Randomized Phase III Intergroup Trial of Etoposide and Cisplatin With or Without Paclitaxel and Granulocyte Colony-Stimulating Factor in Pat
http://www.100md.com
《临床肿瘤学》
the University of Tennessee-Memphis, Memphis
Vanderbilt University, Nashville, TN
Duke University, Durham
Wake Forest University School of Medicine, Winston-Salem, NC
University of Colorado Cancer Center, Denver, CO
Mayo Clinic, Rochester, MN
University of Missouri/Ellis Fischel Cancer Center, Columbia, MO
University of Chicago, Chicago, IL
The Ohio State University Medical Center, Columbus, OH
University of California at San Diego, San Diego, CA
Medical University of South Carolina, Charleston, SC
ABSTRACT
PURPOSE: To determine, in a randomized comparison, whether the addition of paclitaxel to etoposide and cisplatin improves the time to progression and overall survival in patients with extensive small-cell lung cancer (SCLC) compared with standard etoposide and cisplatin and to compare the regimens' toxicity.
PATIENTS AND METHODS: Eligible patients (N = 587) with untreated extensive SCLC were randomly assigned to receive either cisplatin 80 mg/m2 on day 1 and etoposide 80 mg/m2 on days 1 through 3 administered every 3 weeks for six cycles (EP) or cisplatin 80 mg/m2 on day 1, paclitaxel 175 mg/m2 over 4 hours on day 1, and etoposide 80 mg/m2 on days 1 to 3 followed by recombinant human granulocyte colony-stimulating factor on days 4 to 18 administered every 3 weeks for six cycles (PET).
RESULTS: Reporting of demographics, response, and survival included 565 patients, of whom 282 were randomly assigned to receive EP and 283 were assigned to receive PET. Overall response rates were 68% for the EP arm and 75% for the PET arm. Median failure-free survival time was 5.9 months for the EP arm and 6 months for the PET arm (P = .179). Median overall survival time was 9.9 months for patients on EP and 10.6 months for patients on PET (P = .169). Toxic deaths occurred in 2.4% of the patients on EP and 6.5% of patients on PET.
CONCLUSION: PET did not improve the time to progression or survival in patients with extensive SCLC compared with EP alone and was associated with unacceptable toxicity.
INTRODUCTION
Small-cell lung cancer (SCLC) makes up approximately 20% of new lung cancer cases and is estimated to approach 34,380 cases in 2004 in the United States.1 SCLC presents with an abrupt clinical presentation, rapid tumor proliferation, and a median survival of less than 3 months in untreated patients.2 Combination chemotherapy has become the mainstay of therapy for SCLC and in the United States; the most commonly used regimen has been cisplatin and etoposide (EP).3 In patients with extensive SCLC, chemotherapy produces response rates of 50% to 60% and median survival times of 7 to 11 months.4,5 Unfortunately, cures are rare, with 3-year survival rates of less than 3%,6 and despite the use of a variety of strategies, there has been only a modest improvement in survival in these patients over the last three decades.7
In the mid-1990s, data were generated that identified paclitaxel as an active agent in previously untreated patients with extensive SCLC, with response rates of 34% to 41%.8,9 Some evidence of non–cross resistance was also suggested with the report of a 29% response rate to paclitaxel in chemotherapy-refractory extensive SCLC.10 Subsequently, several groups reported the results of phase I and II studies with somewhat similar strategies adding paclitaxel to a platinum and etoposide combination.11-15 These studies resulted in 12% to 25% complete response rates, overall response rates of 65% to 100%, and median survival times of up to 11 months. Most of these trials used etoposide 80 mg/m2 on days 1 to 3, paclitaxel 170 to 175 mg/m2 over 3 hours, and granulocyte colony-stimulating factor (G-CSF) either prophylactically or as needed to decrease the incidence of grade 4 neutropenia. In 1997, we designed a phase III randomized intergroup trial to determine whether the addition of paclitaxel to EP (PET) would improve the outcome of patients with extensive SCLC.
PATIENTS AND METHODS
Cancer and Leukemia Group B (CALGB) 9732 was a prospective randomized phase III trial. A permuted block randomization scheme stratified by performance status and sex was used by the CALGB Statistical Center to assign patients to treatment arms. A sample size of 580 patients was planned to detect a 30% improvement in median survival using a one-sided test with a 0.025 level of significance and 80% power. This increase in the median survival was projected to represent an increase from 8.5 months in the EP treatment arm to 11 months in the PET arm. Patient registration and data collection was managed by the CALGB Statistical Center. Statistical analyses were performed by the CALGB statisticians. The study was activated in April 1998 and closed to patient accrual in July 2001.
All patients had histologically or cytologically documented extensive-stage SCLC. The distinction between limited and extensive disease was made as follows. Limited disease was defined as disease restricted to one hemithorax with regional lymph node metastasis, including hilar mediastinal and supraclavicular nodes, whether ipsilateral or contralateral. The extensive disease classification included all patients with disease sites that were not eligible for the limited-stage classification. All patients had measurable or assessable disease (pleural effusions, bone metastases on bone scan, and bone marrow involvement were not measurable or assessable, but in these patients, at least the primary tumor had to be measurable or assessable).
Patient eligibility criteria included age 18 years, Eastern Cooperative Oncology Group performance status of 0 to 2 initially (amended to a status of 0 to 1 after the first 60 patients were accrued because of the early reports of toxicity in a phase II Southwest Oncology Group study),16 life expectancy greater than 2 months, and nonpregnant status for women. No prior chemotherapy or pelvic or mediastinal radiotherapy was allowed. No previous or concurrent malignancy was allowed other than curatively treated carcinoma-in-situ of the cervix or basal cell carcinoma of the skin or other primary cancer completely resected or treated more than 5 years ago without recurrence. Brain metastases were allowed, but the patient had to have completed radiation therapy and could not be on corticosteroids or phenytoin. Required laboratory parameters included the following: granulocytes 1,500/μL, platelets 100,000/μL, serum creatinine 1.5 mg/dL, bilirubin less than 1.5 mg/dL, and AST less than 2 x normal.
Each patient had to be aware of the nature of his or her disease and had to willingly give written consent after being informed of the experimental nature of the therapy, alternatives, potential benefits, side effects, risks, and discomforts. Other serious medical or psychiatric illness precluded participation in the trial.
Treatment Schedules
Patients were randomly assigned to one of two treatment programs (Fig 1). Arm 1 (EP) consisted of cisplatin 80 mg/m2 by intravenous (IV) infusion over 1 hour in 250 mL of normal saline or 3% sodium chloride solution with aggressive prehydration at the discretion of the investigator on day 1 along with etoposide 80 mg/m2 on days 1, 2, and 3 by IV infusion over 1 hour in 250 mL of dextrose 5% in water. This regimen was repeated every 21 days for a total of six cycles. Arm 2 (PET) consisted of cisplatin 80 mg/m2 as in arm 1 on day 1 along with etoposide 80 mg/m2 on days 1, 2, and 3; paclitaxel was administered on day 1 at a dose of 175 mg/m2 IV over 3 hours in 1,000 mL of normal saline. Premedication for paclitaxel consisted of dexamethasone 20 mg orally 12 hours and 6 hours before paclitaxel or 20 mg IV 30 minutes before paclitaxel, diphenhydramine 50 mg IV 30 minutes before paclitaxel, and cimetidine 300 mg, ranitidine 50 mg, or famotidine 20 mg IV 30 minutes before paclitaxel. Recombinant human G-CSF was administered starting on day 4 and continued daily until the granulocytes were 10,000/μL after day 10 or until day 18 (whichever came first). Treatment on arm 2 was repeated every 21 days for six cycles. The use of G-CSF in subsequent cycles (after the first cycle) in EP was at the discretion of the treating physician. Grade 4 neutropenia lasting longer than 5 days, neutropenic fever requiring hospitalization, or grade 4 thrombocytopenia required a 25% reduction in all drug doses with no re-escalation. Aggressive prehydration was recommended for all patients receiving cisplatin, but the method of prehydration was left to the discretion of the investigator.
Patients who initially presented with brain metastasis were treated with whole-brain radiation first and had to have been off anticonvulsants or corticosteroids before initiating chemotherapy. Patients who developed brain metastasis before completing protocol therapy were to receive whole-brain radiation therapy, were considered as having disease progression, and were removed from the protocol treatment. Patients who achieved a complete response from chemotherapy could be considered for prophylactic irradiation of the brain after six cycles of chemotherapy at the discretion of the investigator.
Toxicity and Response
Toxicity was graded according to the CALGB Expanded Common Toxicity Criteria. Complete response was defined as the disappearance of all measurable or assessable disease and signs, symptoms, and biochemical changes related to the tumor for at least 4 weeks, during which time no new lesions appeared. Partial response was defined as a reduction of greater than 50% in the sum of the products of the perpendicular diameters of all measurable lesions, during which time no new lesions appeared and no existing lesion enlarged. Regression (assessable disease only) was defined as a definite decrease in tumor size agreed on by two independent investigators and without the appearance of new lesions. Stable disease was defined as a less than 50% reduction or a less than 25% increase in the sum of the products of two perpendicular diameters of all measured lesions and the appearance of no new lesions. For assessable disease, stable was defined as no clear disease change in tumor size and no appearance of new lesions. Objective progression or relapse was defined as an increase in the product of two perpendicular diameters of any measured lesion by greater than 25% over the size present at entry onto study or, for patients who responded, the size at the time of maximum regression. For assessable disease, objective progression or relapse was defined as a definite increase in tumor size or the appearance of new lesions.
Monitoring and Statistics
The CALGB Data and Safety Monitoring Board reviewed the progress of CALGB 9732 on a semiannual basis. Group sequential methods (specifically, the Lan and DeMets analog to the O’Brien-Fleming boundaries) were used to maintain the overall level of significance while conducting multiple interim analysis of the data.17-19 Only summary results that combined the two treatment arms were presented in interim reports to CALGB members external to the Data and Safety Monitoring Board. CALGB personnel were responsible for quality assurance for all data submitted for this study to the CALGB Statistical Center. In addition, as part of the group data monitoring program, members of the CALGB Data Audit Committee made periodic site visits to all institutions to verify compliance with federal regulations and protocol requirements, including eligibility, treatment, tumor response, and follow-up evaluation.
Fisher’s exact test and other 2 tests were used to compare treatment groups with respect to toxicity, tumor response, and baseline demographic characteristics. Survival was calculated from the day of study entry until the date of death. Treatment was started within 7 days of randomization. Failure-free survival was defined as the time from study entry to date of progressive disease, relapse after response, or death from any cause. The Kaplan-Meier product-limit estimator was used to estimate survival and failure-free survival for each treatment group.20 The log-rank test was used to compare treatment groups with respect to survival and failure-free survival.21 Cox proportional hazards model was used to evaluate the joint effect of various baseline variables on survival.22 Logistic regression was used to evaluate the joint effect of various patient characteristics on the occurrence of grade 4 toxicity.23
RESULTS
Five hundred eighty-seven patients were registered onto CALGB 9732 between April 1998 and July 2001. Patients were accrued through the CALGB (207 patients), Eastern Cooperative Oncology Group (184 patients), Southwest Oncology Group (121 patients), North Central Cancer Treatment Group (20 patients), and the Expanded Participation project (five patients). The data reflect the CALGB database as of November 6, 2003. Of the 587 patients accrued onto the study, 16 patients were retrospectively determined to be ineligible because of no metastatic disease (n = 9), misdiagnosis (n = 2), prior radiation (n = 2), inadequate data (n = 2), and a performance status of 2 (n = 1). Ten of these ineligible patients were randomly assigned to EP, and six were randomly assigned to PET. Six additional patients were excluded (never started treatment); two patients were assigned to EP, and four patients were assigned to PET. Thus, a total of 22 patients were excluded from summaries of patient demographics, response, and survival. Reporting of demographics, response, and survival included the 565 patients who were eligible and started treatment, of whom 282 were randomly assigned to EP and 283 were randomly assigned to PET. Ineligible patients were included in the toxicity summary. The patient characteristics of the 565 eligible patients are listed in Table 1. The patient characteristics were well balanced for sex, age, race, performance status, weight loss, and symptom duration.
The overall response rates were 68% for EP (95% CI, 62% to 73%) and 75% for PET (95% CI, 69% to 80%; Table 2). The complete response rates were 10% for EP and 16% for PET. The median follow-up times in the EP and PET arms were 23.4 and 22.9 months, respectively. Median failure-free survival time was 5.9 months for the EP arm and 6.4 months for the PET arm, with 1-year failure-free survival rates of 9% and 8%, respectively (Table 3). The overall median survival time was 9.9 months for EP and 10.6 months for PET, with 1-year survival rates of 37% and 38%, respectively; 2-year survival rates of 8% and 11%, respectively; and 3-year survival rates of 4% and 4%, respectively (Table 4). Figures 1 and 2 show the failure-free and overall survival curves for the two treatment groups. Log-rank tests were performed to determine whether treatment had an effect on survival or failure-free survival. The 2 statistics were 0.9 and 0.8, with one-sided P values of .169 and .179, respectively. Patients less than 70 years of age survived longer than patients 70 years old (Table 4), with median survival times of 10.5 v 8.6 months, respectively (one-sided P = .0008; Fig 3). Cox proportional hazards model was used to assess whether treatment differences were consistent across race and sex. These analyses included a main effect for treatment of race and sex and their interaction and showed that the lack of treatment differences were consistent across race (P = .57) and sex (P = .34).
Toxicity
The number of cycles of treatment administered was assessed for 551 patients. Three hundred five patients received all six cycles (Table 5), including 156 (56%) of 275 patients on EP and 149 (54%) of 277 patients on PET. Dose modifications after the second cycle could be assessed in 434 patients. The EP arm required dose modifications after the second cycle in 23 (10.5%) of 219 patients, whereas the PET arm required dose adjustments in 33 (15%) of 215 patients.
Hematologic toxicity (Table 6) was significant, with grade 4 neutropenia and thrombocytopenia occurring equally in both arms. Grade 4 lymphocytopenia occurred more commonly in the PET arm. Grade 3 hearing loss, renal toxicity, and motor sensory neuropathy occurred more frequently in the PET arm (Table 7). Grade 4 nonhematologic toxicity was rare in both arms.
Grade 5 or lethal toxicity was experienced in both arms. The distribution and timing of the events are listed in Table 8. Most of the grade 5 events occurred after the first course of therapy. Seven (2.4%) of 292 assessable patients on EP experienced grade 5 toxicity, with most of these patients dying of neutropenic sepsis. One patient died of a progressive motor paralysis that slowly increased over five courses of chemotherapy, resulting in severe debilitation complicated by pneumonia and eventually resulting in death. Nineteen (6.5%) of 289 patients in the PET arm experienced grade 5 toxicity. The majority of these patients died as a result of neutropenic sepsis, whereas two patients died secondary to tumor lysis syndrome. One patient developed an acute motor neuropathy beginning with the sudden onset of memory loss and progressing to motor paralysis, debilitation, and apnea within 2 weeks of the first cycle of chemotherapy. Tumor lysis syndrome as the cause of death was reported in one patient on the PET arm. This patient presented with massive liver metastasis and, on day 3 of chemotherapy, became very nauseated and was unable to eat or drink for a week. The patient’s urinary output dropped, and his creatinine increased to 6.9 mg/dL. At the same time, the lactate dehydrogenase increased from 1,035 to 2,107 U/L, serum potassium increased to 5.6 mmol/L, and the uric acid increased from normal to 24.6 mg/dL. The patient died several days later.
There was no correlation between grade 5 toxicity and sex, performance status, sites of metastasis, weight loss, tumor burden, serum bilirubin, alkaline phosphatase, or AST and ALT. Age may have played a role in the risk of developing grade 5 toxicity because three (43%) of seven patients in the EP arm and eight (42%) of 19 patients in the PET arm who developed grade 5 toxicity were 70 years of age.
DISCUSSION
This phase III intergroup trial was designed to determine whether the addition of paclitaxel with G-CSF to standard therapy with EP would produce a 30% improvement in survival in patients with extensive SCLC. We found that, despite a higher complete and overall response rate in the PET arm, there was no improvement in failure-free survival, overall survival, or the 1-, 2-, or 3-year survivals for patients receiving the three-drug regimen. One other randomized trial that included patients with both limited and extensive disease has been reported using standard therapy with or without paclitaxel, and the results were similar to the present trial.24 In that trial, patients with extensive disease received either carboplatin (area under the curve of 5) on day 1, etoposide phosphate 125 mg/m2 on days 1 through 3, and vincristine 2 mg on days 1 and 8 or etoposide phosphate 102 mg/m2 on days 1 to 3, carboplatin (area under the curve of 5) on day 4, and paclitaxel 175 mg/m2 on day 4. One hundred fifty-three extensive SCLC patients were randomly assigned to each arm. The complete response rate was better in the paclitaxel arm than the vincristine arm (13.7% v 7.8%, respectively), but the overall response, time to progression, and overall survival were not improved with the addition of paclitaxel.
However, the addition of paclitaxel to EP did result in added toxicity. Hematologic toxicity was similar in both arms, with the exception of more lymphocytopenia in the PET arm. Nonhematologic toxicity, including ototoxicity, renal toxicity, and motor sensory neuropathy, was more common in the PET arm. Grade 5 lethal toxicity was 3 times as common in patients receiving PET. Most of the toxic deaths were caused by neutropenic sepsis occurring after the first course of chemotherapy. No single factor was found that predisposed patients to grade 5 toxicity; however, 42% of the patients experiencing grade 5 toxicity were 70 years of age.
Excess toxicity has been previously observed in other studies using similar paclitaxel combinations. Bunn et al16 reported the preliminary results of the administration of paclitaxel 175 mg/m2 on day 1 with cisplatin 80 mg/m2 on day 1 and etoposide 80 mg/m2 on day 1 followed by 160 mg/m2 orally on days 2 and 3 every 21 days for six cycles in patients with extensive SCLC. G-CSF (5 μg/kg) was administered subcutaneously on days 4 through 14. Sixty patients were assessable for toxicity; five patients died of neutropenic sepsis, whereas one patient died of renal failure, for a toxic death rate of 10%. These investigators were unable to identify any predisposing factors that might predict for a fatal outcome. Mavroudis et al25 attempted a phase III randomized trial in patients with limited and extensive SCLC comparing paclitaxel 175 mg/m2 over 3 hours with cisplatin 80 mg/m2 and etoposide 80 mg/m2 on days 1 to 3 with G-CSF support (5 μg/kg subcutaneously on days 5 through 15) versus cisplatin 80 mg/m2 and etoposide 120 mg/m2 on days 1 to 3. After 133 patients entered the study, a toxic death rate of 13% (eight of 62 patients) was encountered in the three-drug regimen, and the study was prematurely closed.
Overall, patients 70 years of age had significantly poorer survival than younger patients. It has been previously documented that elderly patients with extensive SCLC generally have poorer performance status, have more comorbidity, and are at higher risk for complications from intensive treatment.26 A variety of strategies have been used in an attempt to develop an optimal treatment program for the elderly SCLC patient, including the use of monotherapy,27 chemotherapy as needed to palliate symptoms,28 and less intensive chemotherapy regimens.29 To date, none of these approaches has been found to be optimal in elderly SCLC patients. Newer approaches in the management of elderly SCLC patients are needed, and caution should be exercised when exposing this patient population to combination chemotherapy regimens with the potential of producing substantial levels of neutropenia.
When this study was designed, phase I and II data suggested that the etoposide dose of 80 mg/m2 daily for 3 days was the upper limit that could be used with cisplatin and paclitaxel. Therefore, for the control arm, a regimen was used designed by Ihde et al30 with cisplatin 80 mg/m2 on day one and etoposide 80 mg/m3 day 1 through 3 of a 21 day cycle. Idhe et al30 found that, in 46 patients with extensive SCLC, the median survival time was 10.7 months with acceptable toxicity. The results of the present trial confirm and extend those results, with a median survival time of 9.9 months, acceptable toxicity, and a low toxic death rate of 2.4%. We conclude that the regimen of cisplatin 80 mg/m2 and etoposide 80 mg/m2 on days 1 to 3 is acceptable for use in patients with SCLC.
In conclusion, the addition of paclitaxel to standard doses of EP did not improve the overall survival of patients with extensive SCLC and was associated with an unacceptable toxic death rate. The addition of paclitaxel to etoposide and a platinum cannot be recommended for the routine treatment of patients with extensive SCLC.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Consultant/Advisory Role: Alan B. Sandler, Bristol-Myers Squibb; Karen Kelly, Bristol-Myers Squibb; Everett E. Vokes, Bristol-Myers Squibb. Stock Ownership: Mark R. Green, Amgen. Honoraria: Alan B. Sandler, Bristol-Myers Squibb; Everett E. Vokes, Bristol-Myers Squibb; Mark R. Green, Bristol-Myers Squibb. Research Funding: Everett E. Vokes, Bristol-Myers Squibb. For a detailed description of these categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We thank the Protocol Design Team (David S. Ettinger, MD; Jin S. Lee, MD; Alan B. Sandler, MD; Randolph S. Marks, MD; and Ellen Feigal, MD) and members of the Cancer and Leukemia Group B Respiratory Committee.
NOTES
Supported by Bristol-Myers Pharmaceutical Corp through a CRADA agreement with the National Cancer Institute (Bethesda, MD). Also supported by grant Nos. CA47555, CA47577, CA03927, CA49957, CA21115, CA SWOG, CA25224, CA12046, CA41287, CA77658, CA11789, and CA03927.
Presented at the 38th Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida, May 18-21, 2002 and at the 10th World Congress on Lung Cancer, Vancouver, British Columbia, Canada, August 10-14, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
American Cancer Society: Cancer facts and figures, 2004. http://www.cancer.org/docroot/STT/stt_0_2004.aspsitearea=STT&level=1
Seifter EJ, Ihde D: Therapy of small cell lung cancer: A perspective on two decades of clinical research. Semin Oncol 15:278-299, 1988
Loehrer P, Einhorn L, Greco FA: Cisplatin plus etoposide in small cell lung cancer. Semin Oncol 15:2-8, 1988
Spiegelman D, Mauer LH, Ware JH, et al: Prognostic factors in small-cell carcinoma of the lung: An analysis of 1,521 patients. J Clin Oncol 7:344-354, 1989
Albain KS, Crowley JJ, LeBlanc M, et al: Determinants of improved outcome in small cell lung cancer: An analysis of 2,580 patients in the Southwest Oncology Group data base. J Clin Oncol 8:1563-1574, 1990
Ihde DC, Glatstein EJ, Pass HI : Small-cell lung cancer, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology. Philadelphia, PA, Lippincott-Raven, 1997, pp 911-950
Chute JP, Chen T, Feigal E, et al: Twenty years of phase III trials for patients with extensive-stage small-cell lung cancer: Perceptible progress. J Clin Oncol 17:1794-1801, 1999
Kirshling RJ, Jung SH, Jett JR: A phase II trial of Taxol and G-CSF in previously untreated patients with extensive small cell lung cancer (SCLC). Proc Am Soc Clin Oncol 13:326, 1994 (abstr 1078)
Ettinger DS, Finkelstein DM, Sarma R, et al: Phase II study of paclitaxel in patients with extensive-stage small cell lung cancer: An Eastern Cooperative Oncology Group study. J Clin Oncol 13:1430-1435, 1995
Smitt EF, Fokkema E, Biesma B, et al: A phase II study of paclitaxel in heavily pretreated patients with small cell lung cancer. Br J Cancer 77:347-351, 1998
Hainsworth JD, Gray JR, Stroup SL, et al: Paclitaxel, carboplatin and extended-schedule etoposide in the treatment of small cell lung cancer: Comparison of sequential trials using different dose intensities. J Clin Oncol 15:3464-3470, 1997
Kelly K, Pan Z, Wood ME, et al: A phase I study of paclitaxel, etoposide and cisplatin in extensive stage small cell lung cancer. Clin Cancer Res 5:3419-3424, 1999
Levitan N, McKenney J, Tahsildar H, et al: Results of a phase I dose escalation trial of paclitaxel, etoposide and cisplatin followed by filgrastim in the treatment of patients with extensive stage small cell lung cancer. Proc Am Soc Clin Oncol 14:379, 1995 (abstr 1177)
Glisson BS, Kurie JM, Perez-Solar R, et al: Cisplatin, etoposide and paclitaxel in the treatment of patients with extensive small-cell lung cancer. J Clin Oncol 17:2309-2315, 1999
Niell HB, Perry MC, Clamon J, et al: Carboplatin, etoposide and paclitaxel in the treatment of patients with extensive small cell lung cancer. Clin Lung Cancer 2:204-209, 2001
Bunn PA, Kelley K, Crowley J, et al: Preliminary toxicity results from Southwest Oncology Group Trial: A phase II trial of cisplatin, etoposide and paclitaxel with G-CSF in untreated patients with extensive small cell lung cancer. Proc Am Soc Clin Oncol 18:468a, 1999 (abstr 1807)
O'Fallon JR: Policies for interim analysis and interim reporting of results. Cancer Treat Rep 69:1101-1105, 1985
Lan GGK, DeMets DL: Discrete sequential boundaries for clinical trials. Biometrika 70:659-663, 1983
O'Brien PC, Fleming TR: A multiple testing procedure for clinical trials. Biometrics 35:549-556, 1979
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Peto R, Peto J: Asymptotically efficient rank invariant test procedures. J R Stat Soc A 135:185-198, 1972
Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972
Hosmer DW, Lemeshow S: Applied Logistic Regression. New York, NY, Wiley, 1989
Reck M, von Pawel J, Macka H-N, et al: Randomized phase III trial of paclitaxel, etoposide and carboplatin versus carboplatin, etoposide and vincristine in patients with small cell lung cancer. J Natl Cancer Inst 95:1118-1127, 2003
Mavroudis D, Papadakis E, Veslemes M, et al: The Greek Lung Group Cooperative Group: A multicenter randomized phase III study comparing paclitaxel, cisplatin, etoposide versus cisplatin, etoposide as front line treatment in patients with small cell lung cancer. Proc Am Soc Clin Oncol 19:484a, 2000 (abstr 1894)
Findlay MP, Griffin AM, Raghavan D, et al: Retrospective review of chemotherapy for small cell lung cancer in the elderly: Does the end justify the means Eur J Cancer 27:1597-1601, 1991
Smit EF, Carney DN, Harford P, et al: A phase II study of oral etoposide in elderly patients with small cell lung cancer. Thorax 44:631-633, 1989
Earl HM, Rudd RM, Spiro SG, et al: A randomized trial of planned versus as required chemotherapy in small cell lung cancer: A Cancer Research Campaign trial. Br J Cancer 64:566-572, 1991
Medical Research Council Lung Cancer Working Party: Randomized trial of four-drug versus less intensive two-drug chemotherapy in the palliative treatment of patients with small cell lung cancer and poor prognosis. Br J Cancer 73:406-413, 1996
Ihde DC, Mulshine JL, Kramer BS, et al: Prospective randomized comparison of high-dose and standard dose etoposide and cisplatin chemotherapy in patients with extensive stage small cell lung cancer. J Clin Oncol 12:2022-2034, 1994(Harvey B. Niell, James E.)
Vanderbilt University, Nashville, TN
Duke University, Durham
Wake Forest University School of Medicine, Winston-Salem, NC
University of Colorado Cancer Center, Denver, CO
Mayo Clinic, Rochester, MN
University of Missouri/Ellis Fischel Cancer Center, Columbia, MO
University of Chicago, Chicago, IL
The Ohio State University Medical Center, Columbus, OH
University of California at San Diego, San Diego, CA
Medical University of South Carolina, Charleston, SC
ABSTRACT
PURPOSE: To determine, in a randomized comparison, whether the addition of paclitaxel to etoposide and cisplatin improves the time to progression and overall survival in patients with extensive small-cell lung cancer (SCLC) compared with standard etoposide and cisplatin and to compare the regimens' toxicity.
PATIENTS AND METHODS: Eligible patients (N = 587) with untreated extensive SCLC were randomly assigned to receive either cisplatin 80 mg/m2 on day 1 and etoposide 80 mg/m2 on days 1 through 3 administered every 3 weeks for six cycles (EP) or cisplatin 80 mg/m2 on day 1, paclitaxel 175 mg/m2 over 4 hours on day 1, and etoposide 80 mg/m2 on days 1 to 3 followed by recombinant human granulocyte colony-stimulating factor on days 4 to 18 administered every 3 weeks for six cycles (PET).
RESULTS: Reporting of demographics, response, and survival included 565 patients, of whom 282 were randomly assigned to receive EP and 283 were assigned to receive PET. Overall response rates were 68% for the EP arm and 75% for the PET arm. Median failure-free survival time was 5.9 months for the EP arm and 6 months for the PET arm (P = .179). Median overall survival time was 9.9 months for patients on EP and 10.6 months for patients on PET (P = .169). Toxic deaths occurred in 2.4% of the patients on EP and 6.5% of patients on PET.
CONCLUSION: PET did not improve the time to progression or survival in patients with extensive SCLC compared with EP alone and was associated with unacceptable toxicity.
INTRODUCTION
Small-cell lung cancer (SCLC) makes up approximately 20% of new lung cancer cases and is estimated to approach 34,380 cases in 2004 in the United States.1 SCLC presents with an abrupt clinical presentation, rapid tumor proliferation, and a median survival of less than 3 months in untreated patients.2 Combination chemotherapy has become the mainstay of therapy for SCLC and in the United States; the most commonly used regimen has been cisplatin and etoposide (EP).3 In patients with extensive SCLC, chemotherapy produces response rates of 50% to 60% and median survival times of 7 to 11 months.4,5 Unfortunately, cures are rare, with 3-year survival rates of less than 3%,6 and despite the use of a variety of strategies, there has been only a modest improvement in survival in these patients over the last three decades.7
In the mid-1990s, data were generated that identified paclitaxel as an active agent in previously untreated patients with extensive SCLC, with response rates of 34% to 41%.8,9 Some evidence of non–cross resistance was also suggested with the report of a 29% response rate to paclitaxel in chemotherapy-refractory extensive SCLC.10 Subsequently, several groups reported the results of phase I and II studies with somewhat similar strategies adding paclitaxel to a platinum and etoposide combination.11-15 These studies resulted in 12% to 25% complete response rates, overall response rates of 65% to 100%, and median survival times of up to 11 months. Most of these trials used etoposide 80 mg/m2 on days 1 to 3, paclitaxel 170 to 175 mg/m2 over 3 hours, and granulocyte colony-stimulating factor (G-CSF) either prophylactically or as needed to decrease the incidence of grade 4 neutropenia. In 1997, we designed a phase III randomized intergroup trial to determine whether the addition of paclitaxel to EP (PET) would improve the outcome of patients with extensive SCLC.
PATIENTS AND METHODS
Cancer and Leukemia Group B (CALGB) 9732 was a prospective randomized phase III trial. A permuted block randomization scheme stratified by performance status and sex was used by the CALGB Statistical Center to assign patients to treatment arms. A sample size of 580 patients was planned to detect a 30% improvement in median survival using a one-sided test with a 0.025 level of significance and 80% power. This increase in the median survival was projected to represent an increase from 8.5 months in the EP treatment arm to 11 months in the PET arm. Patient registration and data collection was managed by the CALGB Statistical Center. Statistical analyses were performed by the CALGB statisticians. The study was activated in April 1998 and closed to patient accrual in July 2001.
All patients had histologically or cytologically documented extensive-stage SCLC. The distinction between limited and extensive disease was made as follows. Limited disease was defined as disease restricted to one hemithorax with regional lymph node metastasis, including hilar mediastinal and supraclavicular nodes, whether ipsilateral or contralateral. The extensive disease classification included all patients with disease sites that were not eligible for the limited-stage classification. All patients had measurable or assessable disease (pleural effusions, bone metastases on bone scan, and bone marrow involvement were not measurable or assessable, but in these patients, at least the primary tumor had to be measurable or assessable).
Patient eligibility criteria included age 18 years, Eastern Cooperative Oncology Group performance status of 0 to 2 initially (amended to a status of 0 to 1 after the first 60 patients were accrued because of the early reports of toxicity in a phase II Southwest Oncology Group study),16 life expectancy greater than 2 months, and nonpregnant status for women. No prior chemotherapy or pelvic or mediastinal radiotherapy was allowed. No previous or concurrent malignancy was allowed other than curatively treated carcinoma-in-situ of the cervix or basal cell carcinoma of the skin or other primary cancer completely resected or treated more than 5 years ago without recurrence. Brain metastases were allowed, but the patient had to have completed radiation therapy and could not be on corticosteroids or phenytoin. Required laboratory parameters included the following: granulocytes 1,500/μL, platelets 100,000/μL, serum creatinine 1.5 mg/dL, bilirubin less than 1.5 mg/dL, and AST less than 2 x normal.
Each patient had to be aware of the nature of his or her disease and had to willingly give written consent after being informed of the experimental nature of the therapy, alternatives, potential benefits, side effects, risks, and discomforts. Other serious medical or psychiatric illness precluded participation in the trial.
Treatment Schedules
Patients were randomly assigned to one of two treatment programs (Fig 1). Arm 1 (EP) consisted of cisplatin 80 mg/m2 by intravenous (IV) infusion over 1 hour in 250 mL of normal saline or 3% sodium chloride solution with aggressive prehydration at the discretion of the investigator on day 1 along with etoposide 80 mg/m2 on days 1, 2, and 3 by IV infusion over 1 hour in 250 mL of dextrose 5% in water. This regimen was repeated every 21 days for a total of six cycles. Arm 2 (PET) consisted of cisplatin 80 mg/m2 as in arm 1 on day 1 along with etoposide 80 mg/m2 on days 1, 2, and 3; paclitaxel was administered on day 1 at a dose of 175 mg/m2 IV over 3 hours in 1,000 mL of normal saline. Premedication for paclitaxel consisted of dexamethasone 20 mg orally 12 hours and 6 hours before paclitaxel or 20 mg IV 30 minutes before paclitaxel, diphenhydramine 50 mg IV 30 minutes before paclitaxel, and cimetidine 300 mg, ranitidine 50 mg, or famotidine 20 mg IV 30 minutes before paclitaxel. Recombinant human G-CSF was administered starting on day 4 and continued daily until the granulocytes were 10,000/μL after day 10 or until day 18 (whichever came first). Treatment on arm 2 was repeated every 21 days for six cycles. The use of G-CSF in subsequent cycles (after the first cycle) in EP was at the discretion of the treating physician. Grade 4 neutropenia lasting longer than 5 days, neutropenic fever requiring hospitalization, or grade 4 thrombocytopenia required a 25% reduction in all drug doses with no re-escalation. Aggressive prehydration was recommended for all patients receiving cisplatin, but the method of prehydration was left to the discretion of the investigator.
Patients who initially presented with brain metastasis were treated with whole-brain radiation first and had to have been off anticonvulsants or corticosteroids before initiating chemotherapy. Patients who developed brain metastasis before completing protocol therapy were to receive whole-brain radiation therapy, were considered as having disease progression, and were removed from the protocol treatment. Patients who achieved a complete response from chemotherapy could be considered for prophylactic irradiation of the brain after six cycles of chemotherapy at the discretion of the investigator.
Toxicity and Response
Toxicity was graded according to the CALGB Expanded Common Toxicity Criteria. Complete response was defined as the disappearance of all measurable or assessable disease and signs, symptoms, and biochemical changes related to the tumor for at least 4 weeks, during which time no new lesions appeared. Partial response was defined as a reduction of greater than 50% in the sum of the products of the perpendicular diameters of all measurable lesions, during which time no new lesions appeared and no existing lesion enlarged. Regression (assessable disease only) was defined as a definite decrease in tumor size agreed on by two independent investigators and without the appearance of new lesions. Stable disease was defined as a less than 50% reduction or a less than 25% increase in the sum of the products of two perpendicular diameters of all measured lesions and the appearance of no new lesions. For assessable disease, stable was defined as no clear disease change in tumor size and no appearance of new lesions. Objective progression or relapse was defined as an increase in the product of two perpendicular diameters of any measured lesion by greater than 25% over the size present at entry onto study or, for patients who responded, the size at the time of maximum regression. For assessable disease, objective progression or relapse was defined as a definite increase in tumor size or the appearance of new lesions.
Monitoring and Statistics
The CALGB Data and Safety Monitoring Board reviewed the progress of CALGB 9732 on a semiannual basis. Group sequential methods (specifically, the Lan and DeMets analog to the O’Brien-Fleming boundaries) were used to maintain the overall level of significance while conducting multiple interim analysis of the data.17-19 Only summary results that combined the two treatment arms were presented in interim reports to CALGB members external to the Data and Safety Monitoring Board. CALGB personnel were responsible for quality assurance for all data submitted for this study to the CALGB Statistical Center. In addition, as part of the group data monitoring program, members of the CALGB Data Audit Committee made periodic site visits to all institutions to verify compliance with federal regulations and protocol requirements, including eligibility, treatment, tumor response, and follow-up evaluation.
Fisher’s exact test and other 2 tests were used to compare treatment groups with respect to toxicity, tumor response, and baseline demographic characteristics. Survival was calculated from the day of study entry until the date of death. Treatment was started within 7 days of randomization. Failure-free survival was defined as the time from study entry to date of progressive disease, relapse after response, or death from any cause. The Kaplan-Meier product-limit estimator was used to estimate survival and failure-free survival for each treatment group.20 The log-rank test was used to compare treatment groups with respect to survival and failure-free survival.21 Cox proportional hazards model was used to evaluate the joint effect of various baseline variables on survival.22 Logistic regression was used to evaluate the joint effect of various patient characteristics on the occurrence of grade 4 toxicity.23
RESULTS
Five hundred eighty-seven patients were registered onto CALGB 9732 between April 1998 and July 2001. Patients were accrued through the CALGB (207 patients), Eastern Cooperative Oncology Group (184 patients), Southwest Oncology Group (121 patients), North Central Cancer Treatment Group (20 patients), and the Expanded Participation project (five patients). The data reflect the CALGB database as of November 6, 2003. Of the 587 patients accrued onto the study, 16 patients were retrospectively determined to be ineligible because of no metastatic disease (n = 9), misdiagnosis (n = 2), prior radiation (n = 2), inadequate data (n = 2), and a performance status of 2 (n = 1). Ten of these ineligible patients were randomly assigned to EP, and six were randomly assigned to PET. Six additional patients were excluded (never started treatment); two patients were assigned to EP, and four patients were assigned to PET. Thus, a total of 22 patients were excluded from summaries of patient demographics, response, and survival. Reporting of demographics, response, and survival included the 565 patients who were eligible and started treatment, of whom 282 were randomly assigned to EP and 283 were randomly assigned to PET. Ineligible patients were included in the toxicity summary. The patient characteristics of the 565 eligible patients are listed in Table 1. The patient characteristics were well balanced for sex, age, race, performance status, weight loss, and symptom duration.
The overall response rates were 68% for EP (95% CI, 62% to 73%) and 75% for PET (95% CI, 69% to 80%; Table 2). The complete response rates were 10% for EP and 16% for PET. The median follow-up times in the EP and PET arms were 23.4 and 22.9 months, respectively. Median failure-free survival time was 5.9 months for the EP arm and 6.4 months for the PET arm, with 1-year failure-free survival rates of 9% and 8%, respectively (Table 3). The overall median survival time was 9.9 months for EP and 10.6 months for PET, with 1-year survival rates of 37% and 38%, respectively; 2-year survival rates of 8% and 11%, respectively; and 3-year survival rates of 4% and 4%, respectively (Table 4). Figures 1 and 2 show the failure-free and overall survival curves for the two treatment groups. Log-rank tests were performed to determine whether treatment had an effect on survival or failure-free survival. The 2 statistics were 0.9 and 0.8, with one-sided P values of .169 and .179, respectively. Patients less than 70 years of age survived longer than patients 70 years old (Table 4), with median survival times of 10.5 v 8.6 months, respectively (one-sided P = .0008; Fig 3). Cox proportional hazards model was used to assess whether treatment differences were consistent across race and sex. These analyses included a main effect for treatment of race and sex and their interaction and showed that the lack of treatment differences were consistent across race (P = .57) and sex (P = .34).
Toxicity
The number of cycles of treatment administered was assessed for 551 patients. Three hundred five patients received all six cycles (Table 5), including 156 (56%) of 275 patients on EP and 149 (54%) of 277 patients on PET. Dose modifications after the second cycle could be assessed in 434 patients. The EP arm required dose modifications after the second cycle in 23 (10.5%) of 219 patients, whereas the PET arm required dose adjustments in 33 (15%) of 215 patients.
Hematologic toxicity (Table 6) was significant, with grade 4 neutropenia and thrombocytopenia occurring equally in both arms. Grade 4 lymphocytopenia occurred more commonly in the PET arm. Grade 3 hearing loss, renal toxicity, and motor sensory neuropathy occurred more frequently in the PET arm (Table 7). Grade 4 nonhematologic toxicity was rare in both arms.
Grade 5 or lethal toxicity was experienced in both arms. The distribution and timing of the events are listed in Table 8. Most of the grade 5 events occurred after the first course of therapy. Seven (2.4%) of 292 assessable patients on EP experienced grade 5 toxicity, with most of these patients dying of neutropenic sepsis. One patient died of a progressive motor paralysis that slowly increased over five courses of chemotherapy, resulting in severe debilitation complicated by pneumonia and eventually resulting in death. Nineteen (6.5%) of 289 patients in the PET arm experienced grade 5 toxicity. The majority of these patients died as a result of neutropenic sepsis, whereas two patients died secondary to tumor lysis syndrome. One patient developed an acute motor neuropathy beginning with the sudden onset of memory loss and progressing to motor paralysis, debilitation, and apnea within 2 weeks of the first cycle of chemotherapy. Tumor lysis syndrome as the cause of death was reported in one patient on the PET arm. This patient presented with massive liver metastasis and, on day 3 of chemotherapy, became very nauseated and was unable to eat or drink for a week. The patient’s urinary output dropped, and his creatinine increased to 6.9 mg/dL. At the same time, the lactate dehydrogenase increased from 1,035 to 2,107 U/L, serum potassium increased to 5.6 mmol/L, and the uric acid increased from normal to 24.6 mg/dL. The patient died several days later.
There was no correlation between grade 5 toxicity and sex, performance status, sites of metastasis, weight loss, tumor burden, serum bilirubin, alkaline phosphatase, or AST and ALT. Age may have played a role in the risk of developing grade 5 toxicity because three (43%) of seven patients in the EP arm and eight (42%) of 19 patients in the PET arm who developed grade 5 toxicity were 70 years of age.
DISCUSSION
This phase III intergroup trial was designed to determine whether the addition of paclitaxel with G-CSF to standard therapy with EP would produce a 30% improvement in survival in patients with extensive SCLC. We found that, despite a higher complete and overall response rate in the PET arm, there was no improvement in failure-free survival, overall survival, or the 1-, 2-, or 3-year survivals for patients receiving the three-drug regimen. One other randomized trial that included patients with both limited and extensive disease has been reported using standard therapy with or without paclitaxel, and the results were similar to the present trial.24 In that trial, patients with extensive disease received either carboplatin (area under the curve of 5) on day 1, etoposide phosphate 125 mg/m2 on days 1 through 3, and vincristine 2 mg on days 1 and 8 or etoposide phosphate 102 mg/m2 on days 1 to 3, carboplatin (area under the curve of 5) on day 4, and paclitaxel 175 mg/m2 on day 4. One hundred fifty-three extensive SCLC patients were randomly assigned to each arm. The complete response rate was better in the paclitaxel arm than the vincristine arm (13.7% v 7.8%, respectively), but the overall response, time to progression, and overall survival were not improved with the addition of paclitaxel.
However, the addition of paclitaxel to EP did result in added toxicity. Hematologic toxicity was similar in both arms, with the exception of more lymphocytopenia in the PET arm. Nonhematologic toxicity, including ototoxicity, renal toxicity, and motor sensory neuropathy, was more common in the PET arm. Grade 5 lethal toxicity was 3 times as common in patients receiving PET. Most of the toxic deaths were caused by neutropenic sepsis occurring after the first course of chemotherapy. No single factor was found that predisposed patients to grade 5 toxicity; however, 42% of the patients experiencing grade 5 toxicity were 70 years of age.
Excess toxicity has been previously observed in other studies using similar paclitaxel combinations. Bunn et al16 reported the preliminary results of the administration of paclitaxel 175 mg/m2 on day 1 with cisplatin 80 mg/m2 on day 1 and etoposide 80 mg/m2 on day 1 followed by 160 mg/m2 orally on days 2 and 3 every 21 days for six cycles in patients with extensive SCLC. G-CSF (5 μg/kg) was administered subcutaneously on days 4 through 14. Sixty patients were assessable for toxicity; five patients died of neutropenic sepsis, whereas one patient died of renal failure, for a toxic death rate of 10%. These investigators were unable to identify any predisposing factors that might predict for a fatal outcome. Mavroudis et al25 attempted a phase III randomized trial in patients with limited and extensive SCLC comparing paclitaxel 175 mg/m2 over 3 hours with cisplatin 80 mg/m2 and etoposide 80 mg/m2 on days 1 to 3 with G-CSF support (5 μg/kg subcutaneously on days 5 through 15) versus cisplatin 80 mg/m2 and etoposide 120 mg/m2 on days 1 to 3. After 133 patients entered the study, a toxic death rate of 13% (eight of 62 patients) was encountered in the three-drug regimen, and the study was prematurely closed.
Overall, patients 70 years of age had significantly poorer survival than younger patients. It has been previously documented that elderly patients with extensive SCLC generally have poorer performance status, have more comorbidity, and are at higher risk for complications from intensive treatment.26 A variety of strategies have been used in an attempt to develop an optimal treatment program for the elderly SCLC patient, including the use of monotherapy,27 chemotherapy as needed to palliate symptoms,28 and less intensive chemotherapy regimens.29 To date, none of these approaches has been found to be optimal in elderly SCLC patients. Newer approaches in the management of elderly SCLC patients are needed, and caution should be exercised when exposing this patient population to combination chemotherapy regimens with the potential of producing substantial levels of neutropenia.
When this study was designed, phase I and II data suggested that the etoposide dose of 80 mg/m2 daily for 3 days was the upper limit that could be used with cisplatin and paclitaxel. Therefore, for the control arm, a regimen was used designed by Ihde et al30 with cisplatin 80 mg/m2 on day one and etoposide 80 mg/m3 day 1 through 3 of a 21 day cycle. Idhe et al30 found that, in 46 patients with extensive SCLC, the median survival time was 10.7 months with acceptable toxicity. The results of the present trial confirm and extend those results, with a median survival time of 9.9 months, acceptable toxicity, and a low toxic death rate of 2.4%. We conclude that the regimen of cisplatin 80 mg/m2 and etoposide 80 mg/m2 on days 1 to 3 is acceptable for use in patients with SCLC.
In conclusion, the addition of paclitaxel to standard doses of EP did not improve the overall survival of patients with extensive SCLC and was associated with an unacceptable toxic death rate. The addition of paclitaxel to etoposide and a platinum cannot be recommended for the routine treatment of patients with extensive SCLC.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Consultant/Advisory Role: Alan B. Sandler, Bristol-Myers Squibb; Karen Kelly, Bristol-Myers Squibb; Everett E. Vokes, Bristol-Myers Squibb. Stock Ownership: Mark R. Green, Amgen. Honoraria: Alan B. Sandler, Bristol-Myers Squibb; Everett E. Vokes, Bristol-Myers Squibb; Mark R. Green, Bristol-Myers Squibb. Research Funding: Everett E. Vokes, Bristol-Myers Squibb. For a detailed description of these categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We thank the Protocol Design Team (David S. Ettinger, MD; Jin S. Lee, MD; Alan B. Sandler, MD; Randolph S. Marks, MD; and Ellen Feigal, MD) and members of the Cancer and Leukemia Group B Respiratory Committee.
NOTES
Supported by Bristol-Myers Pharmaceutical Corp through a CRADA agreement with the National Cancer Institute (Bethesda, MD). Also supported by grant Nos. CA47555, CA47577, CA03927, CA49957, CA21115, CA SWOG, CA25224, CA12046, CA41287, CA77658, CA11789, and CA03927.
Presented at the 38th Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida, May 18-21, 2002 and at the 10th World Congress on Lung Cancer, Vancouver, British Columbia, Canada, August 10-14, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
American Cancer Society: Cancer facts and figures, 2004. http://www.cancer.org/docroot/STT/stt_0_2004.aspsitearea=STT&level=1
Seifter EJ, Ihde D: Therapy of small cell lung cancer: A perspective on two decades of clinical research. Semin Oncol 15:278-299, 1988
Loehrer P, Einhorn L, Greco FA: Cisplatin plus etoposide in small cell lung cancer. Semin Oncol 15:2-8, 1988
Spiegelman D, Mauer LH, Ware JH, et al: Prognostic factors in small-cell carcinoma of the lung: An analysis of 1,521 patients. J Clin Oncol 7:344-354, 1989
Albain KS, Crowley JJ, LeBlanc M, et al: Determinants of improved outcome in small cell lung cancer: An analysis of 2,580 patients in the Southwest Oncology Group data base. J Clin Oncol 8:1563-1574, 1990
Ihde DC, Glatstein EJ, Pass HI : Small-cell lung cancer, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology. Philadelphia, PA, Lippincott-Raven, 1997, pp 911-950
Chute JP, Chen T, Feigal E, et al: Twenty years of phase III trials for patients with extensive-stage small-cell lung cancer: Perceptible progress. J Clin Oncol 17:1794-1801, 1999
Kirshling RJ, Jung SH, Jett JR: A phase II trial of Taxol and G-CSF in previously untreated patients with extensive small cell lung cancer (SCLC). Proc Am Soc Clin Oncol 13:326, 1994 (abstr 1078)
Ettinger DS, Finkelstein DM, Sarma R, et al: Phase II study of paclitaxel in patients with extensive-stage small cell lung cancer: An Eastern Cooperative Oncology Group study. J Clin Oncol 13:1430-1435, 1995
Smitt EF, Fokkema E, Biesma B, et al: A phase II study of paclitaxel in heavily pretreated patients with small cell lung cancer. Br J Cancer 77:347-351, 1998
Hainsworth JD, Gray JR, Stroup SL, et al: Paclitaxel, carboplatin and extended-schedule etoposide in the treatment of small cell lung cancer: Comparison of sequential trials using different dose intensities. J Clin Oncol 15:3464-3470, 1997
Kelly K, Pan Z, Wood ME, et al: A phase I study of paclitaxel, etoposide and cisplatin in extensive stage small cell lung cancer. Clin Cancer Res 5:3419-3424, 1999
Levitan N, McKenney J, Tahsildar H, et al: Results of a phase I dose escalation trial of paclitaxel, etoposide and cisplatin followed by filgrastim in the treatment of patients with extensive stage small cell lung cancer. Proc Am Soc Clin Oncol 14:379, 1995 (abstr 1177)
Glisson BS, Kurie JM, Perez-Solar R, et al: Cisplatin, etoposide and paclitaxel in the treatment of patients with extensive small-cell lung cancer. J Clin Oncol 17:2309-2315, 1999
Niell HB, Perry MC, Clamon J, et al: Carboplatin, etoposide and paclitaxel in the treatment of patients with extensive small cell lung cancer. Clin Lung Cancer 2:204-209, 2001
Bunn PA, Kelley K, Crowley J, et al: Preliminary toxicity results from Southwest Oncology Group Trial: A phase II trial of cisplatin, etoposide and paclitaxel with G-CSF in untreated patients with extensive small cell lung cancer. Proc Am Soc Clin Oncol 18:468a, 1999 (abstr 1807)
O'Fallon JR: Policies for interim analysis and interim reporting of results. Cancer Treat Rep 69:1101-1105, 1985
Lan GGK, DeMets DL: Discrete sequential boundaries for clinical trials. Biometrika 70:659-663, 1983
O'Brien PC, Fleming TR: A multiple testing procedure for clinical trials. Biometrics 35:549-556, 1979
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Peto R, Peto J: Asymptotically efficient rank invariant test procedures. J R Stat Soc A 135:185-198, 1972
Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972
Hosmer DW, Lemeshow S: Applied Logistic Regression. New York, NY, Wiley, 1989
Reck M, von Pawel J, Macka H-N, et al: Randomized phase III trial of paclitaxel, etoposide and carboplatin versus carboplatin, etoposide and vincristine in patients with small cell lung cancer. J Natl Cancer Inst 95:1118-1127, 2003
Mavroudis D, Papadakis E, Veslemes M, et al: The Greek Lung Group Cooperative Group: A multicenter randomized phase III study comparing paclitaxel, cisplatin, etoposide versus cisplatin, etoposide as front line treatment in patients with small cell lung cancer. Proc Am Soc Clin Oncol 19:484a, 2000 (abstr 1894)
Findlay MP, Griffin AM, Raghavan D, et al: Retrospective review of chemotherapy for small cell lung cancer in the elderly: Does the end justify the means Eur J Cancer 27:1597-1601, 1991
Smit EF, Carney DN, Harford P, et al: A phase II study of oral etoposide in elderly patients with small cell lung cancer. Thorax 44:631-633, 1989
Earl HM, Rudd RM, Spiro SG, et al: A randomized trial of planned versus as required chemotherapy in small cell lung cancer: A Cancer Research Campaign trial. Br J Cancer 64:566-572, 1991
Medical Research Council Lung Cancer Working Party: Randomized trial of four-drug versus less intensive two-drug chemotherapy in the palliative treatment of patients with small cell lung cancer and poor prognosis. Br J Cancer 73:406-413, 1996
Ihde DC, Mulshine JL, Kramer BS, et al: Prospective randomized comparison of high-dose and standard dose etoposide and cisplatin chemotherapy in patients with extensive stage small cell lung cancer. J Clin Oncol 12:2022-2034, 1994(Harvey B. Niell, James E.)