Single-Agent Versus Combination Chemotherapy in Advanced Non–Small-Cell Lung Cancer: The Cancer and Leukemia Group B (study 9730)
http://www.100md.com
《临床肿瘤学》
the Mount Sinai Comprehensive Cancer Center, Miami Beach, FL
The Cancer and Leukemia Group B Statistical Center, Durham, NC
University of Chicago Cancer Research Center, Chicago, IL
Virginia Commonwealth University, Massey Cancer Center, Richmond, VA
Wake Forest University School of Medicine, Winston-Salem, NC
SUNY Upstate Medical University, Syracuse NY
University of Missouri/Ellis Fischel Cancer Center, Columbia, MO
University of California at San Diego, San Diego, CA
Mount Sinai School of Medicine and Memorial Sloan Kettering Cancer Center, New York, NY
The Dana Farber Cancer Institute, Boston, MA
Medical University of South Carolina, Charleston, SC
ABSTRACT
PATIENTS AND METHODS: A total of 561 eligible patients were randomly assigned to receive paclitaxel alone or in combination with carboplatin.
RESULTS: The response rate was 17% in the paclitaxel arm and 30% in the carboplatin-paclitaxel arm (P < .0001). Median failure-free survival was 2.5 months in the paclitaxel arm and 4.6 months in the carboplatin-paclitaxel arm (P = .0002). Median survival times were 6.7 months (95% CI, 5.8 to 7.8) and 8.8 months (95% CI, 8.0 to 9.9), and 1-year survival rates were 32% (95% CI, 27% to 38%), and 37% (95% CI, 32% to 43%), respectively. The overall survival distributions were not statistically different: hazard ratio = 0.91 (95% CI, 0.77 to 1.17; P = .25). Hematological toxicity and nausea were more frequent in the combination arm, but febrile neutropenia and toxic deaths were equally low in both arms. There was no significant survival difference in elderly patients. Performance status 2 patients treated with combination chemotherapy had a better survival rate than those treated with single-agent therapy (P = .019).
CONCLUSION: Combination chemotherapy improves response rate and failure-free survival compared with single-agent therapy, but there was no statistically significant difference in the primary end point of overall survival. The results in elderly patients were similar to younger patients. Performance status 2 patients had a superior outcome when treated with combination chemotherapy.
INTRODUCTION
During the 1990s, several new chemotherapeutic agents were tested in advanced NSCLC. When used in combination with a platinum analog, these third-generation regimens produced better response and survival compared with the older regimens. Recent trials have shown similar efficacy but different toxicity profiles among these now standard combinations regimens.4-6
Over the last two decades, several trials compared single-agent therapy with combination chemotherapy in advanced NSCLC. A systematic review of 25 of these trials demonstrated higher response rates, but significantly more toxicity, in patients treated with combination chemotherapy. Survival was modestly improved with combination chemotherapy, but not significantly so when a more active drug was used as a single agent.7 With the advent of the newer and more active agents, the question of superiority of combination chemotherapy over optimal single-agent therapy became more relevant. Further, the issue of whether a potential survival advantage provided by combination chemotherapy could be offset by a decrement in quality of life remained unknown. Last, no prospective evaluation of the economic impact of combination chemotherapy as opposed to single-agent therapy was available.
On the basis of this rationale, the Cancer and Leukemia Group B (CALGB) conducted a randomized phase III trial (CALGB 9730) of combination chemotherapy versus single-agent therapy. The quality of life and economic components are not presented in this report.
PATIENTS AND METHODS
Treatment Plan
Patients were randomly assigned to treatment with paclitaxel alone or in combination with carboplatin. Randomization was centralized at the CALGB data management center in Durham, NC. Patient randomization was stratified by stage (IIIB v IV v recurrent), PS (0 to 1 v 2), and age (< 70 v ≥ 70 years of age). Paclitaxel was administered intravenously over 3 hours at a dose of 225 mg/m2, on day 1, in both arms. Carboplatin was administered intravenously over 30 minutes, after paclitaxel, at a dose calculated to produce an area under the concentration-time curve (AUC) of 6.0 mg/mL/min. Both treatments were repeated every 3 weeks for a maximum of six cycles. Patients who developed febrile neutropenia or grade 4 neutropenia lasting more than 5 days received filgrastim in all subsequent cycles. Dose modifications were recommended for selected toxicities. Second-line chemotherapy was given at the physician's discretion, and the regimens used were documented as part of the follow-up. Patients were followed every 3 months for 2 years, then every 6 months for 3 years until disease progression.
Response was assessed by imaging studies every two cycles. Response criteria were standard: a complete response was defined as the absence of disease at all known sites; a partial response was defined as a 50% or greater reduction in the sum of the perpendicular diameters of all measurable lesions, with no new lesions or enlargement of existing lesions; stable disease was defined as less than 50% reduction or less than 25% increase in all measurable lesions with the appearance of no new lesions; progressive disease was defined as a 25% increase in the product of two perpendicular diameters of any measured lesion or the development of new lesions.
Statistical Analysis
CALGB 9730 was designed with 80% power to detect a 30% improvement in median survival from 7.3 months in the paclitaxel arm to 9.5 months in the paclitaxel and paclitaxel-carboplatin arm (hazard ratio [HR] of 1.3). Assuming survival comparisons by the log-rank test, at a two-sided significance level of {alpha} = .05, a total of 458 deaths were required. Truncated O'Brien-Fleming boundaries that were used for interim monitoring required the number of deaths to be increased to 480 deaths. The CALGB Data Safety and Monitoring Board monitored that status and progress of CALGB 9730 semiannually.
Overall survival was calculated from the date of randomization to the date of death. Failure-free survival was calculated from the date of randomization to the date of progression, relapse, or death. All survival distributions were calculated using the Kaplan-Meier method. Comparisons of survival were performed by the log-rank test for censored data. The sample median was defined as the first time at which the sample survival function ≤ 0.5, according to the S-Plus 4 guide. The Cox proportional hazards model was used to assess whether treatment differences were consistent across age groups (< 70 v ≥ 70) and across PS groups (0 to 1 v 2). Relationships among dichotomized data were examined using the Fisher's exact test.
RESULTS
The overall response rate for patients treated with single-agent paclitaxel was 17%, whereas it was 30% for those treated with paclitaxel-carboplatin (P < .001; Table 2). The median failure-free survival was 2.5 months in the paclitaxel arm and significantly longer at 4.6 months in the paclitaxel-carboplatin arm (P = .0005; Fig 1A). Median survival times were 6.7 months and 8.8 months, respectively. The percentage of patients alive at 1 year was 32% in the paclitaxel arm, and 37% in the paclitaxel-carboplatin arm (Fig 1B). When analyzed by the log-rank test, the overall survival distributions were not statistically significant (HR = 0.91; 95% CI, 0.77 to 1.17; P = .25).
Anemia, neutropenia, and thrombocytopenia were greater in the combination arm (Table 3). However, the incidence of febrile neutropenia (5% v 8%) and of toxic deaths (one patient in each arm) was similar between the two treatments. The incidence of nausea or emesis was significantly higher in the combination arm, while the incidence of fatigue, peripheral neuropathy, and other nonhematological toxicities was similar between the two arms.
Data on second-line therapy were obtained from approximately 84% of the patients (Table 2). Thirty-nine percent of patients initially treated with single-agent paclitaxel received second-line therapy compared with 32% for patients initially treated with combination chemotherapy (P = .09). Further, 14% of patients in the single-agent arm received a platinum-based combination as second-line therapy compared with 7% of patients initially treated with paclitaxel-carboplatin (P = .005). The remaining patients were treated with a variety of nonplatinum combinations and single agents. Approximately 45% of patients in the paclitaxel arm and 53% in the paclitaxel-carboplatin arm died without receiving second-line therapy.
Subset Analyses: Special Populations
Patient outcome was analyzed within patient subgroup defined by age and PS, as prespecified in the study design (Table 4). For elderly patients, the 1-year survival for combination chemotherapy was 35% v 31% for single-agent therapy; HR = 0.84, P = .29; the corresponding data for younger patients were 38% and 33%, respectively; HR = 0.98, P = .50. The treatment arm difference observed among older patients was similar to that observed among younger patients (P = .546).
Patients with PS 2 had a significantly worse outcome compared with patients with PS 0 to 1: median survival of 3.0 months and 1-year survival of 14% versus 8.8 months and 38%, respectively (Table 5; Figs 1C and 1D). For patients with PS 2, the 1-year survival for combination chemotherapy was 18% v 10% for single agent; HR = 0.60, P = .016. The corresponding data for PS 0 or PS 1 patients was 41% and 38%; HR = 0.95, P = .59. The treatment arm difference observed among patients with PS 2 is significantly different from that observed among PS 0 or PS 1 patients (P = .019). Rates of toxicities between the two arms in both subgroups were comparable to the general study populations.
DISCUSSION
At the time our study was designed, it remained questionable whether combination chemotherapy was indeed superior to optimal single-agent therapy. Further, in the context of a disease with a median survival measured in months, it was unknown whether a potential survival advantage for combination chemotherapy could be offset by a negative impact in quality of life. And last, the economic impact of combination chemotherapy had not been adequately studied in a prospective manner. Thus, our study was designed to frame the question of optimal management of advanced NSCLC patients in a broader context of quality of life and cost-effectiveness.
Our results showed a modest, nonsignificant trend toward improvement in survival in favor of combination chemotherapy. This effect is evident early in the course of the illness, although it was not sustained at 1 year. There are two possible explanations for this observation. One is the impact of second-line therapy, which has been shown to improve survival compared with no therapy or ineffective therapy.9,10 A higher frequency of second-line therapy using platinum-based doublets may have favorably influenced the outcome of patients initially treated with single-agent therapy. The second explanation is related to the clinical heterogeneity of advanced NSCLC patients. Combination chemotherapy appears to have a greater impact in patients with more aggressive disease who are likely to die rapidly without effective therapy. This is corroborated by the results observed in PS 2 subgroup. In patients with biologically more indolent disease, who seem destined to survive longer, however, the initial treatment strategy is less critical and combination chemotherapy may not provide an advantage over single-agent therapy. Because there are as yet no validated tools that can identify disease behavior at the time of the diagnosis, it is reasonable to recommend combination chemotherapy as initial therapy for most patients with advanced NSCLC.
The observation that elderly patients have a similar outcome compared with younger patients has been made in retrospective analyses of other cooperative group trials.11,12 It is important to emphasize that the elderly patients enrolled in our trial were thought to be appropriate to receive a platinum-based combination and, therefore, may not be representative of all elderly patients with advanced NSCLC. A recent Italian trial in elderly patients with several comorbid conditions compared the combination of vinorelbine and gemcitabine with either one of the two agents given alone, and did not show a survival or a quality-of-life advantage for the combination.13 It is unclear whether this discrepancy is because of patient selection, the chemotherapy utilized, or both. Yet, based on our experience, age alone should not determine the treatment strategy, but rather the patient's functional status and the ability to tolerate combination chemotherapy.14 This recommendation is further supported by the absence of a decline in overall quality of life of elderly patients compared with younger patients.
Previous trials have usually excluded PS 2 patients with advanced NSCLC. A notable exception is the Eastern Cooperative Oncology Group 1594 trial, a comparison of four combination chemotherapy regimens in previously untreated patients with advanced NSCLC, which enrolled 64 eligible patients with PS of 2.15 This subset had a high rate of mortality, leading to early discontinuation of accrual. Although a subsequent analysis showed that the underlying illness and not treatment toxicity was responsible for their poor outcome, the authors concluded that platinum-based chemotherapy should not be recommended to PS 2 patients. In our study, the overall survival of PS 2 patients was indeed much poorer than patients with PS 0 to 1. However, and importantly, PS 2 patients treated with combination chemotherapy had a significantly better survival compared with those who received single-agent therapy, without an increase in toxicity. Therefore, although further trials specifically designed for PS 2 patients are needed, combination chemotherapy can be a reasonable strategy in PS 2 patients who wish to pursue more aggressive treatment.
In summary, combination chemotherapy was associated with an improvement in response rate and failure-free survival, but overall survival was not significantly superior to single-agent therapy. A platinum-based doublet remains a reasonable strategy for patients with advanced NSCLC. This recommendation can also be applied to elderly patients and PS 2 patients who desire aggressive treatment.
Appendix
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Patricia S. Graham for her invaluable assistance in the preparation of this manuscript.
NOTES
The research for CALGB 9730 was supported, in part, by grants from the National Cancer Institute (CA31946) to the Cancer and Leukemia Group B (Richard L. Schilsky, Chairman). The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. This research was also partially supported by Bristol-Myers Squibb Company. This research also supported by CA45564, CA33601, CA41287, CA52784, CA03927, CA21060, CA12046, CA11789, CA04457, CA32291, CA77651, CA03927. Additional information appears in the Appendix.
Presented at the plenary Session, American Society of Clinical Oncology Group Meeting, Orlando, FL, May 2002.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Jemal A, Thomas A, Murray T, et al: Cancer statistics 2003. CA Cancer J Clin 53:5-26, 2003
2. Non-Small Cell Lung Cancer Co-Operative Group. Chemotherapy in non-small-cell lung cancer: A meta-analysis using updated data on individual patients from 52 randomized clinical trials. Br Med J 311:899-909, 1995
3. Cullen M, Billingham J, Woodraffe C, et al: Mitomycin, ifosfamide, and cisplatin in unresectable non-small cell lung cancer: Effects on survival and quality of life. J Clin Oncol 17:3188-3194, 1999
4. Kelly K, Crowley J, Bunn PA, et al: Randomized phase III trial of paclitaxel plus carboplatin versus vinorelbine plus cisplatin in the treatment of patients with advanced non-small cell lung cancer: A Southwest Oncology Group trial. J Clin Oncol 19:3210-3218, 2001
5. 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
6. Rodriguez J, Pawel J, Pluzanska A, et al: A multicenter, randomized phase III study of docetaxel + cisplatin (DC) and docetaxel + carboplatin (DCb) vs. vinorelbine + cisplatin (VC) in chemotherapy-nave patients with advanced and metastatic non-small cell lung cancer. Proc Am Soc Clin Oncol 20:20a, 2001 (abstr 1252)
7. Lilenbaum R, Langenberg P, Dickersin K: Single agent versus combination chemotherapy in patients with advanced non-small cell lung carcinoma: A meta-analysis of response, toxicity, and survival. Cancer 82:116-126, 1998
8. Schiller JH, Bernardo P, Harrington D, et al: Comparison of outcome and patient characteristics in advanced non-small cell lung cancer: Analysis of Eastern Cooperative Oncology Group trials from 1980-2000. Proc Am Soc Clin Oncol 21:304a, 2002 (abstr 1215)
9. Shepherd F, Dancey J, Ramlau R, et al: A prospective randomized trial of docetaxel (Taxotere) versus best supportive care in patients with non-small cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 18:2095-2103, 2000
10. Fossella FV, DeVore R, Kerr RN, et al: Randomized phase III trial of docetaxel versus vinorelbine or ifosfamide in patients with advanced non-small cell lung cancer previously treated with platinum-containing chemotherapy regimens. J Clin Oncol 18:2354-2362, 2000
11. Langer CJ, Manola J, Bernardo P, et al: Cisplatin-based therapy for elderly patients with advanced non-small cell lung cancer: Implications of Eastern Cooperative Oncology Group 5592, a randomized trial. J Natl Cancer Inst 94:173-181, 2002
12. Kelly K, Giarritta S, Hayes S, et al: Should older patients receive combination chemotherapy for advanced non-small cell lung cancer An analysis of Southwest Oncology trials 9509 and 9308. Proc Am Soc Clin Oncol 20:329a, 2001 (abstr 1313)
13. Gridelli C, Perronet F, Gallo C, et al: Chemotherapy for elderly patients with advanced non-small cell lung cancer: The Multicenter Italian Lung Cancer in the Elderly Study (MILES) phase III randomized trial. J Natl Cancer Inst 95:362-372, 2003
14. Bunn P, Lilenbaum R: Chemotherapy for elderly patients with advanced non-small cell lung cancer. J Natl Cancer Inst 95:341-343, 2003
15. Sweeney C, Zhu J, Sandler A, et al: Outcome of patients with a performance status of 2 in Eastern Cooperative Oncology Group Study E1594: A phase III trial in metastatic non-small cell lung carcinoma. Cancer 92:2635-2647, 2001(Rogerio C. Lilenbaum, Jam)
The Cancer and Leukemia Group B Statistical Center, Durham, NC
University of Chicago Cancer Research Center, Chicago, IL
Virginia Commonwealth University, Massey Cancer Center, Richmond, VA
Wake Forest University School of Medicine, Winston-Salem, NC
SUNY Upstate Medical University, Syracuse NY
University of Missouri/Ellis Fischel Cancer Center, Columbia, MO
University of California at San Diego, San Diego, CA
Mount Sinai School of Medicine and Memorial Sloan Kettering Cancer Center, New York, NY
The Dana Farber Cancer Institute, Boston, MA
Medical University of South Carolina, Charleston, SC
ABSTRACT
PATIENTS AND METHODS: A total of 561 eligible patients were randomly assigned to receive paclitaxel alone or in combination with carboplatin.
RESULTS: The response rate was 17% in the paclitaxel arm and 30% in the carboplatin-paclitaxel arm (P < .0001). Median failure-free survival was 2.5 months in the paclitaxel arm and 4.6 months in the carboplatin-paclitaxel arm (P = .0002). Median survival times were 6.7 months (95% CI, 5.8 to 7.8) and 8.8 months (95% CI, 8.0 to 9.9), and 1-year survival rates were 32% (95% CI, 27% to 38%), and 37% (95% CI, 32% to 43%), respectively. The overall survival distributions were not statistically different: hazard ratio = 0.91 (95% CI, 0.77 to 1.17; P = .25). Hematological toxicity and nausea were more frequent in the combination arm, but febrile neutropenia and toxic deaths were equally low in both arms. There was no significant survival difference in elderly patients. Performance status 2 patients treated with combination chemotherapy had a better survival rate than those treated with single-agent therapy (P = .019).
CONCLUSION: Combination chemotherapy improves response rate and failure-free survival compared with single-agent therapy, but there was no statistically significant difference in the primary end point of overall survival. The results in elderly patients were similar to younger patients. Performance status 2 patients had a superior outcome when treated with combination chemotherapy.
INTRODUCTION
During the 1990s, several new chemotherapeutic agents were tested in advanced NSCLC. When used in combination with a platinum analog, these third-generation regimens produced better response and survival compared with the older regimens. Recent trials have shown similar efficacy but different toxicity profiles among these now standard combinations regimens.4-6
Over the last two decades, several trials compared single-agent therapy with combination chemotherapy in advanced NSCLC. A systematic review of 25 of these trials demonstrated higher response rates, but significantly more toxicity, in patients treated with combination chemotherapy. Survival was modestly improved with combination chemotherapy, but not significantly so when a more active drug was used as a single agent.7 With the advent of the newer and more active agents, the question of superiority of combination chemotherapy over optimal single-agent therapy became more relevant. Further, the issue of whether a potential survival advantage provided by combination chemotherapy could be offset by a decrement in quality of life remained unknown. Last, no prospective evaluation of the economic impact of combination chemotherapy as opposed to single-agent therapy was available.
On the basis of this rationale, the Cancer and Leukemia Group B (CALGB) conducted a randomized phase III trial (CALGB 9730) of combination chemotherapy versus single-agent therapy. The quality of life and economic components are not presented in this report.
PATIENTS AND METHODS
Treatment Plan
Patients were randomly assigned to treatment with paclitaxel alone or in combination with carboplatin. Randomization was centralized at the CALGB data management center in Durham, NC. Patient randomization was stratified by stage (IIIB v IV v recurrent), PS (0 to 1 v 2), and age (< 70 v ≥ 70 years of age). Paclitaxel was administered intravenously over 3 hours at a dose of 225 mg/m2, on day 1, in both arms. Carboplatin was administered intravenously over 30 minutes, after paclitaxel, at a dose calculated to produce an area under the concentration-time curve (AUC) of 6.0 mg/mL/min. Both treatments were repeated every 3 weeks for a maximum of six cycles. Patients who developed febrile neutropenia or grade 4 neutropenia lasting more than 5 days received filgrastim in all subsequent cycles. Dose modifications were recommended for selected toxicities. Second-line chemotherapy was given at the physician's discretion, and the regimens used were documented as part of the follow-up. Patients were followed every 3 months for 2 years, then every 6 months for 3 years until disease progression.
Response was assessed by imaging studies every two cycles. Response criteria were standard: a complete response was defined as the absence of disease at all known sites; a partial response was defined as a 50% or greater reduction in the sum of the perpendicular diameters of all measurable lesions, with no new lesions or enlargement of existing lesions; stable disease was defined as less than 50% reduction or less than 25% increase in all measurable lesions with the appearance of no new lesions; progressive disease was defined as a 25% increase in the product of two perpendicular diameters of any measured lesion or the development of new lesions.
Statistical Analysis
CALGB 9730 was designed with 80% power to detect a 30% improvement in median survival from 7.3 months in the paclitaxel arm to 9.5 months in the paclitaxel and paclitaxel-carboplatin arm (hazard ratio [HR] of 1.3). Assuming survival comparisons by the log-rank test, at a two-sided significance level of {alpha} = .05, a total of 458 deaths were required. Truncated O'Brien-Fleming boundaries that were used for interim monitoring required the number of deaths to be increased to 480 deaths. The CALGB Data Safety and Monitoring Board monitored that status and progress of CALGB 9730 semiannually.
Overall survival was calculated from the date of randomization to the date of death. Failure-free survival was calculated from the date of randomization to the date of progression, relapse, or death. All survival distributions were calculated using the Kaplan-Meier method. Comparisons of survival were performed by the log-rank test for censored data. The sample median was defined as the first time at which the sample survival function ≤ 0.5, according to the S-Plus 4 guide. The Cox proportional hazards model was used to assess whether treatment differences were consistent across age groups (< 70 v ≥ 70) and across PS groups (0 to 1 v 2). Relationships among dichotomized data were examined using the Fisher's exact test.
RESULTS
The overall response rate for patients treated with single-agent paclitaxel was 17%, whereas it was 30% for those treated with paclitaxel-carboplatin (P < .001; Table 2). The median failure-free survival was 2.5 months in the paclitaxel arm and significantly longer at 4.6 months in the paclitaxel-carboplatin arm (P = .0005; Fig 1A). Median survival times were 6.7 months and 8.8 months, respectively. The percentage of patients alive at 1 year was 32% in the paclitaxel arm, and 37% in the paclitaxel-carboplatin arm (Fig 1B). When analyzed by the log-rank test, the overall survival distributions were not statistically significant (HR = 0.91; 95% CI, 0.77 to 1.17; P = .25).
Anemia, neutropenia, and thrombocytopenia were greater in the combination arm (Table 3). However, the incidence of febrile neutropenia (5% v 8%) and of toxic deaths (one patient in each arm) was similar between the two treatments. The incidence of nausea or emesis was significantly higher in the combination arm, while the incidence of fatigue, peripheral neuropathy, and other nonhematological toxicities was similar between the two arms.
Data on second-line therapy were obtained from approximately 84% of the patients (Table 2). Thirty-nine percent of patients initially treated with single-agent paclitaxel received second-line therapy compared with 32% for patients initially treated with combination chemotherapy (P = .09). Further, 14% of patients in the single-agent arm received a platinum-based combination as second-line therapy compared with 7% of patients initially treated with paclitaxel-carboplatin (P = .005). The remaining patients were treated with a variety of nonplatinum combinations and single agents. Approximately 45% of patients in the paclitaxel arm and 53% in the paclitaxel-carboplatin arm died without receiving second-line therapy.
Subset Analyses: Special Populations
Patient outcome was analyzed within patient subgroup defined by age and PS, as prespecified in the study design (Table 4). For elderly patients, the 1-year survival for combination chemotherapy was 35% v 31% for single-agent therapy; HR = 0.84, P = .29; the corresponding data for younger patients were 38% and 33%, respectively; HR = 0.98, P = .50. The treatment arm difference observed among older patients was similar to that observed among younger patients (P = .546).
Patients with PS 2 had a significantly worse outcome compared with patients with PS 0 to 1: median survival of 3.0 months and 1-year survival of 14% versus 8.8 months and 38%, respectively (Table 5; Figs 1C and 1D). For patients with PS 2, the 1-year survival for combination chemotherapy was 18% v 10% for single agent; HR = 0.60, P = .016. The corresponding data for PS 0 or PS 1 patients was 41% and 38%; HR = 0.95, P = .59. The treatment arm difference observed among patients with PS 2 is significantly different from that observed among PS 0 or PS 1 patients (P = .019). Rates of toxicities between the two arms in both subgroups were comparable to the general study populations.
DISCUSSION
At the time our study was designed, it remained questionable whether combination chemotherapy was indeed superior to optimal single-agent therapy. Further, in the context of a disease with a median survival measured in months, it was unknown whether a potential survival advantage for combination chemotherapy could be offset by a negative impact in quality of life. And last, the economic impact of combination chemotherapy had not been adequately studied in a prospective manner. Thus, our study was designed to frame the question of optimal management of advanced NSCLC patients in a broader context of quality of life and cost-effectiveness.
Our results showed a modest, nonsignificant trend toward improvement in survival in favor of combination chemotherapy. This effect is evident early in the course of the illness, although it was not sustained at 1 year. There are two possible explanations for this observation. One is the impact of second-line therapy, which has been shown to improve survival compared with no therapy or ineffective therapy.9,10 A higher frequency of second-line therapy using platinum-based doublets may have favorably influenced the outcome of patients initially treated with single-agent therapy. The second explanation is related to the clinical heterogeneity of advanced NSCLC patients. Combination chemotherapy appears to have a greater impact in patients with more aggressive disease who are likely to die rapidly without effective therapy. This is corroborated by the results observed in PS 2 subgroup. In patients with biologically more indolent disease, who seem destined to survive longer, however, the initial treatment strategy is less critical and combination chemotherapy may not provide an advantage over single-agent therapy. Because there are as yet no validated tools that can identify disease behavior at the time of the diagnosis, it is reasonable to recommend combination chemotherapy as initial therapy for most patients with advanced NSCLC.
The observation that elderly patients have a similar outcome compared with younger patients has been made in retrospective analyses of other cooperative group trials.11,12 It is important to emphasize that the elderly patients enrolled in our trial were thought to be appropriate to receive a platinum-based combination and, therefore, may not be representative of all elderly patients with advanced NSCLC. A recent Italian trial in elderly patients with several comorbid conditions compared the combination of vinorelbine and gemcitabine with either one of the two agents given alone, and did not show a survival or a quality-of-life advantage for the combination.13 It is unclear whether this discrepancy is because of patient selection, the chemotherapy utilized, or both. Yet, based on our experience, age alone should not determine the treatment strategy, but rather the patient's functional status and the ability to tolerate combination chemotherapy.14 This recommendation is further supported by the absence of a decline in overall quality of life of elderly patients compared with younger patients.
Previous trials have usually excluded PS 2 patients with advanced NSCLC. A notable exception is the Eastern Cooperative Oncology Group 1594 trial, a comparison of four combination chemotherapy regimens in previously untreated patients with advanced NSCLC, which enrolled 64 eligible patients with PS of 2.15 This subset had a high rate of mortality, leading to early discontinuation of accrual. Although a subsequent analysis showed that the underlying illness and not treatment toxicity was responsible for their poor outcome, the authors concluded that platinum-based chemotherapy should not be recommended to PS 2 patients. In our study, the overall survival of PS 2 patients was indeed much poorer than patients with PS 0 to 1. However, and importantly, PS 2 patients treated with combination chemotherapy had a significantly better survival compared with those who received single-agent therapy, without an increase in toxicity. Therefore, although further trials specifically designed for PS 2 patients are needed, combination chemotherapy can be a reasonable strategy in PS 2 patients who wish to pursue more aggressive treatment.
In summary, combination chemotherapy was associated with an improvement in response rate and failure-free survival, but overall survival was not significantly superior to single-agent therapy. A platinum-based doublet remains a reasonable strategy for patients with advanced NSCLC. This recommendation can also be applied to elderly patients and PS 2 patients who desire aggressive treatment.
Appendix
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Patricia S. Graham for her invaluable assistance in the preparation of this manuscript.
NOTES
The research for CALGB 9730 was supported, in part, by grants from the National Cancer Institute (CA31946) to the Cancer and Leukemia Group B (Richard L. Schilsky, Chairman). The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. This research was also partially supported by Bristol-Myers Squibb Company. This research also supported by CA45564, CA33601, CA41287, CA52784, CA03927, CA21060, CA12046, CA11789, CA04457, CA32291, CA77651, CA03927. Additional information appears in the Appendix.
Presented at the plenary Session, American Society of Clinical Oncology Group Meeting, Orlando, FL, May 2002.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Jemal A, Thomas A, Murray T, et al: Cancer statistics 2003. CA Cancer J Clin 53:5-26, 2003
2. Non-Small Cell Lung Cancer Co-Operative Group. Chemotherapy in non-small-cell lung cancer: A meta-analysis using updated data on individual patients from 52 randomized clinical trials. Br Med J 311:899-909, 1995
3. Cullen M, Billingham J, Woodraffe C, et al: Mitomycin, ifosfamide, and cisplatin in unresectable non-small cell lung cancer: Effects on survival and quality of life. J Clin Oncol 17:3188-3194, 1999
4. Kelly K, Crowley J, Bunn PA, et al: Randomized phase III trial of paclitaxel plus carboplatin versus vinorelbine plus cisplatin in the treatment of patients with advanced non-small cell lung cancer: A Southwest Oncology Group trial. J Clin Oncol 19:3210-3218, 2001
5. 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
6. Rodriguez J, Pawel J, Pluzanska A, et al: A multicenter, randomized phase III study of docetaxel + cisplatin (DC) and docetaxel + carboplatin (DCb) vs. vinorelbine + cisplatin (VC) in chemotherapy-nave patients with advanced and metastatic non-small cell lung cancer. Proc Am Soc Clin Oncol 20:20a, 2001 (abstr 1252)
7. Lilenbaum R, Langenberg P, Dickersin K: Single agent versus combination chemotherapy in patients with advanced non-small cell lung carcinoma: A meta-analysis of response, toxicity, and survival. Cancer 82:116-126, 1998
8. Schiller JH, Bernardo P, Harrington D, et al: Comparison of outcome and patient characteristics in advanced non-small cell lung cancer: Analysis of Eastern Cooperative Oncology Group trials from 1980-2000. Proc Am Soc Clin Oncol 21:304a, 2002 (abstr 1215)
9. Shepherd F, Dancey J, Ramlau R, et al: A prospective randomized trial of docetaxel (Taxotere) versus best supportive care in patients with non-small cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 18:2095-2103, 2000
10. Fossella FV, DeVore R, Kerr RN, et al: Randomized phase III trial of docetaxel versus vinorelbine or ifosfamide in patients with advanced non-small cell lung cancer previously treated with platinum-containing chemotherapy regimens. J Clin Oncol 18:2354-2362, 2000
11. Langer CJ, Manola J, Bernardo P, et al: Cisplatin-based therapy for elderly patients with advanced non-small cell lung cancer: Implications of Eastern Cooperative Oncology Group 5592, a randomized trial. J Natl Cancer Inst 94:173-181, 2002
12. Kelly K, Giarritta S, Hayes S, et al: Should older patients receive combination chemotherapy for advanced non-small cell lung cancer An analysis of Southwest Oncology trials 9509 and 9308. Proc Am Soc Clin Oncol 20:329a, 2001 (abstr 1313)
13. Gridelli C, Perronet F, Gallo C, et al: Chemotherapy for elderly patients with advanced non-small cell lung cancer: The Multicenter Italian Lung Cancer in the Elderly Study (MILES) phase III randomized trial. J Natl Cancer Inst 95:362-372, 2003
14. Bunn P, Lilenbaum R: Chemotherapy for elderly patients with advanced non-small cell lung cancer. J Natl Cancer Inst 95:341-343, 2003
15. Sweeney C, Zhu J, Sandler A, et al: Outcome of patients with a performance status of 2 in Eastern Cooperative Oncology Group Study E1594: A phase III trial in metastatic non-small cell lung carcinoma. Cancer 92:2635-2647, 2001(Rogerio C. Lilenbaum, Jam)