Efficacy of larotrectinib in pediatric cancers with NTRK gene fusions Authors: Peter J. Houghton, Mary-Ann Bjornsti Published: 2025 Larotrectinib targets products of NTRK gene fusions Larotrectinib is a potent, selective ATP-competitive inhibitor of tropomyosin-related kinases (TRKs), encoded by NTRK1, -2, -3 genes and constitutively activated in multiple cancer types usually by chromosomal translocations. In normal growth and development, the neurotrophic genes NTRK1, NTRK2, and NTRK3 encode neurotrophin receptors TRKA, -B, and -C, respectively. In organogenesis, TRKs are involved in neuronal growth, functional development of neuronal synaptic processes and in processes such as memory development, learning (1) and protection of neurons following trauma (2). The kinase activity of neurotrophin receptors is activated by ligand binding to the extracellular domain, and subsequent receptor dimerization leads to kinase activation and downstream signaling. Although TRK receptors bind different ligands, activation of each receptor leads to activation of the mitogen activated protein kinase (MAPK), phosphatidylinositol-3-kinase (PI3K) and phospholipase C-gamma (PLCγ) pathways (Figure 1; further detail can be found in references (3,4)]. Oncogenic drivers comprising NTRK genes involve the fusion of the 5' end of over 80 different genes (5), many encoding oligomerization domains (6), with the 3' end of an NTRK gene that encodes the kinase domain. While oligomerization could account for kinase activation (analogous to ligand induced activation of the wild type receptors), recent studies suggest that fusions of genes encoding proteins both with and without oligomerization domains undergo liquid-liquid phase separation (LLPS) becoming concentrated in membrane-less protein granules (7,8), into which downstream signaling components are recruited. Thus, proteins encoded by NTRK gene fusions bear some similarity to the EWS::FLI fusion protein in Ewing sarcoma that also undergoes LLPS (9,10). A second class of NTRK gene fusions that include the transmembrane domain of the kinase gene may insert into the membranes of the endoplasmic reticulum (Figure 2) (5). Signaling pathways downstream of ligand binding to TRK receptors, and proposed function in brain development. Adapted from Khotskaya et al. and Pramanik et al. (3,4). mRNA, messenger RNA; NT, neurotrophin; TRK, tropomyosin-related kinase Schema showing characteristics of fusion proteins and their distribution in cells. Left: incorporation of fusion proteins into the endoplasmic reticulum (red color) may occur if the break point in the kinase domain of the NTRK gene is 5' to the transmembrane domain, e.g., TPM3::NTRK1 identified in 16 of 41 STS and 8 IFS patients in the cohort. Right: fusion proteins lacking the kinase transmembrane domain localize in membrane-less cytoplasmic inclusions (red color), e.g., ETV::NTRK3 or EML4::NTRK1 fusions identified in 39 and 2 of 49 IFS patients, respectively, in the cohort. Adapted from Zhu et al. (5). IFS, infantile fibrosarcoma; STS, soft tissue sarcoma. Frequency of NTRK gene fusions in cancer The original NTRK fusion identified involved fusion between the tropmyosin-3 gene (TPM3) and NTRK1 (11) leading to NTRK fusions being referred to as TRKs. NTRK gene fusions are relatively rare in cancers with adult pan-cancer frequency estimates of 0.03–0.7% (12). However, there are cancers where NTRK gene fusions are frequent oncogenic drivers such as secretary carcinomas of breast where NTRK gene fusions are identified in ~90% of patients, secretory carcinoma of salivary gland (83–89%), infantile fibrosarcoma (IFS; 70–85%) (12,13), congenital mesoblastic nephroma (CMN), and spindle cell neoplasm as well as other rare pediatric sarcomas. The incidence of IFS and CMN is 5 and 8 per million live births, respectively, while the incidence of NTRK gene fusions in other pediatric sarcomas is around 1% and also at low frequency in papillary thyroid carcinoma (14). NTRK gene fusions are reported in pediatric high-grade glioma (5.3%) with a lower frequency (2.5%) in low-grade glioma (15). Rare examples of NTRK gene fusions have been reported in gastrointestinal stromal tumors (GIST) (16), xanthogranuloma (17), neuroblastoma (18), angiosarcoma (19) and non-small cell lung cancer (NSCLC) (20). Clinical activity of larotrectinib in pediatric cancer The article by Mascarhenus et al. (21) reports the longitudinal outcome of patients, under 18 years of age with NTRK gene fusions, enrolled in two clinical trials who were treated with larotrectinib. Of considerable importance are the single agent activity of larotrectinib and the outcomes for children with elective discontinuation of treatment. This analysis included 91 patients enrolled in the SCOUT phase I/II study (NCT02637687) and a single patient from the NAVIGATE phase II basket trial (NCT02576431). The cohort included 49 patients with IFS and 41 patients with other soft tissue sarcomas (STS) with identified NTRK gene fusions, most commonly spindle cell mesenchymal tumors. Of these patients, only 34% were treatment naïve with the remaining patients having received 1 to ≥3 prior systemic therapies. Previous therapies included surgery (37%), radiation therapy (7%) and systemic therapy (63%). In patients receiving prior systemic therapy the objective response rate for the most recent systemic therapy prior to larotrectinib was 20% (12/60) suggesting chemo-resistance in this population. Larotrectinib was administered twice daily (100 mg/m2) on a 28-day cycle. The overall response rate (ORR), assessed by independent review, was 87% (95% confidence interval: 78% to 93%). Of these 52% had complete responses (CRs) that included 13 pathologic CRs, 35% partial responses (PRs), 8% stable disease (SD) and 3% progressive disease (PD). These results are similar to those in previously untreated IFS and NTRK gene fusion positive tumors published as a companion paper (13), where the ORR was 94% within 6 cycles of treatment. Thus, clearly larotrectinib is a very active agent, with minimal serious toxicity associated with prolonged treatment, a so-called ‘magic bullet’ that targets the oncogenic driver in these malignancies. Although not the primary focus of this article, development of resistance to larotrectinib in this cohort occurred quite frequently with 13 patients (31.7%) in the STS group progressing on treatment with 9 subsequent deaths, whereas in the IFS cohort there were 6 patients (12.2%) who progressed on treatment without mortality up to the cut-off date for analysis. Resistance to larotrectinib, like other adenosine triphosphate (ATP) competitive kinase inhibitors, such as imatinib that inhibits the Abl kinase in the BCR-ABL fusion in chronic myelogenous leukemia (CML) (22), can result from mutation in the kinase domain (23). Understanding the precise mutation(s) resulting in resistance to larotrectinib in pediatric patients will be critical in developing efficacious second generation TRK inhibitors for patients progressing on larotrectinib treatment (23). Outcomes after elective discontinuation of larotrectinib The second aspect of this study that makes it distinct from other phase I/II clinical reports is the follow-up of patients who elected to discontinue larotrectinib treatment. While discontinuation of larotrectinib has occurred due to issues of sustainable access to drug, the protocol criteria that allowed a patient to elect discontinuation were, on-study surgical resection, ongoing nonsurgical CR, PR ≥1 year or SD ≥2 years. This wait-and-see cohort included 30 patients diagnosed with IFS and 17 with STS. Almost half of the patients discontinued larotrectinib after tumor resection with 11 having negative surgical margins, 8 with microscopic residual tumor and 1 with macroscopic residual disease. Tumor recurrence occurred in 16 patients. Time from drug discontinuation to progression ranged from 3 to >24 months (median 3.9 months) with most patients exhibiting tumor progression within 6 months (75%). Median time to progression for either nonsurgical or surgery groups was not reached, and all patients were alive at the time of data cutoff. Lack of tumor recurrence in patients with negative tumor margins is consistent with previous IFS studies (24) where 5-year survival was 100%, whereas a lower 5-year survival (74%) was observed for patients with residual disease following resection (25). Of importance, 15 of the 16 patients who had tumor progression had disease control upon restarting larotrectinib treatment with 11 patients achieving at least PR, demonstrating continued sensitivity to the TRK inhibitor. Conclusions and future challenges Overall, this study and its companion paper (13) demonstrate the remarkable efficacy and tolerability of larotrectinib in patients having NTRK gene fusions. Both studies conclude that surgical resection for local control remains an essential component of successful therapy following tumor response to larotrectinib which facilitates less aggressive or potentially mutilating surgery and avoids potentially serious toxic chemotherapy. Notable is the lack of neurological toxicity when one considers the role of NTRK genes in neural development, despite long-term administration of drug. In part, this may be due to lower penetration to normal brain, larotrectinib being excluded from brain tissue by ABCB1 and ABCG2 transporters (26). However, larotrectinib is active in gliomas with NTRK gene fusions, although the ORR is substantially lower than reported for extracranial diseases such as NTRK fusion positive IFS and STS (27). Nevertheless, despite the remarkable activity of larotrectinib in IFS patients, the response rate and rates of progression on treatment in NTRK gene fusion STS cohort still presents a clinical challenge (21). Understanding mechanisms for intrinsic and acquired resistance to NTRK inhibitors and development of relevant pediatric models with NTRK gene fusions will allow development of second-generation inhibitors that overcome larotrectinib resistance (23). Such models would allow assessment of therapeutic approaches with drug combinations that target multiple steps in the signaling pathways activated by specific NTRK fusions (vertical targeting) as in MAPK-activated sarcomas (28,29). Larotrectinib thus represents the new standard of care for NTRK gene fusion positive pediatric cancers prior to conservative surgery in accordance with guidelines proposed for IFS (30). The Mascarenhas et al. study (21) also shows the power of multinational cooperative trials for rare tumors, with 29 centers from 10 countries contributing to the study. These two clinical studies (21,31) provide a template for future trials should another ‘magic bullet’ be identified for rare childhood cancers (32). The therapeutic activity of larotrectinib highlights the relevance of targeting the genetic basis of the specific cancer in order to develop effective therapeutics with reduced acute and long-term toxicities. Other notable advances in treatment of pediatric cancers include treatment of BRAFV600E-activated pediatric low-grade glioma with trametinib/dabrafenib (33) and treatment of patients with neurofibromatosis type I-associated plexiform neurofibromas (NF1) with selumetinib (34), which was recently approved by FDA. Supplementary The article’s supplementary files as Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-664/prf