Pan keratin

IHC was performed using the Ventana Benchmark XT automated system (Ventana Medical System). Antibodies against the following antigens were used: TP53 (Clone DO-7, N1581, Dako; ready to use), vimentin (Ventana 790-2917; dilution 1:100), and pan-keratin (Ventana 760-2595; dilution 1:100). Aggresome-positivity was considered only for cells exhibiting juxtanuclear staining of vimentin.

Immunohistochemistry was performed on a BenchMark autostainer (Ventana Medical Systems, Tucson, Arizona, USA) according to the manufacturer's protocol. The following monoclonal antibodies were used in the study: MT1-MMP (catalytic domain, clone 114-6G6), 1 : 100, mouse (from Merck Millipore, Darmstadt, Germany); Pan-Keratin (clone AE1/AE3/PCK26), mouse; MLH1 (clone G168–728), mouse; MSH2 (clone G219–1129), mouse; MSH6 (clone 44), mouse; PMS-2 (clone EPR3947), rabbit (prediluted; obtained from Ventana Medical Systems, Tucson, Arizona, USA). For MT1-MMP, cases with weak to strong cytoplasmic and/or membranous immunostaining at the leading edge of the tumor were classified as MT1-MMP positive (Figure 1). Loss of MMR protein expression required lack of nuclear immunostaining for MLH1, MSH2, MSH6, or PMS-2 in tumor cells with retained positivity in nonneoplastic epithelium, stromal and immune cells [27 (link)].

Tissues were fixed in 10% neutral-buffered formalin for 24 h, paraffin-embedded and sections were cut at 4 μm. Sections were deparaffinised in xylenes, re-hydrated in ethanol followed by antigen retrieval in boiling 10 mM citrate buffer (pH 6.0). Slides were blocked with Power Block (BioGenex), incubated primary antibody for 1 h at room temperature with the following antibodies: ER (SP1, 790-4324), HER2 (4B5, 790-2991), Ki67 (30-9, 790-4286), Pan-Keratin (AE1/AE3/PCK26, 760-2135), Vimentin (V9, 790-2917), p53 (DO-7, 790-2912), CK5/6 (D5/16B4, 790-4554), CK8/18 (B22.1 & B23.1, 760-4344), CD45 (RP2/18, 760-2505), p63 (4A4, 790-4509), Synaptophysin (SP11, 790-4407) or E-cadherin (36, 790-4497) (Ventana). This was followed by 3% H₂O₂ for 30 min then by SignalStain Boost (Cell Signaling) secondary antibody for 30 min. The SignalStain DAB substrate kit (Cell Signaling) was used as a detection method prior to counterstaining with Harris’ hematoxylin, dehydration and mounting. Slides were scanned using Aperio-XT slide scanner (Aperio). Sankey diagrams were constructed with SankeyMATIC (BETA).