A total of 20 publications encompassing 61 cases of NTRK-rearranged spindle cell neoplasms in the female genital tract were included in the analysis. Among these, 54 cases were located in the uterine cervix and 7 in the uterine corpus. The clinicopathological features of all 61 patients are summarized in Supplementary Table 2 . Data from the literature were statistically analyzed using SPSS software version 25.0, with a p-value of <0.05 considered statistically significant.

The mean age at diagnosis was 39 years (range: 13–69 years), and the average tumor size was 7.0 cm (range: 1.3–23.0 cm). At the time of diagnosis, most tumors were confined to the uterus, with 88% of patients classified as stage IA or IB according to the FIGO system. The average mitotic count was 12 per 10 HPFs, ranging from 0 to 50. Tumor necrosis was present in 40% of cases, and lymphovascular invasion was identified in 20%. Follow-up data were available for 50 patients. Among them, 26 patients (52%) had no evidence of disease (NED), while 7 patients (14%) died of disease (disease-specific death, DSD). In stage IA patients, 83% (10/12) were NED, and 17% (2/12) were alive with disease (AWD); notably, two patients in this group developed distant metastases involving the pleura and bone. Among patients at stage IB, 57% (12/21) were NED, 19% (4/21) were AWD, and 24% (5/21) died of disease. Only five patients were diagnosed at more advanced stages, including stage IIA (n = 2), stage IIB (n = 2), and stage IVB (n = 1) ( Table 1 ).

Molecular detection revealed that NTRK1 fusions were the most common type (44/59, 75%), followed by NTRK3 fusions (14/59, 24%) and a single NTRK2 fusion (1%). Various fusion types were identified ( Table 1 ), among which TPM3::NTRK1 was the most common type (24/41, 59%). The comparative analysis demonstrated that tumors harboring NTRK3 fusions were significantly larger than those with NTRK1 fusions (mean diameter: 10.4 cm vs. 5.5 cm, p = 0.011) and more often presented at advanced FIGO stages (>IB vs. ≤IB, p = 0.016). Furthermore, chi-square analysis revealed a significantly higher disease-specific mortality rate in patients with NTRK3 fusions compared to those with NTRK1 fusions (33% vs. 6%, p = 0.028), suggesting a poorer prognosis associated with NTRK3 fusion tumors ( Table 2 ). Although NTRK3 fusion tumors also exhibited a higher mitotic index and increased recurrence rates, these differences did not reach statistical significance. No significant differences were observed between the NTRK1 and NTRK3 fusion groups in terms of patient age, presence of lymphovascular invasion, or tumor necrosis.

The three members of the neurotrophic tyrosine receptor kinase (NTRK) family (NTRK1, NTRK2, and NTRK3) encode TRKA, TRKB, and TRKC proteins, respectively (12). These receptors are physiologically expressed in the peripheral and central nervous systems and play critical roles in neural development and function (13–15). Upon ligand binding, the extracellular domains of NTRK receptors undergo dimerization, leading to autophosphorylation and activation of downstream signaling cascades, including the MAPK, PI3K, and PKC pathways. These pathways regulate essential cellular processes such as proliferation, differentiation, and survival (1, 27). NTRK gene rearrangements in tumors can result from either intrachromosomal or interchromosomal translocations. These rearrangements typically involve the fusion of the 5’ end of an NTRK gene with the 3’ end of a partner (chaperone) gene, producing a chimeric fusion protein. When the chaperone gene encodes a domain that facilitates dimerization, the resulting fusion protein can undergo constitutive dimerization without ligand binding. This ligand-independent activation leads to persistent stimulation of the downstream signaling pathways mediated by the TRK kinase domain, thereby promoting oncogenic transformation. This mechanism represents a key driver of tumorigenesis in NTRK-altered malignancies. Frequently reported fusion partners include LMNA, TPM3, PAN3, and ETV6 (14, 15). TRK receptor activation via NTRK fusion is recognized as a pan-cancer oncogenic mechanism, with an estimated incidence of 0.68% to 1% in adult soft tissue sarcomas (1, 8, 14, 27). NTRK rearrangements have been documented in a variety of soft tissue tumors in both adults and children, including infantile fibrosarcoma and congenital mesoblastic nephroma. However, such genetic alterations are exceedingly rare in gynecologic tumors.

To date, few distinct morphological features have been validated as reliable diagnostic indicators for NTRK-rearranged spindle cell neoplasms. Therefore, definitive diagnosis largely depends on IHC analysis and molecular testing.

Among the IHC markers, TRK, CD34 and S-100(especially TRK) are considered valuable for raising suspicion of this tumor type and guiding further molecular investigations. Tumor cells typically exhibit diffuse cytoplasmic positivity for TRK. An exception to this pattern has been reported in a TRK-negative case harboring a SPECC1L::NTRK3 fusion gene, which also demonstrated a loss of SMARCB1 (INI1) expression (23). In addition to the classic cytoplasmic staining, alternative TRK expression patterns have been described. For example, tumors with LMNA::NTRK1 fusions may show nuclear membrane localization, while tumors with TPM3::NTRK1 or ETV6::NTRK3 fusions can exhibit membranous staining patterns (1, 14, 28). Although TRK immunostaining is a useful screening tool, it is important to note that TRK positivity does not always indicate NTRK gene rearrangements. Cases with TRK expression must be confirmed by molecular assays. For instance, some high-grade endometrial stromal sarcomas (ESS) have shown TRK positivity despite lacking any NTRK fusions (22). Therefore, confirmatory testing such as FISH, reverse transcription PCR (RT-PCR), or next-generation sequencing (NGS) is essential to establish the diagnosis. Additionally, the tumor frequently expresses S-100 and CD34 (29). S-100 expression was reported in 43 of 49 cases (88%) and CD34 in 34 of 45 cases (76%) ( Supplementary Table 2 ). Thus, we recommend the use of TRK, CD34, and S-100 immunostaining as an accessible and cost-effective initial screening strategy for all uterine spindle cell tumors, particularly those lacking definitive endometrial stroma or smooth muscle differentiation, to facilitate the identification of NTRK-rearranged neoplasms.

As previously discussed, Many mesenchymal tumors of the uterus present with similar spindle cell morphology, making differential diagnosis based solely on conventional histopathology challenging. In this context, immunohistochemistry is an essential and practical tool to support accurate pathological classification. In our case, a broad range of differential diagnoses need to be considered based on the histological features, including low-grade and high-grade ESS, leiomyosarcoma, and IMT, among others. Low-grade ESS typically expresses CD10 with estrogen and progesterone receptors, whereas high-grade ESS may show immunoreactivity for cyclin D1 and/or BCOR. In contrast, our case was negative for myogenic markers such as desmin, caldesmon, calponin, and MyoD1, as well as for the IMT-associated marker ALK, supporting the exclusion of leiomyosarcoma, rhabdomyosarcoma and IMT. Adenosarcoma with sarcomatous overgrowth should also be excluded, as residual benign endocervical glands were observed among the tumor components in our case, but the characteristic periglandular sarcomatous cuffing typical of adenosarcoma was absent. For rare uterine malignant peripheral nerve sheath tumors (MPNSTs), which may show focal positivity for S-100 and CD34, the key distinction lies in their usual expression of neural markers. Some tumors previously diagnosed as MPNSTs have been reclassified as NTRK-rearranged spindle cell tumors by molecular testing (22).

A recent study by Costigan (7) suggests that several indicators, including both morphological and genetic features, may predict poor prognosis, such as lymphovascular invasion, necrosis, mitotic counts ≥8 per 10 HPFs, and NTRK3 fusion. Interestingly, tumors with NTRK3 fusions were generally larger and exhibited higher mitotic activity compared to those with NTRK1 fusions. They also tended to present at more advanced stages and had a higher likelihood of recurrence, suggesting a poorer outcome for these patients. Tumors exhibiting one or more of these features should be considered high-risk, whereas those lacking all of them may be classified as low-risk (7). According to these criteria, our case should be regarded as high-risk due to the presence of necrosis and active mitosis, and warrants prolonged follow-up despite no evidence of recurrence to date. In the female genital tract, 75% of reported cases harbored NTRK1 fusions, with fusion partners including TPM3, TPR, IRF2BP2, C16ORF72, LMNA, NUMA1, and TRIM67. Notably, all patients with TPM3::NTRK1 fusions (accounting for 59% of cases) were alive at the time of follow-up. These findings suggest that fusion type may correlate with survival outcomes, although further investigation is warranted.

Furthermore, it has been proposed that some NTRK-rearranged sarcomas with high-grade nuclear features may harbor TP53 mutations (7). In general, TP53 mutations result in a mutant-type immunostaining pattern, which may serve as a surrogate marker for underlying genetic alteration. TP53 mutations have been associated with high-grade nuclear features and poor prognosis in other tumors, such as endometrial carcinoma. However, the relationship between p53 expression patterns and prognosis remains unclear in NTRK-rearranged sarcomas of the female genital tract, highlighting the need for further studies exploring the prognostic significance of different p53 staining patterns.

Most patients with NTRK-rearranged spindle cell neoplasms underwent surgical resection followed by adjuvant chemotherapy, with or without radiotherapy. As these tumors are typically negative for hormone receptors and rarely involve the ovaries, ovarian preservation does not appear to adversely affect prognosis (7, 12). The necessity of oophorectomy in young patients with NTRK-rearranged sarcomas remains uncertain. Studies have demonstrated that TRK inhibitors, such as entrectinib and larotrectinib, are highly effective, particularly in patients with advanced or metastatic disease. According to reports in the literature, 109 patients with NTRK gene fusions received larotrectinib treatment, achieving an investigator-assessed overall response rate of 81% (95% CI: 72%–88%). Among them, 63% experienced partial responses, while 17% achieved complete responses. Based on these results, the U.S. Food and Drug Administration (FDA) approved larotrectinib in 2018 for the treatment of solid tumors harboring NTRK gene fusions (4, 6, 10, 16, 17, 30). In our case, the patient underwent radical hysterectomy followed by adjuvant chemotherapy and radiotherapy, but has not received targeted therapy, as the disease was diagnosed at an early stage and no recurrence has been observed to date. Therefore, we are currently unable to assess the efficacy of TRK inhibitors in this case. It is worth noting that follow-up data for this tumor type remain limited, highlighting the need for larger, long-term studies to further evaluate treatment strategies.