Combination;therapy. Several key trials have
been conducted to address whether platinum-based combination therapy is superior to single-agent therapy in recurrent platinum-sensitive disease. The results of 2 phase 3 trials were pooled
and reported together as ICON 4/AGO.62 In this
combined study of these 2 parallel-run trials, 802
patients with platinum-sensitive recurrent ovarian
cancer were enrolled between 1996 and 2002.
Eligible patients were required to previously have
received a platinum-based regimen at the time
of initial diagnosis. The 2 trials differed in terms
of the TFI: the TFI needed to be greater than 12
months for patients to be considered eligible in
the ICON 4 trial and greater than 6 months in the
AGO trial. In the ICON 4 trial, patients were randomized to receive either single-agent platinum or
platinum with paclitaxel. There was an improved
response rate (RR, 66% vs 54%, P = 0.06) and
improved PFS (50% progression-free at 1 year vs
40%, P < 0.001) with the paclitaxel with platinum
regimen; the group receiving this regimen also had
an improved OS (57% alive at 2 years vs 50%,
P = 0.023). In the AGO OVAR 2. 5 study, patients
considered platinum-sensitive were stratified to
carboplatin or carboplatin/gemcitabine. The study
demonstrated an improved RR (47% vs 31%,
P = 0.0016) and PFS ( 8. 6 vs 5. 8 months, P = 0.0038)
but no improvement in survival ( 18 vs 17. 3 months,
P = 0.7349). The study was insufficient to justify
approval of the combination.63
Next GEICO (Grupo Espanol de Investigacion
en Cancer de Ovari) conducted a phase 2 trial
that randomly assigned 81 patients to treatment
with carboplatin or carboplatin plus paclitaxel
and yielded similar results, with a response rate
of 75.6% in the carboplatin/paclitaxel arm versus
50% in the carboplatin alone arm.64 In another ran-
domized phase 3 trial, GCIG (Gynecologic Cancer
Intergroup) compared combination gemcitabine
and carboplatin therapy with carboplatin alone
in platinum-sensitive recurrent disease. Interim
analysis of the data revealed an increased RR
with the combination of gemcitabine/carboplatin
therapy (47.2% vs 30.9%, P = 0.0016).65 The HR
for median OS was 0.96, and the HR for PFS
was 0.72, with a median PFS of 8. 6 months in the
combined therapy arm versus 5. 8 months in the
carboplatin alone arm. Based upon these findings,
the FDA approved gemcitabine in combination with
carboplatin for use in women with advanced ovar-
ian cancer who relapse at least 6 months after
completion of previous platinum-based treatment.
Issues have been raised regarding the results
from the ICON 4/AGO, GEICO, and GCIG trials.
For example, a relatively low number (40%) of
patients in ICON 4 had received a taxane during
their initial therapy. However, 87.2% of the patients
received a taxane as part of their initial therapy in
the GEICO trial.64 In addition, these trials do not
address the possible role of sequential therapy in
comparison with combined therapy. Combination
regimens do result in higher response rates and
will likely benefit the symptomatic patient more rap-
idly, but they also carry a higher rate of toxicity. In
the ICON 4/AGO and GEICO trials, approximately
20% of the combined platinum and paclitaxel
group experienced grade 2 to 4 neurological toxic-
ity, and in the GCIG study, there was a 78.3% inci-
dence of grade 3 or 4 hematologic toxicity with the
combined carboplatin and gemcitabine regimen.65
The GCIG CALYPSO trial is the largest randomized clinical trial evaluating platinum-sensitive
ovarian cancer. A total of 976 women were enrolled into this phase 3 study of PLD (30 mg/m2)
plus carboplatin (AUC 5) every 4 weeks versus
carboplatin (AUC 5) and paclitaxel given every 4
weeks. The primary endpoint was the noninferiority
