SEARCH

SEARCH BY CITATION

Keywords:

  • chemotherapy regimens;
  • nonsmall cell lung carcinoma;
  • platinum regimens;
  • nonplatinum regimens

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

The authors compared the toxicity, response rate, and progression free survival of four chemotherapy regimens for patients with advanced (Stage IIIB and IV) nonsmall cell lung carcinoma.

METHODS

A total of 267 patients entered this randomized Phase II trial on one of four arms: paclitaxel, carboplatin, and gemcitabine (Arm A); paclitaxel, carboplatin, and vinorelbine (Arm B); paclitaxel and gemcitabine (Arm C); and gemcitabine and vinorelbine (Arm D). Patient characteristics were similar in all treatment arms. At the time of tumor progression, patients were removed from study and were treated at the discretion of their physician.

RESULTS

Patients received a median of four courses of chemotherapy in all arms, and there was no difference in the dose delivered. There were no statistical differences in response rates (range, 32–45%), median progression free survival (range, 4.9–6.6 months), or progression free survival at 1 year (range, 8–19%). Actuarial survival in all four arms was not different, with a median survival ranging from 8.7 months to 10.7 months and a 1-year survival rate of 38–44%. Each arm was compared with a historic control with a median survival of 8 months. Arm D (gemcitabine and vinorelbine) approached significance at the 0.05 level.

CONCLUSIONS

Two-drug combinations containing the newer drugs without a platinum drug were less toxic than three-drug, platinum-based regimens. There were no significant differences in objective response rates or progression free survival when the four regimens were compared. The two-drug combination of gemcitabine and vinorelbine was the least toxic and, thus, may be superior. A Phase III trial comparing combined gemcitabine and vinorelbine with combined paclitaxel, carboplatin, and gemcitabine is ongoing. Cancer 2002;95:1279–85. © 2002 American Cancer Society.

DOI 10.1002/cncr.10810

Chemotherapy for patients with nonsmall cell lung carcinoma has evolved slowly during the last 15 years. Several newer cytotoxic agents (paclitaxel, gemcitabine, vinorelbine, docetaxel, and irinotecan) have been studied more recently as single agents and in combination with one another or with a platinum (either cisplatin or carboplatin) and appear to be superior to the older, cisplatin-based regimens.1

Several attempts have been made and currently are underway to exploit the improved efficacy or decreased toxicity of the newer drugs. There is a question whether platinum (cisplatin or carboplatin) is a necessary component of combination chemotherapy. Several two-drug combinations2 and three-drug combinations,3, 4 including the newer drugs, have been tested recently in Phase II trials. Comparisons of two-drug combinations of the new drugs with three-drug combinations containing two of the newer drugs plus a platinum are of interest. Toxicity is a major concern, and further testing of relatively well-tolerated two-drug combinations and comparisons with more aggressive, potentially more efficacious regimens are an important area of investigation. The Minnie Pearl Cancer Research Network conducted a prospective randomized study of four chemotherapy regimens of two two-drug regimens and two three-drug regimens to obtain a better understanding of the relative differences of these therapies. We previously tested all of these regimens in Phase II single-arm studies.3–6 Here, we report the details and results of a randomized prospective Phase II comparative trial.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

This randomized Phase II study was performed by the Minnie Pearl Cancer Research Network, a community-based cooperative group. Patients with Stage IV and IIIB nonsmall cell lung carcinoma were treated by random allocation (random card system) with one of four chemotherapy regimens, all of which contained combinations of the newer drugs: Patients in Arm A received paclitaxel, carboplatin, and gemcitabine (PCG); patients in Arm B received paclitaxel, carboplatin, and vinorelbine (PCV); patients in Arm C received paclitaxel and gemcitabine (PG); and patients in Arm D received gemcitabine and vinorelbine (GV). The drug doses and schedules used in these four regimens are detailed in Table 1. In the PCV regimen (Arm B), vinorelbine was given on either Day 8 or Day 15 (not on both days), depending on adequate granulocyte (≥ 1000/μL) and platelet (≥ 75,000/μL) counts. At the time of disease progression, patients were not treated in a standard fashion but were treated at the discretion of the investigator.

Table 1. Treatment Regimens
  • AUC: area under the concentration time curve.

  • a

    Day 8 or 15, depending on adequate granulocyte (≥ 1000/μL) and/or platelet (≥ 75,000/μL) counts.

Arm A (PCG)
 Paclitaxel 200 mg/m2 Day 1
 Carboplatin (AUC = 5) Day 1
 Gemcitabine 1000 mg/m2 Days 1 and 8
 Repeat every 3 weeks for 6 courses
Arm B (PCV)
 Paclitaxel 200 mg/m2 Day 1
 Carboplatin (AUC = 6) Day 1
 Vinorelbine 20 mg/m2 Days 1 and 8 or Days 1 and 15a
 Repeat every 3 weeks for 6 courses
Arm C (PG)
 Paclitaxel 200 mg/m2 Day 1
 Gemcitabine 1000 mg/m2 Days 1, 8, and 15
 Repeat every 3 weeks for 6 courses
Arm D (GV)
 Gemcitabine 1000 mg/m2 Days 1, 8, and 15
 Vinorelbine 25 mg/m2 Days 1, 8, and 15
 Repeat every 4 weeks for 6 courses

Patients were eligible for this study if they had previously untreated, biopsy-proven nonsmall cell lung carcinoma (either Stage IV or Stage IIIB) and if they were not eligible for combined modality therapy. Additional eligibility criteria included an Eastern Cooperative Oncology Group performance status of 0, 1, or 2; no previous chemotherapy; no brain metastasis; measurable or evaluable tumor; adequate bone marrow function (white blood cells ≥ 3000/μL and platelets ≥ 100,000/μL), serum creatinine < 2.0 mg/dL; and serum bilirubin ≤ 2.0mg/dL. All patients gave written informed consent prior to participating in this clinical trial. This trial was approved by the institutional review board at Centennial Medical Center (Nashville, TN) and by the Institutional Review Boards of participating network sites. Dose modifications were based primarily on the degree of myelotoxicity and were based on blood counts measured on the day of scheduled treatment. Dose reductions were identical to those used previously in our Phase II trials of each of these regimens.3–6 In addition, various Grade 3–4 nonhematologic toxicities required dose reductions or discontinuation.3–6 Cytokines were not used routinely but were allowed as a standard approach if necessary after the first cycle of chemotherapy.

After the completion of two treatment courses, all patients underwent reevaluation, including a repeat of all previously abnormal radiologic studies. Patients were assigned a response category using standard definitions. A complete response required the complete resolution of all clinical evidence of tumor, as determined by two observations not less than 4 weeks apart. A partial response required a decrease ≥ 50% in the size of measurable lesions, as determined by the sum of the greatest tumor dimensions, with no new lesions appearing. Stable disease was defined as a decrease < 50% or an increase < 25% in tumor measurements, with no new lesions appearing. Progressive disease occurred when measurable lesions increased in size by > 25%, evaluable lesions worsened, or new lesions appeared.

Statistical Considerations

The primary end points of this study were comparisons of the response rates and toxicities. A secondary end point was comparison of progression free survival. In addition, each arm of the trial was considered separately in reference to median survival and 1-year survival. Survival was determined by the Kaplan–Meier method.7 A sample size of 65 patients per treatment arm was selected on the basis of an anticipated 10% difference in response in any arm and a 90% probability of correctly selecting the arm with the highest response rate (estimated at 35%). Rates of toxicities were estimated ± 10% at the worst with a 95% confidence interval.

Each arm of this study also was compared with the median survival (8 months) and the 6-month survival (59%) of a historic control.8 The triangle test was used to compare independently the 6-month survival of each arm of the study reported here with the historic control. The triangle test for survival9 is based on the triangle test for tumor response data.10 The time to event estimates was obtained using the Kaplan–Meier method.7 The log-rank statistic was used to compare time to event distributions between treatment arms.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Two hundred sixty-seven patients were entered on this randomized trial from February 1998 through January 2001 by 34 affiliate member groups of the Minnie Pearl Cancer Research Network. The characteristics of patients who received each of the four treatments are compared in Table 2. There were no significant differences in age, gender, performance status, stage, or histology among the four groups. Ninety percent of patients with Stage IIIB disease had pleural effusions.

Table 2. Patient Characteristics (n = 267 patients)
CharacteristicArm A (PCG) (n = 71 patients)Arm B (PCV) (n = 65 patients)Arm C (PG) (n = 64 patients)Arm D (GV) (n = 67 patients)
  1. PCG: paclitaxel, carboplatin, and gemcitabine; PCV: paclitaxel, carboplatin, and vinorelbine; PG: paclitaxel and gemcitabine; GV: gemcitabine and vinorelbine.

Median age in yrs (range)63 (36–81)63 (36–78)64 (42–82)62 (33–80)
Male/female (%)47/24 (66/34)46/19 (71/29)44/20 (69/31)46/21 (69/31)
Performance status (%)    
 0–161 (86)58 (89)55 (86)59 (88)
 210 (14)7 (11)9 (14)8 (12)
Disease stage (%)    
 IIIB20 (28)14 (22)11 (17)18 (27)
 IV51 (72)51 (78)53 (83)49 (73)
Histology (%)    
 Adenocarcinoma40 (56)23 (36)27 (42)31 (46)
 Squamous cell14 (20)19 (29)11 (17)14 (21)
 Large cell8 (11)6 (9)9 (14)6 (9)
 Mixed or not specified9 (13)17 (26)17 (27)16 (24)

A median of four courses of chemotherapy was given in all arms. The percentage of planned doses delivered in the first two courses is outlined in Table 3. There was no difference in the ability to deliver the treatment in any of the arms.

Table 3. Percentage of Planned Doses Delivered in the First Two Courses
RegimenDelivered (%)
Course 1Course 2
  1. PCG: paclitaxel, carboplatin, and gemcitabine; PCV: paclitaxel, carboplatin, and vinorelbine; PG: paclitaxel and gemcitabine; GV: gemcitabine and vinorelbine.

Arm A (PCG)8886
Arm B (PCV)8982
Arm C (PG)9085
Arm D (GV)8781

Toxicity

The three-drug regimens were more myelotoxic than the two-drug regimens, as illustrated in Table 4. Grade 3 and 4 leukopenia was less frequent with the two-drug combinations, and thrombocytopenia and anemia were more frequent with Arm B (PCV) compared with the other three regimens (Table 4).

Table 4. Hematologic Toxicity (Grade 3–4)
ToxicityNo. of patients (% of courses)
Arm A (PCG)Arm B (PCV)Arm C (PG)Arm D (GV)
  • PCG: paclitaxel, carboplatin, and gemcitabine; PCV: paclitaxel, carboplatin, and vinorelbine; PG: paclitaxel and gemcitabine; GV: gemcitabine and vinorelbine; RBC: red blood cells.

  • a

    Arm C versus Arms A and B, P = 0.04; Arm D versus Arms A and B, P = 0.06.

  • b

    Arm B versus Arm A and Arm C versus Arm A, P = 0.04; Arm D versus Arm A, P = 0.01.

  • c

    Arm A versus Arms B–D, P = 0.04.

Leukopeniaa31 (21)41 (36)11 (9)20 (14)
 Grade 31619816
 Grade 4152234
Anemia (all Grade 3)(5)(4)(4)(2)
Thrombocytopeniab30 (17)3 (3)8 (3)2 (1)
 Grade 326262
 Grade 44120
Hospitalizations for neutropenia/fever3 (1)16 (10)6 (2)4 (2)
Platelet transfusions8 (4)2 (1)1 (0.5)0 (0)
RBC transfusionsc20 (11)12 (5)8 (5)9 (5)
Cytokines given (granulocyte or granulocyte-macrophage stimulating factor, erythropoietin)(32)(31)(25)(20)
Septic death (no.)1121

The nonhematologic toxicities are illustrated in Table 5. Arm D (GV) was substantially less toxic than the other regimens, producing less alopecia, nausea, emesis, and arthralgia/myalgia. Both of the two-drug regimens (PG and GV) produced less neuropathy than Arm B (PCV). There were five treatment-related deaths (1.8%), all due to sepsis, and these were distributed equally among patients in the four treatment arms.

Table 5. Nonhematologic Toxicity
ToxicityNo. of patients (% of courses)
Arm A (PCG)Arm B (PCV)Arm C (PG)Arm D (GV)
  • PCG: paclitaxel, carboplatin, and gemcitabine; PCV: paclitaxel, carboplatin, and vinorelbine; PG: paclitaxel and gemcitabine; GV: gemcitabine and vinorelbine.

  • a

    Arm D versus Arm B, P = 0.04.

  • b

    Arm D versus Arms A–C, P < 0.01.

  • c

    Arm D versus Arm C, P = 0.05; Arm D versus Arm B, P = 0.02; Arm D versus Arm A, P = 0.01.

Neuropathya22 (10% (22 pts)32 (21)13 (10)9 (4)
 Grade 22029129
 Grade 32310
 Grade 40000
Alopeciab    
 Grade 1(0)(0)(0)(25)
 Grade 2(100)(100)(100)(0)
Nausea/emesis (Grade 2–3)16 (6)12 (5)16 (8)6 (2)
Diarrhea (Grade 2–3)9 (5)5 (2)7 (3)2 (1)
Myalgia/arthralgiac28 (14)39 (20)27 (19)12 (7)
 Grade 327382611
 Grade 41111

Efficacy

The responses of patients who were treated with each of the four regimens are illustrated in Table 6. Of 267 patients, 26 patients were not evaluable for response. The reasons patients were not evaluable for response did not differ among treatment arms and included toxicity (n = 9 patients), intercurrent illness (n = 7 patients), progressive tumor (n = 6 patients), patient request to stop therapy (n = 3 patients), and investigator error (n = 1 patient). No differences in overall objective response rates were detected among patients who received these regimens. Response rates varied from 32% in Arm C (PG) to 45% in Arm B (PCV). The 95% confidence intervals for the response rates were 32–41% in Arm A, 42–49% in Arm B, 29–38% in Arm C, and 29–38% in Arm D. The median progression free survival ranged from 4.9 months to 6.6 months, and the actuarial progression free survival at 1 year ranged from 8% to 19%. There were no significant differences noted in progression free survival in any of the four treatment arms. However, because the GV regimen was given every 4 weeks, compared with every 3 weeks for the other three regimens, there may have been a cumulative delay of several weeks in the evaluation of patients who received the GV regimen.

Table 6. Response Rates
ResponseNo. of patients (%)a
Arm A (PCG)Arm B (PCV)Arm C (PG)Arm D (GV)
  • PCG: paclitaxel, carboplatin, and gemcitabine; PCV: paclitaxel, carboplatin, and vinorelbine; PG: paclitaxel and gemcitabine; GV: gemcitabine and vinorelbine; CR: complete response; PR: partial response; PD: progressive disease; NE: not evaluated.

  • a

    All P values are ≥ 0.1.

No. of patients71656467
CR2 (3)2 (3)1 (2%)2 (3)
PR24 (34)27 (42)19 (30%)20 (30)
Stable29 (41)19 (29)25 (39%)26 (39)
PD9 (12)10 (15)11 (17)15 (22)
NE7 (10)7 (11)8 (12)4 (6)

Figure 1 shows the actuarial survival curves for patients who were treated with each of the four regimens. The median survival ranged from 8.7 months to 10.7 months, and actuarial 1-year survivals ranged from 38% to 44%. Using log-rank comparisons, no significant differences were detected among the survival curves. The triangle test9 was used to compare each arm with an arbitrary historic control with a 6-month survival of 59% and a median survival of 8 months.8 In this evaluation, only Arm D (GV) approached significance at the 0.05 level.

thumbnail image

Figure 1. Actuarial survival curves for the four therapy arms showed no difference in the median survival, 1-year survival, or overall survival. PCG: paclitaxel, carboplatin, and gemcitabine; PCV: paclitaxel, carboplatin, and vinorelbine; PG: paclitaxel and gemcitabine; GV: gemcitabine and vinorelbine.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

This randomized Phase II comparison was useful in showing that the two-drug regimens, neither of which contained a platinum agent, were less toxic than the three-drug regimens. The GV combination was the best tolerated regimen and was the only regimen that was not associated with severe alopecia. Furthermore, there were no significant differences in objective response rates or progression free survival produced by these four regimens. The actuarial 1-year survival rates were consistent with other multicenter studies using the newer third-generation combination regimens for patients with advanced nonsmall cell lung carcinoma.

The design of this study did not allow us to conclude that there was no difference in survival when comparing the four treatment regimens. The study was not large enough to make that determination. However, the toxicity evaluation and objective response rates comparisons were conclusive. When comparing each arm with an historic control8 with a median survival of 8 months, Arm D (GV) was the most promising therapy to advance to a more definitive Phase III randomized comparison. This is particularly true considering that GV was the least toxic of the regimens tested. For these reasons, we have chosen to continue this trial using a Phase III design, and we have deleted Arm B (PCV) and Arm C (PG). When it is completed, the trial will allow a definitive survival comparison of the efficacy of a well-tolerated, nonplatinum combination (GV) with a more intensive, three-drug regimen (PCG). Because the two-drug regimen already has been proven less toxic, even the demonstration of equal efficacy would have important implications for the treatment of patients with advanced nonsmall cell lung carcinoma as well as patients with earlier stage disease.

Other Phase II studies have been reported using gemcitabine and vinorelbine on either a Day 1, 8, and 15 schedule every 28 days11 or on a Day 1 and 8 schedule every 21 days.12 Krajnik et al.11 used gemcitabine 1200 mg/m2 and vinorelbine 25 mg/m2 on Days 1, 8, and 15 every 28 days and found that this regimen was active and well tolerated; those authors suggested comparing their regimen with platinum-based regimens. Gridelli et al.12 also reported on a Day 1 and 8 schedule every 21 days and felt that the combination was active and tolerable. These results and others13 prompted the completion of a large Phase III study in elderly patients comparing gemcitabine plus vinorelbine with gemcitabine alone and with vinorelbine alone.14 In that study, there was no observed difference in survival in any treatment arm. The role of gemcitabine plus vinorelbine continues to evolve, and the results of the Phase II randomized study reported here show that this two-drug, nonplatinum-containing combination is less toxic and perhaps is as active as more aggressive and toxic platinum-containing regimens.

Other recent Phase III trials already have provided some information regarding nonplatinum doublets and three-drug regimens. A Phase III randomized comparison reported by the European Organization for Research and Treatment of Cancer did not show any efficacy advantage for the combination of gemcitabine plus paclitaxel compared with two cisplatin-based combination regimens,15 although the gemcitabine plus paclitaxel combination showed a trend toward worse survival but had less toxicity. A second Phase III trial by the Spanish Lung Cancer Group also showed no survival benefit, but there was more toxicity with a cisplatin-based, three-drug combination compared with a two-drug, cisplatin-based combination and with a nonplatinum, two-drug combinations given in sequence.16

The development of molecular or biologic-targeted therapies with compounds that specifically inhibit tumor growth factors and angiogenesis represent an exciting potential for improved therapy of patients with several types of malignancies, including nonsmall cell lung carcinoma. The addition of one or more of these novel agents to well-tolerated combinations of newer chemotherapeutic agents seems likely to improve therapy in the near future.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Bunn PA Jr., Kelly K. New chemotherapeutic agents prolong survival and improve quality of life in non-small cell lung cancer: a review of the literature and future directions. Clin Cancer Res. 1998; 4: 10871100.
  • 2
    Murren JR. Rationale for non-platinum chemotherapy in advanced non-small cell lung cancer. Oncology. 2001; 15 (Suppl 7): 2934.
  • 3
    Hainsworth JD, Burris HA, Erland JB, et al. Phase I/II trial of paclitaxel by one-hour infusion, carboplatin, and gemcitabine in the treatment of patients with advanced non-small cell lung cancer. Cancer. 1999; 85: 12691276.
  • 4
    Hainsworth JD, Burris HA, Morrissey LH, et al. Paclitaxel, carboplatin, and vinorelbine in the treatment of advanced non-small cell lung cancer: a phase trial of the Minnie Pearl Cancer Research Network. Cancer J Sci Am. 2000; 6: 151156.
  • 5
    Hainsworth JD, Burris HA, Litchy S, et al. Gemcitabine and vinorelbine in second-line treatment of non-small cell carcinoma patients: a Minnie Pearl Cancer Research Network Phase II trial. Cancer. 2000; 88: 13531358.
  • 6
    Meluch AA, Greco FA, Burris HA, et al. Paclitaxel and gemcitabine chemotherapy for advanced transitional-cell carcinoma of the urothelial tract, a Phase II trial of the Minnie Pearl Cancer Research Network. J Clin Oncol. 2001; 19: 30183024.
  • 7
    Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958; 53: 457481.
  • 8
    Kelly K, Crowley J, Bunn PA Jr., et al. Randomized Phase III trial of paclitaxel plus carboplatin versus vinorelbine plus cisplatin in the treatment of patients with advanced non-small cell lung cancer, a Southwest Oncology Group trial. J Clin Oncol. 2001; 19: 32103218.
  • 9
    Seiferheld W, Scott C, Byhardt RW, Curran WJ. The triangle test reduces time to decision in Phase II trials of survival: results from the radiation therapy oncology group (RTOF). Proc Soc Clin Trials. 1999; 20: 54s.
  • 10
    Bellissant E, Benichou J, Chastang C. Application of the triangular test to Phase II cancer clinical trials. Stat Med. 1990; 9: 907917.
  • 11
    Krajnik G, Mohn-Staudner A, Thaler J, et al. Vinorelbine-gemcitabine in advanced non-small cell lung cancer (NSCLC): an AASLC Phase II trial. Austrian Association for the Study of Lung Cancer. Ann Oncol. 2000; 11: 993998.
  • 12
    Gridelli C, Cigolari S, Gallo C, et al. Activity and toxicity of gemcitabine and gemcitabine + vinorelbine in advanced non-small cell lung cancer elderly patients: Phase II data from the Multicenter Italian Lung Cancer in the Elderly Study (MILES) randomized trial. Lung Cancer. 2001; 31: 277284.
  • 13
    Frasci G, Lorusso V, Panza N, et al. Gemcitabine plus vinorelbine versus vinorelbine alone in elderly patients with advanced non-small cell lung cancer. J Clin Oncol. 2000; 18: 25292536.
  • 14
    Gridelli C, Perrone F, Cigolari S, et al. The MILES (Multicenter Italian Lung Cancer in the Elderly Study) phase 3 trial: gemcitabine + vinorelbine vs vinorelbine and vs gemcitabine in elderly advanced NSCLC patients [abstract 1230]. Proc Am Soc Clin Oncol. 2001; 20: 308a.
  • 15
    Van Meerbeeck JP, Smit EF, Lianes P, et al. An EORTC randomized Phase II trial of three chemotherapy regimens in advanced non-small cell lung cancer [abstract 1228]. Proc Am Soc Clin Oncol. 2001; 20: 308a.
  • 16
    Alberola V, Camps C, Provencia M, et al. Cisplatin/gemcitabine versus cisplatin/gemcitabine/vinorelbine versus sequential doublets of gemcitabine/vinorelbine followed by ifosfamide/vinorelbine in advanced non-small cell lung cancer: results of a Spanish Lung Cancer Group Phase III trial [abstract 1229]. Proc Am Soc Clin Oncol. 2001; 20: 308a.