Circulating biomarkers of clinical response to combination targeted therapies in neuroblastoma

PROJECT SUMMARY/ABSTRACT Children with advanced neuroblastoma (NBL) have poor outcomes despite aggressive, multimodal therapy and new therapeutic approaches are needed. Recent studies have demonstrated that targeted 131I-MIBG radiopharmaceutical therapy combining modulators of somatic epigenetic programs (vorinostat) and GD2- targeted immunotherapy (dinutuximab) show substantial therapeutic promise. Although 131I-MIBG is one of the most effective therapies for advanced NBL, not all patients will respond to this therapy and predictive biomarkers of treatment response and mechanisms of resistance have not been identified. Our preliminary work in NBL show that minimally invasive liquid biopsy strategies for the detection of circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), and cell-free GD2 levels can be successfully employed in longitudinal samples without reliance on surgical procedures in this vulnerable pediatric population. In this proposal, we will apply these technologies to samples collected from a randomized, phase 2 trial designed to identify the optimal agents for combination with 131I-MIBG in relapsed/refractory NBL (NANT 2021-02 [N21-02]) launching soon. Patients are randomized to either (A) vorinostat + 131I-MIBG; (B) dinutuximab + 131I-MIBG; or (C) vorinostat + dinutuximab + 131I-MIBG. Our overall hypothesis is that circulating biomarker levels will be associated with treatment response for each combination being tested in this trial. We further hypothesize that selection for mutations, changes in chromatin signatures, and downregulation of antigen expression will be associated with disease progression. In Aim 1, we will profile serial plasma samples to detect and quantify ctDNA levels and measure the association between ctDNA levels and response to therapy on each arm of N21-02. Furthermore, we will identify mutations associated with the development of treatment resistance through deep sequencing of ctDNA samples. In Aim 2, we will validate our existing chromatin accessibility and gene expression signatures associated with vorinostat treatment in preclinical models and in vorinostat-treated patient samples. We will then leverage serial liquid biopsy samples from patients enrolled on N21-02 treated with and without vorinostat to detect induction of vorinostat signatures to determine whether these signatures are associated with responses to vorinostat-containing therapies. Finally, in Aim 3, we will longitudinally measure soluble GD2 levels and surface expression levels on CTCs in samples collected from patients on dinutuximab-containing treatment arms and compare GD2 levels to patients not receiving dinutuximab therapy. We will test the association of soluble GD2 levels and response to GD2-directed therapy and determine whether changes in GD2 expression on CTCs are associated with the development of disease progression on therapy. At the end of this award, we expect to identify biomarkers of response and mechanisms of treatment resistance in patients with relapsed NBL. These findings will lead to optimized treatment selection and new candidate therapies to prevent the emergence of resistant disease in patients with NBL.