Search

PURPOSE Novel biomarkers are needed to differentiate outcomes in intermediate-risk rhabdomyosarcoma (IR RMS). We sought to evaluate strategies for identifying circulating tumor DNA (ctDNA) in IR RMS and to determine whether ctDNA detection before therapy is associated with outcome. PATIENTS AND METHODS Pretreatment serum and tumor samples were available from 124 patients with newly diagnosed IR RMS from the Children's Oncology Group biorepository, including 75 patients with fusion-negative rhabdomyosarcoma (FN-RMS) and 49 with fusion-positive rhabdomyosarcoma (FP-RMS) disease. We used ultralow passage whole-genome sequencing to detect copy number alterations and a new custom sequencing assay, Rhabdo-Seq, to detect rearrangements and single-nucleotide variants. RESULTS We found that ultralow passage whole-genome sequencing was a method applicable to ctDNA detection in all patients with FN-RMS and that ctDNA was detectable in 13 of 75 serum samples (17%). However, the use of Rhabdo-Seq in FN-RMS samples also identified single-nucleotide variants, such as MYOD1 L122R, previously associated with prognosis. Identification of pathognomonic translocations between PAX3 or PAX7 and FOXO1 by Rhabdo-Seq was the best method for measuring ctDNA in FP-RMS and detected ctDNA in 27 of 49 cases (55%). Patients with FN-RMS with detectable ctDNA at diagnosis had significantly worse outcomes than patients without detectable ctDNA (event-free survival, 33.3% v 68.9%; P = .0028; overall survival, 33.3% v 83.2%; P < .0001) as did patients with FP-RMS (event-free survival, 37% v 70%; P = .045; overall survival, 39.2% v 75%; P = .023). In multivariable analysis, ctDNA was independently associated with worse prognosis in FN-RMS but not in the smaller FP-RMS cohort. CONCLUSION Our study demonstrates that baseline ctDNA detection is feasible and is prognostic in IR RMS.

Purpose: PAX-fusion negative rhabdomyosarcoma (FN RMS) is driven by alterations in the RAS/MAP kinase pathway and is partially responsive to MEK inhibition. Overexpression of IGF1R and its ligands is also observed in FN RMS. Preclinical and clinical studies have suggested that IGF1R is itself an important target in FN RMS. Our previous studies revealed preclinical efficacy of the MEK1/2 inhibitor, trametinib, and an IGF1R inhibitor, BMS-754807, but this combination was not pursued clinically due to intolerability in preclinical murine models. Here, we sought to identify a combination of an MEK1/2 inhibitor and IGF1R inhibitor, which would be tolerated in murine models and effective in both cell line and patient-derived xenograft models of RAS-mutant FN RMS. Experimental Design: Using proliferation and apoptosis assays, we studied the factorial effects of trametinib and ganitumab (AMG 479), a mAb with specificity for human and murine IGF1R, in a panel of RAS-mutant FN RMS cell lines. The molecular mechanism of the observed synergy was determined using conventional and capillary immunoassays. The efficacy and tolerability of trametinib/ganitumab was assessed using a panel of RAS-mutated cell-line and patient-derived RMS xenograft models. Results: Treatment with trametinib and ganitumab resulted in synergistic cellular growth inhibition in all cell lines tested and inhibition of tumor growth in four of six models of RAS-mutant RMS. The combination had little effect on body weight and did not produce thrombocytopenia, neutropenia, or hyperinsulinemia in tumor-bearing SCID beige mice. Mechanistically, ganitumab treatment prevented the phosphorylation of AKT induced by MEK inhibition alone. Therapeutic response to the combination was observed in models without a mutation in the PI3K/PTEN axis. Conclusions: We demonstrate that combined trametinib and ganitumab is effective in a genomically diverse panel of RAS-mutated FN RMS preclinical models. Our data also show that the trametinib/ganitumab combination likely has a favorable tolerability profile. These data support testing this combination in a phase I/II clinical trial for pediatric patients with relapsed or refractory RAS-mutated FN RMS.

Purpose:Half of the patients with high-risk neuroblastoma who receive GD2-targeted mAb do not achieve long-term remissions. Recently, the antibody hu14.18 has been linked to IL2 (hu14.18-IL2) to enhance its efficacy and shown promising preclinical and clinical activity. We developed two new immunocytokines (IC) by linking two other γc cytokines, IL15 and IL21, to hu14.18. The purpose of this study was to compare hu14.18-IL15 and -IL21 with hu14.18-IL2 in their ability to induce antibody-dependent cell-mediated cytotoxicity (ADCC) against neuroblastoma.Experimental Design:We assessed ADCC of hu14.18-IL15 and -IL2 (human cytokines, cross-reactive to mouse) against GD2low and GD2high neuroblastoma cell lines in vitro. T-cell–deficient mice with orthotopic patient-derived xenografts (PDX) and immunocompetent mice with transplantable orthotopic neuroblastoma were used to test all three ICs, including hu14.18-IL21 (murine IL21, not cross-reactive to human). Mechanistic studies were performed using single-cell RNA-sequencing (scRNA-seq).Results:hu14.18-IL15 and hu14.18-IL2 mediated equivalent in vitro ADCC by human NK cells. When combined with chemotherapy, all three ICs similarly controlled the growth of PDXs in nude mice with murine NK effector cells. However, hu14.18-IL15 and -IL21 outperformed hu14.18-IL2 in immunocompetent mice with syngeneic neuroblastoma, inducing complete tumor regressions and extending survival. scRNA-seq data revealed an increase in CD8+ T cells and M1 tumor-associated macrophages and decreased regulatory T cells and myeloid-derived suppressor cells in the tumor microenvironment.Conclusions:Hu14.18-IL15 and Hu14.18-IL21 exhibit robust preclinical activity, warranting further consideration for clinical testing in patients with GD2-expressing neuroblastoma.

PDF file 186K, Loss of heterozygosity on Chromosome 11p15.5

PDF file 92K, Supplementary Legends

Supplemental Methods and Legends

XLSX file 2844K, Supplementary Tables S1-S10

Supplemental Tables

Supplemental Figure 1. PAX3 Fusion in enhancer chromatin states. Supplemental Figure 2. Integration of Topological Domain Boundaries and RNA Expression Datasets Enable Discovery of Novel PAX3-FOXO1 Target Genes. Supplemental Figure 3. MYC and MYCN Expression, Enhancers and Super Enhancers in RMS Cell Lines and Primary Tumors, compared to Normal Skeletal Muscle and Myogenic Samples. Supplemental Figure 4. Genome-wide binding of PAX3-FOXO1, MYOD, MYOG and MYCN and their Functional Interdependence. Supplemental Figure 5. BRD4 at PAX3-FOXO1 Enhancers Predicts Presence of Looping Machinery and Increased Gene Expression. Supplemental Figure 6. Molecular Sensitivity Landscape of PAX3-FOXO1 RMS cells. Supplemental Figure 7. JQ1 potency across RMS cell lines and xenograft cultures. Supplemental Figure 8. Depletion of BET bromodomains in FP-RMS. Supplemental Figure 9. Impact of JQ1 on PAX3-FOXO1 target genes and PAX3-FOXO1 stability. Supplemental Figure 10. Validation of PAX3-FOXO1 Driven ALK SE Luciferase Contruct and In Vivo Efficacy of JQ1.

Pharmacodynamic response to trametinib/ganitumab in xenograft models does not predict tumor shrinkage

Purpose:Rhabdomyosarcoma (RMS) is an aggressive soft-tissue sarcoma, which primarily occurs in children and young adults. We previously reported specific genomic alterations in RMS, which strongly correlated with survival; however, predicting these mutations or high-risk disease at diagnosis remains a significant challenge. In this study, we utilized convolutional neural networks (CNN) to learn histologic features associated with driver mutations and outcome using hematoxylin and eosin (H&E) images of RMS.Experimental Design:Digital whole slide H&E images were collected from clinically annotated diagnostic tumor samples from 321 patients with RMS enrolled in Children's Oncology Group (COG) trials (1998–2017). Patches were extracted and fed into deep learning CNNs to learn features associated with mutations and relative event-free survival risk. The performance of the trained models was evaluated against independent test sample data (n = 136) or holdout test data.Results:The trained CNN could accurately classify alveolar RMS, a high-risk subtype associated with PAX3/7-FOXO1 fusion genes, with an ROC of 0.85 on an independent test dataset. CNN models trained on mutationally-annotated samples identified tumors with RAS pathway with a ROC of 0.67, and high-risk mutations in MYOD1 or TP53 with a ROC of 0.97 and 0.63, respectively. Remarkably, CNN models were superior in predicting event-free and overall survival compared with current molecular-clinical risk stratification.Conclusions:This study demonstrates that high-risk features, including those associated with certain mutations, can be readily identified at diagnosis using deep learning. CNNs are a powerful tool for diagnostic and prognostic prediction of rhabdomyosarcoma, which will be tested in prospective COG clinical trials.

Supplemental Figure S2. Sample partitioning for training and testing a TP53 mutation predictive model using K-fold cross-validation.

Related to Figure 4

Purpose:PAX-fusion negative rhabdomyosarcoma (FN RMS) is driven by alterations in the RAS/MAP kinase pathway and is partially responsive to MEK inhibition. Overexpression of IGF1R and its ligands is also observed in FN RMS. Preclinical and clinical studies have suggested that IGF1R is itself an important target in FN RMS. Our previous studies revealed preclinical efficacy of the MEK1/2 inhibitor, trametinib, and an IGF1R inhibitor, BMS-754807, but this combination was not pursued clinically due to intolerability in preclinical murine models. Here, we sought to identify a combination of an MEK1/2 inhibitor and IGF1R inhibitor, which would be tolerated in murine models and effective in both cell line and patient-derived xenograft models of RAS-mutant FN RMS.Experimental Design:Using proliferation and apoptosis assays, we studied the factorial effects of trametinib and ganitumab (AMG 479), a mAb with specificity for human and murine IGF1R, in a panel of RAS-mutant FN RMS cell lines. The molecular mechanism of the observed synergy was determined using conventional and capillary immunoassays. The efficacy and tolerability of trametinib/ganitumab was assessed using a panel of RAS-mutated cell-line and patient-derived RMS xenograft models.Results:Treatment with trametinib and ganitumab resulted in synergistic cellular growth inhibition in all cell lines tested and inhibition of tumor growth in four of six models of RAS-mutant RMS. The combination had little effect on body weight and did not produce thrombocytopenia, neutropenia, or hyperinsulinemia in tumor-bearing SCID beige mice. Mechanistically, ganitumab treatment prevented the phosphorylation of AKT induced by MEK inhibition alone. Therapeutic response to the combination was observed in models without a mutation in the PI3K/PTEN axis.Conclusions:We demonstrate that combined trametinib and ganitumab is effective in a genomically diverse panel of RAS-mutated FN RMS preclinical models. Our data also show that the trametinib/ganitumab combination likely has a favorable tolerability profile. These data support testing this combination in a phase I/II clinical trial for pediatric patients with relapsed or refractory RAS-mutated FN RMS.

supplementary table legend

Supplemental Description of the Rhabdomyosarcoma Web-based Application

Supplemental Table S1. Whole slide image tissue segmentation statistics by an expert pathologist and probability prediction using a trained convolutional neural network. Supplemental Table S2. Clinical and molecular characteristics of FN-RMS samples used for training models for mutation prediction. Yellow boxes indicate genes included in defining the RAS pathway. Supplemental Table S3. Clinical information with COG risk stratification of FN-RMS samples used for training a prognostication predictive CNN.

Legends for all included Supplementary Figures

Heat map of cell cycle gene expression. Gene expression was normalized and compared between the primary MTC, metastasis one, and metastasis two lesions from the whole transcriptome RNA-seq data. Purple indicates lower and yellow indicates higher expression. Values to generate colors are z-scores across each genes' relative transcript levels [log2(FPKM)].

Using transcriptome data for expression profiling of tumors. Heatmap of transcriptome data allowing for expression profiling and clustering of diagnoses based on the 2000 most highly differentially expressed genes in comparison to the other cohort samples to validate tumor diagnosis. The diagnostic abbreviations are the same as in Figure 1 with the addition of EWLS for Ewing-like sarcoma.

Supplementary Method

Imaging from two clinical vignettes highlighting the importance of sequencing at diagnosis and relapse. A. An epithelioid inflammatory myofibroblastic sarcoma (Patient NCI0244) positive for a RANBP2-ALK fusion leading to activation of ALK was treated with targeted monotherapy. PET scans are shown in the top row. At initial diagnosis, abdominal tumors were located in the left upper quadrant with extensive FDG-avid peritoneal lesions. Clinical response after 8 months of targeted therapy with crizotinib was noted. First relapse with development of a recurrent nodule noted in between the right kidney and duodenum with pulmonary and hepatic metastases was observed after 14 months of crizotinib therapy. A secondary clinical response after treatment with ceritinib was obtained after 1 month of therapy, but was not maintained. The primary tumor sample derived cell line, labeled Tumor.1, did not contain an ALK I1171T mutation. Sanger validation confirmed the presence of an ALK p.I1171T mutation after targeted therapy with crizotinib and ceritinib. The patient's tumor samples all contained a RANBP2-ALK fusion, which was not present in the germline. B. A GNAQ Q209L mutation-positive melanoma (Patient NCI0155), treated with trametinib. The melanoma originated from the left frontotemporal scalp and left orbit, and was metastatic to regional lymph nodes, dura mater, liver, bone and lung as shown on CT (first column, first row) with PET avidity in metastases (first column, second row). WES and WTS of her tumor revealed a GNAQ Q209L mutation. The patient was treated with trametinib and had an initial mixed response, with loss of PET avidity of several liver metastases (second column, second row). However, progressive disease of the primary tumor was noted 5 months after initiating therapy, and PET/CT scans showed increased size of the primary tumor in the scalp (third column, first row)

Long tail distribution of the top 50 altered genes in pediatric sarcoma (S1); Long tail distribution of the top 50 altered genes in pediatric extracranial embryonal tumors (S2); Long tail distribution of the top 50 altered genes in pediatric brain tumors (S3); Long tail distribution of the top 50 altered genes in pediatric heme malignancies (S4); Long tail distribution of the top 50 altered genes in pediatric carcinomas (S5); Long tail distribution of the top 50 altered genes in pediatric gonadal tumors (S6); Known fusions identified in the pediatric cohort (S7); Tileplots showing the pattern of co-occurring alterations with novel MLL3 mutations (S8); Tileplot showing the pattern of co-occurring alterations with a novel PRSS1 mutation (S9); Tileplot showing the pattern of co-occurring alterations with novel DKC1 deletions (S10).

Gene lists for Version 1 of the FoundationOne Assay (S1); Gene lists for Version 2 of the FoundationOne Assay (S2); Gene lists for Version 3 of the FoundationOne Assay (S3); Gene lists for Version 4 of the FoundationOne Assay (S4); Gene lists for Version 5 of the FoundationOne Assay (S5); Distribution of samples across different FoundationOne assays (S6); Filtered variants of unknown significance (S7).

A description of additional supplemental methods

Weight-for-age and height-for-age growth curves. Male patient weights (A), heights (B) and female patient weights (C), and heights (D) were monitored during the follow up period. Patient weight-for-age and height-for-age values were compared to U.S. Center for Disease Control reference growth curves.

Fig S1: Site density cell lines with cytokine analyses of normal healthy donors and patient-derived CD22 CART. Fig S2: Development and evaluation of high-affinity CD22V1 CAR. Fig S3: PKCB2 effects of Bryostatin1 and CD22 modulation with PKC-inhibitors. Fig S4: RNA-seq analysis of Bryostatin1 effects on KOPN8 and SEM Fig S5: Exhaustion markers in Bryostatin1 and CD19 CAR with Bryostatin1 Fig S6: Bioluminescence for Bryostatin1 in vivo experiments Table S1: Enrichment analysis of RNA-seq Bryostatin1-treated cells specific to intrinsic components of the plasma membrane

Supplementary methods to paper

Supplementary Data from Anti-GD2 Antibodies Conjugated to IL15 and IL21 Mediate Potent Antitumor Cytotoxicity against Neuroblastoma

Combined Supplementary Data file containing: Figure S1. Proliferation assays using NCI-737 and NCI-006 in an expanded panel of EWS cell lines. Figure S2. Effects of additional siRNA sequences against LDHA and LDHB in an expanded panel of EWS cell lines. Figure S3. Effects of additional siRNA sequences against EWS-FLI1 on LDHA and LDHB and additional ChIP-seq data. Figure S4. LDH expression, basal activity and in vitro early time course inhibition in EWS cell lines. Figure S5. Steady state levels of pyruvate and lactate and 13C-tracing experiments from 13C-U-glucose. Figure S6. Effects of NCI-737 and NCI-006 on ECAR measurements. Figure S7. Effects of NCI-737 on OCR measurements. Figure S8. Effects of NCI-737 on ECAR measurements of expanded panel of EWS cell lines. Figure S9. Results of oral dosing of LDH inhibitors in EWS xenograft models. Figure S10. Results of IV pilot studies of LDH inhibitors in EWS xenograft models. Figure S11. Additional growth rate and pharmacokinetic data for EWS xenograft models. Figure S12. Additional growth rate and necrosis data for EWS xenograft models. Figure S13. Depiction of of changes in lactate and pyruvate peaks and lactate/pyruvate ratio by hyperpolarized 13C-MR spectroscopy in EWS xenograft models. Figure S14. Measurements of toxicity in animals treated with NCI-737.

Doubling time of calcitonin and CEA for patient #12 was suggestive of progressive disease. Patient #12 experienced best response 434 days after starting vandetanib, with stable lesions in the thyroid bed (showed on T2-weighted MRI of the neck, A) and the mediastinum (showed on CT of the chest, B). In August 2016, the sum of longest diameter (per RECIST) of target lesions was consistent with SD (C,D) (note that not all target lesions are shown). Biomarker response in the two years prior (E) showed CEA doubling time 1.2 years (F) and calcitonin doubling time 1.67 years (G). Patient #12 was confirmed to have PD via RECIST February 2017.

Flowcharts of filtering criteria. A. Flowchart of filtering criteria for germline variants, B. Flowchart of filtering criteria for somatic variants

Purpose: We undertook a multidimensional clinical genomics study of children and adolescent young adults with relapsed and refractory cancers to determine the feasibility of genome-guided precision therapy.Experimental Design: Patients with non-central nervous system solid tumors underwent a combination of whole exome sequencing (WES), whole transcriptome sequencing (WTS), and high-density single-nucleotide polymorphism array analysis of the tumor, with WES of matched germline DNA. Clinically actionable alterations were identified as a reportable germline mutation, a diagnosis change, or a somatic event (including a single nucleotide variant, an indel, an amplification, a deletion, or a fusion gene), which could be targeted with drugs in existing clinical trials or with FDA-approved drugs.Results: Fifty-nine patients in 20 diagnostic categories were enrolled from 2010 to 2014. Ages ranged from 7 months to 25 years old. Seventy-three percent of the patients had prior chemotherapy, and the tumors from these patients with relapsed or refractory cancers had a higher mutational burden than that reported in the literature. Thirty patients (51% of total) had clinically actionable mutations, of which 24 (41%) had a mutation that was currently targetable in a clinical trial setting, 4 patients (7%) had a change in diagnosis, and 7 patients (12%) had a reportable germline mutation.Conclusions: We found a remarkably high number of clinically actionable mutations in 51% of the patients, and 12% with significant germline mutations. We demonstrated the clinical feasibility of next-generation sequencing in a diverse population of relapsed and refractory pediatric solid tumors. Clin Cancer Res; 22(15); 3810–20. ©2016 AACR.

Supplementary Figure from Anti-GD2 Antibodies Conjugated to IL15 and IL21 Mediate Potent Antitumor Cytotoxicity against Neuroblastoma

Purpose: Vandetanib is well-tolerated in patients with advanced medullary thyroid carcinoma (MTC). Long-term outcomes and mechanisms of MTC progression have not been reported previously.Experimental Design: We monitored toxicities and disease status in patients taking vandetanib for hereditary, advanced MTC. Tumor samples were analyzed for molecular mechanisms of disease progression.Results: Seventeen patients [8 male, age 13 (9–17)* years] enrolled; 16 had a RET p.Met918Thr germline mutation. The duration of vandetanib therapy was 6.1 (0.1–9.7+)* years with treatment ongoing in 9 patients. Best response was partial response in 10, stable disease in 6, and progressive disease in one patient. Duration of response was 7.4 (0.6–8.7+)* and 4.9 (0.6–7.8+)* years in patients with PR and SD, respectively. Six patients died 2.0 (0.4–5.7)* years after progression. Median progression-free survival (PFS) was 6.7 years [95% confidence interval (CI): 2.3 years–undefined] and 5-year overall survival (OS) was 88.2% (95% CI: 60.6%–96.9%). Of 16 patients with a RET p.Met918Thr mutation, progression-free survival was 6.7 years (95% CI: 3.1–undefined) and 5-year overall survival was 93.8% (95% CI: 63.2%–99.1%). No patients terminated treatment because of toxicity. DNA sequencing of tissue samples (n = 11) identified an increase in copy number alterations across the genome as a potential mechanism of drug resistance [*median (range)].Conclusions: This study demonstrates that vandetanib is safe and results in sustained responses in children and adolescents with hereditary MTC. Our preliminary molecular data suggest that an increase in copy number abnormalities may be associated with tumor progression in hereditary MTC patients treated with vandetanib. Clin Cancer Res; 24(4); 753–65. ©2017 AACR.

Table S6 shows super-enhancers associated genes identified in siCtrl cells and siHAND2 cells.

Tab 1: Patient Demographics at Enrollment. Tab 2: Details of Multiple Samples from Individual Patients , Tab 3: Sample Data of Patients and Sequencing Data of Each Patient Sample, Tab 4: Sample Depth and Coverage, Tab 5: Somatic SNVs in Tumor Samples, Tab 6: Somatic Indels in Tumor Samples, Tab 7: Germline SNVs and Indels in Normal Samples, and Tab 8: Fusion Genes in Defuse, and Tab 9: Fusion Genes in TopHat