The incidence of lung cancer in China has been estimated as higher than any other cancer in men and the second highest, after breast cancer, in women.1 Nonsmall cell lung cancer (NSCLC) represents the majority of all lung cancers and approximately one-third of patients present with locally advanced disease at the time of diagnosis. The prognosis for patients with clinical stages IIIB and IV is poor, with a 5-year survival rate of only 3%.2
Gemcitabine plus cisplatin is a standard treatment for stages IIIB and IV NSCLC and is used in many countries as the first-line chemotherapy treatment for NSCLC. The combination had a clear synergistic mechanism of action in preclinical studies,3 and has been proven effective in phase III clinical trials.4–7 Gemcitabine plus cisplatin therapy for advanced NSCLC has been tested over both a 4-week and a 3-week schedule. Phases II and III studies in which gemcitabine (usually 1000 mg/m2) was administered weekly on days 1, 8 and 15 and cisplatin (usually 100 mg/m2) was administered once on days 1, 8, or 15 of a 28-day schedule to patients with advanced NSCLC showed response rates ranging from 22% to 54% and median survival times reported from 8.1 to 15.4 months.5–13 The major toxicities observed in the above trials were haematological, with grade 3/4 neutropenia occurring in 40% to 63% of patients and grade 3/4 thrombocytopenia occurring in 15% to 65% of patients. The haematological toxicity, while transient and not life-threatening, frequently resulted in the need to reduce or omit subsequent doses of gemcitabine (most commonly on day 15) which translated into a substantial reduction of planned dose intensity. As such, the feasibility of a 3-week schedule with gemcitabine administered on days 1 and 8 and cisplatin administered on day 1 or day 2 was assessed. The 3-week gemcitabine-cisplatin schedule has been reported in phases II and III studies to be effective in the treatment of advanced NSCLC with manageable toxicity as evidenced by response rates (41% to 65%), median survival times (8.7 to 15.4 months) and incidence of grade 3/4 neutropenia (11% to 64%) and thrombocytopenia (6% to 56%).4,14–16
To date, there are insufficient local data to support use of the 3-week schedule in the Chinese population. Since the 3-week schedule produces similar efficacy and possibly lower toxicity than the 4-week schedule in other populations, studies in the Chinese population are warranted. The objectives of this randomized multicentre study, B9E-GH-JHSE, were to determine the response rate and toxicity profile of a 3-week versus a 4-week regimen of gemcitabine combined with cisplatin as first-line treatment for Chinese patients with advanced NSCLC.
Eligible patients for this study had a histological or cytological diagnosis of NSCLC stage IIIB or stage IV disease not amenable to curative surgery and at least one lesion. No prior systemic chemotherapy or prior radiotherapy for the target organ was allowed. Other inclusion criteria was 18 to 75 years of age; Karnofsky Performance Status (KPS) of 80 or higher; estimated life expectancy of at least 12 weeks; white blood cell count ≥3.5×109/L; absolute neutrophil (segmented and bands) count (ANC) ≥2.0×109/1; platelets ≥100×109/L; haemoglobin ≥9 g/dl; bilirubin ≤2 × upper limit of normal (ULN); alanine transaminase (ALT) or aspartate transaminase (AST) ≤5 × ULN; serum creatinine ≤1.5 × ULN and calculated creatinine clearance (Cockcroft- Gault formula) ≥45 ml/min. Exclusion criteria were histological or cytological diagnosis of lung cancer other than NSCLC; brain metastasis; pregnancy or breast feeding; a second primary malignancy that was clinically detectable at the time of consideration for enrolment and history of a significant neurological or mental disorder. This study was conducted according to ICH Good Clinical Practice guidelines, including obtaining written informed consent from all patients.
Prior to entering the study, all patients underwent a physical examination: electrocardiography, chest × ray, computed tomography (CT) scan of the brain, abdomen and thorax, bone scan or skeletal X-ray, urine pregnancy test and blood testing (chemistry and haematology). Eligible patients were randomized to one of two treatment groups; the randomization scheme was performed by a computerized voice response system at a central location for all sites. In the 3-week group, gemcitabine (1250 mg/m2) was administered via intravenous infusion over approximately 30 minutes on day 1 and day 8 of a 21 day cycle. In the 4-week group, gemcitabine (1000 mg/m2) was administered via intravenous infusion over approximately 30 minutes on day 1, day 8 and day 15 of a 28-day cycle. In each group, cisplatin (75 mg/m2) was administered on day 1 via intravenous infusion over approximately 60 to 90 minutes. On day 1, gemcitabine was administered first, followed by cisplatin. Pre- and post-hydration for cisplatin was administered according to local procedure. Patients received up to 6 cycles of treatment. The actual dose of gemcitabine and cisplatin administered to patients was determined at the beginning of each cycle by calculating the body surface area. A ±5% variation in the calculated total dose was allowed for ease of dose administration. Patients received premedication on day 1 with a 5-HT3 antagonist with or without dexamethasone at standard recommended doses, as well as dexamethasone as an antiemetic for the next 24 to 48 hours. On day 8 and/or day 15 of gemcitabine administration, antiemetic therapy was administered according to local practice for patients receiving mildly to moderately emetogenic chemotherapy agents.
To start any cycle, ANC was required to be ≥1.5×109/L, platelets ≥100×109/L, serum creatinine <1.5 × ULN, serum bilirubin ≤2 × ULN and AST/ALT ≤5 × ULN. The treatment of a patient was postponed for up to 4 weeks if he/she had not recovered from haematological and/or nonhaematological toxicity at the beginning of cycle (day 1). The treatment was restarted immediately after recovery; that is, blood tests as above and the absence of grade 3 or 4 nonhaematological toxicities. Patients were excluded from the study if the beginning of a cycle had to be postponed due to toxicity lasting for more than 4 weeks. Dosage adjustments for gemcitabine on day 8 and day 15 were calculated as follows. The dose of gemcitabine was reduced by 25% in the case of: ANC (1.0–1.49) ×109/L and/or platelets (50–99) ×109/L. Gemcitabine therapy was withheld in the case of: ANC <1.0 ×109/L and/or platelets <50 ×109/L and/or febrile fever >38.5°C. Patients with a grade 0, 1, or 2 nonhaematological toxicity received the full dose of therapy on time. For patients who developed grade 3 nonhaematological toxicity (excluding nausea, vomiting, alopecia), the decision of having their gemcitabine therapy reduced to 75% or withheld depended on the course considered most appropriate from a medical perspective, as judged by the physician/investigator. Patients who developed a grade 4 nonhaematological toxicity that was considered life threatening were excluded from the study.
Efficacy analyses were conducted on an intent to treat basis. Safety analyses were conducted on an intent to treat basis on all patients who received at least one dose of study drug. All tests of treatment effects were conducted at a two-sided alpha level of 0.05 unless otherwise stated.
The primary analysis was to compare the objective response rate (ORR) observed in the 3-week group with that observed in the 4-week group. A Fisher's exact test for proportions was used. The ORR was defined as the proportion of patients with an overall best response of partial response or complete response. Exploratory analyses to support the primary analysis of efficacy included logistic regression analyses to adjust for potential confounding variables in the primary analysis (age, investigator, gender, KPS and disease stage). Survival time was calculated from the date of enrolment to the date of death due to any cause (patient's data were censored at the date of last followup). Progression free survival was calculated from the date of enrolment to the date of death due to any cause or determination of progressive disease (patient's data censored at the date of last followup or the commencement of new anticancer therapy). Kaplan-Meier curves were generated and quartiles and point probabilities reported for overall survival and progression free survival. Differences were assessed by log rank test and hazard ratios were calculated from Cox proportion hazard modelling. All statistical analysis was conducted using SAS v8.2 (SAS Institute Inc., USA).
One hundred patients were enrolled in this study: 51 patients were randomized to receive the 3-week schedule (14 females (27%)) and 49 patients were randomized to receive the 4-week schedule (24 females (49%)). Table 1 shows the baseline demographics of the patients at enrolment. All patients were Chinese, aged from 24 to 75 years. The majority of patients had stage IV disease (65%). Of the 100 randomized patients, 25 had a partial response, with an overall response rate of 25% (95% CI: (16.9–34.7)%; Table 2). From the 3-week and 4-week groups, the response rates were 24% (12/51 patients; 95% CI: (12.8–37.5)%) and 27% (13/49 patients; 95% CI: (15.0–41.1)%) respectively, giving an odds ratio (4-week vs 3-week) of 1.17 (95% CI: 0.47–2.90; P=0.82). After adjusting for baseline variables (gender, KPS, disease stage, investigator, age) the odds ratio (4-week vs 3-week) was 0.69 (95% CI: 0.22–2.21; P=0.54). Of the remaining patients, 29% from the 3-week group (15 patients; 95% CI: 17.5—43.8) and 27% from the 4-week group (13 patients; 95% CI: 15.0—41.1), had stable disease.
Response rates were also examined by disease stage, KPS and gender (Table 3). The response rate for patients who had stage IIIB disease at study entry was 23% (8/35 patients), while those with stage IV disease had a response rate of 26% (17/65 patients). Among patients with KPS ≤80, the response rate was 15% (6/39 patients) and among patients with KPS >80, the response rate was 31% (19/61 patients). When examined by gender, the response rates for female patients and male patients were 24% (9/38 patients) and 26% (16/62 patients) respectively. Logistic regression analyses (both unadjusted and adjusted) demonstrated that response to treatment was not statistically significantly dependent on disease stage or gender. Adjusted logistic regression analysis by KPS status demonstrated that there was a statistically significant difference between whether or not patients responded clinically to treatment in favour of those whose KPS was >80 at baseline (adjusted odds ratio: 0.17; 95% CI: 0.04–0.76; P=0.02).
Median survival time was 12.1 months in the 3-week group and 13.8 months in the 4-week group (hazard ratio (HR): 1.19, 95% CI: 0.68–2.10). Median progression free survival in the 3-week group and the 4-week group were 4.9 months and 6.9 months respectively (HR: 1.40, 95% CI: 0.90–2.19). The proportion of patients surviving for at least 1 year was 43% in the 3-week group and 49% in the 4-week group. While there was a trend towards longer survival and progression free survival in the 4-week group, there were no statistically significant differences between the two treatment groups in any of the time to event measures. Kaplan-Meier curves for survival time and progression free survival are shown in Figures 1 and 2. The overall rate of grade 3/4 toxicity in the 3-week group was 55% significantly less compared with 86% in the 4-week group. There were statistically significant differences between the two treatment groups in the number of patients experiencing study drug related grade 3–4 neutropenia, thrombocytopenia and nonhaematological toxicity. Three patients reported grade 1 alopecia during the study (2 patients (4%) in the 3-week group and 1 patient (2%) in the 4-week group). See Table 4.
The planned dose intensities of gemcitabine were 833 mg/m2 for the 3-week group and 750 mg/m2 for the 4-week group. The mean achieved dose intensities of gemcitabine were 760 mg/m2 for the 3-week group and 644 mg/m2 for the 4-week group (P <0.001), giving mean relative dose intensities of 0.913 and 0.859 (P=0.015). The planned dose intensities of cisplatin were 25 mg/m2 and 18.8 mg/m2 for the 3-week and 4-week groups, respectively. The mean achieved dose intensities of cisplatin were 23.7 mg/m2 and 18.1 mg/m2 (P <0.001) for the 3-week and 4-week groups, respectively, giving mean relative dose intensities of 0.946 and 0.964 (P=0.210).
Patients in the 3-week group and those in the 4-week group received an average of 4.1±1.9 and 4.2±1.7 chemotherapy cycles respectively. In the 3-week group, 33 patients (65%) completed 3 treatment cycles and 29 patients (57%) completed 4 treatment cycles. In the 4-week group, 36 patients (73%) and 29 patients (59%) completed 3 and 4 cycles of treatment respectively. During the study, 32 patients in the 3-week group (63%) compared with 41 patients in the 4-week group (84%) experienced a delay or reduction in their gemcitabine dose. For cisplatin dosing throughout the study, 15 patients in the 3-week group (29%) compared with 12 patients in the 4-week group (24%) experienced a delay or reduction. There were 89 gemcitabine dose reductions out of a possible 202 at day 15 in the cycle of the 4-week group. Two patients from the 3-week group and 4 patients from the 4-week group received at least one blood transfusion during the study.
Both 3-week and 4-week dosing schedules for gemcitabine/cisplatin have been shown to be effective in the treatment of advanced NSCLC.4–16 The current study is the first randomized trial to compare a 3-week versus a 4-week regimen for gemcitabine/cisplatin in Chinese patients with NSCLC. Both treatment regimens demonstrated activity in the Chinese patient population with an overall response rate of 25%, which is in line with previous studies. Objective response rates were comparable between the two treatment groups, with 12 of the 51 patients (24%) in the 3-week group and 13 of the 49 (27%) patients in the 4-week group achieving a partial response. These are lower than the response rates of 42% and 38% observed in a similar study for the 3-week and 4-week gemcitabine/cisplatin schedules respectively, however the dose of gemcitabine was higher and cisplatin triple those used in the previous study (both groups: gemcitabine 1000 mg/m2; cisplatin 70 mg/m2).16
While the response rates were comparable between the two treatment groups, the odds ratio was in favour of the 4-week group. When response rates were examined by each baseline variable (age, investigator, KPS, gender, disease stage and treatment group) in turn, our results showed that response to treatment was not statistically significantly dependent on disease stage or gender. However, after adjusting for the remaining predictive variables, patients with a KPS >80 at baseline were more likely to achieve responder status than those with a KPS ≤80. This is important to consider when analyzing the results from this study because, at baseline, more patients in the 4-week group had a KPS >80 and fewer patients had a KPS ≤80 when compared with the 3-week group. Indeed, after adjusting for baseline variables (age, investigator, KPS, gender, disease stage), the odds ratio was in favour of the 3-week group. This highlights the significance of patient and disease characteristics when evaluating response rates.
Survival time, median time to disease progression and 1 year survival rates, while comparable between the two treatment groups, favoured the 4-week group. Median survival time in this study was 12.1 months in the 3 week group and 13.8 months in the 4-week group which is comparable to from 8.1 to 15.4 months from previous studies.4–15 Soto and coworkers16 observed median survival times of 367 days for the 3-week regimen and 279 days for the 4-week regimen. An observational study investigating the clinical benefit of a 3-week gemcitabine/cisplatin regimen in Chinese patients yielded a median survival time of 7.8 months.18 It is expected that survival times may be shorter for patients in observational studies compared with those participating in clinical trials because patients in observational studies had no visits or procedures other than what they would receive in routine clinical practice.
The mean number of chemotherapy cycles received per patient was similar between the two treatment groups. From our results, it seems that a smaller proportion of patients in the 3-week group experienced delays or reductions in their gemcitabine dose compared with patients in the 4-week group (63% cf 84%), however this difference was probably due to dose reductions experienced by patients in the 4-week group at day 15. The different timing of study visits for each dosing schedule in the present study would have favoured the 3-week group as patients in this group did not receive a third infusion. The mean relative dose intensity of gemcitabine was statistically significantly higher in the 3-week group compared with the 4-week group due to the dose reductions experienced at day 15 in the 4-week group.
Overall, grade 3/4 toxicities were significantly reduced in the 3-week group compared with the 4-week group. Some specific toxicities that were reduced in the 3-week group compared with the 4-week group were study drug related neutropenia and thrombocytopenia. In Soto and coworkers,16 it was noted that the level of grade 3/4 neutropenia experienced with gemcitabine/cisplatin administration was slightly higher in the 3-week group (27.8% cf 22.5%), while grade 3/4 thrombocytopenia occurred more frequently in the 4-week group (29.5% vs. 5.5%), however these differences did not reach statistical significance.16 Quoix and coworkers17 also observed a reduced prevalence of thrombocytopenia in patients receiving single agent gemcitabine treatment over a 3-week dosing schedule compared with a 4-week schedule and relatively low rates of grade 3/4 toxicity overall in both treatment groups. The reduced incidence of thrombocytopenia observed in the 3-week group compared with the 4-week group in the present study (8% cf 31%) is consistent with the trend in the incidence of severe thrombocytopenia reported above.16,17
A strength of this randomized study is that patient assignment was stratified based on investigator site to minimise treatment and tumour assessment bias. However, there was an uneven assignment of patients to the two treatment groups in terms of KPS and gender. It is known that KPS is an independent prognostic factor, so it is possible that the better performance status of patients in the 4-week group at baseline, compared with the 3-week group, could explain some treatment differences observed in this study. Furthermore, due to the different cycle lengths in each treatment group, tumour assessments were carried out approximately every 6-weeks during the 3-week schedule and approximately every 8-weeks during the 4-week schedule, which might have introduced ascertainment bias into the assessment of response.
In conclusion, this study further supports the use of gemcitabine and cisplatin over either a 3-week or a 4-week dosing schedule for the treatment of NSCLC in Chinese patients. While all efficacy endpoints were numerically superior in the 4-week group, the difference was not statistically significant. The incidence of grade 3/4 toxicities was less in the 3-week group.
We thank Ying Ma and Yi Cao for study implementation support.
1. Yang L, Parkin DM, Ferlay J, Li L, Chen Y. Estimates of cancer incidence in China for 2000 and projections for 2005. Cancer Epidemiol Biomarkers Prev 2005; 14: 243–250.
2. Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997; 111: 1710–1717.
3. Bergman AM, Ruiz Van Haperen VWT, Veerman G, Kuiper CM, Peters GJ. Synergistic interaction between cisplatin and gemcitabine in vitro.
Clin Cancer Res 1996; 2: 521–530.
4. Cardenal F, Lopez-Cabrerizo MP, Anton A, Alberola V, Massuti B, Carrato A, et al. Randomized phase III study of gemcitabine-cisplatin versus etoposide-cisplatin in the treatment of locally advanced or metastatic nonsmall-cell lung cancer. J Clin Oncol 1999; 17: 12–18.
5. Crino L, Scagliotti GV, Ricci S, De Marinis F, Rinaldi M, Gridelli C, et al. Gemcitabine and cisplatin versus mitomycin, ifosfamide and cisplatin in advanced nonsmall-cell lung cancer: A randomized phase III study of the Italian Lung Cancer Project. J Clin Oncol 1999; 17: 3522–3530.
6. Sandler AB, Nemunaitis J, Denham C, Von Pawel J, Cormier Y, Gatzemeier U, et al. Phase III trial of gemcitabine plus cisplatin versus cisplatin alone in patients with locally advanced or metastatic nonsmall-cell lung cancer. J Clin Oncol 2000; 18: 122–130.
7. Schiller JH, Harrington D, Belani CP, Langer C, Sandler A, Krook J, et al. Comparison of four chemotherapy regimens for advanced nonsmall-cell lung cancer. N Engl J Med 2002; 346: 92–98.
8. Anton A, Diaz-Fernandez N, Gonzalez Larriba JL, Vadell C, Masutti B, Montalar J, et al. Phase II trial assessing the combination of gemcitabine and cisplatin in advanced nonsmall cell lung cancer (NSCLC). Lung Cancer 1998; 22: 139–148.
9. Abratt RP, Bezwoda WR, Goedhals L, Hacking DJ. Weekly gemcitabine with monthly cisplatin: Effective chemotherapy for advanced nonsmall-cell lung cancer. J Clin Oncol 1997; 15: 744–749.
10. Comella P. Phase III trial of cisplatin/gemcitabine with or without vinorelbine or paclitaxel in advanced nonsmall cell lung cancer. Semin Oncol 2001; 28 (2 Suppl 7): 7–10.
11. Crino L, Scagliotti G, Marangolo M, Figoli F, Clerici M, De Marinis F, et al. Cisplatin-gemcitabine combination in advanced nonsmall-cell lung cancer: A phase II study. J Clin Oncol 1997; 15: 297–303.
12. Jassem J, Krzakowski M, Roszkowski K, Ramlau R, Slominski JM, Szczesna A, et al. A phase II study of gemcitabine plus cisplatin in patients with advanced nonsmall cell lung cancer: clinical outcomes and quality of life. Lung Cancer 2002; 35: 73–79.
13. Steward WP, Dunlop DJ, Dabouis G, Lacroix H, Talbot D. Phase I/II study of gemcitabine and cisplatin in the treatment of advanced nonsmall cell lung cancer: Preliminary results. Semin Oncol 23 (5 Suppl 10): 43–47.
14. Castellano D, Lianes P, Paz-Ares L, Hidalgo M, Guerra JA, Gomez-Martin C, et al. A phase II study of a novel gemcitabine plus cisplatin regimen administered every three weeks for advanced nonsmall-cell lung cancer. Ann Oncol 1998; 9: 457–459.
15. Rinaldi M, Crino L, Scagliotti GV, Mosconi AM, De Marinis F, Gridelli C, et al. A three week schedule of gemcitabine-cisplatin in advanced nonsmall-cell lung cancer with two different cisplatin dose levels: a phase II randomized trial. Ann Oncol 2000; 11: 1295–1300.
16. Soto Parra H, Cavina R, Latteri F, Sala A, Dambrosio M, Antonelli G, et al. Three week versus four week schedule of cisplatin and gemcitabine: results of a randomized phase II study. Ann Oncol 2002; 13: 1080–1086.
17. Quoix E, Breton JL, Ducolone A, Mennecier B, Depierre A, Lemarie E, et al. First line chemotherapy with gemcitabine in advanced nonsmall cell lung cancer elderly patients: a randomized phase II study of 3 week versus 4 week schedule. Lung Cancer 2005; 47: 405–412.
18. Wang L, Liao ML, Li LY, Wan HY, Xu N, Liu JW, et al. Clinical benefit of gemcitabine plus cisplatin 3 week regimen for patients with advanced nonsmall cell lung cancer: a prospective observational study. Chin Med J 2004; 117: 1607–1610.