LB-358

Objectives:RRM1 is the crucial gene for gemcitabine efficacy, and ERCC1 is important for platinum-induced DNA adduct repair. We hypothesized that tumoral RRM1 and ERCC1 expression could be used for selection of a standard, double agent chemotherapy regimen for patients with NSCLC, and that this approach would result in promising response rates and survival.

Methodology: A prospective phase II clinical trial was conducted. A dedicated biopsy was required for collection of fresh tumor specimens. They were instantly frozen in liquid nitrogen. Tumor cells were collected by laser-capture microdissection for determination of RRM1 and ERCC1 gene expression by real-time quantitative RTPCR. Predetermined RRM1 and ERCC1 levels were used to determine if gemcitabine (G) and carboplatin (C) would be used. This approach resulted in four treatment regimens: GC for low RRM1/low ERCC1; GD (D, docetaxel) for low RRM1/high ERCC1; DC for high RRM1/low ERCC1; and DV (V, vinorelbine) for high RRM1/high ERCC1. Eligibility included pathologically confirmed stage IV and wet IIIB NSCLC; PS 0-1; no prior systemic therapy for NSCLC; age >18; adequate organ function; and measurable or evaluable disease by RECIST. Disease response and patient survival were assessed.
 >Results: Sixty eligible patients were enrolled from 2/04 to 12/05, and a data lock was performed in 12/06. Tumor specimens were collected in all patients, and one minor and insignificant complication occurred. Expression analysis was successful in 55 patients. RRM1 expression ranged from 0-1,637, and ERCC1 expression ranged from 1-8,103. The partial response rate was 44% (95% CI 31-59%). Overall survival and progression-free survival were 59% and 14% at 12 months with medians of 13.3 and 6.6 months respectively.
 >Conclusion: Selection of chemotherapy based on tumoral RRM1 and ERCC1 expression is feasible and promising for improvement in outcome of patients with NSCLC.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA