Annals of Oncology 11: 1323-1333, 2000.
© 2000 Kluwer Academic Publishers. Printed in the Netherlands.
Original article
Biweekly irinotecan or raltitrexed plus 6S-leucovorin and bolus 5-fluorouracil
in advanced colorectal carcinoma: A Southern Italy Cooperative Oncology
Group phase II—III randomized trial
P. Cornelia,1 F. DeVita,2 S. Mancarella,3 L. De Lucia,4 M. Biglietto,5 R. Casaretti,1
A. Farris,6 G. P. Ianniello,7 V. Lorusso,3 A. Avallone,1 G. Carteni,5 S. Leo,3
G. Catalano,2 M. De Lena3 & G. Cornelia1
'Division of Medical Oncology A, National Tumor Institute; 22nd University School of Medicine, Naples; 3Oncologic Institute, Bari;
City Hospital, Caserta; Cardarelli Hospital, Naples; 6 University School of Medicine, Sassari; 'City Hospital, Benevento, Italy
A
Summary
CI): 21%—48%) in arm A, including 3 complete responses
(CRs) and 15 partial responses (PRs), 24% (95% CI: 14%38%) in arm B, including 2 CRs and 11 PRs, and 24% (95% CI:
14%-38%), with 2 CRs and 11 PRs, in arm C. After a median
follow-up time of 62 (range 18-108) weeks, the median time to
progression was 38, 25, and 27 weeks for arm A, B, and C,
respectively. With 94 patients still alive, the one-year probability of survival was 61%, 54%, and 59%, respectively. WHO
grade 3 or 4 neutropenia and diarrhoea affected 46% and 16%,
respectively, of patients treated with CPT-11 + LFA-5-FU.
Median relative dose intensity over eight cycles (DI8) was 78%
for CPT-11 and 82% for 5-FU. Severe toxicities of TOM +
LFA-5-FU were neutropenia (16%) and diarrhoea (16%), but
median relative DI8 was 93% for TOM, and 82% for 5-FU.
Conclusions: CPT-11 + LFA-5-FU compares favorably
in term of activity and toxicity with other combination regimens including CPT-11 and continuous infusional 5-FU. The
hypothesis of a RR 15% higher than the MTX + LFA-5-FU
treatment can not be ruled out after this interim analysis. The
TOM + LFA-5-FU regimen showed a RR and a toxicity
profile very close to the MTX + LFA-5-FU combination, and
dose not deserve further evaluation in advanced colorectal
cancer patients.
Purpose: The aim of this randomised trial was to evaluate the
activity and toxicity of a biweekly regimen including 6S-leucovorin-modulated 5-fluorouracil (LFA-5-FU), combined with
either irinotecan (CPT-11 + LFA-5-FU) or raltitrexed (Tomudex®) (TOM + LFA-5-FU), in advanced colorectal cancer
patients, and to make a preliminary comparison of both these
experimental regimens with a biweekly administration of LFA5-FU modulated by methotrexate (MTX + LFA-5-FU).
Patients and methods: One hundred fifty-nine patients with
advanced colorectal carcinoma previously untreated for the
metastatic disease (34 of them previously exposed to adjuvant
5-FU) were randomly allocated to receive: CPT-11, 200 mg/m2
i.v. on day 1, followed on day 2 by LFA, 250 mg/m2 i.v.
infusion and 5-FU, 850 mg/m2 s i.v. bolus (arm A); TOM, 3
mg/m2 i.v. on day 1, followed on day 2 by LFA, 250 mg/m2 i.v.
infusion and 5-FU, 1050 mg/m2 i.v. bolus (arm B); or MTX,
750 mg/m2 i.v. on day 1, followed on day 2 by LFA, 250 mg/m2
i.v. infusion and 5-FU, 800 mg/m2 i.v. bolus (arm C). Courses
were repeated every two weeks in all arms of the trial. Response rate (RR) was evaluated after every four courses. The
sample size was defined to have an 80% power to detect a 35%
RR for each experimental treatment, and to show a difference
of at least 4% in RR with the standard treatment if the true
difference is 15% or more.
Key words: CPT-11 + LFA-5-FU, colorectal cancer, phase I I Results: The RRs were: 34% (95% confidence interval (95% III study,TOM + LFA-5-FU
Introduction
Colorectal cancer is among the most chemoresistant
tumors. Over the past 40 years, 5-fluorouracil (5-FU)
has been the most extensively used drug in post-surgical
adjuvant treatment, and it is still considered a cornerstone in the treatment of metastatic disease, although its
activity rate is only about 10% [1]. Therefore, during the
past decade, several efforts have been made in the field
of the biochemical modulation of 5-FU, in the attempt
to increase its activity by adding other compounds that
might interfere with its metabolic pathway.
Levo-folinic acid (LFA) has been extensively studied
as a modulator of 5-FU. Randomized phase III studies
comparing 5-FU plus FA to 5-FU alone have been
carried out throughout the world; a meta-analysis of
nine of these trials highlighted the significant overall
advantage of the combination in terms of response rate
(23% vs. 11%, P < 0.001), but was unable to show a
survival advantage [2].
During the last decade, there has been increasing
evidence from preclinical observations that methotrexate
(MTX) might enhance the activity of 5-FU. MTX inhibits
purine synthesis, causing an increase of the intracellular
1324
pool of phosphorybosilpyrophosphate (PRPP), which in
turn favours the formation of 5-FU anabolites. In addition, MTX may exert a direct inhibition of thymidylate
synthase (TS). A meta-analysis of eight randomized
trials comparing MTX + 5-FU with 5-FU alone showed
an advantage of the combination in terms of response
rate (19% vs. 10%, P < 0.0001) and a small but statistically significant improvement in survival [3]. The issue of
the double biochemical modulation of 5-FU by means
of MTX and LFA in metastatic colorectal cancer has
recently been addressed by several authors [4-8]. In our
phase II study carried out in advanced colorectal carcinoma, a regimen including MTX 750 mg/m2 (two-hour
i.v. infusion) followed by LFA 250 mg/m2 (two-hour i.v.
infusion) and 5-FU 800 mg/m2 (i.v. bolus) every 2 weeks
obtained a 31% response rate, and a median time to
progression and survival of 27 and 63 weeks, respectively [9]. Therefore, we selected this treatment as a
reference regimen in advanced colorectal cancer.
Irinotecan hydrochloride (CPT-11) represents a novel
therapeutic alternative for the treatment of advanced
colorectal cancer. CPT-11 has been investigated both
in 5-FU-resistant and in chemonaive patients. These
studies have demonstrated a definite activity in first-line
treatment, with response rate (20%-25%) close to that
of standard 5-FU + FA regimens [10, 11]. A moderate
activity (16%-17%) was also observed in second line,
even in patients refractory to 5-FU, suggesting a lack of
cross resistance with 5-FU [12, 13]. Phase II—III studies
are evaluating the combination of CPT-11 with 5-FU +
LFA or other cytotoxic drugs. We recently conducted a
phase I study aimed at evaluating the MTDs of CPT-11
and 5-FU + LFA, given with a 24-hour interval, every
two weeks in patients with advanced colorectal carcinoma. The recommended doses for phase II study were
200 mg/m2 for CPT-11, 850 mg/m2 for 5-FU, and 250
mg/m 2 for LFA. An overall 45% response rate was
reported in that trial [14].
Although 5-FU acts mainly through TS inhibition, it
also has non-specific effects on RNA and DNA. Therefore, direct and specific TS inhibitors represent an attractive class of new agents. Raltitrexed (Tomudex® (TOM))
is the first of these drugs which has reached clinical
development. Phase II studies have demonstrated activity
of TOM in patients with colorectal cancer, even in
patients who had previously received 5-FU-based regimens. In a large multicenter phase II study, a 26%
response rate was reported with TOM [15]. Two out of
three randomized studies comparing TOM alone with
5-FU + FA regimens in colorectal cancer have shown no
differences in response rate, time to treatment failure
and survival between the two treatment arms, and a
better toxicity profile for TOM, while a third study
reported a survival advantage for 5-FU-FA treatment
[16-18]. Furthermore, a synergy between TOM and
5-FU has been demonstrated in colon cancer cell lines
when TOM is followed by 5-FU (as short exposure), and
even a greater synergism was noted with the addition of
leucovorin to 5-FU [19, 20]. Interestingly, a five to six
fold increase of intracellular pool of PRPP, with a 30%
enhancement of 5FdUMP formation, has been demonstrated after TOM exposure [19]. In this way, TOM as
well as MTX, exerts both a direct inhibition and an
indirect increase of the linkage between 5-FUdUMP
and TS. We recently conducted a phase I study in
patients with advanced colorectal or head & neck carcinoma to define the MTDs of both TOM and 5-FU +
LFA, given with a 24-hour interval, every two weeks. A
dose of 3.0 mg/m2 of TOM was safely administered
before 5-FU, 1050 mg/m2, and LFA, 250 mg/m2. Also
pretreated colorectal cancer patients showed a response
to this regimen [21].
With these premises in mind, we decided to carry out
a phase II—III randomised study, with the aim of better
assessing the activity and toxicity of the new regimens
we had devised, and to compare them with our standard
regimen of MTX + 5-FU-LFA, in patients with advanced colorectal carcinoma.
Patients and methods
Selection of patients
To be eligible for this study a patient had to have a histologically
proven diagnosis of locally advanced or metastatic adenocarcinoma of
the colon or rectum; an age > 18 years; a life expectancy > 3 months;
an Eastern Cooperative Oncology Group (ECOG) performance status
<2; presence of measurable indicator lesion; white blood cell (WBC)
count ^3500/mmc, neutrophil count >2000/mmc, platelet (PLT)
count ^ 100,000/mmc, Hb serum level > 10 g/dl; bilirubin serum level
< 1.25 x upper normal limit (UNL), ALT and AST serum level <2.5
UNL in absence of liver metastasis, or bilirubin serum level $1.25,
ALT and AST serum level <5 x UNL in case of liver metastasis;
normal renal function (calculated creatinine clearance ^60 ml/min);
absence of ascite or pleural effusion; written informed consent.
Exclusion criteria were previous administration of chemotherapy for
metastatic disease or adjuvant treatment ended less than six months
before study entry; presence of inflammatory bowel diseases, or
significant diarrhoea in the last week; previous total colectomy or
ileostomy. Also excluded were patients with bowel obstruction/subobstruction, uncontrolled metabolic disorders or active infections,
severe cardiac arrhythmia, uncontrolled congestive cardiac failure, or
acute myocardial infarction within the last six months. The trial was
approved by the Ethical Committee for Clinical Research of the
National Tumor Institute of Naples.
Objective of the study
The primary end-points of this phase II—III trial were to evaluate the
response rate and the acute toxicity profile of a combination chemotherapy in which 5-FU and LFA were combined either with CPT-11 or
TOM in patients with advanced colorectal carcinoma, and to have
preliminary evidence of the difference in response rate between each of
these two experimental regimens (LFA + 5-FU preceded by either
CPT-11 or TOM), and the reference regimen of MTX followed by LFA
+ 5-FU every two weeks.
Secondary end-points were to assess the time to treatment failure,
the time to progression, and the overall survival of treated patients,
and to evaluate their quality of life.
1325
Treatment
Patients that meet all the inclusion criteria were registered at the
Division of Medical Oncology A, National Tumor Institute of Naples.
After stratification for site of primary (colon vs. rectum) and performance status (0 vs. 1-2 of the ECOG scale), patients were randomly
allocated to receive one of the three following treatment: CPT-11 200
mg/m 2 (dissolved in 500 ml of normal saline) administered as 90-min
i.v. infusion on day 1, LFA 250 mg/m 2 (diluted in 1 litre of normal
saline) administered as two-hour i.v. infusion, and 5-FU 850 mg/m 2
given i.v. (bolus) on day 2 every two weeks (arm A); TOM 3.0 mg/m 2
(diluted in 100 ml of normal saline) administered as 15-min i.v.
infusion on day 1, LFA 250 mg/m 2 (diluted in 1 litre of normal saline)
administered as two-hour i.v. infusion, and 5-FU 1050 mg/m 2 given
i.v. bolus on day 2 every two weeks (arm B); or MTX 750 mg/m 2
(diluted in 1 litre of normal saline) administered as two-hour infusion
on day 1 after urine alkalisation, LFA 250 mg/m 2 (diluted in 1 litre
of normal saline) administered as two-hour i.v. infusion, and 5-FU
800 mg/m 2 i.v. bolus on day 2 every two weeks (arm C).
Duration of treatment
Treatment was administered in each arm of the trial every 14 days until
a complete response was achieved (after which 4 more cycles were
administered) or up to a maximum of 14 cycles (or 6 months of
treatment) in the case of partial response or stable disease. Therapy
was discontinued earlier in presence of documented progression of
disease, unacceptable toxicity, patient's refusal, or when the attending
physician judged that it was in the patient's best interest.
After the discontinuation of treatment, patients were controlled
every two months to assess relapse of disease (for responder patients),
or progression of disease (for patients with stable disease).
Dose modification
In each arm of the trials, doses were adjusted according to blood cell
count at nadir: if a grade 4 neutropenia was encountered, or a febrile
neutropenia occurred, the dose of cytotoxic drugs were subsequently
reduced by 25%. If a grade 4 neutropenia again occurred despite this
reduction, a further reduction by 25% was adopted. In case of neutropenia (ANC < 1500/mmc) on day 15, recycling was delayed until
recovery. If a two-week delay was required to reach ANC value
> 1500/mmc, a 25% dose reduction was than adopted.
Dose modifications were also planned for severe non-haematological toxicities: in case of severe diarrhoea (grade 3 or 4 of the WHO
criteria [22]) or other toxicities 3* grade 3 (except alopecia and
anaemia), therapy was delayed for one to two weeks until complete
recovery or ^ grade 1, than re-instituted with a 25% reduction of
cytotoxic drugs.
ciated with grade 3-4 neutropenia were also given fluoroquinolone
therapy. In case of febrile neutropenia (fever ^ 38 °C with concomitant
grade 3-4 neutropenia) or documented infection, patients were hospitalised and a broad spectrum antibiotic was administered i.v. In this
case, the use of haematopoietic growth factors was mandatory. In
presence of grade 4 neutropenia without fever, preventive oral or i.v.
antibiotic were not recommended (except in case of concomitant
diarrhoea), but the use of haematopoietic growth factors was allowed.
Patient study procedures
Initial work-up included a medical history, physical examination,
evaluation of performance status, and registration of symptoms; blood
cell count with white cell count differential, and biochemistry (serum
bilirubin, alkaline phosphatase, ALT, AST, and creatinine, calculated
creatinine clearance, total protein and differential, CEA, CA 19.9). All
patients were submitted to chest X-ray; target lesions were measured
by computed tomography (CT) scan or magnetic nuclear resonance
(MNR) imaging, with addition of ultrasonography if necessary.
During treatment, blood cell count was performed weekly (or more
often in presence of neutropenia), while biochemistry was repeated at
each cycle. Physical status, and evaluation of toxicity was assessed at
each cycle. Toxicity was scored according to WHO criteria [22], and the
worst toxicity suffered by each patient during the whole treatment was
recorded. The occurrence of the cholinergic syndrome was arbitrarily
scored as grade 1 when it did not induce any prophylactic or therapeutic intervention, grade 2 when it required atropine administration, or
grade 3 when a cytotoxic drugs reduction was applied.
Measurements of the disease with CTor MNR scan was repeated
after every four cycles and at the end of treatment. Response to
treatment was defined according to WHO criteria [22]. Extramural
reviewers evaluated the eligibility, assessability, and response in each
patient. A complete response (CR) was defined as no clinical evidence
of residual lesions determined on assessment. To qualify for a partial
response (PR) there had to be a decrease > 50% of the total tumor
mass derived from the sum of the products of the two largest perpendicular diameters of all measurable lesions; a minor response (MR)
was defined as a decrease > 2 5 % but < 5 0 % of tumor mass. The
appearance of a new lesion, or the increase in the total tumor mass of
all measurable lesions by 25% or more, were defined as disease
progression (PD). The lack of objective disease progression, together
with insufficient evidence for partial or minor response were defined
as no change (NC). A response was radiologically confirmed eight
weeks after its first documentation, and only confirmed responses were
reported in each arm of the trial.
Duration of response was calculated from the time of the start of
treatment (in case of partial or minor response), or from the date it was
first documented (in case of complete response) to the date of documented tumor progression.
Assessment of dose intensity
Treatment of specific toxicities
Systematic prophylaxis for cholinergic symptoms caused by CPT-11
was not recommended. If these symptoms occurred and were considered to be troublesome, atropine (0.25 mg SC) was administered
curatively. In this case, preventive treatment with atropine was allowed
in subsequent cycles. No prophylaxis for delayed diarrhoea related to
CPT-11 was given. As soon as the first liquid stool occurred, the patient
immediately started loperamide, 1 capsule (2 mg) orally, every two
hours for at least 12 hours, and up to 12 hours after the last liquid stool,
without exceeding a total treatment duration of 48 hours. Oral rehydration with large volume of water and electrolytes was prescribed
during the whole diarrhoea episode. If diarrhoea persisted for more
than 48 hours despite the recommended loperamide treatment, the
patient was hospitalised for parenteral support. A seven-day oral
therapy with fluoroquinolone was started. Loperamide was replaced
by octreotide at the dose of 0.2 mg s.c. three times a day for at least 48
hours. Patients with grade 4 diarrhoea, diarrhoea with fever or asso-
Dose intensity of each cytotoxic drug was calculated over the first four
(DI4), and eight (DI 8 ) cycles of therapy for each individual patient,
dividing the cumulative dose received by the body surface area and by
the number of days elapsed from treatment onset to the last day of
treatment-free interval following the fourth or eight course, respectively [23].
Assessment of time to treatment failure, time to progression, and
survival
Time to treatment failure (TTF) was calculated from the date of
registration to the date of discontinuation of treatment for any cause
(progression, toxicity, refusal, death, or further anticancer therapy
before documentation of progressive disease, whichever occurred first).
Time to progression (TTP) was calculated from the date of registration
to the date of documented tumor progression, or death. Patients that
1326
discontinued early the treatment because of toxicity, refusal or other
reasons than progression were considered as censored at that time
interval.
Survival time was calculated for all eligible patients from the date
of registration to the date of death for any cause, or patients last
follow-up. After the discontinuation of the treatment planned in this
protocol, the patients were followed every two months to assess the
state of disease, and their survival.
Assessment of quality of life
The assessment of quality of life (QOL) was performed by means of
a simplified Rotterdam symptom checklist questionnaire [24]. Each
patient was asked to fill in this questionnaire just before the start of
therapy, and three months after the start of therapy, or whenever
therapy was definitely discontinued. Five items of this questionnaire
concerned the most frequent physical symptoms related to colorectal
cancer (diarrhoea, weight changes, pain, tiredness, lack of appetite),
while five additional items evaluated the functional, psychological,
social and emotional conditions of the patients. A five-step scale was
used to grade the symptoms and to evaluate the physical, psychological, and social domains, and a total score was attributed to each
patient. An increase or a decrease > 25% of this score during treatment was arbitrarily considered as an improvement or a worsening of
the quality of life, respectively. To avoid the potential bias induced by
missing data, dead patients' missing data were imputed as a zero score.
The median value of the score of all patients at the start, during and at
the end of treatment, as well as the percentage of patients that showed
an increase or a decrease of the score during treatment, was calculated
to assess the impact of treatment on QOL.
Sample size and statistical considerations
To define the sample size, the Simon's minimax two stage design for
phase II clinical trials was utilised [25], setting the a and P errors as 0.05
and 0.20, respectively, and defining the minimum activity of interest
(p0) for the experimental treatments as a response rate of 20%. In
order to test the alternative (p0 hypothesis of a 35% activity, at least 16
responses should be reported among the final sample of 53 patients.
This sample size is also sufficient to make an interim comparison of
activity between experimental and standard treatments, according to
the two-stage design for phase III randomized trials [26]. In fact, with
a hypothesis of a 15% difference in response rate (20% vs. 35%)
between the control and experimental regimen, a difference of at least
4% in response rate should be observed (with an 80% power) among
the first 48 treated patients per arm. If this not the case, the study could
be early terminated. Otherwise, accrual should continue up to reach a
total of 116 patients per arm, and a second comparison will be made on
the final sample size.
Differences in distribution of patients in the three arms of the trial
were evaluated with the chi-square [27], or Fisher exact test [28]. The
exact binomial confidence interval was applied to estimate the response
rates. The probabilities of treatment failure, progression of disease, and
survival were calculated with the Kaplan-Meier method [29].
ended less than six months before registration. The
exclusion of these patients caused an imbalance in proportion of pretreated patients: indeed, twice as many
patients in arms A or C had received 5-FU-based adjuvant chemotherapy as compared with arm B (P = 0.19).
Furthermore, despite randomisation, a slight (not significant) difference also occurred in the distribution of
patients that had recently suffered a significant loss of
body weight (arm A, 24% vs. arm B, 40%, vs. arm C,
38%), and for the proportion of patients with synchronous metastasis (45% vs. 57% vs. 38%, respectively). All
other pre-treatment characteristics resulted well balanced
across the three arms of the trial (Table 1).
Treatment exposure and evaluation of dose intensity
Arm A
The 53 patients allocated in arm A received a total of
467 cycles of chemotherapy, with a median number of
8 (range 1-20) courses per patient. Forty-three (81%)
patients received 5=4 courses, thirty-four patients received >8 courses, eighteen patients received ^12
courses, and eight patients ^16 courses. Considering
the actually delivered dose intensity along the first four
courses of chemotherapy (DI4), median values were
85 (range 55-100) mg/m2/wk for CPT-11, and 350
(range 230-447) mg/m2/wk for 5-FU. Among patients
receiving at least eight cycles, DI8 median values were 78
(range 50-100) and 350 (range 220-417) mg/m2/wk,
respectively. Both these values represent about 80%
of the planned dose intensities for CPT-11 (100
mg/m2/wk), and for 5-FU (425 mg/m2/wk).
ArmB
Three hundred sixty-seven cycles were administered to
the patients of arm B, with a median number of 7 (range
1-14) courses per patient. Forty-one patients (77%)
received ^ 4 courses, twenty-six patients received ^ 8
courses, twelve patients received ^ 12 cycles, and six
patients received fourteen courses. The median TOM
DI4 was 1.4 (range 1.0-1.8) mg/m2/wk (93% of the
planned Dl), while median 5-FU DI4 was 480 (range
330-613) mg/m2/wk (91% of the planned DI). Median
DI8 was again 1.4 (range 1.0-1.6) mg/m2/wk for
TOM, and 430 (range 345-530) mg/m2/wk for 5-FU.
Relative Dl8 median values were 93% and 82%, respectively.
ArmC
The patients in arm C received a total of 393 courses of
treatment. The median number of courses was 7 per
patient (range 1-16). Forty-three patients (81%) received
Patient population
^ 4 courses, twenty-seven patients received ^ 8 courses,
From March 13, 1998 to December 15, 1999, a total 167 fourteen received > 12 cycles, and six received ^ 14
patients were enrolled into this study (54 in arm A, 58 in cycles.2 Median MTX DI4 was 320 (range 154-389)
arm B, and 55 in arm C). Eight patients (1 in arm A, 5 in mg/m /wk,2 and median 5-FU DI4 was 374 (range 210arm B, and 2 in arm C) resulted ineligible because of a 421) mg/m /wk, corresponding to 85% and 93%, remajor protocol violation of entry requirements: all these spectively, of the planned DI. Median MTX DI8 was
patients had received an adjuvant 5-FU-based treatment unchanged over time, while median 5-FU Dig slightly
Results
1327
Table 1. Main pre-treatment characteristics of enrolled patients.
Eligible patients
Sex
Males
Females
Age (years)
Median
Range
Colon or rectosigmoid colon
Rectum
Radical surgery
Previous adjuvant
RT
Previous adjuvant
FA-FU
Median free intervalinmonths
ECOG performance
status
0
1
2
Presence of
symptoms
Weight loss > 5%
Synchronous
metastases
Number of disease
sites
1
2
Arm A, CPT-1
+ LFA-5-FU,
n (%)
Arm B, TOM Arm C, MTX
+ LFA-5-FU, + LFA-5-FU,
n (%)
n (%)
53
53
53
31
22
35
18
26
27
63
34-72
64
37-78
62
39-75
32(60)
21(40)
46(87)
34(64)
19(36)
42(79)
41(77)
12(23)
49(92)
3(6)
3(6)
4(8)
13(24)
7(13)
14(26)
16 (6-28)
21 (6-3 5)
31(6-102)
31(58)
20(38)
2(4)
28(53)
20(38)
5(9)
27(51)
22(42)
4(8)
22(41)
13 (24)
25(47)
21 (40)
24(45)
20 (38)
24(45)
30(57)
20(38)
25(47)
21(40)
7(13)
27(51)
19(36)
7(13)
29(55)
19(36)
5(9)
36(68)
10(19)
39(74)
7(13)
19(36)
13(25)
5(9)
0
13(25)
9(17)
8(15)
2(4)
2(4)
>3
Sites of disease
Liver
40(75)
12(23)
Lung
Local (unresected
13(24)
or recurrent)
Nodes
10(19)
Peritoneum
6(11)
Bone
5(9)
Other
3(6)
CEA value
>5ng/ml
42(79)
CA 19.9 value
>35U/ml
39(73)
4(8)
37(70)
37(72)
29(55)
31(58)
decreased, being 342 (range 169-407) mg/m2/wk, corresponding to 86% of the intended DI.
Evaluation of activity
Arm A
Among the 53 patients allocated to receive the CPT11 +
LFA-5-FU regimen, 18 patients achieved a confirmed
major response, including 3 CRs and 15 PRs. Therefore,
the overall activity rate was 34% (95% confidence interval (95% CI): 21%-48%). A tumor shrinkage that did
not qualify for a major response was observed in seven
additional patients, while 12 patients showed stable disease. Therefore, a control of tumor growth was tempo-
Table 2. Summary of results obtained in the three arms of the trial.
Treatment
Responses
Complete
Partial
Minor
No change
Progression
Not assessed
Overall RR (%) (95% CI)
Time to progression (weeks)
Median survival time
(weeks)
CPT-11+
LFA-5-FU
TOM+
LFA-5-FU
MTX +
LFA-5-FU
3
15
7
12
9
7
34(21-48)
38
2
11
4
15
13
7
24(14-38)
25
2
11
2
13
17
8
24(14-38)
27
74
63
67
rarily achieved in 70% of patients. Seven patients were
not assessed for response for early discontinuation of
chemotherapy, and nine patients had a documented
progression of disease while on treatment (Table 2).
Three CRs were reported in patients affected by one
(1 case), or more (2 cases) metastatic deposits into the
liver. A PR was achieved in patients with one site (5
patients), two sites (9 patients), or three sites (1 patient)
of disease. The median time to reach a documented
response was 11 (range 8-26) weeks from the start of
treatment. It is worth mentioning that, among 42 patients receiving at least 4 courses of chemotherapy, the
RR was 46% for patients receiving a DI4 of both CPT-11
and 5-FU > 80% of those planned, while it decreased to
28% for patients receiving a lower relative DI4 The 3
CRs had a duration 10+, 23 and 28 weeks, while the
duration of PRs ranged between 14 and 66 (median 40)
weeks. Median duration of responses and stable disease
was 47 (range 16-66) weeks.
ArmB
Two CRs and eleven PRs were documented in patients
treated with TOM + LFA-5-FU, giving a 24% (95% CI:
14%—38%) RR. Four patients were classified as having
a minor response, and fifteen patients showed a stabilisation of disease during treatment, giving a control of
tumor growth in 60% of patients. A progression of
disease during treatment was registered in 13 patients.
Seven patients were not assessed for activity (Table 2).
A complete disappearance of disease was documented
in two patients with only one site of disease (liver or
lymph nodes, respectively). The PRs were achieved in
patients with one site (6 cases), two sites (3 cases), or
three sites of disease (2 cases). The median time to reach
a major response was 9 (range 6-23) weeks. Probability
of response was 35% for patients receiving a relative DI4
of both TOM and 5-FU >90%, compared to 19%
registered among patients receiving a lower DI4. One
CR was still persistent at the time of this analysis,
lasting 63+ weeks from its first assessment. The other
patient achieving a CR subsequently relapsed after nine
weeks. However, it should be mentioned that he had
received only four cycles, refusing further therapy after
1328
the first documentation of response. The remaining PRs
had a median duration of 23 (range 11-85+) weeks.
Median duration of responses and stable disease was 35
(range 12-76) weeks.
Arm C
Thirteen confirmed responses, including two CRs and
eleven PRs, were reported among patients enrolled in
arm C, giving an ORR of 24% (95% CI: 14%-38%). Two
additional patients showed a minor tumor shrinkage,
while thirteen patients had a stable disease during treatment. Therefore, 28 (63%) patients showed a control of
tumor growth, while a clear progression of disease while
on treatment was documented in 17 patients, and 8
patients were not assessed (Table 2).
The complete disappearance of tumor was obtained
in a patient with nodal involvement, and in an additional patient with pelvic recurrence. A PR was achieved
in seven patients with one site of disease, in three patients
with two localizations, and in one patients with three
sites. The median time to first documentation of a response was 12 (range 8-20) weeks. The 2 patients achieving a CR were still without evidence of recurrence after
37 and 55 weeks, respectively. PRs lasted a minimum of
8+ weeks to a maximum of 42 weeks, showing a median
length of 30 weeks. Median duration of responses and
stable disease was 41 (range 23-63) weeks.
Evaluation of toxicity
Arm A
Among 53 patients enrolled in Arm A, no toxic deaths
were registered. Three patients received only one cycle of
chemotherapy: two for refusal, and another one for a
disease complication (i.e., bowel obstruction). These
patients were not assessed for toxicity, leaving 50 evaluable patients. Two patients suffered a grade 1, and eight
patient a grade 2 neutropenia. Twelve (24%) patients
showed a grade 3 neutropenia, in five of them requiring
a delay of recycling, and in two a dose reduction. Eleven
(22%) patients showed a grade 4 neutropenia, requiring
dose reduction in eight cases. According to the 467
cycles administered, grade 3 or 4 neutropenia occurred
in 5%, and 3% of them, respectively. However, severe
neutropenia was usually short lasting and uncomplicated.
Indeed, only one patient (2%) suffered a febrile neutropenia. Fourteen patients (28%) showed a mild decrease
(grade 1) of haemoglobin serum level during their treatment, and nine (18%) had a decrease to grade 2. One
patient showed grade 4 anaemia coupled with severe
neutropenia soon after the first cycle of chemotherapy.
Two patients showed a mild decrease of platelet count,
while one patient showed a decrease < 50,000/mmc after
seven cycles of chemotherapy (Table 3).
The most common non-hematologic toxicities were
gastrointestinal disturbances: 34 (68%) patients suffered
from some grade of nausea or vomiting, of grade 3 in
2 (4%) patients (no grade 4 was reported); 26 (52%)
patients were affected by delayed diarrhoea during treat-
Table 3. Acute hematologic and non-hematologic toxicity by patients
according to WHO scale.
Arm A, CPT-11 Arm B, TOM Arm C, MTX
+ FA-5-FU,
+ LFA-5-FU, + LFA-5-FU,
n (%)
n (%)
n (%)
Assessable patients 50
Neutropenia
10(20)
Grade 1-2
Grade 3
12(24)
Grade 4
11 (22)
Febrile neutropenia
2(4)
Anemia
Grade 1-2
23 (46)
Grade 3/4
0/1(2)
Thrombocytopenia
2(4)
Grade 1-2
Grade 3/4
1/0(2)
Nausea or vomiting
Grade 1-2
32 (64)
Grade 3
2(4)
Diarrhoea
18(36)
Grade 1-2
Grade 3/4
3/5(16)
Stomatitis
8(16)
Grade 1-2
Grade 3/4
1/0(2)
Hair loss
Grade 1-2
17(34)
Grade 3
22 (44)
Cholinergic syndrome
Grade 1-2
12(24)
Liver toxicity
Grade 1-2
2(4)
Renal toxicity
Grade 1-2
1(2)
50
50
8(16)
4(8)
4(8)
1(2)
13(26)
4(8)
3(6)
0(-)
12(24)
3/0(6)
14(28)
0/1(2)
2(4)
0/1(2)
6(12)
2/1(6)
24 (48)
4(8)
24 (48)
1(2)
12(24)
7/1(16)
22 (44)
1/1(4)
15(30)
2/1(6)
15(30)
5/1(12)
14(28)
1(2)
6(12)
2(4)
0(-)
0(-)
5(10)
8(16)
0(-)
4(8)
ment, but severity of this side effect was usually mild or
moderate, being of grade 3 in 3 (6%) patients, and of
grade 4 in 5 (10%) patients. Stomatitis was reported in
nine (18%) patients, but it was of grade 3 in only one
case. Thirty-seven (74%) patients showed some hair loss
during treatment, that was complete in twenty-two
(44%) patients. Alopecia usually occurred after more
than four courses of chemotherapy, but it was always
reversible, even in patients still on treatment. Twelve
(24%) patients showed the acute cholinergic syndrome
attributable to irinotecan, in eight cases requiring
administration of atropine sulphate for control and
subsequent prevention. Two (4%) patients showed a
transient derangement of serum liver enzymes, and one
(2%) patient an occasional increase of creatinine serum
level.
ArmB
Among 53 patients enrolled in arm B, 3 patients were
not properly assessed for toxicity: 1 for early discontinuation after 1 cycle, 1 for an early complication that
required the admission of patient to another hospital,
and the third for lack of adequately collected data. Fifty
patients were fully assessable: among them, no toxic
deaths were recorded. Sixteen (32%) patients showed
some degree of neutropenia during treatment: grade 3
1329
occurred in four (8%) patients, one of whom requiring a
delay of recycling; grade 4 neutropenia occurred in four
(8%) patients, for whom a dose reduction was applied in
subsequent cycles. Considering the total of 367 cycles
administered, grade 3 or 4 neutropenia affected 1% each,
respectively. Severe neutropenia did not seem related to
drug exposure. Indeed, 4 of 8 episodes occurred after 5
or fewer cycles, while the remaining were observed in
patients receiving 8-12 cycles. Only one patient showed
an episode of febrile neutropenia. Twelve (24%) patients
showed some kind of mild (8 cases) or moderate (4 cases)
anaemia during their treatment, while 3 (6%) patients
showed a fall of Hb serum level < 8 g/dl, requiring
packed red cell transfusions. Thrombocytopenia was
occasionally detected in three patients, being of grade 4
in one case.
Despite the prophylaxis for emesis, 28 (56%) patients
complained of some kind of vomiting after chemotherapy,
that was severe in 4 cases. Diarrhoea induced by chemotherapy was reported in 20 (40%) patients. It was usually
mild and short-lasting, but it implied a subsequent dose
reduction in six (12%) cases, and caused the refusal of
further cycles in one (2%) patient. Another patient, after
the second cycle, suffered persistent diarrhoea requiring
hospitalisation for forced hydration, and was thereafter
withdrawn from further cytotoxic treatment by the attending physician. Stomatitis of any grade affected 18
(36%) patients. It was of grade 3 in two (4%) cases, and
of grade 4 in one (2%) patient. Fifteen patients showed
some hair loss during treatment, that was complete in
one case. Five patients showed a transient derangement
of liver enzymes, never requiring dose reduction. Eight
(16%) patients showed some rash localised to the trunk
and abdomen, occurring the day of, and lasting three to
five days after, chemotherapy administration. Two (4%)
patients reported a mild or moderate asthenia, and one
complained of arthralgias and myalgias.
Arm C
Three out of fifty-three patients assigned to receive our
reference regimen were not assessed for toxicity: one for
refusal to stay on treatment after one cycle, another one
for early discontinuation because of cardiac complications (not treatment-related) after three courses, and
the third case for incomplete collection of available
information. Therefore, 50 patients were fully assessable
for toxicity. One toxic death was recorded among these
patients: she was a 62-year-old woman that suffered a
severe episode of diarrhoea and stomatitis, coupled with
a grade 4 neutro- and thombocytopenia and anaemia,
occurring soon after the first cycle. Despite a prompt
hospitalisation and the administration of intensive supportive care, the patient died eight weeks after the start
of therapy.
Twenty (40%) patients showed an inter-cycle decrease
of granulocyte count during treatment, but grade 3 or 4
neutropenia occurred in only four (8%), and three (6%)
patients, respectively. Two of the three episodes of grade
4 neutropenia occurred after the first cycle, and they
were accompanied by severe thrombocytopenia and/or
anaemia. Both these patients were females, and aged
more than 60 years. No episodes of neutropenic fever
were reported. A mild or moderate decrease of haemoglobin serum level was detected in 14 (28%) patients
during treatment. Only one patient showed a grade 4
anaemia. Thrombocytopenia was occasionally observed
in nine (18%) patients. Only in three (6%) cases it was
severe (grade 3 or 4).
As far as non-hematologic toxicities are concerned,
nausea or vomiting was reported by 25 (50%) patients at
some time after treatment, but these side effects were
usually mild, only one episode being graded as severe.
Also the episodes of diarrhoea reported by 24 (48%)
patients were commonly graded as mild or moderate;
however, 2 (4%) patients suffered from severe diarrhoea,
that was fatal for the above mentioned patient. Stomatitis
occurred in 21 (42%) patients, and it was severe in
6 (12%). Hair loss was infrequently observed, being
complete only in two patients. A derangement of liver
enzymes due to chemotherapy was reported in eight
cases (16%), while four (8%) patients showed a transient
and reversible increase of creatinine serum level. Additional side effects were: skin rash (in 2 patients), severe
asthenia (1 case), transient conjunctivitis (5 patients).
Time to treatment failure, time to progression, and
survival
At the time of this analysis (April 15, 2000), the median
potential follow-up for all 159 patients entered in this
study was 62 (range 18-108) weeks. At that time, 97
(61%) patients had progressed, and 65 (41%) patients
had died. Five (3%) patients (arm A, 1 patient; arm B, 3
patients; arm C, 1 patient) were lost to follow-up after a
median observation time of 9 (range 2-85) weeks.
Arm A
Ten patients are still on treatment at the time of this
analysis, while forty-three patients are off treatment. The
main reason for discontinuation was achievement of a
response (16 cases), or a disease progression (14 cases),
or no change (1 case). Six patients interrupted the treatment for severe toxicity, two patients for refusal, and one
patient committed suicide. In 2 cases, a disease-related
complication (bowel obstruction) precluded further therapy, while 1 patient after 13 courses of chemotherapy
was referred to surgical removal of residual liver metastasis. The median time to treatment failure for the whole
group of patients was 29 (range 2-73+) weeks. Twentyseven patients had a documented progression of disease,
and the median time to progression was thirty-eight
weeks. Fifteen (29%) patients have subsequently been
submitted to further treatment, consisting of local surgery
or irradiation (1 case each), and of further chemotherapy
(Tomudex alone, 1 case; oxaliplatin-based chemotherapy, 12 cases). Median survival time was 74 weeks, and
one-year and two-year survival was 61% and 46%,
respectively, for patients entered in this arm.
1330
ArmB
At the time of analysis, five patients are still on treatment. Among the remaining 48 patients, the reason for
discontinuation was the achievement of a response or
stable disease (13 patients), or occurrence of progressive
disease (19 patients). An early drop-out occurred for
refusal in four patients, and for toxicity in two patients.
Two patients were forced to definitively interrupt the
treatment because of occurrence of complications
(stroke, 1 case; bowel obstruction, 1 case), and three
patients died early. In five cases, the decision of stopping
treatment was taken by the physician in charge, who
believed it was in the patient's best interest. The time to
treatment failure ranged between 2 and 72+ weeks, and
had a median length of 18 weeks. Progression of disease
occurred in 38 patients. The median time to progression
was 25 weeks. One patient, after the discontinuation of
first-line chemotherapy, underwent liver surgery, another
patient underwent local (pelvic) irradiation, while sixteen (30%) patients received additional chemotherapy
(protracted i.v. infusion of 5-FU, 4 cases; CPT-11, 5
cases; oxaliplatin, 5 cases; or a combination of CPT-11
and oxaliplatin, 2 cases). Median survival time was 63
weeks, and projected probability of survival at 1 and 2
years was 54% and 26%, respectively.
Arm C
Nine patients are still on treatment. Among the remaining 44 patients, reasons for discontinuation of treatment
were: achievement of a response (10 cases) or no change
(5 cases), occurrence of progressive disease (18 cases),
while chemotherapy was interrupted for toxicity (5 cases),
or patient's refusal (3 cases). In three cases therapy was
discontinued because of complications (i.e., bowel obstruction, disseminated intravascular coagulation, and
acute cardiac failure, respectively). The median time to
treatment failure was 19 (range 2-63+) weeks. At the
time of analysis, thirty-two patients showed a progression of disease. The median time to progression was 27
weeks. Twenty (38%) patients received a second-line
treatment, consisting of CPT-11 (11 patients), oxaliplatin
(3 patients), or both (1 patient), 5-FU continuous i.v.
infusion (3 patients), or TOM alone (1 patient), while
one patient was submitted to surgery. The median
survival for this group of patients was 67 weeks, and
actuarial one-year and two-year survival was 59% and
25%, respectively.
Evaluation of quality of life
Only three Centers participating in this trial cooperated
in collecting the questionnaires of QOL of treated patients. Therefore, 118 of 159 (74%) patients have been
evaluated for this purpose (arm A, 41; arm B, 39; arm C,
38).
Among evaluable patients entered in arm A, basal
QOL score ranged between 48 and 94, and had a median
value of 76. At three-month evaluation, median score
was 78, with a range of 62 to 90. Six (15%) patients at
that time interval were considered to have obtained a
significant (>25%) increase of their score, while no
patient showed a significant decrease of QOL score.
Basal median value for patients entered in arm B was
74 (range 40-94). Five (13%) patients significantly improved their QOL score at three months, while seven
(18%) patients totalled a significantly lower score. The
median value was 70.
Basal median value for patients of arm C was 78
(range 38-90). None of these patients significantly improved the QOL score at three-month assessment, and
10 (26%) patients had a significant worsening of their
score. Median value at this time assessment was 60.
Discussion
The main aim of this study was to assess the activity and
toxicity of two new cytotoxic regimens for advanced
colorectal cancer patients, in which a leucovorin-modulated 5-FU i.v. bolus given every two weeks was preceded
by either CPT-11 or TOM. Furthermore, we would also
have an estimate of the activity of both these regimens in
comparison with our standard treatment of double biochemical modulation of 5-FU by means of MTX and
LFA.
As far as the CTP-11 + LFA-5-FU combination is
concerned, this regimen produced major responses in
34% of treated patients. Considering also the patients
that showed a minor tumor response, and those with a
transient stabilisation of disease, overall tumor control
was achieved in 70% of eligible patients. The activity of
this regimen was coupled with a fairly good toxicity
profile. The main acute toxicity was severe neutropenia,
affecting nearly half of treated patients. However,
neutropenia was usually short lasting, as reflected by
the very low occurrence of febrile complications, and by
the actually delivered dose intensity of both drugs along
the first four and eight cycles of chemotherapy, being
about 80% of that intended ones. In addition, giving
careful information of possible side effects to each patient before therapy, and starting appropriate treatment
as soon as required, we were able to limit the occurrence
of severe delayed diarrhoea to 16% of our series.
Although about a quarter of our patients showed some
kind of cholinergic symptoms after the CPT-11 administration, no patient required a dose reduction because
of this side effect. Though limited to a proportion of
the whole series, the evaluation of QOL confirmed the
tolerability of this regimen; indeed, among assessable
patients, none showed a deterioration, and 15% showed
a significant improvement of basal score after three
months of therapy.
Recently, several investigators have assessed the
combination of CPT-11 plus FA-5-FU in the first-line
treatment of advanced colorectal carcinoma. A trial was
carried-out in Europe to compare the addition of CPT11 (80 or 180 mg/m2, respectively) to either weekly or
biweekly modulated 5-FU [30]. The planned DI in the
1331
weekly regimen were 69 mg/m2/wk for CPT-11, and
about 1970 mg/m2/wk for 5-FU (as 24-hour infusion),
while in the biweekly regimen they were 90 mg/m2/wk
for CPT-11, and 400 mg/nr/wk (as bolus) plus 600 mg/
m2/wk (as continuous infusion) for 5-FU. In this study, a
significantly higher proportion of patients receiving the
combined treatment achieved a major response (35%),
and showed a longer time to progression (6.7 months),
and survival (17.4 months), as opposed to patients treated
with modulated infusional 5-FU alone (22%, 4.4 and
14.1 months, respectively). In that series, the one-year
survival was 69% for the experimental arm, and 59% in
the control arm.
On the other hand, a three-arm US trial compared
the activity of CPT-11 alone, of modulated 5-FU monthly
(Mayo Clinic) regimen, or of a combination of CPT-11
(80 mg/m2) plus modulated 5-FU both given as i.v.
bolus weekly for four consecutive weeks and two weeks
of rest [31]. In the combination arm, the ideal DIs were
83 mg/m2/wk for CPT-11, and 333 mg/m2/wk for 5-FU,
respectively. In that trial, the confirmed RR of the
combination arm was significantly higher (33%) than
each single-agent arm (17% and 18%, respectively); the
time to treatment failure was significantly longer (5.0 vs.
3.1 and 3.8 months, respectively). Also in this trial a twomonth significantly longer median survival was reported
with the combined treatment (14.4 vs. 12.6 months).
Our response rate with CTP-11 + LFA-5-FU,
although obtained in a smaller series, seem comparable
to those reported by European and American investigators. Moreover, it is worth noting the actually delivered
DI of both CPT-11 and 5-FU as i.v. bolus in our series
resulted very close to those planned in the US and
European studies. If this observation will be furtherly
confirmed, one could raise the question of whether a
dose-densified (weekly) administration, or the use of
costly and time-consuming infusional devices, are needed
for the palliative management of advanced colorectal
cancer patients. In addition, we would stress that our
regimen does not require an indwelling central venous
catheter, so that any possible related complication is
avoided. At the interim comparison of this novel combination regimen with our standard MTX-based biweekly
regimen, a 10% increase in RR (34% vs. 24%) justify the
completion of the patients' accrual, to make a definitive
comparison on the whole series.
Considering the results reported in arm B of this trial,
we can confirm our previous findings: TOM and leucovorin-modulated 5-FU can be administered together at
full doses without unacceptable toxicity. This biweekly
regimen represents an active and safe treatment for
advanced colorectal cancer patients. The activity seemed
related to the achievement of an adequate DI for both
drugs. As far as tolerability is concerned, it is possible
that the administration of LFA on day 2, besides potentiating the activity of 5-FU, may also rescue the toxicity
of TOM, contributing to the low toxicity profile of this
regimen. However, the RR was in the range of activity
achievable with each cytotoxic drug given in a standard
schedule. Therefore, it seems that only a mild, if any,
additive effect between TOM and 5-FU can be claimed
from our results. Indeed, in randomised trials comparing TOM with standard 5-FU-based regimens, a RR in
about 20% of patients treated with TOM has been
reported [16-18, 32]. It is interesting to note that in our
trial both TOM + LFA-5-FU and MTX + LFA-5-FU
regimens showed a similar activity and toxicity profile.
This observation lends support to the hypothesis that,
like MTX, the pre-exposure of tumor cells to TOM
before the administration of 5-FU may interfere with its
intracellular metabolic pathway. However, likely because
these two effects share the same target enzyme, they do
not seem to increase the overall cytotoxicity on tumor
cells and on normal tissues. As a matter of fact, this
regimen in our opinion does not deserve further evaluation. Indeed, its activity rate and its toxic profile does
not seem to represent any clear advantage for the treatment of advanced colorectal cancer patients. Therefore,
the accrual of patients in this arm of the trial has been
stopped.
Finally, a short comment should be made also about
our standard regimen. In this randomised trial, the
results yielded by MTX + LFA-5-FU were similar to
those obtained in our previous study [9]. Indeed,
although the RR was slightly lower, due to the inclusion
in the present study of confirmed responses only, the
median time to progression, and the probability of
survival of patients receiving this regimen were exactly
the same in this study as well as in the previous one. On
the basis of these findings, and considering that the
main prognostic factors resulted well balanced across
the arms of the trial, we may rule out that a selection
bias may had occurred in the present study. Indeed, the
overall median time to progression in this arm of the
study (about six months) was not shorter (or perhaps
even longer) than those reported in recent trials, once
again lending support to the relevance of achieving a
growth inhibition rather than obtaining a significant
tumor shrinkage.
In conclusion, the CPT-11 + LFA-5-FU regimen is
an active and well tolerated treatment for advanced
colorectal cancer patients. Since the increase in RR
over the MTX + LFA-5-FU regimen at the interim
analysis met the requirements of the statistical design,
the accrual of patients in these two arms of the trial is
still open. The analysis on the final sample size will
assess whether this new regimen significantly improve
the tumour control, and therefore represents a true
advantage in comparison with MTX + LFA-5-FU.
Acknowledgements
This work was supported in part by Aventis Pharma,
Italy. Tomudex® was kindly granted by Astra-Zeneca,
Italy. We acknowledge the participation in this trial of the
following colleagues from the National Tumor Institute
of Naples: M. Montella, A. Crispo (Epidemiology Unit)
1332
contributing in statistical analysis, V. De Rosa, F. Fiore
(Service of Radiology) for imaging evaluation, F.
Cremona, F. Izzo (Division of Surgery C) for care and
referral of patients. We wish also thank Ms S. Caiazzo
for data management, and Ms L. Gallifuoco for her
helpful secretarial assistance.
10.
11.
12.
Appendix 1
Investigators and Institutions participating in this study
13.
This multicenter trial was carried-out by the following investigators
(Institutions) of the Southern Italy Cooperative Oncology Group:
P. Cornelia, R. Casaretti, A. Avallone, G. Cornelia (National Tumor
14.
Institute, Naples); F. De Vita, M. Orditura, A. Gambardella, G.
Catalano (2nd University School of Medicine, Naples); S. Mancarella,
V. Lorusso, M. De Lena (Oncologic Institute, Bari); M. Biglietto,
R. Guarrasi, G. Carteni (Cardarelli Hospital, Naples); L. De Lucia,
(City Hospital, Caserta); A. Farris (University School of Medicine,
15.
Sassari); G. P. Ianniello (Rummo Hospital, Benevento); L. Leopaldi,
L. Maiorino (San Gennaro Hospital, Naples); F. Del Gaizo, M. Belli
(City Hospital, Avellino); S. Leo (IRCCS Castellana Grotte, Bari);
F. Buzzi (City Hospital, Terni); A. Russo, S. Palmeri (University
16.
School of Medicine, Palermo); C. Brunetti (City Hospital, Manduria);
D. Muci (City Hospital, Nardo); L. Di Lullo (City Hospital, Isernia);
E. Greco (City Hospital, Lametia Terme); F. Avino (Pellegrini Hospital, Napoli); A. Iannelli (City Hospital, Siderno); M. Bianco (City
17.
Hospital, Castellammare di Stabia); Italy.
References
1. Moertel CG. Chemotherapy of gastrointestinal cancer. N Engl J
Med 1978; 299: 1049-52.
2. Advanced Colorectal Cancer Meta-Analysis Project. Modulation
of fluorouracil by leucovorin in patients with advanced colorectal
cancer: Evidence in terms of response rate. J Clin Oncol 1992; 10:
896-903.
3. Advanced Colorectal Cancer Meta-Analysis Project. Meta-analysis of randomized trials testing the biochemical modulation of
fluorouracil by methotrexate in metastatic colorectal cancer.
J Clin Oncol 1994; 12: 960-9.
4. Abad A, Garcia P, Gravalos C et al. Sequential methotrexate,
5-fluorouracil (5-FU), and high-dose leucovorin oversus 5-FU
and high-dose leucovorin versus 5-FU alone for advanced colorectal cancer. Cancer 1995; 75: 1238—44.
5. Glimelius B. Biochemical modulation of 5-fluorouracil: A
randomized comparison of sequential methotrexate, 5-fluorouracil and leucovorin versus sequential 5-fluorouracil and leucovorin
in patients with advanced symptomatic colorectal cancer. Ann
Oncol 1993; 4: 235-50.
6. Polyzos A, Tsavaris N, Giannopoulos C et al. Biochemical
modulation of fluorouracil: Comparison of methotrexate, folinic
acid, and fluorouracil versus folinic acid and fluorouracil in
advanced colorectal cancer: A randomized trial. Cancer Chemother Pharmacol 1996; 38: 292-7.
7. Sobrero A, Frassineti GL, Giuliani R et al. Randomized comparison between sequential MTXFU and schedule specific biochemical modulation in advanced colorectal cancer (ACRC). Ann
Oncol 1998; 9: (Suppl 4): 33.
8. Van der Wilt CL, Braakhuis BJM, Pinedo HM et al. Addition of
leucovorin in modulation of 5-fluorouracil with methotrexate:
Potentiating or reversing effect? Int J Cancer 1995; 61: 672-8.
9. Cornelia P, Lorusso V, Casaretti R et al. Concurrent modulation
of 5-fluorouracil with methotrexate and 1-leucovorin: An effective
and moderately toxic regimen for the treatment of advanced
colorectal carcinoma. A multicenter phase II study of the
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Southern Italy Cooperative Oncology Group. Tumori 1999; 85:
465-72.
Conti JA, Kemeny NE, Saltz LB et al. Irinotecan is an active
agent in untreated patients with metastatic colorectal cancer. J
Clin Oncol 1996; 14: 709-15.
Rougier P, Bugat R. CPT-11 in the treatment of colorectal cancer:
Clinical efficacy and safety profile. Semin Oncol 1996; 23 (Suppl
3): 34-41.
Rougier P, Van Cutsem E, Bajetta E et al. Randomized trial of
irinotecan versusfluorouracilby continuous infusion after fluorouracil failure in patients with metastatic colorectal cancer.
Lancet 1988; 352: 1407-12.
Cunningham D, Pyrhonen S, James RD et al. Randomized trial
of irinotecan plus supportive care versus supportive care alone
after fluorouracil failure for patients with metastatic colorectal
cancer. Lancet 1998; 352: 1413-8.
Cornelia P, Casaretti R, De Vita F et al. Concurrent irinotecan
and 5-fluorouracil plus levo-folinic acid given every other week in
the first-line management of advanced colorectal carcinoma: A
phase I study of the Southern Italy Cooperative Oncology Group.
Ann Oncol 1999; 10: 915-21.
Zalcberg JR, Cunningham D, Van Custem E et al. ZD1694: A
novel thymidylate synthase inhibitor with substantial activity in
the treatment of patients with advanced colorectal cancer. J Clin
Oncol 1996; 14: 716-21.
Cunningham D, Zalcberg JR, Rath U et al. Final results of a
randomised trial comparing 'Tomudex' (raltitrexed) with fluorouracil plus leucovorin in advanced colorectal cancer. Ann
Oncol 1996; 7: 961-5.
Cocconi G, Cunningham D, Van Custem E et al. Open, randomized, multicenter trial of raltitrexed versusfluorouracilplus highdose leucovorin in patients with advanced colorectal cancer. J
Clin Oncol 1998; 16: 2943-52.
Pazdur R, Vincent M. Raltitrexed (Tomudex) versus 5-fluorouracil and leucovorin (5-FU + LV) in patients with advanced colorectal cancer (ACC): Results of randomized, multicenter, North
American trial. Proc Am Soc Clin Oncol 1997; 26: 228 (Abstr 801).
Longo GSA, Izzo J, Chang YM et al. Pretreatment of colon
carcinoma cells with Tomudex enhances 5-fluoruracil cytotoxicity.
Clin Cancer Res 1998; 4: 469-73.
Budillon A, Barbarulo D, Di Gennaro E et al. In vitro cytotoxicity
of Tomudex in combinations with 5-fluorouracil and levofolinic
acid in head and neck and colon cancer cell lines. Ann Oncol
1998; 9 (Suppl 4): 133 (Abstr 637).
Caponigro F, Avallone A, McLeod A et al. Phase I and pharmacokinetic study of Tomudex combined with 5-fluorouracil plus
levofolinic acid in advanced head and neck cancer and colorectal
cancer. Clin Cancer Res 1999; 5: 3948-55.
Miller AB, Hoogstraten B, Staquet M, Winkler V. Reporting
results of cancer treatment. Cancer 1981; 47: 207-14.
Hryniuk WH, Goodyear M. The calculation of received dose
intensity. J Clin Oncol 1995; 8: 1935-7.
De Haes JCJM. Measuring psychological and physical distress in
cancer subjects: Structure and application of the Rotterdam
Symptom Checklist. Br J Cancer 1990; 62: 1034-8.
Simon R. Optimal two stage design for phase II clinical trials.
Contr Clin Trial 1989; 10:1-10.
Thall PF, Simon R, Ellenberg SS. A two-stage design for choosing among experimental treatments and a control in clinical
trials. Biometrics 1989: 45: 537-47.
Cochran WG. Some methods for strengthening the common chisquare test. Biometrics 1954; 10: 417-51.
Fisher RA. Statistical Methods for Research Workers. Edinburgh,
UK: Olive and Boyd Editors 1925.
Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457-81.
Douillard JY, Cunningham D, Roth AD et al. Irinotecan combined withfluorouracilcompared withfluorouracilalone as firstline treatment for metastatic colorectal cancer. A multicentre
randomised trial. Lancet 2000; 355: 1041-7.
1333
31. Saltz LB, Locker PK, Pirotta N et al. Weekly irinotecan (CPT-11),
leucovorin (LV), and fluorouracil (FU) is superior to daily x 5
LV-FU in patients (PTS) with previously untreated metastatic
colorectal cancer (CRC). Proc. Am Soc Clin Oncol 1999; 233a
(Abstr 898).
32. Maugham TS, James RD, Kerr D et al. Preliminary results of
a multicentre randomised trial comparing three chemotherapy
regimens (de Gramont, Lokich, and Raltitrexed) in metastatic
colorectal cancer. Proc Am Soc Clin Oncol 1999; 262a (Abstr
1007).
Received 26 June 2000; accepted 21 July 2000.
Correspondence to:
P. Cornelia, MD
Division of Medical Oncology A
National Tumor Institute
Via M. Semmola
80131 Naples
Italy
E-mail: [email protected]