Abstract
Sectioning a whole tissue microarrray (TMA block) and storing the sections maximizes the number of sections obtained, but may impair the antigenicity of the stored sections. We have investigated the impact of TMA section storage on antigenicity. First, we reexamined existing TMA data to determine whether antigenicity in stored sections changes over time. Component scores for each marker, based on cellular compartment of staining and score-type, were evaluated separately. Residual components scores adjusted for grade, tumor size, and node positivity, were regressed on the number of days storage to evaluate the effect of storage time. Storage time ranged from 2 to 1897 days, and the mean change in antigenicity per year ranged from -0.88 (95% confidence interval, -1.11 to -0.65) to 0.035 (95% confidence interval, 0.016-0.054). Further analysis showed no significant improvement in the fit of survival models if storage time adjusted scores were included in the models rather than unadjusted scores. We then compared 3 ways of processing TMA sections after cutting-immediate staining, staining after 1 year, and staining after 1 year coated in wax-on the immunohistochemistry results for: progesterone receptor, a routinely used, robust antibody, and MKI67, which is generally considered less robust. The progesterone receptor scores for stored sections were similar to those for unstored sections, whereas the MKI67 scores for stored sections were substantially different to those for unstored sections. Wax coating made little difference to the results. Biomarker antigenicity shows a small decline over time that is unlikely to have an important effect on studies of prognostic biomarkers.
Highlights
Tissue microarrays (TMAs) constructed using archival, formalin-fixed, paraffin-embedded pathology material, are a standard tool for investigating tumor biomarkers in large-scale clinical epidemiological studies
All but 2 of the marker components showed a decline in antigenicity over time and for 40 of them the decline was significant at a nominal P < 0.05
For fresh compared with dipped sections, weighted kappa values of 0.89, 0.80, 0.64, and 0.44 were obtained for progesterone receptor (PGR) proportion, PGR intensity, MKI67 proportion, and MKI67 intensity, respectively
Summary
Tissue microarrays (TMAs) constructed using archival, formalin-fixed, paraffin-embedded pathology material, are a standard tool for investigating tumor biomarkers in large-scale clinical epidemiological studies. The reliability of TMAs in such studies has been investigated primarily by focussing on the number of cores needed from each case to produce results equivalent to those from whole tissue sections.[1,2,3] for biomarkers based on immunohistochemistry (IHC) the quality of staining will depend on a wide range of factors These include preanalytical variables such as the handling and ischemic time of the fresh tumor sample at the time of surgery, the length of fixation and method of tissue processing used, the duration of storage of the paraffin blocks, and the environment in which they are stored, and the methods for TMA construction, the storage conditions for the TMA, the methods for processing TMA sections; and analytical variables within the protocol for the IHC such as antigen retrieval and staining times. The efficiency of this approach needs to be balanced against the potential for loss of antigenicity over time due to oxidation of the cut sections
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
