 2000;124:
979 94.
(14) Gospodarowicz MK, Miller D, Groome PA, Greene FL, Logan PA, Sobin
LH. The process for continuous improvement of the TNM classification.
Cancer 2004;100:15.
(15) Wong JH, Severino R, Honnebier MB, Tom P, Namiki TS. Number of
nodes examined and staging accuracy in colorectal carcinoma. J Clin Oncol
1999;17:2896 900.
(16) Swanson RS, Compton CC, Stewart AK, Bland KI. The prognosis of T3N0
colon cancer is dependent on the number of lymph nodes examined. Ann
Surg Oncol 2003;10:6571.

(17) Goldstein NS, Sanford W, Coffey M, Layfield LJ. Lymph node recovery
from colorectal resection specimens removed for adenocarcinoma. Trends
over time and a recommendation for a minimum number of lymph nodes
to be recovered. Am J Clin Pathol 1996;106:209 16.
(18) North American Association of Central Cancer Registries. Available at:
http://www.naaccr.org. [Last accessed: August 6, 2004.]

NOTES
1

Editors note: SEER is a set of geographically defined, population-based,

central cancer registries in the United States, operated by local nonprofit
organizations under contract to the National Cancer Institute (NCI). Registry
data are submitted electronically without personal identifiers to the NCI on a
biannual basis, and the NCI makes the data available to the public for
scientific research.

Manuscript received April 23, 2004; revised June 16, 2004; accepted August
6, 2004.

Journal of the National Cancer Institute, Vol. 96, No. 19, October 6, 2004

ARTICLES 1425