Integrated molecular approach including single-cell FISH enables accurate detection of genetic rearrangements, classification, and identification of novel variants in pediatric brain tumors

Integrated molecular approach including single-cell FISH enables accurate detection of genetic rearrangements, classification, and identification of novel variants in pediatric brain tumors | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Integrated molecular approach including single-cell FISH enables accurate detection of genetic rearrangements, classification, and identification of novel variants in pediatric brain tumors Simone Minasi, Chiara Dossena, Francesca Gianno, Marica Ficorilli, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7244481/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract INTRODUCTION: Pediatric brain tumors (PBTs) are a heterogeneous group of neoplasms. The WHO classification highlights the relevance of genetic and epigenetic alterations in defining molecular subtypes. This study investigated key rearrangements ( MYB , MYBL1 , ZFTA , YAP1 , MN1 ) and fusions ( KIAA1549::BRAF, FGFR1::TACC1, PRKCA::SLC44A1 ), which have diagnostic, prognostic, and therapeutic implications. MATERIALS AND METHODS We applied a novel single-cell fluorescence in situ hybridization (FISH) method in 50 PBTs, comparing results to whole-transcriptome sequencing (16/50) and DNA methylation profiling (23/50). Our FISH-based approach, using signal distance measurements and pattern analysis, aims to objectively distinguish normal from rearranged signals, improving accuracy in detecting structural variants and discriminating ambiguous cases. RESULTS FISH-based signal distance analysis identified rearrangements in MYB (5+/10), MYBL1 (3+/10), ZFTA (7+/12), YAP1 (3+/12), MN1 (4+/8), including novel unbalanced patterns. Fusion detection by pattern analysis revealed KIAA1549::BRAF (8+/10), FGFR1::TACC1 (4+/6), and PRKCA::SLC44A1 (1+/4). High concordance was observed with RNA-seq and methylation profiling. RNA-seq identified novel fusions: MYB::CA10 , YAP1::TYR , and PAX5::ATM . Concurrent structural variants were observed in single tumors, including KIAA1549::BRAF with MN1::BEND2 and ZFTA::RELA with YAP1::TYR , revealing genetic heterogeneity of PBTs. CONCLUSIONS Our single-cell FISH approach provides standardized, reproducible criteria for structural variant detection, minimizing interpretive variability. When combined with transcriptomic and epigenetic data, this approach supports accurate subtype classification and reveals clinically relevant intratumoral heterogeneity in PBTs. Pediatric Brain Tumors Structural Variants Detection Single-Cell FISH Multi-Omics Integration Intratumoral Heterogeneity Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Pediatric brain tumors (PBTs) represent a heterogeneous group of central nervous system (CNS) neoplasms, ranging 0–19 years and encompassing a wide range of pathological entities [ 1 ]. The 2021 WHO Classification of CNS tumors emphasizes the importance of molecular diagnostics by integrating genetic and epigenetic data to define clinically relevant tumor subtypes [ 2 – 7 ]. PBTs frequently harbour somatic structural variants (SVs), including duplications, inversions, translocations, and fusions, which play a critical role in tumor classification and biological behavior [ 2 , 8 – 11 ]. Several rare PBTs are characterized by distinctive genetic rearrangements or fusions, which serve as diagnostic hallmarks. These include alterations involving MYB/MYBL1 , ZFTA/YAP1 , MN1 , KIAA1549::BRAF , FGFR1::TACC1 , and PRKCA::SLC44A1 . MYB / MYBL1 -altered gliomas comprise ~ 2% of pediatric low-grade gliomas (pLGG) and exhibit uniform glial morphology and low mitotic activity [ 19 – 21 ], with fusions involving partners such as PCDHGA1 , MMP16 , and MAML2 [ 22 – 24 ]. Pediatric supratentorial ependymomas (pSTE) are divided into ZFTA - and YAP1 -rearranged subtypes, which differ by age distribution and L1CAM/p65 expression. ZFTA fusions (commonly with RELA ) occur mainly in adolescents, whereas YAP1 rearrangements (e.g., with MAML2 , NCOA1/2 ) are more frequent in younger children [ 2 , 25 – 30 ]. MN1 -rearranged astroblastomas (pABM) typically arise in children and young adults, with a female predominance. Histologically, they exhibit perivascular pseudorosettes and are defined by fusions of MN1 with BEND2 or other partners (e.g., CXXC5 , KMT2A ), though MN1 rearrangements are not exclusive to this tumor type [ 32 – 39 ]. KIAA1549::BRAF fusions, affecting the MAPK/ERK pathway, are present in ~ 80% of pilocytic astrocytomas (PA), the most common pediatric glioma [ 40 – 43 ]. These fusions are also common in diffuse leptomeningeal glioneuronal tumors (DLGNT) (~ 65%), a rare category of tumors with oligodendroglial and neuronal features [ 43 – 46 ]. Extraventricular neurocytomas (EVN), which show mixed neuronal-glial features, frequently harbour FGFR1::TACC1 fusions (~ 70%), which are essential for diagnosis [ 47 – 52 ]. Papillary glioneuronal tumors (PGNT), although extremely rare, are defined by the PRKCA::SLC44A1 fusion [ 53 , 56 ]. Unlike other entities, PGNTs may lack a distinct methylation profile [ 52 , 55 ]. Detection of these SVs in formalin-fixed, paraffin-embedded (FFPE) tissue relies on several methods such as fluorescence in situ hybridization (FISH), real-time/digital PCR, RNA sequencing, and DNA methylation profiling [ 2 , 4 , 5 , 8 – 12 ]. Among these, FISH is considered a gold standard for detecting SVs in FFPE samples due to its high sensitivity and single-cell resolution, enabling correlation with tumor morphology even in presence of significant heterogeneity. However, its limitations include restriction to predefined targets, intra-/inter-observer variability, and lack of standardized interpretation criteria for pediatric brain tumors [ 13 – 16 ]. Conversely, sequencing and array-based platforms offer broader detection of genetic variants but are affected by degraded FFPE nucleic acids and intratumoral heterogeneity, reducing analytical sensitivity and consistency [ 5 – 7 , 16 – 18 ]. This study aims to improve FISH-based diagnostics by introducing a novel single-cell approach that incorporates quantitative signal distance measurement and pattern analysis for the detection of SVs in PBTs. We analyze key rearrangements and fusions (Supplementary Table 1) , validating results with transcriptomic and epigenetic data. Our integrated approach enhances the diagnostic accuracy of FISH and enables discovery of novel or rare genetic variants in PBTs. Materials and Methods Cohort-Samples Selection This study analyzed a cohort of 50 PBTs, representing a range of different diagnostic classes. FFPE samples were obtained as part of Italy’s national centralization program for pediatric brain tumors, coordinated by the Neuropathology Unit (Department of pathological anatomy, Sapienza University of Rome). Clinical and histopathological data, including age at diagnosis, gender, and tumor anatomical location, were documented for all patients and are available in Supplementary Tab. 2. Ethical approval was granted in accordance with the Declaration of Helsinki, and informed consent was obtained from patients' families. Anonymization protocols ensured patient confidentiality. Histopathological diagnoses were established through morphological examination of hematoxylin and eosin (H&E)-stained sections, supported by immunohistochemical (IHC) analysis (see Supplementary Methods for details). Our integrative approach combined morphology, immunophenotype, and molecular genetics to achieve tumor classification and identify novel rearrangements. “Single-Cell Signal Distance Measurement” for Rearrangements Detection using Break-Apart Probes A comprehensive description of the FISH protocol, including probe specifications and analytical workflow, is available in the Supplementary Methods. Break-apart probes are designed to hybridize to the 5′ and 3′ regions flanking the genomic breakpoint of interest. In the presence of a rearrangement, these probes yield separated signals—typically one green (G) and one orange (O) signal with or without a residual fusion (1F/1G/1O)—whereas intact alleles in non-rearranged cells display two co-localized fusion signals (2F) (Fig. 1A). I. Challenges and solution for signal distance quantification Although the guidelines recommend that separated signals should be at least twice the diameter of a single signal to confirm a rearrangement, accurately quantifying this distance presents significant challenges in routine diagnostic practice. To overcome this limitation, we developed and implemented a novel single-cell image analysis method using ImageJ (NIH, Bethesda, MD, USA). This technique provides precise and reproducible measurements of the distances between signals, ensuring greater accuracy and standardization in the assessment of genomic rearrangements. II. Image acquisition and pre-processing Images were acquired under standardized conditions using constant exposure times on the fluorescence microscope to ensure data consistency. Ten images per sample were captured at 100× magnification, allowing for the evaluation of at least 200 nuclei. Individual neoplastic cells were isolated using the zoom tool in the ISIS MetaSystems software. Background subtraction was performed to enhance fluorescent signals and reduce nonspecific noise before saving the images in JPG format. III. Signal distance measurement using ImageJ The distance between the orange and green signals, indicative of fused or separated probes, was measured using ImageJ. For each single-cell image acquired, the "line selection" tool was used to draw straight lines between the centers of the corresponding orange/green probes. The line lengths, expressed in pixels, were calculated using the "Analyze → Measure" function (Fig. 1B). To minimize variability, two independent researchers (S.M. and F.R.B.) performed the measurements on the same digital images. Length values (LVs), representing precise measurements of signal distances, were recorded, ranging from 12 pixels (minimum distance indicating signal fusion) to 348 pixels (maximum separation indicating rearrangement). IV. Data Collection, Statistical Analysis, and Interpretation on Control-Samples For each Break-apart probe, ten control samples were analyzed, including five cases with confirmed rearrangements and five negative cases, as determined by FISH and methylation profiling. From each control sample, ten representative cells with variable signal distances were selected and measured using the previously described method, resulting in a total of 200 signal distance (LV) measurements per probe. All LV data were processed using MedCalc statistical software (version 19.2.6, MedCalc Software bv, Belgium, https://www.medcalc.org) and are available in the Supplementary Data. Initially, the data were graphically analyzed to visualize the distribution of distance measurements using density plots. Subsequently, the K-Means elbow method was applied to determine the optimal number of clusters. The K-Means elbow method is a statistical technique used to partition data into distinct groups based on numerical characteristics. It identifies the optimal number of clusters by detecting the point at which adding more clusters no longer significantly improves model performance. The analysis followed these steps: clustering using the K-Means algorithm for various numbers of clusters (up to 10), calculating the Within-Cluster Sum of Squares (WCSS), and identifying the "elbow" point, where further increases in the number of clusters do not substantially enhance the model. Using the K-Means elbow method, the optimal number of clusters was determined to be three (K=3) for each probe. The mean distortion, defined as the average of the squared distances between each data point and the centroid of its respective cluster, decreased significantly before stabilizing (Fig. 1C, representing MYBL1 ). Three distinct groups were observed (Fig. 1D): Cluster 1, which represents normal signals and includes the lowest values; Cluster 2, which includes intermediate values; and Cluster 3, representing rearranged signals and displaying the highest values. This three-cluster distribution pattern was consistently observed across all Break-apart probes. V. Thresholds Establishment for Break-Apart Probes To establish statistically significant cutoff values for distinguishing between rearranged and non-rearranged signals, an independent sample t-test was performed. A thresholding approach based on the mean ± standard deviation (σ) was applied, with cutoffs defined as: mean of rearranged signals - 1σ and mean of non-rearranged signals + 1σ. The separation between the rearranged cluster (green) and the non-rearranged cluster (blue) was statistically significant for all Break-apart probes (p<0.001), with the following LV distributions: MYB (190±64.2 vs. 47.2±15), MYBL1 (189.4±71.9 vs. 49.9±11), ZFTA (148.1±51.5 vs. 50.7±10.9), YAP1 (179.9±72.1 vs. 47.1±11.4), and MN1 (146.5±48.5 vs. 51.8±13.1). Threshold lines drawn in the graphical representation clearly separate the two groups. For instance, in the case of MYBL1 , rearranged signals were defined as ≥ 117 LV, while normal signals were ≤ 60 LV (Fig. 1D). An intermediate range (61–116 for MYBL1 ), defined as "uncertain", includes LV values that are not statistically distinguishable from the other two clusters. The same methodology was applied to all other Break-apart probes, yielding the following thresholds: MYB rearranged ≥ 126 vs. normal ≤ 62, ZFTA rearranged ≥ 96 vs. normal ≤ 61, YAP1 rearranged ≥ 108 vs. normal ≤ 58, and MN1 rearranged ≥ 98 vs. normal ≤ 65 (Supplementary Fig. 1). VI. Analysis and Interpretation of Rearrangements on Cohort-Samples To classify tumors samples in our cohort analyzed with Break-apart probes, LV measurements were obtained from 100 representative neoplastic cells per sample. The mean signal distance value (mean LV) was then calculated for each sample and applied to threshold distribution graphs, in order to determine the presence/absence of genomic rearrangements (see Results section). “Single-Cell Signal Pattern Analysis” for Fusions Detection Using Break-Apart and Fusion Probes To assess inter- and intra-chromosomal genetic fusions in PBTs using FISH, we applied a single-cell signal pattern analysis system. This methodology enables accurate fusion detection and offers standardized evaluation criteria, addressing a critical gap in current FISH diagnostic practices. I. KIAA1549::BRAF Fusion The KIAA1549::BRAF fusion arises from a tandem duplication at the 7q34 region, involving the 5′ portion of KIAA1549 and the 3′ portion of BRAF . Breakpoints are typically observed in exons 15 or 16 of KIAA1549 and exons 9, 10, or 11 of BRAF [21,43,62]. Due to their proximity on the same chromosome (spanning ~1500 Kb), conventional fusion probes are ineffective. To overcome this, we used a Break-Apart Tri-Color Probe that enables the detection of intra-chromosomal fusions by analyzing signal patterns. The probe set included: Aqua-labelled probe (533 Kb) targeting regions upstream and downstream of KIAA1549 , Orange-labelled probe (178 Kb) hybridizing the 3′ flanking region of BRAF , and Green-labelled probe (358 Kb) targeting the 5′ flanking region of BRAF (Supplementary Fig. 2A). At the single-cell level, the presence of the KIAA1549::BRAF intra-chromosomal fusion is indicated by the duplication of Aqua and Orange signals (O/Aq doublets) along with a single Green signal, reflecting the tandem duplication event. Only nuclei displaying a normal signal pattern (1G/1O/1Aq) along with a fusion signal were considered for the analysis. Samples were classified as positive when ≥20% of neoplastic nuclei exhibited this specific doublet-signal pattern (2G/2O+1O/2Aq+1Aq). The absence of the fusion was indicated by a normal pattern with closely spaced single signals (2G/2O/2Aq). II. FGFR1::TACC1 Fusion The FGFR1::TACC1 fusion results from an in-frame tandem duplication at the 8p11.23 region, involving the 5′ portion of TACC1 and the 3′ portion of FGFR1 . Breakpoints typically occur at exon 7 of TACC1 and exon 17 of FGFR1 , leading to the expression of the fusion oncogene [51,63]. Similar to KIAA1549::BRAF , due to the proximity of these genes on the same chromosome (~300 Kb), conventional fusion probes are ineffective. To detect this fusion, a Break-Apart Probe targeting regions flanking FGFR1 was employed (Supplementary Fig. 2B). At the single-cell level, a normal pattern was characterized by two fused Orange/Green signals (2F), while the FGFR1::TACC1 intra-chromosomal fusion was identified by Green signal duplication. Samples were deemed positive if ≥20% of neoplastic nuclei displayed this specific doublet-signal pattern (1F/1F+1G). III. SLC44A1::PRKCA Fusion The t(9;17)(q31;q24) translocation results in the inter-chromosomal fusion of the 5′ portion of SLC44A1 with the 3′ portion of PRKCA , with breakpoints typically occurring between exons 14–15 of SLC44A1 and exons 8–9 of PRKCA , leading to the expression of the fusion oncogene [55,56,64]. The PRKCA::SLC44A1 fusion involves genes located on separate chromosomes, making it detectable using conventional Fusion Probes. The probe set included a Green-labelled probe (525 Kb) targeting PRKCA at 17q24.2 and an Orange-labelled probe (328 Kb) targeting SLC44A1 at 9q31.1 (Supplementary Fig. 2C). At the single-cell level, a normal pattern was defined by separated Green and Orange signals (2G/2O), while the presence of fusion was indicated by two separated and one juxtaposed-fused signals (1F/1O/1G). Samples were considered positive when ≥20% of neoplastic nuclei exhibited this specific fusion-signal pattern. Whole-Transcriptome Sequencing for Evaluation of Genomic Rearrangements and Fusions Sixteen selected samples were prepared from FFPE tissue sections through manual microdissection of neoplastic areas selected from H&E-stained slides. Total RNA was extracted using the miRNeasy FFPE Kit (Qiagen), following the manufacturer's protocol. RNA quality and concentration were evaluated using the Qubit 4 Fluorometer with the RNA Broad Range kit (Thermo Fisher Scientific) and the TapeStation 4150 with the RNA ScreenTape kit (Agilent). Only samples with an RNA Integrity Number (RIN) greater than 2 were included for sequencing. Whole-transcriptome libraries were prepared using the RNA Prep with Enrichment Tagmentation Kit (Illumina). The protocol included RNA denaturation, cDNA synthesis, tagmentation (fragmenting cDNA and adding adapter sequences via bead-linked transposomes), PCR amplification with IDT sample indices, and library purification. The concentration of the libraries was quantified using the Qubit 4 Fluorometer with the dsDNA High Sensitivity kit (Thermo Fisher Scientific), and fragment size distributions were evaluated using the TapeStation 4150 with the High Sensitivity DNA ScreenTape kit (Agilent). The prepared libraries were sequenced on the high-throughput NovaSeq 6000 platform (Illumina) to generate paired-end reads. Raw data were subjected to quality control using FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Reads were aligned to the GRCh38 reference genome using the STAR aligner [66]. Structural variant detection was performed using specialized software, including STAR-Fusion [67] and FusionCatcher [68], to analyze chimeric reads and junction points aligned to the reference genome. Fusion transcripts were annotated to identify the involved genes and exon boundaries, and fusion events were analyzed using the R package chimeraviz (Bioconductor/R package version 1.30.0). Only relevant fusion transcripts were annotated, enabling quantitative assessment of each transcript. True positive (TP) and false positive (FP) events were counted based on a minimum threshold of supporting reads for each fusion. A stringent filtering algorithm was applied to variant calls to determine fusion events, ensuring the highest predictive accuracy. Moreover, genome-wide DNA methylation profiling was performed on 23 of 50 FFPE tumors using the Illumina EPIC 850K array to support tumor classification and CNV analysis, with detailed methodology provided in the Supplementary Methods. Results Single-Cell Signal Distance Analysis of MYB/MYBL1 Rearrangements Using FISH As previously described, to establish detection thresholds, we analyzed 200 single-cell signal distances in control-samples, identifying two statistically significant clusters for MYB : rearranged (190 ± 64.2) and normal (47.2 ± 15), with an intermediate “uncertain” range (63–125) ( Supplementary Fig. 1 ). Subsequently, ten pediatric diffuse astrocytoma cohort-samples, initially diagnosed based on morphological and immunohistochemical features, were analyzed for MYB/MYBL1 rearrangements using Break-apart probes. Mean signal distances (Mean LVs) were calculated across 100 representative neoplastic cells per cohort-sample. Four cases (40%) exhibited a break pattern (1F/1G/1O) in ≥ 30% of tumor cells (Fig. 2 B), with two exceeding the rearrangement threshold (> 126) and two within the uncertain range (LVs: 100 and 94) (Fig. 2 D). RNA sequencing confirmed rearrangements in the latter ( Supplementary Fig. 7A ). Five tumors (50%) showed a normal MYB pattern with two fused signals (2F) and mean LVs below the threshold (< 63) (Fig. 2 C). Interestingly, one tumor (10%) displayed a previously unreported atypical FISH pattern (2F + 1O), characterized by gain of the 3' Orange probe (Fig. 2 E), suggestive of an unbalanced MYB rearrangement, which was confirmed by additional methods. A similar approach was applied to MYBL1 , identifying rearranged (189.4 ± 71.9) and normal (49.9 ± 11) clusters in control-samples, with an uncertain range of 61–116. Subsequently, mean LVs were calculated in the ten cohort-samples, revealing MYBL1 rearrangements in three cases (LV > 117, Fig. 3 B). Seven cases (70%) were not rearranged, with six showing mean LVs below the threshold (Fig. 3 C) and one in the uncertain range (LV: 62, Fig. 3 D). The latter was analyzed by RNA sequencing to verify the absence of rearrangement, which was indeed confirmed ( Supplementary Fig. 7B ). MYB and MYBL1 rearrangements were mutually exclusive across all tumors analyzed. The RNA sequencing validation of uncertain cases refined our cutoff values, establishing more stringent thresholds for the uncertainty range ( MYB rearrangement: >93 vs. non-rearranged: >63; MYBL1 rearrangement: >117 vs. non-rearranged: >63) (Fig. 2 A, Fig. 3 A). Single-Cell Signal Distance Analysis of ZFTA/YAP1 Rearrangements Using FISH As previously described, we conducted the analysis of 200 single-cell signal distance measurements in control-samples to establish thresholds for detecting ZFTA and YAP1 rearrangements, revealing two distinct clusters: ZFTA -rearranged (148.1 ± 51.5) vs. normal signals (50.7 ± 10.9) and YAP1 -rearranged (179.9 ± 72.1) vs. normal signals (47.1 ± 11.4). Signals classified as uncertain were also observed within the range of 62–95 for ZFTA and 59–107 for YAP1 ( Supplementary Fig. 1 ). Subsequently, twelve pediatric supratentorial ependymoma cohort-samples, initially diagnosed based on morphological and immunohistochemical features, were analyzed for ZFTA / YAP1 rearrangements using Break-apart probes. Mean LVs were calculated across 100 representative neoplastic cells per cohort-sample. Five cases (41.6%) exhibited a ZFTA break pattern (1F/1G/1O) in ≥ 30% of tumor cells (Fig. 4 B), with four exceeding the rearrangement threshold (> 96) and one falling within the uncertain range (91). This latter case (Fig. 4 D), classified as borderline positive due to its proximity to the threshold, was confirmed as positive via RNA sequencing (Supplementary Fig. 7E ). Five samples (41.6%) showed no rearrangement (2F) (Fig. 4 C), with four below the normal cutoff (< 55) and one borderline (LV: 66), confirmed as negative by RNA sequencing ( Supplementary Fig. 7F ). Notably, two cases (16.8%) exhibited atypical FISH patterns not previously described in the literature: one with a gain of the 5' Green probe (Fig. 4 E) and one with a gain of the 3' Orange probe (Fig. 4 F). These patterns, indicative of unbalanced ZFTA rearrangements, were both confirmed through additional molecular analyses. Three cases (25%) harboured YAP1 rearrangements (1F/1G/1O) in > 40% of cells (Fig. 5 B). Two had mean LVs > 108, while one was borderline (mean LV: 92). RNA sequencing confirmed rearrangements in the latter ( Supplementary Fig. 7C ). Seven cases (70%) showed no evidence of YAP1 rearrangement, displaying two fused signals (Fig. 5 C). Six of these had mean LVs below the normal threshold (< 58), while one case fell within the uncertain range (62) (Fig. 5 D), confirmed as negative through RNA sequencing ( Supplementary Fig. 7F ). Thresholds were refined through RNA sequencing validation of uncertain cases: ZFTA rearrangement > 90 vs. non-rearranged 91 vs. non-rearranged < 6 (Fig. 5 A). Remarkably, one tumor harboured concurrent ZFTA and YAP1 rearrangements, indicating intratumoral heterogeneity and the necessity of comprehensive molecular profiling in pediatric brain tumors. Single-Cell Signal Distance Analysis of MN1 Rearrangements Using FISH As previously described, single-cell signal distance analysis of control-samples revealed two distinct MN1 clusters: rearranged (146.5 ± 48.5) and normal signals (51.8 ± 13.1), with a statistically significant difference. An additional group of uncertain signals (66–97) was also observed. Subsequently eight pediatric tumors, immunophenotypically consistent with astroblastoma, were analyzed using Break-apart probes. Mean signal distances were calculated on 100 representative neoplastic cells for each cohort-sample. MN1 rearrangement was detected in four cases (50%), characterized by two separated spots (1F/1G/1O) in > 40% of cells (Fig. 6 B). All these exhibited mean LVs exceeding the rearrangement threshold (> 97). The other four cases (50%) showed a normal pattern (2F) (Fig. 6 C), with mean LVs below the normal threshold (< 66). Notably, none of the MN1 analyzed tumors exhibited unbalanced patterns or showed borderline mean LVs (Fig. 6 A), reflecting a higher uniformity of signal distance measurements for MN1 rearrangements using FISH, particularly when compared to other rearrangements. Single-Cell Signal Pattern Analysis of KIAA1549::BRAF Fusion Using FISH We developed a single-cell signal pattern analysis method to establish standardized evaluation criteria for KIAA1549::BRAF fusion, addressing a gap in the current literature. Using this approach, we analyzed ten tumors with clinical, morphological, and immunohistochemical features consistent with pilocytic astrocytomas (7) or diffuse leptomeningeal glioneuronal tumors (3), employing Break-Apart Tri-Color probes as previously described ( Supplementary Fig. 2A ). The characteristic fusion pattern was defined by the presence of juxtaposed Orange and Aqua signals with gain of Orange (3' BRAF ) and Aqua ( KIAA1549 ) signals (e.g., 2G/2O + 1O/2Aq + 1Aq), indicating duplication events (Fig. 7 B, pattern 2 ). In some cases, partial probe binding resulted in patterns lacking duplicated Aqua (2G/2O + 1O/2Aq) (Fig. 7 B, pattern 1 ). In contrast, the normal pattern consisted of two single Green, Orange, and Aqua signals (2G/2O/2Aq), without doublets or signal overlap (Fig. 7 A). To ensure accurate interpretation, fluorescence channels and focal planes were individually acquired and visualized, facilitating clear differentiation between fusion-specific doublets and the single signals of normal pattern. Using this single-cell signal pattern analysis, 200 neoplastic nuclei per cohort-sample were evaluated. Six out of ten cases (60%) showed the fusion pattern in more than 30% of cells. Two samples, with 20–22% positive nuclei respectively, were close to the positivity threshold (20%). These were classified as positive based on FISH pattern analysis and subsequently validated by second-level methods ( Supplementary Fig. 7D ). The remaining two cases were negative for the fusion. This study is the first to establish objective and reproducible FISH-based standardized criteria for detecting KIAA1549::BRAF fusion in pediatric brain tumors. Single-Cell Signal Pattern Analysis of FGFR1::TACC1 Fusion Using FISH We applied the same single-cell analytical approach to define standardized criteria for FGFR1::TACC1 fusion detection, addressing the lack of clear guidelines for assessing this fusion via FISH in the current literature. Six tumors with histopathological features of extraventricular neurocytoma were analyzed using Break-Apart probes targeting the duplicated region implicated in the fusion ( Supplementary Fig. 2B ). The FGFR1::TACC1 fusion displayed a distinctive FISH pattern characterized by Green doublets (1F/1F + 1G), resulting from the in-frame tandem duplication in the region 8p11 (Fig. 8 B). In contrast, the normal pattern showed two fused signals (2F) without duplication (Fig. 8 A). Accurate interpretation of this fusion required the visualization of individual fluorescence channels and focal planes, enabling precise identification of the additional Green signal that formed the doublets. By comparison, fluorescence channels of the normal pattern displayed single probes for each fluorochrome, with no evidence of doublets. Using this approach, 200 neoplastic nuclei per cohort-sample were evaluated. The FGFR1::TACC1 fusion was identified in 4 out of 6 tumors (67%), while 2 cases (33%) were negative. One borderline sample, with 20% of cells exhibiting the fusion pattern, was near the positivity threshold and was subsequently analyzed via RNA sequencing, which failed to detect this fusion target ( Supplementary Fig. 6C ). This lack of detection is likely attributable to high intratumoral heterogeneity and low proportion of cells harbouring the variant. Notably, this is the first study establishing reproducible and standardized FISH criteria for FGFR1::TACC1 , improving the reliability of fusion detection in PBTs. Single-Cell Signal Pattern Analysis of PRKCA::SLC44A1 Fusion Using FISH In contrast to KIAA1549::BRAF and FGFR1::TACC1 , the PRKCA::SLC44A1 fusion results from an inter-chromosomal rearrangement, making it well-suited for detection using conventional dual-color Fusion probes ( Supplementary Fig. 2C ). To enhance diagnostic accuracy, we established standardized evaluation criteria for PRKCA::SLC44A1 detection based on a single-cell signal pattern analysis. We analyzed four samples exhibiting clinical, morphological, and immunohistochemical features suggestive of papillary glioneuronal tumors. The normal pattern displayed two separated Green and Orange signals (2G/2O), reflecting the absence of a fusion (Fig. 9 A). Fusion-positive nuclei were defined by directly juxtaposed and overlapping Orange and Green signals (1F/1O/1G) (Fig. 9 B), confirmed only when direct contact between probes was observed (Fig. 9 C). In addition, to further refine detection accuracy, imaging analysis was conducted across multiple single focal planes to precisely assess the spatial localization of the probes within the nuclei. True-positive signals were consistently captured and observed within the same focal plane ( Supplementary Fig. 3B ). Conversely, false-positive signals, arising from random folding of genomic DNA, were detected across different focal planes ( Supplementary Fig. 3A ). This stringent requirement helped avoid false positives from nuclear DNA folding artifacts, which produce apparent overlap across different focal planes. A total of 200 neoplastic nuclei per cohort-sample were analyzed using this single-cell signal pattern analysis. Of the four tumors, one was positive, with a clear fusion pattern in over 30% of nuclei, while two cases were classified as negative. One sample exhibited the fusion pattern in only 13% of cells, below the diagnostic threshold (20%). Consequently, this tumor was classified as negative for PRKCA::SLC44A1 and was further analyzed by RNA sequencing for additional validation, which did not detect the fusion target ( Supplementary Fig. 7H ). This underscores the limitations of RNA-seq for identifying variants present in low-frequency subclones. Notably, this approach establishes robust and standardized criteria for FISH-based detection of PRKCA::SLC44A1 , addressing the current lack of established guidelines in literature for assessing this rare fusion in PBTs. Correlation Between FISH and Genomic DNA Methylation Profiling Genomic DNA methylation profiling enables accurate subclassification of PBTs based on epigenetic signatures and provides complementary information on copy number variations (CNVs) [ 2 , 5 – 7 ]. Although this method does not directly detect structural variants, it can support the interpretation of fusions and rearrangements identified by FISH. Therefore, methylation profiling was performed on 23 out of 50 tumors (46%) with adequate DNA input (≥ 500 ng), independently of FISH outcomes, to indirectly assess the reliability of the signal distance measurement and pattern analysis. Of these, 21 cases (91.3%) showed complete concordance between FISH-detected rearrangements and methylation-defined epigenetic subtypes (Supplementary Table 3 ). These included cases with unbalanced rearrangements in MYB and ZFTA , which were correctly classified as Low-grade glioma with MYB rearrangement (LGG MYB) and Supratentorial ependymoma with ZFTA::RELA fusion (ST-RELA), respectively. These cases exhibited FISH patterns not previously reported, reinforcing the importance of recognizing unbalanced rearrangements in FISH interpretation. All concordant samples had epigenetic scores > 0.8, emphasizing the reliability of tumor classification based on epigenetic data [ 5 , 65 ]. In two samples (8.7%), the methylation subclass assigned was inconsistent with FISH results: one ZFTA -rearranged ependymoma was misclassified as a low-grade glioma, and one MN1 -rearranged astroblastoma was misclassified as a pleomorphic xanthoastrocytoma (PXA). Both tumors had low epigenetic scores below the threshold of 0.8 (0.55 and 0.46), indicating uncertain predictions due to insufficiently distinctive epigenetic features [ 5 , 65 ]. These discrepancies were resolved through RNA sequencing, which confirmed the rearrangements identified by FISH, underscoring methylation profiling limitations in low-score cases. Methylation clustering for all 23 tumors is showed in Supplementary Fig. 4 . CNVs profiling did not demonstrate correlation with specific rearrangements, revealing that such variants lead to heterogeneous and non-specific copy number alterations. Notably, KIAA1549::BRAF and FGFR1::TACC1 fusions corresponded to focal gains at 7q34 and 8p11 respectively, reflecting the duplications underlying these intra-chromosomal fusions ( Supplementary Fig. 5 ). Correlation Between FISH and Whole Transcriptome Sequencing To validate the accuracy of our single-cell FISH methods, specifically signal distance measurement and signal pattern analysis, we performed whole-transcriptome sequencing on 16 selected samples. These included borderline cases with mean signal distances near the rearranged/normal thresholds, as well as those exhibiting high intratumoral heterogeneity or fusion patterns close to the positivity cutoff. RNA-seq confirmed FISH findings in 13 out of 16 cases (81.2%), supporting strong concordance between the two methods, even in diagnostically challenging samples. Three cases (18.8%) showed discrepancies, as detailed in Supplementary Table 3 . In two samples, one with MYB rearrangement and the other with KIAA1549::BRAF fusion identified by FISH, no corresponding fusion transcripts were detected by RNA-seq ( Supplementary Fig. 6A and 6B ). Notably, genome-wide analysis failed to identify any high-confidence fusion events in either sample, suggesting that the sequencing reaction may have been compromised by RNA degradation or poor sample quality. In the third discordant case, a FGFR1::TACC1 fusion detected by FISH in 20% of tumor cells was not validated by RNA-seq ( Supplementary Fig. 6C ). This likely reflects the low abundance of the fusion-positive subclone and the high degree of intratumoral heterogeneity, underscoring a known limitation of RNA-seq in detecting low-frequency variants in genetically heterogeneous tumors [ 16 – 18 ]. Despite these exceptions, RNA-seq confirmed the majority of borderline/ambiguous cases and supported the refinement of FISH thresholds for signal distances and patterns interpretation. Notably, a novel MYB::CA10 fusion was identified by RNA-seq, resulting from an inter-chromosomal translocation t(6;17)(q23.3;q21.22) and marking the first report of CA10 as fusion partner of MYB . CA10 encodes a protein in the carbonic anhydrase family involved in CNS development and previously associated with colorectal and breast cancers [ 69 – 71 ], suggesting a potential novel fusion partner in MYB -rearranged pediatric astrocytomas. Sequencing analysis confirmed the absence of MYB / MYBL1 rearrangements in a FISH-negative tumor, validating the accuracy of the signal distance measurement method. Among the 17 somatic fusions identified in this sample, the ZFTA::RELA fusion transcript was found, leading to reclassification from low-grade glioma to ZFTA -rearranged supratentorial ependymoma and illustrating the diagnostic value of integrated molecular analysis. Additionally, RNA-seq also confirmed complex fusion patterns identified by FISH. In one tumor, both ZFTA::RELA and YAP1::TYR fusions were detected, demonstrating the coexistence of two relevant structural variants in the same tumor. While ZFTA::RELA is a hallmark of supratentorial ependymomas, the identification of TYR as a YAP1 fusion partner is novel. In another case, RNA-seq validated the KIAA1549::BRAF fusion detected by FISH and also revealed a concurrent MN1::BEND2 fusion, highlighting the potential for multiple concurrent rearrangements within a single tumor. Notably, RNA-seq also identified a novel PAX5::ATM fusion in a ZFTA/YAP1 -negative supratentorial ependymoma, resulting from an inter-chromosomal t(9;11)(p13.2;q22.3) rearrangement. While both PAX5 and ATM alterations are well known in hematologic malignancies [ 72 , 73 , 75 ], this is the first report of PAX5::ATM fusion in a pediatric brain tumor, representing a significant genetic finding in this context. A Circos plot of all fusions is presented in Supplementary Fig. 7 . Overall, our results demonstrate that RNA-seq complements FISH by validating and extending the genomic characterization of PBTs, particularly in cases with borderline signal distances or complex fusion patterns. The integrated application of histopathology, FISH-based single-cell analysis, DNA methylation profiling, and RNA-seq enabled accurate and consistent molecular classification across all analyzed cases, in alignment with current WHO diagnostic criteria [ 2 ]. This combined approach provides a robust framework for the diagnosis and subclassification of genetically heterogeneous pediatric brain tumors. Supplementary Fig. 8 presents a comprehensive overview of the clinical, cytogenetic, epigenetic, and transcriptomic data used for the formulation of the final integrated diagnoses. Discussion and Conclusion Pediatric brain tumors encompass a heterogeneous group of neoplasms defined by distinct genetic and epigenetic alterations. Accurate diagnosis requires a multidisciplinary approach integrating histopathology, immunophenotyping, and molecular data from advanced methodologies [ 2 ]. Among genetic alterations, PBTs are characterized by a relatively low somatic mutation burden but a high prevalence of structural variants, including tandem duplications, inversions, fusions, and translocations, which play critical roles in their pathogenesis [ 8 – 11 ]. A variety of advanced technologies have been employed to detect SVs in PBTs, including reverse transcription or digital PCR, RNA next-generation sequencing, methylation profiling, and FISH [ 2 , 4 , 5 , 8 – 10 , 12 ], with each methodology having inherent advantages and limitations. FISH is rapid and highly sensitive technique capable of detecting genetic alterations at the single-cell level, making it particularly valuable for identifying SVs in heterogeneous tumor cell populations, where subclonal events may otherwise go undetected [ 14 – 16 ]. However, its application is restricted to preselected genomic targets, it is susceptible to intra- and inter-observer variability, and it currently lacks standardized evaluation criteria specific to PBTs [ 13 – 15 ]. Conversely, RNA-seq allows for the comprehensive identification of multiple genomic fusions without prior knowledge of target genes, making it particularly useful for characterizing novel fusion events and complex structural rearrangements. Nevertheless, its effectiveness can be hampered by RNA degradation in formalin-fixed paraffin-embedded samples and by intratumoral genetic heterogeneity, which may obscure the detection of rare or subclonal rearrangements, reducing diagnostic sensitivity [ 5 – 7 , 16 – 18 ]. These complementary strengths and limitations underscore the importance of integrating multiple methodologies to achieve accurate and comprehensive characterization of PBTs, enabling improved diagnostic precision and insights into tumor biology. To address these challenges, we developed novel single-cell imaging methods for FISH, which provide objective and reproducible assessments of rearrangements and fusions. These methods were applied to a cohort of 50 pediatric patients with various CNS tumor subtypes, integrating RNA-seq and methylation profiling as complementary diagnostic tools. Using a newly developed "distance measurement method" for FISH Break-apart probes on control-samples, we established statistically significant thresholds (p < 0.001) to differentiate normal from rearranged signals for key genes recognized as diagnostic markers for specific tumor subtypes in the WHO classification, including MYB , MYBL1 , ZFTA , YAP1 , and MN1 [ 2 ]. By measuring the average signal distance (mean LV) across 100 representative neoplastic cells per cohort sample using FISH and employing additional confirmation methods such as RNA-seq and methylation array, we detected positive and negative rearrangements with the following concordance rates: MYB/MYBL1 (90%), ZFTA/YAP1 (91.5%), and MN1 (100%). In addition, we identified novel unbalanced rearrangements for MYB and ZFTA using FISH, which have not been previously reported in the literature. These unbalanced patterns emphasize the importance of recognizing atypical rearrangements when interpreting FISH results. We also observed mutual exclusivity between MYB and MYBL1 rearrangements, supporting their distinct roles in tumorigenesis. Notably, we documented the co-occurrence of ZFTA and YAP1 rearrangements within a single tumor, underscoring the genetic complexity and significant intratumoral heterogeneity of PBTs. This rare observation suggest that concurrent genetic events may contribute to PBT pathogenesis in specific cases. To further refine the characterization of genetic fusions, we implemented a "single-cell signal pattern analysis" method, enabling precise identification of distinct FISH patterns associated with key fusions, including KIAA1549::BRAF (dual orange/aqua signals), FGFR1::TACC1 (dual green signals), and PRKCA::SLC44A1 (juxtaposed orange/green signals). By analyzing single-cell patterns across 100 representative neoplastic cells per cohort-sample using FISH and employing RNA-seq and methylation array for confirmation, we detected positive and negative fusions with the following concordance rates: KIAA1549::BRAF (90%), FGFR1::TACC1 (83.5%), and PRKCA::SLC44A1 (100%). These methodological advancements provide standardized and objective criteria for the detection of structural variants in PBTs using FISH, significantly enhancing the diagnostic accuracy and minimizing interpretive variability. Integrative analyses demonstrated strong concordance between FISH and methylation profiling, with discrepancies observed in only two tumors exhibiting low methylation scores (< 0.6). These low methylation scores have been previously reported in the literature as potential indicators of misclassification within the epigenetic class and are directly associated with pre-analytical variability or genetic heterogeneity [ 5 – 7 , 65 ]. RNA-seq confirmed FISH findings in 81.2% of cases, with discrepancies observed in only three samples. These discrepancies were attributed to RNA degradation or significant intratumoral heterogeneity, factors that have been reported to contribute to the failure in detecting subclonal alterations by sequencing [ 16 – 18 ]. These results underscore the robustness of the novel FISH methods in detecting SVs in PBTs, even in challenging samples with borderline signal distances, degraded DNA/RNA, or high genetic variability. RNA-seq validation of FISH results in borderline cohort-samples, with mean signal distances near the thresholds, enabled the refinement of the uncertainty interval, establishing more precise and stringent cutoffs for detecting rearrangements. This process also contributed to the standardization of values for signal distances and pattern interpretation. However, additional validation using secondary methodologies is necessary when interpreting borderline FISH results in individual samples. Notably, RNA-seq uncovered novel fusion partners, including MYB::CA10 from t(6;17)(q23.3;q21.22) and YAP1::TYR from an intrachromosomal rearrangement at 11q22.1. These represent the first descriptions of these partners in association with MYB and YAP1 . Additionally, we identified the PAX5::ATM fusion, arising from t(9;11)(p13.2;q22.3), further expanding the spectrum of structural variants in PBTs. Importantly, the integration of FISH and RNA-seq reclassified a tumor initially diagnosed as a low-grade glioma to a ZFTA -rearranged supratentorial ependymoma, highlighting the critical role of a multi-methodologic molecular diagnostic approach. Notably, our identification of coexisting structural variants, such as ZFTA::RELA with YAP1::TYR in one tumor and KIAA1549::BRAF with MN1::BEND2 in another, provides compelling evidence of intratumoral genetic heterogeneity of PBTs. These findings demonstrate the coexistence of distinct genetic subpopulations within single tumors, reflecting the complexity of PBTs biology and emphasizing the importance of comprehensive molecular profiling for accurate tumor classification. In conclusion, our single-cell FISH methodologies establish robust, standardized criteria for detecting structural variants in PBTs, minimizing interpretive variability and improving diagnostic precision. When combined with RNA-seq and methylation profiling, this integrated approach enables comprehensive tumor characterization, enhances subtype classification accuracy, and uncovers clinically relevant genetic intratumoral heterogeneity of PBTs. Although the number of analyzed cases limits our study, it represents the first comprehensive effort to standardize FISH-based detection of genetic rearrangements in pediatric central nervous system tumors, laying a critical foundation for future research and clinical applications. Declarations Funding: This work was supported by Fondazione Italiana per la Lotta al Neuroblastoma ONLUS. Competing Interests: The authors have no relevant financial or non-financial interests to disclose. Author Contributions: All authors contributed to the study conception and design. Material preparation, data collection and FISH analysis were performed by Simone Minasi and Francesca Romana Buttarelli. Chiara Dossena, Marica Ficorilli, and Matilde Oriani performed methylation profiling and RNA sequencing analyses, under the supervision of Loris De Cecco. Francesca Gianno and Manila Antonelli were responsible for all morphological and immuno-phenotypical evaluations. Maura Massimino collected all clinical data of the patients. The first draft of the manuscript was written by Simone Minasi and Francesca Romana Buttarelli, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Conference Presentation: Part of these results will be accepted for abstract publication and poster presentation at the 20th Annual Meeting of the European Association of Neuro-Oncology (EANO 2025), to be held from October 16–19, 2025, in Prague, Czech Republic (Control No.: 2025-A-352-EANO). Data Availability: The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Ethics approval: This study was performed in line with the principles of the Declaration of Helsinki. Anonymization protocols ensured patient confidentiality. Consent to participate: Written informed consent was obtained from all the parents of individual participants included in the study. References Ostrom QT, de Blank PM, Kruchko C et al (2015) Alex's Lemonade Stand Foundation Infant and Childhood Primary Brain and Central Nervous System Tumors Diagnosed in the United States in 2007–2011. Neuro Oncol. 2015;16 Suppl 10(Suppl 10):x1-x36 Louis DN, Perry A, Wesseling P et al (2021) The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Rev Neuro Oncol 23(8):1231–1251 Pollack IF, Agnihotri S, Broniscer A (2019) Childhood brain tumors: current management, biological insights, and future directions. J Neurosurg Pediatr 23:261–273 Lim-Fat MJ, Macdonald M, Lapointe S et al (2022) Molecular testing for adolescent and young adult central nervous system tumors: A Canadian guideline. Front Oncol. 2022;12:960509 Capper D, Jones DTW, Sill M et al (2018) DNA methylation-based classification of central nervous system tumours. Nature 555(7697):469–474 Korshunov A, Sahm F, Zheludkova O et al (2019) DNA methylation profiling is a method of choice for molecular verification of pediatric WNT-activated medulloblastomas. Neuro Oncolol 21(2):214–221 Danielsson A, Nemes S, Tisell M et al (2015) MethPed: a DNA methylation classifier tool for the identification of pediatric brain tumor subtypes. Clin Epigenet. 2015;7:62 Gröbner SN, Worst BC, Weischenfeldt J et al (2018) ICGC PedBrain-Seq Project; ICGC MMML-Seq Project. The landscape of genomic alterations across childhood cancers. Nature. 2018;559(7714):E10 Yang Y, Yang L (2023) Somatic structural variation signatures in pediatric brain tumors. medRxiv. 2023 May 26:2023.05.18.23290139. doi: 10.1101/2023.05.18.23290139. Update in: Cell Rep. 2023;42(10):113276. 10.1016/j.celrep.2023.113276 Zhang Y, Chen F, Donehower LA (2021) A pediatric brain tumor atlas of genes deregulated by somatic genomic rearrangement. Nat Commun. 2021;12(1):937. 10.1038/s41467-021-21081-y Roosen M, Odé Z, Bunt J, Kool M (2022) The oncogenic fusion landscape in pediatric CNS neoplasms. Acta Neuropathol. 2022;143(4):427–451 Struijk RB, Dorssers LCJ, Henneman P et al (2020) Comparing genome-scale DNA methylation and CNV marks between adult human cultured ITGA6 + testicular cells and seminomas to assess in vitro genomic stability. PLoS ONE 15(3):e0230253 Allahyar A, Pieterse M, Swennenhuis J et al (2021) Robust detection of translocations in lymphoma FFPE samples using targeted locus capture-based sequencing. Nat Commun. 2021;12(1):3361 Chrzanowska NM, Kowalewski J, Lewandowska MA (2020) Use of Fluorescence In Situ Hybridization (FISH) in Diagnosis and Tailored Therapies in Solid Tumors. Molecules. 2020;25(8):1864 Horbinski C, Miller CR, Perry A (2011) Gone FISHing: clinical lessons learned in brain tumor molecular diagnostics over the last decade. Brain Pathol. 2011;21(1):57–73 Koduru P, Chen W, Fuda F et al (2024) RNASeq Analysis for Accurate Identification of Fusion Partners in Tumor Specific Translocations Detected by Standard FISH Probes in Hematologic Malignancies. Clin Pathol. 2024;17:2632010X241230262 Allahyar A, Pieterse M, Swennenhuis J et al (2021) Robust detection of translocations in lymphoma FFPE samples using targeted locus capture-based sequencing. Nat Commun. 2021;12(1):3361 Matsunaga H, Arikawa K, Yamazaki M et al (2022) Reproducible and sensitive micro-tissue RNA sequencing from formalin-fixed paraffin-embedded tissues for spatial gene expression analysis. Sci Rep. 2022;12(1):19511 Wefers AK, Stichel D, Schrimpf D (2020) Isomorphic diffuse glioma is a morphologically and molecularly distinct tumour entity with recurrent gene fusions of MYBL1 or MYB and a benign disease course. Acta Neuropathol 139:193–209 Ellison DW, Hawkins C, Jones DTW et al (2019) cIMPACT-NOW update 4: diffuse gliomas characterized by MYB, MYBL1, or FGFR1 alterations or BRAFV600E mutation. Acta Neuropathol 137:683–687 Ryall S, Tabori U, Hawkins C (2020) Pediatric Low-Grade Glioma in the Era of Molecular Diagnostics. Acta Neuropathol Commun 8(1):30 Fabbri VP, Caporalini C, Asioli S, Buccoliero A (2022) Paediatric-type diffuse low-grade gliomas: a clinically and biologically distinct group of tumours with a favourable outcome. Pathologica 114(6):410–421 Cicirò Y, Sala A (2021) MYB oncoproteins: emerging players and potential therapeutic targets in human cancer. Oncogenesis 10:19 Park YW, Vollmuth P, Foltyn-Dumitru M (2023) al al The 2021 WHO Classification for Gliomas and Implications on Imaging Diagnosis: Part 2-Summary of Imaging Findings on Pediatric-Type Diffuse High-Grade Gliomas, Pediatric-Type Diffuse Low-Grade Gliomas, and Circumscribed Astrocytic Gliomas. J Magn Reson Imaging 2023 Apr 17 Pajtler KW, Witt H, Sill M et al (2015) Molecular Classification of Ependymal Tumors across All CNS Compartments, Histopathological Grades, and Age Groups. Cancer Cell 2015;27:728–743 Merchant TE, Bendel AE, Sabin ND et al (2019) Conformal radiation therapy for pediatric ependymoma, chemotherapy for incompletely resected ependymoma, and observation for completely resected, supratentorial ependymoma. J Clin Oncol 37:974–983 Pages M, Pajtler KW, Puget S et al (2019) Diagnostics of pediatric supratentorial RELA ependymomas: integration of information from histopathology, genetics, DNA methylation and imaging. Brain Pathol 29:325–335 Arabzade A, Zhao Y, Varadharajan S et al (2021) ZFTA-RELA Dictates Oncogenic Transcriptional Programs to Drive Aggressive Supratentorial Ependymoma. Cancer Discov. 2021, 11, 2200–2215 Gessi M, Giagnacovo M, Modena P et al (2019) Role of Immunohistochemistry in the Identification of Supratentorial C11ORF95-RELA Fused Ependymoma in Routine Neuropathology. Am J Surg Pathol. 2019;43(1):56–63 Andreiuolo F, Varlet P, Tauziède-Espariat A et al (2019) Childhood Supratentorial Ependymomas with YAP1-MAMLD1 Fusion: An Entity with Characteristic Clinical, Radiological, Cytogenetic and Histopathological Features: YAP1-MAMLD1 Ependymoma. Brain Pathol. 2019, 29, 205–216 Parker M, Mohankumar KM, Punchihewa C et al (2014) C11orf95-RELA fusions drive oncogenic NF-κB signalling in ependymoma. Nature 2014;506:451–455 Burford A, Mackay A, Popov S et al (2018) The ten-year evolutionary trajectory of a highly recurrent paediatric high grade neuroepithelial tumour with MN1:BEND2 fusion. Sci Rep 8:1032 Sughrue ME, Choi J, Rutkowski MJ et al (2011) Clinical features and post-surgical outcome of patients with astroblastoma. J Clin Neurosci. 2011;18(6):750-4 Lehman NL, Usubalieva A, Lin T et al (2019) Genomic analysis demonstrates that histologically defined astroblastomas are molecularly heterogeneous and that tumors with MN1 rearrangement exhibit the most favorable prognosis. Acta Neuropathol Commun 7(1):42 Lehman NL, Spassky N, Sak M et al (2023) Astroblastomas exhibit radial glia stem cell lineages and differential expression of imprinted and X-inactivation escape genes. Nat Commun. 2023;14(1):507 Gopakumar S, McDonald MF, Sharma H et al (2022) Recurrent HGNET-MN1 altered (astroblastoma MN1-altered) of the foramen magnum: Case report and molecular classification. Surg Neurol Int. 2022;13:139 Sturm D, Orr BA, Toprak UH et al (2016) New brain tumor entities emerge from molecular classification of CNS-PNETs. Cell 164:1060–1072 Wood MD, Tihan T, Perry AJ et al (2018) Multimodal molecular analysis of astroblastoma enables reclassification of most cases into more specific molecular entities. Brain Pathol 28:192–202 Brat DJ, Aldape K, Idbaih A et al (2021) Astroblastoma, MN1-altered. In: WHO Classification of Tumours Editorial Board, ed. WHO Classification of the Central Nervous System Tumours. 5th ed. IARC; 2021 Knight J, De Jesus O (2021) Pilocytic Astrocytoma. In StatPearls Publishings. 2021; 30:14–24 Bornhorst M, Frappaz D, Packer RJ (2016) Pilocytic astrocytomas. Handb. Clin. Neurol. 2016; 134:329–44 Jones DT, Gronych J, Lichter P et al (2012) MAPK pathway activation in pilocytic astrocytoma. Cell Molec Life Sci 2012;69:1799Y811 Ryall S, Zapotocky M, Fukuoka K et al (2020) Integrated molecular and clinical analysis of 1,000 pediatric low-grade gliomas. Cancer Cell 37:569–583e5 Appay R, Pages M, Colin C et al (2020) Diffuse leptomeningeal glioneuronal tumor: a double misnomer? A report of two cases. Acta Neuropathol Commun 2020;8, 95 Gardiman MP, Fassan M, Orvieto E et al (2010) Diffuse leptomeningeal glioneuronal tumors: a new entity? Brain Pathol 2010;20:361–66 Deng MY, Sill M, Chiang J et al (2018) Molecularly defined diffuse leptomeningeal glioneuronal tumor (DLGNT) comprises two subgroups with distinct clinical and genetic features. Acta Neuropathol. 2018;136(2):239–253 Zacharoulis S, Morales La Madrid A et al (2016) Central Versus Extraventricular Neurocytoma in Children: A Clinicopathologic Comparison and Review of the Literature. J Pediatr Hematol Oncol. Aug 2016;38(6):479 – 85 Giangaspero F, Cenacchi G, Losi L et al (1997) Extraventricular neoplasms with neurocytoma features. A clinicopathological study of 11 cases. Am J Surg Pathol. Feb 1997;21(2):206 – 12 Xiong Z, Zhang J, Li Z et al (2015) A comparative study of intraventricular central neurocytomas and extraventricular neurocytomas. J Neurooncol. Feb 2015;121(3):521-9 Xu L, Ouyang Z, Wang J et al (2017) A clinicopathologic study of extraventricular neurocytoma. J Neurooncol. Mar 2017;132(1):75–82 Sievers P, Stichel D, Schrimpf D et al (2018) FGFR1:TACC1 fusion is a frequent event in molecularly defined extraventricular neurocytoma. Acta Neuropathol. 2018;136(2):293–302 Kalawi AZ, Malicki DM, Abdullaev Z et al (2022) The role of methylation profiling in histologically diagnosed neurocytoma: a case series. J Neurooncol. 2022;159(3):725–733 Janice S, Ahn MD et al (2021) Papillary Glioneuronal Tumor With a Novel GPR37L1-PRKCA Fusion, Journal of Neuropathology & Experimental Neurology, Volume 80, Issue 10, October 2021, Pages 1004–1006 Hou Y, Pinheiro J, Sahm F et al (2019) Papillary glioneuronal tumor (PGNT) exhibits a characteristic methylation profile and fusions involving PRKCA. Acta Neuropathol 137:837–846 Pages M, Lacroix L, Tauziede-Espariat A et al (2015) Papillary glioneuronal tumors: Histological and molecular characteristics and diagnostic value of SLC44A1-PRKCA fusion. Acta Neuropathol Commun 2015;3:85 Nagaishi M, Nobusawa S, Matsumura N et al (2016) SLC44A1-PRKCA fusion in papillary and rosette-forming glioneuronal tumors. J Clin Neurosci 2016;23:73–5 Goyal-Honavar A, Balasundaram A, Thayakaran IP et al (2022) ZFTA (Zinc Finger Translocation Associated) Fusion in Supratentorial Ependymomas: Low Prevalence in South Asians and No Correlation with Survival. World Neurosurg. 2022;164:e82-e90 Fudaba H, Momii Y, Kawasaki Y et al (2020) Well-differentiated Astroblastoma with Both Focal Anaplastic Features and a Meningioma 1 Gene Alteration. NMC Case Rep J. 2020;7(4):205–210 Dandapath I, Das S, Charan BD et al (2024) Evaluation of KIAA1549::BRAF fusions and clinicopathological insights of pilocytic astrocytomas. Ann Diagn Pathol. 2024;72:152318 Tomić TT, Olausson J, Wilzén A et al (2017) A new GTF2I-BRAF fusion mediating MAPK pathway activation in pilocytic astrocytoma. PLoS One. 2017;12(4):e0175638 Rodriguez FJ, Schniederjan MJ, Nicolaides T et al (2015) High rate of concurrent BRAF-KIAA1549 gene fusion and 1p deletion in disseminated oligodendroglioma-like leptomeningeal neoplasms (DOLN). Acta Neuropathol. 2015;129(4):609–610 Jones DT, Kocialkowski S, Liu L et al (2008) Tandem duplication producing a novel oncogenic BRAF fusion gene defines the majority of pilocytic astrocytomas. Cancer Res. 2008; 68(21):8673–7 Asuzu DT, Desai B, Maggio D et al (2021) FGFR1-TACC1 fusion associated with malignant transformation in a primary spinal cord glioma: a case report. J Spine Surg. 2021;7(3):434–438 Bridge JA, Liu XQ, Sumegi J et al (2013) Identification of a novel, recurrent SLC44A1-PRKCA fusion in papillary glioneuronal tumor. Brain Pathol. 2013;23(2):121-8 Capper D, Stichel D, Sahm F et al (2018) Practical implementation of DNA methylation and copy-number-based CNS tumor diagnostics: the Heidelberg experience. Acta Neuropathol 136:181–210 Dobin A, Davis CA, Schlesinger F et al (2012) STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15–21 Haas BJ, Dobin A, Li B et al (2019) Accuracy assessment of fusion transcript detection via read-mapping and de novo fusion transcript assembly-based methods. Genome Biol 20:213 Nicorici D, Satalan M, Edgren H et al (2014) FusionCatcher – a tool for finding somatic fusion genes in paired-end RNA-sequencing data, bioRxiv, Nov. 2014 Sterky FH, Trotter JH, Lee SJ et al (2017) Carbonic anhydrase-related protein CA10 is an evolutionarily conserved pan-neurexin ligand. Proc Natl Acad Sci U S A. 2017;114(7):E1253-E1262 Schulte I, Batty EM, Pole JC et al (2012) Structural analysis of the genome of breast cancer cell line ZR-75-30 identifies twelve expressed fusion genes. BMC Genomics. 2012;13:719 Novak RL, Harper DP, Caudell D et al (2012) Gene expression profiling and candidate gene resequencing identifies pathways and mutations important for malignant transformation caused by leukemogenic fusion genes. Exp Hematol. 2012;40(12):1016-27 Fouad FM, Eid JI (2023) PAX5 fusion genes in acute lymphoblastic leukemia: A literature review. Medicine (Baltimore). 2023;102(20):e33836 Črepinšek K, Klobučar N, Tesovnik T et al (2024) PAX5 Alterations in a Consecutive Childhood B-Cell Acute Lymphoblastic Leukemia Cohort Treated Using the ALL IC-BFM 2009 Protocol. Cancers (Basel). 2024;16(6):1164 Kanagal-Shamanna R, Bao H, Kearney H et al (2022) Molecular characterization of Novel ATM fusions in chronic lymphocytic leukemia and T-cell prolymphocytic leukemia. Leuk Lymphoma. 2022;63(4):865–875 Additional Declarations No competing interests reported. Supplementary Files SupplementaryFiguresMinasietal..docx SupplementaryTablesMinasietal..docx SupplementaryMethodsMinasietal..docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7244481","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":494739909,"identity":"818b0bce-3012-449f-91e1-c35ac3e976d5","order_by":0,"name":"Simone Minasi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIiWNgGAWjYDACZijNBiISKqAMBgMGHiK1nIFpKWDgwa0HGTC2wVgfGHBaI9/Onfi4oIIhmk+6+fGHh/MO5/FJHz664YcBg4w9Di0Gh3k3G884w5DbJnPMTCJx2+FiNr60tJs9BrgdZsDMu02atw2oRSLBjAGoJbGNh8fsBg8eLfLNIC3/QFrSP39InAPRcvMPHi0Mh0FaGkBacgwkEhsgWm7jswXsF55jEiAtZRIJx9KBWtjSbssYSPDwHMDhsP6zGx/z1Njkzp+RvvnjjxrrxPk9zMduvvljY8/egMtlYCBBhMgoGAWjYBSMAuIBAOF7TrTYStoIAAAAAElFTkSuQmCC","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":true,"prefix":"","firstName":"Simone","middleName":"","lastName":"Minasi","suffix":""},{"id":494739910,"identity":"df0fc05c-9425-40e8-a189-0b6d18dd0209","order_by":1,"name":"Chiara Dossena","email":"","orcid":"","institution":"Fondazione IRCCS Istituto Nazionale dei Tumori","correspondingAuthor":false,"prefix":"","firstName":"Chiara","middleName":"","lastName":"Dossena","suffix":""},{"id":494739911,"identity":"dcf64f60-4e74-4b95-aff3-574e93bdc08a","order_by":2,"name":"Francesca Gianno","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Francesca","middleName":"","lastName":"Gianno","suffix":""},{"id":494739912,"identity":"f3227ec6-da7d-4374-8621-fc61ea405967","order_by":3,"name":"Marica Ficorilli","email":"","orcid":"","institution":"Fondazione IRCCS Istituto Nazionale dei Tumori","correspondingAuthor":false,"prefix":"","firstName":"Marica","middleName":"","lastName":"Ficorilli","suffix":""},{"id":494739913,"identity":"7d1df958-e5ce-4958-af85-b0e65e289684","order_by":4,"name":"Matilde Oriani","email":"","orcid":"","institution":"Fondazione IRCCS Istituto Nazionale dei Tumori","correspondingAuthor":false,"prefix":"","firstName":"Matilde","middleName":"","lastName":"Oriani","suffix":""},{"id":494739914,"identity":"bac2f812-a1d3-4b53-a048-e30d3abcd3b1","order_by":5,"name":"Manila Antonelli","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Manila","middleName":"","lastName":"Antonelli","suffix":""},{"id":494739915,"identity":"5794610b-4643-4488-97d5-0c8dfca5abf0","order_by":6,"name":"Maura Massimino","email":"","orcid":"","institution":"Fondazione IRCCS Istituto Nazionale dei Tumori","correspondingAuthor":false,"prefix":"","firstName":"Maura","middleName":"","lastName":"Massimino","suffix":""},{"id":494739916,"identity":"0e9665fa-5ada-4d05-bd36-34610f09864c","order_by":7,"name":"Loris Cecco","email":"","orcid":"","institution":"Fondazione IRCCS Istituto Nazionale dei Tumori","correspondingAuthor":false,"prefix":"","firstName":"Loris","middleName":"","lastName":"Cecco","suffix":""},{"id":494739917,"identity":"9105f7f8-6221-47c5-bd24-93d60fb1bb5c","order_by":8,"name":"Francesca Romana Buttarelli","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Francesca","middleName":"Romana","lastName":"Buttarelli","suffix":""}],"badges":[],"createdAt":"2025-07-29 14:38:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7244481/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7244481/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88339320,"identity":"e69ae424-e801-4865-a22e-763a31d7c343","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":319380,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e“Signal distance measurement” in single cells using Break-apart probes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) \u003cem\u003eMYBL1\u003c/em\u003e Break-apart probe images show: a rearranged pattern (left), a normal fused pattern (middle), and an ambiguous case with minimal separation (right).\u003c/p\u003e\n\u003cp\u003e(B) Signal distances were measured in ImageJ using the line selection tool (1) to draw lines from the center of the orange probe to the center of the corresponding green probe and the “Analyze → Measure” function (2) to calculate the signal distance in pixels.\u003c/p\u003e\n\u003cp\u003e(C) \u003cem\u003eMYBL1\u003c/em\u003e K-means elbow analysis of 200 LVs from ten control-samples identifies 3 optimal clusters.\u003c/p\u003e\n\u003cp\u003e(D) Representative plot of \u003cem\u003eMYBL1\u003c/em\u003e \u0026nbsp;shows the distribution of 200 LVs measured in single cells using ImageJ. Two main clusters are clearly separated with statistically significant differences (p\u0026lt;0.001): rearranged (≥117, green) and normal (≤60, blue). Centroids (*) mark mean LV values (189 vs. 50). A third cluster (LVs 61–116) represents uncertain signals.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/b7d0705c1871deafdc9c5a47.png"},{"id":88339322,"identity":"ae75fe74-397e-4ce5-bf83-68762b033458","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":498787,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMYB\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003erearrangements by FISH using single-cell signal distance measurement.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Distribution graph of mean LVs for each cohort-sample shows two rearranged (169, 143), six normal (45–59), and two uncertain cases (100, 94), near the rearrangement threshold. RNA-seq confirmed the uncertain cases as rearranged, refining the uncertainty range and defining more stringent cutoffs: ≥93 for rearranged (orange line), ≤63 for normal (black line).\u003c/p\u003e\n\u003cp\u003e(B–E) Representative \u003cem\u003eMYB\u003c/em\u003eimages show: (B) a rearranged pattern (1F/1G/1O); (C) a normal pattern with two fused signals (2F); (D) an uncertain pattern with slightly separated signals; and (E) an atypical pattern (2F+1O), characterized by an additional orange signal (white circle), suggestive of an unbalanced rearrangement.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/1927c89dece87bbf02a51483.png"},{"id":88339330,"identity":"8ba52bb9-b90f-4ad3-940f-b47433eb500b","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":404649,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMYBL1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003erearrangements by FISH using single-cell signal distance measurement.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Distribution graph of mean LVs for each cohort-sample shows three rearranged (132, 153, 137), six normal (47, 48, 40, 53, 57, 45), and one uncertain case (62), near the normal threshold. RNA-seq confirmed the uncertain case as not rearranged, refining the uncertainty interval and establishing more stringent cutoffs: ≥117 for rearranged (orange line), ≤63 for normal (black line).\u003c/p\u003e\n\u003cp\u003e(B–E) Representative \u003cem\u003eMYBL1\u003c/em\u003eimages show: (B) a rearranged pattern (1F/1G/1O); (C) a normal pattern with two fused signals (2F); and (D) an uncertain pattern with minimally separated signals.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/dbef9d7d596ab7fb42e873ce.png"},{"id":88339329,"identity":"779cbb94-8d07-4c51-ad97-316db498818b","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":851023,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eZFTA\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003erearrangements by FISH using single-cell signal distance measurement.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Distribution graph of mean LVs for each cohort-sample shows four rearranged (131, 115, 105, 101), six normal (45, 51, 49, 39, 54, 46), and two uncertain cases (91, 66), with values near the rearranged and normal thresholds, respectively. RNA sequencing validation enabled the refinement of the uncertainty interval and the definition of more stringent cutoffs: ≥90 for rearranged (orange line), ≤66 for normal (black line).\u003c/p\u003e\n\u003cp\u003e(B–E) Representative \u003cem\u003eZFTA\u003c/em\u003eimages show: (B) a rearranged pattern (1F/1G/1O); (C) a normal pattern with two fused signals (2F); (D) an uncertain pattern with minimally separated signals; (E) an atypical pattern (2F+1G), characterized by an additional green signal (white circle), suggestive of an unbalanced rearrangement; and (F) an atypical pattern (2F+1O), characterized by an additional orange signal (white circle), suggestive of an unbalanced rearrangement\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/b9e7164ed9cff045ed6ed3ec.png"},{"id":88339982,"identity":"1959a440-9a9c-44d7-9e3f-f1197ee3036f","added_by":"auto","created_at":"2025-08-05 12:32:31","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":404338,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eYAP1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003erearrangements by FISH using single-cell signal distance measurement.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Distribution graph of mean LVs for each cohort-sample shows two rearranged (138, 165), eight normal (46, 37, 35, 50, 44, 52, 39, 46), and two uncertain cases (92, 62), with values near the rearranged and normal thresholds, respectively. RNA-seq confirmed both uncertain cases, allowing refinement of the uncertainty range and the definition of stricter cutoffs: ≥91 for rearranged (orange line), ≤63 for normal (black line).\u003c/p\u003e\n\u003cp\u003e(B–E) Representative \u003cem\u003eYAP1\u003c/em\u003eimages show: (B) a rearranged pattern (1F/1G/1O); (C) a normal pattern with two fused signals (2F); and (D) an uncertain pattern with minimally separated signals.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/adaae072d2009f4420f31c7d.png"},{"id":88339326,"identity":"6c868e5c-07a8-4aab-99a3-1730cc8a5185","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":284629,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMN1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003erearrangements by FISH using single-cell signal distance measurement.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Distribution graph of mean LVs for each cohort sample shows four rearranged cases (146, 134, 100, 111) and four normal cases (63, 49, 41, 38), with no tumors in the uncertain range. The red line indicates the rearrangement cutoff (≥98), and the black line indicates the normal cutoff (≤63).\u003c/p\u003e\n\u003cp\u003e(B–E) Representative \u003cem\u003eMN1\u003c/em\u003eimages show: (B) a rearranged pattern (1F/1G/1O); and (C) a normal pattern with two fused signals (2F).\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/f093d215a0b0096bd95f394e.png"},{"id":88339331,"identity":"6b23087a-5956-45ab-93f5-bf57670327d3","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":207533,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eKIAA1549::BRAF\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003efusion by FISH using single-cell signal pattern analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Simplified linear representation of the intact 7q34 region, indicating the genomic positions of probes. Normal FISH pattern displays single Green, Orange, and Aqua signals (2G/2O/2Aq) and the absence of doublets.\u003c/p\u003e\n\u003cp\u003e(B) Simplified linear representation of the 7q34 region, showing duplication of 5’ portion of \u003cem\u003eKIAA1549\u003c/em\u003e (partially complementary to Aqua probe) and 3' portion of \u003cem\u003eBRAF\u003c/em\u003e (fully complementary to Orange probe). Depending on partial binding of Aqua probe to the duplicated region, Aqua may form doublets (pattern 2: 2G/2O+1O/2Aq+1Aq) or remain single (pattern 1: 2G/2O+1O/2Aq). In both fusion patterns, Orange and Aqua signals are juxtaposed, forming a characteristic fusion. Individual fluorescence channels clearly show the presence of doublets, which are highlighted by white circles.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/34105dd397c06f356d1339b7.png"},{"id":88339989,"identity":"36c72775-30b5-499c-a2df-99be43099ce3","added_by":"auto","created_at":"2025-08-05 12:32:32","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":228811,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eFGFR1::TACC1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003efusion by FISH using single-cell signal pattern analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA) Simplified linear representation of the intact 8p11 region, showing the genomic positions of probes. \u0026nbsp;Normal FISH pattern displays two fused Green/Orange signals (2F) and the absence of doublets.\u003c/p\u003e\n\u003cp\u003e(B) Simplified linear representation of the genomic regions on 8p11 involved in the \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusion, resulting from an in-frame tandem duplication. This event leads to the duplication of Green signals (1F/1F+1G), forming a characteristic doublet. Individual fluorescence channels demonstrate the presence of doublets, highlighted by white circles.\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/96e51c59b0f823c5db68984a.png"},{"id":88339335,"identity":"d3ca0a15-17ee-4175-935c-8bf9be948f08","added_by":"auto","created_at":"2025-08-05 12:24:32","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":442086,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePRKCA::SLC44A1 \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003efusion by FISH using single-cell signal pattern analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Simplified linear representation of \u003cem\u003eSLC44A1\u003c/em\u003e (9q31.1) and \u003cem\u003ePRKCA\u003c/em\u003e (17q24.2) genomic regions. The normal FISH pattern shows separate signals (2G/2O).\u003c/p\u003e\n\u003cp\u003e(B) Simplified linear representation of genomic regions fused to form the \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003eoncogene. This variant results in the fusion of Green/Orange signals (1F/1O/1G), highlighted by white circles.\u003c/p\u003e\n\u003cp\u003e(C) Representative images show non-juxtaposed signals (yellow arrow), lacking probe contact and classified as negative, versus juxtaposed/overlapping signals (white circle), indicating direct probe contact and classified as positive for the fusion.\u003c/p\u003e","description":"","filename":"Figure9.png","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/f63a86136292fe9fda749dc3.png"},{"id":89226778,"identity":"7b5db565-ccec-4c29-8068-b88e579051b2","added_by":"auto","created_at":"2025-08-17 12:38:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5713464,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/18ada12c-ecd9-4f83-ae8b-2eddee614d4a.pdf"},{"id":88339327,"identity":"6113672d-18a4-4390-8ea3-2504050f655f","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":3993983,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFiguresMinasietal..docx","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/f7f04b5c815c3ea4e37a7df3.docx"},{"id":88339321,"identity":"05411c87-44a4-4821-b3f5-ecc6f435551d","added_by":"auto","created_at":"2025-08-05 12:24:31","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":23463,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTablesMinasietal..docx","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/9bab9eed8bbdd658db4f1bf2.docx"},{"id":88339979,"identity":"95c26c7a-e39b-458d-816a-1bf8b7770845","added_by":"auto","created_at":"2025-08-05 12:32:31","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":17705,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMethodsMinasietal..docx","url":"https://assets-eu.researchsquare.com/files/rs-7244481/v1/028407ab4ad6953006bf0fec.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Integrated molecular approach including single-cell FISH enables accurate detection of genetic rearrangements, classification, and identification of novel variants in pediatric brain tumors","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePediatric brain tumors (PBTs) represent a heterogeneous group of central nervous system (CNS) neoplasms, ranging 0\u0026ndash;19 years and encompassing a wide range of pathological entities [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The 2021 WHO Classification of CNS tumors emphasizes the importance of molecular diagnostics by integrating genetic and epigenetic data to define clinically relevant tumor subtypes [\u003cspan additionalcitationids=\"CR3 CR4 CR5 CR6\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e\u003cp\u003ePBTs frequently harbour somatic structural variants (SVs), including duplications, inversions, translocations, and fusions, which play a critical role in tumor classification and biological behavior [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Several rare PBTs are characterized by distinctive genetic rearrangements or fusions, which serve as diagnostic hallmarks. These include alterations involving \u003cem\u003eMYB/MYBL1\u003c/em\u003e, \u003cem\u003eZFTA/YAP1\u003c/em\u003e, \u003cem\u003eMN1\u003c/em\u003e, \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e, \u003cem\u003eFGFR1::TACC1\u003c/em\u003e, and \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e. \u003cem\u003eMYB\u003c/em\u003e/\u003cem\u003eMYBL1\u003c/em\u003e-altered gliomas comprise\u0026thinsp;~\u0026thinsp;2% of pediatric low-grade gliomas (pLGG) and exhibit uniform glial morphology and low mitotic activity [\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], with fusions involving partners such as \u003cem\u003ePCDHGA1\u003c/em\u003e, \u003cem\u003eMMP16\u003c/em\u003e, and \u003cem\u003eMAML2\u003c/em\u003e [\u003cspan additionalcitationids=\"CR23\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Pediatric supratentorial ependymomas (pSTE) are divided into \u003cem\u003eZFTA\u003c/em\u003e- and \u003cem\u003eYAP1\u003c/em\u003e-rearranged subtypes, which differ by age distribution and L1CAM/p65 expression. \u003cem\u003eZFTA\u003c/em\u003e fusions (commonly with \u003cem\u003eRELA\u003c/em\u003e) occur mainly in adolescents, whereas \u003cem\u003eYAP1\u003c/em\u003e rearrangements (e.g., with \u003cem\u003eMAML2\u003c/em\u003e, \u003cem\u003eNCOA1/2\u003c/em\u003e) are more frequent in younger children [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR26 CR27 CR28 CR29\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. \u003cem\u003eMN1\u003c/em\u003e-rearranged astroblastomas (pABM) typically arise in children and young adults, with a female predominance. Histologically, they exhibit perivascular pseudorosettes and are defined by fusions of \u003cem\u003eMN1\u003c/em\u003e with \u003cem\u003eBEND2\u003c/em\u003e or other partners (e.g., \u003cem\u003eCXXC5\u003c/em\u003e, \u003cem\u003eKMT2A\u003c/em\u003e), though \u003cem\u003eMN1\u003c/em\u003e rearrangements are not exclusive to this tumor type [\u003cspan additionalcitationids=\"CR33 CR34 CR35 CR36 CR37 CR38\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e fusions, affecting the MAPK/ERK pathway, are present in ~\u0026thinsp;80% of pilocytic astrocytomas (PA), the most common pediatric glioma [\u003cspan additionalcitationids=\"CR41 CR42\" citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. These fusions are also common in diffuse leptomeningeal glioneuronal tumors (DLGNT) (~\u0026thinsp;65%), a rare category of tumors with oligodendroglial and neuronal features [\u003cspan additionalcitationids=\"CR44 CR45\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Extraventricular neurocytomas (EVN), which show mixed neuronal-glial features, frequently harbour \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusions (~\u0026thinsp;70%), which are essential for diagnosis [\u003cspan additionalcitationids=\"CR48 CR49 CR50 CR51\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. Papillary glioneuronal tumors (PGNT), although extremely rare, are defined by the \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e fusion [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. Unlike other entities, PGNTs may lack a distinct methylation profile [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eDetection of these SVs in formalin-fixed, paraffin-embedded (FFPE) tissue relies on several methods such as fluorescence in situ hybridization (FISH), real-time/digital PCR, RNA sequencing, and DNA methylation profiling [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9 CR10 CR11\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Among these, FISH is considered a gold standard for detecting SVs in FFPE samples due to its high sensitivity and single-cell resolution, enabling correlation with tumor morphology even in presence of significant heterogeneity. However, its limitations include restriction to predefined targets, intra-/inter-observer variability, and lack of standardized interpretation criteria for pediatric brain tumors [\u003cspan additionalcitationids=\"CR14 CR15\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Conversely, sequencing and array-based platforms offer broader detection of genetic variants but are affected by degraded FFPE nucleic acids and intratumoral heterogeneity, reducing analytical sensitivity and consistency [\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThis study aims to improve FISH-based diagnostics by introducing a novel single-cell approach that incorporates quantitative signal distance measurement and pattern analysis for the detection of SVs in PBTs. We analyze key rearrangements and fusions \u003cb\u003e(Supplementary Table\u0026nbsp;1)\u003c/b\u003e, validating results with transcriptomic and epigenetic data. Our integrated approach enhances the diagnostic accuracy of FISH and enables discovery of novel or rare genetic variants in PBTs.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003ch2\u003eCohort-Samples Selection\u003c/h2\u003e\n\u003cp\u003eThis study analyzed a cohort of 50 PBTs, representing a range of different diagnostic classes. FFPE samples were obtained as part of Italy\u0026rsquo;s national centralization program for pediatric brain tumors, coordinated by the Neuropathology Unit (Department of pathological anatomy, Sapienza University of Rome). Clinical and histopathological data, including age at diagnosis, gender, and tumor anatomical location, were documented for all patients and are available in\u0026nbsp;Supplementary Tab. 2. Ethical approval was granted in accordance with the Declaration of Helsinki, and informed consent was obtained from patients\u0026apos; families. Anonymization protocols ensured patient confidentiality. Histopathological diagnoses were established through morphological examination of hematoxylin and eosin (H\u0026amp;E)-stained sections, supported by immunohistochemical (IHC) analysis (see Supplementary Methods for details). Our integrative approach combined morphology, immunophenotype, and molecular genetics to achieve tumor classification and identify novel rearrangements.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u0026ldquo;Single-Cell Signal Distance Measurement\u0026rdquo; for Rearrangements Detection using Break-Apart Probes\u003c/h2\u003e\n\u003cp\u003eA comprehensive description of the FISH protocol, including probe specifications and analytical workflow, is available in the Supplementary Methods. Break-apart probes are designed to hybridize to the 5\u0026prime; and 3\u0026prime; regions flanking the genomic breakpoint of interest. In the presence of a rearrangement, these probes yield separated signals\u0026mdash;typically one green (G) and one orange (O) signal with or without a residual fusion (1F/1G/1O)\u0026mdash;whereas intact alleles in non-rearranged cells display two co-localized fusion signals (2F) (Fig. 1A).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eI.\u0026nbsp; \u0026nbsp; \u0026nbsp;Challenges and solution for signal distance quantification\u003c/p\u003e\n\u003cp\u003eAlthough the guidelines recommend that separated signals should be at least twice the diameter of a single signal to confirm a rearrangement, accurately quantifying this distance presents significant challenges in routine diagnostic practice. To overcome this limitation, we developed and implemented a novel single-cell image analysis method using ImageJ (NIH, Bethesda, MD, USA). This technique provides precise and reproducible measurements of the distances between signals, ensuring greater accuracy and standardization in the assessment of genomic rearrangements.\u003c/p\u003e\n\u003cp\u003eII.\u0026nbsp; \u0026nbsp;\u0026nbsp;Image acquisition and pre-processing\u003c/p\u003e\n\u003cp\u003eImages were acquired under standardized conditions using constant exposure times on the fluorescence microscope to ensure data consistency. Ten images per sample were captured at 100\u0026times; magnification, allowing for the evaluation of at least 200 nuclei. Individual neoplastic cells were isolated using the zoom tool in the ISIS MetaSystems software. Background subtraction was performed to enhance fluorescent signals and reduce nonspecific noise before saving the images in JPG format.\u003c/p\u003e\n\u003cp\u003eIII.\u0026nbsp; \u0026nbsp;Signal distance measurement using ImageJ\u003c/p\u003e\n\u003cp\u003eThe distance between the orange and green signals, indicative of fused or separated probes, was measured using ImageJ. For each single-cell image acquired, the \u0026quot;line selection\u0026quot; tool was used to draw straight lines between the centers of the corresponding orange/green probes. The line lengths, expressed in pixels, were calculated using the \u0026quot;Analyze \u0026rarr; Measure\u0026quot; function (Fig. 1B). To minimize variability, two independent researchers (S.M. and F.R.B.) performed the measurements on the same digital images. Length values (LVs), representing precise measurements of signal distances, were recorded, ranging from 12 pixels (minimum distance indicating signal fusion) to 348 pixels (maximum separation indicating rearrangement).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIV.\u0026nbsp; \u0026nbsp;Data Collection, Statistical Analysis, and Interpretation on Control-Samples\u003c/p\u003e\n\u003cp\u003eFor each Break-apart probe, ten control samples were analyzed, including five cases with confirmed rearrangements and five negative cases, as determined by FISH and methylation profiling. From each control sample, ten representative cells with variable signal distances were selected and measured using the previously described method, resulting in a total of 200 signal distance (LV) measurements per probe. All LV data were processed using MedCalc statistical software (version 19.2.6, MedCalc Software bv, Belgium, https://www.medcalc.org) and are available in the Supplementary Data.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInitially, the data were graphically analyzed to visualize the distribution of distance measurements using density plots. Subsequently, the K-Means elbow method was applied to determine the optimal number of clusters. The K-Means elbow method is a statistical technique used to partition data into distinct groups based on numerical characteristics. It identifies the optimal number of clusters by detecting the point at which adding more clusters no longer significantly improves model performance. The analysis followed these steps: clustering using the K-Means algorithm for various numbers of clusters (up to 10), calculating the Within-Cluster Sum of Squares (WCSS), and identifying the \u0026quot;elbow\u0026quot; point, where further increases in the number of clusters do not substantially enhance the model.\u003c/p\u003e\n\u003cp\u003eUsing the K-Means elbow method, the optimal number of clusters was determined to be three (K=3) for each probe. The mean distortion, defined as the average of the squared distances between each data point and the centroid of its respective cluster, decreased significantly before stabilizing (Fig. 1C, representing \u003cem\u003eMYBL1\u003c/em\u003e). Three distinct groups were observed (Fig. 1D): Cluster 1, which represents normal signals and includes the lowest values; Cluster 2, which includes intermediate values; and Cluster 3, representing rearranged signals and displaying the highest values. This three-cluster distribution pattern was consistently observed across all Break-apart probes.\u003c/p\u003e\n\u003cp\u003eV.\u0026nbsp; \u0026nbsp;\u0026nbsp;Thresholds Establishment for Break-Apart Probes\u003c/p\u003e\n\u003cp\u003eTo establish statistically significant cutoff values for distinguishing between rearranged and non-rearranged signals, an independent sample t-test was performed. A thresholding approach based on the mean \u0026plusmn; standard deviation (\u0026sigma;) was applied, with cutoffs defined as: mean of rearranged signals - 1\u0026sigma; and mean of non-rearranged signals + 1\u0026sigma;. The separation between the rearranged cluster (green) and the non-rearranged cluster (blue) was statistically significant for all Break-apart probes (p\u0026lt;0.001), with the following LV distributions: \u003cem\u003eMYB\u003c/em\u003e (190\u0026plusmn;64.2 vs. 47.2\u0026plusmn;15), \u003cem\u003eMYBL1\u003c/em\u003e (189.4\u0026plusmn;71.9 vs. 49.9\u0026plusmn;11), \u003cem\u003eZFTA\u003c/em\u003e (148.1\u0026plusmn;51.5 vs. 50.7\u0026plusmn;10.9), \u003cem\u003eYAP1\u003c/em\u003e (179.9\u0026plusmn;72.1 vs. 47.1\u0026plusmn;11.4), and \u003cem\u003eMN1\u003c/em\u003e (146.5\u0026plusmn;48.5 vs. 51.8\u0026plusmn;13.1).\u003c/p\u003e\n\u003cp\u003eThreshold lines drawn in the graphical representation clearly separate the two groups. For instance, in the case of \u003cem\u003eMYBL1\u003c/em\u003e, rearranged signals were defined as \u0026ge; 117 LV, while normal signals were \u0026le; 60 LV (Fig. 1D). An intermediate range (61\u0026ndash;116 for \u003cem\u003eMYBL1\u003c/em\u003e), defined as \u0026quot;uncertain\u0026quot;, includes LV values that are not statistically distinguishable from the other two clusters. The same methodology was applied to all other Break-apart probes, yielding the following thresholds: \u003cem\u003eMYB\u003c/em\u003e rearranged \u0026ge; 126 vs. normal \u0026le; 62, \u003cem\u003eZFTA\u003c/em\u003e rearranged \u0026ge; 96 vs. normal \u0026le; 61, \u003cem\u003eYAP1\u003c/em\u003e rearranged \u0026ge; 108 vs. normal \u0026le; 58, and \u003cem\u003eMN1\u003c/em\u003e rearranged \u0026ge; 98 vs. normal \u0026le; 65 (Supplementary Fig. 1).\u003c/p\u003e\n\u003cp\u003eVI.\u0026nbsp;\u0026nbsp;Analysis and Interpretation of Rearrangements on Cohort-Samples\u003c/p\u003e\n\u003cp\u003eTo classify tumors samples in our cohort analyzed with Break-apart probes, LV measurements were obtained from 100 representative neoplastic cells per sample. The mean signal distance value (mean LV) was then calculated for each sample and applied to threshold distribution graphs, in order to determine the presence/absence of genomic rearrangements (see Results section).\u003c/p\u003e\n\u003ch2\u003e\u0026ldquo;Single-Cell Signal Pattern Analysis\u0026rdquo; for Fusions Detection Using Break-Apart and Fusion Probes\u003c/h2\u003e\n\u003cp\u003eTo assess inter- and intra-chromosomal genetic fusions in PBTs using FISH, we applied a single-cell signal pattern analysis system. This methodology enables accurate fusion detection and offers standardized evaluation criteria, addressing a critical gap in current FISH diagnostic practices.\u003c/p\u003e\n\u003cp\u003eI. \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e Fusion\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e fusion arises from a tandem duplication at the 7q34 region, involving the 5\u0026prime; portion of \u003cem\u003eKIAA1549\u003c/em\u003e and the 3\u0026prime; portion of \u003cem\u003eBRAF\u003c/em\u003e. Breakpoints are typically observed in exons 15 or 16 of \u003cem\u003eKIAA1549\u003c/em\u003e and exons 9, 10, or 11 of \u003cem\u003eBRAF\u003c/em\u003e [21,43,62]. Due to their proximity on the same chromosome (spanning ~1500 Kb), conventional fusion probes are ineffective. To overcome this, we used a Break-Apart Tri-Color Probe that enables the detection of intra-chromosomal fusions by analyzing signal patterns. The probe set included: Aqua-labelled probe (533 Kb) targeting regions upstream and downstream of \u003cem\u003eKIAA1549\u003c/em\u003e, Orange-labelled probe (178 Kb) hybridizing the 3\u0026prime; flanking region of \u003cem\u003eBRAF\u003c/em\u003e, and Green-labelled probe (358 Kb) targeting the 5\u0026prime; flanking region of \u003cem\u003eBRAF\u003c/em\u003e (Supplementary Fig. 2A). At the single-cell level, the presence of the \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e intra-chromosomal fusion is indicated by the duplication of Aqua and Orange signals (O/Aq doublets)\u0026nbsp;along with a single Green signal, reflecting the tandem duplication event. Only nuclei displaying a normal signal pattern (1G/1O/1Aq) along with a fusion signal were considered for the analysis. Samples were classified as positive when \u0026ge;20% of neoplastic nuclei exhibited this specific doublet-signal pattern (2G/2O+1O/2Aq+1Aq). The absence of the fusion was indicated by a normal pattern with closely spaced single signals (2G/2O/2Aq).\u003c/p\u003e\n\u003cp\u003eII. \u003cem\u003eFGFR1::TACC1\u003c/em\u003e Fusion\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusion results from an in-frame tandem duplication at the 8p11.23 region, involving the 5\u0026prime; portion of \u003cem\u003eTACC1\u003c/em\u003e and the 3\u0026prime; portion of \u003cem\u003eFGFR1\u003c/em\u003e. Breakpoints typically occur at exon 7 of \u003cem\u003eTACC1\u003c/em\u003e and exon 17 of \u003cem\u003eFGFR1\u003c/em\u003e, leading to the expression of the fusion oncogene [51,63]. Similar to \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e, due to the proximity of these genes on the same chromosome (~300 Kb), conventional fusion probes are ineffective. To detect this fusion, a Break-Apart Probe targeting regions flanking \u003cem\u003eFGFR1\u003c/em\u003e was employed (Supplementary Fig. 2B). At the single-cell level, a normal pattern was characterized by two fused Orange/Green signals (2F), while the \u003cem\u003eFGFR1::TACC1\u003c/em\u003e intra-chromosomal fusion was identified by Green signal duplication. Samples were deemed positive if \u0026ge;20% of neoplastic nuclei displayed this \u0026nbsp;specific doublet-signal pattern (1F/1F+1G).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIII. \u003cem\u003eSLC44A1::PRKCA\u003c/em\u003e Fusion\u003c/p\u003e\n\u003cp\u003eThe t(9;17)(q31;q24) translocation results in the inter-chromosomal fusion of the 5\u0026prime; portion of \u003cem\u003eSLC44A1\u003c/em\u003e with the 3\u0026prime; portion of \u003cem\u003ePRKCA\u003c/em\u003e, with breakpoints typically occurring between exons 14\u0026ndash;15 of \u003cem\u003eSLC44A1\u003c/em\u003e and exons 8\u0026ndash;9 of \u003cem\u003ePRKCA\u003c/em\u003e, leading to the expression of the fusion oncogene [55,56,64]. The \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e fusion involves genes located on separate chromosomes, making it detectable using conventional Fusion Probes. The probe set included a Green-labelled probe (525 Kb) targeting \u003cem\u003ePRKCA\u003c/em\u003e at 17q24.2 and an Orange-labelled probe (328 Kb) targeting \u003cem\u003eSLC44A1\u003c/em\u003e at 9q31.1 (Supplementary Fig. 2C). At the single-cell level, a normal pattern was defined by separated Green and Orange signals (2G/2O), while the presence of fusion was indicated by two separated and one juxtaposed-fused signals (1F/1O/1G). Samples were considered positive when \u0026ge;20% of neoplastic nuclei exhibited this specific fusion-signal pattern.\u003c/p\u003e\n\u003ch2\u003eWhole-Transcriptome Sequencing for Evaluation of Genomic Rearrangements and Fusions\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eSixteen selected samples were prepared from FFPE tissue sections through manual microdissection of neoplastic areas selected from H\u0026amp;E-stained slides. Total RNA was extracted using the miRNeasy FFPE Kit (Qiagen), following the manufacturer\u0026apos;s protocol. RNA quality and concentration were evaluated using the Qubit 4 Fluorometer with the RNA Broad Range kit (Thermo Fisher Scientific) and the TapeStation 4150 with the RNA ScreenTape kit (Agilent). Only samples with an RNA Integrity Number (RIN) greater than 2 were included for sequencing. Whole-transcriptome libraries were prepared using the RNA Prep with Enrichment Tagmentation Kit (Illumina). The protocol included RNA denaturation, cDNA synthesis, tagmentation (fragmenting cDNA and adding adapter sequences via bead-linked transposomes), PCR amplification with IDT sample indices, and library purification. The concentration of the libraries was quantified using the Qubit 4 Fluorometer with the dsDNA High Sensitivity kit (Thermo Fisher Scientific), and fragment size distributions were evaluated using the TapeStation 4150 with the High Sensitivity DNA ScreenTape kit (Agilent). The prepared libraries were sequenced on the high-throughput NovaSeq 6000 platform (Illumina) to generate paired-end reads. Raw data were subjected to quality control using FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Reads were aligned to the GRCh38 reference genome using the STAR aligner [66]. Structural variant detection was performed using specialized software, including STAR-Fusion [67] and FusionCatcher [68], to analyze chimeric reads and junction points aligned to the reference genome. Fusion transcripts were annotated to identify the involved genes and exon boundaries, and fusion events were analyzed using the R package chimeraviz (Bioconductor/R package version 1.30.0). Only relevant fusion transcripts were annotated, enabling quantitative assessment of each transcript. True positive (TP) and false positive (FP) events were counted based on a minimum threshold of supporting reads for each fusion. A stringent filtering algorithm was applied to variant calls to determine fusion events, ensuring the highest predictive accuracy.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMoreover, genome-wide DNA methylation profiling was performed on 23 of 50 FFPE tumors using the Illumina EPIC 850K array to support tumor classification and CNV analysis, with detailed methodology provided in the Supplementary Methods.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eSingle-Cell Signal Distance Analysis of\u003c/b\u003e \u003cb\u003eMYB/MYBL1\u003c/b\u003e \u003cb\u003eRearrangements Using FISH\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAs previously described, to establish detection thresholds, we analyzed 200 single-cell signal distances in control-samples, identifying two statistically significant clusters for \u003cem\u003eMYB\u003c/em\u003e: rearranged (190\u0026thinsp;\u0026plusmn;\u0026thinsp;64.2) and normal (47.2\u0026thinsp;\u0026plusmn;\u0026thinsp;15), with an intermediate \u0026ldquo;uncertain\u0026rdquo; range (63\u0026ndash;125) (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e). Subsequently, ten pediatric diffuse astrocytoma cohort-samples, initially diagnosed based on morphological and immunohistochemical features, were analyzed for \u003cem\u003eMYB/MYBL1\u003c/em\u003e rearrangements using Break-apart probes. Mean signal distances (Mean LVs) were calculated across 100 representative neoplastic cells per cohort-sample. Four cases (40%) exhibited a break pattern (1F/1G/1O) in \u0026ge;\u0026thinsp;30% of tumor cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB), with two exceeding the rearrangement threshold (\u0026gt;\u0026thinsp;126) and two within the uncertain range (LVs: 100 and 94) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). RNA sequencing confirmed rearrangements in the latter (\u003cb\u003eSupplementary Fig.\u0026nbsp;7A\u003c/b\u003e). Five tumors (50%) showed a normal \u003cem\u003eMYB\u003c/em\u003e pattern with two fused signals (2F) and mean LVs below the threshold (\u0026lt;\u0026thinsp;63) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Interestingly, one tumor (10%) displayed a previously unreported atypical FISH pattern (2F\u0026thinsp;+\u0026thinsp;1O), characterized by gain of the 3' Orange probe (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE), suggestive of an unbalanced \u003cem\u003eMYB\u003c/em\u003e rearrangement, which was confirmed by additional methods.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eA similar approach was applied to \u003cem\u003eMYBL1\u003c/em\u003e, identifying rearranged (189.4\u0026thinsp;\u0026plusmn;\u0026thinsp;71.9) and normal (49.9\u0026thinsp;\u0026plusmn;\u0026thinsp;11) clusters in control-samples, with an uncertain range of 61\u0026ndash;116. Subsequently, mean LVs were calculated in the ten cohort-samples, revealing \u003cem\u003eMYBL1\u003c/em\u003e rearrangements in three cases (LV\u0026thinsp;\u0026gt;\u0026thinsp;117, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Seven cases (70%) were not rearranged, with six showing mean LVs below the threshold (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC) and one in the uncertain range (LV: 62, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). The latter was analyzed by RNA sequencing to verify the absence of rearrangement, which was indeed confirmed (\u003cb\u003eSupplementary Fig.\u0026nbsp;7B\u003c/b\u003e). \u003cem\u003eMYB\u003c/em\u003e and \u003cem\u003eMYBL1\u003c/em\u003e rearrangements were mutually exclusive across all tumors analyzed. The RNA sequencing validation of uncertain cases refined our cutoff values, establishing more stringent thresholds for the uncertainty range (\u003cem\u003eMYB\u003c/em\u003e rearrangement: \u0026gt;93 vs. non-rearranged: \u0026gt;63; \u003cem\u003eMYBL1\u003c/em\u003e rearrangement: \u0026gt;117 vs. non-rearranged: \u0026gt;63) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eSingle-Cell Signal Distance Analysis of\u003c/b\u003e \u003cb\u003eZFTA/YAP1\u003c/b\u003e \u003cb\u003eRearrangements Using FISH\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAs previously described, we conducted the analysis of 200 single-cell signal distance measurements in control-samples to establish thresholds for detecting \u003cem\u003eZFTA\u003c/em\u003e and \u003cem\u003eYAP1\u003c/em\u003e rearrangements, revealing two distinct clusters: \u003cem\u003eZFTA\u003c/em\u003e-rearranged (148.1\u0026thinsp;\u0026plusmn;\u0026thinsp;51.5) vs. normal signals (50.7\u0026thinsp;\u0026plusmn;\u0026thinsp;10.9) and \u003cem\u003eYAP1\u003c/em\u003e-rearranged (179.9\u0026thinsp;\u0026plusmn;\u0026thinsp;72.1) vs. normal signals (47.1\u0026thinsp;\u0026plusmn;\u0026thinsp;11.4). Signals classified as uncertain were also observed within the range of 62\u0026ndash;95 for \u003cem\u003eZFTA\u003c/em\u003e and 59\u0026ndash;107 for \u003cem\u003eYAP1\u003c/em\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e). Subsequently, twelve pediatric supratentorial ependymoma cohort-samples, initially diagnosed based on morphological and immunohistochemical features, were analyzed for \u003cem\u003eZFTA\u003c/em\u003e/\u003cem\u003eYAP1\u003c/em\u003e rearrangements using Break-apart probes. Mean LVs were calculated across 100 representative neoplastic cells per cohort-sample. Five cases (41.6%) exhibited a \u003cem\u003eZFTA\u003c/em\u003e break pattern (1F/1G/1O) in \u0026ge;\u0026thinsp;30% of tumor cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), with four exceeding the rearrangement threshold (\u0026gt;\u0026thinsp;96) and one falling within the uncertain range (91). This latter case (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), classified as borderline positive due to its proximity to the threshold, was confirmed as positive via RNA sequencing \u003cb\u003e(Supplementary Fig.\u0026nbsp;7E\u003c/b\u003e). Five samples (41.6%) showed no rearrangement (2F) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), with four below the normal cutoff (\u0026lt;\u0026thinsp;55) and one borderline (LV: 66), confirmed as negative by RNA sequencing (\u003cb\u003eSupplementary Fig.\u0026nbsp;7F\u003c/b\u003e). Notably, two cases (16.8%) exhibited atypical FISH patterns not previously described in the literature: one with a gain of the 5' Green probe (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE) and one with a gain of the 3' Orange probe (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). These patterns, indicative of unbalanced \u003cem\u003eZFTA\u003c/em\u003e rearrangements, were both confirmed through additional molecular analyses.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThree cases (25%) harboured \u003cem\u003eYAP1\u003c/em\u003e rearrangements (1F/1G/1O) in \u0026gt;\u0026thinsp;40% of cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Two had mean LVs\u0026thinsp;\u0026gt;\u0026thinsp;108, while one was borderline (mean LV: 92). RNA sequencing confirmed rearrangements in the latter (\u003cb\u003eSupplementary Fig.\u0026nbsp;7C\u003c/b\u003e). Seven cases (70%) showed no evidence of \u003cem\u003eYAP1\u003c/em\u003e rearrangement, displaying two fused signals (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Six of these had mean LVs below the normal threshold (\u0026lt;\u0026thinsp;58), while one case fell within the uncertain range (62) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD), confirmed as negative through RNA sequencing (\u003cb\u003eSupplementary Fig.\u0026nbsp;7F\u003c/b\u003e). Thresholds were refined through RNA sequencing validation of uncertain cases: \u003cem\u003eZFTA\u003c/em\u003e rearrangement\u0026thinsp;\u0026gt;\u0026thinsp;90 vs. non-rearranged\u0026thinsp;\u0026lt;\u0026thinsp;66 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), \u003cem\u003eYAP1\u003c/em\u003e rearrangement\u0026thinsp;\u0026gt;\u0026thinsp;91 vs. non-rearranged\u0026thinsp;\u0026lt;\u0026thinsp;6 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Remarkably, one tumor harboured concurrent \u003cem\u003eZFTA\u003c/em\u003e and \u003cem\u003eYAP1\u003c/em\u003e rearrangements, indicating intratumoral heterogeneity and the necessity of comprehensive molecular profiling in pediatric brain tumors.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eSingle-Cell Signal Distance Analysis of\u003c/b\u003e \u003cb\u003eMN1\u003c/b\u003e \u003cb\u003eRearrangements Using FISH\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAs previously described, single-cell signal distance analysis of control-samples revealed two distinct \u003cem\u003eMN1\u003c/em\u003e clusters: rearranged (146.5\u0026thinsp;\u0026plusmn;\u0026thinsp;48.5) and normal signals (51.8\u0026thinsp;\u0026plusmn;\u0026thinsp;13.1), with a statistically significant difference. An additional group of uncertain signals (66\u0026ndash;97) was also observed. Subsequently eight pediatric tumors, immunophenotypically consistent with astroblastoma, were analyzed using Break-apart probes. Mean signal distances were calculated on 100 representative neoplastic cells for each cohort-sample.\u003c/p\u003e\u003cp\u003e\u003cem\u003eMN1\u003c/em\u003e rearrangement was detected in four cases (50%), characterized by two separated spots (1F/1G/1O) in \u0026gt;\u0026thinsp;40% of cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). All these exhibited mean LVs exceeding the rearrangement threshold (\u0026gt;\u0026thinsp;97). The other four cases (50%) showed a normal pattern (2F) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC), with mean LVs below the normal threshold (\u0026lt;\u0026thinsp;66).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNotably, none of the \u003cem\u003eMN1\u003c/em\u003e analyzed tumors exhibited unbalanced patterns or showed borderline mean LVs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA), reflecting a higher uniformity of signal distance measurements for \u003cem\u003eMN1\u003c/em\u003e rearrangements using FISH, particularly when compared to other rearrangements.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSingle-Cell Signal Pattern Analysis of\u003c/b\u003e \u003cb\u003eKIAA1549::BRAF\u003c/b\u003e \u003cb\u003eFusion Using FISH\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe developed a single-cell signal pattern analysis method to establish standardized evaluation criteria for \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e fusion, addressing a gap in the current literature. Using this approach, we analyzed ten tumors with clinical, morphological, and immunohistochemical features consistent with pilocytic astrocytomas (7) or diffuse leptomeningeal glioneuronal tumors (3), employing Break-Apart Tri-Color probes as previously described (\u003cb\u003eSupplementary Fig.\u0026nbsp;2A\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eThe characteristic fusion pattern was defined by the presence of juxtaposed Orange and Aqua signals with gain of Orange (3' \u003cem\u003eBRAF\u003c/em\u003e) and Aqua (\u003cem\u003eKIAA1549\u003c/em\u003e) signals (e.g., 2G/2O\u0026thinsp;+\u0026thinsp;1O/2Aq\u0026thinsp;+\u0026thinsp;1Aq), indicating duplication events (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, \u003cb\u003epattern 2\u003c/b\u003e). In some cases, partial probe binding resulted in patterns lacking duplicated Aqua (2G/2O\u0026thinsp;+\u0026thinsp;1O/2Aq) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, \u003cb\u003epattern 1\u003c/b\u003e). In contrast, the normal pattern consisted of two single Green, Orange, and Aqua signals (2G/2O/2Aq), without doublets or signal overlap (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). To ensure accurate interpretation, fluorescence channels and focal planes were individually acquired and visualized, facilitating clear differentiation between fusion-specific doublets and the single signals of normal pattern.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eUsing this single-cell signal pattern analysis, 200 neoplastic nuclei per cohort-sample were evaluated. Six out of ten cases (60%) showed the fusion pattern in more than 30% of cells. Two samples, with 20\u0026ndash;22% positive nuclei respectively, were close to the positivity threshold (20%). These were classified as positive based on FISH pattern analysis and subsequently validated by second-level methods (\u003cb\u003eSupplementary Fig.\u0026nbsp;7D\u003c/b\u003e). The remaining two cases were negative for the fusion. This study is the first to establish objective and reproducible FISH-based standardized criteria for detecting \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e fusion in pediatric brain tumors.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSingle-Cell Signal Pattern Analysis of\u003c/b\u003e \u003cb\u003eFGFR1::TACC1\u003c/b\u003e \u003cb\u003eFusion Using FISH\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe applied the same single-cell analytical approach to define standardized criteria for \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusion detection, addressing the lack of clear guidelines for assessing this fusion via FISH in the current literature. Six tumors with histopathological features of extraventricular neurocytoma were analyzed using Break-Apart probes targeting the duplicated region implicated in the fusion (\u003cb\u003eSupplementary Fig.\u0026nbsp;2B\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eThe \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusion displayed a distinctive FISH pattern characterized by Green doublets (1F/1F\u0026thinsp;+\u0026thinsp;1G), resulting from the in-frame tandem duplication in the region 8p11 (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB). In contrast, the normal pattern showed two fused signals (2F) without duplication (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). Accurate interpretation of this fusion required the visualization of individual fluorescence channels and focal planes, enabling precise identification of the additional Green signal that formed the doublets. By comparison, fluorescence channels of the normal pattern displayed single probes for each fluorochrome, with no evidence of doublets.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eUsing this approach, 200 neoplastic nuclei per cohort-sample were evaluated. The \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusion was identified in 4 out of 6 tumors (67%), while 2 cases (33%) were negative. One borderline sample, with 20% of cells exhibiting the fusion pattern, was near the positivity threshold and was subsequently analyzed via RNA sequencing, which failed to detect this fusion target (\u003cb\u003eSupplementary Fig.\u0026nbsp;6C\u003c/b\u003e). This lack of detection is likely attributable to high intratumoral heterogeneity and low proportion of cells harbouring the variant.\u003c/p\u003e\u003cp\u003eNotably, this is the first study establishing reproducible and standardized FISH criteria for \u003cem\u003eFGFR1::TACC1\u003c/em\u003e, improving the reliability of fusion detection in PBTs.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSingle-Cell Signal Pattern Analysis of\u003c/b\u003e \u003cb\u003ePRKCA::SLC44A1\u003c/b\u003e \u003cb\u003eFusion Using FISH\u003c/b\u003e\u003c/p\u003e\u003cp\u003eIn contrast to \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e and \u003cem\u003eFGFR1::TACC1\u003c/em\u003e, the \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e fusion results from an inter-chromosomal rearrangement, making it well-suited for detection using conventional dual-color Fusion probes (\u003cb\u003eSupplementary Fig.\u0026nbsp;2C\u003c/b\u003e). To enhance diagnostic accuracy, we established standardized evaluation criteria for \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e detection based on a single-cell signal pattern analysis. We analyzed four samples exhibiting clinical, morphological, and immunohistochemical features suggestive of papillary glioneuronal tumors.\u003c/p\u003e\u003cp\u003eThe normal pattern displayed two separated Green and Orange signals (2G/2O), reflecting the absence of a fusion (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eA). Fusion-positive nuclei were defined by directly juxtaposed and overlapping Orange and Green signals (1F/1O/1G) (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eB), confirmed only when direct contact between probes was observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eC). In addition, to further refine detection accuracy, imaging analysis was conducted across multiple single focal planes to precisely assess the spatial localization of the probes within the nuclei. True-positive signals were consistently captured and observed within the same focal plane (\u003cb\u003eSupplementary Fig.\u0026nbsp;3B\u003c/b\u003e). Conversely, false-positive signals, arising from random folding of genomic DNA, were detected across different focal planes (\u003cb\u003eSupplementary Fig.\u0026nbsp;3A\u003c/b\u003e). This stringent requirement helped avoid false positives from nuclear DNA folding artifacts, which produce apparent overlap across different focal planes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eA total of 200 neoplastic nuclei per cohort-sample were analyzed using this single-cell signal pattern analysis. Of the four tumors, one was positive, with a clear fusion pattern in over 30% of nuclei, while two cases were classified as negative. One sample exhibited the fusion pattern in only 13% of cells, below the diagnostic threshold (20%). Consequently, this tumor was classified as negative for \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e and was further analyzed by RNA sequencing for additional validation, which did not detect the fusion target (\u003cb\u003eSupplementary Fig.\u0026nbsp;7H\u003c/b\u003e). This underscores the limitations of RNA-seq for identifying variants present in low-frequency subclones. Notably, this approach establishes robust and standardized criteria for FISH-based detection of \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e, addressing the current lack of established guidelines in literature for assessing this rare fusion in PBTs.\u003c/p\u003e\u003cp\u003e\u003cb\u003eCorrelation Between FISH and Genomic DNA Methylation Profiling\u003c/b\u003e\u003c/p\u003e\u003cp\u003eGenomic DNA methylation profiling enables accurate subclassification of PBTs based on epigenetic signatures and provides complementary information on copy number variations (CNVs) [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Although this method does not directly detect structural variants, it can support the interpretation of fusions and rearrangements identified by FISH. Therefore, methylation profiling was performed on 23 out of 50 tumors (46%) with adequate DNA input (\u0026ge;\u0026thinsp;500 ng), independently of FISH outcomes, to indirectly assess the reliability of the signal distance measurement and pattern analysis. Of these, 21 cases (91.3%) showed complete concordance between FISH-detected rearrangements and methylation-defined epigenetic subtypes \u003cb\u003e(Supplementary Table\u0026nbsp;3\u003c/b\u003e). These included cases with unbalanced rearrangements in \u003cem\u003eMYB\u003c/em\u003e and \u003cem\u003eZFTA\u003c/em\u003e, which were correctly classified as Low-grade glioma with \u003cem\u003eMYB\u003c/em\u003e rearrangement (LGG MYB) and Supratentorial ependymoma with \u003cem\u003eZFTA::RELA\u003c/em\u003e fusion (ST-RELA), respectively. These cases exhibited FISH patterns not previously reported, reinforcing the importance of recognizing unbalanced rearrangements in FISH interpretation. All concordant samples had epigenetic scores\u0026thinsp;\u0026gt;\u0026thinsp;0.8, emphasizing the reliability of tumor classification based on epigenetic data [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn two samples (8.7%), the methylation subclass assigned was inconsistent with FISH results: one \u003cem\u003eZFTA\u003c/em\u003e-rearranged ependymoma was misclassified as a low-grade glioma, and one \u003cem\u003eMN1\u003c/em\u003e-rearranged astroblastoma was misclassified as a pleomorphic xanthoastrocytoma (PXA). Both tumors had low epigenetic scores below the threshold of 0.8 (0.55 and 0.46), indicating uncertain predictions due to insufficiently distinctive epigenetic features [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. These discrepancies were resolved through RNA sequencing, which confirmed the rearrangements identified by FISH, underscoring methylation profiling limitations in low-score cases. Methylation clustering for all 23 tumors is showed in \u003cb\u003eSupplementary Fig.\u0026nbsp;4\u003c/b\u003e. CNVs profiling did not demonstrate correlation with specific rearrangements, revealing that such variants lead to heterogeneous and non-specific copy number alterations. Notably, \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e and \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusions corresponded to focal gains at 7q34 and 8p11 respectively, reflecting the duplications underlying these intra-chromosomal fusions (\u003cb\u003eSupplementary Fig.\u0026nbsp;5\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003cb\u003eCorrelation Between FISH and Whole Transcriptome Sequencing\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo validate the accuracy of our single-cell FISH methods, specifically signal distance measurement and signal pattern analysis, we performed whole-transcriptome sequencing on 16 selected samples. These included borderline cases with mean signal distances near the rearranged/normal thresholds, as well as those exhibiting high intratumoral heterogeneity or fusion patterns close to the positivity cutoff.\u003c/p\u003e\u003cp\u003eRNA-seq confirmed FISH findings in 13 out of 16 cases (81.2%), supporting strong concordance between the two methods, even in diagnostically challenging samples. Three cases (18.8%) showed discrepancies, as detailed in \u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e. In two samples, one with \u003cem\u003eMYB\u003c/em\u003e rearrangement and the other with \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e fusion identified by FISH, no corresponding fusion transcripts were detected by RNA-seq (\u003cb\u003eSupplementary Fig.\u0026nbsp;6A and 6B\u003c/b\u003e). Notably, genome-wide analysis failed to identify any high-confidence fusion events in either sample, suggesting that the sequencing reaction may have been compromised by RNA degradation or poor sample quality. In the third discordant case, a \u003cem\u003eFGFR1::TACC1\u003c/em\u003e fusion detected by FISH in 20% of tumor cells was not validated by RNA-seq (\u003cb\u003eSupplementary Fig.\u0026nbsp;6C\u003c/b\u003e). This likely reflects the low abundance of the fusion-positive subclone and the high degree of intratumoral heterogeneity, underscoring a known limitation of RNA-seq in detecting low-frequency variants in genetically heterogeneous tumors [\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Despite these exceptions, RNA-seq confirmed the majority of borderline/ambiguous cases and supported the refinement of FISH thresholds for signal distances and patterns interpretation.\u003c/p\u003e\u003cp\u003eNotably, a novel \u003cem\u003eMYB::CA10\u003c/em\u003e fusion was identified by RNA-seq, resulting from an inter-chromosomal translocation t(6;17)(q23.3;q21.22) and marking the first report of \u003cem\u003eCA10\u003c/em\u003e as fusion partner of \u003cem\u003eMYB\u003c/em\u003e. \u003cem\u003eCA10\u003c/em\u003e encodes a protein in the carbonic anhydrase family involved in CNS development and previously associated with colorectal and breast cancers [\u003cspan additionalcitationids=\"CR70\" citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e], suggesting a potential novel fusion partner in \u003cem\u003eMYB\u003c/em\u003e-rearranged pediatric astrocytomas.\u003c/p\u003e\u003cp\u003eSequencing analysis confirmed the absence of \u003cem\u003eMYB\u003c/em\u003e/\u003cem\u003eMYBL1\u003c/em\u003e rearrangements in a FISH-negative tumor, validating the accuracy of the signal distance measurement method. Among the 17 somatic fusions identified in this sample, the \u003cem\u003eZFTA::RELA\u003c/em\u003e fusion transcript was found, leading to reclassification from low-grade glioma to \u003cem\u003eZFTA\u003c/em\u003e-rearranged supratentorial ependymoma and illustrating the diagnostic value of integrated molecular analysis.\u003c/p\u003e\u003cp\u003eAdditionally, RNA-seq also confirmed complex fusion patterns identified by FISH. In one tumor, both \u003cem\u003eZFTA::RELA\u003c/em\u003e and \u003cem\u003eYAP1::TYR\u003c/em\u003e fusions were detected, demonstrating the coexistence of two relevant structural variants in the same tumor. While \u003cem\u003eZFTA::RELA\u003c/em\u003e is a hallmark of supratentorial ependymomas, the identification of \u003cem\u003eTYR\u003c/em\u003e as a \u003cem\u003eYAP1\u003c/em\u003e fusion partner is novel. In another case, RNA-seq validated the \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e fusion detected by FISH and also revealed a concurrent \u003cem\u003eMN1::BEND2\u003c/em\u003e fusion, highlighting the potential for multiple concurrent rearrangements within a single tumor.\u003c/p\u003e\u003cp\u003eNotably, RNA-seq also identified a novel \u003cem\u003ePAX5::ATM\u003c/em\u003e fusion in a \u003cem\u003eZFTA/YAP1\u003c/em\u003e-negative supratentorial ependymoma, resulting from an inter-chromosomal t(9;11)(p13.2;q22.3) rearrangement. While both \u003cem\u003ePAX5\u003c/em\u003e and \u003cem\u003eATM\u003c/em\u003e alterations are well known in hematologic malignancies [\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e, \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e, \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e], this is the first report of \u003cem\u003ePAX5::ATM\u003c/em\u003e fusion in a pediatric brain tumor, representing a significant genetic finding in this context.\u003c/p\u003e\u003cp\u003eA Circos plot of all fusions is presented in \u003cb\u003eSupplementary Fig.\u0026nbsp;7\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eOverall, our results demonstrate that RNA-seq complements FISH by validating and extending the genomic characterization of PBTs, particularly in cases with borderline signal distances or complex fusion patterns. The integrated application of histopathology, FISH-based single-cell analysis, DNA methylation profiling, and RNA-seq enabled accurate and consistent molecular classification across all analyzed cases, in alignment with current WHO diagnostic criteria [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. This combined approach provides a robust framework for the diagnosis and subclassification of genetically heterogeneous pediatric brain tumors. \u003cb\u003eSupplementary Fig.\u0026nbsp;8\u003c/b\u003e presents a comprehensive overview of the clinical, cytogenetic, epigenetic, and transcriptomic data used for the formulation of the final integrated diagnoses.\u003c/p\u003e"},{"header":"Discussion and Conclusion","content":"\u003cp\u003ePediatric brain tumors encompass a heterogeneous group of neoplasms defined by distinct genetic and epigenetic alterations. Accurate diagnosis requires a multidisciplinary approach integrating histopathology, immunophenotyping, and molecular data from advanced methodologies [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Among genetic alterations, PBTs are characterized by a relatively low somatic mutation burden but a high prevalence of structural variants, including tandem duplications, inversions, fusions, and translocations, which play critical roles in their pathogenesis [\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eA variety of advanced technologies have been employed to detect SVs in PBTs, including reverse transcription or digital PCR, RNA next-generation sequencing, methylation profiling, and FISH [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], with each methodology having inherent advantages and limitations. FISH is rapid and highly sensitive technique capable of detecting genetic alterations at the single-cell level, making it particularly valuable for identifying SVs in heterogeneous tumor cell populations, where subclonal events may otherwise go undetected [\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. However, its application is restricted to preselected genomic targets, it is susceptible to intra- and inter-observer variability, and it currently lacks standardized evaluation criteria specific to PBTs [\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Conversely, RNA-seq allows for the comprehensive identification of multiple genomic fusions without prior knowledge of target genes, making it particularly useful for characterizing novel fusion events and complex structural rearrangements. Nevertheless, its effectiveness can be hampered by RNA degradation in formalin-fixed paraffin-embedded samples and by intratumoral genetic heterogeneity, which may obscure the detection of rare or subclonal rearrangements, reducing diagnostic sensitivity [\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. These complementary strengths and limitations underscore the importance of integrating multiple methodologies to achieve accurate and comprehensive characterization of PBTs, enabling improved diagnostic precision and insights into tumor biology. To address these challenges, we developed novel single-cell imaging methods for FISH, which provide objective and reproducible assessments of rearrangements and fusions. These methods were applied to a cohort of 50 pediatric patients with various CNS tumor subtypes, integrating RNA-seq and methylation profiling as complementary diagnostic tools.\u003c/p\u003e\u003cp\u003eUsing a newly developed \"distance measurement method\" for FISH Break-apart probes on control-samples, we established statistically significant thresholds (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) to differentiate normal from rearranged signals for key genes recognized as diagnostic markers for specific tumor subtypes in the WHO classification, including \u003cem\u003eMYB\u003c/em\u003e, \u003cem\u003eMYBL1\u003c/em\u003e, \u003cem\u003eZFTA\u003c/em\u003e, \u003cem\u003eYAP1\u003c/em\u003e, and \u003cem\u003eMN1\u003c/em\u003e [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. By measuring the average signal distance (mean LV) across 100 representative neoplastic cells per cohort sample using FISH and employing additional confirmation methods such as RNA-seq and methylation array, we detected positive and negative rearrangements with the following concordance rates: \u003cem\u003eMYB/MYBL1\u003c/em\u003e (90%), \u003cem\u003eZFTA/YAP1\u003c/em\u003e (91.5%), and \u003cem\u003eMN1\u003c/em\u003e (100%).\u003c/p\u003e\u003cp\u003eIn addition, we identified novel unbalanced rearrangements for \u003cem\u003eMYB\u003c/em\u003e and \u003cem\u003eZFTA\u003c/em\u003e using FISH, which have not been previously reported in the literature. These unbalanced patterns emphasize the importance of recognizing atypical rearrangements when interpreting FISH results. We also observed mutual exclusivity between \u003cem\u003eMYB\u003c/em\u003e and \u003cem\u003eMYBL1\u003c/em\u003e rearrangements, supporting their distinct roles in tumorigenesis. Notably, we documented the co-occurrence of \u003cem\u003eZFTA\u003c/em\u003e and \u003cem\u003eYAP1\u003c/em\u003e rearrangements within a single tumor, underscoring the genetic complexity and significant intratumoral heterogeneity of PBTs. This rare observation suggest that concurrent genetic events may contribute to PBT pathogenesis in specific cases.\u003c/p\u003e\u003cp\u003eTo further refine the characterization of genetic fusions, we implemented a \"single-cell signal pattern analysis\" method, enabling precise identification of distinct FISH patterns associated with key fusions, including \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e (dual orange/aqua signals), \u003cem\u003eFGFR1::TACC1\u003c/em\u003e (dual green signals), and \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e (juxtaposed orange/green signals). By analyzing single-cell patterns across 100 representative neoplastic cells per cohort-sample using FISH and employing RNA-seq and methylation array for confirmation, we detected positive and negative fusions with the following concordance rates: \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e (90%), \u003cem\u003eFGFR1::TACC1\u003c/em\u003e (83.5%), and \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e (100%). These methodological advancements provide standardized and objective criteria for the detection of structural variants in PBTs using FISH, significantly enhancing the diagnostic accuracy and minimizing interpretive variability.\u003c/p\u003e\u003cp\u003eIntegrative analyses demonstrated strong concordance between FISH and methylation profiling, with discrepancies observed in only two tumors exhibiting low methylation scores (\u0026lt;\u0026thinsp;0.6). These low methylation scores have been previously reported in the literature as potential indicators of misclassification within the epigenetic class and are directly associated with pre-analytical variability or genetic heterogeneity [\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. RNA-seq confirmed FISH findings in 81.2% of cases, with discrepancies observed in only three samples. These discrepancies were attributed to RNA degradation or significant intratumoral heterogeneity, factors that have been reported to contribute to the failure in detecting subclonal alterations by sequencing [\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. These results underscore the robustness of the novel FISH methods in detecting SVs in PBTs, even in challenging samples with borderline signal distances, degraded DNA/RNA, or high genetic variability. RNA-seq validation of FISH results in borderline cohort-samples, with mean signal distances near the thresholds, enabled the refinement of the uncertainty interval, establishing more precise and stringent cutoffs for detecting rearrangements. This process also contributed to the standardization of values for signal distances and pattern interpretation. However, additional validation using secondary methodologies is necessary when interpreting borderline FISH results in individual samples.\u003c/p\u003e\u003cp\u003eNotably, RNA-seq uncovered novel fusion partners, including \u003cem\u003eMYB::CA10\u003c/em\u003e from t(6;17)(q23.3;q21.22) and \u003cem\u003eYAP1::TYR\u003c/em\u003e from an intrachromosomal rearrangement at 11q22.1. These represent the first descriptions of these partners in association with \u003cem\u003eMYB\u003c/em\u003e and \u003cem\u003eYAP1\u003c/em\u003e. Additionally, we identified the \u003cem\u003ePAX5::ATM\u003c/em\u003e fusion, arising from t(9;11)(p13.2;q22.3), further expanding the spectrum of structural variants in PBTs. Importantly, the integration of FISH and RNA-seq reclassified a tumor initially diagnosed as a low-grade glioma to a \u003cem\u003eZFTA\u003c/em\u003e-rearranged supratentorial ependymoma, highlighting the critical role of a multi-methodologic molecular diagnostic approach. Notably, our identification of coexisting structural variants, such as \u003cem\u003eZFTA::RELA\u003c/em\u003e with \u003cem\u003eYAP1::TYR\u003c/em\u003e in one tumor and \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e with \u003cem\u003eMN1::BEND2\u003c/em\u003e in another, provides compelling evidence of intratumoral genetic heterogeneity of PBTs. These findings demonstrate the coexistence of distinct genetic subpopulations within single tumors, reflecting the complexity of PBTs biology and emphasizing the importance of comprehensive molecular profiling for accurate tumor classification.\u003c/p\u003e\u003cp\u003eIn conclusion, our single-cell FISH methodologies establish robust, standardized criteria for detecting structural variants in PBTs, minimizing interpretive variability and improving diagnostic precision. When combined with RNA-seq and methylation profiling, this integrated approach enables comprehensive tumor characterization, enhances subtype classification accuracy, and uncovers clinically relevant genetic intratumoral heterogeneity of PBTs. Although the number of analyzed cases limits our study, it represents the first comprehensive effort to standardize FISH-based detection of genetic rearrangements in pediatric central nervous system tumors, laying a critical foundation for future research and clinical applications.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003eThis work was supported by\u0026nbsp;Fondazione Italiana per la Lotta al Neuroblastoma ONLUS.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests:\u003c/strong\u003e The authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u003c/strong\u003e All authors contributed to the study conception and design. Material preparation, data collection and FISH analysis were performed by Simone Minasi and Francesca Romana Buttarelli. Chiara Dossena, Marica Ficorilli, and Matilde Oriani performed methylation profiling and RNA sequencing analyses, under the supervision of Loris De Cecco. Francesca Gianno and Manila Antonelli were responsible for all morphological and immuno-phenotypical evaluations. Maura Massimino collected all clinical data of the patients. The first draft of the manuscript was written by Simone Minasi and Francesca Romana Buttarelli, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConference Presentation:\u0026nbsp;\u003c/strong\u003ePart of these results will be accepted for abstract publication and poster presentation at the 20th Annual Meeting of the European Association of Neuro-Oncology (EANO 2025), to be held from October 16\u0026ndash;19, 2025, in Prague, Czech Republic (Control No.: 2025-A-352-EANO).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability:\u003c/strong\u003e The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval:\u0026nbsp;\u003c/strong\u003eThis study was performed in line with the principles of the Declaration of Helsinki. Anonymization protocols ensured patient confidentiality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate:\u003c/strong\u003e\u0026nbsp; Written informed consent was obtained from all the parents of individual participants included in the study.\u003cbr\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eOstrom QT, de Blank PM, Kruchko C et al (2015) Alex's Lemonade Stand Foundation Infant and Childhood Primary Brain and Central Nervous System Tumors Diagnosed in the United States in 2007\u0026ndash;2011. Neuro Oncol. 2015;16 Suppl 10(Suppl 10):x1-x36\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLouis DN, Perry A, Wesseling P et al (2021) The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Rev Neuro Oncol 23(8):1231\u0026ndash;1251\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePollack IF, Agnihotri S, Broniscer A (2019) Childhood brain tumors: current management, biological insights, and future directions. J Neurosurg Pediatr 23:261\u0026ndash;273\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLim-Fat MJ, Macdonald M, Lapointe S et al (2022) Molecular testing for adolescent and young adult central nervous system tumors: A Canadian guideline. Front Oncol. 2022;12:960509\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCapper D, Jones DTW, Sill M et al (2018) DNA methylation-based classification of central nervous system tumours. Nature 555(7697):469\u0026ndash;474\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKorshunov A, Sahm F, Zheludkova O et al (2019) DNA methylation profiling is a method of choice for molecular verification of pediatric WNT-activated medulloblastomas. Neuro Oncolol 21(2):214\u0026ndash;221\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDanielsson A, Nemes S, Tisell M et al (2015) MethPed: a DNA methylation classifier tool for the identification of pediatric brain tumor subtypes. Clin Epigenet. 2015;7:62\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGr\u0026ouml;bner SN, Worst BC, Weischenfeldt J et al (2018) ICGC PedBrain-Seq Project; ICGC MMML-Seq Project. The landscape of genomic alterations across childhood cancers. Nature. 2018;559(7714):E10\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang Y, Yang L (2023) Somatic structural variation signatures in pediatric brain tumors. medRxiv. 2023 May 26:2023.05.18.23290139. doi: 10.1101/2023.05.18.23290139. Update in: Cell Rep. 2023;42(10):113276. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.celrep.2023.113276\u003c/span\u003e\u003cspan address=\"10.1016/j.celrep.2023.113276\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang Y, Chen F, Donehower LA (2021) A pediatric brain tumor atlas of genes deregulated by somatic genomic rearrangement. Nat Commun. 2021;12(1):937. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-021-21081-y\u003c/span\u003e\u003cspan address=\"10.1038/s41467-021-21081-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoosen M, Od\u0026eacute; Z, Bunt J, Kool M (2022) The oncogenic fusion landscape in pediatric CNS neoplasms. Acta Neuropathol. 2022;143(4):427\u0026ndash;451\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStruijk RB, Dorssers LCJ, Henneman P et al (2020) Comparing genome-scale DNA methylation and CNV marks between adult human cultured ITGA6\u0026thinsp;+\u0026thinsp;testicular cells and seminomas to assess in vitro genomic stability. PLoS ONE 15(3):e0230253\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllahyar A, Pieterse M, Swennenhuis J et al (2021) Robust detection of translocations in lymphoma FFPE samples using targeted locus capture-based sequencing. Nat Commun. 2021;12(1):3361\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChrzanowska NM, Kowalewski J, Lewandowska MA (2020) Use of Fluorescence In Situ Hybridization (FISH) in Diagnosis and Tailored Therapies in Solid Tumors. Molecules. 2020;25(8):1864\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHorbinski C, Miller CR, Perry A (2011) Gone FISHing: clinical lessons learned in brain tumor molecular diagnostics over the last decade. Brain Pathol. 2011;21(1):57\u0026ndash;73\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoduru P, Chen W, Fuda F et al (2024) RNASeq Analysis for Accurate Identification of Fusion Partners in Tumor Specific Translocations Detected by Standard FISH Probes in Hematologic Malignancies. Clin Pathol. 2024;17:2632010X241230262\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllahyar A, Pieterse M, Swennenhuis J et al (2021) Robust detection of translocations in lymphoma FFPE samples using targeted locus capture-based sequencing. Nat Commun. 2021;12(1):3361\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatsunaga H, Arikawa K, Yamazaki M et al (2022) Reproducible and sensitive micro-tissue RNA sequencing from formalin-fixed paraffin-embedded tissues for spatial gene expression analysis. Sci Rep. 2022;12(1):19511\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWefers AK, Stichel D, Schrimpf D (2020) Isomorphic diffuse glioma is a morphologically and molecularly distinct tumour entity with recurrent gene fusions of MYBL1 or MYB and a benign disease course. Acta Neuropathol 139:193\u0026ndash;209\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEllison DW, Hawkins C, Jones DTW et al (2019) cIMPACT-NOW update 4: diffuse gliomas characterized by MYB, MYBL1, or FGFR1 alterations or BRAFV600E mutation. Acta Neuropathol 137:683\u0026ndash;687\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRyall S, Tabori U, Hawkins C (2020) Pediatric Low-Grade Glioma in the Era of Molecular Diagnostics. Acta Neuropathol Commun 8(1):30\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFabbri VP, Caporalini C, Asioli S, Buccoliero A (2022) Paediatric-type diffuse low-grade gliomas: a clinically and biologically distinct group of tumours with a favourable outcome. Pathologica 114(6):410\u0026ndash;421\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCicir\u0026ograve; Y, Sala A (2021) MYB oncoproteins: emerging players and potential therapeutic targets in human cancer. Oncogenesis 10:19\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark YW, Vollmuth P, Foltyn-Dumitru M (2023) al al The 2021 WHO Classification for Gliomas and Implications on Imaging Diagnosis: Part 2-Summary of Imaging Findings on Pediatric-Type Diffuse High-Grade Gliomas, Pediatric-Type Diffuse Low-Grade Gliomas, and Circumscribed Astrocytic Gliomas. J Magn Reson Imaging 2023 Apr 17\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePajtler KW, Witt H, Sill M et al (2015) Molecular Classification of Ependymal Tumors across All CNS Compartments, Histopathological Grades, and Age Groups. Cancer Cell 2015;27:728\u0026ndash;743\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMerchant TE, Bendel AE, Sabin ND et al (2019) Conformal radiation therapy for pediatric ependymoma, chemotherapy for incompletely resected ependymoma, and observation for completely resected, supratentorial ependymoma. J Clin Oncol 37:974\u0026ndash;983\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePages M, Pajtler KW, Puget S et al (2019) Diagnostics of pediatric supratentorial RELA ependymomas: integration of information from histopathology, genetics, DNA methylation and imaging. Brain Pathol 29:325\u0026ndash;335\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArabzade A, Zhao Y, Varadharajan S et al (2021) ZFTA-RELA Dictates Oncogenic Transcriptional Programs to Drive Aggressive Supratentorial Ependymoma. Cancer Discov. 2021, 11, 2200\u0026ndash;2215\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGessi M, Giagnacovo M, Modena P et al (2019) Role of Immunohistochemistry in the Identification of Supratentorial C11ORF95-RELA Fused Ependymoma in Routine Neuropathology. Am J Surg Pathol. 2019;43(1):56\u0026ndash;63\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAndreiuolo F, Varlet P, Tauzi\u0026egrave;de-Espariat A et al (2019) Childhood Supratentorial Ependymomas with YAP1-MAMLD1 Fusion: An Entity with Characteristic Clinical, Radiological, Cytogenetic and Histopathological Features: YAP1-MAMLD1 Ependymoma. Brain Pathol. 2019, 29, 205\u0026ndash;216\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eParker M, Mohankumar KM, Punchihewa C et al (2014) C11orf95-RELA fusions drive oncogenic NF-κB signalling in ependymoma. Nature 2014;506:451\u0026ndash;455\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBurford A, Mackay A, Popov S et al (2018) The ten-year evolutionary trajectory of a highly recurrent paediatric high grade neuroepithelial tumour with MN1:BEND2 fusion. Sci Rep 8:1032\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSughrue ME, Choi J, Rutkowski MJ et al (2011) Clinical features and post-surgical outcome of patients with astroblastoma. J Clin Neurosci. 2011;18(6):750-4\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLehman NL, Usubalieva A, Lin T et al (2019) Genomic analysis demonstrates that histologically defined astroblastomas are molecularly heterogeneous and that tumors with MN1 rearrangement exhibit the most favorable prognosis. Acta Neuropathol Commun 7(1):42\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLehman NL, Spassky N, Sak M et al (2023) Astroblastomas exhibit radial glia stem cell lineages and differential expression of imprinted and X-inactivation escape genes. Nat Commun. 2023;14(1):507\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGopakumar S, McDonald MF, Sharma H et al (2022) Recurrent HGNET-MN1 altered (astroblastoma MN1-altered) of the foramen magnum: Case report and molecular classification. Surg Neurol Int. 2022;13:139\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSturm D, Orr BA, Toprak UH et al (2016) New brain tumor entities emerge from molecular classification of CNS-PNETs. Cell 164:1060\u0026ndash;1072\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWood MD, Tihan T, Perry AJ et al (2018) Multimodal molecular analysis of astroblastoma enables reclassification of most cases into more specific molecular entities. Brain Pathol 28:192\u0026ndash;202\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrat DJ, Aldape K, Idbaih A et al (2021) Astroblastoma, MN1-altered. In: WHO Classification of Tumours Editorial Board, ed. WHO Classification of the Central Nervous System Tumours. 5th ed. IARC; 2021\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKnight J, De Jesus O (2021) Pilocytic Astrocytoma. In StatPearls Publishings. 2021; 30:14\u0026ndash;24\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBornhorst M, Frappaz D, Packer RJ (2016) Pilocytic astrocytomas. Handb. Clin. Neurol. 2016; 134:329\u0026ndash;44\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJones DT, Gronych J, Lichter P et al (2012) MAPK pathway activation in pilocytic astrocytoma. Cell Molec Life Sci 2012;69:1799Y811\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRyall S, Zapotocky M, Fukuoka K et al (2020) Integrated molecular and clinical analysis of 1,000 pediatric low-grade gliomas. Cancer Cell 37:569\u0026ndash;583e5\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAppay R, Pages M, Colin C et al (2020) Diffuse leptomeningeal glioneuronal tumor: a double misnomer? A report of two cases. Acta Neuropathol Commun 2020;8, 95\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGardiman MP, Fassan M, Orvieto E et al (2010) Diffuse leptomeningeal glioneuronal tumors: a new entity? Brain Pathol 2010;20:361\u0026ndash;66\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng MY, Sill M, Chiang J et al (2018) Molecularly defined diffuse leptomeningeal glioneuronal tumor (DLGNT) comprises two subgroups with distinct clinical and genetic features. Acta Neuropathol. 2018;136(2):239\u0026ndash;253\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZacharoulis S, Morales La Madrid A et al (2016) Central Versus Extraventricular Neurocytoma in Children: A Clinicopathologic Comparison and Review of the Literature. J Pediatr Hematol Oncol. Aug 2016;38(6):479\u0026thinsp;\u0026ndash;\u0026thinsp;85\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiangaspero F, Cenacchi G, Losi L et al (1997) Extraventricular neoplasms with neurocytoma features. A clinicopathological study of 11 cases. Am J Surg Pathol. Feb 1997;21(2):206\u0026thinsp;\u0026ndash;\u0026thinsp;12\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXiong Z, Zhang J, Li Z et al (2015) A comparative study of intraventricular central neurocytomas and extraventricular neurocytomas. J Neurooncol. Feb 2015;121(3):521-9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu L, Ouyang Z, Wang J et al (2017) A clinicopathologic study of extraventricular neurocytoma. J Neurooncol. Mar 2017;132(1):75\u0026ndash;82\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSievers P, Stichel D, Schrimpf D et al (2018) FGFR1:TACC1 fusion is a frequent event in molecularly defined extraventricular neurocytoma. Acta Neuropathol. 2018;136(2):293\u0026ndash;302\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKalawi AZ, Malicki DM, Abdullaev Z et al (2022) The role of methylation profiling in histologically diagnosed neurocytoma: a case series. J Neurooncol. 2022;159(3):725\u0026ndash;733\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJanice S, Ahn MD et al (2021) Papillary Glioneuronal Tumor With a Novel GPR37L1-PRKCA Fusion, Journal of Neuropathology \u0026amp; Experimental Neurology, Volume 80, Issue 10, October 2021, Pages 1004\u0026ndash;1006\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHou Y, Pinheiro J, Sahm F et al (2019) Papillary glioneuronal tumor (PGNT) exhibits a characteristic methylation profile and fusions involving PRKCA. Acta Neuropathol 137:837\u0026ndash;846\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePages M, Lacroix L, Tauziede-Espariat A et al (2015) Papillary glioneuronal tumors: Histological and molecular characteristics and diagnostic value of SLC44A1-PRKCA fusion. Acta Neuropathol Commun 2015;3:85\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNagaishi M, Nobusawa S, Matsumura N et al (2016) SLC44A1-PRKCA fusion in papillary and rosette-forming glioneuronal tumors. J Clin Neurosci 2016;23:73\u0026ndash;5\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoyal-Honavar A, Balasundaram A, Thayakaran IP et al (2022) ZFTA (Zinc Finger Translocation Associated) Fusion in Supratentorial Ependymomas: Low Prevalence in South Asians and No Correlation with Survival. World Neurosurg. 2022;164:e82-e90\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFudaba H, Momii Y, Kawasaki Y et al (2020) Well-differentiated Astroblastoma with Both Focal Anaplastic Features and a Meningioma 1 Gene Alteration. NMC Case Rep J. 2020;7(4):205\u0026ndash;210\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDandapath I, Das S, Charan BD et al (2024) Evaluation of KIAA1549::BRAF fusions and clinicopathological insights of pilocytic astrocytomas. Ann Diagn Pathol. 2024;72:152318\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTomić TT, Olausson J, Wilz\u0026eacute;n A et al (2017) A new GTF2I-BRAF fusion mediating MAPK pathway activation in pilocytic astrocytoma. PLoS One. 2017;12(4):e0175638\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRodriguez FJ, Schniederjan MJ, Nicolaides T et al (2015) High rate of concurrent BRAF-KIAA1549 gene fusion and 1p deletion in disseminated oligodendroglioma-like leptomeningeal neoplasms (DOLN). Acta Neuropathol. 2015;129(4):609\u0026ndash;610\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJones DT, Kocialkowski S, Liu L et al (2008) Tandem duplication producing a novel oncogenic BRAF fusion gene defines the majority of pilocytic astrocytomas. Cancer Res. 2008; 68(21):8673\u0026ndash;7\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAsuzu DT, Desai B, Maggio D et al (2021) FGFR1-TACC1 fusion associated with malignant transformation in a primary spinal cord glioma: a case report. J Spine Surg. 2021;7(3):434\u0026ndash;438\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBridge JA, Liu XQ, Sumegi J et al (2013) Identification of a novel, recurrent SLC44A1-PRKCA fusion in papillary glioneuronal tumor. Brain Pathol. 2013;23(2):121-8\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCapper D, Stichel D, Sahm F et al (2018) Practical implementation of DNA methylation and copy-number-based CNS tumor diagnostics: the Heidelberg experience. Acta Neuropathol 136:181\u0026ndash;210\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDobin A, Davis CA, Schlesinger F et al (2012) STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15\u0026ndash;21\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaas BJ, Dobin A, Li B et al (2019) Accuracy assessment of fusion transcript detection via read-mapping and de novo fusion transcript assembly-based methods. Genome Biol 20:213\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNicorici D, Satalan M, Edgren H et al (2014) FusionCatcher \u0026ndash; a tool for finding somatic fusion genes in paired-end RNA-sequencing data, bioRxiv, Nov. 2014\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSterky FH, Trotter JH, Lee SJ et al (2017) Carbonic anhydrase-related protein CA10 is an evolutionarily conserved pan-neurexin ligand. Proc Natl Acad Sci U S A. 2017;114(7):E1253-E1262\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchulte I, Batty EM, Pole JC et al (2012) Structural analysis of the genome of breast cancer cell line ZR-75-30 identifies twelve expressed fusion genes. BMC Genomics. 2012;13:719\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNovak RL, Harper DP, Caudell D et al (2012) Gene expression profiling and candidate gene resequencing identifies pathways and mutations important for malignant transformation caused by leukemogenic fusion genes. Exp Hematol. 2012;40(12):1016-27\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFouad FM, Eid JI (2023) PAX5 fusion genes in acute lymphoblastic leukemia: A literature review. Medicine (Baltimore). 2023;102(20):e33836\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eČrepinšek K, Klobučar N, Tesovnik T et al (2024) PAX5 Alterations in a Consecutive Childhood B-Cell Acute Lymphoblastic Leukemia Cohort Treated Using the ALL IC-BFM 2009 Protocol. Cancers (Basel). 2024;16(6):1164\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKanagal-Shamanna R, Bao H, Kearney H et al (2022) Molecular characterization of Novel ATM fusions in chronic lymphocytic leukemia and T-cell prolymphocytic leukemia. Leuk Lymphoma. 2022;63(4):865\u0026ndash;875\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Pediatric Brain Tumors, Structural Variants Detection, Single-Cell FISH, Multi-Omics Integration, Intratumoral Heterogeneity","lastPublishedDoi":"10.21203/rs.3.rs-7244481/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7244481/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eINTRODUCTION:\u003c/h2\u003e\u003cp\u003ePediatric brain tumors (PBTs) are a heterogeneous group of neoplasms. The WHO classification highlights the relevance of genetic and epigenetic alterations in defining molecular subtypes. This study investigated key rearrangements (\u003cem\u003eMYB\u003c/em\u003e, \u003cem\u003eMYBL1\u003c/em\u003e, \u003cem\u003eZFTA\u003c/em\u003e, \u003cem\u003eYAP1\u003c/em\u003e, \u003cem\u003eMN1\u003c/em\u003e) and fusions (\u003cem\u003eKIAA1549::BRAF, FGFR1::TACC1, PRKCA::SLC44A1\u003c/em\u003e), which have diagnostic, prognostic, and therapeutic implications.\u003c/p\u003e\u003ch2\u003eMATERIALS AND METHODS\u003c/h2\u003e\u003cp\u003eWe applied a novel single-cell fluorescence in situ hybridization (FISH) method in 50 PBTs, comparing results to whole-transcriptome sequencing (16/50) and DNA methylation profiling (23/50). Our FISH-based approach, using signal distance measurements and pattern analysis, aims to objectively distinguish normal from rearranged signals, improving accuracy in detecting structural variants and discriminating ambiguous cases.\u003c/p\u003e\u003ch2\u003eRESULTS\u003c/h2\u003e\u003cp\u003eFISH-based signal distance analysis identified rearrangements in \u003cem\u003eMYB\u003c/em\u003e (5+/10), \u003cem\u003eMYBL1\u003c/em\u003e (3+/10), \u003cem\u003eZFTA\u003c/em\u003e (7+/12), \u003cem\u003eYAP1\u003c/em\u003e (3+/12), \u003cem\u003eMN1\u003c/em\u003e (4+/8), including novel unbalanced patterns. Fusion detection by pattern analysis revealed \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e (8+/10), \u003cem\u003eFGFR1::TACC1\u003c/em\u003e (4+/6), and \u003cem\u003ePRKCA::SLC44A1\u003c/em\u003e (1+/4). High concordance was observed with RNA-seq and methylation profiling. RNA-seq identified novel fusions: \u003cem\u003eMYB::CA10\u003c/em\u003e, \u003cem\u003eYAP1::TYR\u003c/em\u003e, and \u003cem\u003ePAX5::ATM\u003c/em\u003e. Concurrent structural variants were observed in single tumors, including \u003cem\u003eKIAA1549::BRAF\u003c/em\u003e with \u003cem\u003eMN1::BEND2\u003c/em\u003e and \u003cem\u003eZFTA::RELA\u003c/em\u003e with \u003cem\u003eYAP1::TYR\u003c/em\u003e, revealing genetic heterogeneity of PBTs.\u003c/p\u003e\u003ch2\u003eCONCLUSIONS\u003c/h2\u003e\u003cp\u003eOur single-cell FISH approach provides standardized, reproducible criteria for structural variant detection, minimizing interpretive variability. When combined with transcriptomic and epigenetic data, this approach supports accurate subtype classification and reveals clinically relevant intratumoral heterogeneity in PBTs.\u003c/p\u003e","manuscriptTitle":"Integrated molecular approach including single-cell FISH enables accurate detection of genetic rearrangements, classification, and identification of novel variants in pediatric brain tumors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-05 12:24:26","doi":"10.21203/rs.3.rs-7244481/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e3a8debb-f75f-486d-befb-4352865d66a9","owner":[],"postedDate":"August 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-17T12:38:09+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-05 12:24:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7244481","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7244481","identity":"rs-7244481","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}