Phase I study of gemcitabine given weekly as a short infusion for non-small cell lung cancer: results and possible immune system-related mechanisms.

Purpose: To define the maximum tolerated dose (MTD) and the nature of the toxicities associated with gemcitabine given as a short infusion to patients with non-small cell lung cancer (NSCLC). Secondary objectives were to monitor immunologic response, clinical response, and survival.
Patients and Methods: Thirty-two patients diagnosed with advanced inoperable NSCLC and performance status of 0 or 1 participated in this study. Patients consisted of 22 males and 10 females whose median age was 62 years (range 32-79). Gemcitabine was administered as a 30 min infusion once weekly for 3 weeks followed by 1 week of rest. Patients were enrolled at six gemcitabine dose levels ranging from 1000 to 3500 mg/m2. Patients completed a median of four cycles (range 1-17). Responses were evaluated after every two cycles.
Results: Toxicity was evaluated in all 32 patients. The MTD was not reached as gemcitabine was well tolerated at all dose levels. Grade 4 toxicity occurred in three (9%) patients: pulmonary and lymphocytopenia in one patient each, and both neurocortical and cardiac in one patient. Grade 3 toxicity was found in a total of 20 (63%) patients: pulmonary in 10 (31%) patients; pain in 6 (19%) patients; liver toxicity in 6 (19%) patients; leukopenia and lymphocytopenia in 5 (16%) patients each; anemia, nausea, and cardiac toxicity in 3 (9%) patients each; proteinuria and infection in 2 (6%) patients each; and hemorrhage in 1 (3%) patient. Of the 29 patients evaluable for response, seven objective responses were achieved: six at the 2200 mg/m2 dose level and one at the 2800 mg/m2 dose level. The distribution of responses differed significantly by dose (P = 0.0124 by the exact chi-square test for independence). The overall response rate was 24.1% (95% CI, 10.3-43.5%). At 6 h post-infusion, there was a significant increase in spontaneous tumor necrosis factor (TNF) release and stimulated interleukin (IL)-2 production, and significant decreases in total white blood cell and lymphocyte counts (CD3+, CD8+, and CD16+ lymphocytes) and resting and stimulated superoxide production by formyl-methionyl-leucyl-phenylalanine (fMLP), phorbol myristate acetate, and opsonized zymosan (OPS-Z). At 24 h post-infusion, there were significant decreases in total lymphocyte count, lymphocyte subsets (CD3+, CD4-, CD8+, CD56+, CD19+), and in resting and stimulated superoxide production by fMLP and OPS-Z. There also appeared to be an association between the levels of spontaneous TNF release and the severity of both gastrointestinal (GI) and pulmonary toxicities.
Conclusion: Gemcitabine given as a short infusion was well tolerated at the dose levels of 1000-3500 mg/m2. The MTD was not reached. Toxicities appeared to be cumulative with multiple cycles. Gemcitabine appears to have activity against NSCLC. Although there was a differential dose-response rate among dose levels, increasing the gemcitabine dose beyond 2200mg/m2 did not show increased clinical response. Gemcitabine appears to modulate the immune response, which may in turn mediate both response and toxicity, although no statistically significant correlation between immune and clinical response was detected.