Long-Term Survival in Patients
with Locally Advanced Colon
Cancer After En Bloc
ORIGINAL Pancreaticoduodenectomy
CONTRIBUTION and Colectomy
Akio Saiura, M.D.1  Junji Yamamoto, M.D.1  Masashi Ueno, M.D.1 
Rintaro Koga, M.D.1  Makoto Seki, M.D.1  Norihiro Kokudo, M.D.2
1 Department of Gastroenterological Surgery, Cancer Institute Hospital, Tokyo, Japan
2 Division of Hepato-Biliary-Pancreatic Surgery, Department of Surgery, Graduate School of Medicine,
University of Tokyo, Tokyo, Japan
PURPOSE: Surgical indications for colon cancer directly
invading the pancreas head are controversial.
METHODS: Between 1957 and 2007, a total of 12 patients
(8 men) underwent pancreaticoduodenectomy combined
with right hemicolectomy for colon cancer involving the
pancreas head.
RESULTS: Mean age was 58 (range, 3477) years. Fistula
formation was observed in five patients (41 percent)
preoperatively. Tumor involvement was duodenum only
(n=4), duodenum/pancreas (n=3), stomach/pancreas
(n=1), duodenum/stomach (n=2), duodenum/liver
(n=1), and pancreas only (n=1). Only one postoperative
death was encountered. Histologic examination showed
malignant invasion to the pancreas head in nine cases
(75 percent). Overall one-year, three-year and, five-year
survival rates after surgery were 75, 66, and 55 percent,
respectively. Five patients (41 percent) survived for more
than ten 10 years.
CONCLUSIONS: Pancreaticoduodenectomy for advanced
colon cancer invading the pancreas or duodenum
provides favorable long-term survival.
KEY WORDS: Pancreaticoduodenectomy; Colon cancer;
Direct invasion.
n the era of advanced chemotherapy, surgical resection
I remains the only potentially curative treatment for
colorectal cancer.1,2 Adequate margin of resection (R0) is
the optimal treatment for colorectal cancer. Incomplete
Supported by a Grant-in-Aid for Basic Research to Dr. Akio Saiura
from the Ministry of Education, Culture, Sports, Science and
Technology.
Reprints are not available.
Address of correspondence: Akio Saiura, M.D., Department of
Gastroenterological Surgery, Cancer Institute Hospital, 3-10-6 Ariake,
Koto-ku, Tokyo 135-8500, Japan. E-mail: akio.saiura@jfcr.or.jp
1548 DOI: 10.1007/s10350-008-9318-0  VOLUME 51: 15481551 (2008)  THE ASCRS 2008  PUBLISHED ONLINE: 3 MAY 2008
resection is now generally considered as the reason for
high rates of local recurrence and dismal prognosis.3,4
Curative resection of colorectal cancer that has invaded
adjacent structures involves en bloc resection of all
involved structures.
Right-sided colon cancer can involve the duodenum
or pancreas head. In such cases, right hemicolectomy
together with pancreaticoduodenectomy (PD) is indicated
to achieve R0 resection. Some reports have described
improved prognosis after pancreaticoduodenectomy in
patients with colorectal cancer invading the pancreas
head.57 However, long-term results remain unclear. The
present study reviewed 12 patients who underwent
combined pancreaticoduodenectomy and right hemico-
lectomy for colon cancer involving the pancreas head.
PATIENTS AND METHODS
We retrospectively reviewed the database for Cancer
Institute Hospital between January 1957 and April 2007.
During this time, a total of 3,074 patients underwent
surgery for colon cancer. Subjects comprised patients
who underwent potentially curative resection for colon
cancer directly invading the pancreas head. A total of 12
patients (8 men) underwent en bloc resection by
pancreaticoduodenectomy together with colectomy for
locally advanced colon cancer. Patients with secondary
involvement of the pancreas head by bulky lymph node
metastasis or peritoneal seeding were excluded. Clinical
data were carefully reviewed with regard to tumor
histology, histologic extent of tumor invasion, and
overall patient survival.
Statistical Analysis
Overall rate was estimated according to Kaplan-Meier
methods. The log-rank test was used to compare
significant differences. Comparisons between groups were
made by using Fishers exact test, and values of P < 0.05
were considered statistically significant. All statistical
SAIURA ET AL : PD FOR C OLON C ANCER 1549

analyses were conducted by using SPSS version 9.0 TABLE 2. Relationship between histologic type of primary and
software (SPSS, Inc., Chicago, IL). tumor extension
Histologic
type
RESULTS
Muc
Our series comprised 0.39 percent of the 3,074 patients Well or Poor P value
with colon cancer who underwent resection in our Lymph node metastasis
institute between January 1957 and April 2007 (Table 1). present (n=5) 0 5
Mean age was 58 (range, 3477) years. Sites of colon absent (n=7) 6 1 0.004**
cancer were the right colic flexure (n=9), ascending colon Malignant invasion
present (n=9) 3 6
(n=1), and transverse colon (n=2). Direct invasion to the absent (n=3) 3 0 0.231
pancreas head was confirmed histologically in nine Fisutula formation
patients, with five patients developing fistula formation present (n=5) 1 4
between the colon and duodenum (n=3) or colon and absent (n=7) 5 2 0.242
stomach (n=2). Adhesions were inflammatory in three well = well-differentiated adenocarcinoma; muc = mucinous adenocarcinoma; poor =
patients. Histologic examination revealed mucinous poorly differentiated adenocarcinoma.  * Fishers exact test.

adenocarcinoma (n=5), well-differentiated adenocarci-

noma (n=6), and poorly differentiated adenocarcinoma
(n=1). Lymph node metastases were present in five
patients. None of the patients were administered adjuvant
chemotherapy.
Table 2 shows the relationship between histologic
type of the primary tumor and tumor extension. All
patients with well-differentiated adenocarcinoma showed
no lymph node metastases. Rate of lymph node metastasis
was significantly higher in mucinous or poorly differen-
tiated adenocarcinoma (P=0.015). Mucinous or poorly
differentiated adenocarcinoma was more likely to invade
to adjacent structures with malignant invasion and fistula
formation than well-differentiated adenocarcinoma.
One hospital death occurred because of liver failure.
Overall median survival was 5.7 (range, 018) years.
Overall survival rates were 75 percent at one year, 66
percent at three years, and 55 percent at five years (Fig. 1).
Four patients died from recurrent disease, but six patients
survived for more than five years after surgery, and five
patients (41 percent) survived for more than ten years.
Table 3 shows actual numbers of survivors and
univariate analysis for overall survival. Patients with
node-positive or pancreatic invasion displayed significant-
ly poorer survival than node-negative patients. Node-
negative status or well-differentiated adenocarcinoma
seemed to indicate relatively favorable survival. Only one
patient with lymph node metastasis survived for more
than five years, and that patient (Case 9) displayed one
regional lymph node metastasis and two peripancreatic
lymph node metastases. Pancreatic invasion was con-
firmed histologically in four patients. None of the four
patients with true pancreatic invasion survived for more
than five years.

TABLE 1. Patients characteristics

Maximum Serosal Lymph node
Age Site of tumor invasion metastasis of Pathologic
Case (yr)/Sex Site of colon cancer adhesion size (cm) of tumor primary tumor Fistula findings Survival Outcome
1 59 F Right colic flexture Du, Panc 8 si Present Du muc 1 mo dead
2 50 M Right colic flexture St, Panc 5.5 si Absent St well 3 yr 1 mo dead
3 77 F Right colic flexture Du 7 si Absent well 14 yr dead
4 51 F Right colic flexture *
Du* 6.5 ss Absent well 18 yr alive
5 49 M Right colic flexture Du, Panc 11 si Present muc 3 yr 3 mo dead
6 55 M Right colic flexture *
Du* 7 se Absent well 10 yr alive
7 34 M Transverse colon St, Du 9 si Absent St muc 11 yr 9 mo alive
8 68 M Right colic flexture Du, Panc 5 si Present muc 10 mo dead
9 61 M Right colic flexture Du 10 si present Du muc 12 yr 10 mo alive
10 69 M Ascending colon Du, liver 8 si Present Du poor 1 yr 3 mo dead
11 62 M Right colic flexture *
Panc* 5 ss Absent well 6 yr 6 mo alive
12 62 F Transverse colon St, Du 8 si Absent well 3 mo alive
si = tumor invading to adjacent structure; se = tumor exposed beyond serosa; ss = tumor within subserosal layer; muc = mucinous adenocarcinoma; well = well-differentiated
adenocarcinoma; poor = poorly differentiated adenocarcinoma; Du = duodenum; Panc = pancreas; St = stomach.  * Inflammatory adhesion on histologic examination.
1550 SAIURA ET AL : PD FOR C OLON C ANCER

infrequent. Two problems must be addressed: curability

. and safety.
Several reports in a small number of cases have
described favorable results for en bloc resection of colon
.
cancer involving the pancreas.6,7 However, no long-term
survivors of more than ten years have been reported
. previously. In our report, five patients survived for more
than ten years without recurrence. Long-term results after
en bloc resection of right-sided colon cancer involving the
. pancreas head seem equal to those of colorectal cancer
involving other structures or organs.
. Our data showed that patients with node-negative
status or well-differentiated adenocarcinoma experienced
favorable survival outcomes. Eisenberg et al.13 reported 58
cases of locally advanced colorectal carcinoma, and actual
disease-free survival rate was 76 percent in patients with
FIGURE 1. Cumulative overall survival for the 12 patients.
no lymph node metastases compared with 0 percent in
patients with lymph node metastases. The number of
patients in the literature is too small to reach definitive
conclusions regarding prognostic factors. The current
DISCUSSION limited review suggests that biologic characteristics of the
We described 12 patients who underwent en bloc resection of tumor contribute to prognosis. According to our study,
colorectal cancer invading the pancreas head. Of these, five mucinous or poorly differentiated adenocarcinoma seems
patients survived for more than ten years. As for local
control, all patients showed negative margins histologically. TABLE 3. Actual number of survivors and univariate analysis
Direct invasion to adjacent organs often is seen with of overall survival
colorectal cancer. All adhesions between the carcinoma and Survivors
adjacent structures should be assumed to be malignant,
1- 3- 5-
because 33 to 84 percent are malignant on histologic Variable n year year year P value**
examination.8
Overall 9 8 6
Patients who undergo margin-negative multivisceral Age (yr)
resection display the same survival as patients with no <60 6 5 5 3 0.835
adjacent organ involvement on a stage-matched basis.7,9 60 6 3 3 3
Conversely, local recurrence is frequent (70100 percent) Sex
in patients with dissection of adhesions.10,11 En bloc Female 4 2 2 2 0.764
Male 8 6 6 4
resection together with resection of adhesions should be Maximum tumor size (mm)
attempted if the operation can be performed safely and no <80 6 5 5 4 0.377
distant metastases have been identified. 80 6 3 2 2
In our series, histologic examination showed malignant Lymph node metastasis
invasion to the pancreas head in 9 of 12 patients (75 Present 5 2 2 1 0.027*
Absent 7 6 6 5
percent). Interestingly, all mucinous or poorly differentiated Malignant invasion
adenocarcinomas displayed malignant invasion (6/6, 100 Present 9 5 5 3 0.597
percent), but only half of well-differentiated adenocarcino- Absent 3 2 2 2
mas showed malignant invasion histologically (3/6, 50 Pancreatic invasion
percent). Mucinous or poorly differentiated adenocarcinoma Present 4 2 2 0 0.007*
Absent 8 6 6 6
of the colon is reportedly associated with a higher incidence Fistula formation
of local extension, which leads to lower curative and overall Present 5 3 3 2 0.266
resection rates.10,12 Our findings suggest that adhesions in Absent 7 5 5 4
patients with mucinous or poorly differentiated adenocarci- Period of operation
noma are more likely to indicate malignant invasion. 19571990 8 6 6 4 0.695
19912007 4 2 2 2
Right-sided colon cancer can invade directly to the Histologic type
pancreas head or stomach. Pancreaticoduodenectomy Well 6 5 5 4 0.113
together with right hemicolectomy is the most radical Muc/poor 6 3 3 2
approach to this condition. However, the feasibility of the well = well-differentiated adenocarcinoma; muc = mucinous adenocarcinoma; poor =
procedure is poorly understood, because the condition is poorly differentiated adenocarcinoma.  * Log-rank test.
SAIURA ET AL : PD FOR C OLON C ANCER
to be associated with malignant features, such as
malignant invasion and lymph node metastasis. As noted
by Eisenberg et al., lymph node metastasis predicted
dismal prognosis in our study. Our data also demonstrat-
ed poor prognosis in patients with pancreatic invasion.
Pancreatic invasion of colon cancer may have specific
effects on the ability of such aggressive procedures to
achieve cure, because none of the patients with true
pancreatic invasion survived long in this series (median
survival, 23.5 months). Thus, aggressive surgical resection
may have limited impact on survival extension in patients
with pancreatic invasion.
None of our patients were administered adjuvant
chemotherapy, because no effective drugs were available
at that time. In the modern era of advanced chemothera-
pies, adjuvant chemotherapy should be attempted in
patients with node-positive or pancreatic invasion. How-
ever, neoadjuvant chemotherapy is not feasible in most
patients with pancreas invasion because of high rates of
ileus or hemorrhagic ulcer caused by concurrent duode-
nal or stomach invasion.
Another problem is the operative risk that is associated
with PD, although this procedure has become much safer
recently. High-volume centers display mortality rates after
PD of 1 to 6 percent.1416 In our series, one patient died one
month after surgery. This case occurred 30 years ago. From
our recent data, mortality rate is <2 percent. In a high-
volume center, pancreaticoduodenectomy with right hemi-
colectomy can be performed safely with low mortality rate.
Thus, surgeons should not hesitate to perform PD in
patients with adhesion between the tumor and pancreas
head if curative resection is possible.
CONCLUSIONS
Direct invasion of right-sided colon cancer is a rare con-
dition, but en bloc resection by pancreaticoduodenectomy
offers favorable long-term survival. An aggressive surgical
approach offers long-term survival, particularly in the
absence of lymph node metastasis and pancreatic invasion.
REFERENCES
1. Alberts SR. Evolving role of chemotherapy in resected liver
metastases. J Clin Oncol 2006;24:495253.
1551
2. Benoist S, Brouquet A, Penna C, et al. Complete response
of colorectal liver metastases after chemotherapy: does it
mean cure? J Clin Oncol 2006;24:393945.
3. Quirke P, Durdey P, Dixon MF, Williams NS. Local
recurrence of rectal adenocarcinoma due to inadequate
surgical resection. Histopathological study of lateral tu-
mour spread and surgical excision. Lancet 1986;2:9969.
4. de Haas-Kock DF, Baeten CG, Jager JJ, et al. Prognostic
significance of radial margins of clearance in rectal cancer.
Br J Surg 1996;83:7815.
5. Sasson AR, Hoffman JP, Ross EA, Kagan SA, Pingpank JF,
Eisenberg BL. En bloc resection for locally advanced cancer of
the pancreas: is it worthwhile? J Gastrointest Surg 2002;6:147
57.
6. Koea JB, Conlon K, Paty PB, Guillem JG, Cohen AM.
Pancreatic or duodenal resection or both for advanced
carcinoma of the right colon: is it justified? Dis Colon
Rectum 2000;43:4605.
7. Curley SA, Evans DB, Ames FC. Resection for cure of
carcinoma of the colon directly invading the duodenum or
pancreatic head. J Am Coll Surg 1994;179:58792.
8. Orkin BA, Dozois RR, Beart RW Jr, Patterson DE,
Gunderson LL, Ilstrup DM. Extended resection for locally
advanced primary adenocarcinoma of the rectum. Dis
Colon Rectum 1989;32:28692.
9. Rowe VL, Frost DB, Huang S. Extended resection for locally
advanced colorectal carcinoma. Ann Surg Oncol 1997;4:131
6.
10. Bonfanti G, Bozzetti F, Doci R, et al. Results of extended
surgery for cancer of the rectum and sigmoid. Br J Surg
1982;69:3057.
11. McGlone TP, Bernie WA, Elliott DW. Survival following
extended operations for extracolonic invasion by colon
cancer. Arch Surg 1982;117:5959.
12. Kanemitsu Y, Kato T, Hirai T, et al. Survival after curative
resection for mucinous adenocarcinoma of the colorectum.
Dis Colon Rectum 2003;46:1607.
13. Eisenberg SB, Kraybill WG, Lopez MJ. Long-term results of
surgical resection of locally advanced colorectal carcinoma.
Surgery 1990;108:77986.
14. Edge SB, Schmieg RE Jr, Rosenlof LK, Wilhelm MC.
Pancreas cancer resection outcome in American university
centers in 19891990. Cancer 1993;71:35028.
15. Cameron JL, Riall TS, Coleman J, Belcher KA. One
thousand consecutive pancreaticoduodenectomies. Ann
Surg 2006;244:105.
16. Gordon TA, Burleyson GP, Tielsch JM, Cameron JL. The
effects of regionalization on cost and outcome for one general
high-risk surgical procedure. Ann Surg 1995;221:439.