NTRK Fusions in Cancer

by Jason Wasserman MD PhD FRCPC
March 21, 2026

NTRK fusions are chromosomal rearrangements that fuse one of three genes — NTRK1, NTRK2, or NTRK3 — to a partner gene, producing an abnormal fusion protein that continuously drives cancer cell growth. The NTRK genes normally encode a family of receptor proteins called tropomyosin receptor kinases (TRK A, B, and C), which play important roles in the development and maintenance of the nervous system. When these genes are rearranged in cancer cells, the resulting TRK fusion proteins are permanently switched on, transmitting unregulated growth signals. What makes NTRK fusions particularly important is not how common they are — they occur in only 1–3% of most common cancers — but rather how effectively they can be treated: drugs called TRK inhibitors achieve remarkably high response rates in NTRK fusion-positive tumours regardless of where in the body the cancer originated. This tumour-agnostic activity — the same drug working across many different cancer types based on a shared molecular alteration rather than the cancer’s tissue of origin — represents one of the most significant conceptual advances in modern oncology.

What the test looks for

The human genome contains three NTRK genes, each encoding a different TRK receptor protein:

NTRK1 encodes TRKA, which is normally activated by nerve growth factor (NGF).
NTRK2 encodes TRKB, normally activated by brain-derived neurotrophic factor (BDNF).
NTRK3 encodes TRKC, normally activated by neurotrophin-3 (NT-3).

In cancer, chromosomal rearrangements fuse any one of these three genes to a partner gene, creating a fusion protein in which the kinase signalling domain of TRK is constitutively active — continuously transmitting growth and survival signals regardless of whether the cell needs to divide. More than 80 different fusion partners have been identified across the three NTRK genes, spanning a wide variety of cancer types. Despite this enormous diversity of fusion partners and cancer types, all NTRK fusions share the same fundamental mechanism — unregulated TRK kinase activation — and all predict sensitivity to TRK inhibitor drugs.

The frequency of NTRK fusions varies dramatically by cancer type:

Very high frequency (greater than 50–90%): Infantile fibrosarcoma (NTRK3), secretory breast carcinoma (NTRK3), secretory carcinoma of the salivary gland (NTRK3), and congenital mesoblastic nephroma (NTRK3) — a group of rare tumours where NTRK fusions are essentially defining molecular events.
Intermediate frequency (5–25%): Certain thyroid cancers, spitzoid melanomas, gastrointestinal stromal tumours (GISTs) without KIT/PDGFRA mutations, and some brain tumours, including gliomas.
Low frequency (1–3%): Non-small cell lung cancer, colorectal cancer, pancreatic cancer, and many other common solid tumours. Though rare in any individual cancer type, these common cancers collectively account for the largest absolute number of NTRK fusion-positive patients.

NTRK fusions by tumour type: diagnostic and treatment implications

Because NTRK fusions occur across such a wide range of cancers — and because their significance varies enormously from one tumour type to another — it is worth understanding what a positive result means in specific clinical settings. In some cancers, an NTRK fusion is essentially a defining feature of the diagnosis itself. In others, it is an incidental but highly actionable finding discovered during routine molecular profiling. The treatment implications are the same regardless — TRK inhibitor therapy — but the diagnostic weight of the finding, and the urgency with which testing should be performed, differ substantially by tumour type.

Infantile fibrosarcoma

Infantile fibrosarcoma is a rare soft tissue tumour that occurs almost exclusively in infants and very young children, typically presenting as a rapidly growing mass in the limbs or trunk. It is one of the most NTRK fusion-rich tumours known: approximately 90% of cases harbour an ETV6-NTRK3 fusion, making this fusion essentially diagnostic of the condition when found in the appropriate clinical and pathological context. Before TRK inhibitors became available, the standard treatment for infantile fibrosarcoma was aggressive chemotherapy — often followed by surgery, sometimes requiring amputation of a limb. TRK inhibitors have transformed this picture. Larotrectinib and entrectinib achieve response rates exceeding 90% in infantile fibrosarcoma, with many tumours shrinking dramatically or disappearing entirely before surgery. In some patients, surgery can be avoided altogether, or a far less extensive operation can be performed than would otherwise have been necessary. NTRK testing is now considered mandatory at diagnosis for infantile fibrosarcoma, and TRK inhibitor therapy may be used to shrink the tumour before surgery (neoadjuvant therapy) or to treat disease that cannot be completely removed. The identification of an ETV6-NTRK3 fusion in a soft tissue tumour in a young child is therefore not merely a treatment-selection finding — it is a transformative diagnostic event that can change the entire clinical trajectory.

Secretory carcinoma of the breast

Secretory carcinoma of the breast is a rare subtype of breast cancer distinguished by its characteristic microscopic appearance — cells producing a secretory material — and by its very high prevalence of ETV6-NTRK3 fusions, present in approximately 90% of cases. Like its name-twin in the salivary gland, secretory breast carcinoma is defined in part by this fusion, and detecting it helps confirm the diagnosis when the histological appearance is suggestive. Secretory breast carcinoma tends to have a more favourable prognosis than most breast cancers when localised, but when it presents in advanced or metastatic form, TRK inhibitor therapy offers a highly effective option. Response rates to larotrectinib in secretory breast carcinoma have been very high in reported series, reflecting the fact that the ETV6-NTRK3 fusion is a dominant driver in this tumour type. Patients with a diagnosis of secretory breast carcinoma who have advanced disease should be tested for NTRK fusions — and specifically for ETV6-NTRK3 — if this has not already been done.

Secretory carcinoma of the salivary gland

Secretory carcinoma of the salivary gland (also called mammary analogue secretory carcinoma, or MASC) is a tumour of the salivary glands that closely resembles secretory breast carcinoma in its microscopic features and molecular profile. Approximately 90% of cases harbour an ETV6-NTRK3 fusion, and detecting this fusion is now a required part of the diagnostic workup for salivary gland tumours with the appropriate morphology — it helps distinguish secretory carcinoma from other salivary gland tumours that can look similar under the microscope. In advanced or metastatic secretory carcinoma of the salivary gland, TRK inhibitor therapy has shown high response rates, including complete responses in some patients. Given the rarity of this tumour and the limited efficacy of conventional chemotherapy in the metastatic setting, the identification of an ETV6-NTRK3 fusion and the availability of TRK inhibitors represent a major advance for patients with this diagnosis.

Congenital mesoblastic nephroma

Congenital mesoblastic nephroma is a rare kidney tumour that occurs primarily in newborns and very young infants. The cellular variant — the subtype at highest risk of recurrence — harbours an ETV6-NTRK3 fusion in the majority of cases. The fusion helps confirm the diagnosis and identify patients at higher risk. For localised disease, surgery alone is usually curative. For the rare cases that are not completely resectable or that recur, TRK inhibitor therapy offers a targeted option in place of more aggressive chemotherapy — an important consideration given the very young age of these patients and the potential long-term effects of cytotoxic treatment in infants.

Thyroid cancer

NTRK fusions — involving all three NTRK genes — occur in approximately 5–25% of papillary thyroid cancers, with the frequency varying considerably by patient age and geography. NTRK fusions are enriched in papillary thyroid cancers arising in children and young adults, particularly those with a history of radiation exposure (such as survivors of the Chornobyl disaster), where NTRK1 and NTRK3 fusions have been found at rates above 20%. In adults with sporadic papillary thyroid cancer, NTRK fusions are less common but still clinically important when present. In radioiodine-refractory or metastatic thyroid cancer with an NTRK fusion, TRK inhibitor therapy offers a meaningful treatment option with high response rates. It is now incorporated into thyroid cancer treatment guidelines for this subset of patients.

Gliomas and central nervous system tumours

NTRK fusions — particularly involving NTRK2 and NTRK3 — occur in a subset of gliomas, including low-grade gliomas in children and some high-grade gliomas in adults. Paediatric low-grade gliomas are a biologically diverse group in which NTRK fusions represent one of several actionable driver alterations. In children with NTRK fusion-positive brain tumours that have recurred or progressed after prior treatment, TRK inhibitors have shown meaningful activity. Both larotrectinib and entrectinib penetrate the blood-brain barrier to a clinically relevant degree, and intracranial responses have been documented. The identification of an NTRK fusion in a brain tumour is therefore not merely a research finding — it is an actionable result that should prompt discussion of TRK inhibitor therapy, particularly when other treatment options have been exhausted.

Gastrointestinal stromal tumours (GISTs)

Most GISTs are driven by mutations in KIT or PDGFRA and are treated with imatinib and related drugs. However, a subset of GISTs — particularly those arising in the stomach of children and young adults, and those occurring in the context of certain syndromes — lack KIT and PDGFRA mutations and are instead driven by NTRK fusions (most commonly NTRK3) or other alterations. These wild-type GISTs are resistant to imatinib but may respond to TRK inhibitors. Testing for NTRK fusions in GISTs lacking KIT and PDGFRA mutations is recommended, particularly in paediatric patients and in young adults, where this subset is proportionally more common.

Common solid tumours: lung, colorectal, and pancreatic cancer

In non-small cell lung cancer, colorectal cancer, and pancreatic cancer, NTRK fusions occur in approximately 1–3% of cases. Although individually rare, these common cancers collectively account for a large number of affected patients in absolute terms. In these settings, an NTRK fusion is typically an incidental finding on comprehensive molecular profiling rather than a diagnostically expected alteration. However, the treatment implications are identical to any other NTRK fusion-positive cancer: TRK inhibitor therapy is highly effective, with response rates above 70% seen across these tumour types in the clinical trials supporting larotrectinib and entrectinib approval. For patients with advanced lung, colorectal, or pancreatic cancer in whom an NTRK fusion is identified, TRK inhibitor therapy represents one of the most effective targeted treatment options available for any cancer at that stage.

Why is the test done

To determine eligibility for TRK inhibitor therapy. Larotrectinib (Vitrakvi) and entrectinib (Rozlytrek) are both approved on a tumour-agnostic basis for solid tumours with NTRK fusions — meaning they are approved for any cancer type, regardless of where the cancer started, as long as an NTRK fusion is present. This was a landmark regulatory decision, reflecting the principle that the molecular target rather than the tissue of origin should sometimes determine treatment. Both drugs achieve high response rates — consistently above 70% in clinical trials — across a wide range of NTRK fusion-positive tumours.
To avoid ineffective treatment. In cancers where an NTRK fusion is the primary driver, targeted TRK inhibition is substantially more effective than standard chemotherapy. Identifying the fusion ensures these patients are not undertreated.
To characterise an unexpected or rare tumour. In tumour types where NTRK fusions are very common — such as infantile fibrosarcoma or secretory breast carcinoma — NTRK testing is essential for diagnosis and directly determines treatment. In these rare settings, detecting an NTRK fusion can transform a diagnosis from one requiring aggressive chemotherapy to one that responds to a well-tolerated oral targeted drug.
To support clinical trial eligibility. New TRK inhibitors designed to overcome resistance to first-generation drugs are in clinical development, and knowing NTRK fusion status is necessary for trial participation.

Who should be tested

Guidelines on NTRK fusion testing have evolved as tumour-agnostic drug approvals have expanded. Testing recommendations depend on the cancer type:

Universal testing recommended: All patients with advanced or metastatic non-small cell lung cancer, colorectal cancer, and several other common advanced solid tumours should have NTRK fusion testing as part of comprehensive molecular profiling at diagnosis. Major guidelines, including those from ESMO and ASCO, recommend broad NTRK testing in advanced solid tumours, particularly when comprehensive NGS is being performed.
Reflex testing in specific rare tumour types: For tumours with high NTRK fusion prevalence — including infantile fibrosarcoma, secretory carcinoma of the breast or salivary gland, and congenital mesoblastic nephroma — NTRK testing (specifically for NTRK3 fusions) is recommended as a routine part of the diagnostic workup regardless of stage, because the fusion may be diagnostic and because treatment implications are immediate.
Selective testing in intermediate-prevalence tumours: For thyroid cancers, brain tumours, GISTs, and spitzoid melanomas, NTRK testing is recommended in appropriate clinical contexts.

In practice, NTRK fusion testing is most efficiently performed as part of a comprehensive NGS panel that simultaneously assesses all relevant driver genes. Standalone NTRK testing is less efficient and increasingly uncommon at major cancer centres.

How the test is performed

NTRK fusion testing presents a particular technical challenge because of the large number of possible fusion partners — over 80 across the three genes. No single testing method detects all possible fusions with equal sensitivity, and the choice of method has important implications for the reliability of a negative result.

Next-generation sequencing (NGS)

Comprehensive next-generation sequencing (NGS) is the preferred testing approach, particularly RNA-based NGS, which directly sequences the messenger RNA produced by tumour cells and can detect any expressed fusion transcript regardless of the partner gene. RNA-based NGS is the most sensitive and comprehensive method for NTRK fusion detection and is the approach most likely to identify rare or novel fusions. DNA-based NGS panels can also detect NTRK rearrangements by identifying structural variants at the genomic level. Still, they may miss some fusions — particularly those involving complex rearrangements or intragenic deletions — and RNA-based testing is generally preferred when available.

Fluorescence in situ hybridization (FISH)

FISH can detect NTRK rearrangements by identifying the separation of probes flanking each NTRK gene. Separate FISH tests are required for NTRK1, NTRK2, and NTRK3, making comprehensive FISH testing resource-intensive. FISH cannot identify the fusion partner and may have limited sensitivity for certain rearrangement types. It is rarely used as a primary testing method for NTRK fusions at centres with NGS capability.

Immunohistochemistry (IHC)

Immunohistochemistry using pan-TRK antibodies (which detect all three TRK proteins simultaneously) has emerged as a useful and widely available screening tool for NTRK fusions. In tumour types with high NTRK fusion prevalence — particularly infantile fibrosarcoma and secretory carcinomas — strong, diffuse pan-TRK IHC staining is highly predictive of an underlying NTRK fusion. However, IHC has important limitations: some NTRK fusions produce weakly positive or negative IHC staining, while some non-fusion alterations can produce positive staining. For this reason, a positive IHC result should ideally be confirmed with a molecular method (NGS or FISH) before treatment decisions are made, and a negative IHC result does not definitively exclude an NTRK fusion in all contexts.

Liquid biopsy

Cell-free circulating tumour DNA testing can detect NTRK fusions, though — as with other structural rearrangements — sensitivity is lower than for tissue-based methods. Liquid biopsy is most useful when tissue is unavailable or insufficient, or for monitoring during disease progression. A negative liquid biopsy does not exclude an NTRK fusion, and tissue testing should be performed when clinically indicated.

How results are reported

NTRK fusion results are reported as positive (fusion detected) or negative (no fusion detected), with specification of which gene is involved (NTRK1, NTRK2, or NTRK3), the fusion partner, and the specific breakpoint or exon junction where identifiable. A typical positive NGS report might read: “ETV6-NTRK3 fusion detected” or “TPM3-NTRK1 fusion, exon 7-11 confirmed.”

IHC reports will note the staining pattern — cytoplasmic, nuclear, or membranous — and the intensity and extent of staining, using a validated scoring system. Strong diffuse cytoplasmic staining with a pan-TRK antibody is the pattern most strongly associated with an underlying fusion.

Some NGS reports will identify a structural variant involving an NTRK gene with an uncharacterised partner. These results should be discussed with a medical oncologist and, where relevant, a molecular tumour board, as confirmatory testing or functional assessment may be warranted to determine clinical significance.

What the result means

NTRK fusion positive (any of NTRK1, NTRK2, or NTRK3). An NTRK fusion gene is present in the cancer cells. The tumour is driven at least in part by the constitutively active TRK fusion protein and is expected to respond to TRK inhibitor therapy. Both larotrectinib (Vitrakvi) and entrectinib (Rozlytrek) are approved on a tumour-agnostic basis for this finding, meaning approval covers any solid tumour type with an NTRK fusion. In pooled clinical trial data, larotrectinib achieved an overall response rate of approximately 75% across a broad range of tumour types, with durable responses — many lasting well over a year — and a favourable tolerability profile. Response rates are even higher in tumour types where NTRK fusions are the dominant driver: in infantile fibrosarcoma and secretory carcinomas, response rates above 90% have been reported, and complete responses — in which no cancer is detectable on imaging — are observed in a meaningful proportion of patients. In common solid tumours such as lung, colorectal, and pancreatic cancer, response rates above 70% have been observed, which compares very favourably with chemotherapy in these settings. Entrectinib achieved comparable systemic response rates and additionally demonstrated CNS activity, making it a preferred option when brain metastases are present or at risk. Both drugs are taken as daily oral capsules. Your oncologist will discuss which drug is more appropriate given your specific cancer type, CNS involvement, prior treatments, and other individual factors.
NTRK fusion negative. No NTRK fusion was detected in the regions assessed by the test. TRK inhibitor therapy is not indicated based on this result. It is important to understand that the reliability of a negative result depends on the testing method used: RNA-based NGS provides the most comprehensive coverage and the most reassuring negative result, while IHC and DNA-based methods may miss some fusions. If a negative result was obtained by IHC alone or by a limited molecular panel, and clinical suspicion for an NTRK fusion remains high — for example, in a tumour type with high NTRK prevalence — discussion with a pathologist about confirmatory testing may be warranted.
Equivocal IHC result requiring molecular confirmation. Weakly positive or heterogeneous pan-TRK IHC staining may occur in tumours with or without NTRK fusions. In this setting, molecular confirmation with NGS or FISH is recommended before treatment decisions are made. An equivocal IHC result is a prompt for further testing, not a basis for initiating or withholding TRK inhibitor therapy.

TRK inhibitors and resistance

First-generation TRK inhibitors — larotrectinib and entrectinib — achieve high response rates but, as with other targeted therapies, resistance eventually develops in most patients. Resistance mechanisms include:

On-target resistance: Mutations within the TRK kinase domain — particularly at the solvent front (G595R for TRKA, G623R for TRKC) and gatekeeper positions — that impair drug binding while preserving the fusion’s kinase activity. These are detected by repeated molecular testing at progression.
Off-target resistance: Activation of bypass signalling pathways (such as KRAS mutations, MET amplification, or other alterations) that allow the cancer to continue growing independently of TRK signalling.

Second-generation TRK inhibitors — including selitrectinib and repotrectinib — are specifically designed to overcome on-target resistance mutations and are approved or in advanced clinical development. When an NTRK fusion-positive cancer progresses on a first-generation TRK inhibitor, repeat molecular testing with both liquid biopsy and tissue biopsy is recommended to identify the resistance mechanism and guide the choice of subsequent therapy.

NTRK fusions and brain metastases

The propensity for brain metastases varies considerably across the tumour types in which NTRK fusions occur. In lung cancer specifically — where NTRK fusions occur alongside other driver genes that are associated with CNS spread — brain involvement can occur. Entrectinib has demonstrated intracranial activity and may be preferred over larotrectinib when brain metastases are present or at high risk. Repotrectinib, a second-generation TRK/ROS1 inhibitor, also demonstrates CNS penetration. Your oncologist will assess CNS status at diagnosis and recommend brain imaging where appropriate.

NTRK fusions: germline vs. somatic

NTRK fusions found in solid tumours are almost always somatic — they arise within the cancer cells during the patient’s lifetime and are not inherited. Germline NTRK alterations are associated with rare congenital conditions and are distinct from somatic fusions found in cancer. Patients with a somatic NTRK fusion in their tumour do not need to worry about passing it to their children, and family members do not require NTRK screening on this basis.

What happens next

If an NTRK fusion is found: Your oncologist will discuss TRK inhibitor therapy — larotrectinib or entrectinib — as a treatment option. The choice between them will depend on your cancer type, the presence or risk of brain metastases, other molecular findings, and local drug availability. Both are oral daily medications. Your oncologist will review the expected side effects — which are generally manageable — and the monitoring schedule. If your cancer type is one where NTRK fusions are diagnostically significant (such as infantile fibrosarcoma or secretory carcinoma), the presence of the fusion also confirms the diagnosis. It may directly influence whether chemotherapy is needed at all.
If no NTRK fusion is found, TRK inhibitor therapy is not indicated. Other molecular findings, PD-L1 expression, and cancer type will guide treatment.
If the result was obtained by IHC alone and is equivocal, Molecular confirmation with NGS should be arranged before treatment decisions are made.
At progression on a first-generation TRK inhibitor, repeat molecular testing is strongly recommended to identify the resistance mechanism. Second-generation TRK inhibitors are available or accessible through clinical trials and may achieve responses despite resistance to larotrectinib or entrectinib.

Questions to ask your doctor

Has my tumour been tested for NTRK fusions, and what method was used — NGS, FISH, or IHC?
If an NTRK fusion was found, which gene is involved — NTRK1, NTRK2, or NTRK3 — and what is the fusion partner?
Which TRK inhibitor — larotrectinib or entrectinib — is recommended for me, and why?
Do I have brain metastases, and how does that affect the choice of drug?
If my NTRK test result was negative by IHC or a limited panel, should I have confirmatory RNA-based NGS testing?
What side effects should I expect from TRK inhibitor therapy, and how are they managed?
If my cancer progresses on a TRK inhibitor, will re-testing be done, and is a second-generation TRK inhibitor available?
Are there clinical trials studying new TRK-targeted treatments that I might be eligible for?