(P 113) IMMUNOSARC II: A SPANISH SARCOMA GROUP (GEIS) PHASE IB TRIAL OF DOXORUBICIN AND DACARBAZINE PLUS NIVOLUMAB IN FIRST LINE OF ADVANCED LEIOMYOSARCOMA

Objective: The immunogenic cell death (ICD) caused by certain chemotherapy agents consists of molecular changes in tumor dying cells that stimulate immunogenicity and enhance antitumor effects. Doxorubicin is a drug with recognized ICD induction capacity, mainly through calreticulin membrane translocation that elicits immunogenic signals as phagocytosis by dendritic cells. The combination of doxorubicin and pembrolizumab was explored in a phase I/II trial treating patients with metastatic/unresectable sarcomas, achieving Partial Response (PR) in the 22% of patients, stable disease (SD) in the 59%, and progressive disease (PD) in the 19%, with median progression-free survival (mPFS) of 8.1 months, superior to historical controls. We hypothesized that the addition of an anti-PD1 (nivolumab) would increase the antitumor activity of doxorubicin plus dacarbazine, a commonly used upfront polychemotherapy regimen in advanced leiomyosarcoma (LMS) . We present here the phase Ib, cohort 7b of IMMUNOSARC trial.

Methods: Adult patients, with good performance status (ECOG 0-1), naïve of previous anthracycline-containing treatments and with centrally confirmed diagnosis of advanced/metastatic LMS were eligible. Initial dose level 0 (L0) was defined as doxorubicin (DOX) 75 mg/m2/day in 20 min on D1 followed by dacarbazine (DAC) 400 mg/m2/day in 60 min on D1 and 2, plus nivolumab (NIV) 360 mg on D2 after DAC every 21 days with GCSF support. This combo would be given up to 6 courses of 21-day cycles, followed by 1-year NIV maintenance. A -1 dose level (L-1) was defined with the same regimen but with NIV 240 mg. A classic 3+3 phase 1 design was used to determine the recommended phase 2 dose (RP2D) based on dose-limiting toxicities (DLTs), which was the main endpoint, observed during the first 21-day cycle. An expansion of the cohort with RP2D up to 20 evaluable patients was foreseen. Secondary endpoints included overall response rate (ORR), safety profile, translational objectives, among others. A baseline tumoral biopsy and blood sampling baseline and every 6 weeks were mandatory for translational purposes. Levels of expression of HMGB1 were assessed in plasma at week 6 and compared with baseline levels.

Results: Between January 2022 and February 2023, 20 patients (6 Male/14 Female), ECOG 0/1 (15/5), with median age 54 years (range 31-72) were enrolled. All patients were treated with the initial L0 scheme and no DLTs were observed being L0 the RP2D. Grade 3-4 toxicities were neutropenia (20%), anemia (10%), febrile neutropenia, asthenia, alopecia, and GGT increased (5% each). Four patients were not evaluable for efficacy at the moment of the data lock (1 due to uncompliant dosing and in 3 patients the first tumor assessment was not reached yet at the data lock). Of 16 efficacy-evaluable patients, RECIST ORR according to local clinical site assessment was 9 PR (56.3%), 6 SD (37.5%), and 1 PD (6.3%) (2 SD cases showed tumor size reductions >20%). Five patients ended treatment due to progression (4 radiological, 1 clinical) and 15 patients remain under the trial therapy. With a median follow-up of 8 months (2-12), in the 16 evaluable patients, the mPFS was 8.67 months (95% CI 7.96-9.37). In 12 patients, paired blood samples (baseline and week 6) were analyzed for HMGB1 levels. In 10/12 patients HMGB1levels increased at week 6 when compared with baseline. In the univariate analysis, increases in HMGB1levels of at least 11.6% correlated with a better mPFS (8.4 (95% CI 6.1-10.6 vs not reached, p=0.029).

Conclusion: DOX 75 mg/m2/d on D1 followed by DAC 400 mg/m2/d on D1 and 2, plus NIV 360 mg on D2 after DAC Q3W, followed by 1-year NIV maintenance is a feasible and well-tolerated scheme. Clinical activity is encouraging, improving historical efficacy outcomes in first line of advanced LMS, which deserves further testing in phase II trials.