Overcoming Neuroblastoma Tumour HETerogeneity, Resistance and RecurrAnCe (ONTHETRRAC)

Neuroblastoma (NB) remains a deadly disease for 60% of high-risk patients. Existing molecular classifiers/drug targets stem from single biopsies of bulky tumors, but this simplified view of NB homogeneity is inappropriate. Intratumor heterogeneity (ITH) must be taken into account for predictive biomarker identification. ONTHETRRAC has assembled an international expert team in NB molecular genetic and clinical research, and build on their recent discoveries that NBs are both spatially and temporally heterogeneous, including intratumor, primary tumor/distant metastasis and diagnosis/relapse comparisons. We will invest cutting-edgegenomicranscriptomic/proteomic/metabolomic/liquid biopsy technologies to develop strategies overcoming ITH-based sampling inaccuracy and support in-depth predictive biomarker analyses. Clonal evolution and disease progression simulated in mouse and zebrafish models will support biomarker validation. ONTHETRRAC aims to provide clinically relevant recommendations for how to molecularly diagnose and monitor disease course on the sub(clonal) level, taking spatial and temporal ITH into account. The resulting diagnostics and monitoring guidelines will be most valuable to guide new targeted therapeutic interventions and will directly be integrated into the next European high-risk and relapse NB trial protocols by partners participating in SIOPEN and GPOH trial planning boards, covering 99% of high-risk NB patients treated in Europe. ONTHETRRAC has the power to generate a more complete picture of spatialemporal ITH, thereby supporting individualized therapeutic strategies targeted to the most aggressive and resistant tumor clone. Minimally invasive techniques will improve the feasibility of monitoring mutations and disease load during treatment. We propose a highly innovative approach for the scientific foundation necessary for a paradigm shift from diagnostic single-biopsy use towards multiple tumor and liquid biopsies for NB clinical care.

Neuroblastoma (NB) remains a deadly disease for 60% of high-risk patients. Therefore, reliable prognostic/predictive factors are needed to more precisely diagnosereat these patients. Existing molecular classifiers/drug targets stem from frequently tiny biopsies of bulky tumors, assuming (wrongly) that the tumor is genetically homogeneous. But this simplified view of NB homogeneity is inappropriate. Intratumor heterogeneity must therefore be taken into account for prognostic/predictive biomarker identification. We applied cutting- edge omics and liquid biopsy techniques to elucidate different aspects of intratumor heterogeneity both spatially and temporally heterogeneity, including intratumor, primary tumor/distant metastasis and diagnosis/relapse comparisons. Genomic data on tumor and corresponding bone marrow samples from 155 neuroblastoma patients support the hypothesis of a branched clonal evolution and a parallel progression of primary and metastatic tumor cells. Therefore, searching for biomarkers to identify the relapse-seeding clone should involve diagnostic disseminated tumor cells (DTCs) alongside the tumor tissue (Abbasi et al. 2017). The genomic data on tumors and corresponding DTCs were supplemented by the first RNA- Seq data on DTCs. RNA-Seq analysis of NB DTCs taken at different time points revealed their unique expression profile in comparison to the tumors and MNCs, and less pronounced differences between diagnostic and relapse DTCs. The latter preferentially affected downregulation of genes encoded by chromosome 19 (frequently tumor suppressor genes). As these alterations might be associated with treatment failure and disease relapse, further functional studies on DTCs should be considered (Rifatbegovic et al. 2018). Amplification of the MYCN oncogene remains the strongest prognostic marker in NB leading to intensive therapy when present, independent of disease stage. However, in a substantial number of NB patients this aberration is only present in a sub-population of tumor cells, indicated as heterogeneous MYCN amplification (hetMNA). By exploring the genomic background of hetMNA tumors, applying a combination of a single cell analysis technique and pan genomic analysis on 99 tumors, we could identify prognostic subgroups completely unknown until then. We therefore concluded that the integration of age, stage and, importantly, the genetic tumor background are crucial for therapy stratification in hetMNA neuroblastoma. These data will now allow a more precise classification of these patients into genetically defined treatment groups (Bogen et al. 2016, Berbegall et al. 2018). Predictive factors to subcategorize high risk NB patients are needed, but so far, are not available. Therefore, we undertook a comprehensive genomic study on 200 uniformly treated neuroblastoma patients enrolled in the SIOPEN HR-NBL1 clinical trial and combined these data with a functional parameter, i.e. bone marrow clearing after induction therapy. This in this tumor entity until then not addressed combination of functional data and genomic markers (e.g. PTPRD, ATRX intragenic deletions, TERT activation, 1q aberrations) resulted in a clear separation into different prognostic/predictive subgroups, allowing a better disease status prediction which can be used in future clinical trials to find a more adequate therapy (Fiedler et al. in submission). To overcome the frequent restrictions in the pediatric oncology setting - i.e. shortage of blood and bone marrow volumes - different DNA isolation techniques, transport conditions, and preservation techniques were compared to establish the most reliable technique for obtaining as much circulating tumor DNA (ctDNA) from small amounts of peripheral blood and bone marrow plasma samples as possible. The technical prerequisites for identifying genomic aberrations in cfDNA were addressed successfully by members of the consortium and the successful application to neuroblastoma cfDNA samples were implemented (Gerber et al. manuscript under review). Importantly, ddPCR data on consecutive blood plasma samples from high risk neuroblastoma patients showed that liquid biopsy techniques allowed early relapse detection weeks before the recurrence was clinically detected (Gerber et al. in preparation). As a consequence, in the next clinical high risk neuroblastoma study, which is supposed to start this year, the meticulous analysis of genomic aberrations of the tumor, applying state-of-the-art techniques, will be complemented by liquid biopsy techniques to learn about disease evolution (regression or progression) during the course of the treatment and facilitate early relapse detection.