Vol. 4, 2723-2728, November 1998 Clinical Cancer Research 2723 Superiority of Sequential versus Concurrent Administration of Paclitaxel with Etoposide in Advanced Non-Small Cell Lung Cancer: Comparison of Two Phase II Trials’ Enriqueta Felip, Bartomeu Massuti, complete responses and seven partial responses, for an over- Carlos Camps, Dolores Benito, Dolores Isla, all response rate of 37.5% (95% confidence interval, 21- 58%). In conclusion, toxicity and antitumor activity of the Jose Luis Gonz#{225}lez-Larriba, paclitaxel/etoposide combination may be sequence depend- M. Paz L#{243}pez-Cabrerizo, Oscar Salamanca, ent. Our findings suggest that etoposide followed by pacli- Jos#{233}Puerto-Pica, Aifredo Moyano, Jos#{233}Baselga, taxel is well tolerated and has greater activity in NSCLC and Rafael Rosell2 than concurrent administration. Medical Oncology Department, Hospital General Vail d’Hebron, 08035 Barcelona [E. F., J. B.]; Hospital General de Alicante, 03010 INTRODUCTION Alicante [B. M.]; Hospital General de Valencia, 46014 Valencia The treatment of patients with metastatic or locally ad- [C. C.]; Bristol-Myers Squibb Company, 28040 Madrid [D. B., 0. SI; Hospital ClInico Universitario de Zaragoza, 50009 Zaragoza [D. I.]; vanced NSCLC3 remains unsatisfactory (1 , 2). Chemotherapy Hospital ClInico San Carlos, 28040 Madrid [J. L. G-L.]; Hospital trials in NSCLC have often focused on platinum-based chemo- Germans Trias y Pujol, 08916 Badalona, Barcelona [M. P. L-C., therapy, although such regimens have led to only marginal R. R.]; Hospital Infanta Cristina, 06080 Badajoz [J. P-P.] and Hospital improvements in survival (2-7). ln the past few years, several Ram#{243}nyCajal, 28034 Madrid [A. M.], Spain new chemotherapy agents have demonstrated activity in ad- vanced NSCLC. Paclitaxel (Taxol#{174}; Bristol-Myers Squibb ABSTRACT Company, Princeton, NJ) is a novel chemotherapeutic agent that Paclitaxel and etoposide are two chemotherapy agents enhances microtubule assembly, inhibits the depolyrnerization with broad cytotoxic activity and different mechanisms of of tubulin, and has been shown to have promising preclinical action and resistance. Preclinical studies of their combined activity in NSCLC (8, 9). In several published trials. single- cytotoxicity have yielded conflicting results. We performed agent paclitaxel has been reported to have significant activity in two sequential Phase II trials using different sequence NSCLC as first-line therapy, with response rates between 21% schedules of paclitaxel and etoposide as first-line treatment and 36% and 1-year survival rates of -40% (10-12). This level in advanced non-small cell lung cancer (NSCLC). Forty-four of antitumor activity is at least similar to previously available patients with stage 11113 or LV NSCLC were included be- agents, and in addition, paclitaxel can be given in combination tween July 1995 and September 1996. All patients received with other active agents in NSCLC such as carboplatin, cispla- etoposide at 100 mg/m2, given as an i.v. infusion on days 1, tin, and etoposide. The combination of paclitaxel and cisplatin 2, and 3. The first 20 patients (part A) also received pacli- or carboplatin has been tested previously (13-16). Nonetheless, taxel at 175 mg/m2 as a 3-h infusion on day 1, immediately platinum-containing regimens are associated with substantial prior to etoposide. The subsequent 24 patients (part B) were toxicity, and therefore, newer combinations with better toxicity given the same paclitaxel dose, but on day 4. Grade 3-4 profiles and greater efficacy are needed. This trial was designed granulocytopenia was seen in 70% of the patients in part A to explore the combination of paclitaxel and etoposide in ad- vanced NSCLC. Etoposide (VePesid; Bristol-Myers Squibb and in 37% of those in part B (P = 0.04). Twenty-five % of the courses in part A and 4% of the courses in part B were Company) is a topoisomerase II inhibitor that, as a single agent associated with granulocyte nadir 5OO/d (P = 0.00006). administered either p.o. or i.v., has demonstrated activity in No responses were observed in part A, although disease was NSCLC with response rates of 10-15% ( 17). Our rationale for stabilized in 14 patients (70%). In part B, there were two studying paclitaxel in combination with etoposide was based on several observations: both drugs are active in NSCLC, each drug has a different mechanism of action, and both drugs have a toxicity profile that makes them suitable to be studied in com- bination. Recent preclinical studies have highlighted the prob- Received 5/1/98; revised 8/7/98; accepted 8/10/98. lem posed by the combined cytotoxicity of these two agents (18, part The costs of publication of this article were defrayed in by the 19). Perez et a!. (19) evaluated the effects of paclitaxelietopo- payment of page charges. This article must therefore be hereby marked side scheduling in cell growth inhibition in lung human cancer advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. cell lines, reporting that little antagonism was observed when I The authors participated in this study on behalf of the Spanish Lung Cancer Group. This work was supported in part by a grant from Bristol-Myers Squibb Company. 2 To whom requests for reprints should be addressed, at Hospital Ger- mans Trias i Pujol, Medical Oncology Service, Box 72, 08916 Bada- :; The abbreviations used are: NSCLC, non-small cell lung cancer; ANC, lona, Barcelona, Spain. Phone: 34-93-4650409; Fax: 34-93-3954206; absolute neutrophil count; CBC. complete blood cell; PR, partial re- E-mail: [email protected]. sponse; SD, stable disease; Cl, confidence interval. Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 1998 American Association for Cancer Research. 2724 Sequence-dependent Paclitaxel/Etoposide Schedule paclitaxel and etoposide were given concurrently, although syn- of <5004a1 for 7 days or a platelet nadir of <50,000/pA for ergisrn was observed when paclitaxel followed etoposide expo- 7 days, and it was reduced two steps (1 10 mg/rn2) in the event sure. The evidence that sequence-dependent effects of pacli- of febrile neutropenia. The etoposide dose was reduced to 75% taxel/etoposide may be crucial in synergism or antagonism if the neutrophil nadir was <500/pA or the platelet nadir was between the two drugs prompted us to combine paclitaxel and <50,000/pl. If, on day 21, neutrophil count was < 1,500/pi or etoposide in two sequential Phase II trials. In the first Phase II the platelet count was < l00,000/pJ, treatment was delayed 1 study (part A), paclitaxel and etoposide were administered con- week or until recovery. The use of hematopoietic growth factor currently, whereas in the second (part B), etoposide preceded was restricted to patients who developed neutropenic fever. For paclitaxel administration. nonhernatological toxicities, the following criteria were consid- ered for dose reductions: the paclitaxel dose level was reduced one step in the presence of grade 2 stomatitis and two steps for PATIENTS AND METHODS grade 3 neuropathy, and paclitaxel was withheld if grade 4 Patient Eligibility. Entry criteria for these two sequential neuropathy occurred. Phase II studies were identical. All patients had histologically Scans or films of indicator lesions were performed for confirmed stage IIIB or IV NSCLC and were chemotherapy response evaluation after three cycles of therapy or sooner, if the naive. Prior radiotherapy was allowed. provided that measurable patient appeared to have disease progression. Standard Eastern lesions were outside the radiotherapy treatment field. All par- Cooperative Oncology Group criteria for solid tumor response ticipants were required to have age of 18 years, an Eastern were used to evaluate response. All patients received a total of Cooperative Oncology Group performance status of 2, a life six cycles of treatment, unless there was evidence of progressive expectancy of >3 months, and adequate hematological (ANC of disease. On completion of 6 cycles, patients who had complete 2,000/i.l and platelet count of > l00,000/xl), hepatic (total response or PR continued therapy up to a maximum of 10 bilirubin level of 2 mg/dl), and renal (creatinine concentration cycles. Toxicity from treatment was graded according to stand- of < 1 .5 mg/dl) function. Written informed consent was ob- ard WHO criteria. Following completion of treatment, patients tamed from all patients, and the protocol was approved by the were evaluated every 3 months up to 2 years and then every 6 institutional ethics committees of the participating centers. Pa- months. tients were excluded from the trial for any of the following: In part A, 20 eligible patients were to be included. If one or history of prior malignancies, pregnancy, history of atrial or more objective responses were observed, an additional 15 eli- ventricular arrhythmias or congestive heart failure, documented gible patients were to be accrued. Such a design has a 98% myocardial infarction in the preceding 6 months, history of power to detect an objective response rate of 20% after accrual second- or third-degree heart block, preexisting motor or sen- of the first 20 patients. Part A of the trial was, in fact, terminated sory neuropathy greater than grade II according to WHO crite- after the accrual of 20 patients because no response was ob- na, and history of sensitivity to cremophor. Patients with brain served. To have comparable sample size in part B, which was metastases were also excluded. conducted subsequently, it was decided that a similar number of Before protocol enrollment, all patients underwent a com- patients should be accrued. plete history and physical examination. Laboratory evaluation Overall response rate was calculated for all patients en- included a CBC count, electrolytes, liver function tests, urea, tered, with no attempt to exclude patients whose response was creatinine, and urinalysis. Electrocardiography.