Supporting Tables and Figures

Supplementary data for

“Regulator of chromosome condensation 2 identifies high-risk patients within both major phenotypes of colorectal cancer”

Supporting Tables and Figures

Fig. S1. Flow diagram for inclusion of patients in the study.

Fig. S2. RNA secondary structure analysis of the 5’UTR RCC2 mutation compared to wild type.

Fig. S3. Vector construction overview.

Fig. S4. Morphological changes in actin fiber patterns following depletion of RCC2 protein in HCT15 cells.

Fig. S5. Subgroup analyses of in situ protein expression of RCC2 for a consecutive CRC series.

Fig. S6. Representative electropherograms from fragment analysis of the 5’UTR RCC2 mutation.

Table S1. Statistics for relevant parameters for the test series (n = 37).

Table S2. Statistics for relevant parameters for the validation series (n = 122).

Table S3. Statistics for relevant parameters for the validation series (n=60, R0 patients).

Table S4. Primer and PCR details.

1

MATERIALS AND METHODS

Mutation verification by sequencing

Genomic DNA was amplified using new primers, including M13-tails, designed by the Primer3 program with default settings (Table S4). Direct sequencing in both directions was carried out on a 3730 Genetic Analyzer (Applied Biosystems by Life Technologies, CA, USA). An initial PCR was performed in a 25µl reaction using HotStar HiFidelity Polymerase (Qiagen, CA, USA) followed by enzymatic clean-up with ExoSAP-IT (USB Corporation, OH, USA) and sequencing of the PCR product using the BigDye® Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems by Life Technologies, CA, USA), all following the manufacturers’ instructions. Both forward and reverse sequences were analyzed and compared with the genomic reference sequences. Ninety five percent of the mutations identified by fragment analysis were confirmed, giving a false discovery rate of 5%.

Reporter gene assays

The method involves cotransfection of an experimental reporter plasmid (expressing a firefly (Photinus pyralis) luciferase gene) together with a control plasmid (expressing a sea pansy (Renilla reniformis) luciferase gene, pGL4.74[hRluc/TK Vector]) to correct for variation in transfection efficiency. The reporter and control plasmids were mixed at an optimized mass ratio between 5:1 to 20:1, to achieve low levels of Renilla luciferase expression relative to firefly luciferase expression, hence reducing any risk of transactivation. See Fig. S3.

Each cell line was transiently cotransfected with reporter plasmids employing one of the following transfection reagents: Lipofectamine 2000 (Invitrogen by Life Technologies, Paisley, UK) for HCT15, Lipofectamine LTX/Plus Reagent (Invitrogen by Life Technologies, Paisley, UK) for HT29 and FuGene 6 (Roche Diagnostics, Oslo, Norway) for HeLa cells. All transfections were optimized to ensure high transfection efficiency and low toxicity, and carried out according to the manufacturer’s instructions. Luminescence was measured using a BMG LABTECH POLARstar Omega microplate reader with its embedded MARS Data Analysis software (Version 1.2 R2). Within each experiment, five biological replicates for wild type and mutant were used. Replicate variation was generally low, ranging from 2% to 6%.

Construction of plasmids

Design of reporter plasmids was ordered from MedProbe (GeneArt). The wild-type and the mutated 5’UTR fragments of RCC2 were each cloned into a pGL3-Promoter Vector (Cat E1761, Promega, CA, USA) upstream of the firefly luciferase gene and downstream of the constitutive SV40 promoter, between the Hind III (245) and NcoI (278) restriction sites (Fig.S3).

LuciferasemRNA expression analysis

A TaqMan assay was designed to examine if the effects of the RCC2 mutation analyzed with the dual luciferase assay could be caused by transcriptional rather than translational regulation. The expression of firefly luciferase was normalized against the expression of Renilla luciferase to correct for variation in transfection efficiency.

A default TaqMan-assay was used, including 2 minutes at 50°C to remove all unconverted RNA, 10 minutes at 95°C, followed by 40 cycles with 15 seconds at 95°C, and 60 seconds at 60°C. The RT-PCR mix consisted of 10 µl 2X Universal master mix II, with UNG (Applied Biosystems by Life Technologies, CA, USA), 0.2 µM probe, 0.9 µM of each primer, 50 ng template, and dH20 to a total of 20 µl.

A total of 15 independent transfections were analyzed (n = 8 for HCT15, n = 5 for HT29 and n = 2 for HeLa). The samples were run in triplicate for all experiments, and fold difference between mutant luciferase expression and wild type luciferase expression was calculated using ΔΔCt, according to the manufacturer’s instructions (Applied Biosystems by Life Technologies, CA, USA).

The primers used for thefirefly luciferase expression analysis were the following:

LUC_pGL3_fwd: 5’-TGCACATATCGAGGTGGACATC-3’; LUC_pGL3_rev: 5’-TGCCAACCGAACGGACAT-3’; probe, 5’-CTTACGCTGAGTACTTCG-3’.

The primers used for theRenilla luciferase expression analysis were the following:

RLUC_pGL4_v2_fwd: 5’-CAACGCAAACGCATGATCAC-3’; RLUC_pGL4_v2_rev: 5’- GCACGTTCATTTGCTTGCA -3; probe: RLUC_pGL4_v2: 5’-CTCAGTGGTGGGCTC-3’

Immunohistochemistry

Tissue cores were organized into a tissue microarray according to the original method described by Kononen and colleagues in 1998 (Kononen et al, Nat. Med. 1998). The in situ protein expression analysis was done on 3 µm thick sections on microscope slides, and was performed as previously described (Skotheim et al, Neoplasia 2003). In brief, sections were de-paraffinized in xylene for 10 minutes, and then rehydrated. Antigen retrieval was performed in a microwave oven by heating the sections in plastic containers filled with Tris-buffer (pH=9). Staining was performed according to the DAKO Envision protocol, using the reagents supplied with the K5007 kit (Dako, Glostrup, Denmark). A test tissue microarray containing representative tissues from nine human organs and six types of cancer was utilized to optimize staining conditions. A negative control experiment was provided by omitting the primary antibody from one slide. The staining of RCC2 was scored according to the proportion and intensity categories proposed by Allred et al. (Allred et al, Mod Pathol.1998). The proportion score represents the estimated fraction of positive cells (0 = none, 1 = less than 1%, 2 = 1-10%, 3 = 11-33%, 4 = 34-66% and 5 = 67-100%), while the intensity score represents their average staining intensity (0 = negative, 1 = weak, 2 = intermediate, 3 = strong). Expression was scored separately for cytoplasmic and nuclear staining patterns. Tumors with a score above 4 were termed strong, while tumors with score 4 and below were termed weak. The scoring was performed independently by two investigators (JB and MK), blinded to clinical data, in close collaboration with an experienced pathologist. The interobserver agreement was strong with intraclass correlation coefficients (ICC) of 0.89 and 0.67 for nuclear and cytosolic staining, respectively. Due to the limited ability of the Allred scoring system to differentiate proportionately between negative tumors (score 0) and weak tumors (score 4-6, mostly), these ICCs underestimate the true value, especially for cytosolic staining for which tumors largely exhibited a uniform staining with proportion scores of 4 or 5. Calculations were confirmed by cross tabular visualizations. All discrepancies were resolved and reassigned on consensus of opinion.

1

Fig. S1. Flow diagram for inclusion of patients in the study.

aTissue from four patients with endoscopic procedure included. bFrom patients with synchronous tumors only one tumor was randomly selected for mutation analyses. cUnevaluable tissue had insufficient number of epithelial tumor cells, extensive necrosis and/or poor tumor preservation.

1

Fig. S2. RNA secondary structure analysis of the 5’UTR RCC2 mutation compared to wild type

Fig. S3. Vector construction overview. The synthetic gene constructs, (Construct_no1(wt) and Construct_no2(mut)), were assembled from synthetic oligonucleotides and/or PCR products. The fragments were cloned into pGL3-Promoter_A202 using HindIII and Ncol cloning sites. Plasmid DNA was purified from transformed bacteria and concentration determined by UV spectroscopy. The final construct was verified by sequencing from both sides and the sequence congruence within the used restriction sites was 100%.

Construct_no1(wt) and Construct_no2(mut):

HindIII–CGCACATGTGTTTCTGTTTTGTGTTGTAGCATTTGTTCTGGAAGCTCGTATTTACATTTTAAGTGTATCTGGTGAGTGGGCTGGAGCCCTCGTCTGGGCCGGAAAAAAAAAAGCCCTCCGATCCGTCTTTTAGTTGCTTCTCTTCCTTTTTTCTCTCCGGTTTCTCATCACTCCAACCAGCCGCGACC – Ncol

The red base is deleted in the mutated construct.

Fig. S4. Morphological changes in actin fiber patterns following depletion of RCC2 protein in HCT15 cells. The arrows indicate both ventral- and transverse arc stress-like fibers. The asterisks indicate lamellipodia-like structures. The star indicates filopodia-like structures. All indicated cells are near negative for RCC2 protein staining, as shown in the main figure (Fig.5).

Fig. S5. Subgroup analyses of in situprotein expression of RCC2 for a consecutive CRC series. Here RCC2 staining demonstrate ability to stratify high- and low-risk patients with rectal cancer, early stage disease (stage I/II) and old age (>70).

The Kaplan-Meier method (logrank test) was used to draw the plots and determine statistical differences in time-to-recurrence between patients with strong and weak RCC2 staining. Green curves represent strong staining and blue curves represent weak staining. The analyses were restricted to patients with complete resection(R0).

Fig. S6.Representative electropherograms from fragment analyses of the 5’UTR RCC2 mutation. Wild type (upper) and mutant (lower).

Parameter / RCC2 wt (n) / RCC2 mut (n) / P-valuea
Gender / 0.50
male / 9 / 5
female / 11 / 12
Age / 0.78b
median / 65 / 69
range / 26-92 / 33-90
Stage / 1.0b
I / 2 / 3
II / 10 / 6
III / 5 / 6
IV / 3 / 2
Tumor location / 0.09b
proximal colon / 9 / 13
distal colon / 6 / 2
rectum / 5 / 2

Table S1. Statistics for relevant clinical parameters for the test series (n=37).

Abbreviations: ND – no data

aFisher’s exact test, 2-tailed

bWilcoxon rank-sum test, exact, 2-tailed

Parameter / RCC2 wt (n) / RCC2 mut (n) / P-valuea
Gender / 0.12
male / 13 / 17
female / 25 / 67
Age / 0.31b
range / 30-93 / 40-92
median / 74 / 74
Stage / 0.52b
I / 4 / 6
II / 19 / 54
III / 11 / 15
IV / 4 / 9
Tumor location / 0.54b
proximal colon / 30 / 69
distal colon / 5 / 6
rectum / 1 / 3
Histopathologic grade / 0.72b
G1 / 3 / 3
G2 / 20 / 43
G3 / 15 / 32
Mucinous* / 0 / 5
ND* / 0 / 1
Resection / 0.039b
0 (complete) / 29 / 76
1 (marginal) / 2 / 2
2 (intralesional) / 7 / 7

Table S2. Statistics for relevant clinical parameters for the validation series (n=122).Abbreviations: G1 – High differentiation; G2 – Moderate differentiation; G3 – Poor differentiation;ND – no data

aFisher’s exact test, 2-tailed

bWilcoxon rank-sum test, exact, 2-tailed

*Excluded from rank test

Parameter / RCC2 wt (n) / RCC2 mut (n) / P-valuea
Gender / 0.029
male / 7 / 7
female / 8 / 38
Age / 0.66b
range / 30-92 / 50-90
median / 70 / 72
Stage / 0.25b
I / 2 / 5
II / 7 / 32
III / 6 / 7
IV / 0 / 1
Tumor location / 1.0b
proximal colon / 15 / 45
distal colon / 0 / 0
rectum / 0 / 0
Histopathologic grade / 0.72b
G1 / 2 / 2
G2 / 8 / 25
G3 / 5 / 15
Mucinous* / 0 / 2
ND* / 0 / 1

Table S3. Statistics for relevant clinical parameters for the Validation series (n=60, allpatients with complete resection, R0).Abbreviations: G1 – High differentiation; G2 – Moderate differentiation; G3 – Poor differentiation;ND – no data

aFisher’s exact test, 2-tailed

bWilcoxon rank-sum test, exact, 2-tailed

*Excluded from rank test

1

Gene name / Multiplex / Annealing / Number of / [primer] / Fluorescent / Size in
(HGNC) / group / temperature / PCR cycles / in pmol / dye / bp / Forward / Reverse
Fragment analysis primers
ACVR2A / 4 / 58°C / 27 / 1.6 / VIC / 113 / GTTGCCATTTGAGGAGGAAA / CAGCATGTTTCTGCCAATAATC
ADCY1 / 9 / 58°C / 27 / 1.2 / NED / 98 / CCAGAAGCAAATTCACAAGAC / TTTTGCGTGTTCCTTCCTTC
AIM2 / 8 / 60°C / 27 / 1.6 / PET / 76 / TTCTCCATCCAGGTTATTAAGGC / TTAGACCAGTTGGCTTGAATTG
ASTE1 / 3 / 60°C / 27 / 1.0 / VIC / 117 / ATATGCCCCCGCTGAAATA / TTGGTGTGTGCAGTGGTTCT
ATR / 3 / 60°C / 27 / 2.0 / PET / 70 / GCTTCTGTCTGCAAGCCATT / TGAAAGCAAGTTTTACTGGACTAGG
AXIN2 / 2 / 60°C / 27 / 1.6 / 6-FAM / 123 / CCTACCCCTTGGAGTCTGC / CAGGGTCCTGGGTGAACA
BAX / 1 / 58°C / 27 / 0.8 / VIC / 94 / ATCCAGGATCGAGCAGGGCG / ACTCGCTCAGCTTCTTGGTG
BLM / 9 / 58°C / 27 / 1.2 / 6-FAM / 88 / GAGTAGCAACTGGGCTGAAA / GACAGCAGTGCTTGTGAGAA
CASP5 / 5 / 60°C / 27 / 1.2 / PET / 141 / CAGAGTTATGTCTTAGGTGAAGG / ACCATGAAGAACATCTTTGCCCAG
CDC25 / 2 / 60°C / 27 / 1.6 / VIC / 179 / AGCAGTCAGAGCCCTTAACC / GCCATTTAAAGCTAGTTATCTAATCCA
EP300, ex27 / 3 / 60°C / 27 / 1.0 / PET / 156 / ACACAACAGGGCATATTTGG / CATGGACAATACGCTCTGATACA
EP300, ex3 / 5 / 60°C / 27 / 1.2 / VIC / 100 / GCCTTGGTCTCCAGATTCAG / ATGTTGGGCATTCCTCCA
EPHB2 / 5 / 60°C / 27 / 1.2 / VIC / 85 / CACGAGACGTCACCAAGAAA / CCCCTCCCAGGATCTGTT
GRB14 / 7 / 53°C / 27 / 1.6 / 6-FAM / 80 / GGCAATAGTGATATTTATGT / GGCTTAAAGCAGAATCCAT
GRK4 / 2 / 60°C / 27 / 1.6 / NED / 68 / GGAAAAATGTATGCCTGCAA / GCCATAGCTTCACCTTTCCTC
IGF2R / 1 / 58°C / 27 / 1.0 / NED / 111 / GCAGGTCTCCTGACTCAGAA / GAAGAAGATGGCTGTGGAGC
MARCKS / 10 / 58°C / 30 / 3.5 / PET / 109 / GACTTCTTCGCCCAAGGC / GCCGCTCAGCTTGAAAGA
MBD4 / 4 / 58°C / 27 / 2.0 / NED / 98 / TGACCAGTGAAGAAAACAGCC / GTTTATGATGCCAGAAGTTTTTTG
MRE11A / 7 / 53°C / 27 / 1.6 / NED / 122 / AATATTTTGGAGGAGAATCT / AATTGAAATGTTGAGGTTGCC
MSH3 / 9 / 58°C / 27 / 1.2 / VIC / 153 / AGATGTGAATCCCCTAATCAAGC / ACTCCCACAATGCCAATAAAAAT
MSH6 / 8 / 60°C / 27 / 1.2 / 6-FAM / 94 / GGGTGATGGTCCTATGTGTC / CGTAATGCAAGGATGGCGT
OGT / 4 / 58°C / 27 / 1.2 / PET / 116 / TCACTTTTGGCTGGTCAGAG / GGGAGGGAAAGGAGGTAAAG
PA2G4 / 5 / 60°C / 27 / 1.2 / 6-FAM / 93 / GCAAGTCGAAAAACCCAGAA / GGGAGAAAGAGGGATCAGGT
PCNXL2 / 6 / 57°C / 27 / 1.0 / NED / 121 / GGAAAATTATGAACAGCCACAA / GCAGCCAAATGCTTGTTATG
PRDM2 / 7 / 53°C / 27 / 1.2 / VIC / 145 / TCTCACATCTGCCCTTACTG / GTGATGAGTGTCCACCTTTC
PRKDC / 1 / 58°C / 27 / 1.6 / 6-FAM / 113 / GACTCATGGATGAATTTAAAATTGG / TTTGAAAATAACATGTAAATGCATCTC
PTEN, ex7 / 7 / 53°C / 27 / 1.6 / 6-FAM / 236 / CCTGTGAAATAATACTGGTATG / GTTTCTTCTCCCAATGAAAGTAAAGTACA
PTEN, ex8 / 7 / 53°C / 27 / 1.6 / PET / 166 / GTGCAGATAATGACAAGGAATA / GTTTCTTACACATCACATACATACAAGTC
PTHLH / 2 / 60°C / 27 / 1.6 / PET / 113 / TTTCACTTTCAGTACAGCACTTCTG / GAAGTAACAGGGGACTCTTAAATAATG
RAD50 / 3 / 60°C / 27 / 1.0 / NED / 86 / GCGACTTGCTCCAGATAAAC / GCACAAGTCCCAGCATTTCA
RBBP8 / 8 / 60°C / 27 / 2.0 / VIC / 88 / GAATACAAGTTTGTCCCCTTC / GCTAGATATACAAGTGTTGCTA
RCC2 / 6 / 57°C / 27 / 1.6 / VIC / 143 / GCATTTGTTCTGGAAGCTCGT / GTGATGAGAAACCGGAGAGAA
SEC63 / 2 / 60°C / 27 / 1.6 / NED / 145 / GGAGGATGGCAACAGAAGAG / CATTCCCAACGACTCCATTT
SEMG1 / 5 / 60°C / 27 / 1.2 / NED / 152 / ACAACGACCGAAACCCATTA / CCCACAAAAGTCCTGGAGAG
SLC23A2 / 1 / 58°C / 27 / 1.0 / 6-FAM / 61 / GACTACTACGCCTGTGCACG / TGTTTATTGCGTGGATGGG
SPINK5 / 2 / 60°C / 27 / 1.2 / VIC / 99 / TGAGGCGTTTGTTCACTTTG / TGCTCCTGTCTTCATCCTCTT
SYCP1 / 4 / 58°C / 27 / 1.6 / 6-FAM / 153 / CCCCTTCATCTCTAACAACCC / CACTGATTCTCTGAAATTAAACAAATAAC
TAF1B / 6 / 57°C / 27 / 1.2 / VIC / 115 / CCAAATAAAAGCCCTCAACC / TGTCCTGACATCATGAAGGTG
TCF7L2 / 4 / 58°C / 27 / 1.6 / 6-FAM / 75 / GCCTCTATTCACAGATAACTC / GTTCACCTTGTATGTAGCGAA
TGFBR2 / 1 / 58°C / 27 / 1.0 / NED / 73 / CTTTATTCTGGAAGATGCTGC / GAAGAAAGTCTCACCAGG
UVRAG / 10 / 58°C / 30 / 4.5 / PET / 116 / TTTATTTTTAAACATTGTGAGTATG / TTTTTAACTGCAGGCATTCAC
WIPS3 / 10 / 58°C / 30 / 7.0 / PET / 126 / TCTCCCTTTGTTTTAGC / ATTGGTCACCCTGTTAG
ZMYND8 / 6 / 57°C / 27 / 1.2 / 6-FAM / 83 / CAAAGAAGGTTGTCAGATGGAC / CCTCTTTAATCTCCACTGGG
Consensus MSI-primers
BAT25 / NA / 55°C / 27 / 1.2 / NED / 124 / TCGCCTCCAAGAATGTAAGT / TCTGCATTTTAACTATGGCTC
BAT26 / NA / 55°C / 27 / 1.6 / 6-FAM / 122 / TGACTACTTTTGACTTCAGCC / AACCATTCAACATTTTTAACCC
D17S250 / NA / 55°C / 27 / 3.0 / 6-FAM / 152 / GGAAGAATCAAATAGACAAT / GCTGGCCATATATATATTTAAACC
D2S123 / NA / 55°C / 27 / 1.6 / NED / 211 / AAACAGGATGCCTGCCTTTA / GGACTTTCCACCTATGGGAC
D5S346 / NA / 55°C / 27 / 1.2 / VIC / 125 / ACTCACTCTAGTGATAAATCGGG / AGCAGATAAGACAGTATTACTAGTT
Sequencing primers
ACVR2A / NA / 58°C / 40 / 10.0 / NA / 311 / CCAGTTTGAAAGTCAGGAGGA / TGTGAAGATCACCTTCCAGAAA
AIM2 / NA / 58°C / 40 / 10.0 / NA / 348 / GATCCAAGGCAGACCAATGT / TTCTGAATTTGTTTATCCAGCAA
ASTE1 / NA / 58°C / 40 / 10.0 / NA / 395 / TCCTGTTGCACTGAATTACTTCTT / AACTGAGTTTTATTCAATGTTGGAG
AXIN2 / NA / 58°C / 40 / 10.0 / NA / 398 / CCCAGTTTCTTTCCTTCTGTTTT / TTCTCATGGGAGGGTTTGAG
BLM / NA / 58°C / 40 / 10.0 / NA / 392 / CCCTATGGAGGGTGATTCCT / CCCAGTCATCATCATCATCAA
EPHB2 / NA / 58°C / 40 / 10.0 / NA / 322 / CTCGGCTCACCTCTTCCTC / TGGACACATCGCATGAATCT
GRK4 / NA / 58°C / 40 / 10.0 / NA / 298 / CCTAAGAAATGCCAGGTGGA / AATGACTTCCACGGCTTCAG
MBD4 / NA / 58°C / 40 / 10.0 / NA / 302 / CTGCATTTCTGATGCTGGAG / TTGGTGAGCAGTTGTTGTCC
PTHLH / NA / 58°C / 40 / 10.0 / NA / 370 / CCACGGCTAAAAGACCTGAG / GATGGGTTTGCCAGCTTAAA
RAD50 / NA / 58°C / 40 / 10.0 / NA / 276 / CCCCGTTTGTCAGAGAGTTT / GGCATACCAGCTCAGAGTCC
RBBP8 / NA / 58°C / 40 / 10.0 / NA / 398 / TTCCTCTGCTTTTCCCCTTC / TGATGTGTGAAAAGGGCACTA
RCC2 / NA / 58°C / 40 / 10.0 / NA / 328 / CCGCACATGTGTTTCTGTTT / ATCCGCCTTCCTTCCTCTT
SEMG1 / NA / 58°C / 40 / 10.0 / NA / 246 / AGTGATCGTCATTTGGCACA / GGGGAGGCTCATCTTCCTAC
SLC23A2 / NA / 58°C / 40 / 10.0 / NA / 341 / ACTGGTCCTGGTCACTTTCG / GTCGCTAGAGTCCTGCTGCT
SYCP1 / NA / 58°C / 40 / 10.0 / NA / 228 / TGTACTCAGGCCCCTTCATC / TTTGGCTCTGGCAAATAAGAA
TAF1B / NA / 58°C / 40 / 10.0 / NA / 266 / GGTCCTGGCACTCACAATTT / TGTCCTGACATCATGAAGGTG
WISP3 / NA / 58°C / 40 / 10.0 / NA / 350 / ATTTTGGGGTTGGGAAAGAG / CTGTCGCAAGGCTGAATGTA
ZMYND8 / NA / 58°C / 40 / 10.0 / NA / 376 / CCACATGCTGGGTTTTCTTT / TTTATGGGGTGAGGTGAAGG

Table S4. Primer and PCR details.The fragments were amplified in multiplex PCR. One primer per fragment was labeled with a fluorescent dye (6-FAM, VIC, NED or PET) from the G5 dye set from Applied Biosystems. The size standard used was GS500 LIZ (orange). All fragments were analyzed on a 3730 DNA Analyzer (Applied Biosystems by Life Technologies, CA, USA) using default microsatellite analysis settings and the GeneMapper 3.7 software.

Abbreviations: HGNC – Human Genome Nomenclature Committee; NA – not applicable; pmol – picomol; bp – base pair.

1

Test series (n=37) / Validation series (n=108)a
Gene / n / KM estimate (S.E.) / P-value (log rank) / n / KM estimate (S.E.) / P-value (log rank)
ACVR2A / 0.022 / 0.48
WT / 5 / 40(22) / 8 / 50(18)
Mut / 32 / 83(7) / 85 / 60(6)
ND / 15
ADCY1 / 0.67
WT / 17 / 73(11)
Mut / 20 / 80(9)
AIM2 / 0.98
WT / 14 / 77(12)
Mut / 23 / 77(9)
ASTE1 / 0.29
WT / 12 / 67(14)
Mut / 25 / 83(8)
ATR / 0.46
WT / 35 / 76(8)
Mut / 2 / 100
AXIN2 / 0.036 / 0.79
WT / 31 / 83(7) / 93 / 60(5)
Mut / 6 / 50(20) / 14 / 62(13)
ND / 1
BAX / 0.32
WT / 23 / 71(10)
Mut / 14 / 86(9)
BLM / 0.13
WT / 30 / 71(9)
Mut / 7 / 100
CASP5 / 0.49
WT / 14 / 71(12)
Mut / 23 / 81(9)
CDC25 / 0.81
WT / 32 / 77(8)
Mut / 5 / 75(22)
EP300, ex27 / 0.003 / 0.48
WT / 25 / 91(6) / 86 / 54(6)
Mut / 12 / 50(14) / 4 / 75(22)
ND / 18
EP300, ex3 / 0.61
WT / 36 / 77(7)
Mut / 1 / 100
EPHB2 / 0.67
WT / 25 / 75(9)
Mut / 11 / 80(13)
GRB14 / 0.57
WT / 24 / 74(9)
Mut / 13 / 83(11)
GRK4 / 0.22
WT / 32 / 73(8)
Mut / 5 / 100
IGF2R / 0.10
WT / 27 / 70(9)
Mut / 10 / 100
MARCKS / 0.46
WT / 15 / 71(12)
Mut / 22 / 81(9)
MBD4 / 0.46
WT / 23 / 73(10)
Mut / 14 / 85(10)
MRE11A / 0.010 / 0.39
WT / 10 / 50(16) / 15 / 47(13)
Mut / 27 / 88(7) / 72 / 58(6)
ND / 21
MSH3 / 0.76
WT / 17 / 75(11)
Mut / 20 / 79(9)
MSH6 / 0.62
WT / 31 / 79(8)
Mut / 6 / 67(20)
OGT / 0.011 / 0.74
WT / 22 / 62(11) / 55 / 62(7)
Mut / 15 / 100 / 52 / 58(7)
ND / 1
PA2G4 / 0.61
WT / 35 / 76(7)
Mut / 2 / 100
PCNXL2 / 0.40
WT / 13 / 69(13)
Mut / 24 / 82(8)
PRDM2 / 0.67
WT / 25 / 75(9)
Mut / 12 / 82(12)
PRKDC / 0.63
WT / 24 / 75(9)
Mut / 13 / 82(12)
PTEN, ex7 / 0.46
WT / 35 / 76(8)
Mut / 2 / 100
PTEN, ex8 / 0.28
WT / 33 / 74(8)
Mut / 4 / 100
PTHLH / 0.11
WT / 11 / 60(16)
Mut / 26 / 84(7)
RAD50 / 0.65
WT / 21 / 75(10)
Mut / 16 / 80(10)
RBBP8 / 0.69
WT / 29 / 79(8)
Mut / 8 / 71(17)
RCC2 / 0.045 / 0.0115
WT / 20 / 65(11) / 30 / 43(9)
Mut / 17 / 93(6) / 75 / 68(6)
ND / 3
SEC63 / 0.41
WT / 22 / 73(10)
Mut / 15 / 85(10)
SEMG1 / 0.54
WT / 25 / 74(9)
Mut / 12 / 83(11)
SLC23A2 / 0.21
WT / 20 / 85(8)
Mut / 17 / 67(12)
SPINK5 / 0.46
WT / 35 / 76(8)
Mut / 2 / 100
SYCP1 / 0.40
WT / 32 / 80(7)
Mut / 5 / 60(22)
TAF1B / 0.13
WT / 10 / 60(16)
Mut / 27 / 84(7)
TCF7L2 / 0.34
WT / 22 / 71(10)
Mut / 15 / 86(9)
TGFBR2 / 0.39
WT / 9 / 67(16)
Mut / 28 / 81(8)
UVRAG / 0.29
WT / 26 / 72(9)
Mut / 11 / 90(10)
WIPS3 / 0.92
WT / 26 / 76(9)
Mut / 10 / 78(14)
ZMYND8 / 0.61
WT / 36 / 77(7)
Mut / 1 / 100

Table S5. Association between frameshift mutation in 42 genes and time-to-recurrence in the Test series (n=37). The six genes found to be significantly associated with time-to-recurrence in the Test series were tested in the Validation series (n=108).aA set of samples (n=16) with no determined MSI status waslater reanalyzed in order toextend the MSI series for more robust RCC2 mutation analyses, yielding a total of 128 MSI positive tumors (of which 122 with RCC2mutation status).

1

Variable / Patients n (%) / Univariate analysisa / Multivariate analysisb
5-year TTR (%) / P / HR / 95% CI / P
Total
Gender
Female
Male
Agec
RCC2
Wild type
Mutated
Stage
I+II
III+IV / 37(100)
23 (62)
14 (38)
20 (54)
17 (46)
21 (57)
16 (43) / 76.2
78.6
65.0
93.3
95.2
50.0 / 0.95
-
0.045
0.0019 / 1
0.98
1.01
1
0.12
1
15.6 / 0.23-4.15
0.96-1.05
0.01-1.03
1.86-132 / 0.97
0.71
0.054
0.011

Table S6. Univariate and multivariate analyses of RCC2 5’UTR mutation status in the Test series. aKaplan-Meier estimate (logrank test). bCox regression model (Wald test) - all included parameters are displayed in the table. cHazard ratios are given per year of age.

1

1