DNA repair functionality modulates the clinical outcome of patients with advanced sarcoma treated with trabectedin (ET-743).

9522 Background and Methods: The sensitivity to trabectedin (Yondelis, ET-743) in experimental cancer models correlates with functional transcription-coupled nucleotide excision repair (TC-NER) and with deficient homologous recombination repair (HRR) activity. In order to assess the clinical relevance of this observation we have characterized the mRNA expression levels of ERCC1, XPD, BRCA1 and BRCA2 by RT-PCR in historical tumor samples from 92 sarcoma patients (pts) treated with the agent.,The overall objective response (CR+PR) rate to Yondelis in this group was 9%, which confirms the previous clinical experience with the compound in unselected, pre-treated patients with sarcoma. The Kaplan-Meier estimates for progression-free survival at 6 months (PFS6) was 23% and for median survival 8 months (mo). 26% of pts were alive at 24 mo. Correlative Study: Pts with high expression levels (> median) of ERCC1 had better PFS6 (32% vs 15%, p = 0.07) and better median survival (12 vs 7 mo) as compared to pts with low expression ( ≤ to median). Pts with low expression levels of BRCA1 had better PFS6 (33 vs 11%, p = 0.02) and superior median survival (15 vs 5 mo, p = 0.0003) than pts with high expression. Expression levels of XPD and BRCA2 (58 pts) had no significant impact on PFS and survival. The analysis of co-expression of ERCC1-BRCA1 identified a highly sensitive group of pts, characterized by high ERCC1 and low BRCA1 expression levels, with a PFS6 of 50% (p=0.0003) and median survival of 20.4 mo (p = 0.0005). A Cox regression analysis demonstrated that ERCC1 (HR=0.52, p = 0.02) and BRCA1 (HR=2.73, p = 0.0006) are independent variables for PFS, while BRCA1 (HR= 2.57, p = 0.0005) is also an independent variable for survival.,This exploratory retrospective study supports the hypothesis that the sensitivity to Yondelis in pts with sarcoma correlates with a functional TC-NER and with deficient HRR. Prospective studies in sarcoma and other potentially sensitive tumor types are required to confirm and validate these findings. [Table: see text].