Objective: Gemcitabine in low dose in prolonged infusion is a treatment with documented activity against a variety of tumors. We here report the first randomized trial to compare standard brief and low-dose prolonged infusion of gemcitabine.
Patients and methods: Eligible patients had non-small cell lung cancer in stage IIIB (wet) or IV, Karnofsky performance status 100 to 70 (Eastern Cooperative Oncology Group 0-2), measurable disease, were chemonaïve and fulfilled the standard criteria for chemotherapy. In arm A (standard treatment), gemcitabine was given at 1250 mg/m(2) in 20 to 30 minutes and in arm B (prolonged infusion) at 250 mg/m(2) in 6 hours infusion. All patients received gemcitabine on days 1 and 8 and cisplatin at 75 mg/m(2) on day 2 of a 3-week cycle for four cycles, followed by two cycles of gemcitabine as monotherapy.
Results: A total of 249 patients (188 men and 61 women, median age 58 years) were randomized between arm A (125 patients) and arm B (124 patients). Adenocarcinoma (53.9%) was the predominant histologic type; 92% of patients were in stage IV. The two groups were balanced for prognostic factors; however, group A had fewer patients with significant weight loss and no patient with lung cancer as a second malignancy or after radiotherapy for brain metastases. Grade 3 or greater toxicity was rare: anemia in 0.8 and 3.2%, neutropenia in 21.6 and 22.6%, thrombocytopenia in 0 and 1.6%, and nausea/vomiting in 4 and 8.1% for arms A and B, respectively. Alopecia was seen in 54.5% of patients in arm B, as compared with 9.7% in arm A. No patient died of treatment-related toxicity. During cycle 5, 47.7% of patients in arm A and 60.7% in arm B reported improved well-being, as compared with the status before chemotherapy. Patients in arm A had no complete remission, 32.8% partial responses, 48% minimal responses or stable disease, 13.6% progressions, and 5.6% were not evaluable. For arm B, the corresponding figures are as follows: complete remission 0.8%, partial responses 46% (for overall response rate of 46.8%), minimal responses or stable disease 36.3%, progression 12.1%, and not evaluable 4.8%. Median progression-free survival was 5.5 and 6.0 months, median overall survival was 10.1 and 10.0 months, and 1-year survival was 46.6 and 41.1% for arms A and B, respectively. For the 71 patients with squamous carcinoma, arm B seems superior to arm A, as seen by the higher overall response rate (51.3 versus 35.5%), longer median progression-free survival (6.2 versus 4.9 months), and longer median survival (11.3 versus 8.5 months). However, because of the small number of patients, these differences did not reach the level of statistical significance.
Conclusion: In the treatment of advanced non-small cell lung cancer, gemcitabine in low dose in prolonged infusion in combination with cisplatin has low toxicity and has activity comparable with gemcitabine in higher dose in standard brief infusion. Low-dose gemcitabine may be preferred for incurable cancer among economically deprivileged patients. In addition, apparent superior activity against squamous carcinoma opens new perspectives and deserves further research.