This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bonomi, P. Articles by Johnson, D. Search for Related Content PubMed PubMed Citation Articles by Bonomi, P. Articles by Johnson, D.

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

From the Rush-Presbyterian St. Luke’s Medical Center, Chicago; Illinois Oncology Research Association, Peoria; and Evanston Northwestern Health Care, Northwestern University, Evanston, IL; University of Wisconsin Comprehensive Cancer Center, Madison, WI; American Medical Center Cancer Research Center, Denver, CO; Cancer Therapy and Research Center, San Antonio, TX; Dana-Farber Cancer Center, Boston, MA; and Vanderbilt University, Nashville, TN.

Address reprint requests to Philip Bonomi, MD, Rush Medical College, 1725 West Harrison St, Ste 821, Chicago, IL; email bonomi.phil{at}jimmy.harvard.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: Treatment with cisplatin-based chemotherapy provides a modest survival advantage over supportive care alone in advanced non–small-cell lung cancer (NSCLC). To determine whether a new agent, paclitaxel, would further improve survival in NSCLC, the Eastern Cooperative Oncology Group conducted a randomized trial comparing paclitaxel plus cisplatin to a standard chemotherapy regimen consisting of cisplatin and etoposide.

PATIENTS AND METHODS: The study was carried out by a multi-institutional cooperative group in chemotherapy-naive stage IIIB to IV NSCLC patients randomized to receive paclitaxel plus cisplatin or etoposide plus cisplatin. Paclitaxel was administered at two different dose levels (135 mg/m² and 250 mg/m²), and etoposide was given at a dose of 100 mg/m² daily on days 1 to 3. Each regimen was repeated every 21 days and each included cisplatin (75 mg/m²).

RESULTS: The characteristics of the 599 patients were well-balanced across the three treatment groups. Superior survival was observed with the combined paclitaxel regimens (median survival time, 9.9 months; 1-year survival rate, 38.9%) compared with etoposide plus cisplatin (median survival time, 7.6 months; 1-year survival rate, 31.8%; P = .048). Comparing survival for the two dose levels of paclitaxel revealed no significant difference. The median survival duration for the stage IIIB subgroup was 7.9 months for etoposide plus cisplatin patients versus 13.1 months for all paclitaxel patients (P = .152). For the stage IV subgroup, the median survival time for etoposide plus cisplatin was 7.6 months compared with 8.9 months for paclitaxel (P = .246).

With the exceptions of increased granulocytopenia on the low-dose paclitaxel regimen and increased myalgias, neurotoxicity, and, possibly, increased treatment-related cardiac events with high-dose paclitaxel, toxicity was similar across all three arms. Quality of life (QOL) declined significantly over the 6 months. However, QOL scores were not significantly different among the regimens.

CONCLUSION: As a result of these observations, paclitaxel (135 mg/m²) combined with cisplatin has replaced etoposide plus cisplatin as the reference regimen in our recently completed phase III trial.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
DISSEMINATED OR locally recurrent disease will occur in the majority of non–small-cell lung (NSCLC) cancer patients. This year approximately 135,000 deaths will occur from NSCLC in the United States.¹ This number is virtually equal to the total number of deaths from breast, prostate, and colon cancers, the second, third, and fourth leading causes of cancer deaths combined.¹ The outcome for untreated NSCLC patients with widespread disease is disappointingly predictable, with a median survival time of 4 months, and a 1-year survival rate of approximately 10% to 15%.^2,3 Over the past 2 decades, Eastern Cooperative Oncology Group (ECOG) investigators have tested a series of chemotherapy combinations in an effort to identify the most effective regimen for advanced NSCLC. Unfortunately, the relatively high response rates observed in earlier single-institution trials were not confirmed,^4-6 and no combination chemotherapy regimen provided an obvious survival advantage.^3-6 The overall median survival duration and the 1-year survival rate were 6 months and 19%, respectively.⁵

Given these results, ECOG investigators decided to use a strategy of new drug testing in chemotherapy naive patient populations. Through this program, a 21% response rate was observed, with paclitaxel given at a dose of 250 mg/m² over 24 hours every 3 weeks.⁷ Simultaneously, investigators at M.D. Anderson Cancer Center noted a 24% response rate with 200 mg/m² of paclitaxel every 21 days.⁸ The 1-year survival rate in each trial was approximately 40%.

Encouraged by the relatively high 1-year survival rates observed in the single-agent trials, a phase III study was designed to test the effect of paclitaxel on survival. Paclitaxel combined with cisplatin was compared with a standard cisplatin-containing regimen because modestly improved survival has been observed in advanced NSCLC patients treated with cisplatin-based chemotherapy.^2,3 Etoposide plus cisplatin was chosen as the reference regimen because it had produced the highest 1-year survival rate (25%) in previous ECOG phase III trials.⁹ However, the primary objective of this trial was to compare survival of three treatment arms, with particular interest in the role of paclitaxel compared with etoposide plus cisplatin. Secondary objectives included comparison of serial quality-of-life (QOL) measurements, response rates, and toxicity profiles, as well as comparison of the same parameters for two dose levels of paclitaxel (135 mg/m² v 250 mg/m²).

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patient entry onto this trial occurred between August 1993 and December 1994. Eligibility requirements included histologically or cytologically confirmed NSCLC (identification of small-cell lung cancer excluded patients); bidimensionally measurable or assessable stage IIIB or IV disease; ECOG performance status of 0 or 1; no previous history of malignant disease, with the exception of skin cancer or carcinoma-in-situ of the cervix; and no brain metastases. In addition, patients were required to have adequate organ function defined as follows: leukocyte count 4,000/µL, platelets 100,000/µL, bilirubin 1.5 mg/dL, and serum creatinine 1.5 mg/dL. Patients with active infections and patients who had received previous chemotherapy were excluded.

Patients were allowed to have had previous radiation, provided it had been completed 2 weeks before trial entry. However, patients whose only measurable lesion was within a previously radiated field were ineligible. Additional exclusion criteria included uncontrolled diabetes mellitus (defined as blood glucose > 200 mg/dL), uncontrolled hypertension, unstable angina, congestive heart failure, myocardial infarction within the previous year, and evidence of pre-existing peripheral neuropathy. This protocol was reviewed and approved by the human investigation committee at each participating institution. All patients gave written informed consent before study entry.

Patients were randomized to one of three chemotherapy regimens: (1) etoposide/cisplatin (EC) regimen: cisplatin 75 mg/m² intravenously (IV) over 1 hour on day 1 plus etoposide 100 mg/m² IV over 45 minutes on days 1, 2, and 3; (2) high-dose paclitaxel (PCG) regimen: cisplatin 75 mg/m² IV over 1 hour on day 2 preceded by paclitaxel 250 mg/m² IV as a 24-hour infusion on day 1, plus filgrastim 5 mcg/kg subcutaneously beginning on day 3 and continuing until the granulocyte count was 10,000/µL; and (3) low-dose paclitaxel (PC) regimen: cisplatin 75 mg/m² IV over 1 hour on day 2 preceded by paclitaxel 135 mg/m² IV as a 24-hour infusion starting on day 1. Each regimen was repeated every 21 days provided there was recovery from toxicity and no evidence of disease progression.

Stratification parameters included: (1) ECOG performance status of 0 versus 1; (2) weight loss during the previous 6 months less than 5% versus 5%; (3) stage IIIB versus stage IV disease; and (4) bidimensionally measurable disease versus assessable disease.

History and physical examination, complete blood count, serum chemistries, and tumor measurements were performed before each treatment cycle. Toxicity and tumor responses were defined according to ECOG criteria.¹⁰ The accrual goal of the study (585 patients) was estimated to have more than 90% power to detect a 50% increase in median survival time from 6 months on the EC regimen to 9 months on either of the paclitaxel-containing regimens at an overall, experiment-wise significance level of 5% to be monitored for up to four times using an O’Brien-Fleming type-group sequential method.^11,12 Fisher’s exact test¹³ was used to compare response rates, and the Kruskal-Wallis¹⁴ test was used to compare degrees of toxicity. Survival estimates were calculated by the Kaplan-Meier¹⁵ method, and the log-rank¹⁶ test was used for survival comparisons.

During the first two chemotherapy cycles, blood (7 to 10 mL) was collected in heparinized tubes from patients during the last hour of the 24-hour paclitaxel infusion. The blood was immediately centrifuged, and the plasma was stored at -20°C and subsequently packed in dry ice for shipment to the central laboratory for pharmacologic assay. Paclitaxel concentrations were measured by a high-performance liquid chromatography assay.¹⁷ End-of-infusion paclitaxel concentrations have been shown to be nearly equal to steady-state concentration values.¹⁸

QOL was assessed with version 2 of the Functional Assessment of Cancer-Lung (FACT-L) instrument,¹⁹ which consists of 35 questions, each of which is scored using a 5-point scale (not at all = 0, a little bit = 1, somewhat = 2, quite a bit = 3, and very much = 4). The questions addressed six areas: physical well-being, functional well-being, lung cancer symptoms and concerns, social well-being, emotional well-being, and relationship with the doctor.

The 21 questions relating to physical well-being, functional well-being, and lung cancer symptoms were combined and designated as the Trial Outcome Index (TOI), which, based on previous studies, is considered to be the best summary indicator of the physical component of QOL.¹⁹ Patients were asked to complete the FACT-L immediately before receiving the first course of chemotherapy and again 6,12, and 26 weeks later. The numeric values for each item were added to give the total score and the scores for subsets of questions, such as the TOI. Higher scores are associated with better QOL.¹⁹

A joint mixed effects and survival model was used to estimate the change in QOL scores as a continuous variable over 6 months and to compare the three treatment regimens.²⁰ Deterioration of QOL was ultimately expected in our population of incurable NSCLC patients. To determine whether there might be short-term improvement in QOL, the physical aspect of QOL (TOI) was analyzed as a discontinuous variable. Each patient was classified according to his QOL response. The QOL response category for individual patients was determined by calculating the difference between the baseline TOI score and the TOI scores at 6,12, and 26 weeks. QOL response categories were defined as follows: improved QOL = an increase of 5 units; stable QOL = a change of less than 5 units; worsened QOL = a decrease of 5 units; and unknown = missing data. Based on previously validated work, a change of 5 units was considered clinically significant.¹⁹ The ² test of independence was used to compare the frequency of QOL categories at the 6-,12-, and 26-week time points.

This study was formally reviewed by the ECOG Data Monitoring Committee in March and September of 1995 and in March of 1996 for possible early termination based on the primary efficacy comparison of blinded data. Observed differences at these interim analyses were not sufficiently significant for early termination of the study.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Between August 1993 and December 1994, 599 patients were entered onto the trial: 200 on the EC arm, 201 on the PCG arm, and 198 on the PC arm. Eleven patients were withdrawn for the following reasons: withdrawal of consent (five patients), death before starting treatment (three patients), discovery of brain metastases before starting protocol therapy (two patients), and infection (one patient). Fourteen patients were classified as ineligible for the following reasons: performance status 2 (five patients), required tests performed more than 2 weeks before protocol entry (two patients), multiple primary cancers (two patients), inadequate organ function (one patient), less than stage IIIB disease (one patient), radiation to only measurable site (one patient), small-cell histology (one patient), and radiation given concurrently with start of protocol therapy (one patient). The number of eligible and analyzed patients for each regimen was 193 on EC, 191 on PCG, and 190 on PC.

Patient characteristics are listed in Table 1. The median patient age was 61.8 years. Slightly more than one third of the patients were women, and approximately 13% of the patients were not white. In addition, one third of the patients were asymptomatic, and approximately 30% of the patients had lost more than 5% of their usual body weight during the 6 months before study entry. Stage IV disease was identified in 81% of the patients, whereas the remaining 19% had stage IIIB disease. The majority of the patients (79%) had measurable disease. The baseline patient characteristics were comparable among the three treatment groups.

Survival
Kaplan-Meier survival estimates were calculated after a median follow-up of 28.5 months and after 91.6% of the eligible patients had expired. Median survival durations and 1-year survival rates were as follows: EC = 7.6 months and 31.8%; PCG = 10.0 months and 40.3%; PC = 9.5 months and 37.4%, respectively. The Kaplan-Meier survival estimates for each treatment regimen are depicted in Fig 1. There was marginally improved survival for each paclitaxel arm compared with EC (P = .097 for PCG v EC and P = .090 for PC v EC). Survival between the two paclitaxel regimens was not different P = .931. The median survival time for all patients receiving paclitaxel was 9.9 months, with a 1-year survival rate of 38.9%. Survival for paclitaxel-treated patients (PCG and PC) was significantly longer than survival for EC-treated patients (Fig 2, P = .048). The P value for the comparison between EC and the combined paclitaxel arms was not adjusted. Survival results and log-rank comparisons for stage IIIB and stage IV patient subsets, according to treatment regimen, are listed in Table 2. For stage IIIB patients, the median survival duration and the 1-year survival rate for EC were 7.9 months and 40.0%, respectively, compared with 13.1 months and 54.9%, respectively, for PC and PCG combined (P = .152).

View larger version (15K): [in this window] [in a new window]   Fig 1. The Kaplan-Meier survival estimates for each treatment regimen are depicted above. The log-rank comparisons are as follows: EC v PCG: P = .097; EC v PC: P = .90; PCG v PC: P = .931.

View larger version (15K): [in this window] [in a new window]   Fig 2. The Kaplan-Meier survival estimates for all paclitaxel-treated patients (PC + PCG) and for patients treated with EC are shown in this graph. Survival is significantly longer for paclitaxel patients (log-rank, P = .048).

Failure-Free Survival
Progressive disease has been observed in 97.0% of the patients. The median survival time and 1-year failure-free survival results are as follows: EC = 2.8 months and 8.8%; PCG = 5.0 months and 13.1%; PC = 4.4 months and 11.1%; and PC + PCG = 4.8 months and 12.1%, respectively. The following results were observed for the differences in failure-free survival according to the log rank test: EC versus PCG, P = .007; EC versus PC, P = .067; PC versus PCG, P = .294; and EC versus PC + PCG, P = .007.

Toxicity
Hematologic and nonhematologic toxicity data are listed in Tables 3 and 4, respectively. The results of toxicity comparisons using the Kruskal-Wallis test are listed in Table 5. The major toxicity was granulocytopenia, with grade 4 toxicity occurring in the majority of patients. The incidence of proven infections and of temperature elevations ( 101^o) during periods of granulocytopenia was 7.4% and 9%, respectively.

Grade 3 neurologic toxicity, which was manifested predominantly as sensory abnormalities, occurred in 21% to 40% of the patients and was significantly more common on the high-dose paclitaxel arm. Severe nausea and vomiting were relatively infrequent. Grade 3 myalgias were significantly more common on the paclitaxel regimens, particularly with the high-dose regimen.

Fatal cardiac events, which were possibly related to treatment, were observed in 0.5% of patients on EC, in 0.5% of patients on PCG, and in 2% of patients on PC. According to Kruskal-Wallis tests, there were significantly more (P = .026) serious and possibly treatment-related cardiac events on the high-dose paclitaxel regimen but not (P = .143) on the low-dose paclitaxel regimen (Table 5).

Grade 5 cardiac events included sudden death in three patients, myocardial infarction in two patients, and hypotension associated with acute pericarditis in one patient. There was a history of cardiovascular disease in four of the patients who seemed to succumb as a result of cardiovascular disease; two patients had known coronary artery disease, one had hypertension, and one was being treated for cardiac arrhythmia.

One patient who expired suddenly had no known risk factors. The patient who was found to have acute pericarditis had received previous radiation, which included the heart. At autopsy, the pericardium was free of tumor.

There were three additional deaths that seemed to be treatment related. Two patients died of respiratory insufficiency, which may have been related to protocol therapy. One received the low-dose paclitaxel regimen, whereas the other was treated with the high-dose paclitaxel regimen. The remaining fatality was caused by renal insufficiency in a patient on the high-dose paclitaxel arm.

QOL
The baseline FACT-L was completed by 94% of the patients; and the compliance rates (defined as the percentage of expected evaluations) among the surviving patients at 6,12, and 26 weeks were 72%, 60%, and 50%, respectively. The changes over 6 months in the Trial Outcome Index (TOI) and the total FACT-L scores for each regimen were as follows: EC = 9.3 and 11.8; PCG = 10.2 and 11.4; and PC = 8.3 and 9.9, respectively. All three regimens demonstrated significant decreases in the scores over 6 months, but there were no significant differences between the regimens (P = .45 for TOI scores; P = .59 for the total FACT-scores). Short-term QOL responses (difference in TOI scores at baseline and 6 weeks) are listed in Table 6. A higher percentage of patients treated with paclitaxel were classified as having improved QOL, but the difference was not significant in the analysis, which excluded patients who failed to complete the 6-week FACT-L (P = .46).

Pharmacodynamics
Steady-state paclitaxel concentrations were measured during the first and second courses of therapy in 71 patients treated with PC and 75 patients treated with PCG. The mean (± SD) steady-state concentrations (courses 1 and 2) for high- (250 mg/m²) and low-dose (135 mg/m²) paclitaxel were 0.94 ± 0.59 mmol and 0.44 ± 0.45 mmol, respectively, which is a significant difference (P < .0001). However, no significant relationships were observed between steady-state paclitaxel concentrations and response (P = .37), survival (P = .72), worst degree of leukopenia (P = .08), neuromotor toxicity (P = .20), or neurosensory toxicity (P = .50).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Comparison of survival on each paclitaxel treatment arm versus survival on the EC regimen was one of the primary objectives of the current trial. The median survival for patients treated with EC was 6 months in a previous ECOG study.⁵ An increase of 50% in median survival from 6 to 9 months was considered a clinically significant survival improvement. Based on these considerations an accrual goal of 585 patients was established for the present study.

The median survival on each paclitaxel regimen exceeded 9 months. However, the median survival of 7.6 months that was observed in EC-treated patients was better than expected. Comparison of each paclitaxel regimen to the reference regimen revealed survival differences, which approached, but did not reach, statistical significance (P = .09 for EC v PC; P = .097 for EC v PCG).

Failure to achieve a significant survival difference for each paclitaxel treatment arm compared with EC seems to be related to the relatively long survival results observed with EC. The survival results observed in the PC-treated patients were virtually equivalent to the results that were anticipated based on the results of the phase II paclitaxel trials.^7,8 We should have expected better survival in patients treated with EC in the current trial because our more recent patients had more favorable prognostic characteristics. Had these factors been considered, the accrual goal for the current study would have been higher.

Why is survival for EC-treated patients longer in the current trial? The most likely explanation seems to be related to patient selection. It seems that patients in the current trial had a better prognosis than did patients in earlier ECOG trials. Poor performance status and more advanced disease stage are associated with shorter survival.²¹ In an earlier ECOG study, 17% of patients treated with EC were classified as having nonambulatory performance status (ECOG performance status of 2),⁵ whereas nonambulatory patients were excluded from the current trial. In addition, eligibility in our earlier study was limited to stage IV patients,⁵ but our present trial included patients who had less advanced disease, with 19% of the patients being classified as having stage IIIB disease.

European investigators have apparently observed better survival results with a similar, older chemotherapy regimen, teniposide plus cisplatin; and they have also surmised that the favorable shift in survival was secondary to the selection of better patients in their most recent study.²²

PCG treatment was associated with significantly longer progression-free survival compared with EC (P = .001). In contrast, comparison of overall survival for these regimens failed to show a significant difference (P = .097). Although the difference in progression-free survival for PC versus EC was not significant, this difference approached significance more closely than the difference in overall survival (P = .067 for progression-free survival; P = .09 for overall survival). These observations suggest that overall survival may be affected by second-line chemotherapy containing newer agents, and that it may be important to collect information regarding time to progression and the use of salvage therapy in phase III NSCLC trials. Results from a recent randomized study seem to support this possibility. Patients who had been treated with cisplatin chemotherapy were subsequently treated with docetaxel or with supportive care only. Median survival was longer in the patients who were treated with docetaxel (5.9 v 4.9 months).²³

Because the major objective of this trial was to assess the effect of paclitaxel on survival, EC and the combined paclitaxel regimens were compared in terms of survival. This analysis showed significantly longer survival for the combined paclitaxel regimens. The median survival time for PC + PCG was 9.8 months versus 7.6 months for EC, and the 1-year survival rate for PC + PCG was 38.9% versus 31.8% for EC (P = .048).

Analogous to the comparisons for the single paclitaxel regimens versus EC, the advantage for the combined paclitaxel regimens seems to be more pronounced for progression-free survival compared with overall survival (P = .007 for progression-free survival; P = .048 for overall survival).

Results of meta-analyses have shown that platinum-based chemotherapy is associated with a modest improvement in survival compared with supportive care only.^2,3 Although treatment with paclitaxel did not result in a 50% improvement in median survival, it seems that treatment with paclitaxel has produced a further, modest survival improvement compared with an older platinum-based regimen.

Paclitaxel has been tested in three additional phase III trials in which the drug was given as a 3-hour infusion,^22-24,25 in contrast to the 24-hour infusion rate used in our study. Paclitaxel was combined with cisplatin in two trials^22,24 and with carboplatin in the remaining trial.²⁵ The control arm was cisplatin alone in one trial²⁵ and teniposide plus cisplatin in a second trial²² In the remaining study, the investigators compared paclitaxel plus carboplatin with the same etoposide plus cisplatin regimen tested in our study.²² Each study was relatively large, with 179 to 206 patients entered onto each treatment arm. Mature data have been reported for the trial in which paclitaxel-cisplatin was compared with teniposide plus cisplatin,^24,25 but only preliminary results are available for the remaining two studies.^23,24 Response rates for the paclitaxel and platinum regimens ranged from 21% to 44%.^22-24,25

Similar to our results, two groups of investigators observed significantly higher response rates for the paclitaxel regimen,^22,24 whereas the remaining group observed a marginally significant higher response rate for the paclitaxel regimen.²⁵ Complete remission rates were less than 5% in each of the trials.

In contrast to our observations with respect to survival, the other studies showed no evidence of a survival advantage for paclitaxel.^22-24,25 Progression-free survival was longer with the paclitaxel treatments in our study and in the study reported by Gatzemeier et al.²⁴ However, the other trials failed to detect significant differences in progression-free survival.^22,25

The explanation for the conflicting survival results in the phase III paclitaxel trials is not readily apparent. One possibility for the discordant survival results may be related to the duration of the paclitaxel infusion. In our study, paclitaxel was infused over 24 hours because paclitaxel had been infused over 24 hours in the phase II trials,^7,8 which served as the basis for our phase III study. However, in each of the other trials, the drug was infused over 3 hours.^22-24,25 Preclinical observations suggest that prolonged exposure to paclitaxel results in greater cytotoxicity.²⁶ Paclitaxel dose may be another factor that affected survival. Our observations with paclitaxel given as a 24-hour infusion have shown that paclitaxel given at a dose of 250 mg/m² results in a two-fold increase in the plasma paclitaxel level compared with a paclitaxel dose of 135 mg/m². However, the lower paclitaxel dose is associated with less toxicity and similar survival results. Based on these observations, it seems that paclitaxel doses greater than 135 mg/m² are not clinically beneficial when the drug is given as a 24-hour infusion.

The optimal dose of paclitaxel given as a 3-hour infusion in the treatment of lung cancer is less well-defined. Phase I and II trials in which paclitaxel was combined with carboplatin have shown trends for lower response rates and shorter survival with doses less than 175 mg/m².^27,28 It is possible that 175 mg/m² is close to the minimum effective dose for paclitaxel given as a 3-hour infusion.

A shorter infusion time combined with a relatively low dose of paclitaxel might explain the failure to observe significantly improved survival with the paclitaxel regimens in two of these phase III studies.^22,24

However, this argument does not provide an explanation for the results of the third trial in which similar survival results were observed for patients treated with carboplatin combined with paclitaxel given at a dose of 225 mg/m² over 3 hours versus patients treated with EC.²⁸ Although the use of carboplatin rather than cisplatin may have influenced survival, the findings of this study increase concerns regarding the effectiveness of shorter paclitaxel infusion rates. This issue should be resolved because the same paclitaxel (225 mg/m² over 3 hours) plus carboplatin regimen has been included in three phase III studies, two of which have been completed. Carboplatin combined with paclitaxel, given as a 3-hour infusion, was compared with vinorelbine plus cisplatin in the recent Southwest Oncology Group trial.²⁹ Vinorelbine plus cisplatin was selected as the reference regimen in this study trial because this two-drug combination produced superior survival compared with cisplatin alone.³⁰ Survival results for paclitaxel plus carboplatin and for vinorelbine plus cisplatin were virtually identical, and toxicity results and patient compliance seem to favor paclitaxel plus carboplatin for use in future studies. The recently completed phase III ECOG trial in which paclitaxel at a dose of 225 mg/m² given over 3 hours was compared with paclitaxel at a dose of 135 mg/m² given over 24 hours will provide more information regarding the efficacy of shorter paclitaxel infusions.

Patterns of toxicity observed in our trial were consistent with the results reported from the other randomized paclitaxel trials.^22-24,25 Increased rates of myalgias and neurotoxicity were observed in paclitaxel-treated patients. In our trial, the rate of grade 3 neurotoxicity was significantly more frequent for the high-dose regimen (PCG) but not for the low-dose regimen (PC). It is somewhat surprising to observe a lack of correlation between paclitaxel serum levels and neurotoxicity. With serum paclitaxel levels being determined in approximately 50% of patients, this observation may have occurred because of low sample size. Two groups of investigators observed significantly worse nausea and vomiting on the reference regimen (cisplatin 100 mg/m²²⁴ and cisplatin plus etoposide).²⁵ Nausea and vomiting were similar on our three regimens, each of which included cisplatin at a dose of 75 mg/m². Significantly higher rates of infection and febrile neutropenia were observed in European patients treated with teniposide plus cisplatin.²² The differences in the rates of infection and febrile neutropenia were not significantly different in our study. There was a significantly higher rate of possibly treatment-related adverse cardiac events in our patients treated with PCG but not in patient’s treated with PC. None of the other investigators reported an increased rate of adverse cardiac events with the paclitaxel regimen.

Serial QOL measurements were collected in our trial and in the study conducted by Giaccone et al.²² The QOL instruments were different with the FACT-L instrument used in our study¹⁹ and the European Organization for Research and Treatment of Cancer Quality of Life Core Questionnaire-C30 and Lung Cancer Module-13 used in the European trial.²² Comparison of the studies reveals decreasing rates of compliance over time. It is difficult to determine whether this observation is related to lack of investigator enthusiasm or to rapid disease progression with concomitant deterioration in the patients’ functional status. We suspect that the latter situation is the major reason for the relatively low compliance. The declining compliance rates in both trials emphasize the difficulty of performing QOL studies in NSCLC patients. Nevertheless, both of these trials have provided a relatively large amount of serial QOL data in NSCLC patients treated on randomized clinical trials. Results of the QOL analysis from both studies failed to show significant differences in overall QOL scores or in individual components of QOL at the 3- and 6-month time points.

However, the European Organization for Research and Treatment of Cancer investigators observed significantly better QOL with respect to emotional well-being, cognitive function, social function, and global health state in paclitaxel patients 6 weeks after starting protocol treatment.²² Similarly, we observed a higher percentage of improvement in the physical aspect of QOL (TOI) 6 weeks after starting protocol treatment in paclitaxel-treated patients. However, the difference in the TOI for paclitaxel- versus nonpaclitaxel-treated patients was not significant. The lack of significant QOL difference seems to result from the fact that a significantly higher percentage of EC-treated patients did not complete the QOL questions at the 6 week time point. Although patients may have failed to complete the QOL questionnaire because of progressive disease and worsening symptoms, this possibility cannot be tested because data regarding the reasons for the failure to complete the FACT-L are not available.

Paclitaxel combined with cisplatin has produced a modest survival improvement compared with etoposide plus cisplatin without producing negative effects on QOL, and there was a trend for short-term improvement in the physical aspect of QOL in paclitaxel-treated patients. Based on these descriptions paclitaxel, infused over 24 hours at a dose of 135 mg/m², plus cisplatin has replaced EC as the reference regimen in the most recent phase III ECOG NSCLC trial. Results from the recently completed phase III ECOG trial, as well as data from other phase III studies testing shorter paclitaxel infusions, will provide additional information regarding the optimal use of paclitaxel in advanced NSCLC.

	ACKNOWLEDGMENTS

 
Supported in part by Public Health Service grants no. CA25988, CA23318, CA49957, CA66636, and CA21115 from the National Cancer Institute, National Institutes of Health, and the Department of Health and Human Services.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Landis SH, Murray T, Bolden S, et al: Cancer statistics. CA Cancer J Clin 49:8-31, 1999[Abstract/Free Full Text]

2. Grilli R, Oxman AD, Julian JA: Chemotherapy for advanced non–small-cell lung cancer: How much benefit is enough. J Clin Oncol 11:1866-1872, 1993[Abstract/Free Full Text]

3. 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. Br Med J 311:899-909, 1995[Abstract/Free Full Text]

4. Ruckdeschel JC, Finkelstein DM, Mason BA, et al: Chemotherapy for metastatic non-small cell bronchogenic carcinoma: EST:2575, generation V—A randomized comparison of four cisplatin-containing regimens. J Clin Oncol 3:72-79, 1985[Abstract]

5. Ruckdeschel JC, Finkelstein DN, Ettinger D, et al: A randomized trial of the four most active regimens for metastatic non–small-cell lung cancer. J Clin Oncol 4:14-22, 1986[Abstract]

6. Bonomi P, Finkelstein DN, Ruckdeschel JC, et al: Combination chemotherapy versus single agents followed by combination chemotherapy in stage IV non–small-cell lung cancer: A study of the Eastern Cooperative Oncology Group. J Clin Oncol 7:1602-1613, 1989[Abstract]

7. Chang AT, Kim K, Glick J, et al: Phase II study of Taxol, merbarone, and piroxantrone in stage IV non-small cell lung cancer: The Eastern Cooperative Oncology Group results. J Natl Cancer Inst 85:388-392, 1993[Abstract/Free Full Text]

8. Murphy WL, Fossella FV, Winn RJ, et al: Phase II study of Taxol in patients with untreated advanced non-small cell lung cancer. J Natl Cancer Inst 85:384-387, 1993[Abstract/Free Full Text]

9. Finkelstein DM, Ettinger DS, Ruckdeschel JC: Long-term survivors in metastatic non–small cell lung cancer: An Eastern Cooperative Group study. Oncol 4:702-709, 1986

10. Oken MM, Creech RH, Tormey DC, et al: Toxicity and response criteria of the Eastern Cooperative Oncology Group. Oncol 15:649-655, 1982

11. O’Brien PC, Fleming TR: A multiple testing procedure for clinical trials. Biometrics 35:549-556, 1979[Medline]

12. Lan KKG, DeMets DL: Discrete sequential boundaries for clinical trials. Biometrika 70:659-663, 1983[Abstract/Free Full Text]

13. Fisher RA: Statistical methods for research workers. Oliver and Boyd, Edinburgh, United Kingdom, 1925

14. Kruskal WH, Wallis WA: Use of ranks in one criterion variance analysis. J Am Stat Assoc 47:583-612, 1952

15. Kaplan FL, Meier P: Non-parametric estimation from incomplete estimation. J Am Stat Assoc 53:457-481, 1958

16. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966[Medline]

17. Longnecker SM, Donehower RC, Cates WE, et al: High performance liquid chromatographic assay for Taxol (NSC 125973) in human plasma and urine pharmacokinetics in a phase I trial. Cancer Treat Rep 71:53-59, 1981

18. Rowinsky EK, Chaundry V, Forestierre A, et al: Phase I and pharmacologic study of paclitaxel and cisplatin with granulocyte colony-stimulating factor: Neuromuscular toxicity is dose-limiting. J Clin Oncol 11:2010-2020, 1993[Abstract/Free Full Text]

19. Cella DG, Bonomi AE, Lloyd SR, et al: Reliability and validity of the functional assessment of cancer therapy-lung (FACT-L) quality of life instrument. Lung Cancer 12:199-220, 1995[Medline]

20. Fairclough DL, Peterson H, Cella D, et al: Comparison of model based methods dependent on missing data mechanism in two clinical trials of cancer therapy. Stat Med 17:781-796, 1998[Medline]

21. Aisner J, Hansen HH: Commentary: Current status of chemotherapy for non-small cell lung cancer. Cancer Treat Rep 65:979-986, 1981[Medline]

22. Giaccone G, Splinter TAW, Debruyne C, et al: Randomized study of paclitaxel-cisplatin versus cisplatin-teniposide in patients with advanced non–small-cell lung cancer. J Clin Oncol 16:2133-2141, 1998[Abstract]

23. Shepherd F, Ramlau R, Mattson K, et al: Randomized study of taxotere (tax) versus best supportive care (BSC) in non–small-cell lung cancer (MSCLC) patients previously treated with platinum-containing chemotherapy. Proc Am Soc Clin Oncol 18:463a, 1999 (abstr 1784)

24. Gatzemeier U, Von Pawel J, Gottfield M, et al: Phase III comparative study of high-dose cisplatin versus a combination regimen of paclitaxel and cisplatin in patients with advanced non–small-cell lung cancer. Proc Am Soc Clin Oncol 17:454a, 1998 (abstr 1748)

25. Belani CP, Natale RB, Lee JS, et al: Randomized phase III trial comparing cisplatin-etoposide versus carboplatin-paclitaxel in advanced and metastatic non–small-cell lung cancer. Clin Oncol 17:455a, 1998 (abstr 1751)

26. Raymond E, Hanauske A, Faivre S, et al: Effects of prolonged versus short-term exposure paclitaxel on human tumor colony forming units. Anticancer Drugs 8:379-385, 1997[Medline]

27. Kelly K, Pan Z, Murphy J, et al: A phase I trial of paclitaxel plus carboplatin in untreated patients with advanced non-small cell lung cancer. Clin Cancer Res 3:1117-1123, 1997[Abstract]

28. Huizing MT, Giaccone G, Van Warmerdam LJ, et al: Pharmacokinetics of paclitaxel and carboplatin in a dose escalating and dose-sequencing studies in patients with non–small-cell lung cancer. J Clin Oncol 15:317-329, 1997[Abstract/Free Full Text]

29. Kelly K, Crowley J, Bunn PA, et al: A randomized phase III trial of paclitaxel plus carboplatin (PC) versus vinorelbine plus cisplatin (VC) in untreated advanced non-small cell lung cancer (NSCLC): A Southwest Oncology Group trial. Clin Oncol 18:461a, 1999 (abstr 1777)

30. Wozniak AJ, Crowley JJ, Balcerzak SP, et al: Randomized trial comparing cisplatin with cisplatin-vinorelbine in the treatment of advanced non–small-cell lung cancer: A Southwest Oncology Group study. J Cln Oncol 16:2459-2465, 1998

Submitted March 2, 1999; accepted September 16, 1999.

This article has been cited by other articles:

				A. Miyanaga, A. Gemma, R. Noro, K. Kataoka, K. Matsuda, M. Nara, T. Okano, M. Seike, A. Yoshimura, A. Kawakami, et al. Antitumor activity of histone deacetylase inhibitors in non-small cell lung cancer cells: development of a molecular predictive model Mol. Cancer Ther., July 1, 2008; 7(7): 1923 - 1930. [Abstract] [Full Text] [PDF]

				M. J. Edelman, D. Watson, X. Wang, C. Morrison, R. A. Kratzke, S. Jewell, L. Hodgson, A. M. Mauer, A. Gajra, G. A. Masters, et al. Eicosanoid Modulation in Advanced Lung Cancer: Cyclooxygenase-2 Expression Is a Positive Predictive Factor for Celecoxib + Chemotherapy--Cancer and Leukemia Group B Trial 30203 J. Clin. Oncol., February 20, 2008; 26(6): 848 - 855. [Abstract] [Full Text] [PDF]

				S. S. Ramalingam, S. E. Dahlberg, C. J. Langer, R. Gray, C. P. Belani, J. R. Brahmer, A. B. Sandler, J. H. Schiller, and D. H. Johnson Outcomes for Elderly, Advanced-Stage Non Small-Cell Lung Cancer Patients Treated With Bevacizumab in Combination With Carboplatin and Paclitaxel: Analysis of Eastern Cooperative Oncology Group Trial 4599 J. Clin. Oncol., January 1, 2008; 26(1): 60 - 65. [Abstract] [Full Text] [PDF]

				D. -W. Kim, S. -Y. Kim, H. -K. Kim, S. -W. Kim, S. W. Shin, J. S. Kim, K. Park, M. Y. Lee, and D. S. Heo Multicenter phase II trial of Genexol-PM, a novel Cremophor-free, polymeric micelle formulation of paclitaxel, with cisplatin in patients with advanced non-small-cell lung cancer Ann. Onc., December 1, 2007; 18(12): 2009 - 2014. [Abstract] [Full Text] [PDF]

				M. A. Socinski, R. Crowell, T. E. Hensing, C. J. Langer, R. Lilenbaum, A. B. Sandler, and D. Morris Treatment of Non-small Cell Lung Cancer, Stage IV: ACCP Evidence-Based Clinical Practice Guidelines (2nd Edition) Chest, September 1, 2007; 132(3_suppl): 277S - 289S. [Abstract] [Full Text] [PDF]

				G. R. Simon and A. Turrisi Management of Small Cell Lung Cancer: ACCP Evidence-Based Clinical Practice Guidelines (2nd Edition) Chest, September 1, 2007; 132(3_suppl): 324S - 339S. [Abstract] [Full Text] [PDF]

				Y. Fujiwara, I. Sekine, K. Tsuta, Y. Ohe, H. Kunitoh, N. Yamamoto, H. Nokihara, K. Yamada, and T. Tamura Effect of Platinum Combined with Irinotecan or Paclitaxel against Large Cell Neuroendocrine Carcinoma of the Lung Jpn. J. Clin. Oncol., July 24, 2007; (2007) hym053v1. [Abstract] [Full Text] [PDF]

				G. Weiss, R Rosell, F Fossella, M Perry, R Stahel, F Barata, B Nguyen, S Paul, P McAndrews, N Hanna, et al. 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 Ann. Onc., March 1, 2007; 18(3): 453 - 460. [Abstract] [Full Text] [PDF]

				P Comella, G Filippelli, G De Cataldis, B Massidda, G Frasci, L Maiorino, C Putzu, S Mancarella, S Palmeri, R Cioffi, et al. Efficacy of the combination of cisplatin with either gemcitabine and vinorelbine or gemcitabine and paclitaxel in the treatment of locally advanced or metastatic non-small-cell lung cancer: a phase III randomised trial of the Southern Italy Cooperative Oncology Group (SICOG 0101) Ann. Onc., February 1, 2007; 18(2): 324 - 330. [Abstract] [Full Text] [PDF]

				A. D. Colevas Chemotherapy Options for Patients With Metastatic or Recurrent Squamous Cell Carcinoma of the Head and Neck J. Clin. Oncol., June 10, 2006; 24(17): 2644 - 2652. [Abstract] [Full Text] [PDF]

				C. A. Dooms, Y. N. Lievens, and J. F. Vansteenkiste Cost-utility analysis of chemotherapy in symptomatic advanced nonsmall cell lung cancer Eur. Respir. J., May 1, 2006; 27(5): 895 - 901. [Abstract] [Full Text] [PDF]

				L. J. Fallowfield, J. M. Bliss, L. S. Porter, M. H. Price, C. F. Snowdon, S. E. Jones, R. C. Coombes, and E. Hall Quality of Life in the Intergroup Exemestane Study: A Randomized Trial of Exemestane Versus Continued Tamoxifen After 2 to 3 Years of Tamoxifen in Postmenopausal Women With Primary Breast Cancer J. Clin. Oncol., February 20, 2006; 24(6): 910 - 917. [Abstract] [Full Text] [PDF]

				M. J. Edelman, R. Smith, P. Hausner, L. A. Doyle, K. Kalra, J. Kendall, M. Bedor, and S. Bisaccia Phase II Trial of the Novel Retinoid, Bexarotene, and Gemcitabine Plus Carboplatin in Advanced Non-Small-Cell Lung Cancer J. Clin. Oncol., August 20, 2005; 23(24): 5774 - 5778. [Abstract] [Full Text] [PDF]

				B. Besse, J. C. Soria, and T. Le Chevalier Front-line doublets in advanced non-small cell lung cancer: The golden age for second line chemotherapy Ann. Onc., July 1, 2005; 16(7): 997 - 998. [Full Text] [PDF]

				C. P. Belani, J. S. Lee, M. A. Socinski, F. Robert, D. Waterhouse, K. Rowland, R. Ansari, R. Lilenbaum, and R. B. Natale Randomized phase III trial comparing cisplatin-etoposide to carboplatin-paclitaxel in advanced or metastatic non-small cell lung cancer Ann. Onc., July 1, 2005; 16(7): 1069 - 1075. [Abstract] [Full Text] [PDF]

				T. Winton, R. Livingston, D. Johnson, J. Rigas, M. Johnston, C. Butts, Y. Cormier, G. Goss, R. Inculet, E. Vallieres, et al. Vinorelbine plus Cisplatin vs. Observation in Resected Non-Small-Cell Lung Cancer N. Engl. J. Med., June 23, 2005; 352(25): 2589 - 2597. [Abstract] [Full Text] [PDF]