An individual patient data meta-analysis of adjuvant therapy with uracil-tegafur (UFT) in patients with curatively resected rectal cancer.

Abstract

Uracil-Tegafur (UFT), an oral fluorinated pyrimidine chemotherapeutic agent, has been used for adjuvant chemotherapy in curatively resected colorectal cancer patients. Past trials and meta-analyses indicate that it is somewhat effective in extending survival of patients with rectal cancer. The objective of this study was to perform a reappraisal of randomised clinical trials conducted in this field. We designed an individual patient-based meta-analysis of relevant clinical trials to examine the benefit of UFT for curatively resected rectal cancer in terms of overall survival (OS), disease-free survival (DFS), and local relapse-free survival (LRFS). We analysed individual patient data of five adjuvant therapy randomised clinical trials for rectal cancer, which met the predetermined inclusion criteria. These five trials had a combined total of 2091 patients, UFT as adjuvant chemotherapy compared to surgery-alone, 5-year follow-up, intention-to-treat-based analytic strategy, and similar endpoints (OS and DFS). In a pooled analysis, UFT had significant advantage over surgery-alone in terms of both OS (hazard ratio, 0.82; 95% confidence interval (CI), 0.70-0.97; P=0.02) and DFS (hazard ratio, 0.73; 95%CI, 0.63-0.84; P<0.0001). This individual patient-based meta-analysis demonstrated that oral UFT significantly improves both OS and DFS in patients with curatively resected rectal cancer.

Full text

Page 1

An individual patient data meta-analysis of adjuvant therapy with
uracil–tegafur (UFT) in patients with curatively resected
rectal cancer
J Sakamoto*,1, C Hamada1, S Yoshida1, S Kodaira1, M Yasutomi1, T Kato1, K Oba1, H Nakazato1, S Saji1and
Y Ohashi1
1Meta-Analysis Group of the Japanese Society for Cancer of the Colon and Rectum; Secretariat, Department of Epidemiological & Clinical Research
Information Management, Kyoto University, Graduate School of Medicine, Kyoto, Japan
Uracil–Tegafur (UFT), an oral fluorinated pyrimidine chemotherapeutic agent, has been used for adjuvant chemotherapy in curatively
resected colorectal cancer patients. Past trials and meta-analyses indicate that it is somewhat effective in extending survival of patients
with rectal cancer. The objective of this study was to perform a reappraisal of randomised clinical trials conducted in this field. We
designed an individual patient-based meta-analysis of relevant clinical trials to examine the benefit of UFT for curatively resected rectal
cancer in terms of overall survival (OS), disease-free survival (DFS), and local relapse-free survival (LRFS). We analysed individual
patient data of five adjuvant therapy randomised clinical trials for rectal cancer, which met the predetermined inclusion criteria. These
five trials had a combined total of 2091 patients, UFT as adjuvant chemotherapy compared to surgery-alone, 5-year follow-up,
intention-to-treat-based analytic strategy, and similar endpoints (OS and DFS). In a pooled analysis, UFT had significant advantage
over surgery-alone in terms of both OS (hazard ratio, 0.82; 95% confidence interval (CI), 0.70–0.97; P¼0.02) and DFS (hazard ratio,
0.73; 95%CI, 0.63–0.84; Po0.0001). This individual patient-based meta-analysis demonstrated that oral UFT significantly improves
both OS and DFS in patients with curatively resected rectal cancer.
British Journal of Cancer (2007) 96, 1170–1177. doi:10.1038/sj.bjc.6603686
Published online 20 March 2007
& 2007 Cancer Research UK
www.bjcancer.com
Keywords: rectal cancer; UFT; adjuvant chemotherapy; randomised clinical trials; individual patient data meta-analysis
????????????????????????????????????????????????
Colorectal cancer accounts for 10–15% of all cancers and is the
second leading cause of cancer deaths in developed countries
(Pisani et al, 1993). In Japan alone, nearly 56000 new cases are
diagnosed and this disease causes 36000 deaths every year
(Statistics and information department, Ministry of Health and
Welfare, 1996). Surgical treatment is the primary management of
colorectal cancers, with 75–80% of the patients being operable at
the time of diagnosis (Boring et al, 1991; Vernaba et al, 1994).
However, even if a curative resection is performed, those patients
with regional lymph node involvement (Dukes’ C, Stage III) have a
40–50% 5-year survival rate.
Recently, in the field of Stage III colon cancer treatment,
adjuvant chemotherapy by 5-fluorouracil (5-FU)/levamisole was
proved to be superior to surgery-alone therapy, and then various
5-FU/leucovorin (LV) regimens were confirmed to be effective
from the results of numerous large-scale randomised trials and
from the pooled analysis of clinical trials (Wolmark et al, 1993;
International Multicentre Pooled Analysis of Colon Cancer Trials
(IMPACT) investigators, 1995; O’Connell et al, 1997). In 2004,
results from the Multicenter International Study of Oxaliplatin/5-
FU/Leucovorin in the Adjuvant Treatment of Colon Cancer
(MOSAIC) trial demonstrated that combination chemotherapy
with 5-FU/LV (de Gramont regimen) plus oxaliplatin was
significantly superior to 5-FU/LV alone (Andre ´ et al, 2004). With
regard to adjuvant chemotherapy for colon cancer, therefore, solid
evidence has been accumulated from relevant clinical trials, and
steady evolution of the new treatment modalities has been
achieved.
However, the situation is still uncertain focusing on adjuvant
therapy for rectal cancer. Despite apparently curative surgery,
rectal cancer recurs in more than 55% of the patients, including
local recurrence rates of 25% (Vernaba et al, 1994). Despite the
recommendation of the consensus conference by the National
Institute of Health (NIH consensus conference, 1990) that
concluded that adjuvant radiotherapy and chemotherapy should
be given to all patients with locally advanced rectal cancer, recent
findings by a large-scale randomised trial and meta-analysis have
failed to prove significant benefit of radiotherapy for survival
(Fisher et al, 1988; Vernaba et al, 1994). In this regard, the quest
for an effective adjuvant treatment with a robust advantage on the
Received 1 December 2006; revised 19 February 2007; accepted 19
February 2007; published online 20 March 2007
*Correspondence: Dr J Sakamoto, Young Leaders Program, Department
of Social Life Science, Nagoya University Graduate School of Medicine,
65 Tsurumaicho, Showaku, Nagoya 466-8550, Japan;
E-mail: sakamjun@med.nagoya-u.ac.jp
Previous presentation: The 41st annual meeting of American Society of
Clinical Oncology, Orlando (Proc Am Soc Clin Oncol 2005; 23: 253s)
British Journal of Cancer (2007) 96, 1170–1177
& 2007 Cancer Research UKAll rights reserved 0007– 0920/07$30.00
www.bjcancer.com
Clinical Studies

Page 2

outcome of resected rectal cancer remain an important task for
gastrointestinal oncologists.
In Japan, mesorectal excision is standard surgical procedure.
Radiotherapy is not routinely performed as adjuvant therapy.
In Japan, adjuvant therapy after resection of colorectal cancer
was developed primarily using oral fluorinated pyrimidines (O-
FPs). A meta-analysis of three old trials (Sakamoto et al, 1999) and
a more sophisticated analysis of four recent pivotal randomised
trials (Sakamoto et al, 2004) demonstrated a statistically significant
benefit of O-FPs on the outcome of colorectal cancers over surgery
alone. However, the survival benefit shown in that meta-analysis
was more pronounced in colon cancers. The risk reduction in
terms of rectal cancer was only 8% and the result of those previous
meta-analyses that analysed various types of oral fluorinated
pyrimidine clinical trials was not sufficient to show a significant
effect on survival.
Uracil–tegafur (UFT) is one of the O-FPs. In colon cancer, the
majority of recurrences occurred in the liver, whereas in rectal
cancer many recurrences occurred in the lung and locally in
addition to the liver. Treatment effect may thus differ between
colon cancer and rectal cancer. As the previous meta-analysis, two
trials of UFT in patients with rectal cancer have been reported. The
present study focused on rectal cancer, which lacked a clear-cut
survival benefit in our previous meta-analysis. Unlike oral
fluoropyrimidines such as carmofur and capecitabine, the
formulation of UFT includes a dihydropyrimidine dehydrogenase
inhibitor (Diasio, 1999), designed to enhance the bioavailability of
FU. This combination of uracil and tegafur was shown, in an
animal tumour system, to increase the anti-tumour activity
compared with tegafur alone (Ooi et al, 2001). UFT also produced
an enhanced intratumoural concentration of fluorinated pyrimi-
dine, 5–10 times greater than that achieved with Tegafur alone
(Fukunaga et al, 1987). Preclinical studies established that the
optimal molar ratio of uracil to Tegafur is 4:1, which resulted in
the highest 5-FU tumour: blood and tumour: normal tissue
partition coefficients (Kawaguchi et al, 1980). UFT has now been
clinically tested for lung cancer (Kato et al, 2004), breast cancer
(Noguchi et al, 2005), and for gastric cancer (Kinoshita et al, 2005)
in an adjuvant setting in Japan. Recently, UFT has also been tested
in Western countries, regarding its efficacy for both advanced and
curatively resected colon cancer (Carmichael et al, 2002; Douillard
et al, 2002; Lembersky et al, 2006).
Here, we present an individual patient data meta-analysis of five
centrally randomised trials recently performed in Japan to
compare rectal cancer patients treated with UFT, with the
surgery-alone control group. This meta-analysis includes data
from more than 2000 patients and therefore provides a more
reliable assessment of the effect of UFT on the survival, disease-
free survival (DFS), and local relapse-free survival (LRFS) of the
patients with rectal cancer than is available from any of the
individual studies.
PATIENTS AND METHODS
Selection of trials
Trials that randomly assigned patients to either long-term (12
months) administration of UFT or surgery-alone treatment after
curative resection of rectal cancer were eligible for meta-analysis.
The randomisation technique used in these trials was the
centralised randomisation that precluded the possibility of prior
knowledge of the treatment to be allocated.
Five relevant trials identified as Japanese Foundation for Multi-
disciplinary Treatment of Cancer (JFMC) 7-1 (Kodaira et al, 1998),
JFMC15-1, JFMC15-2 (Watanabe et al, 2004), Tokai Adjuvant
Chemotherapy Study Group for Colorectal Cancer (TAC–CR) (Kato
et al, 2002), and National Surgical Adjuvant Study of Colorectal
Cancer (NSAS-CC) (Akasu et al, 2006) were included in the meta-
analysis involving a total of 2091 patients. In trials JFMC7-1, JFMC15-
1, and JFMC 15-2, patients who were randomly assigned to the
experimental group received intravenous mitomycin C (6mgm?2) at
1 week and once monthly for 6 months. In the JFMC15-1 and 15-2
trials, patients who were randomly assigned to the experimental
group additionally received an induction course of intravenous 5-FU
(250mgdaily?1) during 7 postoperative days (Table 1).
Protocol and data collection for the meta-analysis
In December 2003, a protocol for the meta-analysis, describing the
study rationale, statistical methods, and guidelines for publication,
was distributed to the principal investigators of the five trials.
Investigators were asked to provide individual data for every
randomised patient, whether eligible or not, assessable or not, and
Table 1Details of the randomized controlled trials included in the individual patient data meta-analysis
Category JFMC7-1JFMC15-1 JFMC15-2 TAC-CRNSAS-CC Total
Additional chemotherapy
Radiotherapy
UFT dose/day
Period
Dates of accrual
No. of patients
Duration of accrual, months
Mitomycin C
None
400mg
12 months
1986–1988
834
35
Mitomycin C+FU IV
None
400mg
12 months
1989
447
24
Mitomycin C+FU IV
None
400mg
12 months
1990
391
24
None
None
400mg
24 months
1991–1994
143
36
None
None
600mga
12 months
1996–2001
276
54
—
—
—
—
—
—
—
Sex, No. of patients (male–female ratio)
Male
Female
521 (62.4%)
313 (37.6%)
260 (58.1%)
187 (41.9%)
244 (62.4%)
147 (37.6%)
93 (65.0%)
50 (35.0%)
167 (60.5%)
109 (39.5%)
1285 (61.4%)
806 (38.9%)
Duke’s stage, No. of patients
A
B
C
135
326
373
67
175
205
62
139
189
12
53
78
0
0
276
693
1121 276
Median age
Upper age limit, years
57
70
60
75
59
75
62
75
58
75
58
—
JFMC¼Japanese Foundation for Multidisciplinary Treatment of Cancer; NSAS-CC¼National Surgical Adjuvant Study of Colorectal Cancer; TAC–CR¼Tokai Adjuvant
Chemotherapy for Colorectal Cancer; UFT¼Uracil–Tegafur.a400mgm?2day?1for 5 days every 7 days.
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1171
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies

Page 3

properly followed up or not. Items requested for every patient
were as follows: patient identification, date of surgery, eligibility,
allocated treatment by random assignment, age, sex, primary
tumour site, Dukes’ stage, induction chemotherapy, dates of
recurrence, death, or last visit. Disease-free survival was calculated
from the date of surgery to the date of recurrence, second primary
cancer or death, whichever occurred first. Survival was calculated
from the date of surgery to the date of death, regardless of the
cause of death. Local relapse-free survival was calculated from the
date of surgery to the date of local recurrence. Data from patients
with only distant recurrence and those who were died without
recurrence were censored. Patients enrolled in these trials had
been followed up for 5–7 years. Toxicity data were not collected,
because detailed analysis of side effects can be found in the
published reports of the individual trials (Kodaira et al, 1998; Kato
et al, 2002; Watanabe et al, 2004; Akasu et al, 2006).
All investigators and the Clinical Trial Committee of all the trials
agreed to join in the meta-analysis. Individual patient data were
received by the independent secretariat by February 2004 and
October 2006.
Pretreatment patient characteristics
All 2091 patients had curatively resected rectal cancer without
evidence of distant metastasis by diagnostic imaging criteria or by
macroscopic examination of the abdominal organs during surgery.
Patients with severe postoperative complications were excluded
from all trials, as were patients with any previous chemotherapy or
radiotherapy or with a synchronous or metachronous second
cancer. Median patient age was 61 years at the time of random
assignment. The male/female ratio was approximately 3:2.
Performance status was less than 2 on the Japan Clinical Oncology
Group scale for all patients.
Statistical analysis
The method used for the meta-analysis and the format for the
presentation of the results have been described in detail elsewhere
(Advanced Colorectal Cancer Meta-Analysis Project, 1992). All
analyses were based on individual patient data. Treatment effects
on DFS, LRFS, and survival were first estimated within each trial
and then combined using classical meta-analytic methods (Color-
ectal Cancer Collaborative Group, 2001). Treatment effects were
displayed as hazard ratios. These ratios were estimated by
univariate Cox’s proportional model as relative risks of having
an event in the UFT group as compared with having the same
event in the surgery-alone control group. A ratio less than unity
indicates benefit from UFT, and this benefit is statistically
significant when the 95% confidence interval (CI) of the ratio
does not include unity. The overall effect of treatment was assessed
through a w1
a w4
analyses were carried out to determine which of the following
prognostic features, if any, were predictive of the treatment effect:
Dukes’ stage (A vs B vs C), sex (male vs female), and age (three
groups of increasing age). Tests for interaction were applied to
detect departures from the homogeneity of treatment effects.
Multivariate analyses were performed with the use of the Cox
proportional hazards regression model for DFS, LRFS, and survival
to assess the robustness of the observed effects to adjustments for
important covariates and the magnitude of interaction between
treatment effect and covariate (Advanced Colorectal Cancer Meta-
Analysis Project, 1992). All P-values resulted from use of two-sided
statistical tests. The significance level was set at 5% for all tests.
2d.f. and the heterogeneity between five trials through
2d.f. (Colorectal Cancer Collaborative Group, 2001). Additional
RESULTS
Survival
Survival hazard ratios for all the trials are presented in Figure 1.
The overall hazard ratio was 0.82 (95% CI, 0.70–0.97; P¼0.02)
with no significant heterogeneity between the treatment effects in
different trials (w4
significant effect on survival of curatively resected rectal cancers
with a 5-year survival benefit of approximately 5%.
Figure 2 shows the breakdown of the survival hazard ratio
stratified by various patient characteristics. There was a slight
trend toward larger treatment benefits in earlier Dukes’ stages
(Hazard ratio; Dukes’ A¼0.60, Dukes’ B¼0.79, Dukes’ C¼0.86)
but heterogeneity tests did not show any significant difference
(w2
difference in sex (w1
for interaction¼0.22; P¼0.898).
Figure 3 shows survival curves by treatment and disease stage.
These curves confirm the hazard ratio analysis shown in Figure 2
and point to favourable effects of UFT in all Dukes’ stages.
2for heterogeneity¼4.31; P¼0.37). UFT showed
2¼1.41; P¼0.495). There was no statistically significant
2for interaction¼1.62; P¼0.204) or age (w2
2
Disease-free survival
Disease-free survival hazard ratios are presented in Figure 4 for all
the trials. These figure show a somewhat larger effect of treatment
on DFS than on survival, with an overall DFS ratio of 0.73 (95%CI,
Study
JFMC
Control
Hazard Ratio
(Treatment / Control)
Total
Test for heterogeneity
P = 0.37
No. Events / No. Entered
Treatment
HR; 0.82
Treatment better
Treatment effect P = 0.02
Control better
0 12
7–1
15–1
15–2
TAC-CR
NSAS-CC
124 / 416
68 / 218
63 / 269
15 / 72
20 / 140
290 /1115
143 / 418
77 / 229
32 / 122
22 / 71
37 / 136
311 / 976
O-E
–8.9
–3.5
–2.6
–3.8
–9.5
–28.5
V
66.7
36.2
20.3
9.2
14.2
146.6
Hazard
Redn
13%
9%
12%
34%
50%
18%
Figure 1
Expected number of events; V¼variance of (O–E); Hazard Redn¼hazard reduction; SE¼standard error of hazard reduction).
Survival hazard ratios by individual trial (Abbreviations: O/N¼observed number of events/number of patients; O–E¼Observed minus
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1172
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies

Page 4

0.63–0.84; Po0.0001) with a 5-year DFS benefit of 9.7%, but
demonstrating some heterogeneity among the treatment effects in
different trials (w4
ally, random effect model assuming the variation between trials
was applied. The results of the random effect model still revealed
highly significant differences owing to the relatively high effect in
TAC–CR and NSAS-CC trials.
Figure 5 lists the DFS hazard ratios by various patient and
treatment characteristics.
2for heterogeneity¼7.85; P¼0.097). Addition-
Figure 6 shows DFS curves by treatment and disease stage. These
curves again point to benefits of UFT in Dukes’ A, B and C stages.
Roughly identical effect extended across all Dukes’ stages: the DFS
benefits at 5 years in terms of risk reduction were 0.42, 0.33, 0.23.
Local relapse free survival
The overall hazard ratio was 0.68 (95%CI, 0.53–0.87; P¼0.0026),
and demonstrating some heterogeneity among the treatment
Control
Hazard Ratio
(Treatment / Control)
Treatment better
Treatment
Hazard
Redn
O-EV
01
2
Control better
10 / 148
66 / 377
213 / 589
16 / 128
73 / 316
222 / 532
–3.9
–9.5
–17.2
6.5
34.5
108.3
Test for interaction ?2=1.41 P=0.495
2
Stage
Dukes A
Dukes B
Dukes C
40%
21%
14%
Test for interaction ?2=1.62 P=0.204
1
Sex
Male
Female
192 / 681
98 / 434
195 / 604
116 / 372
–12.4
–19.2
96.4
53.0
10%
29%
Test for interaction ?2=0.22 P=0.898
2
Survival age
<54
55–64
65+
103 / 401
114 / 439
73 / 275
104 / 328
125 / 382
82 / 266
–11.3
–14.9
–5.8
51.2
59.4
38.7
18%
21%
12%
Test for interaction ?2=3.78 P=0.151
2
Induction
mit
mit+5FUiv
none
124 / 416
131 / 487
35 / 212
143 / 418
109 / 351
59 / 207
–9.0
–8.1
–13.5
66.7
58.3
23.5
13%
13%
44%
No. Events / No. Entered
Figure 2Survival hazard ratios by patient and treatment characteristics (Abbreviations as in Figure 1).
304
236
357
299
316
258
388
339
328
273
438
374
343
291
494
442
364
310
560
507
377
316
589
532
Treatment
Control
Treatment
Control
Dukes B
Dukes C
Treatment
Control
No. at risk
147
125
148
128
137
114
139
118
145
121
148
125
Dukes A
Years
01234
5
Survival
93.2%
87.5%
82.4%
76.8%
64.1%
Treatment
Control
1.00
0
0.50
0.75
0.25
Dukes A
59.2%
Dukes B
Dukes C
Figure 3 Survival curves by tumour stage and by treatment.
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1173
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies

Page 5

effects in different trials (w4
UFT also showed significant effect on LRFS of curatively resected
rectal cancers.
2for heterogeneity¼8.82; P¼0.0658).
DISCUSSION
Extensive preclinical and clinical research led to the optimisation
of 5-FU administration, with 5-FU bolus in combination with LV
as standard therapy both in metastatic disease (Advanced Colo-
rectal Cancer Meta-Analysis Project, 1992) and after curative
resection of Stage III (Dukes’ C) colon cancer (International
Multicentre Pooled Analysis of Colon Cancer Trials (IMPACT)
investigators, 1995; O’Connell et al, 1997; Wolmark et al, 1999).
However, the toxicity of bolus 5-FU/LV regimen, especially the risk
of haematologic toxicity and mucositis, could not have been
negligible.
Continuous-infusion 5-FU modulated by LV, utilised mostly in
European countries, showed somewhat better efficacy and
definitely better tolerance than bolus 5-FU in advanced diseases
(de Gramont et al, 1997; Meta-Analysis Group In Cancer, 1998a,b;
Schmoll et al, 2000). In the adjuvant setting, one of the conti-
nuous regimens (LV5-FU2) was shown to have low toxicity than
the bolus regimen, but no difference was shown in terms of
survival (Andre ´ et al, 2003). Recently, combination of continuous
5-FU/LV and oxaliplatin (FOLFOX 4) was demonstrated to have
significant effect on DFS, and is now considered as the standard
adjuvant regimen for colon cancer in the Western world.
Treatment better
Treatment effect P < 0.0001
Control better
Study
Hazard Ratio
(Treatment / Control)
Test for heterogeneity
P = 0.097
ControlTreatment
HR; 0.73
012
JFMC
7–1
15–1
TAC-CR
NSAS-CC
Total
15–2
142 / 416
78 / 218
17 / 72
42 / 140
364 /1115
85 / 269
180 / 418
88 / 229
38 / 71
58 / 136
411 / 976
47 / 122
O-EV
Hazard
Redn
–21.6
–4.6
–13.0
–11.8
–58.9
–7.9
80.4
41.4
13.6
24.8
187.7
27.5
24%
10%
63%
38%
27%
24%
No. Events / No. Entered
Figure 4 Disease-free survival hazard ratios by individual trial (Abbreviations as in Figure 1).
Control
Hazard Ratio
(Treatment / Control)
Treatment better Control better
Treatment
Hazard
Redn O-EV
01
2
Test for interaction ?2=1.30 P=0.522
2
Stage
Dukes A
Dukes B
Dukes C
14 / 148
86 / 377
263 / 589
21 / 128
106 / 316
284 / 532
–5.0
–20.2
–37.3
8.7
47.4
135.3
42%
33%
23%
Test for interaction ?2=0.67 P=0.414
1
Sex
Male
Female
234 / 681
130 / 434
261/ 604
150 / 372
–34.0
–27.5
122.8
68.8
23%
33%
Test for interaction ?2=5.77 P=0.056
2
Induction
mit
mit+5FUiv
none
142 / 416
163 / 487
59 / 212
180 / 418
135 / 351
96 / 207
24%
17%
48%
–21.6
–13.4
–24.8
80.4
71.8
38.4
Test for interaction ?2=0.00 P=0.999
2
DFS age
<54
55–64
65+
133 / 401
140 / 439
91 / 275
139 / 328
160 / 382
112 / 266
–21.4
–24.5
–16.1
66.7
74.3
50.5
26%
27%
28%
No. Events / No. Entered
Figure 5Disease-free survival hazard ratios by patient and treatment characteristics (Abbreviations as in Figure 1).
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1174
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies

Page 6

The recent development of O-FPs has therefore opened new
perspectives. Oral fluorinated pyrimidines may mimic continuous
regimens without its technical inconvenience and deterring
patients’ quality of life. In patients with advanced colorectal
cancer, the efficacy of UFT (typical and most prescribed O-FP)
plus oral LV (Carmichael et al, 2002; Douillard et al, 2002) or of
capecitabine alone (Hoff et al, 2001; Van Cutsem et al, 2001) seems
comparable in terms of the efficacy with significantly less
significant severe haematologic toxicities and/or stomatitis. The
risk of severe hand-foot syndrome is lower in UFT than with
capecitabine, but the risk of severe diarrhoea and other gastro-
intestinal symptoms is higher in UFT and in UFT/oral LV
treatment for Western patients.
In Japan, UFT have been administered for many years especially
for patients with curatively resected colorectal cancers. For some
unknown reason, severe gastrointestinal toxicities are much less
frequent in Japanese patients, and patients usually prefer oral
chemotherapy especially in an adjuvant setting (Borner et al,
2002).
Furthermore, with regard to rectal cancer, it is a difficult
objective for a clinical trial to accrue enough patients, compared to
colon cancer, and despite the fact that several attempts of
determining a standard adjuvant treatment for rectal cancer,
almost no clinical trial has succeeded in showing a relevant
survival benefit of adjuvant treatment, except one with preopera-
tive radiotherapy (Swedish Rectal Cancer Trial, 1997).
In this context, several Japanese groups conducted randomised
clinical trials comparing UFT with surgery alone for curatively
resected rectal cancers. Five such trials were identified after a
meticulous search, and are included in the present meta-analysis.
This meta-analysis was restricted to trials that had been
randomised centrally and from which no patient had been
excluded for any reason. It represents the largest series of properly
randomly assigned patients receiving the single oral adjuvant O-FP
agent, that is, UFT, for rectal cancer comparing with patients
receiving no therapy after curative tumour resection.
This meta-analysis found a statistically significant benefit of
UFT with regard to overall survival (OS) (hazard ratio¼0.82;
P¼0.02) as well as DFS (hazard ratio¼0.73; Po0.0001), and LRFS
(hazard ratio¼0.68; P¼0.0026). As can seen by comparing the
data in Figures 1 and 4, the data from the NSAS-CC and TAC–CR
study show benefits that are, apparently, larger than the others.
As shown in Table 1, the dosage and duration of treatment
with UFT in the NSAS-CC and TAC–CR trials differed from
those in the other three trials; the dose intensity of UFT was higher
in the former two trials. Several studies have reported that a high-
dose intensity of UFT improves survival in patients given
postoperative adjuvant chemotherapy for gastric cancer (Sugimachi
et al, 1997; Danno et al, 2001). The higher dose intensity of UFT in
the NSAS-CC and TAC–CR trials may have influenced the
outcomes.
Most of the Japanese rectal cancer patients did not receive pre-
or postoperative radiotherapy in any of the trials. Although
radiotherapy has been considered one of the standard adjuvant
treatments in the Western countries, significant survival benefit
has not been shown with reproducibility (Wolmark et al, 2000;
Colorectal Cancer Collaborative Group, 2001). The ostensible
advantage of adjuvant radiotherapy is to decrease local recurrence
of rectal cancers. As compared with postoperative chemo-
radiotherapy, preoperative chemoradiotherapy does not improve
OS, but inhibits local recurrence and reduces toxicity (Sauer et al,
2004). In our study, however, LRFS was also significantly better in
the UFT group compared to surgery alone group. As far as our
results are concerned, UFT might also be useful in preventing local
recurrence in Japanese patients who usually do not receive
radiotherapy in an adjuvant setting.
Also, there is still a debate whether adjuvant chemotherapy for
early stage rectal cancer is feasible (Buyse and Piedbois, 2001). In
terms of numbers needed to treat, these benefits imply that
approximately 20 patients need to be treated for one more patient
to survive 5 years, and approximately 10 to be treated for one fewer
patient to suffer a cancer recurrence within 5 years, regardless of
disease stage. Our results show that the therapy is beneficial in
Stage II patients not only Stage III patients with nodal involvement
(Mamounas et al, 1999; Gray et al, 2004). As for early stage disease,
further investigations are needed to assess potential benefits of
treatment because events were infrequent and hazard ratios were
small.
Regardless of the disease stage and patient background
characteristics, there is a need for further trials involving UFT
and new agents that are effective in advanced disease, such as
irinotecan, oxaliplatin, and monoclonal antibodies.
282
206
298
227
291
217
320
251
303
237
354
277
315
250
403
323
347
280
487
396
377
316
589
532
Treatment
Control
Treatment
Control
Dukes B
Dukes C
Treatment
Control
No. at risk
143
119
148
128
132
108
134
111
140
115
147
124
Dukes A
Years
0
1234
5
Disease-free survival
90.5%
83.5%
77.1%
Treatment
Control
1.00
0
0.50
0.75
0.25
Dukes A
66.4%
55.0%
46.5%
Dukes B
Dukes C
Figure 6Disease-free survival curves by tumour stage and by treatment.
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1175
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies

Page 7

ACKNOWLEDGEMENTS
This work is supported in part, by non-profit organization
Epidemiological & Clinical Research Information Network (ECRIN),
Kyoto University EBM Collaborating Center, and grant-in aid from
Japanese Society for Cancer of the Colon and Rectum, from
Japanese Society for Strategies of Cancer Therapy, and from Osaka
Cancer Foundation. The authors are indebted to Professor J Patrick
Barron of the International Medical Communications Center of
Tokyo Medical University for his review of this article.
REFERENCES
Advanced Colorectal Cancer Meta-Analysis Project (1992) Modulation of
fluorouracil by leucovorin in patients with advanced colorectal cancer:
evidence in terms of response rate. J Clin Oncol 10: 896–903
Akasu T, Moriya Y, Ohashi Y, Yoshida S, Shirao K, Kodaira S, NSAS-CC
Group (2006) Adjuvant chemotherapy with uracil-tegafur for patholo-
gical stage III rectal cancer after mesorectal excision with selective lateral
pelvic lymphadenectomy: a multicenter randomized controlled trial. Jpn
J Clin Oncol 36: 237–244
Andre ´ T, Boni C, Mounedji-Boudiaf L, Navarro M, Tabernero J, Hickish T,
Topham C, Zaninelli M, Clingan P, Bridgewater J, Tabah-Fisch I, de
Gramont A (2004) Multicenter International Study of Oxaliplatin/5-
Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer
(MOSAIC) Investigators. Oxaliplatin, fluorouracil, and leucovorin as
adjuvant treatment for colon cancer. N Engl J Med 350: 2343–2351
Andre ´ T, Colin P, Louvet C, Gamelin E, Bouche O, Achille E, Colbert N,
Boaziz C, Piedbois P, Tubiana-Mathieu N, Boutan-Laroze A, Flesch M,
Buyse M, de Gramont A (2003) Bi-monthly regimen of 5-fluorouracil and
leucovorin (LV5FU2) as adjuvant therapy in stage II and III colon cancer:
4-year results of a randomised trial. J Clin Oncol 21: 2896–2903
Boring CC, Squires TS, Tong T (1991) Cancer statistics. CA Cancer J Clin
41: 19–51
Borner MM, Schoffski P, de Wit R, Caponigro F, Comella G, Sulkes A,
Greim G, Peters GJ, van der Born K, Wanders J, de Boer RF, Martin C,
Fumoleau P (2002) Patient preference and pharmacokinetics of oral
modulated UFT vs intravenous fluorouracil leucovorin: a randomised
crossover trial in advanced colorectal cancer. Eur J Cancer 38: 349–358
Buyse M, Piedbois P (2001) Should Dukes’ B patients receive adjuvant
chemotherapy? A statistical perspective. Seminars Oncol 28(Suppl 1):
20–24
Carmichael J, Popiela T, Radstone D, Falk S, Borner M, Oza A, Skovsgaard
T, Munier S, Martin C (2002) Randomized comparative study of tegafur/
uracil and oral leucovorin vs parenteral fluorouracil and leucovorin in
patients with previously untreated metastatic colorectal cancer. J Clin
Oncol 20: 3617–3627
Colorectal Cancer Collaborative Group (2001) Adjuvant radiotherapy for
rectal cancer: a systematic overview of 8,507 patients form 22
randomised trials. Lancet 358: 1291–1304
Danno M, Shiroto H, Kunii Y, Ishibiki K, Yamamura Y, Sakamoto Y,
Tamura Y, Kunitomo K, Kumashiro R (2001) Study on the intensity of
MMC and UFT in postoperative adjuvant chemotherapy for gastric
cancer-study report of JFMTC Study No. 10. Jpn J Cancer Chemother 28:
195–203
de Gramont A, Bosset JF, Milan C, Rougier P, Bouche O, Etienne PL,
Morvan F, Louvet C, Guillot T, Francois E, Bedenne L (1997)
Randomized trial comparing monthly low-dose leucovorin and fluor-
ouracil bolus with bimonthly high-dose leucovorin and fluorouracil
bolus plus continuous infusion for advanced colorectal cancer. J Clin
Oncol 15: 808–815
Diasio RB (1999) Clinical implications of dihydropyrimidine dehydro-
genase inhibition. Oncology 13: 17–21
Douillard JY, Hoff PM, Skillings JR, Eisenberg P, Davidson N, Harper P,
Vincent MD, Lembersky BC, Thompson S, Maniero A, Benner SE (2002)
Multicenter phase III study of uracil/tegafur and oral leucovorin vs
fluorouracil and leucovorin in patients with previously untreated
metastatic colorectal cancer. J Clin Oncol 20: 3605–3616
Fisher B, Wolmark N, Rockette H, Redmond C, Deutsch M, Wickerham DL,
Fisher ER, Caplan R, Jones J, Lerner H, Gordon P, Feldman M, Cruz A,
Legault-Poisson S, Wexler M, Lawrence W, Robidoux A, Other NSABP
Investigators (1988) Postoperative adjuvant chemotherapy or radiation
therapy for rectal cancer: results from NSABP protocol R-01. J Natl
Cancer Inst 80: 21–29
Fukunaga H, Katsumi M, Aoki Y, Oka S, Konishi R, Yukawa H, Kawaguchi
T, Tamaki Y, Takifuji K (1987) 5-FU concentration in tumor tissue and
the antitumor effect in patients with gastric cancer after oral adminis-
tration of UFT. Jpn J Cancer Chemother 14: 2735–2739 (in Japanese)
Gray RG, Barnwell J, Hills R, McConkey C, Williams N, Kerr D,
for the QUASAR Collaborative Group (2004) QUASAR: A randomized
study of adjuvant chemotherapy (CT) vs observation including
3238 colorectal cancer patients. Proc Am Soc Clin Oncol 23: 246 (abstr
3501)
Hoff PM, Ansari R, Batist G, Cox J, Kocha W, Kuperminc M, Maroun J,
Walde D, Weaver C, Harrison E, Burger HU, Osterwalder B, Wong AO,
Wong R (2001) Comparison of oral capecitabine vs intravenous
fluorouracil plus leucovorin as first-line treatment in 605 patients with
metastatic colorectal cancer: results of a randomized phase III study.
J Clin Oncol 19: 2282–2292
International Multicentre Pooled Analysis of Colon Cancer Trials
(IMPACT) investigators (1995) Efficacy of adjuvant fluorouracil and
folinic acid in colon cancer. Lancet 345: 939–944
Kato H, Ichinose Y, Ohta M, Hata E, Tsubota N, Tada H, Watanabe Y,
Wada H, Tsuboi M, Hamajima N, Ohta M, Japan Lung Cancer Research
Group on Postsurgical Adjuvant Chemotherapy (2004) A randomized
trial of adjuvant chemotherapy with uracil-tegafur for adenocarcinoma
of the lung. N Engl J Med 350: 1713–1721
Kato T, Ohashi Y, Nakazato H, Koike S, Saji S, Suzuki H, Takagi H, Nimura
Y, Hasumi A, Baba S, Manabe T, Maruta M, Miura K, Yamaguchi A
(2002) Efficacy of oral UFT as adjuvant chemotherapy to curative
resection of colorectal cancer: multicenter prospective randomized trial.
Langenbecks Arch Surg 386: 575–581
Kawaguchi Y, Nagayama S, Masuda H, Yasuda A (1980) Studies on the
metabolism of 1-(2-tetrahydrofuryl)-5-fluorouracil and uracil co-admi-
nistered orally to tumor- bearing rats. Gann 71: 889–899
Kinoshita T, Nakajima T, Ohashi Y, National Surgical Adjuvant Study
Group of Gastric Cancer (2005) Survival benefit of adjuvant chemother-
apy with Uracil-tegafur (UFT) for serosa negative advanced gastric
cancer (AGC): Results of a randomized trial by National Surgical
Adjuvant Study of Gastric Cancer. Proc Am Soc Clin Oncol 23S: 313s
(abstr 4021)
Kodaira S, Kikuchi K, Yasutomi M, Takahashi T, Hojo K, Kato T, Tominaga
T, Kunii Y (1998) Postoperative adjuvant chemotherapy with mitomycin
C and UFT for curatively resected rectal cancer. Results from the
Cooperative Project No. 7 Group of the Japanese Foundation for
Multidisciplinary Treatment of Cancer. Int J Clin Oncol 3: 357–364
Lembersky BC, Wieand HS, Petrelli NJ, O’Connell MJ, Colangelo LH, Smith
RE, Seay TE, Giguere JK, Marshall ME, Jacobs AD, Colman LK, Soran A,
Yothers G, Wolmark N (2006) Oral uracil and tegafur plus leucovorin
compared with intravenous fluorouracil and leucovorin in stage II and
III carcinoma of the colon: results from National Surgical Adjuvant
Breast and Bowel Project Protocol C-06. J Clin Oncol 24: 2059–2064
Mamounas E, Wieand S, Wolmark N, Bear HD, Atkins JN, Song K, Jones J,
Rockette H (1999) Comparative efficacy of adjuvant chemotherapy in
patients with Dukes’ B vs Dukes’ C colon cancer: results from four
National Surgical Adjuvant Breast and Bowel Project adjuvant studies (C-
01, C-02, C-03, and C-04). J Clin Oncol 17: 1349–1355
Meta-Analysis Group In Cancer (1998a) Toxicity of 5-fluorouracil in
patients with advanced colorectal cancer: effect of administration
schedule and prognostic factors. J Clin Oncol 16: 1–6
Meta-Analysis Group In Cancer (1998b) Efficacy of intravenous continuous
infusion of fluorouracil compared with bolus administration in advanced
colorectal cancer. J Clin Oncol 16: 301–308
NIH Consensus Conference (1990) Adjuvant therapy for patients with
colon and rectal cancer. JAMA 264: 1444–1450
Noguchi S, Koyama H, Uchino J, Abe R, Miura S, Sugimachi K, Akazawa K,
Abe O (2005) Postoperative adjuvant therapy with tamoxifen, tegafur
plus uracil, or both in women with node-negative breast cancer: a pooled
analysis of six randomized controlled trials. J Clin Oncol 23: 2172–2184
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1176
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies

Page 8

O’Connell MJ, Mailliard JA, Kahn MJ, Macdonald JS, Haller DG, Mayer RJ,
Wieand HS (1997) Controlled trial of fluorouracil and low-dose
leucovorin given for 6 months as postoperative adjuvant therapy for
colon cancer. J Clin Oncol 15: 246–250
Ooi A, Ohkubo T, Higashigawa M, Kawasaki H, Kakito H, Kagawa Y,
Kojima M, Sakurai M (2001) Plasma, intestine and tumor levels of 5-
fluorouracil in mice bearing L1210 ascites tumor following oral
administration of 5-fluorouracil, UFT (mixed compound of tegafur and
uracil), carmofur and 50-deoxy-5-fluorouridine. Biol Pharm Bull 24:
1329–1331
Pisani P, Parkin DM, Ferlay J (1993) Estimates of the worldwide mortality
from eighteen major cancers in 1985. Implications for prevention and
projections of future burden. Int J Cancer 55: 891–903
Sakamoto J, Hamada C, Kodaira S, Nakazato H, Ohashi Y (1999) Adjuvant
therapy with oral fluoropyrimidines as main chemotherapeutic agents
after curative resection for colorectal cancer: Individual patient data
meta-analysis of randomized trials. Jpn J Clin Oncol 29: 78–86
Sakamoto J, Ohashi Y, Hamada C, Buyse M, Burzykowski T, Piedbois P
(2004) Meta-Analysis Group of the Japanese Society for Cancer of the
Colon and Recum; Meta-Analysis Group in Cancer. Efficacy of oral
adjuvant therapy after resection of colorectal cancer: 5-year results from
three randomized trials. J Clin Oncol 22: 484–492
Sauer R, Becker H, Hohenberger W, Rodel C, Wittekind C, Fietkau R,
Martus P, Tschmelitsch J, Hager E, Hess CF, Karstens JH, Liersch T,
Schmidgerger H, Raab R, German Rectal Cancer Study Group (2004)
Preoperative vs postoperative chemoradiotherapy for rectal cancer. N
Engl J Med 351: 1731–1740
Schmoll HJ, Ko ¨hne CH, Lorenz M, Scho ¨ffski P, Voigtmann R, Bokemeyer C,
Lutz MP, Kleeberg U, Ridwelski K, Souchon R, El-Sarafi M, Weiss U,
Couvreur ML, Baron B, Wils JA (2000) Weekly 24h infusion of high-dose
(HD) 5-fluorouracil (5-FU24h) with or without folinic acid (FA) vs bolus
5-FU/FA (NCCTG/Mayo) in advanced colorectal cancer (CRC): a
randomized phase III study of the EORTC GITCCG and the AIO. Proc
Am Soc Clin Oncol 19: 241a (abstr 935)
Statistics and information department, Ministry of Health and Welfare
(1996) Deaths and death rates by sex and causes of death: Japan 1995 and
1994. In: Vital Stastics of Japan, vol 1, p 465. Health and Welfare
Statistics Association: Tokyo
Sugimachi K, Maehara Y, Ogawa M, Kakegawa T, Tomita M (1997) Dose
intensity of uracil and tegafur in postoperative chemotherapy for
patients with poorly differentiated gastric cancer. Cancer Chemother
Pharmacol 40: 233–238
Swedish Rectal Cancer Trial (1997) Improved survival with preoperative
radiotherapy in resectable rectal cancer. N Engl J Med 336: 980–987
Van Cutsem E, Twelves C, Cassidy J, Allman D, Bajetta E, Boyer M, Bugat R,
Findlay M, Frings S, Jahn M, McKendrick J, Osterwalder B, Perez-Manga
G, Rosso R, Rougier P, Schmiegel WH, Seitz JF, Thompson P, Vieitez JM,
Weitzel C, Harper P (2001) Oral capecitabine compared with with
intravenous 5-fluorouracil plus leucovorin (Mayo Clinic regimen) in
patients with metastatic colorectal cancer: results of a large phase III
study. J Clin Oncol 19: 4097–4106
Vernaba AM, Longo WE, Virgo KS, Coplin MA, Wade TP, Johnson FE
(1994) Current follow up statistics after resection of colorectal cancer.
Dis Colon Rectum 37: 573–583
Watanabe M, Nishida O, Kunii Y, Kodaira S, Takahashi T, Tominaga T,
Hojyo K, Kato T, Niimoto M, Kunitomo K, Isomoto H, Ohashi Y,
Yasutomi M (2004) Randomized controlled trial of the efficacy of
adjuvant immunochemotherapy and adjuvant chemotherapy for colo-
rectal cancer, using different combinations of the intracutaneous
streptococcal preparation OK-432 and the oral pyrimidines 1-hexylcar-
bamoyl-5-fluorouracil and uracil/tegafur. Int J Clin Oncol 9: 98–106
Wolmark N, Rockette H, Fisher B, Wickerham DL, Redmond C, Fisher ER,
Jones J, Mamounas EP, L Ore L, Petrelli NJ (1993) The benefit of
leucovorin-modulated fluorouracil as postoperative adjuvant therapy for
primary colon cancer: results from National Surgical Adjuvant Breast
and Bowel Project Protocol C-03. J Clin Oncol 11: 1879–1887
Wolmark N, Rockette H, Mamounas E, Jones J, Wieand S, Wickerham DL,
Bear HD, Atkins JN, Dimitrov NV, Glass AG, Fisher ER, Fisher B (1999)
Clinical trial to assess the relative efficacy of fluorouracil and leucovolin,
fluorouracil and levamisole, and fluorouracil, leucovorin, and levamisole
in patients with Dukes’B and C carcinoma of the colon : Results from
national surgical adjuvant breast and bowel project C-04. J Clin Oncol 17:
3553–3559
Wolmark N, Wieand HS, Hyams DM, Colangelo L, Dimitrov NV, Romond
EH, Wexler M, Prager D, Cruz Jr AB, Gordon PH, Petrelli NJ, Deutsch M,
Mamounas E, Wickerham DL, Fisher ER, Rockette H, Fisher B (2000)
Randomized trial of postoperative adjuvant chemotherapy with or
without radiotherapy for carcinoma of rectum: National Surgical
Adjuvant Breast and Bowel Project Protocol R-02. J Natl Cancer Inst
92: 388–396
Meta-analysis of UFT for rectal cancer
J Sakamoto et al
1177
British Journal of Cancer (2007) 96(8), 1170–1177
& 2007 Cancer Research UK
Clinical Studies