© 2007 European Society for Medical Oncology
lung cancer |
The impact of induction chemotherapy on the outcome of second-line therapy with pemetrexed or docetaxel in patients with advanced non-small-cell lung cancer
1 Division of Medical Oncology, University of Colorado Health Sciences Center, Aurora, CO, USA
2 Catalan Institute of Oncology, Barcelona, Spain
3 MD Anderson Cancer Center, Houston, TX
4 University of Missouri, Ellis Fischel Cancer Center, Columbia, MO, USA
5 University Hospital Division of Medical Oncology, Zurich, Switzerland
6 Centro Hospitalar de Coimbra, Coimbra, Portugal
7 Eli Lilly and Company, Indianapolis, IN
8 Indiana University Medical Center, Indianapolis, IN, USA
* Correspondence to: Dr P. A. Bunn Jr, Division of Medical Oncology, University of Colorado Health Sciences Center, Aurora, CO, USA. Tel: +1 303-724-3152; Fax: +1 303-724-3162; E-mail: paul.bunn{at}uchsc.edu
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Background: Using data from a large phase III study of previously treated advanced non-small-cell lung cancer (NSCLC) that showed similar efficacy for pemetrexed and docetaxel, this retrospective analysis evaluates the impact of first-line chemotherapy on the outcome of second-line chemotherapy.
Patients and methods: In all, 571 patients with advanced NSCLC were randomly assigned to receive pemetrexed 500 mg/m2 or docetaxel 75 mg/m2 on day 1 of a 21-day cycle. Comparisons were made based on type of first-line therapy [gemcitabine + platinum (GP), taxane + platinum (TP), or other therapies (OT)], response to initial therapy, time since initial therapy, and clinical characteristics. The two second-line treatment groups were pooled for this analysis due to similar efficacy and were assumed to have no interaction with the first-line therapies.
Results: Baseline characteristics were generally balanced. By multivariate analysis, gender, stage at diagnosis, performance status (PS), and best response to first-line therapy significantly influenced overall survival (OS). Additional factors by univariate analysis, histology, and time elapsed from first- to second-line therapy significantly influenced OS.
Conclusions: Future trials in the second-line setting should stratify patients by gender, stage at diagnosis, PS, and best response to first-line therapy.
Key words: chemotherapy, docetaxel, non-small-cell lung cancer, pemetrexed, survival
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Current American Society of Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) guidelines recommend treatment of advanced non-small-cell lung cancer (NSCLC) with platinum-based doublet combination chemotherapy in the first-line setting and with nonplatinum-based doublets as a reasonable alternative [1, 2]. These recommendations are based on multiple randomized clinical trials comparing various platinum-based doublets and nonplatinum-based doublets including trials of the Southwest Oncology Group and Eastern Cooperative Oncology Group (ECOG) [3–9].
Preliminary results of a randomized phase II/III trial, ECOG 4599 were reported at ASCO 2005, evaluating the role of bevacizumab (B) in addition to carboplatin and paclitaxel (PC) in selected NSCLC patients. The addition of B to PC is significantly superior in response (27% versus 10%, P < 0.0001), progression-free survival (6.4 months versus 4.5 months, P < 0.0001), and median survival (12.5 months versus 10.2 months, P = 0.0075) compared with the PC arm alone [10].
ASCO and NCCN guidelines recommend docetaxel, pemetrexed, or erlotinib in the second-line setting. These recommendations were based on randomized clinical trials comparing these agents with best supportive care or placebo plus best supportive care, respectively, in two randomized trials. Shepherd et al. [11] compared docetaxel 75–100 mg/m2 with best supportive care in an international study of 204 patients. Median survival was 7.5 months versus 4.6 months that translated into a superior 1-year survival rate of 37% versus 12%, P = 0.003, favoring docetaxel 75 mg/m2 over the control arm. In another study of 373 patients, Fossella et al. [12] compared docetaxel 75 or 100 mg/m2 with the control group of either vinorelbine or ifosfamide. Subjects treated with docetaxel had significantly longer time to progression (TTP) versus the control group (P = 0.046). One-year survival was 32% for docetaxel 75 mg/m2 versus 19% for vinorelbine or ifosfamide (P = 0.025); however, the overall survival (OS) time was not significantly different.
A large randomized phase III comparison trial of pemetrexed to docetaxel showed similar survival with median survival times (MSTs) of 8.3 and 7.9 months, respectively [hazard ratio (HR), 0.99; 95% confidence interval (CI), 0.82 to 1.2] [13]. Objective and subjective response rates and TTP were also similar in the two arms. There were, however, significantly more grade 3/4 hematologic toxicity, febrile neutropenia, and drug-related hospitalizations on the docetaxel arm. On the basis of this trial and other trial data supporting pemetrexed results, the Food and Drug Administration (FDA) approved the use of pemetrexed in the second-line setting.
The BR.21 trial [14] was a large phase III trial of 731 patients who underwent 2 : 1 randomization to 150 mg oral erlotinib daily or placebo after failing one or two prior chemotherapy regimens. The erlotinib arm had a response rate of 8.9% versus <1% on the placebo arm (P < 0.001), and an OS rate of 6.7 months versus 4.7 months, respectively (HR 0.70; P < 0.001). On the basis of these data, docetaxel and pemetrexed are FDA-approved in the second-line setting, and erlotinib is approved in the second- or third-line setting.
On the basis of the aforementioned data, docetaxel and pemetrexed have received regulatory approval worldwide in the second-line setting, and erlotinib is approved for both the second- and third-line setting. With at least three agents approved for advanced NSCLC patients in the second-line setting and the development of new agents, it is probable that many randomized phase III trials will be comparing agents in the second-line setting. Optimal study design will require appropriate stratification on the basis of established prognostic factors. Factors that were prospectively stratified for in the randomized trial comparing pemetrexed with docetaxel in the second-line setting included the following: performance status (PS) at start of second-line therapy, prior platinum or paclitaxel use, number of prior chemotherapy regimens, time since last chemotherapy, best response to last chemotherapy, stage at diagnosis, baseline plasma homocysteine level, and center. In this retrospective analysis of the randomized trial, we evaluate age, gender, stage, PS at start of second-line therapy, type of first-line chemotherapy, response to first-line chemotherapy, and time elapsed from first- to second-line chemotherapy on survival outcome in the second-line setting [13]. Baseline values for lactate dehydrogenase, percent change in weight loss (<10%), and smoking history were not available.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The major efficacy and toxicity findings from the phase III trial comparing pemetrexed (Alimta, Eli Lilly and Company, Indianapolis, IN, USA) with docetaxel (Taxotere, Sanofi-Aventis, Bridgewater, NJ, USA) in the second-line setting were previously published [13]. Briefly, all patients had confirmed stage III or IV NSCLC; had received no more than one prior chemotherapy regimen for the treatment of advanced disease; had an ECOG PS of zero to two; were
18 years of age, with measurable or evaluable disease; and had adequate bone marrow, hepatic, and renal function. Exclusion criteria included the following: prior docetaxel or pemetrexed treatment, significant weight loss (10% body weight during the previous 6 weeks), grade 3 peripheral neuropathy, symptomatic or uncontrolled brain metastases, uncontrolled pleural effusions, or an inability to interrupt nonsteroidal anti-inflammatory drugs. All patients provided written informed consent before treatment. The protocol was approved by each institution's ethical review board. From March 2001 through February 2002, 571 patients were randomly assigned to receive either docetaxel 75 mg/m2 (n = 288) or pemetrexed 500 mg/m2 (n = 283) every 3 weeks. All 571 patients were assessable for survival and TTP analyses; 538 patients qualified for objective tumor response evaluation. Time elapsed since first-line therapy was available on 563 patients and PS at start of second-line therapy was available on 538 patients. Best response rate to first-line treatment was available on 545 patients and used the reported data collected at study entry and did not undergo independent radiologic confirmation.
statistical analysis
The retrospective statistical analysis compared the survival data for all patients combined and each of the three first-line treatment groups from the randomized, phase III, advanced NSCLC study based on first-line therapy received: gemcitabine + platinum (GP) (n = 182), taxane + platinum (TP) (n = 113), and other therapies (OT) (n = 276). The OT group consisted of vinorelbine + platinum (35%), etoposide + platinum (19%), gemcitabine + vinorelbine (11%), and other regimens (35%).
In the phase III study, the efficacy measures of pemetrexed and docetaxel were similar. There was no interaction between the first-line therapies and the second-line therapy. Therefore, the outcome data for these two treatment groups were pooled for this analysis. Unless otherwise stated, all tests of hypotheses were conducted at the 
= 0.05 level, with a 95% CI.
Cox proportional hazards model was used for the comparison of survival and TTP (data not shown) among the first-line therapies in baseline characteristics, subgroups when applicable (e.g. age and gender), and treatment and subgroup by treatment interaction. Both unadjusted and adjusted comparisons for the predictors were analyzed for survival. Kaplan–Meier estimates were used to assess medians and percentiles. Comparison of the tumor response rates among the first-line therapies was made using the Fisher's exact test with 95% CI calculated using the method of Leemis and Trivedi [15].
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Baseline characteristics are shown in Table 1. Median survival by baseline characteristics with univariate and multivariate analysis is shown in Table 2. Median survival by baseline characteristics and treatment group is shown in Table 3.
|
|
|
age
The median age was 58 (range 22–87) in the entire patient population and 57 (range 22–79), 58 (range 40–87), 59 (range 30–81) in the GP, TP, and OT groups, respectively (P = 0.177) (Table 1). Age was not a significant prognostic factor [16]. Patients who were elderly (
70 years old) had a median survival of 8.8 months compared with 7.9 months for those <70 years (P = 0.809) (Table 2).
gender
There were 411 (72%) males and 160 (28%) females in the entire group. In the GP, TP, and OT groups, there were 71%, 59%, 78% males, respectively (P < 0.001) (Table 1). TP had significantly more women than the other two groups. Gender was a significant prognostic factor as the median survival for females was 9.4 months versus 7.2 months for males (P = 0.001) (Table 2).
stage at diagnosis
There were 144 (25%) with stage III disease and 427 (75%) with stage IV disease at diagnosis for the entire population. In the GP, TP, and OT groups, there were 82%, 77%, and 69% with stage IV disease at diagnosis, respectively (P = 0.008) (Table 1). Stage at diagnosis was also a significant prognostic factor as the median survival for stage III disease at diagnosis was 9.5 months compared with 7.8 months for stage IV disease (P = 0.036) (Table 2).
histology
For the entire population, 296 (52%) had adenocarcinoma, 171 (30%) had squamous cell carcinoma, and 104 (18%) had other or mixed histology. In GP, TP, and OT groups, 53%, 63%, and 46% had adenocarcinomas, respectively (P = 0.047) (Table 1). TP had more adenocarcinoma histology than the other two groups. Histology was a significant prognostic factor as the median survival for patients with adenocarcinoma was 9.1 months, 6.5 months for squamous cell carcinoma, and 7.8 months for other or mixed histology (P = 0.004) (Table 2).
PS at start of second-line treatment
Of the 538 patients evaluated for PS, 100 (19%) had ECOG PS of zero, 374 (70%) had ECOG PS of one, and 64 (12%) had ECOG PS of two. [Unknown was not included in analyses (n = 33)]. For GP, TP, and OT, the percentages were similar (P = 0.849) (Table 1). PS was a significant prognostic factor. For all patients, median survival was 12.7 months for ECOG PS of zero, 8.3 months for ECOG PS of one, and 2.6 months for ECOG PS of two (P < 0.001) (Table 2). Figures 1A–D show OS based on PS for the entire population, GP, TP, and OT, respectively.
|
best overall response to first-line treatment
For the whole study population, best responses to first-line therapy were 16 (3%) complete response (CR), 190 (35%) partial response (PR), 199 (37%) stable disease (SD), and 140 (26%) progressive disease (PD). The objective response rate was 41% for GP, 35% for TP, and 37% for OT (P = 0.714) (Table 1). The disease control rate (CR/PR + SD) was 76% for GP, 76% for TP, and 72% for OT. Initial response to chemotherapy was a significant prognostic factor. For all patients, median survival was 15.8 months for CR/PR, 10.5 months for SD, and 4.6 months for PD (P < 0.001) (Table 2). Figures 2A–D show OS based on best response to first-line therapy for the entire population, GP, TP, and OT, respectively.
|
time elapsed from first- to second-line therapy
Patients were analyzed on the basis of time elapsed from first- to second-line therapy and grouped as
3 months, 3–6 months, and 6 months. Time elapse to initiation of second-line therapy for the study population was 277 (49%) 3 months, 117 (21%) 3–6 months, and 169 (30%) 6 months elapse to initiation of second-line therapy. Time since initial chemotherapy was a significant prognostic factor as median survival was 6.9 months, 9.2 months, and 9.3 months for 3 months, 3–6 months, and 6 months groups, respectively (P = 0.001) (Table 2). Figures 3A–D show OS based on time elapsed from first- to second-line therapy for the entire population, GP, TP, and OT, respectively.
|
first-line regimen
Of the 571 patients randomly assigned and treated in the second-line study, 182 patients received GP, 113 received TP, and 276 received OT for first-line treatment. Patients treated with first-line GP had a numerically superior MST (9.1 months) relative to TP (7.4 months) or OT (7.8 months), but these differences failed to reach statistical significance (P = 0.626) (Table 2 and Figure 4).
|
Thus, by univariate analyses the following characteristics significantly influenced OS as follows: gender, stage at diagnosis, histology, PS, best overall response to first-line therapy, and time elapsed from first- to second-line therapy (Table 2). By multivariate analysis, the following characteristics significantly influenced OS as follows: gender, stage at diagnosis, PS, and best overall response to first-line therapy (Table 2).
In Table 3, the analysis of the subgroups within the baseline characteristics by treatment (GP, TP, or OT) did not demonstrate significant interaction. For each subgroup, there were no significant median OS differences by treatment group.
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
In this randomized trial comparing pemetrexed with docetaxel in the second-line setting of advanced NSCLC, we found that the following variables were associated with survival outcome: gender, histology, stage at diagnosis, PS at start of second-line therapy, type of initial therapy, best response to initial therapy, and time since initial therapy. Of these, type of initial therapy was not statistically significant in univariate analysis, while the following factors were statistically significant by multivariate analyses: gender, stage at diagnosis, PS at start of second-line therapy, and best response to initial therapy. Thus, these four factors should be used as stratification factors in future randomized trials. It should be noted that interpretation of best response to first-line therapy by imaging was not confirmed in a standardized fashion. It is not surprising that status and gender have a significant impact on survival [17–18]. Appetite and weight loss have been reported as prognostic factors in the first-line setting [17], however, this data is not available for our analyses.
Other factors that should be documented when enrolling patients on trials are tumor histology and time elapsed from first- to second-line therapy. In the phase III study, the time elapse groups were divided into <3 or 3 months. In this analysis, the time elapse groups were further divided (3, 3–6, and 6 months). While there was significant survival improvement with longer time elapse by univariate analysis, this trend did not persist in the multivariate analysis. Smoking history in pack-years was not available in our data, but would be of interest when enrolling patients into trials beyond first-line therapy to see if this is an independent prognostic factor.
This retrospective analysis is the first to show the potential influence of first-line chemotherapy on the outcome of second-line cytotoxic chemotherapy. First-line therapy with GP has a numerically higher survival time compared with TP or OT. This outcome was similar regardless of best response to first-line treatment or length of time since discontinuing therapy.
To our knowledge, the effect of first-line chemotherapy on outcomes using targeted chemotherapy in advanced NSCLC has not been previously published. As more randomized trials are designed to answer treatment options in second-line therapy and beyond, conducting subset and retrospective analyses will help guide future study and hypotheses. Three additional areas that may be worth pursuing in data collection would be quality-of-life (QoL) improvement, and toxicity patients experienced within the first-line setting prior to enrollment in a second-line or beyond chemotherapy trial. Larger numbers of patients would have to be pooled to look for trends, but results of analysis might shed light on toxicity-prone individuals or profiles of patients most likely to benefit by QoL improvement despite poor prognostic factors or PS.
In conclusion, future trials in the second-line setting should stratify patients by gender, stage at diagnosis, PS at start of second-line therapy, and best response to first-line therapy. Advanced NSCLC patients have several first-line chemotherapy options. Selection of the most appropriate agents should be discussed with the treating physician.
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
We would like to thank Takashi Nakamura, Associate Senior Statistical Analyst at Eli Lilly, for his support with the statistical analyses and also Pete Fairfield, Scientific Communications Associate at Eli Lilly, for his editorial support. This study was supported in part by a grant from Eli Lilly and Company, Indianapolis, IN. This work was presented in part at the 11th World Conference on Lung Cancer, July 3–6, 2005, Barcelona, Spain. FF—Eli Lilly: research funding >$100 000; BN—Eli Lilly: employment, stock $10 000–100 000; SP—Eli Lilly: employment, stock $10 000–100 000; PM—Eli Lilly: employment, stock <$10 000; NH—Eli Lilly: consultant/advisory role <$10 000, honoraria $10 000–100 000, research funding >$100 000, and Sanofi-Aventis: consultant/advisory role <$10 000, honoraria <$10 000; KK—Eli Lilly: consultant/advisory role <$10 000, honoraria <$10 000, and Sanofi-Aventis: consulant/advisory role <$10 000; PAB—Eli Lilly: consultant/advisory role <$10 000, honoraria <$10 000, and Sanofi-Aventis: consulant/advisory role <$10 000, honoraria <$10 000.
Received for publication July 16, 2006. Revision received October 29, 2006. Accepted for publication October 31, 2006.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
1. Pfister DG, Johnson DH, Azzoli CG, et al. (2004) American Society of Clinical Oncology treatment of unresectable non-small-cell lung cancer guideline: update 2003. J Clin Oncol 22:330–353.
2. National Comprehensive Cancer Network. Non-small cell lung cancer: clinical practice guidelines in oncology. Version 2. 2006. http://nccn.org/professionals/physician_gls/PDF/nscl.PDF (13 July 2006, date last accessed).
3. Wozniak AJ, Crowley JJ, Balcerzak SP, et al. (1998) Randomized trial comparing cisplatin with cisplatin plus vinorelbine in the treatment of advanced non-small-cell lung cancer: a Southwest Oncology Group study. J Clin Oncol 16:2459–2465.[Abstract]
4. Bonomi P, Kim K, Fairclough D, et al. (2000) Comparison of survival and quality of life in advanced non-small-cell lung cancer patients treated with two dose levels of paclitaxel combined with cisplatin versus etoposide with cisplatin: results of an Eastern Cooperative Oncology Group trial. J Clin Oncol 18:623–631.
5. Cardenal F, Lopez-Carbrerizo MP, Anton A, et al. (1998) Randomized trial comparing cisplatin with cisplatin plus vinorelbine in the treatment of advanced non-small cell lung cancer: a Southwest Oncology Group Study. J Clin Oncol 16:2459–2465.[Abstract]
6. Kosmidis PA, Kalofonos C, Syrigos K, et al. (2005) Paclitaxel and gemcitabine vs carboplatin and gemcitabine. A multicenter, phase III randomized trial in patients with advanced inoperable non-small cell lung cancer (NSCLC). J Clin Oncol 23:16S (June 1 Supplement) 7000.
7. Pujol JL, Breton JL, Gervais R, et al. (2005) Gemcitabine-docetaxel versus cisplatin-vinorelbine in advanced or metastatic non-small-cell lung cancer: a phase III study addressing the case for cisplatin. Ann Oncol 16:602–610.
8. Schiller JH, Harrington D, Belani CP, et al. (2002) Comparison of four chemotherapy regimens for advanced non-small cell lung cancer. N Engl J Med 346:92–98.
9. Kelly K, Crowley J, Bunn PA Jr, et al. (2001) 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.
10. Sandler AB, Gray R, Brahmer J, et al. (2005) Randomized phase II/III trial of paclitaxel (P) plus carboplatin (C) with or without bevacizumab (NSC #704865) in patients with advanced non-squamous non-small cell lung cancer (NSCLC): an Eastern Cooperative Oncology Group (ECOG) Trial—E4599. J Clin Oncol 23:16S (June 1 Supplement) 2005: LBA4.
11. Shepherd FA, Dancey J, Ramlau R, et al. (2000) Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 18:2095–2103.
12. Fossella F, Pereira JR, von Pawel J, et al. (2003) Randomized, multinational, phase III study of docetaxel plus platinum combinations versus vinorelbine plus cisplatin for advanced non-small-cell lung cancer: the TAX 326 study group. J Clin Oncol 21:3016–3024.
13. Hanna N, Shepherd FA, Fossella FV, et al. (2004) Randomized phase III trial of pemetrexed versus docetaxel in patients with non-small-cell lung cancer previously treated with chemotherapy. J Clin Oncol 22:1589–1597.
14. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. (2005) Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 353:123–132.
15. Leemis LM and Trivedi KS. (1996) A comparison of approximate interval estimators for the Bernoulli parameter. Am Stat 50:63–68.[CrossRef]
16. Weiss GJ, Langer C, Rosell R, et al. (2006) Elderly patients benefit from second-line cytotoxic chemotherapy: a subset analysis of a randomized phase III trial of pemetrexed compared with docetaxel in patients with previously treated advanced non-small-cell lung cancer. J Clin Oncol 24:4405–4411.
17. Jiroutek M, Johnson D, Blum R, et al. (1998) Prognostic factors in advanced NSCLC: analysis of ECOG trials from 1981–1992. Proc Am Soc Clin Oncol 17:1774.
18. Neugut AI and Jacobson JS. (2006) Women and Lung Cancer. Gender Equality at a Crossroad? JAMA 296:218–219.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. H. Cullen, P. Zatloukal, S. Sorenson, S. Novello, J. R. Fischer, A. A. Joy, M. Zereu, P. Peterson, C. M. Visseren-Grul, and N. Iscoe A randomized phase III trial comparing standard and high-dose pemetrexed as second-line treatment in patients with locally advanced or metastatic non-small-cell lung cancer Ann. Onc., May 1, 2008; 19(5): 939 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. I. Na, B. H. Byun, H. J. Kang, G. J. Cheon, J. S. Koh, C. H. Kim, D. H. Choe, B.-Y. Ryoo, J. C. Lee, S. M. Lim, et al. 18F-Fluoro-2-Deoxy-Glucose Uptake Predicts Clinical Outcome in Patients with Gefitinib-Treated Non-Small Cell Lung Cancer Clin. Cancer Res., April 1, 2008; 14(7): 2036 - 2041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Stinchcombe and M. A. Socinski Considerations for Second-Line Therapy of Non-Small Cell Lung Cancer Oncologist, January 1, 2008; 13(suppl_1): 28 - 36. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||