NTRK Fusion Test: Cancer Targeted Therapy, Gene Fusion, and Results

An NTRK fusion test looks for an abnormal joining of NTRK1, NTRK2, or NTRK3 with another gene. The fusion creates a continuously active TRK protein that can drive cancer growth regardless of where the tumor began. Although NTRK fusions are rare across common adult cancers, they are frequent in several uncommon tumors, including infantile fibrosarcoma, secretory breast carcinoma, secretory salivary carcinoma, and cellular congenital mesoblastic nephroma. A confirmed fusion can qualify adults or children with advanced solid tumors for tumor-agnostic TRK inhibitor therapy when the clinical indication and regulatory requirements are met. The test must identify a true expressed, in-frame fusion with an intact kinase domain; an NTRK point mutation, amplification, or nonspecific TRK protein stain is not equivalent. RNA-based next-generation sequencing is often the most direct broad method, while immunohistochemistry can screen selected tumors and FISH can confirm specific settings. Negative results require attention to assay design because DNA panels may miss difficult NTRK2 or NTRK3 intronic breakpoints.

• A true NTRK1, NTRK2, or NTRK3 gene fusion can predict benefit from TRK inhibitor therapy across many solid-tumor types.
• NTRK point mutations and amplifications usually do not carry the same treatment meaning as an activating fusion.
• Pan-TRK immunohistochemistry is a screening tool; positive staining often needs molecular confirmation.
• RNA sequencing can confirm that the fusion is expressed and preserve the kinase domain.
• A negative DNA panel may need RNA-based testing when the tumor type or clinical pattern strongly suggests an NTRK fusion.

Table of Contents

• What an NTRK Fusion Is
• Who Should Be Tested
• Testing Methods
• Interpreting Results
• Tumors Associated With NTRK Fusions
• TRK-Targeted Therapy
• Resistance and Repeat Testing
• Limitations and Questions to Ask

What an NTRK Fusion Is

NTRK1, NTRK2, and NTRK3 encode the TRKA, TRKB, and TRKC receptor tyrosine kinases. In normal nervous-system development and function, these receptors respond to neurotrophins such as nerve growth factor and activate pathways controlling survival, differentiation, and growth. Their kinase activity is normally regulated by ligand binding.

A cancer-driving fusion joins the 3′ portion of an NTRK gene, including its tyrosine kinase domain, to the 5′ portion of a partner gene. The partner often provides a dimerization domain or strong promoter. The resulting fusion protein activates continuously and signals through MAPK, PI3K–AKT, and PLCγ pathways.

Examples include ETV6::NTRK3, TPM3::NTRK1, LMNA::NTRK1, and many rarer partners. More than 80 partners have been described. The exact partner can support diagnosis in a characteristic tumor, but the presence of an intact NTRK kinase domain and demonstrated oncogenic structure is more important for treatment eligibility.

Not every NTRK alteration is a fusion. Tumor panels may report point mutations, splice variants, copy-number gain, or amplification. Most do not have the established predictive value of a canonical fusion. A report should avoid the vague phrase “NTRK positive” and state the gene, partner, breakpoints or exons, reading frame, method, and classification.

NTRK fusions are somatic in most cancers. They do not usually imply inherited cancer risk. The fusion can appear in a child’s tumor without being present in the child’s other cells. Germline NTRK variants are a separate topic and require a dedicated inherited-disorder evaluation.

Who Should Be Tested

Testing strategy depends on tumor type and fusion frequency. In rare cancers where NTRK fusions are common, focused testing early in the diagnostic workup is efficient. In common cancers where prevalence is usually below 1%, NTRK is best included in broad genomic profiling for advanced disease, especially when standard driver tests are negative.

High-frequency or characteristic settings include:

• infantile fibrosarcoma;
• cellular congenital mesoblastic nephroma;
• secretory carcinoma of the breast;
• secretory carcinoma of salivary glands;
• some pediatric thyroid cancers and high-grade gliomas;
• selected spindle-cell tumors with S100/CD34 co-expression and distinctive morphology.

Lower-frequency settings include lung, colorectal, pancreatic, biliary, thyroid, sarcoma, melanoma, and many other solid tumors. In these cancers, testing may be prioritized when the tumor lacks common drivers. For example, an NTRK fusion is more plausible in a lung adenocarcinoma without EGFR, KRAS, ALK, ROS1, RET, BRAF, or MET alterations than in a tumor already driven by one of them.

Current molecular programs often use a comprehensive genomic profiling test that includes RNA fusion analysis. This reduces the need for sequential single-gene tests and can identify other tumor-agnostic biomarkers at the same time.

Testing may also be ordered at progression when an earlier panel lacked RNA coverage or when a histologic diagnosis is reconsidered. A secretory carcinoma initially labeled as another salivary or breast subtype may be clarified by ETV6::NTRK3 detection.

Testing Methods

No single method is perfect for every tumor. The most reliable pathway considers tumor type, available tissue, expected fusion prevalence, and whether a broad or focused answer is needed.

Method	Role	Strength	Limitation
RNA-based NGS	Broad fusion detection and confirmation	Shows an expressed transcript and identifies partner	RNA can degrade in fixed tissue
DNA-based NGS	Broad genomic profiling	Tests mutations, copy number, and some fusions together	Large NTRK2/3 introns may be incompletely covered
Pan-TRK immunohistochemistry	Screening	Fast, inexpensive, and tissue-sparing	False positives and false negatives occur; pattern matters
FISH	Gene-specific rearrangement testing	Works on fixed tissue and can be rapid	Separate assays are needed for NTRK1, 2, and 3 and partner is unknown
RT-PCR	Known fusion testing	Sensitive and fast	Misses unexpected partners or breakpoints

RNA-based anchored multiplex PCR can detect fusions without knowing the partner in advance. It is often preferred when a DNA panel is negative but suspicion remains. Whole-transcriptome RNA sequencing offers broader discovery but may require more material and complex analysis.

Pan-TRK immunohistochemistry detects C-terminal TRK protein shared by the three genes. Staining can be nuclear, cytoplasmic, membranous, or perinuclear depending on the partner. Neural and smooth-muscle tissues can show physiologic expression, and some tumors express wild-type TRK without a fusion. Molecular confirmation is generally important before treatment, especially in a low-prevalence cancer.

FISH break-apart probes show rearrangement of one NTRK locus but do not prove the resulting transcript is in-frame or expressed. In tumors where ETV6::NTRK3 is a defining event, an ETV6 or NTRK3 FISH result may be highly useful. A gene fusion test should be selected with the likely biology in mind.

Interpreting Results

A pathogenic NTRK fusion detected result should identify NTRK1, NTRK2, or NTRK3 and its partner. The report may describe the joined exons and whether the kinase domain is retained. An in-frame, expressed fusion with a recognized oncogenic structure supports TRK inhibitor eligibility when the tumor is advanced and other clinical criteria are satisfied.

A rearrangement detected by FISH is supportive but may need RNA or DNA confirmation if the pattern is atypical, the tumor type has low fusion prevalence, or treatment depends on the result. Likewise, pan-TRK positive is not a complete molecular diagnosis. The positive predictive value of IHC is much higher in a characteristic high-prevalence tumor than in a common carcinoma.

A not-detected result is only as broad as the assay. A DNA panel can miss NTRK3 fusions because breakpoints lie in large introns not fully captured. An RNA assay can fail because RNA quality is poor. The report should state quality-control metrics and whether fusion detection was technically successful.

A fusion of uncertain significance may be out of frame, lack the kinase domain, involve an unusual orientation, or appear at very low read count. Such a result should not be treated as actionable without confirmation. The molecular tumor board may request orthogonal testing and review the tumor’s morphology and other drivers.

An NTRK mutation or amplification should not be automatically substituted for a fusion. Rare activating mutations and resistance mutations exist, but treatment evidence and approved indications may differ. The report’s therapy section should match the exact alteration, not the gene name alone.

Tumors Associated With NTRK Fusions

NTRK fusions are unusual because prevalence is concentrated at two extremes. They are very common in a small set of rare tumors and very rare in many common cancers. This distribution shapes testing efficiency.

ETV6::NTRK3 is a defining or near-defining event in infantile fibrosarcoma, secretory breast carcinoma, secretory salivary carcinoma, and cellular congenital mesoblastic nephroma. These tumors can occur in very different organs yet share fusion-driven biology. Detecting the fusion can resolve diagnosis as well as treatment.

In differentiated thyroid cancer, NTRK fusions are enriched in children, young adults, and radiation-associated cases, particularly when BRAF and RAS are absent. In lung cancer, they are rare but actionable and generally occur without another dominant oncogenic driver. In colorectal cancer, NTRK fusions are enriched in a small subset of tumors that are RAS/BRAF wild type and may also be MSI-H because of MLH1 methylation.

Gliomas and sarcomas require careful interpretation because many fusion partners and histologic entities exist. An NTRK fusion may define a recently recognized tumor family or coexist with other changes that influence grading. Central pathology review can be valuable in children and rare tumors.

Frequency alone should not determine whether a result is real. A rare fusion in a common tumor can be genuine if it is structurally sound and confirmed. Conversely, a weak IHC stain in a low-prevalence setting can be nonspecific. The method, morphology, and molecular architecture must agree.

TRK-Targeted Therapy

Larotrectinib and entrectinib are TRK inhibitors developed for NTRK fusion-positive solid tumors, with tumor-agnostic approvals that include adults and children under specified conditions. They can produce high response rates across many histologies, and responses may be durable. Entrectinib also inhibits ROS1 and ALK and has central nervous system activity; larotrectinib is highly selective for TRK.

Eligibility generally requires a solid tumor with an NTRK gene fusion, no known acquired resistance mutation, advanced or metastatic disease or surgery likely to cause severe morbidity, and no satisfactory alternative treatment or progression after treatment, depending on the jurisdiction and label. The oncology team should use the current product label and guideline rather than relying on a generic “NTRK positive” statement.

Common adverse effects can include fatigue, dizziness, nausea, constipation or diarrhea, weight gain, liver-enzyme elevation, anemia, and neurologic symptoms. Because TRK signaling has normal nervous-system roles, dizziness, gait changes, sensory symptoms, and withdrawal pain can occur. Pediatric dosing and monitoring require specialist experience.

Tumor shrinkage can make surgery less mutilating in selected locally advanced fusion-positive tumors. Neoadjuvant use should be planned by a multidisciplinary team because treatment duration, timing of surgery, and long-term pediatric effects need careful consideration.

A response is not guaranteed. Tumor lineage, co-alterations, drug exposure, brain involvement, and resistance mechanisms matter. Still, a confirmed NTRK fusion is one of the clearest examples of a molecular finding that can connect many different cancers to the same targeted treatment class.

Resistance and Repeat Testing

Resistance can be present from the start or develop after an initial response. On-target resistance mutations alter the TRK kinase domain at solvent-front, gatekeeper, or xDFG positions, reducing binding of first-generation inhibitors. Examples include NTRK1 G595R and NTRK3 G623R solvent-front mutations. Off-target resistance can reactivate MAPK signaling through alterations in KRAS, BRAF, MET, or other pathways.

At progression, a new tissue biopsy or plasma ctDNA test can identify the mechanism. Tissue also shows whether the tumor transformed histologically or whether a progressing lesion is unrelated. Plasma is less invasive and can sample several metastases, but some tumors shed little DNA.

Later-generation TRK inhibitors are designed to overcome selected kinase-domain resistance mutations. Access and approval differ by country, age, tumor type, and prior therapy. A resistance result should be matched to current clinical evidence or a trial rather than assumed to predict every newer drug.

Repeat testing is also reasonable when an earlier test was technically incomplete. A DNA-only panel from years ago may not have covered NTRK introns adequately. RNA testing on another block can reveal a fusion that was not truly absent.

Limitations and Questions to Ask

NTRK testing is vulnerable to both false negatives and false positives. DNA capture gaps, degraded RNA, low tumor percentage, physiologic TRK expression, unusual FISH patterns, and bioinformatic filtering can all affect the result. Confirmation is most important when the finding is unexpected or treatment-changing.

Ask:

• Is the finding a fusion, point mutation, or amplification?
• Which NTRK gene and partner are involved?
• Is the fusion in frame, expressed, and retaining the kinase domain?
• Was the result found by DNA NGS, RNA NGS, FISH, RT-PCR, or IHC?
• Does a screening result need orthogonal confirmation?
• Did the assay pass RNA quality controls and cover NTRK2 and NTRK3 adequately?
• Does the current drug label fit the patient’s age, disease stage, and prior treatment?
• Should tissue or plasma be retested at progression for resistance?
• Would the fusion change the pathologic diagnosis?

Keep the complete report, including fusion exons, partner, method, read support, and interpretation. This detail becomes essential when a later resistance mutation is found or when a clinical trial requires central confirmation.

A confirmation pathway for an unexpected result

When pan-TRK staining is positive in a common carcinoma, the pathologist first reviews the staining pattern and internal controls. Diffuse strong nuclear staining can support certain NTRK3 fusions, while diffuse cytoplasmic staining can occur with several partners. Weak focal staining in neural, smooth-muscle, or inflammatory elements may be nonspecific. A molecular test should target RNA when possible.

If DNA NGS reports a rearrangement, the reviewer checks whether both breakpoints are covered, the orientation can create a transcript, and the NTRK kinase domain is retained. An intronic event with an unknown partner may be a structural change without an expressed oncogenic product. RNA confirmation can distinguish a true fusion from a passenger rearrangement.

If RNA NGS is negative, quality metrics decide the next step. Failure of control transcripts means the assay was technically uninformative, not biologically negative. Another block, a fresh biopsy, or FISH may help. If RNA quality was strong and broad sequencing found no fusion, a weak screening stain is more likely false positive.

Planning treatment and follow-up

Before starting a TRK inhibitor, document baseline neurologic symptoms, liver tests, weight, concomitant medicines, and disease measurements. Dizziness, gait imbalance, cognitive changes, and sensory symptoms can overlap with brain metastases or prior therapy, so a baseline makes later attribution clearer. Drug interactions may alter exposure, particularly with medicines that affect CYP3A.

Imaging intervals follow tumor type and disease tempo. A dramatic early response can occur, but some lesions change density or become cystic before shrinking. In children and young adults with locally advanced tumors, surgeons and oncologists should agree in advance on when to reassess resectability and how much response is needed to reduce morbidity.

When treatment stops, some patients experience withdrawal pain because normal TRK signaling has been chronically inhibited. The care team may taper or manage symptoms rather than assuming sudden pain always means rapid cancer progression. New neurologic or hepatic symptoms still require prompt evaluation.

At progression, classify it as isolated, oligoprogressive, central nervous system, or widespread. Local therapy can sometimes control one resistant site while the TRK inhibitor continues, whereas widespread progression prompts molecular reassessment and a new systemic plan. Tissue and plasma can be complementary: tissue confirms histology and local resistance, while plasma can show multiple kinase-domain clones.

The full fusion report should be preserved for trials. Eligibility forms may require the exact partner, exons, assay accreditation, and proof that the event was present before TRK therapy. A one-line portal summary often lacks this information and can delay access.

Special considerations in children and rare tumors

In infants with a locally advanced spindle-cell tumor, rapid fusion confirmation can avoid highly morbid surgery or intensive chemotherapy. The pathology team may use morphology, pan-TRK staining, and an RNA fusion assay together. Because tissue is small, preplanning prevents repeated anesthesia and biopsy. Pediatric dosing is based on body surface area or weight according to the drug label, and growth, development, neurologic function, and long-term effects require follow-up.

Secretory breast carcinoma and secretory salivary carcinoma usually have indolent behavior when localized, so surgery remains central. The presence of ETV6::NTRK3 does not mean every early tumor needs a TRK inhibitor. Targeted therapy becomes most relevant when disease is unresectable, metastatic, recurrent, or surgery would cause severe morbidity.

In thyroid cancer, an NTRK fusion can be both diagnostic and therapeutic. Pathologists should distinguish a true fusion from NTRK expression in normal nerve tissue and should report coexisting mutations. Children with radiation exposure or fusion-positive papillary thyroid cancer may have extensive nodal disease but can still respond well to appropriate multidisciplinary care.

What a molecular tumor board reviews

A tumor board checks the structural plausibility of the fusion, reading frame, kinase-domain retention, expression evidence, tumor type, and absence of a stronger competing driver. It also reviews whether the assay is approved or validated for the proposed treatment and whether confirmatory testing is required by payer or trial.

For an unusual partner, the board may inspect raw read support and public databases. Recurrent partners with a known dimerization domain are easier to classify. A one-off event with few reads, an out-of-frame junction, or a breakpoint after the kinase domain is unlikely to be a functional driver.

Treatment planning includes brain imaging when clinically appropriate because some NTRK-positive cancers spread to the central nervous system and drug CNS penetration differs. Baseline electrocardiogram and medication review may be needed for entrectinib. Each inhibitor has a distinct safety profile despite sharing the TRK target.

When a patient has a rapid and deep response, treatment generally continues until progression or unacceptable toxicity under the label. Stopping because imaging looks normal can permit regrowth from microscopic disease. Any planned interruption for surgery, pregnancy, toxicity, or adherence should be coordinated with the oncology team.

A confirmed fusion can also settle difficult pathology terminology. For example, a salivary tumor with ETV6::NTRK3 is classified as secretory carcinoma rather than acinic cell carcinoma. Correct naming improves surgical planning, prognosis discussion, and family communication even when targeted therapy is not currently needed.

Not every NTRK rearrangement is an actionable fusion. The report should show that the event is in frame, retains the NTRK kinase domain, and is expressed when possible. Identifying both partners and the supporting reads helps distinguish a functional driver from an uncertain structural finding.

References

• NTRK gene fusion testing and management in lung cancer 2024 (Review)
• NTRK insights: best practices for pathologists 2022 (Review)
• Tissue-Agnostic Targeting of Neurotrophic Tyrosine Receptor Kinase Fusions: Current Approvals and Future Directions 2024 (Review)
• Larotrectinib Sulfate 2025 (Official Drug Information)
• Entrectinib 2025 (Official Drug Information)

Disclaimer

This article provides general education about NTRK fusion testing and does not confirm treatment eligibility. A molecular pathologist and oncology team must review the exact fusion, assay, tumor diagnosis, disease stage, current drug label, and alternatives. Never start or stop a TRK inhibitor based only on a screening stain or abbreviated portal result.