VOLUME

27

NUMBER

FEBRUARY

2009

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Phase III Study of Immediate Compared With Delayed Docetaxel After Front-Line Therapy With Gemcitabine Plus Carboplatin in Advanced NonSmall-Cell Lung Cancer
Panos M. Fidias, Shaker R. Dakhil, Alan P. Lyss, David M. Loesch, David M. Waterhouse, Jane L. Bromund, Ruqin Chen, Maria Hristova-Kazmierski, Joseph Treat, Coleman K. Obasaju, Martin Marciniak, John Gill, and Joan H. Schiller
From the Massachusetts General Hospital, Boston, MA; Penrose Cancer Center of Kansas, Wichita, KS; Missouri Baptist Cancer Center, St Louis, MO; Central Indiana Cancer Centers; Eli Lilly & Co, Indianapolis, IN; Oncology Hematology Care, Inc, Cincinnati, OH; and University of Texas Southwestern Medical Center, Dallas, TX. Submitted March 27, 2008; accepted September 17, 2008; published online ahead of print at www.jco.org on December 15, 2008. Supported by Eli Lilly & Co, Indianapolis, IN. Presented at the 42nd Annual Meeting of the American Society of Clinical Oncology, June 2-6, 2006, Atlanta, GA, and the 43rd Annual Meeting of the American Society of Clinical Oncology, June 1-5, 2007, Chicago, IL. Authors disclosures of potential conicts of interest and author contributions are found at the end of this article. Clinical Trials repository link available on JCO.org. Corresponding author: Panos M. Fidias, MD, Massachusetts General Hospital, Center for Thoracic Cancers, 55 Fruit St, Boston, MA 02114; e-mail: pdias2@partners.org. 2008 by American Society of Clinical Oncology 0732-183X/09/2704-591/$20.00 DOI: 10.1200/JCO.2008.17.1405

Purpose Gemcitabine plus carboplatin (GC) is active as front-line treatment for advanced nonsmall-cell lung cancer (NSCLC). For patients without progression, timing of second-line chemotherapy for optimum clinical benet remains uncertain. This phase III, randomized trial assessed the efcacy and safety of docetaxel administered either immediately after GC or at disease progression. Patients and Methods The chemotherapy-nave patients enrolled had either stage IIIB NSCLC with pleural effusion or stage IV NSCLC. Gemcitabine (1,000 mg/m2) was administered on days 1 and 8 followed by carboplatin (area under the curve 5) on day 1. After four 21-day cycles, patients who did not have progression were randomly assigned either to an immediate docetaxel group (docetaxel 75 mg/m2 on day 1 every 21 days, with maximum of six cycles) or to a delayed docetaxel group. The primary end point was overall survival (OS) measured from random assignment. Additional analyses included tumor response, toxicity, progression-free survival (PFS), and quality of life (QOL). Results Enrollment totaled 566 patients; 398 patients completed GC; 309 patients were randomly assigned equally to the two docetaxel treatment groups. Toxicity proles were generally comparable for the docetaxel groups. Median PFS for immediate docetaxel (5.7 months) was signicantly greater (P .0001) than for delayed docetaxel (2.7 months). Median OS for immediate docetaxel (12.3 months) was greater than for delayed docetaxel (9.7 months), but the difference was not statistically signicant (P .0853). QOL results were not statistically different (P .76) between docetaxel groups. Conclusion We observed a statistically signicant improvement in PFS and a nonstatistically signicant increase in OS when docetaxel was administered immediately after front-line GC, without increasing toxicity or decreasing QOL. J Clin Oncol 27:591-598. 2008 by American Society of Clinical Oncology

INTRODUCTION

Nonsmall-cell lung cancer (NSCLC) is the leading cause of cancer-related deaths in the United States.1 Current treatment strategy for advanced NSCLC includes front-line chemotherapy using a platinumbased regimen, with the addition of the targeted agent bevacizumab for certain patients.2 A number of chemotherapy agents, including gemcitabine, docetaxel, and paclitaxel, are active in front-line treatment as part of a platinum-based combination.3,4 The combination of gemcitabine plus carboplatin (GC), with known activity in advanced NSCLC,5-8 was used as front-line chemotherapy in the current trial.

On the basis of reports of several randomized trials evaluating the duration of front-line chemotherapy, full efcacy benet for NSCLC patients is achieved after three to four treatment cycles.9 A high percentage of treated patients will respond or have their disease stabilized. However, prolonged frontline, platinum-based chemotherapy does not seem to provide additional benet.10 After front-line therapy, historical practice has included monitoring NSCLC patients who do not have progression and initiating treatment with a non cross-resistant agent on progression of disease (PD). Several chemotherapy agents, including docetaxel, erlotinib, and pemetrexed, have shown efcacy and been approved by the US Food and Drug
2008 by American Society of Clinical Oncology

591

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Fidias et al

Administration for second-line treatment of patients with locally advanced or metastatic NSCLC.11-14 A number of studies have evaluated regimens using either sequential or maintenance chemotherapy as second-line treatment for NSCLC patients who are not experiencing disease progression. A review of these studies suggests that optimal timing and duration of second-line therapy remain unclear.15 The current randomized phase III trial compared the efcacy and safety of docetaxel administered for a xed duration to patients without progression either immediately after completion of front-line GC therapy or at the time of PD. Our primary end point was overall survival (OS). Secondary end points were response rate, progressionfree survival (PFS), toxicity, and quality of life (QOL).
PATIENTS AND METHODS
Patient Eligibility Enrolled patients were 18 years old with histologically or cytologically conrmed stage IIIB plus pleural effusion or stage IV NSCLC; life expectancy of 12 weeks; Eastern Cooperative Oncology Group performance status (PS) of 0, 1, or 2; and adequate renal, hepatic, and bone marrow function. Exclusion criteria included prior adjuvant or neoadjuvant chemotherapy for NSCLC, active or ongoing infection, and symptomatic brain metastases (treated stable brain metastases were allowed). Radiation therapy ( 25% of bone marrow) was allowed 3 weeks before enrollment provided that patients had recovered from all adverse effects. All patients provided written informed consent. Institutional review boards approved the trial protocol before patient enrollment. This study complied with the principles of good clinical practice, the Helsinki Declaration, and federal and institutional guidelines. A data safety monitoring board at Dana-Farber/Harvard Cancer Center provided interim reviews of safety and efcacy. Treatment Plan During the GC phase, gemcitabine 1,000 mg/m2 was administered as a 30- to 60-minute intravenous (IV) infusion on days 1 and 8 followed by IV carboplatin (area under the curve 5.0) over 30 to 60 minutes on day 1. Cycles were repeated every 21 days for four cycles. Patients with no progression (ie, patients without PD after four GC cycles) were randomly assigned to one of two docetaxel treatment arms. Patients entering the immediate docetaxel arm received IV docetaxel 75 mg/m2 over 60 minutes on day 1 of a 3-week cycle until PD, to a maximum of six cycles. Patients who entered the delayed docetaxel arm received best supportive care (BSC) until rst evidence of PD. Patients t for chemotherapy at that time received the same docetaxel regimen as in the immediate docetaxel arm. Patients with PD who were unt for chemotherapy received BSC and continuous follow-up. Patients were premedicated with oral dexamethasone 4 mg twice daily on the day before, the day of, and the day after docetaxel treatment. Dose adjustments were based on absolute neutrophil count, platelets, and nonhematologic toxicities. Granulocyte colony-stimulating factor was used only for patients with absolute neutrophil count less than 0.5 109/L, febrile neutropenia, or documented infection with neutropenia. Use of erythropoietin was allowed. Patients who had drug withheld during GC treatment resumed therapy if toxicities resolved to grade 2. Patients who had drug withheld during docetaxel therapy resumed therapy at a 25% dose reduction if nonhematologic toxicities were resolved to grade 1. During docetaxel therapy, treatment was discontinued if patients experienced grade 3 neurotoxicity, grade 4 hypersensitivity reaction, or grade 3 aberration in liver function lasting more than 3 weeks. Patient Evaluations Before entry, patients received a complete history; physical examination; baseline evaluations of PS, QOL, and toxicity symptoms; comprehensive laboratory tests; and radiologic imaging plus physical evaluation of palpable
592
2008 by American Society of Clinical Oncology

lesions to assess known disease. Tumor response was evaluated using Response Evaluation Criteria in Solid Tumors16 for measurable disease in combination with assessments of nonmeasurable disease. Patients underwent disease assessment before cycle 3 of GC treatment and at restaging after cycle 4. Patients experiencing PD during or immediately after GC treatment were discontinued from the trial and observed for survival. For patients assigned to immediate docetaxel therapy, docetaxel treatment was initiated from day 21 up to day 35 after the start of cycle 4 during the GC phase, and disease scans obtained during docetaxel treatment were compared with baseline scans administered before GC therapy. Patients assigned to delayed docetaxel therapy were given standard evaluations including physical examinations; laboratory tests; and PS, toxicity, and QOL assessments every 3 weeks after random assignment until PD. Patients in the delayed docetaxel arm underwent disease assessment every 3 months until PD. During docetaxel therapy, standard evaluations were performed for each arm at the time of each cycle visit, and disease assessments were completed immediately after cycles 2 and 4 and 21 to 25 days after completion of cycle 6. Patients experiencing PD during docetaxel treatment were observed for survival, and patients discontinuing for reasons other than PD were observed for PD and survival. Patients were assessed for QOL using the Lung Cancer Symptom Scale (LCSS) questionnaire17 2 weeks before starting GC and at restaging after GC cycle 4. For the delayed docetaxel arm, LCSS was administered every 3 weeks until PD. During docetaxel therapy, LCSS was administered before cycles 2 to 6. Patients who completed six docetaxel cycles and achieved complete response (CR), partial response (PR), or stable disease (SD) were given the LCSS 21 to 30 days after last dose and at 3 months after last dose unless the patient received subsequent therapy or PD was noted. Statistical Considerations Before trial, the expected OS for delayed second-line docetaxel therapy was 9 months. To measure a meaningful improvement (4 months) in OS, 238 death events were required to achieve 80% statistical power with a two-sided .05. Anticipating that 40% of GC-treated patients would have PD, an enrollment of 550 patients to GC treatment was expected to deliver 294 patients for random assignment to the docetaxel treatment arms. Interim efcacy analysis using Pocock stopping rules18 was performed when 50% of patients in each arm had died, were lost to follow-up less than 24 months after random assignment, or were observed for 24 months after random assignment. Patients were centrally randomly assigned by Ingenix Pharmaceutical Services (Basking Ridge, NJ). Patients not randomly assigned because of PD, toxicity, or other reasons were included in the description of patient discontinuations. Randomly assigned patients who did not receive docetaxel treatment, for whatever reason, were incorporated into the intent-to-treat analysis, encompassing the phase of the study from random assignment onward. Patients receiving at least one dose of treatment were included in the safety analysis. OS and PFS were estimated from the date of random assignment using the Kaplan-Meier method,19 and results for each treatment arm were compared using the log-rank test.20 For response rate, 95% CIs were calculated using the exact method based on the binomial distribution. Distributions of changes from baseline in the average symptom burden index (ASBI) of the patient LCSS for each treatment arm were compared using the MantelHaenszel 2 test.21 Patient characteristics and toxicities were summarized in descriptive statistics. Toxicities were dened using the National Cancer Institute Common Toxicity Criteria, version 2.0.22

RESULTS

Patient Characteristics From February 2002 through October 2005, 566 patients were enrolled at 51 institutions in the United States. Figure 1 provides a ow diagram for patient disposition throughout the trial. Three patients were excluded before starting GC; one patient died, a second patient
JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Immediate v Delayed Docetaxel After Front-Line GC in NSCLC

Enrolled to GC Phase (N = 566)

Excluded (n = 3) Reasons: Lost to follow up, death, unknown Received cycle 1 (n = 563) Received cycle 2 (n = 507) Received cycle 3 (n = 432) Received cycle 4 (n = 398) Off study Reasons: Disease progression Death Adverse event Patient decision Investigator decision Other Missing (n = 254) (n = 147) (n = 32) (n = 22) (n = 19) (n = 11) (n = 19) (n = 4)

Nonprogessors randomly assigned (n = 309)

Immediate Docetaxel Allocated to intervention (n = 153) Received cycle 1 (n = 145) Did not receive allocated intervention (n = 8) Reasons: Patient decision (n = 5) Disease progression (n = 2) Adverse event (n = 1)

Delayed Docetaxel Allocated to intervention (n = 156) Received cycle 1 (n = 98) Did not receive allocated intervention (n = 58) Reasons: Disease progression (n = 25) Patient decision (n = 12) Death (n = 5) Investigator decision (n = 4) Study closure (n = 2) Adverse event (n = 1) Other (n = 6) Missing (n = 3) Lost to follow-up Discontinued intervention Reasons: Disease progression Adverse event Patient decision Death Investigator decision Other Analyzed Excluded from analysis (n = 1) (n = 121) (n = 66) (n = 6) (n = 19) (n = 9) (n = 4) (n = 16) (n = 156) (n = 0)

Fig 1. Patient disposition diagram. GC, gemcitabine and carboplatin.

Lost to follow-up Discontinued intervention Reasons: Disease progression Adverse event Patient decision Death Investigator decision Other Analyzed Excluded from analysis

(n = 0) (n = 79) (n = 38) (n = 15) (n = 13) (n = 3) (n = 2) (n = 8) (n = 153) (n = 0)

was lost to follow-up, and the reason was unknown for the third patient. Table 1 lists patient characteristics for all treatment groups. A total of 398 patients received four cycles of GC therapy, and 309 patients with no progression were randomly assigned to the two docetaxel treatment groups. For the 254 patients who were discontinued from the study during the GC phase, the most common reason for discontinuation was PD (n 147). Of the 153 patients starting docetaxel therapy immediately after random assignment, 145 patients (94.8%) received at least one treatment cycle. However, only 98 patients (62.8%) of the 156 patients randomly assigned to the delayed docetaxel arm received the rst cycle of docetaxel therapy. The most common reasons for discontinuation before delayed docetaxel treatment were PD (n 25), patient or investigator decision (n 16), and death (n 5). Patient characteristics were well balanced between the two docetaxel treatment arms (Table 1). Treatment Administration and Toxicity Table 2 lists treatment administration results. The mean number of cycles delivered in the immediate docetaxel arm (4.4 cycles) was
www.jco.org

slightly higher than in the delayed docetaxel arm (3.8 cycles) or in the GC phase (3.4 cycles). There was a higher percentage of dose modications in the GC phase compared with either the immediate or delayed docetaxel arm. The median dose-intensities for all treatment drugs were similar in each group. Hematologic and nonhematologic toxicities are listed in Table 3. The most common grade 3 and 4 hematologic toxicities in the GC phase were neutropenia (29.0%), thrombocytopenia (25.0%), and anemia (13.1%). During the GC phase, grade 4 thrombocytopenia was observed in 47 patients (8.3%), and grade 3 febrile neutropenia was observed in 10 patients (1.8%). Rates of grade 3 and 4 neutropenia observed in each docetaxel arm (immediate, 27.6%; delayed, 28.6%) were similar to the rate in the GC phase. No grade 3 or 4 thrombocytopenia was observed in either docetaxel arm, and only one patient in each docetaxel arm experienced grade 4 febrile neutropenia. The most common grade 3 and 4 nonhematologic toxicities in all treatment groups were fatigue (GC, 6.2%; immediate docetaxel, 9.7%; delayed docetaxel, 4.1%) and dyspnea (GC, 6.9%; immediate docetaxel, 2.8%; delayed docetaxel, 4.1%), with the notable exception that the delayed
2008 by American Society of Clinical Oncology

593

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Fidias et al

docetaxel group had a higher rate of grade 3 and 4 diarrhea (5.1%). There was no observed grade 3 or 4 neuropathy in any treatment group. Tumor Response Of 563 patients treated with GC, seven patients (1.2%) experienced CR, 168 patients (29.8%) had PR, 244 patient (43.3%) had SD, 68 patients (12.1%) had PD, and 76 patients (13.5%) were unknown. The overall response rate (ORR) for the GC phase was 31.1%. For the immediate docetaxel arm (n 145), seven patients (4.8%) experienced CR, 45 (31.0%) had PR, 53 (36.6%) had SD, 19 (13.1%) had PD, and 21 (14.5%) were unknown. The ORR for the

immediate docetaxel arm was 35.9%. Because response assessments in the immediate docetaxel arm represent the best cumulative response observed over both GC and docetaxel therapy periods, 35 patients in the immediate docetaxel arm had either a CR or PR carryover from the GC phase of treatment. Thus, the actual ORR for single-agent docetaxel on the immediate docetaxel arm was 11.7% (17 of 145 patients). For the delayed docetaxel arm (n 98), response assessments were based on single-agent docetaxel therapy only. In the delayed arm, there were no CRs, 11 patients (11.2%) had PR, 44 patients (44.9%) had SD, 26 patients (26.5%) had PD, and 17 patients (17.3%) were unknown. The ORR for the delayed docetaxel arm was 11.2%.

Table 1. Patient Characteristics GC Phase (N 566) Characteristic Age, years Median Range Sex Male Female Race White African American East/Southeast Asian Hispanic ECOG performance status 0, 1 2 Clinical stage IIIB (with pleural effusion) IV Missing Histology Adenocarcinoma Squamous Bronchoalveolar Large-cell undifferentiated NSCLC, NOS Other Missing No. of radiotherapy treatments, this cancer 0 1 2 3 Missing No. of prior surgeries, this cancer 0 1 2 3 Missing Response at random assignment CR/PR Stable disease No. of Patients 65.2 35.8-87.1 352 214 490 47 14 15 505 61 80 484 2 273 98 7 22 143 22 1 62.2 37.8 86.6 8.3 2.5 2.7 89.2 10.8 14.1 85.5 0.4 48.2 17.3 1.2 3.9 25.3 3.9 0.2 95 58 128 18 6 1 144 9 27 126 0 84 25 0 8 33 3 0 % Immediate Docetaxel (n 153) No. of Patients 65.4 40.9-86.9 62.1 37.9 83.7 11.8 3.9 0.7 94.1 5.9 17.6 82.4 0.0 54.9 16.3 0.0 5.2 21.6 2.0 0.0 97 59 134 12 4 6 140 16 24 130 2 73 29 3 5 40 5 1 % Delayed Docetaxel (n 156) No. of Patients 65.5 36.0-84.7 62.2 37.8 85.9 7.7 2.6 3.8 89.7 10.3 15.4 83.3 1.3 46.8 18.6 1.9 3.2 25.6 3.2 0.6 %

464 89 10 2 1 92 377 72 23 2

82.0 15.7 1.8 0.4 0.2 16.3 66.6 12.7 4.1 0.4

126 23 2 1 1 22 108 16 6 1 62 87

82.4 15.0 1.3 0.7 0.7 14.4 70.6 10.5 3.9 0.7 40.5 56.9

134 20 2 0 0 24 106 21 5 0 77 75

85.9 12.8 1.3 0.0 0.0 15.4 67.9 13.5 3.2 0.0 49.4 48.1

Abbreviations: GC, gemcitabine and carboplatin; ECOG, Eastern Cooperative Oncology Group; NSCLC, nonsmall-cell lung cancer; NOS, not otherwise specied; CR, complete response; PR, partial response.

594

2008 by American Society of Clinical Oncology

JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Immediate v Delayed Docetaxel After Front-Line GC in NSCLC

Table 2. Dose Administration Parameter Number of cycles Mean Median Total cycles delivered Median dose-intensity, % Dose adjustments No. of patients % Dose reductions No. of patients % Dose delays No. of patients % GC Phase (N 563) 3.4 4 1,900 G: 97.8; C: 99.8 330 58.6 250 46.2 178 31.6 Immediate Docetaxel (n 145) 4.4 6 638 98.5 42 29.0 14 9.7 34 23.4 Delayed Docetaxel (n 98) 3.8 4 373 98.9 31 31.6 14 14.3 22 22.4

Abbreviation: GC, gemcitabine and carboplatin.

Survival At trial end, 256 death events (immediate docetaxel, n 124; delayed docetaxel, n 132) had been recorded. Figure 2 shows both PFS and OS results. Median PFS was signicantly greater (P .0001) in the immediate docetaxel arm (5.7 months) compared with the delayed docetaxel arm (2.7 months). Median OS for all registered patients receiving at least one dose of GC (n 563) was 10.1 months (95% CI, 9.4 to 10.8 months). OS for the 254 patients who discontinued the trial before random assignment was 4.7 months (95% CI, 4.2 to 5.4 months). OS for randomly assigned patients was greater by 2.6 months in the immediate docetaxel arm (12.3 months, n 153) compared with the delayed docetaxel arm (9.7 months, n 156). The difference in OS between docetaxel arms did not reach statistical signicance (P .0853). OS for patients in the delayed arm who actually received docetaxel therapy (n 98) was 12.5 months (95% CI, 9.6 to 14.6 months), which was identical to the OS observed for the

safety population (n 145) in the immediate docetaxel arm (12.5 months; 95% CI, 10.6 to 15.8). The 12-month survival rate was 51.1% (95% CI, 43.0% to 59.3%) for immediate docetaxel and 43.5% (95% CI, 35.5% to 51.4%) for delayed docetaxel. QOL The LCSS questionnaire was completed by 250 patients before random assignment. An equal number (n 109) of patients who completed the LCSS after random assignment in each docetaxel treatment arm were evaluated for QOL. The compliance rate for completion of the LCSS questionnaire was 85.8% for the immediate docetaxel arm and 71.9% for the delayed docetaxel arm. The same number of patients (n 17, 15.6%) showed improved ASBI in each docetaxel arm. The majority of patients in both docetaxel arms had stable ASBI; stable ASBI was slightly more prevalent in the immediate docetaxel arm (n 64, 58.7%) compared with the delayed docetaxel arm

Table 3. Grade 3 and 4 Toxicities GC Phase (N 563) Grade 3 Toxicity Hematologic Neutropenia Thrombocytopenia Anemia Febrile neutropenia Nonhematologic Fatigue Dyspnea Nausea Vomiting Diarrhea Asthenia Myalgia Neuropathy No. of Patients 106 94 68 10 % 18.8 16.7 12.1 1.8 Grade 4 No. of Patients 57 47 6 0 % 10.1 8.3 1.1 0.0 Immediate Docetaxel (n 145) Grade 3 No. of Patients 12 0 1 4 % 8.3 0.0 0.7 2.8 Grade 4 No. of Patients 28 0 0 1 % 19.3 0.0 0.0 0.7 Delayed Docetaxel (n 98) Grade 3 No. of Patients 8 0 0 1 % 8.2 0.0 0.0 1.0 Grade 4 No. of Patients 20 0 0 1 % 20.4 0.0 0.0 1.0

34 34 14 12 1 10 1 0

6.0 6.0 2.5 2.1 0.2 1.8 0.2 0.0

1 5 1 1 1 2 0 0

0.2 0.9 0.2 0.2 0.2 0.4 0.0 0.0

13 2 0 0 1 2 1 0

9.0 1.4 0.0 0.0 0.7 1.4 0.7 0.0

1 2 0 0 0 1 0 0

0.7 1.4 0.0 0.0 0.0 0.7 0.0 0.0

3 3 1 1 4 1 1 0

3.1 3.1 1.0 1.0 4.1 1.0 1.0 0.0

1 1 0 0 1 0 0 0

1.0 1.0 0.0 0.0 1.0 0.0 0.0 0.0

Abbreviation: GC, gemcitabine and carboplatin.

www.jco.org

2008 by American Society of Clinical Oncology

595

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Fidias et al

A
Progression-Free Survival (probability)

1.0 0.8 0.6 0.4 0.2

Delayed Immediate

9 12 15 18 21 24 27 30 33 36 39 42 45 48

Time (months)
No. of patients at risk Delayed 156 59 28 18 13 Immediate 153 106 72 42 26 6 5 1 2

B
Overall Survival (probability)

1.0 0.8 0.6 0.4 0.2

Delayed Immediate

12

18

24

30

36

42

48

54

60

Time (months)
No. of patients at risk Delayed 156 109 65 42 21 Immediate 153 119 73 49 28 6 13 2 2

Fig 2. (A) Progression-free survival (PFS). Median PFS time was 5.7 months (95% CI, 4.4 to 6.9 months) for the immediate docetaxel am and 2.7 months (95% CI, 2.6 to 2.9 months) for the delayed docetaxel group (log-rank test, P .0001); seven (4.6%) and nine patients (5.8%), respectively, were censored from each group. (B) Overall survival (OS). Median OS time was 12.3 months (95% CI, 10.4 to 15.2 months) for the immediate docetaxel group and 9.7 months (95% CI, 8.4 to 12.5 months) for the delayed group (log-rank test, P .0853). Median 12-month survival rates were 51.1% (95% CI, 43.0% to 59.3%) for the immediate docetaxel group and 43.5% (95% CI, 35.5% to 51.4%) for the delayed group; 29 (19.0%) and 24 patients (15.4%) were censored, respectively.

(n 58, 53.2%). Conversely, there were more patients in the delayed docetaxel arm (n 20, 18.4%) with worsened ASBI compared with the immediate docetaxel arm (n 12, 11.0%). However, overall ASBI results were not statistically different (P .76) between the two docetaxel arms.
DISCUSSION

Any attempt to evaluate the timing of second-line chemotherapy for patients with advanced NSCLC will be judged by balancing improvement in patient OS against increased toxicities or diminished QOL. The goal of the current randomized phase III trial was to assess this balance in NSCLC patients without progression after GC treated with docetaxel either immediately after front-line GC therapy or with the onset of PD.
596
2008 by American Society of Clinical Oncology

Our results show that front-line GC therapy was well tolerated, with myelosuppression being the major cause of toxicities and only 22 patients (3.9%) in the GC phase discontinuing treatment as a result of adverse events. These results are consistent with previous reports for GC therapy5-8 and support an earlier analysis that area under the curve of 5 may be the optimal dose of carboplatin delivered as part of front-line GC therapy for NSCLC patients in terms of both response and control of thrombocytopenia.23 Previous studies of NSCLC patients who had received prior platinum-based chemotherapy reported manageable toxicities for docetaxel administered as second-line therapy.11,13 Our results show that immediate docetaxel therapy had a manageable toxicity prole compared with delayed docetaxel therapy. These observations are also consistent with the toxicity results of an earlier study using a 28-day GC regimen with equivalent GC dosages followed by docetaxel at 75 mg/m2.24 Another important consideration for the timing of second-line therapy in advanced NSCLC is the possible impact on patient QOL. Previous studies have shown that providing second-line docetaxel therapy to patients with advanced NSCLC enhanced QOL compared with BSC.11,13 Our results showed no statistical difference in QOL between patients in the two docetaxel treatment arms; these results are consistent with the observed toxicity proles. Earlier studies have been unable to demonstrate a signicant improvement in OS for NSCLC patients treated with sequential or maintenance chemotherapy after front-line treatment.25-34 Similar to the current trial design, several reports included evaluation of patients randomly assigned after completion of front-line chemotherapy. Buccheri et al27 evaluated continued treatment with a regimen of methotrexate, doxorubicin, cyclophosphamide, and lomustine compared with BSC in NSCLC patients with stable disease after front-line treatment with methotrexate, doxorubicin, cyclophosphamide, and lomustine. There was no signicant improvement in OS, whereas toxicity increased and QOL diminished in the group continuing treatment. In another trial, Westeel et al28 compared maintenance vinorelbine with BSC in NSCLC patients who responded to a front-line regimen of mitomycin, ifosfamide, and cisplatin either alone or combined with radiation therapy. There was no signicant improvement in either OS or PFS with maintenance therapy compared with BSC. In another phase III trial, Brodowicz et al29 showed that gemcitabine administered to NSCLC patients without progression after front-line treatment with a gemcitabine/cisplatin combination provided significant improvement in time to progression compared with BSC. However, OS was not signicantly lengthened, except for a limited subset of patients with a Karnofsky performance score of more than 80. More recently, Park et al35 evaluated NSCLC patients without progression who were treated with two cycles of a platinum doublet therapy and randomly assigned to continue with either two or four additional cycles of treatment. Although there were no differences in toxicity or OS between the two groups, there was signicant improvement in PFS for the group receiving the greater amount of therapy. Our study also showed a statistically signicant improvement in PFS by 3 months (P .0001) for patients receiving immediate docetaxel therapy. However, our study is one of the rst to show a trend toward improvement in OS for maintenance therapy, with OS increased by 2.6 months for patients in the immediate docetaxel arm. The improvement in OS did not reach statistical signicance (P .0853), perhaps because our study was not powered to detect a
JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Immediate v Delayed Docetaxel After Front-Line GC in NSCLC

difference in survival of less than 4 months. It is also possible that the observed increase in OS might be related to the use of a non crossresistant agent as maintenance therapy. In most of the earlier trials, the maintenance regimen was an extension of the same chemotherapy administered during front-line treatment. One interesting observation of our trial was the notable difference between treatment arms in the number of patients receiving docetaxel therapy. In the immediate docetaxel arm, most patients (145 of 153 patients; 94.8%) received docetaxel, whereas in the delayed docetaxel arm, a large number of patients (58 of 156 patients; 37.2%) never received docetaxel treatment. This observation is further underscored by results showing that the median OS time (12.5 months) was identical for the safety populations in each docetaxel arm. Thus, it seems that patients beneted from docetaxel therapy and the randomly assigned patients in the immediate docetaxel arm trended toward improved OS because more patients were able to receive treatment. The major reasons for discontinuation before delayed docetaxel treatment were PD and patient decision. Additional patient review revealed that many of these patients experienced signicant symptomatic deterioration by the time they reached PD and were unable to receive docetaxel therapy. These results suggest that NSCLC patients may be healthier and more likely to proceed to additional therapy if it is offered immediately after front-line treatment. These observations could merit further consideration in the design of future clinical trials for the treatment of advanced NSCLC. On the basis of these encouraging results, evaluation of new regimens using non cross-resistant agents delivered as maintenance therapy to NSCLC patients in appropriately powered studies may be warranted. A large phase III trial comparing maintenance pemetrexed plus BSC versus placebo plus BSC after front-line therapy was recently completed.36 This trial was statistically powered to evaluate an OS benet for maintenance therapy in advanced NSCLC.
AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following author(s) indicated a nancial or other interest that is relevant to the subject

matter under consideration in this article. Certain relationships marked with a U are those for which no compensation was received; those relationships marked with a C were compensated. For a detailed description of the disclosure categories, or for more information about ASCOs conict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conicts of Interest section in Information for Contributors. Employment or Leadership Position: Jane L. Bromund, Eli Lilly & Co (C); Ruqin Chen, Eli Lilly & Co (C); Maria Hristova-Kazmierski, Eli Lilly & Co (C); Joseph Treat, Eli Lilly & Co (C); Coleman K. Obasaju, Eli Lilly & Co (C); Martin Marciniak, Eli Lilly & Co (C); John Gill, Eli Lilly & Co (C) Consultant or Advisory Role: David M. Loesch, AstraZeneca (U); Joan H. Schiller, Eli Lilly & Co (C), Sano-aventis (C) Stock Ownership: Jane L. Bromund, Eli Lilly & Co; Ruqin Chen, Eli Lilly & Co; Maria Hristova-Kazmierski, Eli Lilly & Co; Joseph Treat, Eli Lilly & Co; Coleman K. Obasaju, Eli Lilly & Co; Martin Marciniak, Eli Lilly & Co; John Gill, Eli Lilly & Co Honoraria: Panos M. Fidias, Eli Lilly & Co, Genentech; David M. Loesch, AstraZeneca Research Funding: Joan H. Schiller, Eli Lilly & Co Expert Testimony: None Other Remuneration: None

AUTHOR CONTRIBUTIONS
Conception and design: Jane L. Bromund, Maria Hristova-Kazmierski, Coleman K. Obasaju, Joan H. Schiller Administrative support: Jane L. Bromund, Maria Hristova-Kazmierski Provision of study materials or patients: Panos M. Fidias, Shaker R. Dakhil, Alan P. Lyss, David M. Loesch, David M. Waterhouse, Joan H. Schiller Collection and assembly of data: Panos M. Fidias, Shaker R. Dakhil, Alan P. Lyss, David M. Waterhouse, Jane L. Bromund, Maria Hristova-Kazmierski Data analysis and interpretation: Panos M. Fidias, Jane L. Bromund, Ruqin Chen, Joseph Treat, Coleman K. Obasaju, Martin Marciniak, John Gill, Joan H. Schiller Manuscript writing: Panos M. Fidias, David M. Waterhouse, Jane L. Bromund, Ruqin Chen, Joseph Treat, Martin Marciniak, John Gill, Joan H. Schiller Final approval of manuscript: Panos M. Fidias, Shaker R. Dakhil, Alan P. Lyss, David M. Loesch, David M. Waterhouse, Jane L. Bromund, Ruqin Chen, Maria Hristova-Kazmierski, Joseph Treat, Coleman K. Obasaju, Martin Marciniak, John Gill, Joan H. Schiller
11. Shepherd FA, Dancey J, Ramlau R, et al: Prospective randomized trial of docetaxel versus best supportive care in patients with nonsmall-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 18:2095-2103, 2000 12. Fossella FV, DeVore R, Kerr RN, et al: Randomized phase III trial of docetaxel versus vinorelbine or ifosfamide in patients with advanced nonsmallcell lung cancer previously treated with platinumcontaining chemotherapy regimens. J Clin Oncol 18:2354-2362, 2000 13. Hanna N, Shepherd FA, Fossella FV, et al: Randomized phase III trial of pemetrexed versus docetaxel in patients with nonsmall-cell lung cancer previously treated with chemotherapy. J Clin Oncol 22:1589-1597, 2004 14. Shepherd FA, Pereira JR, Ciuleanu T, et al: Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 353:123-132, 2005 15. Grossi F, Aita M, Follador A, et al: Sequential, alternating and maintenance/consolidation chemotherapy in advanced non-small cell lung cancer: A 597

REFERENCES
1. Jemal A, Siegel R, Ward E, et al: Cancer statistics, 2007. CA Cancer J Clin 57:43-66, 2007 2. National Comprehensive Cancer Network: Clinical Practice Guidelines in Oncology: Non-small cell lung cancer. http://www.nccn.org/professionals/ default.asp 3. Schiller JH, Harrington D, Belani CP, et al: Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 346:92-98, 2002 4. Scagliotti GV, De Marainis F, Rinaldi M, et al: Phase III randomized trial comparing three platinumbased doublets in advanced non-small-cell lung cancer. J Clin Oncol 20:4285-4291, 2002 5. Carrato A, Alberola V, Massuti B, et al: Combination of gemcitabine and carboplatin as rst line treatment in non-small cell lung cancer. 23rd Congress of the European Society for Medical Oncology, Athens, Greece, November 6-10, 1998 (abstr 431P) www.jco.org

6. Parente B, Barroso A, Conde S, et al: A randomized phase III study of gemcitabine and carboplatinum versus vinorelbine and carboplatinum in advanced non small-cells lung cancer (NSCLC). Lung Cancer 29:61, 2000 (suppl 1) 7. Masters G, Argiris A, Hahn E, et al: A randomized phase II trial using two different treatment schedules of gemcitabine and carboplatin in patients with advanced non-small-cell lung cancer. J Thorac Oncol 1:19-24, 2006 8. Obasaju C, Ye Z, Bloss L, et al: Gemcitabine/ carboplatin in patients with metastatic non-small cell lung cancer: Phase II study of 28-day and 21-day schedules. Clin Lung Cancer 7:202-207, 2005 9. Lustberg MB, Edelman MJ: Optimal duration of chemotherapy in advanced non-small cell lung cancer. Curr Treat Options Oncol 8:38-46, 2007 10. Socinski MA, Stinchcombe TE: Duration of rst-line chemotherapy in advanced non-small-cell lung cancer: Less is more in the era of effective subsequent therapies. J Clin Oncol 25:5155-5157, 2007

2008 by American Society of Clinical Oncology

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.

Fidias et al

review of the literature. Oncologist 12:451-464, 2007 16. Therasse P, Arbuck S, Eisenhauer E, et al: New guidelines to evaluate the response to treatment in solid tumors: European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205-216, 2000 17. Hollen PJ, Gralla RJ, Kris MG, et al: Measurement of quality of life in patients with lung cancer in multicenter trials of new therapies: Psychometric assessment of the Lung Cancer Symptom Scale. Cancer 73:2087-2098, 1994 18. Sankoh AJ: Interim analyses: An update of an FDA reviewers experience and perspective. Drug Inform J 33:165-176, 1999 19. Kaplan E, Meier P: Nonparametric estimation of incomplete observations. J Am Stat Assoc 53: 457-481, 1958 20. Bland JM, Altman DG: The logrank test. BMJ 328:1073, 2004 21. Agresti A: Categorical Data Analysis. New York, NY, John Wiley & Sons, Inc, 1990, pp 230-234 22. National Cancer Institute: National Cancer Institute Common Toxicity Criteria Version 2.0. http: ctep.cancer.gov/forms/CTCv20_4-30-992.pdf 23. Treat J, Belani CP, Schiller J, et al: Gemcitabine (G) plus carboplatin at AUC 5 demonstrates reduced grade 4 thrombocytopenia rate compared to AUC 5.5 in rst-line therapy of patients with advanced stage NSCLC. J Clin Oncol 24:396s, 2006 (suppl; abstr 7130)

24. Chiappori A, Simon G, Williams C, et al: Phase II study of rst-line sequential chemotherapy with gemcitabine-carboplatin followed by docetaxel in patients with advanced non-small cell lung cancer. Oncology 68:382-390, 2005 25. Smith IE, OBrien ME, Talbot DC, et al: Duration of chemotherapy in advanced non-small-cell lung cancer: A randomized trial of three versus six courses of mitomycin, vinblastine, and cisplatin. J Clin Oncol 19:1336-1343, 2001 26. Socinski MA, Schell MJ, Peterman A, et al: Phase III trial comparing a dened duration of therapy versus continuous therapy followed by secondline therapy in advanced-stage IIIB/IV non-small-cell lung cancer. J Clin Oncol 20:1335-1343, 2002 27. Buccheri GF, Ferrigno D, Curcio A, et al: Continuation of chemotherapy versus supportive care alone in patients with inoperable non-small cell lung cancer and stable disease after two or three cycles of MACC. Cancer 63:428-432, 1989 28. Westeel V, Quoix E, Moro-Sibilot D, et al: Randomized study of maintenance vinorelbine in responders with advanced non-small-cell lung cancer. J Natl Cancer Inst 97:499-506, 2005 29. Brodowicz T, Krzakowski M, Zwitter M, et al: Cisplatin and gemcitabine rst-line chemotherapy followed by maintenance gemcitabine or best supportive care in advanced non-small cell lung cancer: A phase III trial. Lung Cancer 52:155-163, 2006 30. Edelman MJ, Gandara DR, Lau DHM, et al: Sequential chemotherapy in patients with advanced nonsmall cell lung carcinoma: Carboplatin and gemcitabine followed by paclitaxel. Cancer 92:146-152, 2001

31. Edelman MJ, Clark JI, Chansky K, et al: Randomized phase II trial of sequential chemotherapy in advanced non-small cell lung cancer (SWOG 9806): Carboplatin/gemcitabine followed by paclitaxel or cisplatin/vinorelbine followed by docetaxel. Clin Cancer Res 10:5022-5026, 2004 32. Belani CP, Barstis J, Perry MC, et al: Multicenter, randomized trial for stage IIB or IV nonsmallcell lung cancer using weekly paclitaxel and carboplatin followed by maintenance weekly paclitaxel or observation. J Clin Oncol 21:2933-2939, 2003 33. Depierre A, Quoix E, Mercier M, et al: Maintenance chemotherapy in advanced non-small cell lung cancer (NSCLC): A randomized study of vinorelbine (V) versus observation (OB) in patients (pts) responding to induction therapy (French Cooperative Oncology Group). Proc Am Soc Clin Oncol 20:309a, 2001 (abstr 1231) 34. Sculier J, Latte J, Lecomte J, et al: A phase III randomized trial comparing sequential to standard chemotherapy in advanced non-small cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 24:367s, 2006 (suppl; abstr 7012) 35. Park JO, Kim SW, Ahn JS, et al: Phase III trial of two versus four additional cycles in patients who are nonprogressive after two cycles of platinumbased chemotherapy in nonsmall-cell lung cancer. J Clin Oncol 25:5233-5239, 2007 36. Ciuleanu T, Brodowicz T, Belani CP, et al: Maintenance pemetrexed plus best supportive care (BSC) versus placebo plus BSC: A phase III study. Proc Am Soc Clin Oncol 26:426s, 2008 (suppl; abstr 8011)

Acknowledgment We thank Elizabeth Zobre, Jian Yu, and Andrea Koehler for their support.

598

2008 by American Society of Clinical Oncology

JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on May 14, 2012. For personal use only. No other uses without permission. Copyright 2009 American Society of Clinical Oncology. All rights reserved.