Sorting Out Mix-Ups:<i>The Provenance of Tissue Sections May Be Confirmed by PCR Using Microsatellite Markers</i>

Standard identification systems usually ensure that biopsy material is correctly associated with a given patient. Sometimes, as when a tumor is unexpectedly found, the provenance (proof of origin) of a tissue sample may be questioned; the tissue may have been mislabelled or contaminated with tissue from another patient. Techniques used to confirm tissue provenance include comparing either tissue markers of gender or ABO blood groups; however, these methods have weak confirmatory power. Recently, the use of DNA-based polymerase chain reaction (PCR) techniques has been reported. Paired, formalin-fixed, paraffin-embedded, 10 microns tissue sections were selected from 17 patients, 8 of whom had carcinoma, either by dividing a biopsy section, using sequential biopsies, or sequential biopsy and autopsy tissue. The resulting 36 samples were coded before analysis. In two additional cases, 1-mm fragments of tumor from one patient were included in the tissue block of benign tissue from another patient, the tumor fragments were identified on hematoxylin-and-eosin-stained sections, separately scraped off the glass slide, and analyzed. Tissue from two clinical cases, one of suspected mislabelling and one with a suspected carry-over of malignant tissue were also investigated. Short tandem repeat sequences (STR) or microsatellites, are 2-5 base pair repeats that vary in their repeat number between individuals. This variation (polymorphism) can be assessed using a PCR. A panel of markers of 3 STRs; ACPP, INT 2, and CYP 19 (on chromosomes 3, 11, and 15, respectively) were used. DNA was isolated from the samples after xylene deparaffinization and proteinase digestion, and was then amplified in a radioactive PCR using primers selected to give a product size ranging from 136-178 bases. Amplified products were electrophoresed on denaturing polyacrylamide gels, dried, and autoradiographed. DNA segments were successfully extracted from all samples but one, which was fixed in Bouin's fluid. By comparing allele sizes from the panel, all tissue pairs (other than the Bouin's pair) were successfully matched, the 1-mm tumor fragments were correctly assigned, and the two clinical problems were solved. STRs are highly informative and robust markers, well suited to PCR of small portions of tissue sections, and are an effective method to confirm the provenance of benign and malignant biopsy and autopsy material.