Retinoic acid fails to induce cell cycle arrest with myogenic differentiation in rhabdomyosarcoma

Background: Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in children. Current treatment strategies do not cure most children with recurrent or high-risk disease, underlying the need for novel therapeutic approaches. Retinoic acid has been shown to induce differentiation in a variety of cells including skeletal myoblasts and neuroblasts. In the setting of minimal residual disease, retinoic acid improves survival in neuroblastoma, another poorly differentiated childhood tumor. Whether such an approach is useful for rhabdomyosarcoma has not yet been investigated. Several in vitro studies have demonstrated an appreciable effect of retinoic acid on human RMS cellular proliferation and differentiation. Procedure: We assessed the efficacy of ATRA on rhabdomyosarcoma, in vitro and in vivo, using cell lines and xenografts. Results: ATRA slowed RMS cell proliferation, and promoted a more differentiated myogenic phenotype in both alveolar and embryonal RMS cell lines. Treatment of cultured murine myoblasts with retinoids increased Myogenin expression, but did not induce cell cycle arrest. Despite the favorable in vitro effects, ATRA failed to delay relapse of minimal residual disease using human RMS xenografts in immuno-suppressed NOD-SCID (NSG) mice. Interestingly, tumors that recurred after ATRA treatment showed evidence of enhanced muscle differentiation. Conclusion: Our results indicate that ATRA could increase the expression of some genes associated with muscle differentiation in rhabdomyosarcoma cells, but there was no benefit of single-agent therapy in an MRD model, likely because cell cycle arrest was uncoupled from the pro-differentiation effects of retinoids.