MYH9::LTK Fusion in a Pediatric Acral Soft Tissue Spindle Cell Neoplasm

We read with great interest the recent article by Yeung et al., which described two pediatric patients with acral low-grade spindle cell soft tissue tumors harboring novel LTK fusions [1]. Here, we report a case of a 2-year-old male who presented at 3 months of age with an infiltrative mass involving almost the entire musculature of the right leg. He underwent surgical resection at 7 months, followed by two re-excisions due to local recurrences. Histologically, the tumor was composed of a well-vascularized proliferation of relatively monomorphic spindle cells arranged in short fascicles and infiltrating the adipose tissue. Occasional perivascular hyalinization was noted (Figure 1A,B). Mitoses were seen, but there was no marked cytological atypia (Figure 1C,D). By immunohistochemistry, the neoplastic cells were variably positive for CD34 and S100 (Figure 1E,F). ALK-1 immunostain was negative (not shown). The morphologic features, along with the immunohistochemical staining profile, suggested a diagnosis of NTRK-rearranged spindle cell neoplasm. Fluorescent in situ hybridization (FISH) assay targeting ETV6 (surrogate test for ETV6::NTRK3 fusion) and NTRK1 genes (break-apart probes) was negative for rearrangements. Whole transcriptome sequencing identified an in-frame fusion between MYH9 (NM_002473.5) exon 35 and LTK (NM_002344.6) exon 11, with a 3 amino acid intervening insertion (Figure 2A). The chimeric protein product retains an intact kinase domain of LTK. RNA expression of LTK was increased in comparison with more than 900 solid tumors (Figure 2B), supporting constitutive kinase activation of LTK by the fusion event. On unsupervised t-distributed stochastic neighbor embedding (t-SNE) clustering analysis of whole transcriptome expression data performed with the St. Jude Cloud RNA-Seq Expression Classification analysis workflow [2], the patient's tumor clustered nearby dermatofibroma protuberans (DFSP) samples, but also close to NTRK1-fused soft tissue tumors (Figure 2C). Targeted genomic sequencing using a hybrid capture-based next-generation sequencing (NGS) assay that interrogates molecular alterations (single nucleotide variants and deletion–insertion variants) in DNA of 362 genes was negative for pathogenic variants. There has been no tumor recurrence since the third surgery in March 2023. The morphological features, immunoprofile (CD34+, S100+), and outcome in our case are similar to those reported by Yeung et al. [1], although the age of presentation in our patient was much younger than in the other two reported patients. The transcriptional similarities noted by RNA expression clustering analysis between this tumor and DFSP recapitulate the DNA methylation clustering profile seen in the previously described MYH9::LTK-fused soft tissue tumor [1]. The LTK gene shares a high homology with the ALK gene. Interestingly, CD34-positive spindle cell neoplasms morphologically similar to DFSP with alterations in the ALK gene have been described [3]. Downstream signaling of both ALK and PDGF through the Erk MAP-kinase pathway may at least in part explain the findings from the clustering analysis, although studies of larger cohorts are needed for further elucidation of the potential biological connection of these tumors. In summary, we present an additional case of MYH9::LTK-fused low-grade spindle cell neoplasm occurring at an acral site of a pediatric patient. LTK-fused tumors share morphological and phenotypical features with an emerging group of soft tissue neoplasm molecularly characterized by having activating kinase fusions (e.g., NTRK1, NTRK3, RET, RAF1, BRAF, ALK, and ROS1). Recognizing the role of LTK in soft tissue tumorigenesis is important from a diagnostic standpoint, as this gene may not be included in targeted sequencing panels. NGS assays for fusion detection are not necessarily equivalent. Therefore, a negative NGS panel test result in a context suspicious for a kinase-fused tumor should not abbreviate the diagnostic workup or be considered a definitive answer. Comprehensive RNA sequencing analysis may be a more effective strategy in this setting. Larissa V. Furtado and Teresa Santiago wrote and edited the manuscript. Roxana Martinez, Soad Fuentes-Alabi, and Alberto Pappo provided clinical data. Teresa Santiago and Ana C. Polanco provided histopathology data. Larissa V. Furtado and Zonggao Shi analyzed the molecular data. All authors reviewed and approved the manuscript. The authors declare no conflicts of interest. The authors have nothing to report.