NASHVILLE, Tennessee-"The
need to find effective treatment for patients
with recurrent colorectal cancer following
treatment with IFL" (irinotecan
[CPT-11, Camptosar], fluorouracil(Drug information on fluorouracil) [5-
FU], and leucovorin) was the driving force
leading to a recent three-arm trial conducted
at institutions in Canada and in
the United States. Data from the trial were
presented by Mace L. Rothenberg, MD,
Ingram Professor of Cancer Research at
Vanderbilt-Ingram Cancer Center in
Nashville, Tennessee (ASCO abstract
1011).
For the past 3 years, IFL "has been
widely used in North America as first-line
chemotherapy in patients with metastatic
colorectal cancer," Dr. Rothenberg
pointed out in his presentation.
Therapeutic Options for
Progressive Disease
"However, in the United States and
Canada no standard therapy has been
available for patients with progressive disease
following this first-line therapy. Because
there are no data in this setting, we
extrapolated from phase II trials performed
in patients with progressive disease
following 5-FU and leucovorin. This
suggested potential activity for three therapeutic
options."
Dr. Rothenberg listed those options
as:
"Given the similarity of results between
these three alternatives in terms of
objective response rate, time to tumor
progression, and immediate survival, and
the need to find effective treatment for
patients with recurrent colorectal cancer
following IFL treatment, we designed this
three-armed trial," he explained.
Eligibility criteria for this trial included
a diagnosis of metastatic colorectal
adenocarcinoma, progressive disease during
or within 6 months following IFL, and
the ability to complete a tumor-related
questionnaire. Patients were randomized
to the three arms as outlined above. Patients
receiving 5-FU/leucovorin were
considered the control group and those
experiencing progressive disease while receiving
5-FU/leucovorin were later given
the opportunity to participate in the oxaliplatin
treatment-access program.
Study Design
The trial was designed to accrue 786
patients and have a primary end point of
overall survival, and secondary end points
of objective response rate, time to tumor
progression, safety, and time to tumor
symptom worsening.
Baseline patient characteristics were
well-balanced between all three arms in
terms of median age (approximately 60
years old), gender (55% to 60% of patients
being male), and race.
There was also equivalent distribution
of baseline stratification factors between
treatment arms. Karnofsky performance
status (KPS) was 70 to 100 in about
95% of patients. Approximately 65% had
two or more sites of metastases and onethird
of patients had elevated LDH. About
70% of patients had tumors arising in the
colon and 20% in the rectum.
By independent radiologic review, the
objective response rate was 0.7% for patients
treated with 5-FU and leucovorin
and 9.6% for patients treated with FOLFOX4.
The difference was significant with
a P value of less than .0001. Single-agent
oxaliplatin was associated with a response
rate of 1.1% which was not significantly
different from the control arm. Similar
results were obtained from using investigator-
assessed responses.
Determination of time to tumor progression
by the independent radiologic
review panel was based entirely on radiographic
data. In this analysis, time to tumor
progression was 2.6 months for the
5-FU/leucovorin control arm and 5.6
months for FOLFOX4, a difference that
again was statistically significant with a P
value less than .0001. Interestingly, patients
treated with single-agent oxaliplatin
did not fare as well, with a median
time to tumor progression of 1.9 months.
This difference was significant in favor of
the control arm with a
P value less than
.008.
Time to tumor progression determined
by investigators included not only
radiographic events of progressive disease
but also clinical parameters of disease progression such as death or clinical
worsening due to tumor-related symptoms.
Median time to tumor progression
was 1.9 months for the single-agent oxaliplatin,
2.6 months for 5-FU/leucovorin,
and 5.6 months for FOLFOX4.
Approximately two-thirds of patients
had tumor-related symptoms at baseline.
To meet the criteria for relief of tumorrelated
symptoms, patients had to have at
least one of the following: an improvement
in KPS by at least 20 points, a decrease
in pain intensity by at least 20 millimeters
on a 100 millimeter scale, decrease
in analgesic consumption by at least 50%,
or weight gain of at least 5%.
The proportion that had improvement
in at least one of these parameters lasting
for at least 4 weeks was 15% of patients
treated with 5-FU/leucovorin, and 28%
for patients treated with FOLFOX4. The
difference was significant with a P value of
less than .002, once again in favor of FOLFOX4.
Patients treated with single-agent
oxaliplatin had a 10% rate of relief from
tumor-related symptoms, which was not
significantly different from the control
arm.
Survival
Median survival in patients treated
with 5-FU/leucovorin was 8.7 months,
and patients treated with FOLFOX4 had a
median survival of 9.8 months. This difference
did not reach statistical significance
with a two-sided stratified log-rank
value of 0.07. Although patients treated
with FOLFOX4 had an overall 16% reduction
in death hazard ratio compared
to patients in the control arm, the 95%
confidence intervals overlap, indicating
that improved survival may not be one of
the advantages conferred by this therapy
in this second-line setting. Median survival
in patients treated with single-agent
oxaliplatin was 8.1 months.
Dr. Rothenberg posited three possible
explanations for the similarities in survival.
"First, the survival of the control group
was better than expected while survival of
the FOLFOX4 group was not as great as
expected. Second, salvage treatment with
oxaliplatin could have affected the survival
in the 5-FU/leucovorin control group.
About 42% of patients randomized to the
control group received oxaliplatin through
the treatment access program. However
the exact impact of this salvage therapy on
survival is difficult to assess accurately.
Finally it is possible that the additive effect
of FOLFOX4 on survival is brief when
used in a second-line setting and not detected
by the log rank method, which
gives greatest weight to late events."
Safety and Toxicity Data
Compared to the control arm, patients
who received FOLFOX4 had a greater
incidence of grade 3/4 diarrhea (12%
vs 2%), nausea (11% vs 3%), and vomiting
(9% vs 2%). Grade 3/4 neutropenia
occurred more commonly in patients
treated with FOLFOX4 (48%) than
in those patients treated with
5-FU/leucovorin (6%).
The clinical manifestations of neutropenia,
mainly neutropenic fever, were relatively uncommon, occurring in 5% of
patients treated with FOLFOX4 and 1%
treated with 5-FU/leucovorin. Grade 3/4
thrombocytopenia was also more common
with FOLFOX4 (5% vs 0), but there
were no episodes of severe bleeding in any
patients.
Preliminary results from this trial were
used as the basis for accelerated FDA approval
of oxaliplatin in August 2002.
