Download Eastern Cooperative Oncology Group Experience

Cooperative Oncology
Group Experience With
Chemotherapy in Advanced
Non-small Cell Lung Cancer*
of new agents were started, and the results of these studies
led to the most recent phase III trial comparing the
survival results for two dose levels of paclitaxel combined
with cisplatin and those obtained with etoposide/cisplatin.
Each of these phases will be discussed briefly in this
Philip Bonomi, AID
Early Phase III Trials
In the late 1970s and early 1980s, a variety of cisplatinand noncisplatin-containing combination regimens were
reported to produce response rates >30% in patients with
advanced NSCLC.1 The ECOG systematically evaluated
each of these regimens in relatively large phase III trials.
The results of two of these studies, summarized in Tables
1 and 2,2-3 showed none of the regimens was associated
with significantly superior survival. However, the MVP
(mitomycin/vinblastine/cisplatin) regimen produced the
best response rate in the first trial and a slightly but
significantly higher response rate among patients with
either adenocarcinoma or squamous cell carcinoma in the
second trial. These observations resulted in MVP being
selected as the reference regimen for future ECOG trials.
The major objective of the phase III trials was to
identify a regimen that significantly improved the survival
of patients with advanced NSCLC. Unfortunately, this
objective was not met. Similarly disappointing was the fact
that response rates in the ECOG trials were considerably
lower than those observed in smaller single-institution
phase II trials.1
Eastern
Eastern Cooperative Oncology Group (ECOG) inves¬
have tested a variety of single-agent and
combination regimens in patients with non-small cell
lung cancer (NSCLC) during the last 2 decades. The
following observations have been made. (1) The
tigators
mitomycin/vinblastine/cisplatin regimen produced a
trend for higher response rates in two studies and a
significantly higher response rate in a third study.
Survival, however, tended to be shorter in patients
receiving this regimen. (2) Carboplatin produced a
9% overall response rate and a median survival of
31.7 weeks, which was slightly but significantly
longer than the median survivals obtained with three
combination chemotherapy regimens. (3) Paclitaxel
produced an overall response rate of 21% and a
1-year survival rate of 40% in previously untreated
NSCLC patients. This observation led to a phase III
trial in which paclitaxel (135 mg/m2 and 250 mg/m2)
was combined with cisplatin and compared with
etoposide/cisplatin. Response rates for each of the
regimens (26% for 135 mg/m2
paclitaxel/cisplatin
for
250 mg/m2) were significantly
and
31%
paclitaxel
higher than the response rate for etoposide/cisplatin
(12%), but response between the two paclitaxel/
this
cisplatin arms was not significantly different. inAteach
a
toward
survival
there
is
trend
longer
point,
of the paclitaxel/cisplatin arms, but the final survival
have not been completed. In the next phase
analyses
III trial, ECOG will evaluate paclitaxel (135 mg/m2)
plus cisplatin in comparison to three other regi¬
mens.docetaxel/cisplatin,
and carboplatin/paclitaxel.
gemcitabine/cisplatin,
(CHEST 1998; 113:13S-16S)
T\ uring the past 16 years, the Eastern Cooperative
*-^ Oncology Group (ECOG) has conducted a series of
phase III and randomized phase II trials in advanced
non-small cell lung cancer (NSCLC). This period can be
divided into four phases, each of which is characterized by
a particular study design and philosophical approach.
III trials testing promising combina¬
Initially, large phase
regimens were conducted. Then ECOG began a
transitional phase during which single agents were com¬
pared with combination regimens and, simultaneously,
databases were analyzed. Next, randomized phase II trials
tion
*From the Section of Medical
Chicago.
Oncology, Rush Medical College,
MD, Director, Medical Oncol¬
Reprint requests: Philip Bonomi,
IL
1725 W Harrison
ogy, Rush Medical
60612-3838
College,
St,
Chicago,
review.
A Time of Transition.Single Agents vs
Combination Chemotherapy Regimens
Based on results from the phase III trials, ECOG investi¬
gators decided to test new agents in patients with previously
untreated advanced NSCLC. Considerable controversy ex¬
isted regarding the effect of chemotherapy on the survival of
patients with advanced NSCLC while this study design was
being developed. Randomized trials comparing chemother¬
apy with best supportive care had not been completed,4 and
ECOG investigators were concerned that initial treatment
with an ineffective single agent might have a detrimental
effect on survival. Thus, the first ECOG trial design called for
previously untreated patients to receive primary treatment
with one of three combination regimens or one of two single
agents followed by the MW regimen.5 The combination
regimens consisted of MVP (the reference regimen) vs
vinblastine/cisplatin, vs MVP alternating with CAMP (cycloCar¬
phosphamide/doxorubicin/methotrexate/procarbazine).
and
were
selected
as
the
single agents in
boplatin iproplatin
this trial. As had been observed in earlier trials, the MVP
regimen produced the best response rate, but in this trial, the
difference was statistically significant for all cell types (20%;
addition, both of the single agents produced
p=0.03). In lower
rates (ie, 9% for carboplatin
significantlyand 6% response
for iproplatin [p=0.001]). Surprisingly,
[p=0.001]
patients receiving carboplatin experienced a slightly but
significantly longer median survival (31.7 weeks; p=0.008),
whereas patients given MVP had a shorter median survival
(22.7 weeks; p=0.09). Median survival for iproplatin-treated
patients was 26.1 weeks, which was slightly longer than that
CHEST/113/1 /JANUARY, 1998 SUPPLEMENT
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21758/ on 04/29/2017
13S
1.Response Rates and Survival Data: Four
Cisplatin Regimens*
Table
Regimen
No. of
Patients
Response,
%
Median Survival,
wk
104
107
109
112
27
24
18
22
23.7
23.5
21.6
22.1
MVP
CAP
AFP
CBP
*CAP=cyclophosphamide/doxorubicin/cisplatin;
5-fluorouracil/cisplatin; CBP=cyclophosphamide/bleomycin-cisplatin. Adapted from Ruckdeschel et al.2
achieved with MVP and virtually identical to the median
survival observed with vinblastine/cisplatin and with MVP
alternating with CAMP. Response rates and median survival
durations for this study are summarized in Table 3.
Critics of this study have suggested that the superior
survival results obtained with carboplatin are purely coin¬
cidental, and certainly this is possible. It should be noted,
however, that this was a relatively large study, and not only
was carboplatin associated with slightly longer survival, but
survival in iproplatin-treated patients was virtually equiv¬
alent to that observed with the combination chemotherapy
regimens.
Iproplatin and carboplatin were also tested in a
randomized phase II trial conducted by the Cancer and
Leukemia Group B (CALGB).6 These investigators
that were similar to the findings of the
reported resultsThe
ECOG study.
response rates for carboplatin and
were 16% and 7%, respectively, and median
iproplatinwere
6.5 and 5 months, respectively. Although
survivals
in the CALGB trial were slightly higher
rates
response
and median survival was slightly shorter than those in
the ECOG trial, it is interesting to note that in both
trials, carboplatin achieved a higher response rate than
iproplatin, and that the differences in median survival
for carboplatin- vs iproplatin-treated patients in the
ECOG and CALGB studies were 5 and 6 weeks,
ECOG trial, both carboplatin and
respectively. In the
associated with significantly less lifeiproplatin were
threatening and lethal toxic reactions than the combi¬
nation regimens (p<0.0001). Although there is no
clear-cut explanation as to why carboplatin-treated pa¬
tients had a significantly lower response rate and a
significantly longer survival, these observations suggest
2.Response Rates and Survival Data: Four
Most Active Regimens*
Regimen
No. of
Patients
Response,
Median Survival,
wk
121
126
124
115
31
25
20
17
22.0
26.0
26.6
25.1
MVP
VP
EP
CAMP
*VP=vindesine/cisplatin; EP=etoposide/cisplatin. Adapted
Ruckdeschel
14S
et
al.3
Regimen
MVP
VP
MVP/CAMP
AFP=doxorubicin/
Table
Table 3.Response Rates and Survival Data:
Combination Chemotherapy vs Single Agents*
from
No. of
Patients
Response,
%
Median Survival,
wk
176
175
172
88
20
13
13
9
6
22.7
25.1
25.0
31.7
26.1
Carboplatin/MVP
Iproplatin/MVP
Adapted from Bonomi et al.5
explanation of abbreviations.
*
See text and Table 2 footnotes for
that increased toxic reactions from chemotherapy may
exert a detrimental effect on survival or that chemo¬
therapy may exert a positive effect on survival without
producing a high rate of tumor shrinkage.
Database Analyses
While the randomized trial comparing single agents and
combination chemotherapy regimens was ongoing, data
from 893 patients who had participated in the earlier
phase III trials23 were being analyzed.7 The following
observations were made: (1) the overall 1-year survival rate
for patients receiving chemotherapy was 19%; (2) the
etoposide/cisplatin regimen achieved the highest 1-year
survival rate (25%), although none of the regimens signif¬
icantly improved 1-year survival; and (3) although the
MVP regimen was associated with higher response rates in
each trial (Tables 1 and 2), MVP-treated patients had the
lowest 1-year survival rate (12%; p=0.08). These results
are similar to those observed in the transitional protocol in
which MVP was compared with other combination regi¬
mens and with single agents.5 The discordance between
response rate and survival both for the MVP regimen235
and for single-agent carboplatin5 emphasizes the need to
test promising treatments in phase III trials and should
alert clinicians to exercise caution when selecting treat¬
ments based on data from smaller phase II trials.
Examination of the toxicity data from the ECOG phase
III trial comparing the four most active regimens2 showed
that nonambulatory patients (ECOG performance status
of 2) experienced a 10% rate of lethal toxicity. In compar¬
ison, patients whose ECOG performance status was 0 or 1
experienced a 3% rate of lethal toxicity. Based on these
observations, ECOG has excluded patients with an ECOG
performance status of 2 from subsequent trials.
Drug Discovery
observed
no apparent detrimental effect of
Having
in the carboplatin trial,5 ECOG
on
survival
single agents
initiated a series of phase II trials of new agents in
NSCLC patients. Acivicin, the first
previously untreated
drug tested in the ECOG Drug Discovery Program, was
compared with etoposide/cisplatin (the reference regi¬
men) in a randomized phase II trial.8 The rationale for
the reference regimen was to assess the con¬
including
stancy of the ECOG patient population. In early studies,1-2
the overall response rates for MVP were 27% and 31%,
International
Symposium on Thoracic Malignancies
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21758/ on 04/29/2017
while in a more recent trial,5 the response rate was 20%.
The observations suggest a trend toward lower response
rates in more recent studies. In fact, in the trial comparing
acivicin with etoposide/cisplatin,8 the overall response rate
for etoposide/cisplatin was 11%. The reasons for the
apparent lower response rates are not readily recognized,
but it is possible that responses are being defined more
rigorously because of the increasing use of CT scans.
In the Drug Discovery Program, the ECOG observed
no significant activity with six consecutive single agents9
(Alex Chang, MD; personal communication; April
1996), including acivicin.8 However, paclitaxel at a dose
of 250 mg/m2 given IV over 24 h produced a 21% overall
response rate in 24 patients.9 Investigators at M.D.
Anderson Cancer Center, using a 24-h infusion of
paclitaxel at a dose of 200 mg/m2, observed a 24%
response rate in 25 patients.10 Perhaps more important,
both studies910 reported a 40% 1-year survival rate.
Although the confidence intervals for the 1-year survival
rates observed in these small trials are wide, these
results are nevertheless encouraging considering the
1-year survival rate observed with previously tested
combination regimens was 19%.7
Testing the Impact of New Agents on Survival
Encouraged by the 40% 1-year survival rate observed
with paclitaxel,9 ECOG investigators decided to test the
survival effect of paclitaxel in a large phase III trial.
Etoposide/cisplatin was compared with paclitaxel, 135
mg/m2, combined with cisplatin, 75 mg/m2; and pacli¬
taxel, 250 mg/m2, combined with cisplatin, 75 mg/m2,
plus granulocyte colony-stimulating factor, 5 |xg/kg
given daily following chemotherapy.11 Paclitaxel was
given as a 24-h IV infusion. Etoposide/cisplatin was
chosen as the reference regimen because it had pro¬
duced the highest 1-year survival rate (25%) in a prior
ECOG trial.7 The lower dose of paclitaxel (135 mg/m2)
was selected because it had shown significant activity
and acceptable toxicity when combined with cisplatin,
75 mg/m2, in women with ovarian carcinoma.12 The
higher dose of paclitaxel (250 mg/m2) combined with
cisplatin was selected based on the results of a phase I
trial conducted at Johns Hopkins.13 In this study, the
use of granulocyte colony-stimulating factor overcame
the dose-limiting toxic reaction of granulocytopenia and
fever. At doses higher than 250 mg/m2, however, neurotoxicity became prohibitive.
Accrual to the ECOG phase III study was completed
rapidly (600 patients in 16 months). The paclitaxel/cispla¬
tin regimens produced significantly higher response rates
(26% for 135 mg/m2 and 31% for 250 mg/m2 paclitaxel)
than etoposide/cisplatin (12%).n There was, however, no
significant difference in response rates between the pacli¬
taxel-containing
regimens. Preliminary analyses have
shown a trend toward longer survival in patients receiving
each of the paclitaxel-containing regimens. Additional data
are being collected, and final survival analyses will be
completed shortly. Significant grade 4 granulocytopenia
was observed with each regimen, but the incidence of
febrile episodes during granulocytopenia was relatively
low (<10%).n In addition to the major objective of this
trial (evaluation of survival, response, and toxicity), serial
quality of life data have been collected and are currently
being analyzed.
Future Directions
During the last 5 years, at least five new agents with
response rates >20% in advanced NSCLC have been
identified.14 Two of these compounds, docetaxel and
gemcitabine, will be combined with cisplatin and tested in
the next phase III trial. In addition, the carboplatin/
paclitaxel regimen, which has produced relatively high
response rates and a median survival of approximately 1
year in two relatively large phase II trials,1516 will be
evaluated in this study. Paclitaxel (135 mg/m2 given IV
over 24 h) combined with cisplatin (75 mg/m2) will serve
as the reference regimen in this study, and comparison of
survival will be the primary end point.
References
1 Bonomi P. Brief overview of combination chemotherapy in
non-small cell lung cancer. Sernin Oncol 1986; 13(suppl
3):89-91
2 Ruckdeschel JC, Finkelstein DN, Ettinger DS, et al. Chemo¬
therapy for metastatic non-small cell bronchogenic carci¬
noma: EST:2575, generation V-a randomized comparison of
four
cisplatin-containing regimens. J
Clin Oncol 1985;
3:72-79
JC, Finkelstein DN, Ettinger DS, et al. A
randomized trial of the four most active regimens for meta¬
static non-small cell lung cancer. J Clin Oncol 1986; 4:14-22
4 Grilli R, Oxman AD, Julian JA. Chemotherapy for advanced
non-small cell lung cancer: how much benefit is enough?
J Clin Oncol 1993; 11:1866-72
5 Bonomi P, Finkelstein DN, Ruckdeschel JC, et al. Combina¬
tion chemotherapy versus single agents followed by combi¬
nation chemotherapy in stage IV non-small cell lung cancer:
a study of the Eastern Cooperative Oncology Group. J Clin
7:1602-13
6 Kreisman H, Ginsberg S, Propert KS, et al. Carboplatin or
iproplatin in advanced non-small cell lung cancer: a Cancer
and Leukemia Group B study. Cancer Treat Rep 1987;
3 Ruckdeschel
Oncol'l989;
71:1049-52
7 Finkelstein DN,
Ettinger DS, Ruckdeschel JC. Long-term
lung cancer: an Eastern
Cooperative Oncology Group study. J Clin Oncol 1986;
survivors in metastatic non-small cell
4:702-09
P, Finkelstein D, Chang A. Phase II trial of acivicin
etoposide-cisplatin in non-small cell lung cancer: an
Eastern Cooperative Oncology Group study. Am J Clin Oncol
1994; 17:215-17
K, Glick J, et al. Phase II study of Taxol,
Chang AY, Kim
and
in
IV non-small cell
8 Bonomi
versus
9
merbarone,
stage
lung
Cooperative Oncology Group results.
J Natl Cancer Inst 1993; 85:388-94
Murphy WK, Fossella FV, Winn RJ, et al. Phase II study of
Taxol in patients with untreated advanced non-small cell lung
cancer. J Natl Cancer Inst 1993; 85:384-88
cancer:
10
piroxantrone
the Eastern
K, Chang A, et al. Phase III trial comparing
etoposide-cisplatin versus Taxol with cisplatin-G-CSF versus
Taxol-cisplatin in advanced non-small cell lung cancer: an
Eastern Cooperative Oncology7 Group trial [abstract]. Proc
11 Bonomi P, Kim
Am Soc Clin Oncol
1996; 15:382
CHEST/113/1 /JANUARY, 1998 SUPPLEMENT
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21758/ on 04/29/2017
15S
WJ, Brady MF, et al. Cyclophospha¬
with paclitaxel and cisplatin in
cisplatin compared
patients with stage III and stage IV ovarian cancer. N Engl
J Med 1996; 334:1-6
Rowinsky EK, Chandra V, Forastierre A, et al. Phase I and
and cisplatin with granulo¬
pharmacologic study of paclitaxel
cyte colony stimulating factor: neuromuscular toxicity is dose
limiting, j'clin Oncol 1993; 11:2010-20
Lilenbaum RC, Green MP. Novel chemotherapeutic agents
in treatment of non-small cell
12 McGuire WP, Hoskins
mide and
13
14
16S
11:1391-1402
15
Langer CJ, Leighton JC,
carboplatin
lung cancer. J Clin Oncol 1993;
Comis Rl, et al. Paclitaxel and
in combination in the treatment of advanced
non-small cell lung
cancer: a
phase II response and survival
analysis. J Clin Oncol 1995; 13:1860-70
16 Muggia EM, Vofai D, Natale R, et al. Paclitaxel 3 hour infusion
given alone and combined with carboplatin: preliminary results
of dose escalation studies. Sernin Oncol 1995; 15:382-87
International
Symposium on Thoracic Malignancies
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21758/ on 04/29/2017