Prognosis of Node-Positive Colon Cancer

Alfred M. Cohen, MD, Salvatore Tremiterra, MD, Frank Candela, MD,

Howard T. Thaler, PhD, and Elin R. Sigurdson, MD, PhD
The most recent American Joint Committee on Cancer/International Union Against
Cancer (AJCC/UICC) staging system subgroups patients into one to three and four
or more positive nodes. However, the Gastrointestinal Study Group and the
National Surgical Adjuvant Breast and Bowel Project divides node-positive patients
into one to four and five or greater. A Cox multi-variate retrospective analysis was
done of the overall survival of node-positive colon cancer patients with the specific
objective of determining the most appropriate subcategorization. Data on 306
patients with node-positive colon cancer who underwent potentially curative
surgery from 1970 to 1984 were analyzed retrospectively. No patient received
adjuvant chemotherapy. Also excluded were patients with synchronous resected
metastatic disease or those with rectal primaries. The median follow-up was 6
years, and the median survival for the entire group was 8.6 years. By univariate
analysis, the following were significant prognostic features: number of positive
nodes ( P < O.OOOl), degree of differentiation ( P < 0.0001), colon primary site (P
= 0.009), tumor stage (P = O.OOl), and tumor size (P < 0.0001). Lymphatic/blood
vessel invasion and a mucinous histology were not significant. By Cox multivariate
analysis the number of positive lymph nodes remained the best discriminant of
survival (P = 0.0001). The number of positive nodes was related inversely to
prognosis with the optimal dichotomization between one to three (66% 5-year
survival) and four or greater nodes (37% 5-year survival).
Cancer 67:1859-1861,1991.

American Joint Committee on

Cancer/International Union Against Cancer (AJCC/
UICC) staging system subgroups patients into one to three
and four or more positive nodes. However, the Gastrointestinal Study G r o ~ pand
~ , the
~ National Surgical Adjuvant Breast and Bowel Project4groups node positive patients into one to four and five or greater. We did a Cox
multi-variate analysis of the overall survival of node-positive colon cancer patients with the specific objective of
determining the optimum dichotomization.

adjuvant chemotherapy. Also excluded were patients with

synchronous resected metastatic disease or those with
rectal primaries. All patients underwent elective surgery.
Data on the exact tumor site in the colon, adherent contiguous organs, operative procedure, and postoperative
complications were tabulated. Pathologic data included
the tumor size (length and width) and configuration,depth
of penetration, total number of nodes resected, number
of positive nodes, degree of tumor differentiation, the
presence of mucinous or colloid histology, and vascular
or lymphatic invasion. The distinction between T3 and
T4 penetration with true mesothelial penetration5 was
unreliable on routine pathologic review unless direct invasion of a contiguous organ was documented. Our data
did not allow a distinction of the apical node.697
The Cox
proportional hazard multi-variate analysis method was
used to make statistical inferences.

HE MOST RECENT

'

Patients and Methods

Data on 306 patients with node-positive colon cancer

who underwent potentially curative surgery from 1970 to
1984 were analyzed retrospectively. No patients received
From the Colorectal Service, Department of Surgery, Memorial SloanKettering Cancer Center, New York, New York.
The authors thank H. Friedlaner-Klar for excellent data analysis.
Address for reprints: Alfred M. Cohen, MD, Memorial Sloan-Kettering
Cancer Center, 1275 York Avenue, New York, NY 10028.
Accepted for publication September 15, 1990.

Results

The median follow-up was 6 years, and the median

survival for the entire group was 8.6 years. One hundred

1859

CANCER
April 1 1991

1860

thirty-four patients died. By univariate analysis, the following were significant prognostic features: number of
positive nodes (P < 0.001), degree of differentiation (P
< 0.0001), colon primary site (P = 0.009), tumor stage
(P= 0.00 l), postoperative complications (P = 0.00 15),
and the size of the tumor (P= 0.0001). Lymphatic/blood
vessel invasion or a mucinous histology were not significant. By multi-variate Cox analysisthe number of positive
lymph nodes was the best discriminant of survival (Table
1). Although most patients with four or less positive nodes
survived 5 years, using the Cox model, the optimum single
dichotomization of node-positive patients was obtained
by dividing patients into three or less and four or greater
positive nodes. The Kaplan-Meier 5-year survival with
one to three positive nodes was 66%, and with four or
greater was 37% (Fig. I). Table 2 lists the actuarial 5-year
survival for each nodal group. The number of patients
having greater than six nodes positive was too small to
allow separate analysis.

VOl. 67

TICK MARK(!)

.oo

( 2 2 1 PTS.
( 85 PTS.

1-3 P O S I T I V E NODES
It POSITIVE NODES

143 CENSORED)
29 CENSORED)

I N D I C A T E S L A S T FOLLOW-UP

2.00

4.00

6.00

8.00

10.00

12.00

YEARS

Discussion

FIG. 1. Kaplan-Meier survival for colon cancer with 1-3 and 4 or

Multiple prognostic factors affect the survival of patients

with primary colorectal cancer. The presence of distant
metastatic disease is paramount. In patients with colon
cancer the second most important indicator is the presence
of lymph node metastases. Additional pathologic variables
identified in univariate analyses include the extent of primary tumor mural penetration, primary site in the colon,
degree of differentiation,presence of a colloid or mucinous
histology, tumor configuration, and lymphatic vessel invasion.
In addition to the presence or absence of nodal metastases, the number of nodes involved and their location in
relation to the primary tumor have been analyzed. In one
of the initial publications by Dukes,627metastasis to the
apical node was referred to as Stage C2. The 5-year survival for rectal cases with apical nodal involvement was
only 14% compared with 43% when nodal spread (C1
cases) was present adjacent to the primary tumor. This
subgrouping of node-positive cases has been used by GriTABLE1. Significant Covariates of Prognosis by Bivariate and
Multivariate Cox Proportional Hazards Model
Variable

Comments

P value

Nodal mets
T stage
Degree of differentiation
Tumor size
Colon site

1-3,4, 5,6+
1,2, 3 , 4
Length and width
Rectosigmoid/sigmoid/
proximal
Years
Yes/no

<0.0001
0.0 13
0.010
0.002
0.001

Age
Postoperative
complications

0.00 1
0.003

more positive node dichotomy.

nell el al. at Columbia-Presbyterian9and supported by

the American Society of Colon and Rectal Surgeons Staging Committee.8 Payne" indicated that 90% of patients
with five or greater nodes involved also have apical nodal
involvement. Since the presence of tumor in the apical
node is not only influenced by tumor biology but also by
the extent of the proximal lymphadenectomy,we focused
on the total number of nodes involved with metastatic
disease. This covariate is also influenced by the extent of
the surgical lymphadenectomy' I and by the technique of
the pathologic examination.'*
Many authors confirm the impact of multiple regional
lymph node spread on the prognosis of resectable colon
~ a n c e r . ~ ,The
' ~ , most
' ~ recent AJCC/UICC staging system
subgroups patients into one to three or four or more positive nodes. However, the Gastrointestinal Tumor Study
the National Surgical Adjuvant Breast/Bowel
P r ~ j e c t Jass
, ~ and colleague^'^ at St. Marks, Payne" in
Australia, and the Japanese Research Society for Cancer
TABLE
2. Actuarial 5-Year Survival and Nodal Status
No. of positive nodes
(N cases)

5-year
survival (96)

69
61
63
56
42
27

No. 7

NODE-POSITIVE
COLONCANCER

of the Colon and Rectum subcategorize node positive

patients into one to four or five or greater. Our multivariate data analysis confirms that the number of positive
nodes was related inversely to prognosis and supports the
I987 AJCC/UICC system with the optimal dichotomization of three or less positive nodes or four or greater.
In addition, our data confirm the University of Chicago
and Gastrointestinal Tumor Study Group experience that
indicate appropriate surgery cures most patients with four
or fewer positive
Clinical trials of adjuvant therapy of colon cancer must continue to be stratified by the
number of involved nodes.
REFERENCES
1. American Joint Committee on Cancer. Manual for Staging of Cancer. Philadelphia: JB Lippincott, 1987.
2. Gastrointestinal Tumor Study Group. Adjuvant therapy of colon
cancer: Results of a prospectively randomized trial. N Engl JMed 1984;
310:737-743.
3. Steinberg SM, Barkin JS, Kaplan RS ef al. Prognostic indicators
of colon tumors: The Gastrointestinal Tumor Group experience. Cancer
1986; 57:1866-1870.
4. Wolmark N, Fisher B, Wieand HS. The prognostic value of the

Cohen et al.

1861

modifications of the Dukes C class of colorectal cancer. Ann Surg 1986;

203:l 15-122.
5. Newland RC, Chapuis PH, Smyth EJ. The prognostic value of
substaging colorectal carcinoma: A prognostic study of I 117 cases with
standardized pathology. Cancer 1987; 60352-857.
6. Dukes CE. The classification of cancer of the rectum. J Pathol
1932; 35:323-332.
7. Dukes CE. Cancer of the rectum: An analysis of 1000 cases. J
Pathol 1940; 50527-539.
8. Fielding LP. Clinical-pathologic staging of large-bowel cancer: A
report of the ASCRS committee. Dis Colon Rectum 1988; 31:204-209.
9. Slanetz CA Jr, Herter FP. The large intestine. In: Haagensen CD,
Feind CR, Herter FP, eds. Lymphatics in Cancer. Philadelphia: WB
Saunders, 1972; 489-586.
10. Payne JE. International colorectal carcinoma staging and grading.
Dis Colon Rectum 1989; 32:282-285.
1 I . Enker WE, Laffer UT, Block GE. Enhanced survival of patients
with colon and rectal cancer is based upon wide anatomic resection.
Ann Surg 1979; 190:350-360.
12. Herrera-Ornelas L, Justiniano J, Castillo N, Petrelli NJ, Stulc JP,
Mittelman A. Metastases in small lymph nodes from colon cancer. Arch
Surg 1987; 122:1253-1256.
13. Fielding LP, Phillips RK, Frey JS el al. The prediction of outcome
after curative resection for large bowel cancer. Lancer 1986; 2:904-907.
14. Jass JR, Love SB, Northover JM. A new prognostic classification
of rectal cancer. Lancet 1987; 1:1303-1306.
15. Japanese Society for Research for Cancer of the Colon and Rectum. General rules for clinical and pathological studies on cancer of the
colon, rectum and anus. Jpn JSurg 1983; 13557-573.