Correlation Between a New Growth Modulation Index (Gmi)-Based Geistra Score and Efficacy Outcomes in Patients (PTS) with Advanced Soft Tissue Sarcomas (ASTS) Treated with Trabectedin (T): A Spanish Group for Research on Sarcomas (GEIS-38 Study).

11070 Background: The GMI is a marker of drug activity and represents an intra-patient comparison of successive time to progression (TTP), defined as the TTP ratio between the second (or later) line (TTPn) of therapy divided by the prior line (TTPn−1). Defining a clinical profile of pts with GMI >1.33 could help to identify pts who can gain greater benefit from T. Methods: We retrospectively evaluated the concordance between the GMI and the efficacy outcomes and clinical profiles of 198 pts with ASTS treated with trabectedin 1.5 mg/m² (24-h infusion q3w) as a 2nd or further-line treatment from Jan 2007 to Jun 2016. Results: After a median follow-up of 58 months (m; range: 18-172) median overall survival from ASTS diagnosis (MOS) and from T (MT-OS) were 27.5 m (23-32.1) and 10.8 m (8.9-12.7), respectively, while median TTP from T (MT-TTP) and T-1 were 3.4 m (2.8-4) and 3.5 m (2.8-4.2). Overall, 106 pts (53%) had a GMI <1; 22 (11%) a GMI=1-1.33 and 70 pts (35%) a GMI >1.33. A high GMI (<1.33 vs >1.33) correlated with favorable efficacy outcomes: MT-OS: 23 vs 36 m (p<0.001), MT-TTP 2.3 vs 8.2 m (p<0.001) and clinical benefit (objective response + stable disease) 23% vs 68% (p=0.001). The multivariate analysis identified L-type sarcoma (Odds ratio: 1.99, 95% CI 1.06-3.71), metastatic free interval (MFI) from initial diagnosis > 8.1 m (2.24, 95%CI 1.19-4.18) and Karnofsky >80 (2.3, 95%CI 1.00-5.28) as factors independently associated to GMI> 1.33. Based on these 3 variables we defined a new GEISTRA score assigning 1 point for each adversely affected variable: non L-Sarcoma, MFI<8.1m or Karnofsky <80. This score showed a strong correlation with MT-TTP (p<0.001) and MT-OS (p<0.001). Conclusions: Based on the high GMI we defined a new GEISTRA score, which is strongly associated with favorable efficacy outcomes in pts with ASTS treated with T. Thus, GEISTRA score could be a potentially useful predictable clinical tool for T benefit. [Table: see text]