Neuroblastoma (NB) is the most common extracranial solid tumor among pediatric cancers and is responsible for 15% of childhood cancer deaths [1, 2, 3]. The process typically begins with primary tumors arising in the adrenal medulla or along the sympathetic nervous system. NB is notable for its biological heterogeneity, and the course of the disease can exhibit diverse clinical courses, including spontaneous regression, differentiation, or metastasis [4, 5].

TRK proteins contain three main domains: an extracellular domain, a transmembrane domain, and an intracellular domain. The tyrosine kinase domain is the most important part of the receptor. The active kinase domain initiates biological responses by triggering various signaling pathways within the cell, including RAS/MAPK, PI3K/AKT, and PLC-$\gamma$ [6].

Neurotrophic receptor tyrosine kinases (NTRKs) are genes that play critical roles in the development and function of the nervous system. NTRK genes (NTRK1, NTRK2, and NTRK3) encode transmembrane receptors with tyrosine kinase activity [7, 8]. These receptors activate signaling pathways through interaction with neurotrophic factors, regulating biological processes such as cell proliferation, differentiation, and survival. However, abnormalities in these genes such as fusions, mutations, or overexpression, can affect tumor growth and metastasis, altering sensitivity to treatment [5, 7].

The role of NTRK genes in NB is important both for elucidating the underlying biological mechanisms and for developing therapeutic approaches. NTRK fusions or aberrant expression are being evaluated as potential biomarkers and therapeutic targets for targeted therapy strategies [8, 9]. In recent years, studies on the role of NTRK genes in NB have been an important step in understanding genetic heterogeneity.

Neurotrophic tyrosine receptor kinase (NTRK) inhibitors target abnormal tropomyosin receptor kinase (TRK) fusion proteins resulting from NTRK gene fusions, blocking their tyrosine kinase activity and thereby inhibiting the growth of cancer cells [10, 11]. Drugs such as larotrectinib and entrectinib bind with high affinity to TrkA, TrkB, and TrkC receptors, inhibiting their phosphorylation and thus downstream signaling pathways. This inhibition inhibits tumor growth by stopping cell proliferation and promoting apoptosis [12, 13].

Accordingly, examining NTRK gene expression and genetic alterations in NB not only provides a better understanding of the disease’s biology but also allows for the development of more effective treatment approaches. The potential of NTRK genes in targeted therapies, particularly the clinical success of tumor-agnostic TRK inhibitors, highlights the importance of NTRK fusions and other abnormalities as therapeutic targets. However, the rarity of NTRK fusions and factors such as genetic heterogeneity necessitate more comprehensive investigation of other alterations in these genes, particularly mutations and aberrant expression, to understand their clinical significance.

This study aimed to investigate neurotrophic tyrosine kinases at the genomic and protein levels in 173 patients diagnosed with NB, obtained from the Turkish Pediatric Oncology Group (TPOG) Association’s completed TPOG Neuroblastoma 2009 Protocol and the ongoing TPOG Neuroblastoma 2020 Protocol. For this purpose, paraffin block samples from patients diagnosed with NB were used to investigate pan-tyrosine kinase protein expression by immunohistochemistry (IHC); NTRK1, NTRK2, and NTRK3 gene fusions by RNA next generation sequencing (NGS) analysis; and single nucleotide variants (SNVs) in the NTRK1, NTRK2, and NTRK3 genes by DNA NGS analysis.

This study investigated the distribution and clinical significance of NTRKs in 173 NB cases using point mutations in the NTRK1, NTRK2, and NTRK3 genes from DNA samples, fusion analyses from RNA samples, and IHC expression rates. The study results revealed a high percentage of immunohistochemical expression of NTRKs (67.9%) and NTRK point mutations (78.26%) in NB, while the percentage of NTRK fusions (3.88%) was low. NTRK missense mutations were commonly seen in tyrosine kinase domains of NTRKs. A study has determined whether the c.1810C$>$T mutation is commonly found in neuroblastoma patients younger than 18 months of age with NTRK1 non-MYCN amplification and is associated with an unfavorable prognosis [16].

In our study, NTRK mutations were found at a high frequency in neuroblastoma patients. According to the literature, NTRK mutations are rare and, while most are polymorphisms, are rarely seen in cancers. Furthermore, studies have reported that NTRK point mutations (especially found in tyrosine kinase domain) have oncogenic potential and can cause TRK inhibitor resistance [17, 18]. NTRK1 L564H, V573M and G667C mutations or NTRK3 F617L, G623E, G623R and G696A were detected by NGS and resistant to Cabozantinib, Larotrectinib and Entrectinib that are TRK inhibitors [17, 18, 19]. Joshi et al. [20] identified somatic NTRK2 and NTRK3 mutations in hematological neoplasms and functionally demonstrated to be transforming/oncogenic that NTRK2 and NTRK3 point mutations. NTRK1 p.G595R and p.G667C mutations have been associated with entrectinib/larotrectinib resistance in tumors with NTRK fusions (e.g., colorectal cancer). These mutations were generally identified as resistance mutations located within the ATP-binding pocket of the TRK fusion protein [21]. In our study, we found NTRK1 p.G667C and NTRK3 F617L mutations.

NTRK gene fusions were first identified in a colorectal adenocarcinoma cell line in 1982 [22]. The concept of tumor-agnostic markers refers to markers used as therapeutic targets in tumors with specific genetic or molecular characteristics (e.g., a specific biomarker or mutation), regardless of cancer type. This concept represents modern oncological approaches, where cancer treatment is based on the molecular profile of the tumor rather than its location [23]. The approval of larotrectinib as a tumor-agnostic treatment by the FDA in 2018 and by the European Medicines Agency (EMA) in 2019 has provided access to treatment for patients with NTRK gene fusion regardless of age and tumor type [9, 24, 25].

In our cohort, NTRK fusions were identified in 4 of 103 tested patients (3.9%), comprising 2 patients having NTRK1 and 2 patients having NTRK3 fusions. This detection ratio is concordant with the low prevalence of NTRK alterations reported across solid tumors, yet it is notably higher than the 8.7% fusion prevalence observed by Mohamed et al. [26] in CNS tumors when performing a combined NGS and Pan-TRK IHC workflow. Importantly, while Mohamed et al. [26] demonstrated that Pan-TRK expression lacked specificity—detecting fusion–negative tumors in 82% of Pan-TRK–positive CNS cases—our study relied exclusively on RNA-based NGS for fusion detection, thereby avoiding false-positive immunoreactivity and ensuring high analytical accuracy. The absence of NTRK2 fusions in our cohort also contrasts with their study, where both confirmed fusions were NTRK2-driven, underscoring potential biological differences between tumor types and patient populations [26]. Collectively, these findings highlight the substantial variability of NTRK fusion across tumor groups and strengthen RNA-based NGS as a more trusted diagnostic method compared with IHC, particularly in settings where physiologic TRK expression or tumor heterogeneity may confound IHC interpretation.

In a study by Cocco et al. [7], TRKA and TRKC expression in NB patients was generally a good prognostic indicator, while TRKB was primarily expressed in high-grade tumors that also harbor MYCN amplification and was associated with a poor prognosis. This study used a pan-TRK antibody containing all three proteins. In the study by Vingiani et al. [14], they detected NTRK fusion in only 11 of 30 (37%) cases with positive pan-TRK IHC and in 1 of 87 (1.1%) cases with negative pan-TRK IHC. This study included a cohort of 124 cancer patients (carcinomas, 93 cases; soft tissue sarcomas, 19; primary central nervous system tumors, 10; and neuroblastomas, only 2 cases and negative for fusion). Based on this information, IHC sensitivity was calculated as 91.7% and specificity as 81.9% [14]. However, when we examined the positive predictive value, it was low at 36.7%, while the negative predictive value was high at 98.8%. This indicates that there are many false-positive results. When we considered IHC results according to histological types, higher concordance was seen in carcinomas (94.2%), while the concordance rate was low in nervous system tumors and neuroblastomas (18.2%). It has been reported that IHC expression patterns of the detected TPR-NTRK1 fusion showed cytoplasmic and membranous staining patterns [14]. Compared to these data, in our study, of all cases showing positive IHC, only 44 had fusion studies and 46 had point mutation studies. Percentage-wise, fusions were observed in 6.81% of cases with immunohistochemical expression. When point mutations were evaluated, clinically significant mutations were detected in 34.78% of positive cases, while no mutations were detected in 65.21%. The high NTRK rate in our study was associated with the fact that a large proportion of the patients included were advanced-stage, relapsed-refractory patients.

In another study, Zhao et al. [27] investigated the frequency of NTRK fusions in pediatric tumors. In a cohort of 1347 pediatric patients, NTRK fusions were detected in 2.22% of all tumors and 3.08% of solid tumors. NTRK1 and NTRK3 fusions were commonly observed in papillary thyroid carcinomas, while NTRK2 fusions were more common in central nervous system tumors, and these tumors were mostly low-grade gliomas. In soft tissue tumors, fusions such as ETV6-NTRK3 were found in rare tumors such as spindle cell tumors and infantile fibrosarcoma. TRK inhibitors have emerged as an effective method to support surgery in these types of tumors. No NTRK fusion was detected in any of the 79 NB cases examined in the study [27].

In Cox model analysis, NTRK positivity was not independently associated with overall survival, but it showed a non-significant trend toward improved event-free survival. NTRK single-nucleotide variants were not independently associated with either event-free or overall survival, although a trend toward poorer event-free survival that was not significant. The lack of statistical significance may result from the low frequency of NTRK expression and the limited number of events.

Light et al.’s study [28] evaluated the impact of NTRK gene family expression levels on the clinical features and outcomes of NB patients. 814 NB patient samples were analyzed using Quantitative Real-Time Reverse Transcriptase PCR (qRT-PCR). NTRK1 expression was associated with all established risk factors; Statistically significant differences were found in patients younger than 18 months, with low stage, MYCN amplification negative, hyperploidy, and good histology. In patients with high NTRK1 expression, 5-year EFS and OS rates were significantly higher than in patients with low expression levels (p 0.0001). NTRK2 was found to be significantly associated with age, MYCN, and ploidy status, but not with survival outcomes. However, NTRK3 was not significantly associated with risk factors but was associated with worse OS [28]. Our own research findings indicate that NTRK-IHC negative patients had longer EFS and OS times than positive patients. Although NTRK positivity had an effect on EFS and OS, this was not found to be statistically significant.

As suggested by Märkl et al. [29], in cases with a high NTRK splicing rate, such as congenital fibrosarcoma, FISH, RT-PCR, and NGS are recommended for direct or positive results after NTRK IHC. In cases with low frequency, direct or panTRK-IHC tests were recommended if they included multiple gene assessments, and FISH and NGS were recommended for positive results. For negative results, no other analyses were required [29].

Tyrosine kinases are tumor-agnostic markers, and targeted therapies are available for them. Detecting the presence of NTRK fusions in cancers is crucial for initiating treatment. However, the importance of point mutations, gain-of-function mutations, and missense mutations for the application of NTRK-targeted therapy in NB has not been fully elucidated. In this study, IHC staining of NTRK showed NTRK expression in the majority of NB patients. Therefore, it was concluded that it cannot be a leading biomarker for fusion.

In this study, some samples were not suitable for IHC or NGS analyses due to lack of tissue in the paraffin block, inadequate tissue follow-up, or limited sample size. Furthermore, budget constraints precluded the application of NGS to all samples. However, the relatively large sample size increased statistical power. Tumor mutational burden assessment or broad copy-number analysis are the limitations of our study in terms of NGS panel that we used in this study. One of the limitations of this study is that the detected SNVs were previously interpreted and classified in clinical databases and did not include functional validation. The significant heterogeneity of neuroblastomas is another limitation that must be considered in the interpretation of the findings.

In conclusion, evaluation for TRK inhibition is needed in all cases of NB, a tumor of neural origin. NTRK fusions and NTRK point mutations or other SNVs should be both evaluated to determine possible resistance to NTRK inhibitors. Our data supported that Pan-NTRK immunohistochemistry is not a good biomarker for the selection of further NGS studies. In this study, no concordance was found with nuclear immunohistochemical NTRK expression in the four cases in which fusions were detected.

NTRK fusions are rare in NB, as confirmed in our study and the literature. As an agnostic therapeutic target, RNA fusion and multigene DNA panels containing NTRK genes should be evaluated using NGS in relapsed, refractory, high-risk, poor clinical conditions, and chemotherapy-resistant NB cases.

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

SA and TÇA designed the study. DK, NI, SMÖ, AE, and ÖEG collected the patient samples and data. NI, SMÖ, AE, and ÖEG conducted the research and laboratory experiments. SA, TÇA, and SMÖ performed the statistical analysis of the data. All authors participated in the writing and editing of the manuscript. All authors read and approved the final version of the manuscript. All authors participate in the study, taking responsibility for the adequacy of the content and contributing to the accurate and complete completion of the work.

Ethical Approval was taken from Dokuz Eylul University Non-Interventional Research Ethics Committee on 07.05.2023 with decision number 2023/22-14. Approval was obtained by having the patients’ parents sign Informed Consent forms. The study was carried out in accordance with the guidelines of the Declaration of Helsinki.

We gratefully acknowledge the Turkish Pediatric Oncology Group – Neuroblastoma team.

This research was supported by the Turkish Pediatric Oncology Group Association 2024 Research Projects Support Program for the immunohistochemical step. This study was also partially supported by Dokuz Eylül University Scientific Research Council for the NGS analysis step (Project no: TSA-2023-3007).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/FBS46963.