Histopathology Based AI Model Predicts Anti-Angiogenic Therapy Response in Renal Cancer Clinical Trial.

Background: Predictive biomarkers of treatment response are lacking for metastatic clearcell renal cell carcinoma (ccRCC), a tumor type that is treated with angiogenesis inhibitors, immune checkpoint inhibitors, mTOR inhibitors and a HIF2 inhibitor. The Angioscore, an RNA-based quantification of angiogenesis, is arguably the best candidate to predict anti-angiogenic (AA) response. However, the clinical adoption of transcriptomic assays faces several challenges including standardization, time delay, and high cost. Further, ccRCC tumors are highly heterogenous, and sampling multiple areas for sequencing is impractical. Approach: Here we present a novel deep learning (DL) approach to predict the Angioscore from ubiquitous histopathology slides. In order to overcome the lack of interpretability, one of the biggest limitations of typical DL models, our model produces a visual vascular network which is the basis of the model's prediction. To test its reliability, we applied this model to multiple cohorts including a clinical trial dataset. Results: Our model accurately predicts the RNA-based Angioscore on multiple independent cohorts (spearman correlations of 0.77 and 0.73). Further, the predictions help unravel meaningful biology such as association of angiogenesis with grade, stage, and driver mutation status. Finally, we find our model is able to predict response to AA therapy, in both a real-world cohort and the IMmotion150 clinical trial. The predictive power of our model vastly exceeds that of CD31, a marker of vasculature, and nearly rivals the performance (c-index 0.66 vs 0.67) of the ground truth RNA-based Angioscore at a fraction of the cost. Conclusion: By providing a robust yet interpretable prediction of the Angioscore from histopathology slides alone, our approach offers insights into angiogenesis biology and AA treatment response.

Molecular underpinnings of dedifferentiation and aggressiveness in chromophobe renal cell carcinoma.

Sarcomatoid dedifferentiation is common to multiple renal cell carcinoma (RCC) subtypes, including chromophobe RCC (ChRCC), and is associated with increased aggressiveness, resistance to targeted therapies, and heightened sensitivity to immunotherapy. To study ChRCC dedifferentiation, we performed multiregion integrated paired pathological and genomic analyses. Interestingly, ChRCC dedifferentiates not only into sarcomatoid but also into anaplastic and glandular subtypes, which are similarly associated with increased aggressiveness and metastases. Dedifferentiated ChRCC shows loss of epithelial markers, convergent gene expression, and whole genome duplication from a hypodiploid state characteristic of classic ChRCC. We identified an intermediate state with atypia and increased mitosis but preserved epithelial markers. Our data suggest that dedifferentiation is initiated by hemizygous mutation of TP53, which can be observed in differentiated areas, as well as mutation of PTEN. Notably, these mutations become homozygous with duplication of preexisting monosomes (i.e., chromosomes 17 and 10), which characterizes the transition to dedifferentiated ChRCC. Serving as potential biomarkers, dedifferentiated areas become accentuated by mTORC1 activation (phospho-S6) and p53 stabilization. Notably, dedifferentiated ChRCC share gene enrichment and pathway activation features with other sarcomatoid RCC, suggesting convergent evolutionary trajectories. This study expands our understanding of aggressive ChRCC, provides insight into molecular mechanisms of tumor progression, and informs pathologic classification and diagnostics.

Melanoma antigens in pediatric medulloblastoma contribute to tumor heterogeneity and species-specificity of group 3 tumors.

Background: Medulloblastoma (MB) is the most malignant childhood brain cancer. Group 3 MB subtype accounts for about 25% of MB diagnoses and is associated with the most unfavorable outcomes. Herein, we report that more than half of group 3 MB tumors express melanoma antigens (MAGEs), which are potential prognostic and therapeutic markers. MAGEs are tumor antigens, expressed in several types of adult cancers and associated with poorer prognosis and therapy resistance; however, their expression in pediatric cancers is mostly unknown. The aim of this study was to determine whether MAGEs are activated in pediatric MB. Methods: To determine MAGE frequency in pediatric MB, we obtained formalin-fixed paraffin-embedded tissue (FFPE) samples of 34 patients, collected between 2008 - 2015, from the Children's Medical Center Dallas pathology archives and applied our validated reverse transcription quantitative PCR (RT-qPCR) assay to measure the relative expression of 23 MAGE cancer-testis antigen genes. To validate our data, we analyzed several published datasets from pediatric MB patients and patient-derived orthotopic xenografts, totaling 860 patients. We then examined how MAGE expression affects the growth and oncogenic potential of medulloblastoma cells by CRISPR-Cas9- and siRNA-mediated gene depletion. Results: Our RT-qPCR analysis suggested that MAGEs were expressed in group 3/4 medulloblastoma. Further mining of bulk and single-cell RNA-sequencing datasets confirmed that 50-75% of group 3 tumors activate a subset of MAGE genes. Depletion of MAGEAs, B2, and Cs alter MB cell survival, viability, and clonogenic growth due to decreased proliferation and increased apoptosis. Conclusions: These results indicate that targeting MAGEs in medulloblastoma may be a potential therapeutic option for group 3 medulloblastomas. Key Points: Several Type I MAGE CTAs are expressed in >60% of group 3 MBs. Type I MAGEs affect MB cell proliferation and apoptosis. MAGEs are potential biomarkers and therapeutic targets for group 3 MBs. Importance of the Study: This study is the first comprehensive analysis of all Type I MAGE CTAs ( MAGEA , -B , and -C subfamily members) in pediatric MBs. Our results show that more than 60% of group 3 MBs express MAGE genes, which are required for the viability and growth of cells in which they are expressed. Collectively, these data provide novel insights into the antigen landscape of pediatric MBs. The activation of MAGE genes in group 3 MBs presents potential stratifying and therapeutic options.

L-2-hydroxyglutarate remodeling of the epigenome and epitranscriptome creates a metabolic vulnerability in kidney cancer models.

Tumor cells are known to undergo considerable metabolic reprogramming to meet their unique demands and drive tumor growth. At the same time, this reprogramming may come at a cost with resultant metabolic vulnerabilities. The small molecule L-2-hdroxyglutarate (L-2HG) is elevated in the most common histology of renal cancer. Similar to other oncometabolites, L-2HG has the potential to profoundly impact gene expression. Here, we demonstrate that L-2HG remodels amino acid metabolism in renal cancer cells through the combined effects on histone methylation and RNA N6-methyladenosine (m6A). The combined effects of L-2HG result in a metabolic liability that renders tumors cells reliant on exogenous serine to support proliferation, redox homeostasis, and tumor growth. In concert with these data, high L-2HG kidney cancers demonstrates reduced expression of multiple serine biosynthetic enzymes. Collectively, our data indicate that high L-2HG renal tumors could be specifically targeted by strategies that limit serine availability to tumors.

Comparative genomics incorporating translocation renal cell carcinoma mouse model reveals molecular mechanisms of tumorigenesis.

Translocation renal cell carcinoma (tRCC) most commonly involves an ASPSCR1-TFE3 fusion, but molecular mechanisms remain elusive and animal models are lacking. Here, we show that human ASPSCR1-TFE3 driven by Pax8-Cre (a credentialed clear cell RCC driver) disrupted nephrogenesis and glomerular development, causing neonatal death, while the clear cell RCC failed driver, Sglt2-Cre, induced aggressive tRCC (as well as alveolar soft part sarcoma) with complete penetrance and short latency. However, in both contexts, ASPSCR1-TFE3 led to characteristic morphological cellular changes, loss of epithelial markers, and an epithelial-mesenchymal transition. Electron microscopy of tRCC tumors showed lysosome expansion, and functional studies revealed simultaneous activation of autophagy and mTORC1 pathways. Comparative genomic analyses encompassing an institutional human tRCC cohort (including a hitherto unreported SFPQ-TFEB fusion) and a variety of tumorgraft models (ASPSCR1-TFE3, PRCC-TFE3, SFPQ-TFE3, RBM10-TFE3, and MALAT1-TFEB) disclosed significant convergence in canonical pathways (cell cycle, lysosome, and mTORC1) and less established pathways such as Myc, E2F, and inflammation (IL-6/JAK/STAT3, interferon-γ, TLR signaling, systemic lupus, etc.). Therapeutic trials (adjusted for human drug exposures) showed antitumor activity of cabozantinib. Overall, this study provides insight into MiT/TFE-driven tumorigenesis, including the cell of origin, and characterizes diverse mouse models available for research.

Osteosarcoma Explorer: A Data Commons With Clinical, Genomic, Protein, and Tissue Imaging Data for Osteosarcoma Research.

PURPOSE: Osteosarcoma research advancement requires enhanced data integration across different modalities and sources. Current osteosarcoma research, encompassing clinical, genomic, protein, and tissue imaging data, is hindered by the siloed landscape of data generation and storage. MATERIALS AND METHODS: Clinical, molecular profiling, and tissue imaging data for 573 patients with pediatric osteosarcoma were collected from four public and institutional sources. A common data model incorporating standardized terminology was created to facilitate the transformation, integration, and load of source data into a relational database. On the basis of this database, a data commons accompanied by a user-friendly web portal was developed, enabling various data exploration and analytics functions. RESULTS: The Osteosarcoma Explorer (OSE) was released to the public in 2021. Leveraging a comprehensive and harmonized data set on the backend, the OSE offers a wide range of functions, including Cohort Discovery, Patient Dashboard, Image Visualization, and Online Analysis. Since its initial release, the OSE has experienced an increasing utilization by the osteosarcoma research community and provided solid, continuous user support. To our knowledge, the OSE is the largest (N = 573) and most comprehensive research data commons for pediatric osteosarcoma, a rare disease. This project demonstrates an effective framework for data integration and data commons development that can be readily applied to other projects sharing similar goals. CONCLUSION: The OSE offers an online exploration and analysis platform for integrated clinical, molecular profiling, and tissue imaging data of osteosarcoma. Its underlying data model, database, and web framework support continuous expansion onto new data modalities and sources.

Serum D-2-hydroxyglutarate and the ratio of D-2HG/L-2HG predict IDH mutation in acute myeloid leukemia.

This study investigates whether serum D-2HG (D-2-hydroxyglutarate) produced by the mutated isocitrate dehydrogenase (IDH) can predict IDH mutations in acute myeloid leukemia (AML) at diagnosis. D-2HG and L-2HG are measured by liquid chromatography-tandem mass spectrometry. D-2HG, total 2HG and the D/L ratio (D-2HG/L-2HG) are significantly higher in IDH mutated cases than in IDH wild cases. The optimal cutoff values to predict IDH mutations at 100% sensitivity (specificity 91%-94%) are >588 ng/mL for D-2HG and >2.33 for the D/L ratio. Our study indicates that elevated serum D-2HG and the D/L ratio may serve as noninvasive biomarkers of IDH mutation in AML.

Target-directed microRNA degradation regulates developmental microRNA expression and embryonic growth in mammals.

MicroRNAs (miRNAs) are post-transcriptional regulators of gene expression that play critical roles in development and disease. Target-directed miRNA degradation (TDMD), a pathway in which miRNAs that bind to specialized targets with extensive complementarity are rapidly decayed, has emerged as a potent mechanism of controlling miRNA levels. Nevertheless, the biological role and scope of miRNA regulation by TDMD in mammals remains poorly understood. To address these questions, we generated mice with constitutive or conditional deletion of Zswim8, which encodes an essential TDMD factor. Loss of Zswim8 resulted in developmental defects in the heart and lungs, growth restriction, and perinatal lethality. Small RNA sequencing of embryonic tissues revealed widespread miRNA regulation by TDMD and greatly expanded the known catalog of miRNAs regulated by this pathway. These experiments also uncovered novel features of TDMD-regulated miRNAs, including their enrichment in cotranscribed clusters and examples in which TDMD underlies "arm switching," a phenomenon wherein the dominant strand of a miRNA precursor changes in different tissues or conditions. Importantly, deletion of two miRNAs, miR-322 and miR-503, rescued growth of Zswim8-null embryos, directly implicating the TDMD pathway as a regulator of mammalian body size. These data illuminate the broad landscape and developmental role of TDMD in mammals.

Target-directed microRNA degradation regulates developmental microRNA expression and embryonic growth in mammals.

MicroRNAs (miRNAs) are post-transcriptional regulators of gene expression that play critical roles in development and disease. Target-directed miRNA degradation (TDMD), a pathway in which miRNAs that bind to specialized targets with extensive complementarity are rapidly decayed, has emerged as a potent mechanism of controlling miRNA levels. Nevertheless, the biological role and scope of miRNA regulation by TDMD in mammals remains poorly understood. To address these questions, we generated mice with constitutive or conditional deletion of Zswim8 , which encodes an essential TDMD factor. Loss of Zswim8 resulted in developmental defects in heart and lung, growth restriction, and perinatal lethality. Small RNA sequencing of embryonic tissues revealed widespread miRNA regulation by TDMD and greatly expanded the known catalog of miRNAs regulated by this pathway. These experiments also uncovered novel features of TDMD-regulated miRNAs, including their enrichment in co-transcribed clusters and examples in which TDMD underlies 'arm switching', a phenomenon wherein the dominant strand of a miRNA precursor changes in different tissues or conditions. Importantly, deletion of two miRNAs, miR-322 and miR-503, rescued growth of Zswim8 null embryos, directly implicating the TDMD pathway as a regulator of mammalian body size. These data illuminate the broad landscape and developmental role of TDMD in mammals.

Testicular relapse of acute lymphoblastic leukemia.

Contemporary chemotherapy regimens have led to improved survival and decreased incidence of testicular relapse for pediatric patients with acute lymphoblastic leukemia (ALL). Local therapies to the testes, such as radiotherapy and orchiectomy, are often not necessary given that high-dose chemotherapy agents can overcome the relative blood-testis barrier. However, urologists should be aware of clinical scenarios involving ALL which still warrant testicular biopsy to guide management. Here, we present a case of a 12-year-old boy with high-risk pre-B cell ALL presenting with a testicular relapse and a clinical presentation overlapping with non-infectious epididymo-orchitis.

Biallelic variants in NUDCD2 associated with a multiple malformation syndrome with cholestasis and renal failure.

NudC-like protein 2 (NUDCD2) is a 4-exon protein-coding gene at 5q34. The protein appears to act in concert with other genes regulating cell migration and microtubule extension. Early studies in model organisms show associations with LIS1, HERC2, and cohesin subunits via a co-chaperone function with Heat shock protein 90 (Hsp90). It is a candidate gene for human pathology. We present two unrelated patients with biallelic variants in NUDCD2. Their phenotypes comprise similar dysmorphic facies, midline brain hypoplasia, hypothyroidism, pulmonary and aortic valve stenosis, severe dysfunction of the liver and kidneys, profound hypotonia, and early death. The cellular analysis demonstrates the absence of the NUDCD2 protein in fibroblasts of one patient with biallelic loss-of-function variants. The data suggest that NUDCD2 deficiency causes this recognizable syndrome that has features of a ciliopathy with additional complications.

Ectopic PLAG1 induces muscular dystrophy in the mouse.

Even though various genetic mutations have been identified in muscular dystrophies (MD), there is still a need to understand the biology of MD in the absence of known mutations. Here we reported a new mouse model of MD driven by ectopic expression of PLAG1. This gene encodes a developmentally regulated transcription factor known to be expressed in developing skeletal muscle, and implicated as an oncogene in certain cancers including rhabdomyosarcoma (RMS), an aggressive soft tissue sarcoma composed of myoblast-like cells. By breeding loxP-STOP-loxP-PLAG1 (LSL-PLAG1) mice into the MCK-Cre line, we achieved ectopic PLAG1 expression in cardiac and skeletal muscle. The Cre/PLAG1 mice died before 6 weeks of age with evidence of cardiomyopathy significantly limiting left ventricle fractional shortening. Histology of skeletal muscle revealed dystrophic features, including myofiber necrosis, fiber size variation, frequent centralized nuclei, fatty infiltration, and fibrosis, all of which mimic human MD pathology. QRT-PCR and Western blot revealed modestly decreased Dmd mRNA and dystrophin protein in the dystrophic muscle, and immunofluorescence staining showed decreased dystrophin along the cell membrane. Repression of Dmd by ectopic PLAG1 was confirmed in dystrophic skeletal muscle and various cell culture models. In vitro studies showed that excess IGF2 expression, a transcriptional target of PLAG1, phenocopied PLAG1-mediated down-regulation of dystrophin. In summary, we developed a new mouse model of a lethal MD due to ectopic expression of PLAG1 in heart and skeletal muscle. Our data support the potential contribution of excess IGF2 in this model. Further studying these mice may provide new insights into the pathogenesis of MD and perhaps lead to new treatment strategies.

Integrated genomic analysis reveals aberrations in WNT signaling in germ cell tumors of childhood and adolescence.

Germ cell tumors (GCTs) are neoplasms of the testis, ovary and extragonadal sites that occur in infants, children, adolescents and adults. Post-pubertal (type II) malignant GCTs may present as seminoma, non-seminoma or mixed histologies. In contrast, pre-pubertal (type I) GCTs are limited to (benign) teratoma and (malignant) yolk sac tumor (YST). Epidemiologic and molecular data have shown that pre- and post-pubertal GCTs arise by distinct mechanisms. Dedicated studies of the genomic landscape of type I and II GCT in children and adolescents are lacking. Here we present an integrated genomic analysis of extracranial GCTs across the age spectrum from 0-24 years. Activation of the WNT pathway by somatic mutation, copy-number alteration, and differential promoter methylation is a prominent feature of GCTs in children, adolescents and young adults, and is associated with poor clinical outcomes. Significantly, we find that small molecule WNT inhibitors can suppress GCT cells both in vitro and in vivo. These results highlight the importance of WNT pathway signaling in GCTs across all ages and provide a foundation for future efforts to develop targeted therapies for these cancers.

Randomized Phase III Trial of Ganitumab With Interval-Compressed Chemotherapy for Patients With Newly Diagnosed Metastatic Ewing Sarcoma: A Report From the Children's Oncology Group.

PURPOSE: Monoclonal antibodies directed against insulin-like growth factor-1 receptor (IGF-1R) have shown activity in patients with relapsed Ewing sarcoma. The primary objective of Children's Oncology Group trial AEWS1221 was to determine if the addition of the IGF-1R monoclonal antibody ganitumab to interval-compressed chemotherapy improves event-free survival (EFS) in patients with newly diagnosed metastatic Ewing sarcoma. METHODS: Patients were randomly assigned 1:1 at enrollment to standard arm (interval-compressed vincristine/doxorubicin/cyclophosphamide alternating once every 2 weeks with ifosfamide/etoposide = VDC/IE) or to experimental arm (VDC/IE with ganitumab at cycle starts and as monotherapy once every 3 weeks for 6 months after conventional therapy). A planned sample size of 300 patients was projected to provide 81% power to detect an EFS hazard ratio of 0.67 or smaller for the experimental arm compared with the standard arm with a one-sided α of .025. RESULTS: Two hundred ninety-eight eligible patients enrolled (148 in standard arm; 150 in experimental arm). The 3-year EFS estimates were 37.4% (95% CI, 29.3 to 45.5) for the standard arm and 39.1% (95% CI, 31.3 to 46.7) for the experimental arm (stratified EFS-event hazard ratio for experimental arm 1.00; 95% CI, 0.76 to 1.33; 1-sided, P = .50). The 3-year overall survival estimates were 59.5% (95% CI, 50.8 to 67.3) for the standard arm and 56.7% (95% CI, 48.3 to 64.2) for the experimental arm. More cases of pneumonitis after radiation involving thoracic fields and nominally higher rates of febrile neutropenia and ALT elevation were reported on the experimental arm. CONCLUSION: Ganitumab added to interval-compressed chemotherapy did not significantly reduce the risk of EFS event in patients with newly diagnosed metastatic Ewing sarcoma, with outcomes similar to prior trials without IGF-1R inhibition or interval compression. The addition of ganitumab may be associated with increased toxicity.

VGLL2-NCOA2 leverages developmental programs for pediatric sarcomagenesis.

Clinical sequencing efforts are rapidly identifying sarcoma gene fusions that lack functional validation. An example is the fusion of transcriptional coactivators, VGLL2-NCOA2, found in infantile rhabdomyosarcoma. To delineate VGLL2-NCOA2 tumorigenic mechanisms and identify therapeutic vulnerabilities, we implement a cross-species comparative oncology approach with zebrafish, mouse allograft, and patient samples. We find that VGLL2-NCOA2 is sufficient to generate mesenchymal tumors that display features of immature skeletal muscle and recapitulate the human disease. A subset of VGLL2-NCOA2 zebrafish tumors transcriptionally cluster with embryonic somitogenesis and identify VGLL2-NCOA2 developmental programs, including a RAS family GTPase, ARF6. In VGLL2-NCOA2 zebrafish, mouse, and patient tumors, ARF6 is highly expressed. ARF6 knockout suppresses VGLL2-NCOA2 oncogenic activity in cell culture, and, more broadly, ARF6 is overexpressed in adult and pediatric sarcomas. Our data indicate that VGLL2-NCOA2 is an oncogene that leverages developmental programs for tumorigenesis and that reactivation or persistence of ARF6 could represent a therapeutic opportunity.

Rhabdomyosarcoma With Epithelioid Features And NSD3::FOXO1 Fusion: Evidence For Reconsideration Of Previously Reported FOXO1::FGFR1 Fusion.

Epithelioid rhabdomyosarcoma is a rare rhabdomyosarcoma variant for which no diagnostic recurrent driver genetic events have been identified. Here we report a rapidly progressive and widely metastatic rhabdomyosarcoma with epithelioid features that arose in the thigh of a male infant. Conventional cytogenetics revealed a t(8;13)(p11.2;q14) translocation. Fluorescence in situ hybridization studies showed rearrangement of FOXO1 and amplification of its 3" end, and rearrangement of NSD3 and amplification of its 5` end. Next generation sequencing identified a NSD3::FOXO1 fusion, which is a previously unreported gene fusion. We also review the historic report of a FOXO1::FGFR1 fusion in a solid variant of alveolar rhabdomyosarcoma and propose that NSD3::FOXO1 fusion may have been the more appropriate interpretation of the data presented in that report.

Exon skipping in genes encoding lineage-defining myogenic transcription factors in rhabdomyosarcoma.

Rhabdomyosarcoma (RMS) is a childhood sarcoma composed of myoblast-like cells, which suggests a defect in terminal skeletal muscle differentiation. To explore potential defects in the differentiation program, we searched for mRNA splicing variants in genes encoding transcription factors driving skeletal muscle lineage commitment and differentiation. We studied two RMS cases and identified altered splicing resulting in "skipping" the second of three exons in MYOD1. RNA-Seq data from 42 tumors and additional RMS cell lines revealed exon 2 skipping in both MYOD1 and MYF5 but not in MYF6 or MYOG. Complementary molecular analysis of MYOD1 mRNA found evidence for exon skipping in 5 additional RMS cases. Functional studies showed that so-called MYODΔEx2 protein failed to robustly induce muscle-specific genes, and its ectopic expression conferred a selective advantage in cultured fibroblasts and an RMS xenograft. In summary, we present previously unrecognized exon skipping within MYOD1 and MYF5 in RMS, and we propose that alternative splicing can represent a mechanism to alter the function of these two transcription factors in RMS.

Intratumoral Resolution of Driver Gene Mutation Heterogeneity in Renal Cancer Using Deep Learning.

Intratumoral heterogeneity arising from tumor evolution poses significant challenges biologically and clinically. Dissecting this complexity may benefit from deep learning (DL) algorithms, which can infer molecular features from ubiquitous hematoxylin and eosin (H&E)-stained tissue sections. Although DL algorithms have been developed to predict some driver mutations from H&E images, the ability of these DL algorithms to resolve intratumoral mutation heterogeneity at subclonal spatial resolution is unexplored. Here, we apply DL to a paradigm of intratumoral heterogeneity, clear cell renal cell carcinoma (ccRCC), the most common type of kidney cancer. Matched IHC and H&E images were leveraged to develop DL models for predicting intratumoral genetic heterogeneity of the three most frequently mutated ccRCC genes, BAP1, PBRM1, and SETD2. DL models were generated on a large cohort (N = 1,282) and tested on several independent cohorts, including a TCGA cohort (N = 363 patients) and two tissue microarray (TMA) cohorts (N = 118 and 365 patients). These models were also expanded to a patient-derived xenograft (PDX) TMA, affording analysis of homotopic and heterotopic interactions of tumor and stroma. The status of all three genes could be inferred by DL, with BAP1 showing the highest sensitivity and performance within and across tissue samples (AUC = 0.87-0.89 on holdout). BAP1 results were validated on independent human (AUC = 0.77-0.84) and PDX (AUC = 0.80) cohorts. Finally, BAP1 predictions correlated with clinical outputs such as disease-specific survival. Overall, these data show that DL models can resolve intratumoral heterogeneity in cancer with potential diagnostic, prognostic, and biological implications. SIGNIFICANCE: This work demonstrates the potential for deep learning analysis of histopathologic images to serve as a fast, low-cost method to assess genetic intratumoral heterogeneity. See related commentary by Song et al., p. 2672.

Spindle Cell/Sclerosing Rhabdomyosarcoma With PAX8::PPARG Fusion.

The spindle cell/sclerosing subtype of rhabdomyosarcoma is classified based on genetic features into the three categories of MYOD1-mutated, gene fusion-driven, and a subset without a currently identified genetic driver event. The gene fusion-driven spindle cell/sclerosing rhabdomyosarcomas are heterogenous and characterized by increasing numbers of gene fusions, the most common gene partners being VGLL2, NCOA2, and TFCP2. Here we report a spindle cell/sclerosing rhabdomyosarcoma that arose in the orbit of a 4-year-old male. This tumor harbored a unique PAX8::PPARG fusion. PAX8::PPARG fusions have previously only been described in follicular thyroid carcinoma and follicular variant of papillary thyroid carcinoma. Our report adds to the growing number of gene fusions in spindle cell/sclerosing rhabdomyosarcomas.