Detecting Gene Rearrangements in Patient Populations

Detecting Gene Rearrangements in Patient Populations Through a 2-Step Diagnostic Test Comprised of Rapid IHC Enrichment Followed by Sensitive Next-Generation Sequencing (NGS).

Murphy DA, Ely HA, Shoemaker R, Boomer A, Culver BP, Hoskins I, Haimes JD, Walters RD, Fernandez D, Stahl JA, Lee J, Kim KM, Lamoureux J, Christiansen J. Appl Immunohistochem Mol Morphol. 2016 Mar 29. Pubmed

  • This paper describes a novel approach to detect NTRK fusions, along with ROS1 and ALK.
  • The method is part of an ongoing NTRK fusion clinical trial.
  • The method uses IHC for screening followed by mRNA NGS.
  • This two step process determines if a gene fusion might be driving a particular cancer.
  • The name of this method is Trailblaze Pharos has recently recently received an investigational device exemption (IDE) from the FDA.

Step 1  Immunohistochemistry (IHC) detects a receptor tyrosine kinase where it does not belong.

Step 2 RNA based next generation sequencing (NGS) identifies kinase fusion genes.

Find a kinase where it does not belong

Finding a protein, particularly a cell growth controlling kinase, that is not normally expressed in a tissue is a good sign that something is amiss.  In fusion genes, kinase domains are not only fused to protein coding regions of partner genes, but also the other gene’s promoter.

Some proteins are more abundant than others

The following is an example of why IHC can be used to detect kinase fusion gene products.  Tropomyosin 3, coded for the by the TPM3 gene, is a common housekeeping gene often found fused to ROS1, ALK, and Trk kinases.  Many of these kinases are not expressed in most tissues of the body (

  • The kinase TrkC is expressed at 1 ppm in the brain, but not found in the oral epithelium.
  • Tryopomyosin 3 is expressed at 1000 ppm in the brain and oral epithelium.

If TrkC is expressed like tropomyosin 3, it should be evident with IHC

Estimated proetin expression of cancer genes ALK, ROS1, TrkB and TrkC in various human cells and tissues. Tropomyosin 3 protein expression is shown as control.

Expression of four kinases capable of driving cancers (left) and a typical N-terminal fusion partner (right).


Immunohistochemistry can reduce the cost of identifying fusion proteins by NGS

If a tissue section tests positive with pan-RTK antibodies, it is then screened with antibodies against ALK, ROS1, and pan-Trk.  In the example of a colorectal adenocarcinoma patient (Murphy, 2016) , her tumor only had abnormal Trk expression.  RNA was isolated from this sample for NGS with primers specific Trk fusion proteins.  She had a TPM3-NTRK1 fusion.  (Trk is the kinase, NTRK is the gene.)

Immunohisto chemistry slide showing prescreen of tissue samples using ALK, ROS1 and panTrk antibodies.

Work flow for IHC pre-screening for NGS to identify fusion kinases.

Next Generation to identify kinase fusion proteins

The first challenge is ensuring that input mRNA is of high quality.   The second challenge is ensuring reproducible results.

Pre-sequencing mRNA quality control :  Ct of a housekeeping gene

The Ct (cycle threshold) is the number of cycles required for the fluorescent signal to cross the threshold of background fluorescence in real time PCR. Ct levels are inversely proportional to the amount of target nucleic acid in the sample. Contamination will also increase the number of cycles required for fluorescence to rise above background.

Extraction of RNA from FFPE samples might result in shearing.  Murphy and coworkers found that Ct of a housekeeping gene could predict samples rendered useless for NGS via too much shearing in the extraction process.

Murphy and coworkers also examined some very old blocks of FFPE preserved tissue.  Generally, FFPE samples used for clinical use are less than a year old.  The inverse correlation between Ct and age indicated that Ct is a good indicator of age dependent changes.

Finally, they examined the interplay between quality and quantity of input DNA and came up with a value of 200 ng.  It may be possible to use smaller amounts of input material from recent clinical testing specimens.   For retrospective studies using older archival tissues, it may be better advised to start with a higher initial test input of 200 ng of RNA.

Accuracy and Precision of the NGS

  • A total of 66 FFPE tumor samples as well as FFPE preserved cell lines with established fusion mutations were used.
  • The testing cohort included at least one known positive sample for each target receptor (NTRK1, NTRK2, NTRK3, ROS1, and ALK).
  • Sanger sequencing confirmed all of the 15 samples found to be positive by NGS sequencing, resulting in 100% positive agreement across all specimens.
  • Samples were run by two different operators on two different days and assessed for gene rearrangements in each replicate; results were 100% precise.

Future prospects:

Improvements over FISH

Fluorescence in situ hybridization (FISH) only detects very specific breaks in DNA sequences.  The red fluorescence probe close to a green fluorescence probe equals yellow fluorescence says little.

Fluorescence in situ hybridization Picture showing NTRK1 gene breakage.

The colorectal adenocarcinoma

This image was taken from Murphy, 2016.  The thin yellow arrow points to what appears to be a yellow dot indicating an intact NTRK1 gene.  The thin white arrow points to a chromosome break of the NTRK1 gene.

  • As you will learn from reading this website, some cancer driving fusion mutations occur in the same region of a chromosome. It is conceivable to have a situation like the hollow thick arrow in which the red and green fluorescent probes are close enough together to yield yellow.
  • Just because a chromosome fusion has occurred does not mean that the reading frame is going to result in a cancer driving fusion protein when translated by the ribosome.
  • As you will learn from reading this website, there are multiple possible breakpoints in the kinase genes that recombine to form fusion genes.

Cost savings

  1. In identifying the fusion.
  2. Even in a very valuable tumor sample, pan-RTK only takes one section.
  3. A few sections for pan-Trk, ALK, and ROS1 IHC and NGS should be set aside.
  4. In the probable event that the pan-RTK is negative, plenty of tissue will be left over for other purposes.
  5. If pan-RTK is positive, one may proceed with the second panel of antibodies and then NGS.

In understanding the cancer biology of kinase fusion proteins

Murphy and coworkers emphasized the cost savings and efficiency of their NGS.  Some mention was made of their ability of their approach to detect fusion mutations in very old FFPE preserved tumors.  For now it is generally assumed that all kinase fusion proteins can drive cancer as long as the kinase domain is in the fusion protein. On this website we are exploring this hypothesis.  Thus far, every kinase domain fusion protein we’ve examined is linked to an oligomerization region of a house keeping gene.  Some of these fusion partners even function as adaptor proteins for downstream targets.  Are these downstream targets phosphorylated on residues recognized by Trk, ALK, and/or ROS1?    Phosphorylation of these downstream targets can be detected with IHC as well as proteomics techniques.


NTRK, ALK, and ROS1 fusions, while comparatively rare, are drivers of many cancers.  The efforts of Murphy and coworkers have made these fusions much easier, inexpensive, and accurate to detect.  There is an open NTRK fusion clinical trial that is actively enrolling.  Eligibility is based on any solid tumor with NTRK, ALK, or ROS1 fusions (STARTRK-2).   More information is available on the NTRK trial website.


Murphy DA, Ely HA, Shoemaker R, Boomer A, Culver BP, Hoskins I, Haimes JD, Walters RD, Fernandez D, Stahl JA, Lee J, Kim KM, Lamoureux J, Christiansen J.(2016)Detecting Gene Rearrangements in Patient Populations Through a 2-Step Diagnostic Test Comprised of Rapid IHC Enrichment Followed by Sensitive Next-Generation Sequencing. Appl Immunohistochem Mol Morphol. 2016 Mar 29. [Epub ahead of print]  Click Murphy 2016 for a free version of this paper.

Leave a Reply