Clone 6g11

Five μm slices were cut and mounted on aminoacetylsilane coated glass slides (Starfrost, Berlin, Germany). Slides were deparaffinized in xylene and rehydrated in decreasing ethanol steps. Endogenous peroxidase was blocked in 1% hydrogen peroxide in Phosphate Buffered Saline (PBS) for 20 min. Antigen retrieval was performed with TRIS (hydroxymethyl) aminomethane-EDTA (pH 9.0, Klinipath, Duiven, The Netherlands) in a microwave (MicroMed T/T Mega, Milestone, Sorisole, Italy) for 15 min. Slides were incubated overnight at 4°C with primary antibodies targeting N-cadherin (1:50, clone 6G11, Dako, Glostrup, Denmark), vimentin (1:500, clone Vim3B4, Dako, Glostrup, Denmark), fibronectin (1:500, ab2413, Abcam, Cambridge, U.K.) or E-cadherin (1:200, clone NCH-38, Dako, Glostrup, Denmark) diluted in phosphate-buffered normal antibody diluent (Immunologic, Duiven, The Netherlands). The specificity of all antibodies had been verified using Western blot analysis (data not shown). Secondary antibodies were incubated for 30 min. at room temperature and chromogenic visualization was performed with the EnVision Dako kit (Dako, Glostrup, Denmark). Slides were counterstained with haematoxylin, washed, dehydrated, cleared in xylene and mounted in Entellan® new (Merck Millipore, Billerica, U.S.A.). All immunohistochemical stainings included negative controls by omitting the primary antibody.

In the immunohistochemistry cohort, IHC was performed with mouse monoclonal antibodies against Wnt5a (Sigma-Aldrich, clone 3A4, dilution 1:50), N-cadherin (Dako, clone 6G11, dilution 1:30), and E-cadherin/NCH-38 (Dako, clone NCH-38, dilution 1:100) and polyclonal rabbit antibodies against β-catenin/CTNNB1 (PRESTIGE antibodies Sigma, dilution 1:300). The sections were counter-stained with Haematoxylin. Assessment was performed manually, and all the IHC sections were evaluated based on the average staining intensity (0-3) multiplied by the percentage of positive cancer cells (0-3), obtaining a total staining index (SI) (0-9). A SI of 0 was regarded as negative, 1-2 as weak positive staining; 3-6 as moderate, and 9 as strong positive staining (Supplementary Table 7). An experienced pathologist (AMB) validated the scoring.