SUPPLEMENTARY INFORMATION

Methods

Immunohistochemistry

Two cores were included for each case and, as for SCP, one of each histological pattern was taken. Whole 2.5 µm sections of the tissue microarray (TMA) were incubated with the panel of antibodies: FSCN1, HPCA, EphB1, HIF1α and PTCH1. Immunohistochemical assessment was made without knowledge of the histological diagnosis.

Table S1. Antibodies, equipment and immunohistochemical procedures.

Antibody / Equipment / Ab purveyor / Code / Antigenretrievalconditions / Ab dilution / Incubation / Clone
buffer / pH / Tempe-rature (ºC) / (min) / (min)
FSCN11 / Autostainer link48 Dako
(Via Real, CA, USA) / Dako
(Dako, Glostrup, Denmark) / M356701-8 / citrate / 6.1 / 95 / 20 / 1:100 / 40 / 55K-2
HPCA / Abgent
(San Diego, CA, USA) / AP1564A / Tris/EDTA / 9 / 95 / 20 / 1:100 / 90 / polyclonal
HIF1a / Abgent
(San Diego, CA, USA) / 2015-1 / Tris/EDTA / 9 / 95 / 20 / 1:150 / O/N / EP1215Y
PTCH1 / Santa Cruz
Biotech, Inc. / sc-6149 / Tris/EDTA / 9 / 95 / 20 / 1:100 / O/N / polyclonal
EPHB2 / R&D System / AF467 / Tris/EDTA / 9 / 95 / 20 / 1:100 / O/N / Polyclonal

Ab: Antibody

Endogenous peroxidase activity was blocked using 0.5% H2O2 for 5 minutes. After primary antibody incubation sections were subsequently treated with peroxidase-labelled polymer (Envision FLEX/HRP, Dako) for 20 minutes at room temperature. For visualisation of the antigen, the sections were immersed for five minutes each in two consecutive fresh 3,3'-diaminobenzidine (DAB) baths (Envision Flex DAB+ chromogen, Dako) and counterstained with Harris´ haematoxylin for 5 minutes. Immunostained slides were evaluated independently by two pathologists (JGS and MGS). Inter-observer discordance (<15%) was resolved using a multi-headed microscope and a consensus score was reached.

DNA extraction and oncogene mutation analysis

From the original paraffin blocks, five sections (4µm each) showing tumor in more than 70% of the tissue were selected and cut by two pathologists (JGS, MPG). Genomic DNA extraction was performed using the QIAmp DNA minikit (cat: 51306) and the QiaCube automatic nucleic acid extractor (Qiagen, Hilden, Germany) according to the manufacturers´ instructions. Mutation in KRAS (12, 13 codons), BRAF (600 codon) and PIK3CA (exon 9) were determined by direct Sanger sequencing as previously described [9].

Microsatellite instability

MSI determination was based on the use of fluorescence-labeled primers for co-amplification of seven markers including five mononucleotide repeat (BAT-25, BAT-26, NR-21, NR-24 and MONO-27) and two pentanucleotide (Penta C and Penta D), as previously described [9]. The cases were categorized as MSI-H or microsatellite stable (MSS)/low-level MSI (MSI-L) according to the NIH criteria [13].

Results

Table S2. Distribution of serrated and conventional carcinomas and serrated, conventional and SCP according to the points obtained after applying the molecular β-score based on KRAS and BRAF mutation status.

Points / Serrated carcinoma
n=98 / Conventional carcinoma
n=92 / Serrated polyp
n=47 / Conventional polyp
n=56 / SCP
n=3
0 / 3 (3.1%) / 0 / 1 (2.1%) / 0 / 0
1 / 20 (20%) / 2 (2.2%) / 30 (64%) / 5 (8.9%) / 0
3 / 41 (42%) / 31 (34%) / 10 (21%) / 13 (23%) / 2 (67%)
4 / 34 (35%) / 59 (64%) / 6 (13%) / 38 (68%) / 1 (33%)

17 HP, 17 SSA, 13 TSA, 3 SCP, 12 VA, 29 TA and 15 TVA

Table S3. Distribution of the different serrated and conventional polyp types according to the points obtained after applying the molecular β predictive score based on KRAS and BRAF mutation status.

Pt. / PH
n=17 / SSA
n=17 / TSA
n=13 / VA
n=12 / TA
n=29 / TVA
n=15 / SCP
n=3
0 / 1 (5.9%) / 0 / 0 / 0 / 0 / 0 / 0
1 / 11 (65%) / 14 (82%) / 5 (39%) / 1 (8.3%) / 2 (6.9%) / 2 (13%) / 0
3 / 3 (18%) / 2 (12%) / 5 (39%) / 4 (33%) / 3 (10%) / 6 (40%) / 2 (67%)
4 / 2 (12%) / 1 (5.9%) / 3 (23%) / 7 (58%) / 24 (83%) / 7 (47%) / 1 (33%)

3