(P 86) RESPONSE TO MIFAMURTIDE AND TUMOR MICROENVIRONMENT: A DECONVOLUTION NANOSTRING ANALYSIS IN LOCALIZED OSTEOSARCOMA PATIENTS TREATED WITHIN THE ISG/OS2 CLINICAL TRIAL

Objective: According to retrospective osteosarcoma series, P-glycoprotein (Pgp) overexpression predicts for poor outcome. A risk-adapted treatment strategy, the ISG/OS-2 trial, evaluated the use of mifamurtide, an EMA-approved innate immunity-modulator, in Pgp-positive patients (pts) (NCT01459484). Here, we present a correlative study to develop a predictive classifier based on tumour immune microenvironment gene profiling.

Methods: 62 localized osteosarcoma pts were enrolled at diagnosis. RNA was extracted from pre-treatment FFPE and non-decalcified tissues and analyzed using PanCancer Immune profiling panel (NanoString Technologies, Seattle, WA,US), which includes 730 immune-genes. 33/62 pts (53%) were Pgp-positive and underwent chemotherapy (CT) and adjuvant mifamurtide, Pgp-negative received CT alone. Univariate Cox regression and ROC curve analysis were conducted to stratify pts and evaluate the model performance. Findings were validated in 2 independent sets. The CIBERSORTx algorithm for deconvolution of NanoString data was used to estimate the proportions of immune-related cell composition.
Primary objective was the identification of prognostic signatures derived from tumour immune microenvironment gene profiling in osteosarcoma pts at diagnosis and in pts undergoing mifamurtide.

Results: No significant differences in terms of overall survival (OS) and event-free survival (EFS) were shown between Pgp-positive and Pgp-negative pts. Therefore, tumour immune microenvironment gene profiles of all 62 pts were analyzed, irrespective of treatment.
First, we identified a 21-gene signature able to stratify all pts into high- and low-risk in terms of OS : 5-year OS for high-risk pts 47%, and 92% for the low-risk (pval = 3e-06, HR=10.60). This 21-signature significantly discriminated between low- and high-risk patients also in terms of EFS (Fig.1a,b).
Next, gene expression profiling data from 62 osteosarcoma cases were subjected to gene deconvolution analysis using CIBERSORT algorithm in order to evaluate the presence and proportions of tumor-infiltrating immune cells.
Of interest, we found that, the tumour immune microenvironment of low-risk vs high-risk pts, was significantly enriched in T-regs, CD8 T-cells and NK activated cells, possibly promoting a better response to therapy (Figure 2).
Robustness of the identified 21-gene signature was confirmed in two independent validation sets: Target-OS TCGA (n=62) and GSE33382 (n=57), demonstrating consistent results despite different molecular platforms were used (Affymetrix Human Genome U133A Array and Illumina Human-6 v2.0). Kaplan-Meier analysis indicated worse OS and EFS in high-risk patients of both validation sets (Figure 3a-d).
Finally, we focused on the subgroup of pts Pgp-positive treated with CT+mifamurtide and identified a 31-gene signature, able to distinguish between high- and low-risk patients in terms of OS (pval = 1e-09, HR=24.51) and EFS (pval = 3e-06, HR= 7.80) (Fig.3a,b).

Conclusion: Prognostic gene signatures derived from the tumor immune microenvironment have been identified for risk stratification of osteosarcoma pts, regardless of mifamurtide treatment. Importantly, we have also developed a mifamurtide-specific signature that predicts both OS and EFS. This promising tool might be used to select pts who could benefit from adjuvant mifamurtide.
CIBERSORT analysis underscored that tumor immune infiltrate significantly affects prognosis in localized patients with osteosarcoma.
CIBERSORT analysis allowed us to dissect the complex immune milieu of localized osteosarcoma and identify tumor-immune interactions possibly implicated in disease progression and treatment response.
Prognostic gene signatures derived from the tumor immune microenvironment have been identified for risk stratification of osteosarcoma pts, regardless of mifamurtide treatment. Importantly, we have also developed a mifamurtide-specific signature that predicts both OS and EFS. This promising tool might be used to select pts who could benefit from adjuvant mifamurtide.
CIBERSORT analysis underscored that tumor immune infiltrate significantly affects prognosis in localized patients with osteosarcoma.
CIBERSORT analysis allowed us to dissect the complex immune milieu of localized osteosarcoma and identify tumor-immune interactions possibly implicated in disease progression and treatment response.

EP, MRS equally contributed
AL, TI equally contributed