Genomic characterization and development of therapies for pediatric sarcoma

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

Project Summary 1. Efforts in Rhabdomyosarcoma Rhabdomyosarcoma (RMS) is a myogenic cancer that is the most common soft tissue sarcoma of childhood. With the development of multimodal chemotherapy regimens, relapse-free survival rates have improved to 70-80% in patients with localized disease, albeit with significant toxicity. Unfortunately, despite aggressive therapy, the 5-year survival rate for patients with metastatic disease remains only 30%. Therapy assignment is currently based on clinicopathologic features and using these criteria, three distinct subgroups of patients can be identified (low, intermediate and high risk). However, many patients fall into the intermediate risk category (which accounts for about 50% of all patients) and have a heterogeneous clinical outcome. This suggests that some of these children could be treated with less aggressive therapy or alternatively should be considered to have more aggressive disease. In an effort to further characterize the genetic events underlying this tumor type, our group in collaboration with the Children's Oncology Group (COG) performed a large sequencing effort using a combination of whole-genome, whole-exome and whole-transcriptome sequencing along with high resolution SNP arrays to characterize the landscape of somatic alterations in 147 tumor/normal pairs. Our findings describe a heterogenous group of genetic events appears to drive RMS most notably the PAX 3/7-FOXO1 fusion in the alveolar subtype and mutation of multiple RAS pathway genes in fusion negative tumors including recurrent genetic lesions in 10 cancer consensus genes (NRAS, KRAS, HRAS, PIK3CA, BCOR, TP53, NF1, FGFR4, FBXW7, CTNNB1). While the majority of these mutations appear to be mutually exclusive, a subset of tumors appears to have coexisting lesions within the same tumor; perhaps indicating a biologically relevant progression in these tumors. Unfortunately, much of the clinical annotation for these cases was incomplete, severely limiting our ability to derive prognostic information from this data set. To overcome this, we performed a more focused retrospective analysis is needed to determine the prognostic significance of the discovered mutations. In collaboration with the COG and the Shipley group in the United Kingdom, we sequenced 675 clinically annotated rhabdomyosarcoma cases using a clinically relevant capture-based sequencing assay. We have validated that our assay can reliably detect mutations even at low frequency and that we can accurately detect gene copy number changes. In addition, we developed a bioinformatics pipeline and visualization portal that enables a user-friendly interface with the generated data. Clinical outcome correlation analysis was performed in concert with the COG and we have demonstrated the clinical impact of the presence or absence of each genetic lesion. This year we published the initial results of this work (Shern et al 2021) and report that mutations of MYOD1 and TP53 were found to be associated with a worse clinical outcome in fusion negative rhabdomyosarcoma. These results were presented at the American Society of Clinical Oncology Annual meeting as well as the Childrens Oncology Group annual meeting. Current efforts are focused on using these results to enlighten prospective clinical trials. This includes implementation of a CLIA certified assay and informatics pipeline to centrally profile all rhabdomyosarcoma patients enrolled on Children's Oncology Group trials. This effort is being coordinated with the molecular profiling protocol associated with the Childhood Cancer Data Initiative (CCDI) and the COG Project:EveryChild protocol. In addition to these efforts, this year my group continued to generate molecular data associated with patient tumor samples using retrospective samples collected on intermediate and high-risk clinical trials. A current focus of our efforts is to understand the genomic differences between rhabdomyosarcoma that occurs in the adolescent and young adult populations when compared to younger patients. In this work, we are attempting to understand if there are molecular differences between these groups that would explain the observed survival differences between older and younger patients. In addition, we are working to understand the genomic changes that occur in the tumor cell that underlie metastatic and refractory disease. 2. Efforts in Malignant Peripheral Nerve Sheath Tumor (MPNST) The goal of this work is to provide further understanding of the genetic, epigenetic and transcriptomic mechanisms of the molecular oncogenesis underlying the transformation of a Plexiform Neurofibroma to a Malignant Peripheral Nerve Sheath Tumor (MPNST). MPNSTs are a devastating sarcoma that frequently occurs in patients with Neurofibromatosis Type 1 (NF1). Leveraging the expertise and patient population within the Pediatric Oncology Branch, in this project we are using preclinical model systems to dissect the genetic and epigenetic changes that occur as NF1 tumors transition from benign precursors to the aggressive MPNST. Specifically, we are studying the effects of mutation or deletion in one of the polycomb complex (PRC2) members SUZ12 or EED as the final step in the transformation to MPNST. The PRC2 complex is a major transcriptional repressor within the cell and recurrent alterations of members of this complex have been discovered in genomic sequencing studies from patient samples. Within the current fiscal year, we used inducible in vitro model systems and performed ChIPseq, RNAseq and ATACseq to nominate genes whose expression is altered upon loss of these genes with the primary objective to identify genes downstream of the PRC2 loss that are potential therapeutic targets. Not surprisingly, there are hundreds of genes whose expression is altered upon reassembly of the PRC2 complex and includes groups of genes whose expression decreases and those whose expression increases. Pathway analysis of this group of genes demonstrated a marked enrichment for genes involved in PI3K signaling, WNT signaling and cell cycle. Interestingly, the transcriptional changes driven by induction of competent PRC2 imperfectly overlapped with gain of the H3K27me3 mark, whereby only a distinct subset of genes appeared to be directly regulated at the RNA level, by the gain of the repressive mark. Analysis of this interesting group of 145 genes discovered a remarkable enrichment of bivalent genes as the primary downstream targets of PRC2 whose expression changes in response to reintroduction of the competent SUZ12. Of interest are several lineage specific transcription factors that appear to be unique vulnerabilities in MPNST cells. We are currently further interrogating these genes for functional relevance to MPNST using biochemical, molecular biology and CRISPR technologies.