Nanostring-based screening for tyrosine kinase fusions in inflammatory myofibroblastic tumors

Abstract

Gene expression imbalances were measured for tyrosine kinase (TK) genes using Nanostring in 19 samples of inflammatory myofibroblastic tumor (IMT). All cases were immunohistochemically stained with anaplastic lymphoma kinase (ALK) and pan-tropomyosin-related-kinase (pan-Trk) antibodies. Five cases with imbalanced ALK expression, reported with Nanostring, were tested using fluorescence in situ hybridization (FISH); two cases with imbalanced neurotrophic tyrosine receptor kinase 3 (NTRK3) expression were tested using reverse transcription-polymerase chain reaction (RT-PCR). One case with imbalanced expression for ROS proto-oncogene 1 (ROS1) was tested using RNA sequencing and RT-PCR. TK fusions were detected in all cases with imbalanced TK expression. RNA sequencing detected a FN1–ROS1 fusion gene in an adult IMT case. IMT with ALK rearrangement showed myofibroblast-dominant features. IMT with ETV6–NTRK3 fusion showed prominent lymphoplasmacytic infiltration with scattered myofibroblasts. Pan-Trk IHC revealed only scattered positively stained cells in IMT with ETV6–NTRK3 fusion gene. ROS1-positive IMT showed myofibroblast-dominant features.

Uterine inflammatory myofibroblastic tumor: First report of a ROS1 fusion

“…Patients are typically young, with a median age of 12 (mean 21; range 1‐77) years, and only 8 (22%) are 30 years or older. Histologically, they resemble other IMTs; however, over 20% (n = 8) are reported as being at least focally collagenized (hyalinized pattern), 3,13,15,17,19,24 which is a very rare morphologic feature in uterine IMTs, and only very focally present in the current case. Details regarding myogenic marker expression are limited in previous reports of ROS1‐ rearranged IMTs, but desmin, smooth muscle actin, and/or caldesmon appear to show variable expression 2,15,19,22,24,25 .…”

“…Only four extrauterine IMTs (two lung, two head and neck) have been reported with a FN1‐ROS1 fusion. Three involved exon 32 of ROS1 , 13,22,24 while the location of the fourth is unknown 21 . In contrast, our IMT involved exon 34, but is predicted to behave similarly to those involving exon 32 as both the ROS1 kinase and transmembrane domains are retained at the C‐terminus of the fusion protein.…”

“…Although there appears to be significant pathological overlap between this NTRK ‐rearranged IMT and NTRK ‐rearranged uterine spindle cell neoplasm, the latter generally lacks plasma cells, has at most a focal myxoid matrix, may be positive for S100 and CD34, and typically is ER and PR negative 11 . Furthermore, ETV6 has not yet been identified as a partner gene in NTRK‐ rearranged uterine spindle cell neoplasms, but has been reported in rare extrauterine IMTs 12,13 . Nonetheless, due to the rarity of spindle cell lesions with NTRK rearrangements, studies are warranted to further determine whether they represent a spectrum of the same disease vs distinct entities.…”

“…Outside of the gynecologic tract, ROS1 ‐rearranged IMTs account for approximately 10% 1,2,14 of tumors, with 37 cases reported to date 1‐3,9,13‐24 . Most occur in the thoracic cavity (n = 14), gastrointestinal tract/liver (n = 10), or soft tissues (n = 10), with very rare examples reported in the head and neck (n = 2) or placental membranes (n = 1).…”

NTRK fusion in Japanese colorectal adenocarcinomas

“…Nanostring assay. Nanostring (NanoString Technologies, Inc., Seattle, Washington, USA) assays were performed as previously described 17,18 , to confirm that positive staining by pan-TRK immunohistochemistry led to unbalanced expression of NTRK gene exons, as expected of gene fusions. The custom-designed probe set contained 90 tyrosine kinase genes.…”