Interference of Tissue-Marking Dyes With Fluorescence In Situ Hybridization Assays

To the Editor.—Tissue-marking dyes (TMDs) are widely used in pathology for identifying tissue margins in surgical excisions and as an additional step to prevent tissue mix-up in biopsies. Although the College of American Pathologists (CAP) does not require TMDs to be used during grossing of biopsies, many laboratories use various colored inks on biopsy tissues including prostate and breast. There is limited literature evaluating TMD endurance and color fidelity in identifying margins and validating use of these inks with immunohistochemical stains.^1,2  However, TMDs have not been validated for fluorescence in situ hybridization (FISH) assays. The number of FISH tests we offer at our national reference laboratory has steadily increased over the years as more probes have become available, not only for diagnostic but also for predictive tests for targeted therapies. In our laboratory, we have occasionally seen orange ink used on breast biopsy tissues submitted for HER2 FISH to autofluoresce, interfering with red signal counting. More recently, the number of other tissue types, such as primary or metastatic lung cancers, that are inked with different colors and submitted for FISH testing has been increasing. We tested 10 colors (orange, yellow, green, black, gold, purple, blue, red, magenta, and teal) from Cancer Diagnostics Incorporated (Morrisville, North Carolina), 8 colors (orange, yellow, lime, red, purple, green, blue, and black) from Davidson Marketing System (Bloomington, Minnesota), and 9 colors (orange, yellow, gold, red, magenta, green, blue, teal, and black) from Royal Marker (Fort Lauderdale, Florida) to evaluate autofluorescence. The inks were used to mark all surfaces of excisional tissue samples. Autofluorescence was evaluated with an Olympus BX53 microscope using orange (excitation wavelength, 540/20 nm; emission wavelength, 575/40 nm), green (excitation wavelength, 475/30 nm; emission wavelength, 530/40 nm), Texas red (excitation wavelength, 560/40 nm; emission wavelength, 630/75 nm), and dual orange/green (excitation wavelengths, 470/30 and 575/20 nm; emission wavelengths, 530/30 and 610/20 nm) filters. The results are shown in the Table.

An example of metastatic lung adenocarcinoma to the sacrum that failed testing with multiple probes because of autofluorescence of orange ink is shown in Figure 1, A and B. We do not know which brand of ink was used by the referring laboratory, but in the image shown, orange autofluorescence interfered with the detection and enumeration of ALK signal.

Most FISH tests use locus-specific or centromeric probes that are labeled with orange or green fluorescent dyes. Orange, lime, and some yellow colored inks may show orange or green autofluorescence, interfering with interpretation of commonly used clinical FISH probes. Uninterpretable FISH tests may, in turn, lead to repeat testing attempts and/or repeat biopsies, leading to increased expense and/or patient morbidity.

We strongly recommend that CAP issue a warning and/or discourage inking of tissue biopsies with these colors unless validated for FISH.