Résumé : This study evaluated the efficacy and safety of a 3-hour infusion of trabectedin (1.3 mg/m2) once every 3 weeks in hormone receptor–positive, HER-2–negative advanced breast cancer patients according to tumor xeroderma pigmentosum G gene (XPG) mRNA expression levels. Of 44 patients, 4 of 21 with high XPG and 6 of 23 with low XPG experienced progression-free survival at 16 weeks. Tumor XPG mRNA expression was not a predictive marker for trabectedin efficacy. The safety of trabectedin was similar to that shown in other indications. Background: Preclinical and clinical data suggest that xeroderma pigmentosum G gene (XPG) status might predict trabectedin efficacy. This phase 2 study evaluated the efficacy of trabectedin at a dose of 1.3 mg/m2 as a 3-hour intravenous infusion every 3 weeks in hormone receptor–positive, HER-2 (human epidermal growth factor receptor 2)-negative, advanced breast cancer patients according to the tumor level of XPG mRNA expression. Patients and Methods: Patients were stratified into high-XPG (>3) or low-XPG (≤ 3) groups. The primary efficacy end point was progression-free survival (PFS) rate at 16 weeks (PFS4); secondary efficacy end points were overall response rate (ORR), duration of response, PFS, overall survival, and safety of trabectedin in this patient population. Results: Forty-four patients were treated, 21 with high XPG and 23 with low XPG. Four high-XPG and 6 low-XPG patients experienced PFS4; the criterion for further recruitment (> 6 patients experienced PFS4) was thus not met, and recruitment was stopped. One high-XPG patient had a partial response (ORR, 5%). One low-XPG patient had a complete response, and 2 low-XPG patients had partial responses (ORR, 13%). Comparison of efficacy parameters between high-XPG and low-XPG patients showed no statistically significant differences. ORR in the efficacy population was 9.3%, median PFS was 1.9 months, and overall survival was 11.8 months. The safety of trabectedin in breast carcinoma was similar to that shown in other indications. Conclusion: Trabectedin as single agent had limited activity in hormone-positive, HER-2–negative advanced breast cancer. XPG mRNA expression was not predictive of trabectedin efficacy.