Despite advances in clinical management, osteosarcoma and Ewing sarcoma, the two most frequent malignant primary bone tumors at pediatric age, still have a poor prognosis for high-risk patients (i.e., relapsed or metastatic disease). Triggering a TRAIL pro-apoptotic pathway represents a promising therapeutic approach, but previous studies have described resistance mechanisms that could explain the declining interest of such an approach in clinical trials. In this study, eight relevant human cell lines were used to represent the heterogeneity of the response to the TRAIL pro-apoptotic effect in pediatric bone tumors and two cell-derived xenograft models were developed, originating from a sensitive and a resistant cell line. The DR5 agonist antibody AMG655 (Conatumumab) was selected as an example of TRAIL-based therapy. In both TRAIL-sensitive and TRAIL-resistant cell lines, two signaling pathways were activated following AMG655 treatment, the canonical extrinsic apoptotic pathway and a non-apoptotic pathway, involving the recruitment of RIPK1 on the DR5 protein complex, activating both pro-survival and pro-proliferative effectors. However, the resulting balance of these two pathways was different, leading to apoptosis only in sensitive cells. In vivo, AMG655 treatment reduced tumor development of the sensitive model but accelerated tumor growth of the resistant one. We proposed two independent strategies to overcome this issue: (1) a proof-of-concept targeting of RIPK1 by shRNA approach and (2) the use of a novel highly-potent TRAIL-receptor agonist; both shifting the balance in favor of apoptosis. These observations are paving the way to resurrect TRAIL-based therapies in pediatric bone tumors to help predict the response to treatment, and propose a relevant adjuvant strategy for future therapeutic development.
Keywords: Ewing sarcoma; TNF-related apoptosis inducing ligand (TRAIL); TRAIL-based therapies; non-apoptotic pathways; osteosarcoma; pediatric bone tumor; resistance.