An integrated single-cell RNA-seq map of human neuroblastoma tumors and preclinical models uncovers divergent mesenchymal-like gene expression programs.

BACKGROUND: Neuroblastoma is a common pediatric cancer, where preclinical studies suggest that a mesenchymal-like gene expression program contributes to chemotherapy resistance. However, clinical outcomes remain poor, implying we need a better understanding of the relationship between patient tumor heterogeneity and preclinical models. RESULTS: Here, we generate single-cell RNA-seq maps of neuroblastoma cell lines, patient-derived xenograft models (PDX), and a genetically engineered mouse model (GEMM). We develop an unsupervised machine learning approach ("automatic consensus nonnegative matrix factorization" (acNMF)) to compare the gene expression programs found in preclinical models to a large cohort of patient tumors. We confirm a weakly expressed, mesenchymal-like program in otherwise adrenergic cancer cells in some pre-treated high-risk patient tumors, but this appears distinct from the presumptive drug-resistance mesenchymal programs evident in cell lines. Surprisingly, however, this weak-mesenchymal-like program is maintained in PDX and could be chemotherapy-induced in our GEMM after only 24 h, suggesting an uncharacterized therapy-escape mechanism. CONCLUSIONS: Collectively, our findings improve the understanding of how neuroblastoma patient tumor heterogeneity is reflected in preclinical models, provides a comprehensive integrated resource, and a generalizable set of computational methodologies for the joint analysis of clinical and pre-clinical single-cell RNA-seq datasets.

Pediatric Erythroid Sarcoma Diagnostically Confirmed by Identification of a Recurrent NFIA::CBFA2T3 Fusion.

Erythroid sarcoma (ES) is exceedingly rare in the pediatric population with only a handful of reports of de novo cases, mostly occurring in the central nervous system (CNS) or orbit. It is clinically and pathologically challenging and can masquerade as a nonhematopoietic small round blue cell tumor. Clinical presentation of ES without bone marrow involvement makes diagnosis particularly difficult. We describe a 22-month-old female with ES who presented with a 2-cm mass involving the left parotid region and CNS. The presence of crush/fixation artifact from the initial biopsy made definitive classification of this highly proliferative and malignant neoplasm challenging despite an extensive immunohistochemical workup. Molecular studies including RNA-sequencing revealed a NFIA::CBFA2T3 fusion. This fusion has been identified in several cases of de novo acute erythroid leukemia (AEL) and gene expression analysis comparing this case to other AELs revealed a similar transcriptional profile. Given the diagnostically challenging nature of this tumor, clinical RNA-sequencing was essential for establishing a diagnosis.

An integrated single-cell RNA-seq map of human neuroblastoma tumors and preclinical models uncovers divergent mesenchymal-like gene expression programs.

Neuroblastoma is a common pediatric cancer, where preclinical studies suggest that a mesenchymal-like gene expression program contributes to chemotherapy resistance. However, clinical outcomes remain poor, implying we need a better understanding of the relationship between patient tumor heterogeneity and preclinical models. Here, we generated single-cell RNA-seq maps of neuroblastoma cell lines, patient-derived xenograft models (PDX), and a genetically engineered mouse model (GEMM). We developed an unsupervised machine learning approach ('automatic consensus nonnegative matrix factorization' (acNMF)) to compare the gene expression programs found in preclinical models to a large cohort of patient tumors. We confirmed a weakly expressed, mesenchymal-like program in otherwise adrenergic cancer cells in some pre-treated high-risk patient tumors, but this appears distinct from the presumptive drug-resistance mesenchymal programs evident in cell lines. Surprisingly however, this weak-mesenchymal-like program was maintained in PDX and could be chemotherapy-induced in our GEMM after only 24 hours, suggesting an uncharacterized therapy-escape mechanism. Collectively, our findings improve the understanding of how neuroblastoma patient tumor heterogeneity is reflected in preclinical models, provides a comprehensive integrated resource, and a generalizable set of computational methodologies for the joint analysis of clinical and pre-clinical single-cell RNA-seq datasets.

Discovery of immunotherapy targets for pediatric solid and brain tumors by exon-level expression.

Immunotherapy with chimeric antigen receptor T cells for pediatric solid and brain tumors is constrained by available targetable antigens. Cancer-specific exons present a promising reservoir of targets; however, these have not been explored and validated systematically in a pan-cancer fashion. To identify cancer specific exon targets, here we analyze 1532 RNA-seq datasets from 16 types of pediatric solid and brain tumors for comparison with normal tissues using a newly developed workflow. We find 2933 exons in 157 genes encoding proteins of the surfaceome or matrisome with high cancer specificity either at the gene (n = 148) or the alternatively spliced isoform (n = 9) level. Expression of selected alternatively spliced targets, including the EDB domain of fibronectin 1, and gene targets, such as COL11A1, are validated in pediatric patient derived xenograft tumors. We generate T cells expressing chimeric antigen receptors specific for the EDB domain or COL11A1 and demonstrate that these have antitumor activity. The full target list, explorable via an interactive web portal ( https://cseminer.stjude.org/ ), provides a rich resource for developing immunotherapy of pediatric solid and brain tumors using gene or AS targets with high expression specificity in cancer.

Discovery of immunotherapy targets for pediatric solid and brain tumors by exon-level expression.

Immunotherapy with CAR T cells for pediatric solid and brain tumors is constrained by available targetable antigens. Cancer-specific exons (CSE) present a promising reservoir of targets; however, these have not been explored and validated systematically in a pan-cancer fashion. To identify CSE targets, we analyzed 1,532 RNA-seq datasets from 16 types of pediatric solid and brain tumors for comparison with normal tissues using a newly developed workflow. We found 2,933 exons in 157 genes encoding proteins of the surfaceome or matrisome with high cancer specificity either at the gene (n=148) or the alternatively spliced (AS) isoform (n=9) level. Expression of selected AS targets, including the EDB domain of FN1 (EDB), and gene targets, such as COL11A1, were validated in pediatric PDX tumors. We generated CAR T cells specific to EDB or COL11A1 and demonstrated that COL11A1-CAR T-cells have potent antitumor activity. The full target list, explorable via an interactive web portal (https://cseminer.stjude.org/), provides a rich resource for developing immunotherapy of pediatric solid and brain tumors using gene or AS targets with high expression specificity in cancer.

Expanding the Clinical Utility of Targeted RNA Sequencing Panels beyond Gene Fusions to Complex, Intragenic Structural Rearrangements.

Gene fusions are a form of structural rearrangement well established as driver events in pediatric and adult cancers. The identification of such events holds clinical significance in the refinement, prognostication, and provision of treatment in cancer. Structural rearrangements also extend beyond fusions to include intragenic rearrangements, such as internal tandem duplications (ITDs) or exon-level deletions. These intragenic events have been increasingly implicated as cancer-promoting events. However, the detection of intragenic rearrangements may be challenging to resolve bioinformatically with short-read sequencing technologies and therefore may not be routinely assessed in panel-based testing. Within an academic clinical laboratory, over three years, a total of 608 disease-involved samples (522 hematologic malignancy, 86 solid tumors) underwent clinical testing using Anchored Multiplex PCR (AMP)-based RNA sequencing. Hematologic malignancies were evaluated using a custom Pan-Heme 154 gene panel, while solid tumors were assessed using a custom Pan-Solid 115 gene panel. Gene fusions, ITDs, and intragenic deletions were assessed for diagnostic, prognostic, or therapeutic significance. When considering gene fusions alone, we report an overall diagnostic yield of 36% (37% hematologic malignancy, 41% solid tumors). When including intragenic structural rearrangements, the overall diagnostic yield increased to 48% (48% hematologic malignancy, 45% solid tumor). We demonstrate the clinical utility of reporting structural rearrangements, including gene fusions and intragenic structural rearrangements, using an AMP-based RNA sequencing panel.

Genome-wide mapping of cancer dependency genes and genetic modifiers of chemotherapy in high-risk hepatoblastoma.

A lack of relevant genetic models and cell lines hampers our understanding of hepatoblastoma pathogenesis and the development of new therapies for this neoplasm. Here, we report an improved MYC-driven hepatoblastoma-like murine model that recapitulates the pathological features of embryonal type of hepatoblastoma, with transcriptomics resembling the high-risk gene signatures of the human disease. Single-cell RNA-sequencing and spatial transcriptomics identify distinct subpopulations of hepatoblastoma cells. After deriving cell lines from the mouse model, we map cancer dependency genes using CRISPR-Cas9 screening and identify druggable targets shared with human hepatoblastoma (e.g., CDK7, CDK9, PRMT1, PRMT5). Our screen also reveals oncogenes and tumor suppressor genes in hepatoblastoma that engage multiple, druggable cancer signaling pathways. Chemotherapy is critical for human hepatoblastoma treatment. A genetic mapping of doxorubicin response by CRISPR-Cas9 screening identifies modifiers whose loss-of-function synergizes with (e.g., PRKDC) or antagonizes (e.g., apoptosis genes) the effect of chemotherapy. The combination of PRKDC inhibition and doxorubicin-based chemotherapy greatly enhances therapeutic efficacy. These studies provide a set of resources including disease models suitable for identifying and validating potential therapeutic targets in human high-risk hepatoblastoma.

Neonatal osteoblastic tumor with a novel PTBP1::FOSB fusion.

Congenital/neonatal bone neoplasms are extremely rare. We present the case of a patient with a neonatal bone tumor of the fibula that had osteoblastic differentiation and a novel PTBP1::FOSB fusion. FOSB fusions are described in several different tumor types, including osteoid osteoma and osteoblastoma; however, these tumors typically present in the second or third decade of life, with case reports as young as 4 months of age. Our case expands the spectrum of congenital/neonatal bone lesions. The initial radiologic, histologic, and molecular findings supported the decision for close clinical follow-up rather than more aggressive intervention. Since the time of diagnosis, this tumor has undergone radiologic regression without treatment.

Progressive metastatic infantile fibrosarcoma with multiple acquired mutations.

Infantile fibrosarcoma is the most common soft-tissue sarcoma in children under the age of 1 yr and is defined molecularly by NTRK fusion proteins. This tumor is known to be locally invasive; however, although rare, metastases can occur. The NTRK fusion acts as a driver for tumor formation, which can be targeted by first- and second-generation TRK inhibitors. Although NTRK gatekeeper mutations have been well-described as mechanisms of resistance to these agents, alternative pathway mutations are rare. Here, we report the case of a patient with infantile fibrosarcoma treated with chemotherapy and TRK inhibition that developed metastatic, progressive disease with multiple acquired mutations, including TP53, SUFU, and an NTRK F617L gatekeeper mutation. Alterations in pathways of SUFU and TP53 have been widely described in the literature in other tumors; however, not yet in infantile fibrosarcoma. Although most patients have a sustained response to TRK inhibitors, a subset will go on to develop mechanisms of resistance that have implications for clinical management, such as in our patient. We hypothesize this constellation of mutations contributed to the patient's aggressive clinical course. Taken together, we report the first case of infantile fibrosarcoma with ETV6::NTRK3 and acquired SUFU, TP53, and NTRK F617L gatekeeper mutation along with detailed clinical course and management. Our report highlights the importance of genomic profiling in recurrent infantile fibrosarcoma to reveal actionable mutations, such as gatekeeper mutations, that can improve patient outcomes.

EGFR internal tandem duplications in fusion-negative congenital and neonatal spindle cell tumors.

Next-generation sequencing (NGS) assays can sensitively detect somatic variation, and increasingly can enable the identification of complex structural rearrangements. A subset of infantile spindle cell sarcomas, particularly congenital mesoblastic nephromas with classic or mixed histology, have structural rearrangement in the form of internal tandem duplications (ITD) involving EGFR. We performed prospective analysis to identify EGFR ITD through clinical or research studies, as well as retrospective analysis to quantify the frequency of EGFR ITD in pediatric sarcomas. Within our institution, three tumors with EGFR ITD were prospectively identified, all occurring in patients less than 1 year of age at diagnosis, including two renal tumors and one mediastinal soft tissue tumor. These three cases exhibited both cellular and mixed cellular and classic histology. All patients had no evidence of disease progression off therapy, despite incomplete resection. To extend our analysis and quantify the frequency of EGFR ITD in pediatric sarcomas, we retrospectively analyzed a cohort of tumors (n = 90) that were previously negative for clinical RT-PCR-based fusion testing. We identified EGFR ITD in three analyzed cases, all in patients less than 1 year of age (n = 18; 3/18, 17%). Here we expand the spectrum of tumors with EGFR ITD to congenital soft tissue tumors and report an unusual example of an EGFR ITD in a tumor with cellular congenital mesoblastic nephroma histology. We also highlight the importance of appropriate test selection and bioinformatic analysis for identification of this genomic alteration that is unexpectedly common in congenital and infantile spindle cell tumors.

Haploinsufficiency of the lysosomal sialidase NEU1 results in a model of pleomorphic rhabdomyosarcoma in mice.

Rhabdomyosarcoma, the most common pediatric sarcoma, has no effective treatment for the pleomorphic subtype. Still, what triggers transformation into this aggressive phenotype remains poorly understood. Here we used Ptch1+/-/ETV7TG/+/- mice with enhanced incidence of rhabdomyosarcoma to generate a model of pleomorphic rhabdomyosarcoma driven by haploinsufficiency of the lysosomal sialidase neuraminidase 1. These tumors share mostly features of embryonal and some of alveolar rhabdomyosarcoma. Mechanistically, we show that the transforming pathway is increased lysosomal exocytosis downstream of reduced neuraminidase 1, exemplified by the redistribution of the lysosomal associated membrane protein 1 at the plasma membrane of tumor and stromal cells. Here we exploit this unique feature for single cell analysis and define heterogeneous populations of exocytic, only partially differentiated cells that force tumors to pleomorphism and promote a fibrotic microenvironment. These data together with the identification of an adipogenic signature shared by human rhabdomyosarcoma, and likely fueling the tumor's metabolism, make this model of pleomorphic rhabdomyosarcoma ideal for diagnostic and therapeutic studies.

Molecular Heterogeneity in Pediatric Malignant Rhabdoid Tumors in Patients With Multi-Organ Involvement.

Rhabdoid tumors (RTs) of the brain (atypical teratoid/rhabdoid tumor; AT/RT) and extracranial sites (most often the kidney; RTK) are malignant tumors predominantly occurring in children, frequently those with SMARCB1 germline alterations. Here we present data from seven RTs from three pediatric patients who all had multi-organ involvement. The tumors were analyzed using a multimodal molecular approach, which included exome sequencing of tumor and germline comparator and RNA sequencing and DNA array-based methylation profiling of tumors. SMARCB1 germline alterations were identified in all patients and in all tumors. We observed a second hit in SMARCB1 via chr22 loss of heterozygosity. By methylation profiling, all tumors were classified as rhabdoid tumors with a corresponding subclassification within the MYC, TYR, or SHH AT/RT subgroups. Using RNA-seq gene expression clustering, we recapitulated the classification of known AT/RT subgroups. Synchronous brain and kidney tumors from the same patient showed different patterns of either copy number variants, single-nucleotide variants, and/or genome-wide DNA methylation, suggestive of non-clonal origin. Furthermore, we demonstrated that a lung and abdominal metastasis from two patients shared overlapping molecular features with the patient's primary kidney tumor, indicating the likely origin of the metastasis. In addition to the SMARCB1 events, we identified other whole-chromosome events and single-nucleotide variants in tumors, but none were found to be prognostic, diagnostic, or offer therapeutic potential for rhabdoid tumors. While our findings are of biological interest, there may also be clinical value in comprehensive molecular profiling in patients with multiple rhabdoid tumors, particularly given the potential prognostic and therapeutic implications for different rhabdoid tumor subgroups demonstrated in recent clinical trials and other large cohort studies.

A developmentally prometastatic niche to hepatoblastoma in neonatal liver mediated by the Cxcl1/Cxcr2 axis.

BACKGROUND AND AIMS: Hepatoblastoma (HB) is the most common pediatric liver cancer. Its predominant occurrence in very young children led us to investigate whether the neonatal liver provides a protumorigenic niche to HB development. APPROACH AND RESULTS: HB development was compared between orthotopic transplantation models established in postnatal day 5 (P5) and 60 (P60) mice (P5Tx and P60Tx models). Single-cell RNA-sequencing (sc-RNAseq) was performed using tumor and liver tissues from both models and the top candidate cell types and genes identified are investigated for their roles in HB cell growth, migration, and survival. CONCLUSIONS: We found that various HB cell lines including HepG2 cells were consistently and considerably more tumorigenic and metastatic in the P5Tx model than in the P60Tx models. Sc-RNAseq of the P5Tx and P60Tx HepG2 models revealed that the P5Tx tumor was more hypoxic and had a larger number of activated hepatic stellate cells (aHSCs) in the tumor-surrounding liver that express significantly higher levels of Cxcl1 than those from the P60Tx model. We found these differences were developmentally present in normal P5 and P60 liver. We showed that the Cxcl1/Cxcr2 axis mediated HB cell migration and was critical to HB cell survival under hypoxia. Treating HepG2 P60Tx model with recombinant CXCL1 protein induced intrahepatic and pulmonary metastasis and CXCR2 knockout (KO) in HepG2 cells abolished their metastatic potential in the P5Tx model. Lastly, we showed that in tumors from patients with metastatic HB, there was a similar larger population of aHSCs in the tumor-surrounding liver than in localized tumors, and tumor hypoxia was uniquely associated with prognosis of patients with HB among pediatric cancers. We demonstrated that the neonatal liver provides a prometastatic niche to HB development through the Cxcl1/Cxcr2 axis.

KRIT1-positive hyperkeratotic cutaneous capillary venous malformation.

Cerebral cavernous malformations (CCM) may present in sporadic or familial forms, with different cutaneous manifestations including deep blue nodules, capillary malformations, and hyperkeratotic cutaneous capillary venous malformations (HCCVM). We report the case of an infant with a KRIT1-positive HCCVM associated with familial CCM. Moreover, histopathology showed positive immunohistochemical stain with GLUT1, further expanding the differential diagnosis of GLUT1-positive vascular anomalies.

Clinicopathologic Characterization of IgG4-Rich Pediatric Head and Neck Lesions.

CONTEXT.­: Immunoglobulin G4 (IgG4)-related disease is rare but well characterized in adults; however, the clinical and histologic manifestations in children may differ. OBJECTIVE.­: To review the clinical and histologic features of IgG4-rich head and neck lesions in a pediatric population. DESIGN.­: Retrospective search for cases with IgG4 immunohistochemical staining performed at our institution from 2011 to 2019. Review of clinical courses, serology profiles, histologic patterns, and immunohistochemical staining patterns. RESULTS.­: Four pediatric IgG4-rich lesions were identified and showed distinct histologic patterns from adult IgG4-related disease, including absence of pathognomonic findings associated with the latter. One case showed intralesional immunoglobulin light-chain restriction. Clinical review showed serum IgG4 elevation in 2 of 4 cases, presence of additional autoantibody positivity, and a generally benign/treatment-responsive clinical course. CONCLUSIONS.­: Pediatric IgG4-related disease shows distinct clinical, serologic, and histologic features from its adult counterpart. Pediatric IgG4-related disease involving the orbit has unique clinical characteristics, including frequently normal serum IgG4 levels and female predominance. Awareness of and evaluation for these features may improve diagnosis and treatment.

Discovery of clinically relevant fusions in pediatric cancer.

BACKGROUND: Pediatric cancers typically have a distinct genomic landscape when compared to adult cancers and frequently carry somatic gene fusion events that alter gene expression and drive tumorigenesis. Sensitive and specific detection of gene fusions through the analysis of next-generation-based RNA sequencing (RNA-Seq) data is computationally challenging and may be confounded by low tumor cellularity or underlying genomic complexity. Furthermore, numerous computational tools are available to identify fusions from supporting RNA-Seq reads, yet each algorithm demonstrates unique variability in sensitivity and precision, and no clearly superior approach currently exists. To overcome these challenges, we have developed an ensemble fusion calling approach to increase the accuracy of identifying fusions. RESULTS: Our Ensemble Fusion (EnFusion) approach utilizes seven fusion calling algorithms: Arriba, CICERO, FusionMap, FusionCatcher, JAFFA, MapSplice, and STAR-Fusion, which are packaged as a fully automated pipeline using Docker and Amazon Web Services (AWS) serverless technology. This method uses paired end RNA-Seq sequence reads as input, and the output from each algorithm is examined to identify fusions detected by a consensus of at least three algorithms. These consensus fusion results are filtered by comparison to an internal database to remove likely artifactual fusions occurring at high frequencies in our internal cohort, while a "known fusion list" prevents failure to report known pathogenic events. We have employed the EnFusion pipeline on RNA-Seq data from 229 patients with pediatric cancer or blood disorders studied under an IRB-approved protocol. The samples consist of 138 central nervous system tumors, 73 solid tumors, and 18 hematologic malignancies or disorders. The combination of an ensemble fusion-calling pipeline and a knowledge-based filtering strategy identified 67 clinically relevant fusions among our cohort (diagnostic yield of 29.3%), including RBPMS-MET, BCAN-NTRK1, and TRIM22-BRAF fusions. Following clinical confirmation and reporting in the patient's medical record, both known and novel fusions provided medically meaningful information. CONCLUSIONS: The EnFusion pipeline offers a streamlined approach to discover fusions in cancer, at higher levels of sensitivity and accuracy than single algorithm methods. Furthermore, this method accurately identifies driver fusions in pediatric cancer, providing clinical impact by contributing evidence to diagnosis and, when appropriate, indicating targeted therapies.