Skip to content
Prof. Dr. Mustafa Özdoğan · Medical Oncologist

What Is Neurotrophic Tyrosine Receptor Kinase (NTRK) Fusion? Should Immunohistochemistry or NGS Be Performed?

What Is Neurotrophic Tyrosine Receptor Kinase (NTRK) Fusion? Should Immunohistochemistry or NGS Be Performed?
In this article

Looking at the Tumor's Genes, Not Its Organ: NTRK Fusion and Choosing the Right Test

An NTRK fusion is a rare genetic change in which two genes are joined at the wrong place, and it can be treated with targeted drugs regardless of the organ in which the tumor arose. It occurs in fewer than one in a hundred common cancers; yet when it is found, TRK inhibitors markedly shrink the tumor in about 60 percent of patients. The challenge is catching this rare change with the right test: pan-TRK immunohistochemistry (IHC) is a fast and inexpensive screening method, whereas next-generation sequencing (NGS) shows the fusion directly. In this guide, we explain what an NTRK fusion is, in which tumors it is sought, the strengths and weaknesses of IHC and NGS, the testing algorithm recommended by ESMO and how treatment is planned if a fusion is found. We covered the basics of genetic testing in our article on next-generation sequencing (NGS) and the concept of tissue-agnostic treatment in our article on tissue-agnostic cancer treatments.

0.3%frequency of NTRK fusion across all adult and pediatric solid tumors
>90%frequency in rare tumors such as secretory breast carcinoma and infantile fibrosarcoma
60%response rate to larotrectinib (339-patient pooled analysis)
43.3 monthsmedian duration of response to larotrectinib
Quick Answer · Physician Summary
  • What is an NTRK fusion? It is the joining of the NTRK1, NTRK2 or NTRK3 gene with another gene. The resulting fusion protein stays permanently switched on and sends the cell a constant growth signal.
  • How common is it? About 3 in 1,000 across all solid tumors; but more than 90% in rare tumors such as secretory breast carcinoma, secretory carcinoma of the salivary gland and infantile fibrosarcoma.
  • IHC or NGS? For a treatment decision, the fusion should be demonstrated by NGS, preferably RNA-based. Pan-TRK IHC can be used for screening when access to NGS or tissue is limited; a positive result is confirmed by NGS, while a negative result does not exclude fusions, particularly NTRK3 fusions.
  • If a fusion is found? Larotrectinib, entrectinib and repotrectinib have tissue-agnostic FDA approval. With larotrectinib, the objective response rate is 60% and the median duration of response is 43.3 months; in children, the response rate is 84%.

Why does this matter?

For many years, cancer treatment was planned according to the organ in which the tumor arose: one drug for lung cancer, another for colon cancer. NTRK fusion is one of the first examples that changed this approach. In 2018, larotrectinib became the first targeted drug approved solely on the basis of a genetic change, regardless of the tumor's organ. That is why knowing whether a patient carries an NTRK fusion can change treatment options across very different tumors, from lung to thyroid and from colon to sarcoma (the importance of genetic alterations on the ESCAT scale).

However, NTRK fusion is very rare in common cancers; thousands of patients must be tested to find a few. Moreover, not every test catches this fusion equally well: some methods miss a portion of fusions, while others can give a positive result even without a fusion. That is why patients and physicians who see "pan-TRK positive" on a pathology report or "no NTRK fusion detected" on a genetic report need to know what the result does and does not show. For other biomarkers on pathology reports, see our articles on the PD-L1 test and HER2 concepts.

Is it a first? A short history of TRK
1982–1986
The first fusion, found in colon cancer
Mariano Barbacid's team showed that, in a colon cancer sample, the tropomyosin gene had joined an unknown kinase gene. The name "tropomyosin receptor kinase" (TRK) comes from this first fusion.
1986–1991
The link to nerve growth factor
Rita Levi-Montalcini and Stanley Cohen, who discovered nerve growth factor (NGF), received the Nobel Prize in 1986; in 1991, TRK was found to be the receptor for this factor.
1998
The secret of a tumor in infants
The ETV6-NTRK3 fusion was identified in infantile (congenital) fibrosarcoma; the same fusion was later also found in secretory breast carcinoma and salivary gland tumors.
November 2018
The first tissue-agnostic targeted drug
Larotrectinib received accelerated FDA approval for all NTRK fusion-positive solid tumors; entrectinib followed in 2019 and repotrectinib in 2024.
April 2025
Full approval
Based on a 339-patient pooled analysis, larotrectinib's approval was converted to full approval. In December 2025, the next-generation drug zurletrectinib was approved in China.

What is NTRK fusion?

An NTRK fusion occurs when the NTRK1, NTRK2 or NTRK3 gene, which encodes the TRK receptors, breaks and joins another gene. The resulting fusion protein stays permanently switched on without waiting for a signal and sends the cell a constant signal to divide. It is rare, but when found, it can be treated with targeted drugs called TRK inhibitors.

What do TRK receptors normally do?

TRK receptors (TRKA, TRKB and TRKC) are found mainly in the nervous system. They switch on briefly when signaling molecules such as nerve growth factor bind to them; they contribute to the development and survival of nerve cells, to the perception of pain and balance, and to the regulation of appetite and body weight. The characteristic side effects of TRK inhibitors, such as dizziness, weight gain and pain that appears when the drug is stopped, also stem from these normal functions.

In a fusion, the kinase portion of the NTRK gene (the "motor" that transmits the signal) joins the front portion of another gene. The front portion of this fusion partner gene usually makes proteins pair up; as a result, the TRK kinase stays switched on spontaneously and permanently, without any outside signal. More than 80 different fusion partner genes that join NTRK genes have been identified to date; the best known are ETV6-NTRK3, TPM3-NTRK1 and LMNA-NTRK1. Because the fusion alone can drive tumor growth, drugs that block this signal often produce strong and durable responses.

Figure 1 · The normal TRK receptor and the TRK fusion protein

From a receptor that waits for a signal to a "motor" that is always on

The normal TRK receptor and the TRK fusion proteinextracellularintracellularNGFkinasenucleusNormal TRK receptorswitches on only with a signalno ligand, no extracellular partpartner geneTRK kinasenucleusTRK fusion proteinalways on, without a signalAt the gene level: two genes join at the wrong placeETV6 gene (chromosome 12)NTRK3 gene (chromosome 15)ETV6-NTRK3 fusion gene: a TRK kinase that is always on
How to read: On the left, the normal TRK receptor sits in the cell membrane; when nerve growth factor (NGF, yellow) binds, its kinase portion switches on briefly. On the right is the TRK fusion protein: the extracellular portion has been lost; the piece from the fusion partner gene (orange) is joined to the TRK kinase (blue), and the proteins pair up and stay permanently on without a signal. The diagram at the bottom shows the gene level: the ETV6 gene on chromosome 12 and the NTRK3 gene on chromosome 15 join at their breakpoints (red lines) to form the ETV6-NTRK3 fusion gene.

In which tumors does NTRK fusion occur?

NTRK fusions occur in two very different groups. The first group comprises certain rare tumors: secretory breast carcinoma, mammary analogue secretory carcinoma of the salivary gland, infantile fibrosarcoma of early childhood and cellular congenital mesoblastic nephroma (a type of kidney tumor in infants). More than 90% of these tumors carry a fusion; indeed, the fusion helps establish the diagnosis itself. The second group includes common cancers such as lung, colon, breast, thyroid and head and neck cancer, melanoma, sarcomas and brain tumors; in these, the fusion is mostly found in less than 1% (types of carcinoma).

How often does NTRK fusion occur, and in which tumors?

Proportion within each tumor type, %. Values reported as "over 90%" in rare tumors are shown as 90.

Secretory breast carcinoma
90%
Mammary analogue secretory carcinoma of the salivary gland
90%
Infantile fibrosarcoma
90%
Cellular congenital mesoblastic nephroma
90%
Lung, colon, breast, head and neck, melanoma
5%
All adult solid tumors
0.31%
All pediatric solid tumors
0.34%

In common cancers, NTRK fusion occurs in less than 5%, usually less than 1%. For this reason, in common cancers the fusion is not tested on its own; it is sought as part of a broad genetic panel. Source: ESMO NTRK guidance and OncologyPRO review.

Among common cancers, subgroups in which NTRK fusion is somewhat more frequent have also been described. In colon cancer, fusions are concentrated particularly in tumors with high microsatellite instability (MSI-H) and normal RAS and BRAF genes (colon cancer, the MSI-H concept). In thyroid cancer, they have been reported more often, especially in younger patients and those exposed to radiation. In lung cancer, they are sought in tumors lacking other driver alterations such as EGFR, ALK, ROS1 or KRAS (lung cancer). The fusion rate is also higher in pediatric tumors and high-grade brain tumors than in adult carcinomas (brain tumors).

Group Example tumors NTRK fusion frequency Testing approach
Enriched (rare) tumors Secretory breast carcinoma, secretory carcinoma of the salivary gland, infantile fibrosarcoma, congenital mesoblastic nephroma Over 90% Confirmation: FISH, RT-PCR or RNA panel
Common cancers Lung, colon, breast, head and neck, melanoma, thyroid, sarcoma, brain tumors Mostly below 1%; all below 5% Screening: broad NGS panel (preferably including RNA) or pre-screening with IHC
All solid tumors Adults and children 0.31% in adults, 0.34% in children As part of a genetic panel in advanced disease

Which tests are used to look for NTRK fusion?

There are four main ways to demonstrate an NTRK fusion, and each looks at a different level. Pan-TRK immunohistochemistry shows the TRK protein with a brown stain on a tissue section. FISH (fluorescence in situ hybridization) looks under the microscope for a break in a single NTRK gene. DNA-based NGS reads hundreds of genes at once. RNA-based NGS, in turn, reads the messenger RNA transcribed from genes and shows the fusion junction directly (guide to RNA sequencing). All of these tests can be performed on tumor tissue obtained by biopsy or surgery and stored in paraffin blocks.

Figure 2 · Four testing methods

Looking at the protein, gene or RNA level

Four methods that show an NTRK fusioneach has different strengths and weaknesses Pan-TRK IHCstains the protein;fast, cheap, low tissue;does not prove fusion FISHlooks for a gene break;one gene, one test;partner not shown DNA-based NGSmany genes together;can miss fusionsin long introns RNA-based NGSreads the fusiontranscript directly;most reliable method Bottom line: IHC screens, NGS decidestreatment requires the fusion to be shown by NGS
How to read: Pan-TRK IHC stains the protein; it is fast and inexpensive and needs little tissue, but it does not prove a fusion. FISH looks for a break in a single gene; it does not show the fusion partner gene. DNA-based NGS examines many genes together, but it can miss the fusion when the breakpoint lies in long introns. RNA-based NGS reads the fusion transcript directly and is the most reliable method. In short, IHC screens; NGS decides.
Method What does it look at? Strength Weakness
Pan-TRK IHC TRK protein Results in 1–2 days, inexpensive, available in every pathology laboratory, needs little tissue Does not prove a fusion; low sensitivity for NTRK3; false positives in tumors with neural or smooth muscle features
FISH A break in a single gene Suitable for confirmation in enriched tumors A separate test for each NTRK gene; does not show the fusion partner gene; can miss some fusions
DNA-based NGS The DNA of hundreds of genes Also looks at other alterations such as mutations, TMB and MSI Can miss fusions in the long introns of NTRK2 and NTRK3
RNA-based NGS Fusion transcript (mRNA) Shows the junction and the fusion partner gene directly; proves that the fusion is active Depends on RNA quality; may fail in old or poorly preserved tissue; more expensive

Pan-TRK immunohistochemistry: strengths and weaknesses

The antibody used in pan-TRK IHC recognizes the kinase region shared by all three TRK proteins. Because very little TRK protein is normally present in adult tissues, distinct brown staining in tumor cells suggests a fusion protein. Where in the cell the staining appears can also offer a clue: staining may be seen in the nucleus with ETV6-NTRK3 fusions, in the cytoplasm with some NTRK1 fusions and at the nuclear membrane with the LMNA-NTRK1 fusion.

Figure 3 · Examples of pan-TRK staining

Negative, cytoplasmic and nuclear staining

Pan-TRK immunohistochemistry: what is seen?Negativeno stainingCytoplasmice.g. NTRK1 fusionsNucleare.g. ETV6-NTRK3Brown staining shows the TRK proteinit does not prove a fusion; confirm positives by NGSstaining may be weak or absent in NTRK3 fusionsneural and smooth muscle tissue can stain without fusionthe staining location can vary with the fusion partner
How to read: In the negative example on the left, the cells show no brown staining; the nuclei appear blue from the counterstain. In the middle example, the cytoplasm is brown (for example, some NTRK1 fusions). In the example on the right, the nuclei are brown (for example, ETV6-NTRK3). Brown staining shows the presence of TRK protein but does not prove a fusion; a positive result should be confirmed by NGS. The drawing is simplified.

The largest validation study of IHC to date came from Memorial Sloan Kettering Cancer Center in the United States. Among the 33,997 patients examined in this study, an NTRK fusion was found in 87 (about 2.6 per 1,000). With RNA-based NGS as the reference, the overall sensitivity of pan-TRK IHC was 87.9% and its specificity was 81.1%. In other words, IHC caught about 88 of every 100 fusion-positive tumors, while giving a false-positive result in about 19 of every 100 tumors without a fusion.

How accurate are the tests? MSK analysis of 33,997 cases

Sensitivity (detecting the fusion) and specificity (testing negative when there is no fusion), %, with RNA-based NGS as the reference.

Pan-TRK IHC sensitivity · NTRK1
96%
Pan-TRK IHC sensitivity · NTRK2
100%
Pan-TRK IHC sensitivity · NTRK3
79%
Pan-TRK IHC specificity · overall
81.1%
DNA-based NGS sensitivity
81.1%
DNA-based NGS specificity
99.9%

Immunohistochemistry can miss about one in five NTRK3 fusions; tumors with neural or smooth muscle features can stain even without a fusion. DNA-based NGS can miss a fusion when the breakpoint lies in long introns not covered by the panel. Source: Solomon et al., Modern Pathology 2020.

Behind these overall figures lie two important details. First, while IHC almost always caught NTRK1 and NTRK2 fusions, it missed about one in five NTRK3 fusions (sensitivity 79%). The ETV6-NTRK3 fusion in infantile fibrosarcoma and secretory carcinomas falls squarely in this group; therefore, a negative IHC does not exclude a fusion in these tumors. Second, in tumors with neural or smooth muscle features (such as some sarcomas and brain tumors), normal TRK protein can stain even without a fusion; in these tumors, IHC is not a reliable screening method.

DNA or RNA? Why can't NGS catch every fusion?

Our genes consist of segments called exons, which carry the recipe for the protein, and the intervening regions between them called introns. In fusions, the break usually occurs within introns. In the NTRK2 and NTRK3 genes, some introns are tens of thousands of letters long; for economic reasons, most DNA-based panels do not read the whole of these long introns. If the break falls in an unread region, the fusion does not appear on the DNA test. In the same MSK study, the sensitivity of DNA-based NGS compared with the RNA test was 81.1%; that is, about one in five fusions was missed by the DNA panel.

With RNA-based NGS, the situation is different: when the cell produces messenger RNA, it already removes the introns. As a result, the junction between the fusion partner gene and the NTRK gene sits side by side in the fusion RNA and is read directly. The RNA test also shows that the fusion is actually "working," meaning that RNA destined to be translated into protein is being produced. Its weakness is that RNA is much more fragile than DNA; the test may fail in samples that are old, have been left in formalin for a long time or contain very few tumor cells.

Figure 4 · The difference between DNA-based and RNA-based NGS

When the breakpoint hides in an intron

DNA or RNA? Where does the fusion hide?DNA-based NGS: not the whole gene is read✗long intron (tens of thousands of letters)regions read by the panelbreakpointif the break is in an unread part, the fusion is missedRNA-based NGS: the junction is read directlyfusion mRNA✓green: reads spanning the junctionRNA-based NGS is more sensitive for fusionsRNA can degrade in old or poorly preserved tissue
How to read: Top, DNA-based NGS: the panel reads the exons and short regions of the NTRK gene (green strips), but most of the long intron is not read. If the breakpoint (red line) lies in the unread region, the fusion is missed. Bottom, RNA-based NGS: in the fusion mRNA, from which the introns have been removed, the fusion partner gene (orange) and the NTRK exons (blue) sit side by side; reads spanning the junction (green) show the fusion directly.

Liquid biopsy (ctDNA) tests, which look at tumor DNA circulating in the blood, can also look for NTRK fusions; however, because they are DNA-based and the amount of tumor DNA in the blood can be low, they are more likely to miss fusions. Finding a fusion on a liquid biopsy is meaningful; not finding one does not exclude a fusion and should, where possible, be followed by tissue testing.

Immunohistochemistry or NGS: which should be done?

The short answer: for a treatment decision, the fusion must be demonstrated by NGS, preferably with a panel that includes RNA; IHC is a screening tool along the way, not a decision tool. ESMO's 2019 recommendations base the choice on tumor type. In rare tumors where the fusion is very common, the goal is confirmation, and FISH, RT-PCR or targeted RNA panels are sufficient. In common cancers where the fusion is rare, the goal is screening; the best option is a broad NGS panel that reliably detects NTRK fusions, and RNA-based sequencing is preferred if RNA quality allows.

The place of pan-TRK IHC in this algorithm is as a pre-screen when NGS cannot be done promptly, tissue is scarce or cost is decisive: if IHC is negative and the tumor has no neural or smooth muscle features, a fusion is unlikely; if IHC is positive, the result must be confirmed by NGS. Today, because many alterations such as EGFR, ALK, KRAS, BRAF, HER2, MSI and TMB need to be assessed together in many advanced cancers, a broad NGS panel that includes NTRK is often the route that makes the best use of both tissue and time.

Figure 5 · Decision pathway for NTRK testing

Simplified algorithm based on the 2019 ESMO recommendations

Advanced solid tumor: NTRK testing neededfirst consider tumor type and the remaining tissue Enriched tumor typesecretory carcinoma,infantile fibrosarcomaconfirm with FISH,RT-PCR or RNA panel Common cancerlung, colon, breast,thyroid, sarcoma, etc.broad NGS panel,preferably with RNA NGS not possible or tissue scarce?screening with pan-TRK IHC is an optiona positive result must be confirmed by NGS DNA-NGS negative but suspicion high?re-examine with RNA-based NGS Fusion confirmed: TRK inhibitorlarotrectinib, entrectinib or repotrectinib
How to read: In an advanced solid tumor, the tumor type and the remaining tissue are considered first. In enriched tumors where the fusion is very common (secretory carcinomas, infantile fibrosarcoma), confirmation is done with FISH, RT-PCR or an RNA panel. In common cancers, a broad NGS panel, preferably including RNA, is used. If NGS cannot be done promptly or tissue is scarce, screening can be done with pan-TRK IHC; a positive result is confirmed by NGS. If DNA-based NGS is negative but suspicion remains high, the tumor is re-examined with RNA-based NGS. If the fusion is confirmed, a TRK inhibitor is considered.
Reading the pathology and genetic report
  • "Pan-TRK positive" suggests that a fusion may be present, but on its own it is not sufficient for treatment; NGS confirmation is required.
  • "Pan-TRK negative" lowers the likelihood of a fusion but does not fully exclude NTRK3 fusions or tumors with neural/smooth muscle features.
  • "No NTRK fusion detected (DNA panel)": check the report to see whether the panel covers the NTRK introns or RNA.
  • A fusion report usually also states the fusion partner gene and the exons involved (for example, ETV6 exon 5 - NTRK3 exon 15). A preserved kinase domain indicates that the fusion is likely to respond to the drug.

Who should be tested for NTRK fusion?

Because the drugs approved for NTRK fusion are currently used in solid tumors that are advanced (metastatic) or whose surgical removal would cause serious problems, testing is mainly recommended for these patients. Since it is impossible to know in advance in which tumor a fusion will turn up, it is recommended that NTRK be tested as part of a genetic panel in patients with advanced solid tumors. In some situations, however, the likelihood of a fusion is clearly higher and testing becomes a priority (what is metastasis?).

Figure 6 · Who should be tested for NTRK fusion?

Priority groups and situations that call for attention

Who should be tested for NTRK fusion?recommended for every advanced solid tumor Prioritysecretory carcinomas,infantile fibrosarcoma,pediatric tumors Also consideryounger patients;tumors with no otherdriver alteration Colon cancermore common in MSI-Htumors with normalRAS and BRAF Thyroid, sarcomabrain tumors andlung cancer: partof a broad panel Not routinely recommended in early stagedrugs approved in advanced or hard-to-operate disease
How to read: NTRK fusion testing should be considered in every advanced solid tumor. Secretory carcinomas, infantile fibrosarcoma and pediatric tumors are priority groups. The likelihood of a fusion increases in younger patients and in tumors without another known driver alteration. In colon cancer, fusions are more common in MSI-H tumors with normal RAS and BRAF. In thyroid cancer, sarcoma, brain tumors and lung cancer, NTRK is tested as part of a broad panel. Routine testing is not recommended in early-stage disease.

If a fusion is found: TRK inhibitors

The drugs used in patients with a confirmed NTRK fusion are oral targeted drugs that switch off the fusion protein by fitting into the signaling pocket of the TRK kinase (how to take oral targeted pills). All three are FDA-approved regardless of the tumor's organ. The approvals are based on single-arm "basket" trials; that is, patients with different tumor types were enrolled in the same trial, and no comparison group was used. Because the fusion is so rare, a randomized trial is almost impossible to conduct.

Response rates to TRK inhibitors in NTRK fusion-positive tumors

Proportion of patients whose tumor shrank markedly (objective response), %. Single-arm trials.

Larotrectinib · adults and children, 339 patients
60%
Larotrectinib · children only, 93 patients
84%
Entrectinib · adults, 121 patients
61.2%
Repotrectinib · TRK drug-naive, 40 patients
58%
Repotrectinib · prior TRK drug, 48 patients
50%

With larotrectinib, the complete response rate (tumor disappearing on imaging) is 24% and the median duration of response is 43.3 months. With entrectinib, the median duration of response is 20 months and the intracranial response rate in brain metastases is 63.6% (11 patients). None of the trials had a comparison group. Comparisons across different studies are indirect; patient populations, prior treatments and follow-up durations differ.

Larotrectinib (Vitrakvi) is the first drug to target TRK alone. It received accelerated approval in November 2018 and full approval in April 2025. In the 339-patient analysis underlying the full approval, the objective response rate was 60%, the complete response rate (the tumor disappearing on imaging) was 24% and the median duration of response was 43.3 months. In a 93-patient analysis in children, the response rate reached 84%; the 3-year overall survival rate was 93%. Because it is available as capsules and as a liquid, it can also be used in infants.

Entrectinib (Rozlytrek) targets ROS1 and ALK in addition to TRK, and it crosses into the brain well. It was approved in August 2019. In an updated analysis of 121 adults, the response rate was 61.2%, the median duration of response was 20 months and the median progression-free survival was 13.8 months. In 11 patients with measurable brain metastases, the intracranial response rate was 63.6% (ALK-targeted treatments).

Repotrectinib (Augtyro) is a next-generation drug designed against resistance mutations. It received accelerated approval in June 2024. The response rate was 58% in 40 patients who had not previously received a TRK drug and 50% in 48 patients who had previously received a TRK drug and whose disease had progressed; in this second group, the median duration of response was 9.9 months. It is therefore an important option, particularly for patients whose disease progresses after first-generation drugs.

Drug FDA approval Targets Key feature
Larotrectinib (Vitrakvi) November 2018 (accelerated), April 2025 (full) TRK only Longest follow-up data; high response in children; liquid formulation available
Entrectinib (Rozlytrek) August 2019 TRK, ROS1, ALK Efficacy data in brain metastases
Repotrectinib (Augtyro) June 2024 (accelerated) TRK, ROS1 Efficacy against resistance mutations
Zurletrectinib China, December 2025; no FDA approval TRK (next generation) High response rate reported in a Chinese trial; no US or European data

The side effects of TRK inhibitors are mostly mild or moderate, but some are distinctive because they relate to TRK's normal functions in the nervous system: dizziness and balance problems, weight gain, altered taste, numbness and tingling, muscle aches, constipation and elevated liver enzymes. Bone fractures have also been reported with larotrectinib. Because some patients may experience temporary pain when the drug is stopped, treatment should not be discontinued without consulting the physician.

What happens if resistance develops?

Responses to TRK inhibitors often last a long time, but resistance can develop over time. The most common cause of resistance is new mutations that alter the pocket in which the drug sits: the NTRK1 G595R or NTRK3 G623R mutations in the "solvent front" region, and mutations in the "gatekeeper" and xDFG regions. Repotrectinib and other next-generation drugs in development may be effective against a substantial share of these mutations. Less often, the tumor bypasses TRK and begins using another signaling pathway, such as KRAS, BRAF or MET; in this "off-target" resistance, a new TRK drug alone may not be sufficient (how do cancer cells become resistant to treatment?).

Figure 7 · TRK inhibitors and resistance

The drug fits the pocket, a mutation narrows it, a next-generation drug fits again

How TRK inhibitors work and how resistance develops✗✓TRK kinaseTRK kinaseTRK kinase1. drug fits pocketsignal stops2. mutation arisespocket narrows3. next-gen drugsmaller molecule1st generationmutation (e.g. G595R)repotrectinibOff-target resistance: another pathway opensKRAS, BRAF or MET alterationsa new TRK drug alone may then not be enoughOn progression, resistance is re-examinednew biopsy or a blood test for tumor DNA
How to read: (1) A first-generation TRK inhibitor (green) fits into the kinase pocket and stops the signal. (2) When a resistance mutation (red; for example, G595R) develops in the pocket, the drug can no longer fit. (3) Next-generation, smaller-molecule drugs such as repotrectinib (purple) can fit into the mutated pocket again. In off-target resistance, by contrast, another pathway opens through KRAS, BRAF or MET alterations. When the disease progresses, the cause of resistance is investigated with a new biopsy or a blood test for tumor DNA.

The situation in Türkiye

Access to testing and drugs in Türkiye
  • Pan-TRK immunohistochemistry and NGS panels that include the NTRK genes are available at many university hospitals and private pathology and genetics laboratories in Türkiye. Access to RNA-based fusion panels varies from center to center; before testing, ask the laboratory whether the panel includes RNA.
  • The conditions under which NGS tests are covered by SGK (the Turkish Social Security Institution) vary by test type and cancer type; current conditions should be confirmed with the hospital and laboratory.
  • The licensing and reimbursement status of larotrectinib and entrectinib in Türkiye could not be verified at the time of writing. If a drug is not licensed or the indication is not reimbursed, an application can be made through the import of medicines from abroad with a physician's report (cancer drugs imported from abroad).
  • If a fusion result is unclear or difficult to interpret, review by a molecular tumor board and, if needed, a second opinion are recommended.

Questions to ask your doctor

  • Has my tumor been tested for NTRK fusion? If so, by which method (IHC, DNA-NGS, RNA-NGS)?
  • Does the NGS panel used include RNA, or does it cover the NTRK introns?
  • If my pan-TRK result is positive, has it been confirmed by NGS? If it is negative, does this result completely exclude a fusion?
  • Is there enough tissue for testing, or is a new biopsy needed? Is a liquid biopsy an option?
  • If a fusion is found, how can I access a TRK inhibitor? Could I take part in a clinical trial?
  • If the disease progresses during treatment, will the cause of resistance be investigated again?
Common misconceptions
  • "If pan-TRK is positive, the drug is started right away." No. IHC does not prove a fusion; the fusion must be confirmed by NGS before treatment.
  • "If IHC is negative, there is no fusion." Not always. About one in five NTRK3 fusions may not show up on IHC.
  • "A genetic panel was done, so NTRK has been checked." Not every panel detects NTRK fusions equally well; DNA panels can miss some fusions. Check the report to confirm that NTRK and RNA were covered.
  • "An NTRK mutation and an NTRK fusion are the same thing." No. TRK inhibitors are effective in fusions; they are not approved for point mutations or copy number gains in the NTRK gene.
  • "A tissue-agnostic drug works in every cancer." No. The drug is independent of the organ but dependent on the fusion; no effect is expected in a tumor without the fusion.

Frequently Asked Questions

What is NTRK fusion?
An NTRK fusion occurs when the NTRK1, NTRK2 or NTRK3 gene breaks and joins another gene. The resulting fusion protein stays permanently switched on and sends the cell a constant growth signal. It is found in about 3 in 1,000 of all solid tumors, but in more than 90% of certain rare tumors.
Should immunohistochemistry or NGS be used to test for NTRK fusion?
For a treatment decision, the fusion must be demonstrated by NGS, preferably with a panel that includes RNA. Pan-TRK immunohistochemistry can be used for screening when access to NGS or tissue is limited; a positive result is confirmed by NGS, while a negative result does not fully exclude fusions, particularly NTRK3 fusions.
Which drugs are used if an NTRK fusion is found?
Larotrectinib, entrectinib and repotrectinib are TRK inhibitors with FDA approval regardless of the tumor's organ of origin. With larotrectinib, the objective response rate is 60% and the median duration of response is 43.3 months; repotrectinib is also effective in patients who developed resistance to a previous TRK drug.
Who should be tested for NTRK fusion?
In patients with advanced or metastatic solid tumors, testing for NTRK as part of a genetic panel is recommended. Secretory carcinomas, infantile fibrosarcoma, pediatric tumors, younger patients and tumors without another known driver alteration are priority groups.

Glossary

Gene fusion
The breaking and joining of two different genes to form a new, abnormal gene.
TRK receptor
A protein encoded by the NTRK genes that transmits signals such as nerve growth factor into the cell.
Kinase
An enzyme that can be switched on and off and transmits signals by adding phosphate to other proteins.
Pan-TRK immunohistochemistry
Staining of a tissue section with an antibody that recognizes all three TRK proteins at once.
NGS (next-generation sequencing)
A genetic testing method that reads many genes at the same time.
Exon and intron
The segments of a gene that carry the protein recipe (exons) and the intervening regions between them (introns).
FISH
A test that uses fluorescently labeled probes to show a break or rearrangement in a gene under the microscope.
Sensitivity and specificity
The proportion of people with a condition whom a test correctly detects (sensitivity) and of people without it whom the test correctly finds negative (specificity).
Tissue-agnostic approval
Approval of a drug based on the genetic alteration a tumor carries rather than the organ in which it arose (also called tumor-agnostic approval).
Solvent front mutation
A type of mutation that prevents the drug from fitting into the kinase pocket and causes resistance to TRK inhibitors.

DROZDOGAN Academy Commentary

NTRK fusion is the best example of an alteration that is "rare but must not be missed." For the patient who carries the fusion, the drug often provides a rapid and durable response; the real challenge is finding that patient.

⚡ Why it matters

Most of these tumors respond only modestly to standard treatments; when a fusion is found, however, responses lasting years can be achieved with an oral drug. In infants and children, disabling surgery can be avoided.

✅ Strengths

Response rates are high regardless of tumor type, and responses are durable. Larotrectinib received full approval based on 339-patient data; next-generation drugs are available for use when resistance develops. ESMO has a clear testing algorithm.

⚠️ Limitations

The data come from single-arm basket trials; there are no comparative data against standard treatment. IHC and DNA-based NGS can miss some fusions; access to RNA-based tests is not equal everywhere. The drugs are expensive, and access varies from country to country.

🩺 In practice

For our patients with advanced solid tumors, we order a broad NGS panel that includes NTRK and add RNA-based fusion analysis whenever possible. We use pan-TRK IHC as a pre-screen when tissue is scarce or NGS will be delayed, and we always confirm a positive result with NGS. If the DNA panel is negative but suspicion remains high, we go back to RNA testing.

❓ Open questions

Should TRK inhibitors be used first-line or after standard treatment? What will be the role of preoperative use in early-stage disease? How much can next-generation drugs delay resistance? How should weight gain and bone health be managed during long-term use?

Bottom line: With NTRK fusion, the right question is not "IHC or NGS?" but "how do we find the fusion without missing it, and confirm it?" The answer is a testing pathway chosen according to tumor type and confirmed by NGS, preferably at the RNA level, before any treatment decision.

Transparency note
  • This article was prepared in October 2026.
  • The testing approach is based on the 2019 ESMO recommendations and the 33,997-case MSK validation study. Treatment data are taken from FDA approval announcements, peer-reviewed publications and conference presentations; zurletrectinib data are based on the manufacturer's press release.
  • The licensing and reimbursement status of the drugs in Türkiye could not be verified; the information reflects the date of writing.
  • Response rates from different studies are shown in the same chart; because patient populations differ, this comparison is indirect. Illustrations are simplified and not to scale. The Academy commentary is our own.
  • This article is not medical advice; make your treatment decisions together with your physician.

References

  1. Marchiò C, Scaltriti M, Ladanyi M, et al. ESMO recommendations on the standard methods to detect NTRK fusions in daily practice and clinical research. Ann Oncol 2019;30:1417-1427. synapse.mskcc.org
  2. Solomon JP, Linkov I, Rosado A, et al. NTRK fusion detection across multiple assays and 33,997 cases: diagnostic implications and pitfalls. Mod Pathol 2020;33:38-46. doi:10.1038/s41379-019-0324-7. synapse.mskcc.org
  3. ESMO OncologyPRO. Overview of NTRK gene fusion in specific tumours. oncologypro.esmo.org
  4. ESMO OncologyPRO. Algorithm for NTRK gene fusion testing. oncologypro.esmo.org
  5. Targeted Oncology. FDA awards full approval to larotrectinib in NTRK+ solid tumors. April 10, 2025. targetedonc.com
  6. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of larotrectinib in TRK fusion-positive cancers in adults and children. N Engl J Med 2018;378:731-739. nejm.org
  7. OncLive. Larotrectinib shows continued tumor-agnostic efficacy in pediatric TRK fusion cancer (ASCO 2022). onclive.com
  8. Demetri GD, De Braud F, Drilon A, et al. Updated integrated analysis of the efficacy and safety of entrectinib in patients with NTRK fusion-positive solid tumors. Clin Cancer Res 2022;28:1302-1312. experts.umn.edu
  9. The ASCO Post. FDA grants accelerated approval to repotrectinib for NTRK fusion-positive solid tumors. June 13, 2024. ascopost.com
  10. InnoCare. InnoCare announces approval of the next-generation TRK inhibitor zurletrectinib in China. December 11, 2025. biospace.com
  11. Martin-Zanca D, Hughes SH, Barbacid M. A human oncogene formed by the fusion of truncated tropomyosin and protein tyrosine kinase sequences. Nature 1986;319:743-748. pubmed.ncbi.nlm.nih.gov
  12. Knezevich SR, McFadden DE, Tao W, Lim JF, Sorensen PH. A novel ETV6-NTRK3 gene fusion in congenital fibrosarcoma. Nat Genet 1998;18:184-187. pubmed.ncbi.nlm.nih.gov
DROZDOGAN AI

Explore this article together

Ask about the article and read answers alongside published sources from our site.

DROZDOGAN AI

Drozdogan Assistant

What Is Neurotrophic Tyrosine Receptor Kinase (NTRK) Fusion? Should Immunohistochemistry or NGS Be Performed?

Ask about the article and read answers alongside published sources from our site.

Conversation and privacy information

Drozdogan Assistant answers this conversation. Authorized site administrators may review it for quality. Do not share personal information.

Replies use OpenAI infrastructure. Questions and replies are processed by that service; conversation records are stored on our site. Do not share identifying, contact or personal health information.

0 / 1000

For information only; no personal diagnosis or treatment advice.

Related articles

Explore other articles on this topic.

Stay informed

Receive new articles and scientific updates by email.

Google Analytics runs on public pages independently of this window and may use analytics cookies. Your choice here controls external services only.

Cookie policy