Chemotherapy is still considered a controversial issue in the treatment
of non-small-cell lung cancer (NSCLC). As single agents, relatively few
of the older drugs achieve a more than 15% response rate in advanced NSCLC,
but, among them, cisplatin (Platinol) has received considerable
attention. Other drugs with some activity are ifosfamide (IFEX), the vinca
alkaloids (eg, vinblastine [Velban] and vindesine), mitomycin (Mutamycin),
and the epipodophyllotoxins (eg, etoposide [VePesid] and teniposide [Vumon]).
The response rate associated with the established active agents in NSCLC
is in the range of 15% to 25%.
Cisplatin is considered the backbone agent for use in combination chemotherapy
against this disease. Combination chemotherapy regimens including cisplatin
have yielded response rates up to 50% and usually include two or three
of the drugs mentioned. One combination chemotherapy regimen commonly
used in Europe and the United States is cisplatin/etoposide, which produces
response rates in the range of 25% to 35%, with tolerable toxicity.
In the European Organization for the Research and Treatment of Cancer
(EORTC) Lung Cancer Cooperative Group, we recently completed a randomized
study in which the activity of single-agent teniposide, an analogue of
etoposide, was compared with that of combination teniposide/cisplatin.
In that same study, teniposide was tested at two different schedules: the
same total dose was given either in 1 day or in 3 refracted days. The study
accrued over 200 patients and was a 2 x 2 factorial design study.
The results of that study indicated that combination teniposide/cisplatin
was superior to teniposide alone in terms of response rate and both progression-free
and overall survival. Further, the treatment arm in which teniposide
was administered in 1 day only had a 0% response rate, suggesting that
the drug has a schedule-dependent activity, as shown for etoposide.
For these reasons, we elected to use the teniposide/cisplatin combination
as the standard arm for the current randomized study in patients with advanced
In recent years, several interesting new drugs have been developed that
exhibit significant activity in advanced NSCLC; among them are the taxanes,
paclitaxel (Taxol) and docetaxel (Taxotere); gemcitabine (Gemzar); a new
vinca alkaloid, vinorelbine (Navelbine); and water-soluble topoisomerase
I inhibitors such as irinotecan. Paclitaxel has shown a response rate of
20% to 25% in previously untreated patients with advanced NSCLC, with a
remarkably long 1-year survival. In these studies, paclitaxel was given
as a 24-hour infusion, with premedication with three antiallergic agents.
We decided to compare our standard arm of cisplatin/teniposide with
the new combination of paclitaxel/cisplatin. A report suggesting a possible
pharmacologic interaction between paclitaxel and cisplatin, with higher
toxicity when paclitaxel was given after cisplatin, prompted us to administer
paclitaxel before cisplatin in this study. Further, we decided to administer
paclitaxel as a 3-hour infusion instead of a 24-hour infusion, based on
a report of lower hematologic toxicity but similar activity with this schedule
compared with the more prolonged infusion time, in patients with advanced
Standard eligibility criteria were required to enter this study: Patients
were aged 18 to 75, had had no prior chemotherapy, had histologically proven
inoperable non-small-cell lung cancer, and were in relatively good condition
(Eastern Cooperative Oncology Group performance status 2 or less and adequate
marrow and organ functions). Prior radiotherapy was allowed to areas not
used as marker lesions for response assessment, but it had to have been
terminated at least 4 weeks before
study entry. Patients with brain metastases were not eligible. Informed
consent was obtained according to the regulations of each participating
Physical examination and complete blood counts and chemistries were
performed before study entry, along with imaging procedures appropriate
to adequately stage and assess the tumor deposits. Weekly blood counts
were performed to determine hematologic toxicity, and blood chemistries,
electrocardiogram, and thoracic x-ray were performed before every cycle
of therapy. Additional imaging techniques were repeated to evaluate the
response to treatment after two, four, and six cycles, and approximately
every 2 months thereafter. At treatment completion, patients were seen
at approximately 2-month intervals.
Patients were randomized to receive one of two cisplatin-based combination
chemotherapy regimens (Figure 1):
- Paclitaxel 175 mg/m² given on day 1 by 3-hour infusion, followed
by 1- to 2-hour infusion of cisplatin 80 mg/m² or
- 1- to 2-hour infusion of cisplatin 80 mg/m² on day 1, followed
by teniposide 100 mg/m² given on days 1, 3, and 5.
Cycles were repeated every 3 weeks in both arms.
Paclitaxel was dissolved in 500 mL 5% dextrose and infused before cisplatin;
teniposide was dissolved in 500 mL normal saline and infused over 1 hour,
administered after cisplatin. Cisplatin was dissolved in 500 mL saline
or 5% dextrose and infused in a program of forced diuresis consisting of
at least 2 L of fluid. Paclitaxel was administered after standard antiallergic
premedication, consisting of oral dexamethasone 20 mg 12 and 6 hours prior
to infusion and diphenhydramine 50 mg intravenously (IV) and cimetidine
300 mg IV 30 minutes before paclitaxel infusion. Prophylactic antiemetic
medication was left to the discretion of the investigators, but usually
included ondansetron (Zofran) 8 mg bid IV in combination with dexamethasone
8 mg bid IV during the cisplatin administration.
The duration of treatment was six cycles in patients whose disease did
not progress during treatment. In the paclitaxel arm only, however, investigators
were allowed to continue administration of paclitaxel alone to a total
of 10 cycles of chemotherapy during the initial randomized, phase II part
of the study. Following observation of a substantial increase of peripheral
neurotoxicity after more than six cycles with this treatment, the recommendation
was made to terminate chemotherapy after a maximum of six cycles. This
was implemented in the phase III part of the study.
Dose-reduction schemas and treatment delays were provided based on occurrence
of severe hematologic toxicities and nonhematologic side effects.
Trial Conduct and Statistical Analysis
Patients were stratified according to institution, performance status
(0 vs 1 vs 2), and extent of disease (stage III vs IV). Because cisplatin/paclitaxel
was a novel combination at the time of
design of this study, we started with a randomized phase II study. After
analysis of the results of the initial phase, which included a total of
80 patients, it was concluded that the standard arm produced similar
activity and toxicity to those observed in the previous study and that
the new experimental arm was both feasible and active.
Based on results of the randomized phase II study, it was then decided
to proceed with the randomized phase III study for which the major end
point was comparison of survival between the arms. An additional end point
of the phase III trial was comparison of the two arms for toxicity, response
rate, and progression-free survival. At selected centers, quality of life
was assessed by the Quality-of-Life-30 instrument with the lung module
in the phase III part of this study. Consequently, a secondary end point
of the phase III trial became comparison of quality of life issues in the
To ensure detectability of an improvement in median survival of 3 months
(from 7 to 10 months), it was planned that a total of 288 patients would
be entered into this trial; this number of patients would also suffice
to detect an increase in response rate from 24% to 42%.
A total of 332 patients (166 in each arm) were randomized by the EORTC
Lung Cancer Cooperative Group from July 1993 to February 1996, and this
report represents a preliminary analysis of the results of this phase III
trial. These results must be viewed with some caution, however, as external
review of radiologic responses is still ongoing and site visits have not
been completed yet; further, longer follow-up may reveal some changes in
the survival figures.
Overall, the two arms were well balanced for major prognostic factors:
approximately 70% of patients were male and had lost less than 5% of their
body weight, 90% had a performance status of 0 or 1, 50% had adenocarcinoma,
and 60% had stage IV disease. The major characteristics of the eligible
patients are reported in Table 1. Thirteen
patients were considered ineligible.
Hematologic toxicity was definitively more severe and more common in
arm B (cisplatin/teniposide) than in arm A (paclitaxel/cisplatin): In the
control arm (B) approximately 50% of patients developed grade 3/4 leukopenia,
almost 80% developed grade 3/4 neutropenia, and almost 40% developed
severe thrombocytopenia. The severe myelosuppression also led to treatment-related
infections in a substantial number of patients. These results are in agreement
with our prior experience with this regimen.
In contrast, the new investigational arm (A) was associated with relatively
mild hematologic toxicity, with 18%, 54%, and 3% grade 3/4 leukopenia,
neutropenia, and thrombocytopenia, respectively. Primarily due to myelosuppression,
there were more treatment delays and dose reductions in the standard arm
than in the investigational arm.
The nonhematologic side effects were mainly represented by alopecia
and nausea and vomiting in both arms. Due to the introduction of more potent
antiemetic medications, emesis was better controlled in this study than
in our previous study. A slightly higher incidence of peripheral neurotoxicity
and arthralgia/myalgia was reported in the paclitaxel arm. Peripheral neurotoxicity
did not represent a serious problem for patients who interrupted treatment
after a maximum of six cycles of chemotherapy. Low and similar frequencies
of severe hypersensitivity reactions and cardiac toxicity were observed
in both arms of the study.
Response and Survival
Among 264 patients evaluable so far, responses have been assessed in
47% of those in the paclitaxel-based treatment arm and 29% of those given
cisplatin/teniposide. Extramural radiologic response evaluation is still
being carried out, however, and these numbers are expected to decrease
slightly in both arms. In both arms, the vast majority of responses were
Given the preliminary nature of this analysis, it is premature to draw
conclusions about the survival comparisons; so far, however, survival appears
to be similar in the two arms, with a median survival time in the range
of 9 months. At this time, the number of deaths is nearly 200 in total.
Longer follow-up is necessary to allow meaningful comparisons.
The introduction of new active drugs to treat NSCLC, such as paclitaxel,
clearly offers new treatment options in this disease. We tested the new
combination of paclitaxel/cisplatin in patients with advanced, previously
untreated NSCLC and compared it with our best standard arm, a combination
of teniposide and cisplatin.
Preliminary analysis of this phase III trial indicates quite clearly
a difference in response rate and toxicity between the two arms: Both differences
favor the paclitaxel arm. So far, however, no major differences in survival
are evident between the two arms. Quality of life has been assessed in
selected centers and the data are being analyzed; this analysis will certainly
provide additional information useful for comparing the two arms. It should
be kept in mind, however, that pallia
tion is the major objective of these treatments in patients with advanced
NSCLC. In light of the present preliminary results, it is tantalizing to
consider the investigational arm of paclitaxel/cisplatin as the new standard
arm for the next randomized study in this disease.
1. Giaccone G: New drugs in non-small cell lung cancer, an overview.
Lung Cancer 12(suppl 1):S155-S162, 1995.
2. Splinter TAW, Sahmoud T, Festen J, et al: Two schedules of teniposide
with or without cisplatin in advanced non-small-cell lung cancer: A randomized
study of the EORTC lung cancer cooperative group. J Clin Oncol 14:127-134,
3. Slevin MC, Clark PI, Joel JP, et al: A randomized trial to evaluate
the effect of schedule on the activity of etoposide in small cell lung
cancer. J Clin Oncol 7:1333-1340, 1989.
4. Rowinsky EK, Gilbert MR, McGuire WP, et al: Sequence of taxol and
cisplatin: A phase I and pharmacologic study. J Clin Oncol 9:1692-1703,
5. Eisenhauer EA, ten Bokkel Huinink WW, Swenerton KD, et al: European-Canadian
randomized trial of paclitaxel in relapsed ovarian cancer: High-dose versus
low-dose and long versus short infusion. J Clin Oncol 12:2654-2666,
6. Postmus PE, Giaccone G, Debruyne C, et al: Results of the phase II
EORTC study comparing paclitaxel/cisplatin with teniposide/cisplatin in
patients with non-small cell lung cancer. Semin Oncol 23(suppl 12):10-13,