17101
Background: G and V have demonstrated clinical efficacy as single agents and in combination in pts with metastatic NSCLC. Methods: This single-institution, prospective, phase II trial evaluated a biweekly combination of G and V in NSCLC pts who had received 1 prior chemotherapeutic regimen and had disease progression. Eligibility criteria included: Age ≥ 18; PS ≤ 2; and adequate organ function. Pts with stable brain metastases were allowed. A Simon two-stage design specified enrollment of 15 evaluable pts in stage I. If 1 response was seen, 20 more pts would be enrolled. The power of this study to detect an increase in response from 10% to 30% was 90% with a two-sided alpha of 5%. G (1200 mg/m
2
IV over 30 min.) was followed by V (30 mg/m
2
IV over 6–10 min.) on days 1 and 15 of each 28 day cycle. Chemotherapy was given for 6 cycles unless disease progression or unacceptable toxicity was seen. Results: From 11/1998 to 10/2000, 15 of 20 pts enrolled (6 males, 9 females) were evaluable for response. Two pts had grade 4 neutropenia, and 1 pt had grade 4 thrombocytopenia. The only non-hematologic grade 3 toxicities were fatigue, phlebitis, and arthralgias. No objective responses were observed, but 11 pts had stable disease for a mean of 6 mos. The median survival time was 9.4 mos (95% CI = 4.2, 14.8), with a median time to progression of 4.2 mos (95% CI = 1.9, 5.6). The one year survival was 47%. Conclusions: While this schedule of GV was well tolerated, it was felt to be inactive. In vitro and pharmacokinetic studies published after the completion of our trial, suggest G followed by V may have antagonistic effects leading to lower dose delivery of both drugs. Our study was the only study of GV in 2
nd
-line NSCLC in the literature (n = 6) without an objective response. Our study, as well, was the only 2
nd
-line study that administered G prior to V. 1
st
-line studies in the literature (n = 12) that administered V prior to G had, on average, a 1.7 month higher median survival (10.0 vs 8.3 mos) which was statistically significant (p-value < 0.001). Based on this data, sequence of GV should be a consideration in future clinical trials. Because of the lack of response, further studies using this drug sequence, dose, and schedule for GV are not recommended.
[Table: see text]