(P 278) CLINICAL MARKERS PREDICTIVE OF TRABECTEDIN RESPONSE IN SOFT TISSUE SARCOMA

Objective: To identify factors predictive of trabectedin response in patients with metastatic soft tissue sarcoma.

Methods: A total number of 93 patients treated with trabectedin at the Karolinska University Hospital Sarcoma Centre between 2012 and 2020 were identified retrospectively. Eligibility criteria for inclusion in the study included age over 18 years and that the patients received at least one cycle of trabectedin. Available patient, biochemical and tumor-related data from medical records were collected. Primary end points were progression-free survival (PFS), defined as the time from the first administration of trabectedin to the date of documented disease progression or death and overall survival (OS), defined as the time from the first administration of trabectedin to the date of death from any cause. Descriptive statistics was performed for the entire study population and Kaplan-Meier and the log-rank test was used for determination of statistical differences in survival curves between groups. Cox regression analyses were performed to evaluate the independent effects of the candidate variables on PFS and OS respectively. All p-values were 2-sided and the significance level set to α=0.05. As this was an explorative study, no adjustment for multiple testing has been performed. All statistical analyses were performed with IBM SPSS Statistics 25 software.

Results: Clinical outcomes
The median PFS and OS in the entire study population was 4.4 months (95% confidence interval (CI) 2.3-6.6) and 16.2 months (95% CI 12.3-20.1) respectively.

Predictive factors of PFS and OS
Five of the explored variables were significantly associated (p < 0.05) with either PFS and/or OS; histological subtype, hemoglobin, albumin, white blood cell count (WBCC) and absolute neutrophil count (ANC).

Progression-free survival
In the univariable analysis of PFS, three of the variables were significantly (p < 0.05) associated with PFS; albumin, WBCC and ANC.

Higher baseline albumin level was associated with better PFS (Hazard Ratio [HR]=0.95; 95% CI 0.90-1.00, p=0.03) whereas higher level of WBCC as well as higher level of ANC were associated with worse PFS (HR=1.05; 95% CI 1.00-1.10, p=0.04) and (HR=1.06; 95% CI 1.01-1.11, p=0.01) respectively.

Median PFS for albumin < 35 (g/L) versus albumin ≥35 (g/L) was 2.5 months (95% CI 0.8-4.2) versus 5.8 months (95% CI 3.3-8.3) (log rank test, p=0.02). Median PFS for ANC ≤7.5 (10^9 cells/L) versus ANC >7.5 (10^9 cells/L) was 5.4 months (95% CI 3.1-7.7) versus 1.8 months (95% CI 1.6-2.0) (log rank test, p=0.02).

Overall survival
In the univariable analysis of OS, the same baseline biochemical factors as in PFS analysis were associated with OS, but with the addition of hemoglobin also being associated. Furthermore, histological subtype was associated with OS.

Other histological subtype versus L-sarcoma was associated with worse OS (HR=1.79; 95% CI 1.09-2.95, p=0.02). Higher baseline level of hemoglobin was associated with better OS (HR=0.98; 95% CI 0.97-1.00, p=0.04).

Median OS for histological subtype other versus histological subtype L-sarcoma was 12.5 months (95% CI 6.9-18.0) versus 17.8 months (95% CI 13.1-22.5).

Conclusion: The results from this study support previously published results of L-sarcoma being associated with better outcome (OS) of trabectedin compared to other histological subtypes in patients with metastatic soft-tissue sarcoma. Moreover, we demonstrate that the baseline biochemical factors hemoglobin, albumin, ANC and WBCC can predict response to trabectedin in patients with metastatic soft-tissue sarcoma in terms of PFS and OS. There are hopeful indications on easily collected biochemical baseline data having the potential to serve as decision support for the selection of patients in favor of this treatment.