NTRK1-3 fusions in sarcomas: prevalence, significance, and clinical implications – a systematic review Authors: Shahd Hamran, Zaineh Alnoubani, Aly Mostafa Hassan, Amani A Al-Rajhi, Humam Emad Rajha, Munirah Altaissan, Giridhara R Babu, Zoran Gatalica, Semir Vranić Published: 2025 ABSTRACT ABSTRACTAimGene fusions involving Neurotrophic Receptor Tyrosine Kinase (NTRK) genes lead to Tropomyosin Receptor Kinases (TRK) overexpression. Detecting NTRK1-3 fusions through advanced molecular techniques has revolutionized cancer care through personalized medicine, such as TRK inhibitors.MethodsWe conducted a comprehensive search of databases (PubMed/MEDLINE, SCOPUS, Web of Science) and extracted data on study characteristics, molecular characteristics, and clinical outcomes. Data synthesis involved narrative and thematic analysis and study quality assessment using the Methodological Standard for Epidemiological Research (MASTER) scale.ResultsWe included 136 studies with 18,077 patients. The most common tumor categories were unclassified soft tissue sarcomas (11.09%), gynecological sarcomas (5.64%), and liposarcomas (3.37%). Most tumors were gynecologic (5.64%), followed by the limbs (1.45%). Genomic sequencing was the predominant diagnostic method used in 110 studies. Overall, 551 patients with sarcoma tested positive for NTRK1-3 gene fusions, primarily involving NTRK1 and NTRK3 (~93%), with ETV6-NTRK3 fusion being the most frequently reported fusion. Larotrectinib was used in 142 patients, demonstrating an 83.80% response rate, with low mortality (2.82%) and recurrence (2.11%) rates. Entrectinib had a lower response rate of 63.64%.ConclusionsWe confirm the rarity of NTRK1-3 fusions in sarcomas. TRK inhibitors show high efficacy in sarcomas, emphasizing the necessity of genomic testing in all cases.Protocol registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024563594.KEYWORDS: Soft tissue tumors, Sarcomas, NTRK1-3 fusions, TRK inhibitors, systematic review Introduction The Tropomyosin Receptor Kinase (TRK) family of receptor tyrosine kinases comprises three types: TRKA, TRKB, and TRKC transmembrane proteins, which are encoded by Neurotrophic Receptor Tyrosine Kinase (NTRK) genes 1, 2, and 3, respectively [1]. These TRKs, found in human neuronal tissue, are vital for cell survival and the development of the nervous system. TRKs are activated by neurotrophins. Neurotrophins, such as nerve growth factor for TRKA, brain-derived growth factor and neurotrophin-4/5 for TRKB, and neurotrophin-3 for TRKC, serve as specific ligands for these receptors. Each of the three TRKs has an extracellular domain for ligand binding, a transmembrane region, and an intracellular domain containing a kinase domain [2]. NTRK1-3 gene fusions with their partner genes are the key oncogenic mechanism, leading to the overexpression of TRK proteins [3]. NTRK1-3 fusions are present in nearly 90% of rare cancers, such as infantile fibrosarcoma and secretory breast carcinoma, but occur in less than 1% of common cancers [4]. These fusions are also rare in some other common cancers, such as uterine soft tissue tumors, some variants of malignant melanoma, papillary thyroid carcinoma, and pancreatic adenocarcinoma. However, evidence on NTRK1-3 fusion prevalence in other cancer types remains limited [5]. NTRK1-3 gene fusions or their proteins are often detected using immunohistochemistry (IHC), fluorescence in situ hybridization (FISH), reverse transcription-polymerase chain reaction (RT-PCR), and next-generation sequencing (NGS) [6]. Agnostic therapy refers to treatment decisions based solely on the tumor’s genomic features [7], enabling personalized and more effective treatment options. This approach has transformed cancer treatment [7]. Precision medicine relies on detecting genomic alterations to guide treatment, using drugs that target tumor-driving mutations rather than tumor anatomic location or type [8,9]. One of the most significant advancements in this field is the development of TRK inhibitors, designed to block the oncogenic TRK signaling [10]. First-generation TRK inhibitors, larotrectinib and entrectinib, target TRKA, TRKB, and TRKC. Larotrectinib is highly selective for TRKs, whereas entrectinib also inhibits ROS1 and ALK [10,11]. These approved first-generation TRK inhibitors have demonstrated significant antitumor activity in clinical trials for treating solid tumors with NTRK1-3 fusions [12,13]. Data from a pooled analysis of three clinical trials showed that patients receiving larotrectinib have a response rate between 81% and 100% [14]. In comparison, data from the analysis of three other trials showed that entrectinib has an objective response rate (ORR) of 57% and durable responses, including in patients with brain metastases [15]. Other inhibitors, such as repotrectinib (TPX-0005), crizotinib, and cabozantinib, are under investigation and show promise in overcoming resistance due to acquired mutations [11]. NTRK1-3 fusion detection is essential for TRK inhibitor eligibility, facilitating personalized and more effective targeted therapies. Molecular testing is integral to precision medicine’s diagnostic and therapeutic process, significantly improving patient outcomes, contributing to the broader understanding of cancer biology, and developing novel therapeutic approaches. This study aims to systematize the knowledge regarding the frequency and types of NTRK1-3 gene fusions across a diverse range of soft tissue sarcomas beyond commonly associated sarcomas and infantile fibrosarcoma in pediatrics, underscoring the importance of comprehensive genomic profiling to guide personalized treatment strategies for patients with advanced or metastatic disease. Methods Study design This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Supplementary File S1). The protocol for this systematic review was registered on the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42024563594). Data sources and search methods We conducted a comprehensive database search (PubMed/MEDLINE, SCOPUS, Web of Science) using NTRK1-3 and sarcoma-related terms, spanning from their inception to May 2024. Additionally, relevant proceedings from the major oncologic scientific conferences, including the American Society of Clinical Oncology (ASCO), the European Society for Medical Oncology (ESMO), and the American Association for Cancer Research (AACR) from January 2023 to May 2024 were reviewed. The search was updated on 14 July 2024. Specific search terms related to NTRK1-3 and sarcoma were employed to identify pertinent studies. These terms encompass various synonyms and related expressions for NTRK receptors and a broad spectrum of sarcoma types. The search terms used for different databases are provided in Supplementary File S2. Procedure for study selection The articles retrieved from the database searches were uploaded into EndNote software to remove duplicates. Following deduplication, titles and abstracts were screened using the Rayyan systematic review management platform (https://www.rayyan.ai/). Full-text articles meeting initial inclusion criteria underwent a detailed eligibility assessment by each of the two independent reviewers. Any discrepancies were resolved through consensus or consultation with a third reviewer (SH). Eligibility Eligible studies included case series, case reports, and observational analytical studies investigating NTRK1-3 gene fusions in soft tissue sarcomas. Relevant abstracts without full text were also considered if they reported applicable outcomes. Studies that included a diagnosis of soft tissue sarcoma, reported the frequencies and types of NTRK1-3 fusions, and provided detailed descriptions of NTRK1-3 fusion variants were included. Studies exclusively focusing on non-sarcoma malignancies or non-English publications were excluded. Key definitions Sarcomas are malignant tumors of mesenchymal origin [16]. Neurotrophic Receptor Tyrosine Kinase (NTRK) fusion genes are formed by the fusion of a part of the NTRK1-3 genes with different partner genes, resulting in constitutive activation of the TRK signaling [17,18]. Outcomes The primary outcome was to determine the prevalence of NTRK1-3 gene fusions in various types of sarcomas, assessed by the proportion of cases with identified fusions using molecular diagnostic techniques. Secondary outcomes included comparing fusion type distributions, evaluating detection method efficacy, assessing the safety and efficacy of TRK inhibitors, and examining fusion distribution across different sarcoma subtypes. Data extraction Two independent authors extracted data using Microsoft Excel. The extracted data included study design, patient demographics, tumor type and location, molecular details of NTRK1-3 fusions, detection methods, treatment approaches, and survival outcomes. To avoid duplication of cases reported across multiple publications, a third reviewer (ZA) independently cross-checked all included studies. This reviewer carefully screened patient characteristics and origin to identify and exclude any overlapping cases. Unique cases were retained for analysis. Whenever possible, the histologic subtype of sarcomas was reported. Data synthesis The results were synthesized using narrative synthesis methods. Tabular presentations were summarized, including the proportion of fusion, variants’ distribution, and clinicopathological correlations. Thematic analysis will explore molecular mechanisms and therapeutic implications. Subgroup analyses were done to assess variations across different sarcoma subtypes. The quality of the studies was assessed using the MethodologicAl STandard for Epidemiological Research (MASTER) scale, which includes 36 safeguards categorized into seven domains [19]. These domains are equal recruitment, equal retention, equal ascertainment, equal implementation, equal prognosis, sufficient analysis, and temporal precedence. We used Stata version 18 (College Station, TX, USA) for the analysis. Ethics This is a systematic review of published literature and thus did not require institutional ethical approval. Results Search results Seven hundred eighty-seven studies were identified from the databases. After removing 186 duplicates, 387 study records were excluded through manual abstract and title screening, and the remaining 214 studies were screened using full text. After full-text screening, 19 studies were excluded because they were reviews, 14 because they did not assess sarcomas, and 10 papers were not available in English. Other reasons for exclusion are illustrated in the PRISMA flowchart (Figure 1). The final review included 136 studies [4,14,15,20–152]. Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flowchart for the search results. Quality assessment of included studies Seven of the 136 articles included were abstracts [77,83,90,111,125,134,141], and one paper was an integrated analysis of unpublished studies [15], so their quality could not be assessed. The average MASTER scale score for the remaining 128 studies we assessed was 25 out of a maximum score of 36, ranging from 20 in some studies [23,36,98,118,119,137] up to 29 in others [42,47,50,70,71,73,95,97,121,124,146], indicating a moderate quality with a low-moderate risk of bias. The 128 studies did not have blinding or allocation concealment, which downgraded the studies in the “Equal ascertainment” (domain 3) and the “Equal prognosis” (domain 5) categories of the MASTER scale. Additionally, there were 40 multi-centered studies [4,14,20–24,30, 34, 40, 44, 51, 53, 57, 59, 61, 62,78–80,85,89,94,98,107,108,110,114,115,118,119,129,131–133,136–138,147,148], so the fourth domain, “Equal implementation,” was not fulfilled. Further details of the MASTER scale assessment can be seen in Supplementary File S3. Characteristics of included studies The baseline characteristics in the analyzed studies exhibit considerable heterogeneity in the study design, duration, location, patient demographics, and gender distribution. Fifty out of 136 papers were case reports [25, 28, 31, 32, 38, 39, 42, 47, 49, 53, 55, 58, 60,65–68,70,71,73,76,81–84,86,88,91–93,95–97,99–101,103,106,109,112,116,120,121,126,128,130,140,145,146,150], 26 were case series [20, 27, 35, 43, 45, 46, 52, 56, 61, 69, 72, 74, 75, 80, 101, 110, 113, 115, 117, 118, 123, 125,132–135,138,152], 54 were observational retrospective studies [4,14,21–24,26, 29, 30, 33, 34, 36, 37, 40, 41, 44, 48, 51, 54, 57, 59,62–64,77–79,85, 87, 89, 90, 94, 98, 102, 104, 108, 111, 114, 122, 124, 129, 131, 133, 136, 137, 139,141–144,147–149,151], one was a prospective cohort [134], two were retrospective and prospective cohort studies [107,119], two were pooled analyses of three clinical trials [15,127], and one was a study with a partitioned survival model [105] (Supplementary File S3). Notably, 85 studies did not specify the exact length of the study or follow-up period. Nineteen out of 51 studies were older more than ten years. Twelve out of 19 studies [22,30,34,40,54,59,79,89,102,139,144,151] were descriptive utilizing archived materials and clinical records to analyze tumor biopsies, while the remaining seven [20,45,69,97,117,123,135] were case series and case reports with long-term patient follow-ups. Additionally, 25 studies [4, 14, 37, 47, 52, 55, 56, 62, 77, 80, 81, 86, 90, 98, 107, 108, 113, 114, 118, 119, 129, 136, 141, 143, 149] had durations ranging from one to five years, including a mixture of case series, case reports, and retrospective studies. Most of the included studies were conducted in the USA (n = 46), followed by China (n = 17), and Japan (n = 14). Eight studies did not specify the geographic location. The remaining studies were distributed across the globe (Figure 2 and Supplementary File S3). Geographical location of included studies. Forty-four studies did not report gender distribution. The remaining 92 studies [21,22,24,25,28–31,35,36,38,40,42,46,47, 49–61,63–76,78,80–84,86–89,91–93,95–97,100–105,109,110,112,114,116–123,126–128,130–132,135,136,138–140,143,145,147,148,150] had a combined sample size of 1,824 patients; 1,139 of them were women (62.45%). Regarding the age of patients, nine studies did not specify the age group, 43 studies [20–23,25,28–34,38,39,42,44–46,48,49,54,55,59,60,67–70,75, 76, 82, 92, 102, 106, 112, 117, 120, 130, 139, 142, 143, 150, 152] focused exclusively on pediatric patients (those under 18 years), 38 studies [24,26,36,37,40,43,51,52,62–64,77–80,89, 90, 98, 104, 105, 107, 113, 114, 119,122–124,129,131–133,136–138,144, 147, 149, 151] included both pediatric and adult populations, and 46 studies [4,15,35,47,50,53,56–58,61,65,66,71–74,81,83,84,86–88,91,93–97,99–101,103,108–110,116,118,121,126–128,140, 141, 145, 146, 148] focused solely on adults. The population’s age range varied from neonates only a few days old to elderly patients (Supplementary File S3). Sarcoma types This systematic review revealed 18,077 sarcoma patients in 136 studies. Of these, 17 studies [4, 14, 62, 63, 77, 90, 94, 98, 110, 111, 114, 133, 134, 139, 141, 142, 144] (representing 11,108 patients) did not mention the specific type of soft tissue tumor detected. Among the remaining 6,969 cases, 2,004 were diagnosed as soft tissue sarcoma, not otherwise specified. Additionally, 1,019 women presented with gynecologic sarcomas, including 591 with leiomyosarcomas. Other soft tissue tumors included 409 gastrointestinal stromal tumors (GIST), 145 synovial sarcomas, 302 osteosarcomas, 162 angiosarcomas, 37 lipofibromatosis-like neural tumors, 609 liposarcomas, and 151 patients with congenital or infantile fibrosarcomas. Further details on other sarcoma types are provided in Supplementary Table S1. Twenty-two studies [39, 51, 53, 57, 65, 70, 73, 76, 84, 88, 96, 97, 100, 101, 103, 107, 109, 112, 121, 140, 145, 147] reported only spindle cell sarcomas (n = 72 patients), while 13 studies [20,25,31,32,38,42,55,67,75,106,117,120,130] reported only congenital or infantile fibrosarcomas (n = 23 patients), and three studies [35,37,58] focused on patients with GIST (n = 197 patients). Further details on studies reporting specific sarcomas are described in Supplementary Table S2. Tumor location Thirty-nine out of 136 studies did not specify tumor location. A total of 1,019 sarcomas were gynecologic tumors (uterus, cervix, or vagina), 262 tumors were in the limbs, and 80 tumors were within the head and neck region. Other anatomic locations of sarcomas are described in Supplementary Table S3. Supplementary Table S4 also summarizes the number of studies and patients for each sarcoma location. Genomic alteration detection methods Two studies [96,105], with a total sample size of 251 patients, did not mention the method used to detect NTRK1-3 alterations. Thirty-eight studies [14, 15, 37, 44, 53, 55, 58, 60, 62, 65, 66, 75, 77, 80, 81, 83, 86, 88, 89, 94, 101, 104, 106, 111, 113, 116, 118, 119, 121, 123, 126, 128, 134, 138,140–143,149], encompassing a total of 12,404 patients out of 18,077 (68.62%), used only genomic sequencing techniques for NTRK1-3 fusion detection. These methods included RNA-based sequencing, DNA-based sequencing, whole-genome sequencing (WGS), whole-exome sequencing (WES), Sanger sequencing, comprehensive genomic profiling (CGP), NGS, or combinations of these approaches. FISH alone was used only in two studies [38,52], adding up to 11 patients (0.06%), and IHC alone was also used in two studies [54,85], involving a total of 400 patients (2.21%). Seven studies [23,24,27,29,30,32,67], with 73 patients (0.40%), used RT-PCR alone. Combined diagnostic methods were also commonly used; for instance, genomic sequencing with IHC was used in 18 studies [4,41,68,72,79,92,93,97,98,103,114,133,137,144,146,148,150,152] with a total of 2,263 patients (12.52%). Genomic sequencing with RT-PCR was used in eight studies [26, 47, 49, 64, 87, 100, 125, 129] (401 patients, 2.22%); genomic sequencing with FISH in 11 studies [43,51,63,69,71,76,109,115,127,130,132] (431 patients, 2.38%), FISH and IHC in four studies [40,50,70,110] (18 patients, 0.10%), and FISH with RT-PCR in two studies [35,117] (13 patients, 0.07%). Some studies used multi-modal diagnostic tests; for example, genomic sequencing combined with FISH and IHC was utilized in 17 studies [45, 56, 59, 73, 74, 78, 82, 90, 91, 99, 108, 112, 124, 131, 135, 145, 147] (1,329 patients, 7.35%); genomic sequencing combined with FISH and RT-PCR in seven studies [39,46,61,95,120,139,151] (113 patients, 0.63%); genomic sequencing combined with IHC and RT-PCR in four studies [42,84,102,136] (189 patients, 1.05%). Moreover, five studies [15,48,57,107,122] utilized all four detection methods, genomic sequencing, IHC, RT-PCR, and FISH; giving a total of 124 patients (0.69%). “Other methods,” including varied diagnostic techniques such as cytogenetic analysis, karyotyping, Northern blotting, Western blotting, and Southern blotting, were employed in nine studies [20–22,25,28,31,33,34,36] involving 57 patients (0.32%). These methods illustrate the diverse diagnostic approaches used across studies, with genomic sequencing as an essential tool for detecting NTRK1-3 genomic alterations, utilized in 110 out of 136 studies with 17,269 patients, either alone or in combination with other techniques (Figure 3). The type of diagnostic methods used for diagnosing NTRK1-3 fusions in sarcomas. Fluorescence in Situ Hybridization (FISH), Immunohistochemistry (IHC), Not Available (NA), Genomic Sequencing (GS), Reverse Transcriptase Polymerase Chain Reaction (RT-PCR), Neurotrophic Receptor Tyrosine Kinase (NTRK). Prevalence of NTRK1-3 alterations in sarcomas Out of 18,077 patients with sarcoma, 821 patients had a genomic alteration in one of the NTRK1-3 genes (4.54%). Forty-eight of the NTRK-positive patients had alterations detected solely through IHC, FISH, or both, without subsequent genomic confirmation; the remaining cases were confirmed using molecular techniques, including RT-PCR and genomic sequencing. NTRK1-3 fusions were the most common (551/821 patients; 67.11%), followed by a positive TRK protein on IHC staining (57/821 patients; 6.94%), NTRK gene rearrangement (10/821 patients; 1.22%), NTRK gene amplification (6/821 patients; 0.73%), or NTRK gene mutation (5/821 patients; 0.61%). Moreover, seven studies [90,94,111,139,141,142,151] with 165/821 patients (20.10%) did not specify the NTRK1-3 genomic alterations. Further details on NTRK1-3 alterations can be found in Supplementary Table S3. NTRK1-3 fusion partners The NTRK1-3 gene fusions mainly were either in the NTRK1 gene (251/551 patients; 45.56%) or the NTRK3 gene (259/551 patients; 47.01%), with only 15 out of 551 fusions affecting the NTRK2 gene (2.72%). The NTRK1 gene was mostly found to be fused with either the TPM3 (91/251 NTRK1 gene fusions; 36.25%) or the LMNA genes (89/251 NTRK1 gene fusions; 35.46%). The NTRK3 gene, on the other hand, was mostly found to be fused with the ETV6 gene (174/259 NTRK3 gene fusions; 67.18%). Further details on NTRK1-3 gene fusion partners can be found in Supplementary File S3. Twenty-nine studies [20–26,28,29,31–35,37, 38, 42, 50, 67, 95, 100, 102, 106, 116, 117, 120, 130, 134, 137], involving 83 patients (0.46% of patients with sarcoma), reported ETV6-NTRK3 gene fusion only. Twelve studies [39, 49, 53, 55, 58, 60, 71, 73, 92, 103, 118, 150], with 12 patients, reported LMNA-NTRK1 gene fusion only (0.07% of patients with sarcoma). Moreover, two studies [48,59], with 51 patients, reported TPM3-NTRK1, LMNA-NTRK1, MIR548F1-NTRK1, SQSTM1-NTRK1, TPR-NTRK1, STRN-NTRK2, ETV6-NTRK3, and EML4-NTRK3 gene fusions (0.28% of patients with sarcoma). Further details are shown in Supplementary Table S5. Treatment outcomes with TRK inhibitors The approved indications for TRK inhibitors in the clinical setting include advanced, unresectable, or metastatic NTRK fusion – positive tumors [153]. Most studies investigated the efficacy and safety of larotrectinib alone; 24 studies explored larotrectinib monotherapy with a cumulative follow-up of 622 months [37, 38, 59, 67, 70, 72, 75, 78, 80, 88, 97, 101, 117, 121, 127, 128, 130, 133, 136, 139, 142, 143, 147, 151]. A total of 1,106 patients with sarcoma were included in those studies, of which 142 had NTRK1-3 gene fusions and received larotrectinib therapy. In patients treated with larotrectinib, the response rate to therapy was 83.80% (119/142), with 2.82% (4/142) deaths, 8.45% (12/142) metastases, and 2.11% (3/142) recurrences after treatment. Entrectinib was explored in six studies [15,79,82,99,140,150] where 33 patients were treated with a total follow-up period of 114 months. Entrectinib showed a response rate of 63.64% (21/33). Moreover, five studies [74,86,104,119,126] explored the effects of combined larotrectinib and entrectinib therapies, reporting a response rate of 41.67% (10/24), metastasis rate of 41.67% (10/24), recurrence rate of 25% (6/24), and a death rate of 20.83% (5/24). One study [138] included 22 patients taking triple therapy of larotrectinib, entrectinib, and selitrectinib, who were followed up on for an average of 70 months. The response rate was 31.82% (7/22), with 9.09% recurrence (2/22) and a mere 4.55% (1/22) reported metastasis and death [138]. Limited data are available for combining larotrectinib with selitrectinib or repotrectinib, with three studies [68,106,148] and one study [120] on these combinations, respectively. In these smaller cohorts, outcomes suggest moderate responses with some recurrences and metastases. All above mentioned cases had confirmed 441 gene fusions. Further details are depicted in Table 1. Table 1.The therapeutic outcomes from studies investigating the efficacy of various TRK inhibitors, including larotrectinib, entrectinib, and combinations with other inhibitors, such as selitrectinib and repotrectinib, in the context of NTRK-positive sarcomas.TRK inhibitorNo. of studiesNo. of patientsFollow-up (months)OutcomeTotalOn TRK inhibitorsResponse to therapyDeathMetastasisRecurrenceLarotrectinib241106142622119/1424/14212/1423/14283.80%2.82%8.45%2.11%Entrectinib65213311421/330/7*0/7*1/7*63.64%0.0%0.0%14.29%Larotrectinib, Entrectinib4942415910/245/2410/246/2441.67%20.83%41.67%25.0%Larotrectinib, Entrectinib, Selitrectinib12222707/221/221/222/2231.82%4.55%4.55%9.09%Larotrectinib, Repotrectinib111NA0/10/11/11/10.0%0.0%100.0%100.0%Larotrectinib, Selitrectinib355253/51/53/53/560.0%20.0%60.0%60.0%*Although 33 patients received entrectinib, detailed data on death, metastasis, and recurrence were available for only seven patients in our review.NA: not available; TRK: Tropomyosin Receptor Kinases; NTRK: Neurotrophic Receptor Tyrosine Kinase. The therapeutic outcomes from studies investigating the efficacy of various TRK inhibitors, including larotrectinib, entrectinib, and combinations with other inhibitors, such as selitrectinib and repotrectinib, in the context of NTRK-positive sarcomas. *Although 33 patients received entrectinib, detailed data on death, metastasis, and recurrence were available for only seven patients in our review. NA: not available; TRK: Tropomyosin Receptor Kinases; NTRK: Neurotrophic Receptor Tyrosine Kinase. Discussion In this systematic review, we synthesized and analyzed available evidence on NTRK1-3 fusions in soft tissue tumors/sarcomas. Our findings indicate that NTRK1-3 fusions are present in a small subset of sarcomas, with most fusions being NTRK1 or NTRK3. Tumors harboring NTRK1-3 fusions often exhibit distinct clinical and pathological characteristics, such as younger age, and distinctive morphologic features, such as densely packed spindle cells and myxoid stroma [154]. These tumors commonly affect specific anatomical sites, like the extremities in infantile fibrosarcoma and the lungs or abdomen in inflammatory myofibroblastic tumors [154]. NTRK1-3 fusions involve gene rearrangements leading to constitutive activation of the TRK signaling pathway, promoting tumorigenesis [104,154]. The clinical behavior of these cancers may substantially vary, with some showing favorable responses to TRK inhibitors, highlighting the importance of routine comprehensive genomic testing for personalized treatment [10]. Recent mechanistic studies have clarified that NTRK1-3 fusions drive oncogenesis through fusion-partner – specific effects rather than uniform TRK activation. Classic work by Knezevich et al. first demonstrated that the ETV6-NTRK3 fusion activates TRK signaling via a coiled-coil – mediated dimerization domain [155]. Subsequent studies of LMNA-NTRK1 and TPM3-NTRK1 revealed that the structural length and rigidity of the 5’partner modulate dimerization strength, subcellular localization, and signaling output [73,154]. Functional and phospho-proteomic analyses show consistent activation of MAPK/ERK, PI3K/AKT, and PLCγ cascades, but with distinct amplitudes; ETV6-NTRK3 strongly drives ERK phosphorylation, whereas TPM3-NTRK1 favors AKT activation [156,157]. Transcriptomic profiling further indicates fusion-specific oncogenic programs, including cell-cycle progression, extracellular-matrix remodeling, and apoptosis resistance. Cryo-EM studies confirm that partner-derived dimerization interfaces induce unique kinase-domain conformations that affect inhibitor binding and resistance emergence [158]. Our systematic review provides an updated and comprehensive synthesis of the literature on NTRK1-3 fusions in sarcomas, addressing gaps left by previous reviews. It builds upon prior work by incorporating studies published in recent years, which reflect advancements in diagnostic techniques and the identification of additional NTRK fusion-positive cases. Unlike the review by Assi et al. [157] that was current to 2020, our review includes studies up to mid-2024. Moreover, several prior reviews focused on specific sarcoma subtypes or did not cover the entire spectrum of sarcomas, thereby limiting the generalizability of their conclusions [159,160]. Other reviews focused on one population demographic, like reporting NTRK1-3 fusions in the pediatric population only [102]. Our systematic review attempted to cover the entire spectrum of soft tissue sarcomas and incorporate the newest publications, providing a more complete and contemporary understanding of the prevalence and clinical implications of NTRK1-3 fusions. This systematic review confirms that the NTRK1-3 fusions are rare in soft tissue sarcomas. Our descriptive analysis shows that ETV6-NTRK3 fusion is the most prevalent, with documentation in 174 patients. Following this, a cohort of NTRK1-related fusions, such as TMP3-NTRK1 and LMNA-NTRK1, indicates a notable presence in NTRK-positive sarcomas. Our review identified 1,019 female patients with gynecological sarcomas. The molecular characterization study by Gatalica et al. supports these findings, highlighting ETV6-NTRK3 and TMP3-NTRK1 as the most prevalent fusions [4]. Furthermore, a systematic review and meta-analysis by Forsythe et al. revealed the prevalence of NTRK1-3 fusions to be 90.56% in 91 patients with infantile fibrosarcomas and 0.34% in 93 patients with uterine sarcomas [159]. However, due to the rarity of these tumors, no large studies have been able to rigorously explore or confirm the association between NTRK1-3 fusions and specific sarcoma subtypes beyond the few known rare entities. Various diagnostic techniques are employed to detect NTRK1-3 fusions, each with advantages and limitations. Traditional methods like FISH and RT-PCR are highly sensitive and specific for detecting known, pre-specified fusions but are inherently limited in their ability to identify novel or rare fusion partners [5]. In contrast, pan-TRK immunohistochemistry has emerged as a valuable, cost-effective, and widely accessible screening tool, capable of suggesting the presence of an NTRK fusion across different sarcoma subtypes [161]. However, current research highlights significant diagnostic challenges unique to sarcomas, as pan-TRK IHC may yield false-positive results in tumors with inherent neural or smooth-muscle differentiation, necessitating confirmatory testing [162,163]. Consequently, genomic sequencing has solidified its role as the gold standard method and is the most commonly used diagnostic modality, offering comprehensive genomic profiling and the ability to detect a wide range of genomic alterations, including novel and rare fusions [11]. DNA-based next-generation sequencing (NGS) panels offer comprehensive genomic profiling, though they can miss fusions with large intronic breaks or complex rearrangements [164,165]. For this reason, RNA-based NGS is increasingly recognized for its superior accuracy, as it directly detects the expressed fusion transcript, making it ideal for identifying cryptic or novel partners. Nevertheless, RNA-based approaches face their own limitations, including susceptibility to RNA degradation in commonly available formalin-fixed paraffin-embedded samples and higher cost considerations [166]. To overcome these hurdles, ongoing research is evaluating integrated diagnostic pathways and technological advances, such as hybrid DNA/RNA sequencing platforms and anchored multiplex PCR methods, which are proving more robust in detecting the full spectrum of NTRK1-3 fusions in sarcomas [167,168]. Despite its higher cost and technical complexity, genomic sequencing is increasingly preferred in clinical settings due to its robust diagnostic capabilities and was recommended by the ESMO guideline for routine testing in patients with advanced cancers, including sarcomas [169]. The identification and targeting of NTRK1-3 fusions have several critical clinical implications. For oncologists, incorporating routine genetic testing for NTRK1-3 fusions into diagnostic workflows can significantly enhance the precision of treatment plans. Patients diagnosed with NTRK1-3 fusion-positive sarcomas can be directed toward TRK inhibitors, which offer a more effective and less toxic alternative compared to traditional chemotherapy, as shown in clinical trials [170,171]. Second, it provides important prognostic information, as NTRK fusion-positive sarcomas have distinct clinical behaviors and outcomes, as previously mentioned. Lastly, the detection of NTRK1-3 fusions can guide the development of personalized treatment plans, improving overall patient management and outcomes. TRK inhibitors, such as larotrectinib and entrectinib, have revolutionized the treatment landscape for patients with NTRK1-3 fusion-positive tumors. Clinical trials have demonstrated significant responses, with high overall response rates and durable responses in many patients [172]. Although with limited evidence due to the infrequency of NTRK gene fusion mutation and the new emergence of TRK inhibitor drugs, we observed that larotrectinib was the most utilized TRK inhibitor, employed as monotherapy in 24 studies [37, 38, 59, 67, 70, 72, 75, 78, 80, 88, 97, 101, 117, 121, 127, 128, 130, 133, 136, 139, 142, 143, 147, 151], and demonstrated a response rate of 83.80% (119/142 patients). This high response rate reinforces the clinical value of identifying NTRK fusions and initiating targeted therapy. Among the 142 cases, only four (2.82%) patients died, 12 (8.45%) developed metastasis, and three (2.11%) exhibited local recurrence. Similarly, entrectinib was also predominantly used as monotherapy in six studies [15,79,82,99,140,150] with 33 patients, yielding a response rate of 63.64%. While our findings suggest that larotrectinib may offer a marginal advantage over entrectinib, these results should be interpreted with caution, given the differences in sample sizes between the two treatment groups. Nevertheless, the overall findings of our review are consistent with those reported by Suh et al. [173]. However, the therapeutic landscape is rapidly evolving beyond first-generation inhibitors. A critical area of current research focuses on overcoming resistance, which inevitably develops in some patients. Primary or acquired resistance can occur through on-target kinase domain mutations, such as TRKA G595R or TRKC G623R, or through off-target bypass activation of alternative pathways like MAPK and MET [174]. This understanding has directly fueled the development of second-generation TRK inhibitors, such as selitrectinib (LOXO-195) and repotrectinib (TPX-0005), which are designed to maintain activity against these resistance mutations and are currently undergoing phase II/III evaluation [175,176]. Furthermore, innovative combination strategies, such as TRK and MEK inhibition, are being explored to preempt or delay the emergence of resistance [177,178]. Beyond efficacy, there is a growing emphasis on understanding the long-term impact of these therapies. While our data and clinical trials report satisfactory response rates, real-world evidence on long-term outcomes, durability of response, and quality-of-life impact in sarcoma-specific populations remains limited, as most evidence is derived from mixed tumor cohorts. Establishing multi-institutional sarcoma registries dedicated to collecting this real-world data will be crucial for generating robust, comparative evidence against traditional modalities and for optimizing the lifelong management of these patients. Thus, with the evidence presented here, we recommend screening for NTRK1-3 fusion genes in patients with soft tissue sarcomas unresponsive to common modalities of management, including surgery, chemotherapy, and radiotherapy [179]. Based on the literature review, for patients with a high suspicion of NTRK1-3 fusion, genomic DNA- or RNA-based sequencing is recommended. RNA-based genomic sequencing provides the highest sensitivity and specificity in detecting NTRK1-3 fusions, with specificity reaching up to 100% [180]. In cases with a lower suspicion of NTRK1-3 fusion, such as in sarcomas other than congenital fibrosarcoma, the initial step should be to perform pan-TRK IHC. If the IHC results are positive, then RNA-based NGS should be used to confirm the presence of an NTRK fusion [181]. Moreover, TRK inhibitors are highly effective in treating soft tissue sarcomas in adult and pediatric populations, with low rates of non-adherence, severe adverse effects, and mortality. Some limitations of our studies include variations in study design, patient populations, detection methods, and reporting standards, which make it difficult to combine and compare results and draw firm conclusions. Additionally, there is limited availability of data regarding the clinical utility of TRK inhibitors; therefore, a direct comparative analysis between first- and second-generation TRK inhibitors was not feasible, as well as the specific types of sarcomas, which complicated subtype-specific interpretation of the prevalence of NTRK1-3 fusions. Conclusion This systematic review provides an updated synthesis of the clinicopathologic spectrum, diagnostic approaches, and therapeutic implications of NTRK1-3 fusions in soft tissue sarcomas. We recommend incorporating routine genetic testing for NTRK1-3 fusions in soft tissue sarcomas with poor response to traditional management approaches, such as chemotherapy and radiotherapy. Genomic sequencing, DNA or RNA-based, is the gold standard owing to its strong diagnostic features and hence is the recommended diagnostic modality to be used in high clinical suspicion. However, pan-TRK IHC is recommended to be used as an initial step in cases of lower clinical suspicion. The availability of TRK inhibitors, which have good efficacy, underscores the importance of integrating routine NTRK1-3 genetic testing into the diagnostic workflow. The results of this study should be interpreted with caution, as the generalizability is limited by insufficient data for certain sarcoma subtypes and inconsistent fusion detection and reporting methods. Future perspectives NTRK1-3 fusions are rare in soft tissue sarcomas. Our systematic review provides further evidence for using TRK inhibitors in clinical settings, aiding clinicians in decision-making and precision medicine options for advanced/refractory sarcoma patients. Given the notable clinical efficacy of TRK inhibitors in sarcomas, genomic testing remains essential in all advanced sarcomas to identify these targetable alterations. Molecular testing for NTRK1-3 fusions should be incorporated into diagnostic workflows in the era of precision oncology. Supplementary Material Supplementary File S1.docx Supplementary File S2.docx Suppl tables revised clean.docx Supplementary file S3.xlsx Acknowledgments The preliminary data from the current study were presented at the 9th QU Health Sector Annual Conference at Qatar University, Doha, Qatar, in May 2025, and the European Society of Medical Oncology (ESMO) Annual Congress, October 17–21, 2025, Berlin, Germany. Open Access funding provided by the Qatar National Library. Funding Statement This paper was not funded. Article highlights The oncogenic Neurotrophic Receptor Tyrosine Kinase (NTRK) 1–3 gene fusions result in the overexpression of Tropomyosin Receptor Kinases (TRK) proteins, driving tumorigenesis. The oncogenic Neurotrophic Receptor Tyrosine Kinase (NTRK) 1–3 gene fusions result in the overexpression of Tropomyosin Receptor Kinases (TRK) proteins, driving tumorigenesis. This study aims to review NTRK1-3 gene fusions across different soft tissue sarcomas and highlight the role of genomic profiling in guiding personalized treatment. This study aims to review NTRK1-3 gene fusions across different soft tissue sarcomas and highlight the role of genomic profiling in guiding personalized treatment. Out of 18,077 patients with sarcoma, 821 (4.54%) harbored a genomic alteration in one of the NTRK1–3 genes, with gene fusions being the most frequent type (551/821; 67.1%). Out of 18,077 patients with sarcoma, 821 (4.54%) harbored a genomic alteration in one of the NTRK1–3 genes, with gene fusions being the most frequent type (551/821; 67.1%). Larotrectinib was given to 142 patients, achieving an 83.8% response rate, with low mortality (2.8%) and recurrence (2.1%) rates, whereas entrectinib showed a lower response rate of 63.6%. Larotrectinib was given to 142 patients, achieving an 83.8% response rate, with low mortality (2.8%) and recurrence (2.1%) rates, whereas entrectinib showed a lower response rate of 63.6%. NTRK1-3 gene fusions are rare in soft tissue sarcomas, and with the availability of targeted therapies, genomic testing remains essential for their identification and the initiation of appropriate treatment. NTRK1-3 gene fusions are rare in soft tissue sarcomas, and with the availability of targeted therapies, genomic testing remains essential for their identification and the initiation of appropriate treatment. Author contributions SH – Conceptualization, Investigation, Formal analysis, Validation, Visualization, Data Curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing, Funding acquisition, Supervision, Project administration; ZA – Conceptualization, Investigation, Formal analysis, Validation, Visualization, Data Curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing, Supervision, Project administration; AMH – Conceptualization, Investigation, Formal analysis, Validation, Visualization, Data Curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing; AAR – Conceptualization, Investigation, Validation, Visualization, Data Curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing; HER – Conceptualization, Investigation, Validation, Visualization, Data Curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing; MAA – Conceptualization, Investigation, Validation, Visualization, Data Curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing; GRB – Conceptualization, Investigation, Formal analysis, Validation, Visualization, Data curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing, Funding acquisition, Supervision; ZG – Investigation, Formal analysis, Validation, Visualization, Data curation, Resources, Software, Methodology, Writing – Review & Editing; SV – Conceptualization, Investigation, Formal analysis, Validation, Visualization, Data curation, Resources, Software, Methodology, Writing – Original Draft, Writing – Review & Editing, Funding acquisition, Supervision, Project administration, Guarantor of Review Disclosure statement The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript. Reviewer disclosures Peer reviewers on this manuscript have no relevant financial or other relationships to disclose. Supplementary material Supplemental data for this article can be accessed online at https://doi.org/10.1080/14796694.2025.2597405 References Papers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers. Associated Data Supplementary Materials Supplementary File S1.docx Supplementary File S2.docx Suppl tables revised clean.docx Supplementary file S3.xlsx