(P 415) TRABECTEDIN PLUS OLAPARIB IN ADVANCED LEIOMYOSARCOMA PROGRESSING AFTER STANDARD TREATMENTS FOR ADVANCED DISEASE: A POOLED ANALYSIS OF TOMAS AND TOMAS2 STUDIES FROM THE ITALIAN SARCOMA GROUP.

Objective: Patients with advanced soft tissue sarcomas (STS) still represent an unmet medical need. The combination of trabectedin plus olaparib (T+O) was assessed in two investigator-initiated studies conducted by the Italian Sarcoma Group (ISG): a phase 1b (TOMAS) and a phase 2 randomized trial (TOMAS2). In both studies, the combination of T+O showed activity especially in patients affected by advanced leiomyosarcoma (LMS). In the present work, we pooled LMS patients treated in both trials to deepen into combination activity in this histotype.

Methods: TOMAS was an open-label, phase 1b study that recruited patients with unresectable STS. Following a typical 3+3 design, patients received trabectedin (TR) as 24-h infusion every 3 weeks and olaparib (OL) tablets orally twice daily across 6 dose-levels (TR 0.675-1.3 mg/m2, OL 100-300 mg BID). The selected recommended phase 2 dose (RP2D) was TR 1.1 mg/m2 every 3weeks plus OL tablets 150 mg BID. TOMAS2 was an open-label, phase 2 randomized trial, in which patients with unresectable STS progressing after ≥1 treatment(s) were randomized in a 1:1 ratio to T+O at the RP2D, or TR 1.5 mg/m2 every 3 weeks. In this pooled analysis, only patients with histologically confirmed LMS treated in both studies were included. Patients from the TOMAS study were included if they received a starting dose of TR ≥0.92 mg/m2 and OL ≥150 mg BID. Key endpoints used to assess T+O activity were progression-free survival (PFS), PFS at 6-month (PFS6m), PFS12m, PFS24m, RECIST 1.1 overall response rate (ORR), evaluation of activity in uterine-LMS (uLMS) and non-uterine LMS (stsLMS). Exploratory endpoints: correlation of treatment outcomes with clinical and molecular biomarkers.

Results: 70 patients affected by unresectable LMS were included in this pooled analysis: median age 58 (IQR 53-67), male/female 15/55, uLMS/stsLMS 35/35. Of these patients, 42 received T+O combination while 28 received TR. With a median follow up of 13.5 months (95%CI 11.1-15.9), PFS6m and median PFS (mPFS) were 44% (95%CI 31-62%) and 4.3 months (1.0-7.7) with T+O compared to 38% (23-61%) and 4.0 months (3.1-4.9) with TR (log-rank p=0.23). Notably, PFS12m and PFS24m were 31% (20-49%) and 9% (3-27%) with T+O vs 10% (3-35%) and 0% with TR. ORR was 22% with T+O vs 11% with TR (odds ratio 2.34, 95%CI 0.57-9.58). Median PFS in patients achieving a partial response was 16.1 months (12.9-19.3) vs 8.6 (2.1-15.0) with T+O and TR, respectively (log-rank p=0.006). In stsLMS T+O did not show a significant benefit compared to TR (mPFS 6.1 vs 5.9; PFS6m 50 vs 48%, PFS12m 30 vs 20%; log-rank p=0.935). Conversely, in uLMS we observed a better long-term outcome with the combination: mPFS, PFS6m, PFS12m, and PFS24m were 4.2 months, 38%, 32%, and 21% with T+O vs 3.9 months, 29%, 0% and 0% with TR (log-rank p=0.138), respectively. Furthermore, LMS patients with a genomic instability score (GIS) computed by Myriad algorithm on TruSightTM Oncology 500 Illumina panel above median values or damaging alterations in key genes involved in homologous recombination pathway showed a better outcome with T+O compared to TR: mPFS, PFS6m, PFS12m, and PFS24m were 6.1 months, 52%, 38%, and 19% with T+O vs 4.0 months, 38%, 0% and 0% with TR (p=0.036).

Conclusion: This pooled analysis of LMS patients treated with the combination of T+O in the TOMAS and TOMAS2 trials highlighted a subgroup of patients with uLMS who experienced prolonged benefit from T+O, with about one third of patients free from progression for more than one year. Furthermore, patients with defects in the homologous recombination pathway benefitted from T+O as well. These results are in line with similar studies involving PARP1 inhibitors and chemotherapy in uLMS, supporting the further development of PARP1 inhibitors and DNA-damaging agents combination in the treatment of advanced uLMS.