1 Department of Medicine II, Okayama University Medical School, 2-5-1, Shikata-cho, Okayama, 700-8558; 2 Department of Epidemiology and Prevention, Aichi Cancer Center Research Institute, 1-1, Kanokoden, Chikusa-ku, Nagoya, 464-8681, Japan
* Correspondence to: Dr K. Hotta, Department of Medicine II, Okayama University Medical School, 2-5-1, Shikata-cho, Okayama, 700-8558, Japan. Tel: +81-86-235-7227; Fax: +81-86-232-8226; Email: khotta{at}md.okayama-u.ac.jp
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods: We carried out a literature search to identify trials, conducted between 1994 and 2003, comparing a doublet of platinum plus a new agent with a new agent alone in previously untreated patients with advanced NSCLC. Outcomes analysed were response, survival and toxicity.
Results: Eight trials encompassing 2374 patients were identified. Platinum-based doublets produced an approximately two-fold higher overall (complete and partial) response rate than the new agent alone [odds ratio = 2.32; 95% confidence interval (CI)=1.683.20]. Platinum-based doublet therapy was also associated with a 13% prolongation of survival (hazard ratio = 0.87; 95% CI = 0.800.94, P <0.001). Despite significant increases in the frequencies of various toxic effects in patients receiving platinum-based doublets, no significant difference in treatment-related mortality was observed.
Conclusion: This is the first published meta-analysis demonstrating the importance of combining platinum with single new agents in the treatment of advanced NSCLC.
Key words: doublets, non-small-cell lung cancer, platinum, single-agent therapy
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Recently, new agents with novel mechanisms [paclitaxel (Taxol), docetaxel (Taxotere), irinotecan, gemcitabine and vinorelbine] have been developed and some of them have already been reported to produce a significant survival advantage as a single-agent over the best supportive care alone in patients with advanced NSCLC [3, 4
]. Furthermore, doublets consisting of CDDP plus one of these new agents have been shown to improve survival compared to CDDP plus existing agents such as vindesine or etoposide in patients with advanced NSCLC [5
, 6
].
Several randomised trials have thus compared single new agent treatment with doublets consisting of platinum plus one of the new agents [716
]. However, these trials have yielded conflicting survival results. Accordingly, we carried out a meta-analysis to compare the effects of platinum plus a single new agent with single new agent therapy alone on overall survival as well as on overall (complete and partial) response rate and toxicity in patients with advanced NSCLC.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Selection of trials
If at least one platinum-based doublet and one single new agent therapy were included in a randomised trial, it was considered to be eligible. A platinum-based doublet included one platinum agent and one new agent (paclitaxel, docetaxel, irinotecan, gemcitabine, or vinorelbine), and the single agent had to be one of these new agents. Trials were excluded from our analysis if the new agents used in the platinum-based doublet were different from the single-agent therapies. Patients with pathologically confirmed advanced NSCLC who had not previously received chemotherapy were enrolled in these trials.
Validity assessment
We carried out an open assessment of the trials and used the instrument reported by Jadad et al. [17]. However, no evident differences were observed among the trials. Therefore, the result of the validity assessment was not considered in the meta-analysis.
Data abstraction
To avoid bias in the data abstraction process, two observers (K.H. and H.U.) independently abstracted the data from the trials and subsequently compared the results. All data were checked for internal consistency, and disagreements were resolved by discussion among the investigators. We tried to contact principal investigators of the trials to confirm or update both published and unpublished data.
Quantitative data synthesis
We applied odds ratios (ORs) to assess objective response rate and toxic events. We constructed 2 x 2 tables from abstracted data for responses and for each toxic event. ORs and their variances for the subjects who received a platinum-based doublet relative to those receiving single new agent therapy alone were calculated from the tables. For OR calculations, we excluded ineligible subjects from each evaluation.
Hazard ratios (HR) were applied to assess the survival advantage of platinum-based doublets compared with that of single-agent therapy alone. Crude log HR and its variance for each trial were calculated using the abstracted survival probabilities at each time point according to the methods proposed by Parmar et al. [18]. Minimum and maximum follow-up times were used to estimate censored subjects under the assumption that censoring happens constantly throughout follow-up. If the minimum follow-up time was not available, time zero was substituted for it. As we assumed a constant hazard for the two types of therapy within an individual trial, all of the survival probabilities available in each trial were used to obtain a representative HR for each trial instead of limiting time points to specified times. HRs were calculated to represent how many times higher the probability of death was from any cause in patients receiving a platinum-based doublet compared with those receiving single-agent therapy. Therefore, a HR below unity indicates that the platinum-based doublet is better than single-agent therapy.
A general variance-based method was used to estimate the summary HR, ORs and their 95% confidence intervals (CIs). We looked for heterogeneity among the trials based on standard methods [19]. We also calculated the between-study variation (
2) from the Q statistic according to the method described by DerSimonian and Laird [20
]. Regardless of the statistical significance of the Q test, we applied a random effect model which allows meta-analyses to take between-study variation into consideration. We also used Begg's funnel plots [21
] and Egger's test [22
] to detect possible publication bias. Meta-regression analysis was applied to detect the source of heterogeneity in the survival analysis. The factors examined in meta-regression analysis were study quality score [17
], starting year of the trial, proportion of patients with performance status (PS) 01, using the World Health Organization criteria or others proportion of stage IV patients, proportion of male patients, and inclusion of carboplatin. A cumulative meta-analysis was planned in order to take trial quality into consideration when the trial showed quality score heterogeneity [17
].
All statistical analyses were conducted with STATA ver. 8 software (College Station, TX, USA). We defined a statistical test with a P value less than 0.05 as significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In this study, we used data from trials comparing platinum plus one of the new agents with the new agent alone; three trials evaluated in our study were also included in Lilenbaum's study [7, 8
, 11
]. Our results indicate that the addition of platinum to single new agents is important, if they have adequate organ function and good performance status. However, it remains unclear which new drug should be combined with platinum in a platinum-based doublet and further investigation is necessary.
Two meta-analyses on the addition of a second drug for recently advanced NSCLC have been presented [24, 25
]. Both were based on literature data. Using 33 trials with 7872 patients, Delbaldo et al. [24
] demonstrated a significant increase in response rate and survival in favour of two drugs (OR = 0.39; 95% CI = 0.350.45, P <0.001 and HR = 0.79; 95%CI = 0.750.83, P <0.001). Baggstrom et al. [25
] identified 17 trials randomising 4421 patients with advanced NSCLC to one drug or two drugs, and demonstrated that the doublet chemotherapy is superior in terms of overall survival as well as response. Although the eligible trials and patient populations were different among the studies, all three studies including our study indicate that a one-drug regimen is inferior in terms of response and survival.
We included the two trials in which the effect of carboplatin was investigated [12, 13
]. Recently, we carried out another meta-analysis of the trials that compared CDDP-based chemotherapy with carboplatin-based chemotherapy, which revealed that combination chemotherapy consisting of CDDP plus a new agent yields a substantial survival advantage compared with carboplatin plus a new agent in patients with advanced NSCLC [26
]. Thus, inclusion of trials using carboplatin will give conservative P values for summary statistics. However, the highly significant association found in the current analysis indicates that CDDP is important in the treatment of advanced NSCLC. In addition, meta-regression analysis in this study showed that inclusion of trials using carboplatin did not change the results in our study, which suggests that the inclusion of the two trials using carboplatin did not alter our main conclusion. Further investigation will be needed to clarify the role of carboplatin in the treatment of advanced NSCLC.
Our study has several limitations. First, one major problem is that our analyses were based on abstracted data, since an individual patient data based meta-analysis would give a more robust estimate of the association [27]. Therefore, physicians should interpret our results carefully. Second, some of the trials we identified were reported in abstract form only, which made it difficult to extract complete data for our meta-analysis. However, additional updated data fully adequate for this meta-analysis were obtained in several cases by contacting the principal investigators. Third, as is often the case with meta-analysis, one must still be cautious in interpreting our results because of the substantial effect of heterogeneity, although we applied a random-effect model to obtain summary statistics. The significant results from the heterogeneity tests for response rate, neutropenia and nausea/vomiting represent potential heterogeneity and may modify the association we found in our study. Possible publication bias is also a potential harm in our study, though we did not detect it statistically. Finally, the dose of new agent used in the platinum-based doublet was different from that in the single-agent therapy in Negoro's trial [10
]. Irinotecan was administered at a dose of 100 mg/m2 in the single-agent therapy, whereas 60 mg/m2 of irinotecan was combined with cisplatin in the combination arm. It might be problematic to analyse the importance of platinum in our study. However, we considered that the effect would be very small, if any, because it occurred in only one of the eight trials and because the same administration schedule for irinotecan was used in both arms.
In conclusion, this is the first published meta-analysis, to our knowledge, of randomised trials of platinum-based doublet versus single new agent therapy alone. Although modest, the survival improvement obtained with the platinum-based doublet in comparison to a single new agent therapy indicates the importance of platinum in the treatment of advanced NSCLC.
![]() |
Acknowledgements |
---|
![]() |
Notes |
---|
Received for publication April 11, 2004. Revision received July 6, 2004. Accepted for publication July 19, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Non-small cell lung cancer collaborative group. Chemotherapy in non-small cell lung cancer: A meta-analysis using updated data on individual patients from 52 randomized clinical trials. BMJ 1995; 311: 899909.
3. Bunn PA. Chemotherapy for advanced non-small-cell lung cancer: who, what, when, why? J Clin Oncol 2002; 20: 23s33s.
4. Ranson M, Davidson N, Nicolson M et al. Randomized trial of paclitaxel plus supportive care versus supportive care for patients with advanced non-small-cell lung cancer. J Natl Cancer Inst 2000; 92: 10741080.
5. Baggstrom MQ, Socinski MA, Hensing TA, Poole C. Third generation chemotherapy regimens (3GR) improve survival over second generation regimens (2GR) in stage IIIB/IV non-small cell lung cancer (NSCLC): a meta-analysis of the published literature. Proc Am Soc Clin Oncol 2002; 21: 306a (Abstr 1222).
6. Yana T, Takada M, Origasa H et al. New chemotherapy agent plus platinum for advanced non-small cell lung cancer: a meta-analysis. Proc Am Soc Clin Oncol 2002; 21: 328a (Abstr 1309).
7. Le Chevalier T, Brisgand D, Douillard J-Y et al. Randomized study of vinorelbine and cisplatin versus vindesine and cisplatin versus vinorelbine alone in advanced non-small-cell lung cancer: Results of a European multicenter trial including 612 patients. J Clin Oncol 1994; 12: 360367.[Abstract]
8. Depierre A, Chastang C, Quoix E et al. Vinorelbine versus vinorelbine plus cisplatin in advanced non-small cell lung cancer: a randomized trial. Ann Oncol 1994; 5: 3742.[Abstract]
9. Lorusso V, Pezzella G, Catino AM et al. Results of a clinical multicentric randomized phase II study of non-small cell lung cancer treated with vinorelbinecisplatin versus vinorelbine alone. Int Oncol 1995; 6: 6568.[ISI]
10. Negoro S, Masuda N, Takada Y et al. Randomized phase III trial of irinotecan combined with cisplatin for advanced non-small-cell lung cancer. Br J Cancer 2003; 88: 335341.[CrossRef][ISI][Medline]
11. Deza EG, Balbiani L, Coppola F et al. Phase III study of navelbine (NVB) vs NVB plus cisplatin in non small cell lung cancer (NSCLC) stage IIIB or IV. Proc Am Soc Clin Oncol 1996; 15: 394 (Abstr 1193).
12. Lilenbaum RC, Herndon J, List M et al. Single-agent versus combination chemotherapy in advanced NSCLC: A CALGB trial of efficacy, quality of life, and cost-effectiveness. Proc Am Soc Clin Oncol 2002; 21: 1a (Abstr 2).
13. Sederholm C. Gemcitabine compared with gemcitabine plus carboplatin in advanced non-small cell lung cancer (NSCLC): a phase III study by Swedish Lung Cancer Study Group (SLUSG). Proc Am Soc Clin Oncol 2002; 21: 291a (Abstr 1162).
14. Georgoulias V, Ardavanis A, Agelidou M et al. Preliminary analysis of a multicenter phase III trial comparing docetaxel versus docetaxel/cisplatin in patients with inoperable advanced and inoperable advanced and metastatic non-small cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 2002; 21: 291a (Abstr 1163).
15. Monnier A, Zatloukal P, Siffnerova H et al. Docetaxel (D) versus vinorelbine and cisplatin (VC) in chemotherapy-naïve patients with advanced non-small cell lung cancer (NSCLC) and with a normal level of serum alpha 1 acid glycoprotein (AAG): Interim results of a multinational, randomised phase II. Ann Oncol 2002; 13 (Suppl 5): 133 (Abstr 487PD).
16. Berardi R, Porfiri E, Massidda B et al. Gemcitabine (GEM) and cisplatin (PL) versus gemcitabine alone in stage IV non small cell lung cancer: Preliminary results of a randomized multicenter phase III study. Proc Am Soc Clin Oncol 2001; 20: 347a (Abstr 1385).
17. Jadad AR, Moore RA, Carroll D et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 112.[ISI][Medline]
18. Parmar MB, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival end points. Stat Med 1998; 17: 28152834.[CrossRef][ISI][Medline]
19. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 15391558.[CrossRef][ISI][Medline]
20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177188.[CrossRef][ISI][Medline]
21. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50: 10881101.[ISI][Medline]
22. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315: 629634.
23. Lilenbaum RC, Langenberg P, Dickersin K. Single agent versus combination chemotherapy in patients with advanced nonsmall cell lung carcinoma. A meta-analysis of response, toxicity, and survival. Cancer 1998; 82: 116126.[CrossRef][ISI][Medline]
24. Delbaldo C, Syz N, Michiels S, Le Chevalier T et al. Adding a second or a third drug to a chemotherapy regimen in patients with advanced non-small-cell lung carcinoma (NSCLC): A meta-analysis of the literature. Proc Am Soc Clin Oncol 2003; 22: 623 (Abstr 2507).
25. Baggstrom MQ, Socinski MA, Hensing TA, Poole C. Addressing the optimal number of cytotoxic agents in stage IIIB/IV non-small cell lung cancer (NSCLC): A meta-analysis of the published literature. Proc Am Soc Clin Oncol 2003; 22: 624 (Abstr 2510).
26. Hotta K, Matsuo M, Ueoka H et al. Meta-analysis of randomized clinical trials comparing cisplatin to carboplatin in patients with advanced non-small-cell lung cancer. J Clin Oncol 2004; 22: 38523859.
27. Clarke MJ, Stewart LA. Obtaining individual patient data from randomized controlled trials. In Egger M, Smith GD, Altman DG (eds): Systematic Reviews in Health Care. London, UK: BMJ Publishing Group 2001; 109121.
28. Ten Bokkel HWW, Bergman B, Chemaissani A et al. Single-agent gemcitabine: an active and better tolerated alternative to standard cisplatin-based chemotherapy in locally advanced or metastatic non-small cell lung caner. Lung Cancer 1999; 26: 8594.[CrossRef][ISI][Medline]
29. Vansteenkiste JF, Vandebroek JE, Nackaerts KL et al. Clinicalbenefit response in advanced non-small-cell lung cancer: A multicentre prospective randomised phase III study of single agent gemcitabine versus cisplatin-vindesine. Ann Oncol 2001; 12: 11911193.[CrossRef][ISI][Medline]
30. Perng RP, Chen YM, Ming-Liu J et al. Gemcitabine versus the combination of cisplatin and etoposide in patients with inoperable non-small-cell lung cancer in a phase II randomized study. J Clin Oncol 1997; 15: 20972102.[Abstract]
31. Takenaka M, Tanio Y, Okishio K et al. A randomized phase II of weekly carboplatin and CPT-11 versus docetaxel for relapsed non-small cell lung cancer. Proc Am Soc Clin Oncol 2001; 20: 263b (Abstr 2805).
32. Splinter TAW, Sahmoud T, Festen J et al. Two schedules of teniposide with or without cisplatin in advanced non-small cell lung cancer: randomized study of the European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group. J Clin Oncol 1996; 14: 127134.[Abstract]