Prospective randomised phase II study of docetaxel versus paclitaxel administered weekly in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy

E. Esteban+, L. González de Sande, Y. Fernández, N. Corral, J Fra, I. Muñiz, J. M. Vieitez, I. Palacio, J. L. Fernández, E. Estrada and A. J. Lacave

Servicio de Oncología Médica, Hospital Central de Asturias, Oviedo, Asturias, Spain

Received 9 February 2003; revised 19 June 2003; accepted 8 August 2003


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

Docetaxel and paclitaxel have activity in the second-line treatment of non-small-cell lung cancer (NSCLC), and can be administered as weekly schedules. This phase II randomised study was designed to test the efficacy and toxicity of both taxanes in patients with NSCLC previously treated with platinum-based chemotherapy.

Patients and methods:

Patients (n = 71) with documented NSCLC were randomised to receive docetaxel (n = 35 patients; 36 mg/m2) or paclitaxel (n = 36 patients; 80 mg/m2) as a 1 h weekly infusion for 6 weeks followed by a 2-week rest. The cycles were repeated until disease progression or non-acceptable toxicities occurred.

Results:

Treatment achieved partial response of one versus five patients, median time-to-progression of 74 versus 68 days, and overall survival of 184 versus 105 days, with docetaxel and paclitaxel, respectively. The most common non-haematological toxicities were (docetaxel versus paclitaxel): grade 3/4 pulmonary toxicity in seven versus one patient; grade 2/3 diarrhoea in nine versus five; and grade 3/4 haematological toxicities occurred in two versus four patients. There were no treatment-related deaths.

Conclusions:

Docetaxel and paclitaxel administered weekly have discrete efficacy in patients with NSCLC previously treated with platinum-based chemotherapy. The higher non-haematological toxicity of docetaxel, particularly pulmonary toxicity and diarrhoea, is of concern and warrants further investigation.

Key words: docetaxel, NSCLC, paclitaxel, second-line therapy, weekly administration


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Treatment of patients diagnosed as having metastatic or locally advanced non-small-cell lung cancer (NSCLC) requires the use of chemotherapy combinations that, on the results of comparative studies and meta-analyses [1, 2], include platinum and its derivatives. However, the usefulness of second-line chemotherapy in NSCLC following initial treatment failure with platinum-based regimens has not been definitively established. Most studies undertaken to-date have been phase II trials, and in most of these the overall response rate has been <15% with the median survival and 1-year survival seldom being reported [3]. One possible exception has been the data generated using the taxanes as second-line treatment in patients previously treated with platinum or its combinations. As such, not only docetaxel but also paclitaxel have been demonstrated, in non-comparative studies, to have therapeutic efficacy of ~20% objective response, with median and 1-year survival of >7 months and 30%, respectively. In schemes involving 3-weekly administration schedules, the haematological toxicities have been of concern [4, 5]. Phase I/II trials have shown that not only paclitaxel, but also docetaxel, can be used in a weekly schedule, which appears to have a reduced toxicity while preserving therapeutic efficacy. The doses recommended for subsequent phase II trials were 80 mg/m2 for paclitaxel [68] and 36 mg/m2 for docetaxel [9].

The studies that explored the individual use of paclitaxel and of docetaxel prompted us to design this present randomised phase II trial using both agents singly as weekly administrations in order to test directly their efficacy and toxicities in patients with NSCLC who had been previously treated with platinum-based chemotherapy.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Eligibility criteria
For this multicentred, open-label, randomised trial, the inclusion criteria were histologically confirmed advanced NSCLC previously treated with a platinum-containing chemotherapy (cisplatin or carboplatin) but naïve to taxanes (and no more than one or two previous chemotherapy regimens). The patients had to be aged >18 years, have a Karnofsky performance status >=60%, life expectancy of at least 12 weeks, measurable or evaluable disease, adequate bone marrow reserve (neutrophil count >=2 x 109/l, platelet count >=100 x 109/l, haemoglobin >=10 g/dl, haematocrit >=30%). Adequate liver function [bilirubin <=1.5 mg/dl, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) below 1.5x the upper limit of normal (ULN), but including patients with liver metastases] and renal function (creatinine <=1.4 mg/dl) were also required for inclusion. Patients needed to be resident within the hospital’s geographical remit so as to ensure compliance with the study protocol. Patients who had received prior radiation therapy or chemotherapy were eligible provided that at least 21 days had elapsed since completion of the previous treatment. Patients with brain metastases that had been treated were eligible for entry provided that they were neurologically stable. Patients were excluded on the grounds of severe uncontrolled co-morbidities and second malignancies, or peripheral neuropathy grade >1 on the World Health Organization (WHO) definition. The study was conducted according to the principles stated in the latest version of the Declaration of Helsinki. Signed informed consent was obtained from all patients prior to recruitment into the study.

Baseline assessments and treatment plan
Baseline evaluations included medical history, physical examination, complete blood count, liver function chemistries and creatinine, electrocardiogram, and radiological staging [chest X-ray or computed tomography (CT) scan of the thorax, mediastinum and abdomen, or echography]. Patients were stratified according to Karnofsky performance status (60% versus >=70%) and then randomised to one of two treatment arms: docetaxel 36 mg/m2 or paclitaxel 80 mg/m2 administered as a 1-h i.v. infusion weekly for 6 weeks followed by a 2-week rest. Patients were prescribed prophylactic i.v. ranitidine (50 mg), diphenhydramine (25 mg) and dexamethasone (8 mg) prior to the taxane administration. The treatment cycles were repeated until disease progression, or unacceptable toxicity occurred.

A complete treatment cycle was defined as the administration of six consecutive weeks of the drugs under study. Prior to commencement of each treatment cycle, a complete blood count and biochemistry measurements of renal and liver function were performed. Haematology blood count was repeated before the administration of each dose of the treatment. Docetaxel or paclitaxel dosage was modified according to the blood count recorded on the day of treatment. The decision to continue with treatment administration was taken only in the case of neutrophil count >=1.5 x 109/l and platelet count >=100 x 109/l. In the event of these levels not being reached, treatment was delayed for 1 week until the blood counts recovered, and the dose was also reduced by 20%. This same dose reduction was performed in the event of there occurring, at some time during the study, neutropenia grade 4 or other non-haematological toxicity >2 (except for alopecia or nausea/vomiting). In all cases, the dose reduction was continued for subsequent cycles. Those patients with febrile neutropenia, or grade 4 non-haematological toxicity (except for alopecia or nausea/vomiting) or requiring a delay in treatment beyond 3 weeks, were withdrawn from the study. Those patients in whom evaluation by physical exploration or chest X-ray was possible had these evaluations repeated before each cycle of treatment. Other methods of evaluation, such as CT scan or echography, were repeated every two cycles of treatment, or if disease progression was suspected.

The criteria followed for the evaluation of toxicity and therapeutic efficacy were those defined by the WHO [10]. A patient was considered evaluable for toxicity if one treatment dose had been received and a patient was evaluable for response if at least two successive doses of the treatment had been administered. The intention-to-treat population for determining response rate included patients who had received at least one dose of the chemotherapy under investigation. Only the most severe grade of toxicity was recorded for each patient. Time-to-disease progression and survival rates were calculated from the date of randomisation up to the time of disease progression and until the date of death, respectively.

Statistical methods
The primary end points of the study were activity and toxicity. The secondary end point of the study was survival. With respect to response, this randomised phase II trial was treated, statistically, as two simultaneous phase II studies and the Simon two-stage design was applied separately for each treatment arm [11]. For a total of 36 subjects with NSCLC previously treated with platinum combinations, 16 had to be recruited during stage 1, and 20 recruited during stage 2. Each treatment arm was considered ‘non-informative’ if the percentage response was <=1% (0.01). Hence, the null hypothesis for comparison is H0 = P (response) <=0.01 against H1 = P (response) >0.01. The significance level of the design was set at 0.05 and the power of the contrast, when H1 = P (response) = 0.10, was 0.8962.

If less than one response was observed during the first stage, then the trial would be stopped early. Should there be three or fewer responses recorded by the end of the trial, no further investigation of the drug would be warranted. The response rate was estimated on all evaluable patients in each of the two treatment arms, and a corresponding 95% confidence interval (CI) for the response rate was calculated. If both schedules were between the pre-established ranges of response rate, the less toxic regimen would be chosen.

For evaluation of toxicity, the two groups were compared using Fisher’s exact text [12]. Continuous data were compared using Student’s t-test [13] when distributions were normal or using Kruskal–Wallis test when they were not [14]. Time-to-progression and survival were estimated by the Kaplan–Meier product limit method [15]. The log-rank, Tarone–Ware and Breslow tests were used for comparing mortality in the two treatment groups and to check the effect of censored data [16].


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patient characteristics
Between July 1999 and May 2002, a total of 71 patients were enrolled in this phase II prospective randomised study. The characteristics of the patients on entry into the trial are listed in Table 1. The two treatment groups had randomly assigned to them 36 patients to receive paclitaxel and 35 to receive docetaxel; the groups were well balanced except for a tendency towards a younger age and better performance status in the patients assigned to docetaxel treatment arm. The predominant histology was adenocarcinoma in both groups with a tendency towards a higher frequency in the paclitaxel group. Both groups were well balanced with respect to previous chemotherapy, as well as to key parameters such as type of previous regimens received and progressive disease status with previous platinum therapy.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics on entry into the trial
 
Two patients in the docetaxel treatment arm received only one dose and so were considered evaluable for toxicity but not for response. In the paclitaxel group, one patient received treatment with a schedule of 175 mg/m2 every 3 weeks and another received concomitant treatment with radiotherapy, and both were excluded from the safety and response analysis.

Treatment administration
After a total of 260 weeks (range 1–18 per patient) of treatment administered in the docetaxel group and 330 weeks (range 1–29 per patient) in the paclitaxel group, the median number of weeks of chemotherapy administered in either group was six. After 42 cycles of treatment in the docetaxel group (range 0–5 per patient) and 45 cycles (range 0–6 per patient) in the paclitaxel group, the median number of cycles of chemotherapy administered in either group was one. The median total dose was 60 mg/week (range 44–72) in the docetaxel group and 136 mg/week (range 90–175) in the paclitaxel group of patients. Of the patients in the docetaxel group, 16 (46%) required dose reduction or delay in treatment due to toxicity, as did nine (26%) in the paclitaxel group of patients.

Efficacy
The overall efficacy was based on an intention-to-treat analysis and included patients with measurable and as well as evaluable lesions who had received at least one chemotherapy infusion. Only patients who received at least two consecutive doses of treatment were considered evaluable for response (Table 2). The objective response rate (partial response) registered in this study was 3% (95% CI 0% to 6%; one of 35 eligible patients) in the docetaxel group and 14% (95% CI 9.6% to 18.4%; five of 36 eligible patients) in the paclitaxel group. In 65% (23 patients) of the docetaxel group and 47% (17 patients) of the paclitaxel group, stable disease status was achieved. In 26% (nine patients) and 33% (12 patients) in the docetaxel and paclitaxel treatment arms, respectively, there was disease progression (P = 0.19, Fisher’s exact test). In each of the treatment arms there were two patients who were not considered evaluable for activity assessment. In the docetaxel group, one patient had received only one dose of the treatment and the other was due to premature death. In the paclitaxel group, one was due to incorrect dose and another because of concomitant chest radiotherapy. Nevertheless, all of these patients were included in the denominator in the response evaluation.


View this table:
[in this window]
[in a new window]
 
Table 2. Summary of response, time-to-progression and survival
 
The median time-to-progression was similar in both treatment groups; 74 days (95% CI 51–96) for docetaxel and 68 days (95% CI 56–79) for paclitaxel (P = 0.17). The median duration of survival was 184 days (95% CI 89–278) and 105 days (95% CI 70–155) for the docetaxel and the paclitaxel treatment groups, respectively (P = 0.47). The 1-year survival rate was 6% in both groups. There were 15 patients (47%) in the docetaxel group and seven (22%) in paclitaxel group who received another line of chemotherapy following progression of the disease, or who were withdrawn from the present study (Table 1). Data on response rate and features of all patients with objective response are presented in Tables 2 and 3. Survival curves of time-to-progression and overall survival are depicted in Figures 1 and 2.


View this table:
[in this window]
[in a new window]
 
Table 3. Features of the patients who responded to treatment with paclitaxel or docetaxel
 


View larger version (15K):
[in this window]
[in a new window]
 
Figure 1. Kaplan–Meier curves depicting time-to-progression in the paclitaxel (triangles) and the docetaxel (circles) treatment arms (P = 0.17).

 


View larger version (15K):
[in this window]
[in a new window]
 
Figure 2. Kaplan–Meier curves depicting overall survival in the paclitaxel (triangles) and docetaxel (circles) treatment arms (P = 0.47).

 
Toxicity
Toxicities and safety of the schemes were assessed in 35 and 34 patients who had received at least one infusion of weekly docetaxel and paclitaxel, respectively. Two patients in the paclitaxel treatment arm were excluded from this evaluation; one due to having received treatment with a schedule of 175 mg/m2 every 3 weeks and the other due to having received concomitant chest radiotherapy. Haematological and non-haematological toxicities are summarised in Table 4. Grade 3 or 4 haematological toxicity was rarely seen in either treatment group. Anaemia and neutropenia grade 3 was observed in three (8%) and one (3%) patients, respectively, in the paclitaxel group, while there was one episode (3%) of grade 4 neutropenia and one of grade 3 thrombocytopenia noted in the docetaxel group.


View this table:
[in this window]
[in a new window]
 
Table 4. Summary of haematological and non-haematological toxicities in the two treatment arms
 
As with published data in other phase II trials using these agents with the weekly administration scheme, the most frequently observed toxicities in the docetaxel group as well as the paclitaxel group were non-haematological toxicities (Table 4). However, the most notable toxicity observed in our study, in contrast to other studies, was grade 3–4 pulmonary toxicity, seen in seven patients (20%) treated with docetaxel and in one patient (3%) treated with paclitaxel. The pulmonary toxicity involved bilateral interstitial pulmonary infiltrates with progressive moderate-to-intense fatigue, and resting hypoxia in the absence of fever. This clinical condition failed to improve with antibiotics. In one patient a bronchoscopy was performed and a transbronchial biopsy showed atypical squamous cells, foci of alveolar damage, interstitial inflammation and fibrosis with some obliterating bronchiolitis, suggestive of tissue damage by drugs. This toxicity appeared at a median of 6 weeks (range 2–12) following the commencement of the treatment. A common factor among all these patients was that they had received prior treatment with gemcitabine. Of additional note are the episodes of grade 2/3 diarrhoea in nine patients (26%) in the docetaxel group and grade 2 diarrhoea in five patients (14%) in the paclitaxel group. The pulmonary damage and the diarrhoea were considered as reversible phenomena, since resolution was achieved by suspending the treatment and, in the case of the pulmonary toxicity, by the prescription of additional corticoids.

Other non-haematological toxicities were less frequent in both treatment arms, and were grade <3, except for asthenia, allergic reaction, neuropathy, alopecia and stomatitis, for which at least one grade 3 episode was registered in each of the treatment arms. For example, two patients (6%) in the docetaxel treatment arm developed skin toxicities (Table 4) and two patients (6%) in the paclitaxel group acquired respiratory infection with fever grade 2 without neutropenia, but with a clinical course that responded to treatment with antibiotics and did not require hospital admission.

Overall, there was a tendency towards higher non-haematological toxicities in the docetaxel group. This necessitated a higher percentage of dose reductions or treatment deferment in the docetaxel group (46% of the cases) compared with the paclitaxel group (26% of cases). Furthermore, these toxicities were responsible for interruption of treatment in five patients (14%) in the docetaxel group compared with one patient (3%) in the paclitaxel group, despite there being no disease progression in these patients.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Until recently, the majority of clinical oncology specialists had no clear second-line treatment for the patients with NSCLC following the failure of treatment with platinum or its derivatives. Recently, in the comparative studies of Shepherd et al. [17] and Fossella et al. [18], docetaxel as a single agent was shown to have a small but significant advantage compared with best support treatment [17] or with vinorelbine/ifosfamide [18]. The major inconvenience, however, was the considerable haematological toxicity observed with docetaxel when administered in the 3-weekly scheme and at doses of 100–175 mg/m2.

Previous phase I and II trials had shown that docetaxel and paclitaxel can be administered using a weekly infusion scheme, which had a better haematological toxicity profile while appearing to preserve the therapeutic efficacy and, as such, was a viable alternative to the 3-weekly administration schedule [69].

Hence, the aim of the present randomised phase II trial was to evaluate the efficacy and toxicity profiles of docetaxel and paclitaxel administered weekly in non-selected patients who were refractory to previous platinum therapy, or its combinations.

This study was designed to detect a percentage of objective response of at least 10%. This was achieved with paclitaxel but not with docetaxel; 3% activity was registered in the latter. These results are in accord with the activity published in non-comparative studies of weekly paclitaxel [6] as well as the comparative study by Shepherd et al. [17] using docetaxel in a 3-weekly administration scheme in patients previously treated with platinum. The apparent advantage in objective response of paclitaxel over docetaxel achieved in the present study does not, however, translate into equivalence in the median time-to-progression, i.e. there were no differences between the two treatment groups with respect to this parameter of efficacy. Because of a possible antiangiogenic mechanism of action of taxanes used in a weekly schedule [1921], one can hypothesise that the percentage response to the drug under investigation is not as important as the possible delay in tumour size increase that the drug may induce. This could be of particular importance in patients with disease stages as advanced as those included in the present study. Surprisingly, there appeared to be an advantage with respect to median survival in favour of the patients treated with docetaxel. Although not part of the remit of the present study, one possible explanation could be that a greater number of patients treated with docetaxel had been transferred out of the study due to toxicity and proceeded to receive other lines of treatment such as irinotecan.

The majority of the clinical and demographic characteristics of the patients entered into the present study are similar in both treatment groups and, as such, do not appear to be responsible for differences in the results obtained (Table 1). It needs to be highlighted that all of the patients who achieved an objective response had a histology diagnosis of adenocarcinoma and, furthermore, all except one had disease progression with the previous platinum therapy. The fact that the paclitaxel treatment group had a tendency towards a greater number of patients with adenocarcinoma could also have had an influence on the number of objective responses observed.

With respect to toxicity, attention needs to be drawn to the pattern and distribution of the toxicities observed. Following a median of 6 weeks of the weekly schedule of treatment in both groups, it was of note that haematological toxicities were reduced compared with other administration schedules. Although this was slightly higher in the paclitaxel treatment group (especially in the red blood cells), a possible influence of the previous platinum therapy’s toxicity can be discarded. Conversely, the non-haematological toxicities, although not causing any toxic deaths, were significantly higher in the docetaxel treatment arm. This is seen clearly in the higher percentage of patients with mucosal toxicities (diarrhoea/stomatitis) and, especially, the pulmonary toxicity observed in the docetaxel group of patients (Table 4). In one of these patients there was histological confirmation of the tissue damage by the drug. However, in all of these patients the radiological findings and the clinical symptoms were completely reversible when the medication was suspended and corticoid therapy was implemented. It is of concern that this pulmonary effect had not been described as a dose-limiting toxicity in phase I trials with weekly docetaxel [9], although it had been noted with weekly paclitaxel at a dose of 175 mg/m2 [6] and also in the 3-weekly scheme [22, 23]. The only factor in common in the patients in our study who developed this toxicity was that all had been treated previously with gemcitabine. This is an agent that is known to be related to the development of pulmonary toxicity in some patients [24] and, as such, could be related in some way to the phenomenon observed in the docetaxel treatment group. It is possible that there is a relationship between docetaxel and gemcitabine, since lung toxicity has been described in other studies in which the two drugs were used in combination [25]. Also, there could have been an association between this toxicity and the pre-medication used in the present study. Dexamethasone (8 mg) was administered just before each infusion of docetaxel as well as paclitaxel, but in the experience of other authors [26] this dose appears to be appropriate and sufficient as a pre-medication for a weekly scheme of docetaxel not exceeding 36 mg/m2. More recently, there have been several studies comparing docetaxel administered weekly or 3-weekly [27, 28] either as a phase II study with paclitaxel [29, 30] or docetaxel alone [31] in patients previously treated with platinum. Although the toxicities reported (albeit from preliminary data) are concordant with our observations, none of these studies make any reference to lung toxicity.

The present study did not achieve the high levels of activities observed by some of the trials reported in the literature. However, other trials, such as those of Gervais et al. [28] and Canfield et al. [31] coincide with our study in describing a low percentage of objective response with docetaxel. Of note is that some other preliminary studies with docetaxel [27, 28] indicate that the activity is similar whether administered as a weekly or as a 3-weekly schedule, and that the only differences are in relation to the toxicity profile.

In summary, in the present study, conducted in patients with NSCLC previously treated with platinum, not only docetaxel but also paclitaxel demonstrated discrete therapeutic benefits without any clear advantage between the agents administered in the weekly schedule. The principal difference encountered was a higher non-haematological toxicity with docetaxel, which needs to be confirmed by other studies. Taking into account all the uncertainties described above, there need to be more studies designed to investigate whether there are real differences between docetaxel and paclitaxel with respect to efficacy and toxicity and the manner of their administration.


    Footnotes
 
+ Correspondence to: Dr E. Esteban-González, Servicio de Oncología Médica, Hospital Central de Asturias, Julián Clavería s/n, 33006 Oviedo, Asturias, Spain. Tel: +34-9851-06121; Fax: +34-9852-34498; E-mail: eestebang{at}seom.org Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. 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 randomised clinical trials. BMJ 1995; 311: 899–909.[Abstract/Free Full Text]

2. Gandara DR, Crowley J, Livingston RB et al. Evaluation of cisplatin intensity in metastatic non-small cell lung cancer: a phase III study of the Southwest Oncology Group. J Clin Oncol 1993; 11: 873–878.[Abstract]

3. Fosella FV, Lee JS, Hong WK. Management strategies for recurrent non-small cell lung cancer. Semin Oncol 1997; 24: 455–462.[ISI][Medline]

4. Hainsworth JD, Thompson SD, Greco FA. Paclitaxel by 1-hour infusion: an active drug in metastatic non-small cell lung cancer. J Clin Oncol 1995; 13: 1609–1614.[Abstract]

5. Fosella FV, Lee JS, Shin DM et al. Phase II study of docetaxel for advanced or metastatic platinum-refractory non-small cell lung cancer. J Clin Oncol 1995; 13: 645–651.[Abstract]

6. Akerley W, Glantz M, Choy H et al. Phase I trial of weekly paclitaxel in advanced lung cancer. J Clin Oncol 1998; 16: 153–158.[Abstract]

7. Fennelly D, Aghajanian C, Shapiro F et al. Phase I and pharmacologic study of paclitaxel administed weekly in patients with relapse ovarian cancer. J Clin Oncol 1997; 15: 187–192.[Abstract]

8. Seidman AD, Murphy B, Hudis C et al. Activity of taxol by weekly 1 hour infusion in patients with metastatic breast cancer: a phase II and pharmacologic study. Proc Am Soc Clin Oncol 1997; 16: 148a (Abstr 517).

9. Hainsworth D, Burris HA, Erland JB et al. Phase I trial of docetaxel administered by weekly infusion in patients with advanced refractory cancer. J Clin Oncol 1998; 16: 2164–2168.[Abstract]

10. World Health Organization. Handbook for Reporting Results of Cancer Treatment. WHO offset publication no. 48. Geneva: WHO 1979.

11. Simon R. Optimal two-stage designs for phase II clinical trials. Control Clin Trials 1989; 10: 1–10.[ISI][Medline]

12. Mehta CR, Patel NR. A network algorithm for performing Fisher’s exact test in r x c contingency tables. J Am Stat Assoc 1983; 78: 427–434.[ISI]

13. Winer BJ. Statistical Principles in Experimental Design. London: McGraw-Hill 1970.

14. Kruskal-Wallis. In Gibbons JD, Chakraborti, S (eds): Nonparametric Statistical Inference. New York, NY: Marcel Dekker 1992.

15. Kaplan E, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

16. Peto R, Pike MC. Conservation of the approximation {sum} (O – E)2/E in the log rank test for survival data or tumor incidence data. Biometrics 1973; 29: 579–584.[ISI][Medline]

17. Shepherd FA, Dancey J, Ramlau R et al. Prospective randomised trial of docetaxel versus best supportive care in patients with non-small cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 2000; 18: 2095–2103.[Abstract/Free Full Text]

18. Fossella FV, De Vore 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. The TAX 320 Non-Small Cell Lung Cancer Study Group. J Clin Oncol 2000; 18: 2354–2362.[Abstract/Free Full Text]

19. Vacca A, Ribatti D, Iurlaro M et al. Docetaxel versus paclitaxel for antiangiogenesis. J Hematother Stem Cell Res 2002; 11: 103–118.[CrossRef][ISI][Medline]

20. Lau DH, Xue L, Young LJ et al. Paclitaxel (Taxol): an inhibitor of angiogenesis in a highly vascularized transgenic breast cancer. Cancer Biother Radiopharm 1999; 14: 31–36.[ISI][Medline]

21. Sweeney CJ, Miller KD, Sissons SE et al. The antiangiogenic property of docetaxel is synergistic with a recombinant humanized monoclonal antibody against vascular endothelial growth factor or 2-methoxyestradiol but antagonized by endothelial growth factors. Cancer Res 2001; 61: 3369–3372.[Abstract/Free Full Text]

22. Golberg HL, Vanice SB. Pneumonitis related to treatment with paclitaxel. J Clin Oncol 1995; 13: 534–535.[ISI][Medline]

23. Ramanathan RK, Reddy VV, Holbert JM. Pulmonary infiltrates following administration of paclitaxel. Chest 1996; 110: 289–291.[Abstract/Free Full Text]

24. van der Els N, Millar V. Successful treatment of gemcitabine toxicity with a brief course of oral corticosteroid therapy. Chest 1998; 114: 1779–1781.[Abstract/Free Full Text]

25. Hosoe S, Komuta K, Shibata K et al. Gemcitabine and vinorebine followed by docetaxel in patients with advanced non-small cell lung cancer: final results of multi-institutional phase II trial of sequential non-platinum triplet combination chemotherapy (JMTO LCOO-02). Proc Am Soc Clin Oncol 2002; 21: 315a (Abstr 1259).

26. Stemmler J, Mair W, Stauch M et al. Weekly docetaxel with or without corticosteroid premedication as first or second-line treatment in patients with metastatic breast cancer. Proc Am Soc Clin Oncol 2002; 21: 58a (Abstr 231).

27. Shütte W, Ángel S, Lautenschläger C, Serke M. Randomized phase II study weekly versus three-weekly docetaxel as second-line chemotherapy for advanced non-small cell lung cancer. Proc Am Soc Clin Oncol 2002; 21: 308a (Abstr 1228).

28. Gervais R, Ducolone AMD, Breton JL et al. Multicenter, randomised phase II trial of docetaxel 75 mg/m2 q3w versus 40 mg/m2 weekly in patients with pretreated non small cell lung cancer. Proc Am Soc Clin Oncol 2002; 21: 310a (Abstr 1238).

29. Juan Vidal O, Albert A, Campos JM et al. Low-dose weekly paclitaxel as second-line treatment for advanced non-small cell lung cancer: a phase II study. Proc Am Soc Clin Oncol 2002; 21: 320a (Abstr 1277).

30. Suzuki R, Taniguchi H, Kondoh Y et al. Second-line weekly paclitaxel in resistant or relapsed non-small cell lung cancer treated with docetaxel and carboplatin. Proc Am Soc Clin Oncol 2002; 21: 325a (Abstr 1299).

31. Canfield VA, Garfield DH, Gregurich MA et al. Phase II trial of single-agent, weekly docetaxel as second-line therapy in patients with advanced non-small cell lung cancer. Proc Am Soc Clin Oncol 2002; 21: 239b (Abstr 2775).