Chromosomal Instability in BRAF Mutant, Microsatellite Stable Colorectal Cancers (original) (raw)

Abstract

The BRAF oncogene is mutated in 15% of sporadic colorectal cancers. Approximately half of these BRAF mutant cancers demonstrate frequent frameshift mutations termed microsatellite instability (MSI), but are diploid and chromosomally stable. BRAF wild type cancers are typically microsatellite stable (MSS) and instead acquire chromosomal instability (CIN). In these cancers, CIN is associated with a poor outcome. BRAF mutant cancers that are MSS, typically present at an advanced stage and have a particularly poor prognosis. We have previously demonstrated clinical and molecular similarities between MSS cancers with or without a BRAF mutation, and therefore hypothesised that CIN may also be frequent in BRAF mutant/MSS cancers. BRAF mutant/MSS (n = 60), and BRAF wild type/MSS CRCs (n = 90) were investigated for CIN using loss of heterozygosity analysis over twelve loci encompassing chromosomal regions 5q, 8p, 17p and 18q. CIN was frequent in BRAF mutant/MSS cancers (41/57, 72%), which was comparable to the rate found in BRAF wild type/MSS cancers (74/90, 82%). The greatest loss in BRAF mutant/MSS cancers occurred at 8p (26/44, 59%), and the least at 5q (19/49, 39%). CIN in BRAF mutant/MSS cancers correlated with advanced stage (AJCC III/IV: 15/17, 88%; p = 0.02); showed high rates of co-occurrence with the CpG Island Methylator Phenotype (17/23, 74%); and CIN at 18q and 8p associated with worse survival (p = 0.02, p<0.05). This study demonstrates that CIN commonly occurs in advanced BRAF mutant/MSS colorectal cancers where it may contribute to poorer survival, and further highlights molecular similarities occurring between these and BRAF wild type cancers.

Introduction

Sporadic colorectal cancer (CRC) is a diverse disease which results from the progression of differing types of precursor lesions that are molecularly and morphologically distinct. These lesions acquire genetic alterations associating with one of at least two recognized molecular pathways leading to tumorigenesis. The ‘traditional pathway’ is the most well characterized and involves the progression of a conventional type adenoma that may acquire mutation or loss of APC, mutation of KRAS and p53, and chromosomal instability prior to the formation of a carcinoma [1] which is typically BRAF wild type. The more recently described ‘serrated’ pathway involves the progression of a serrated lesion to cancer [2], [3], [4], [5]. This is accompanied by an early mutation of the BRAF oncogene [6], [7], and acquisition of the CpG Island Methylator Phenotype (CIMP) which involves widespread promoter hypermethylation and subsequent silencing of key tumour suppressor genes [8], [9].

BRAF is an integral component of the mitogen-activated protein kinase (MAPK) signalling cascade which promotes cellular proliferation and anti-apoptotic effects [10]. The BRAF mutation is considered a marker for the serrated pathway and is found in approximately 10–15% of CRC [11], including the majority of those showing CIMP. Approximately half of these cancers will have hypermethylation and silencing of a DNA mismatch repair gene, MLH1, due to CIMP [8], [12]. This results in multiple mutations within DNA repeat tracts, termed microsatellite instability (MSI) [13]. These MSI, BRAF mutated cancers have previously been well described as diploid [14], [15], [16], more commonly occurring in older females and the proximal colon, are often mucin producing and poorly differentiated [2], [15], [17], [18]. The remaining half of the BRAF mutant lesions of the serrated pathway that do not have methylation of MLH1, are microsatellite stable (MSS) [19], [20]. In contrast to MSI cancers, these have not been well characterized.

Interestingly, the two BRAF mutant subgroups that differ by microsatellite instability status, confer significantly contrasting prognoses. Whilst BRAF mutant/MSI cancers correlate with an excellent patient outcome, BRAF mutant/MSS cancers are associated with a very poor outcome that is even worse than the BRAF wild type/MSS cancers arising via the ‘traditional pathway’ [20], [21], [22]. The molecular mechanisms underlying this disparity are unknown.

Two distinct forms of genomic instability are known to occur in CRC: MSI and chromosomal instability (CIN) [23], [24]. MSI affects genomic integrity at the DNA level and develops at the polyp/carcinoma transition of BRAF mutant serrated polyps which silence MLH1 by DNA methylation through the CIMP phenotype [13], [25]. Alternatively, CIN which affects approximately 70% of CRCs [18], acts on a wider genomic scale as it refers to losses and/or gains of whole or part chromosomal regions, and has been associated with poorer survival [16], [26], [27], [28]. CIN develops in conventional adenomas which are BRAF wild type and progress towards malignancy via the ‘traditional pathway’. Loss of heterozygosity (LOH), which describes either loss or reduction of one of the two parental alleles at a particular chromosomal location, indicates the presence of CIN. The chromosomal regions 18q, 17p, 5q, and 8p, which harbour key tumour suppressor genes, have been found to show extensive LOH in CRC [1], [23].

It is unclear whether BRAF mutant serrated polyps which become malignant but do not methylate MLH1 and do not develop MSI, manifest chromosomal instability. For example, some recent studies have suggested that CIMP and CIN are mutually exclusive [29], [30]. Whilst this is concordant with the diploid nature of BRAF mutant/MSI serrated pathway cancers with CIMP which progress via a high mutational rate, it does not explain how BRAF mutant/MSS serrated pathway cancers progress and especially the fact that their prognosis is particularly poor [16], [29], [30], [31], [32].

We hypothesized that although CIN may be mutually exclusive with MSI, it is not mutually exclusive with CIMP and may be common in BRAF mutant/MSS cancers. MSI cancers have been excluded for LOH analysis in this and previous studies due to the documented low levels of CIN previously found, and because the degree of instability at microsatellite markers present in MSI cancers does not allow for informative assessment [26], [30], [31], [32].

Materials and Methods

Ethics Statement

Written, informed consent was obtained from each patient involved in this study. This research was approved by the Royal Brisbane and Women’s Hospital and Bancroft Human Research Ethics Committees.

Patient Samples

1052 sporadic CRC and matched normal mucosa samples were obtained after surgical excision at the Royal Brisbane and Women’s Hospital (RBWH), Queensland, Australia. Clinical data, including patient age of onset and gender, anatomical location of cancer, cancer stage and patient outcome were collected where possible from review of patient charts, pathology reports or the Queensland Death Registry. Anatomical location was specified as either proximal (proximal to splenic flexure) or distal. Cancer stage was classified according to both the American Joint Committee on Cancer (AJCC) and Tumour Node Metastases (TNM) staging systems [33].

MSI status, CIMP, BRAF, KRAS and p53 mutation detection

Samples had previously been investigated for MSI according to the National Cancer Institute’s 5 marker panel consisting of two mononucleotide markers, BAT25, BAT26; and three dinucleotide markers, D5S346, D2S123 and D17S250 [34]. MSI was classified if at least one of the mononucleotide markers and at least one other marker (either mononucleotide or dinucleotide) was positive [34], [35]. CIMP was previously investigated using MethyLight with a five marker panel consisting of CACNA1G, IGF2, NEUROG1, RUNX3, SOCS1 [36], [37]_ENREF_35_ENREF_35. Cancers were considered CIMP positive if 3 or more markers were methylated. All samples had been screened for the BRAF V600E (a1796t) mutation using an allelic discrimination assay, and KRAS mutations in codons 12 and 13 using Sequenom MassARRAY technology [37]. The incidence of p53 mutation across exons 4 to 8 has previously been reported for all cancers included in this study [37].

Loss of Heterozygosity Analysis

Chromosomal instability (CIN) was assessed by loss of heterozygosity (LOH) analysis over a total of 12 loci covering 4 chromosomal regions known to harbour key tumour suppressor genes implicated in CRC. These were centred over 5q22.2 (containing the APC gene); 8p22 (linked to a CRC suppressor region); 17p13.1 (spanning the p53 locus), and 18q (for SMAD2, SMAD4 and DCC loci). Three markers were investigated over each of the four regions (5q: D5S346, D5S1466, D5S489; 8p: D8S258, D8S254, D8S1121; 17p: 17S261, 17S578, D17S926; and 18q: D18S55, D18S460, D18S487). Primer sequences for each of the markers were obtained from the Ensembl website (http://www.ensembl.org/index.html). (For details of LOH PCR reactions refer to Table 1 in Supplementary Data S1). Paired normal and cancer PCR products were electrophoresed on a 5% polyacrylamide gel and observed radiologically.

Table 1. Clinicopathological and Molecular Characteristics of the _BRAF_mut/MSS and _BRAF_wt/MSS cohorts.

LOH at individual loci was scored if one of the two alleles of the cancer sample had at least a 40% reduction in intensity compared to its paired normal sample. A polymorphic marker where the two parental alleles were able to be visualised was considered informative. The overall extent of CIN was observed over the 12 loci, and CIN positivity (CIN) was assigned if at least 1 informative marker was observed with LOH. CIN negativity was scored if at least 40% (5 of 12) of markers were informative, and none of these demonstrated LOH. For analysis of CIN at individual regions, the 3 loci were observed and scored as CIN if at least one showed LOH, and CIN negative if 2 of the 3 loci were informative and neither demonstrated LOH.

The fractional allelic loss (FAL) has previously been described [38]. This approach was utilised as further analysis of the presence of CIN, where the total number of allelic loss events was divided by the total number of informative markers per sample and per cohort. We used this to verify the overall rate of allelic loss, and correlations with FAL and stage at presentation, cancer location and gender, and molecular features of p53 and KRAS mutation, and CIMP were determined.

Statistical Analysis

Significant relationships between categorical data were assessed using Fisher’s Exact Test (SPSS version 19; Graphpad software). Differences in continuous data were assessed using t-tests. Cox proportional hazard models were utilised to investigate the effects of various factors on survival rate where death due to cancer was taken to be the event of interest. A log rank test was used to assess the equivalence of distributions when applicable. P-values ≤0.05 were considered significant.

Results

BRAF Mutation in MSS Colorectal Cancers

Of 1052 cancers, 128 (12.2%) had a BRAF mutation. Out of these BRAF mutants, 60 (4.7%) were MSS (_BRAF_mut/MSS) and formed the experimental cohort. This was compared to a randomly selected control cohort of 90 MSS, BRAF wild type cancers (_BRAF_wt/MSS). _BRAF_mut/MSS cancers were predominantly located in the proximal colon (33/48, 69%) compared to the _BRAF_wt/MSS cancers (25/82, 30.5%) (p<0.0001). These proximal, BRAF mut/MSS cancers more commonly arose in females (22/33, 66.7% females; p = 0.01) and there was a non-significant trend toward older age (68.6 years for proximal cancers versus 63.1 years for distal cancers; p = 0.1). The _BRAF_mut/MSS cancers showed no difference in rate of CIMP in the proximal (17/27, 63.0%) versus distal bowel (6/12, 50%), rate of p53 mutation in the proximal (14/31, 45.2%) or distal bowel (6/14, 42.9%), or in stage at diagnosis in the proximal (10/21, 47.6%) versus distal bowel (4/9, 44.4%).

Overall, the BRAFmut/MSS cohort did not differ by age of onset, gender distribution or stage at diagnosis to the _BRAF_wt/MSS cohort (Table 1). Molecularly, the _BRAF_mut/MSS cancers did not differ from _BRAF_wt/MSS cancers with extent of p53 mutation (21/57, 37% versus 42/90, 47%), but were significantly more likely to demonstrate CIMP (27/47, 57% versus 3/79, 4%) (p<0.0001) (Table 1). In the _BRAF_wt/MSS cohort, a KRAS mutation was present in 41/90 (46%) cancers (Table 1).

Overall Frequency of Chromosomal Instability (CIN)

The _BRAF_mut/MSS and _BRAF_wt/MSS cancer cohorts displayed highly informative markers for LOH at 95% (57 of 60) and 100% (90 of 90) respectively.

Both cohorts demonstrated a high rate of CIN regardless of BRAF mutation status, with 41/57 (72%) BRAF mutant, and 74/90 (82%) BRAF wild type showing CIN (Table 2). The fractional allelic loss (FAL) [38] which describes the frequency of allelic loss events amongst the total number of informative events, was analysed as a further method of quantifying the overall degree of CIN within a cohort. _BRAF_mut/MSS cancers showed a slightly lower average FAL compared to _BRAF_wt/MSS cancers (132/428 = 0.308, versus 286/745 = 0.384 respectively), but this difference was not statistically significant (p = 0.1) (Table 3).

Table 2. Molecular Characteristics of Cohorts Relative to Presence or Not of Chromosomal Instability (CIN).

Table 3. Fractional Allelic Loss (FAL) describing the number of loss events over the total number of informative events for a sample per cohort.

CIN at Individual Chromosomal Regions

The _BRAF_mut/MSS cancers showed CIN at the highest frequency at the 8p chromosomal region (26 of 44, 59%), and the least at the 5q region (19 of 49, 39%). The _BRAF_wt/MSS cancers had the highest frequency of CIN at 17p (43 of 65, 66%), and the lowest rates of CIN+ at 5q (43 of 80, 54%) and 8p (39 of 72, 54%) (Table 2).

The _BRAF_wt/MSS cancers had a higher frequency of CIN events compared to the _BRAF_mut/MSS cancers occurring at the individual chromosomal regions of 5q, 17p and 18q. This relationship was statistically significant at 18q, with 60% (50 of 83) of _BRAF_wt/MSS cancers showing CIN, compared to 40% (19 of 48) of _BRAF_mut/MSS cancers with CIN at this loci (p = 0.03) (Table 2).

p53 Mutation Rate and Presence of CIN

Mutation of p53 was correlated with deletion of the p53 locus (chromosome 17p) (Table 2). The frequency of p53 mutation in relation to CIN was therefore investigated with the exclusion of 17p data in order to avoid over-estimating an association that may be driven by inclusion of LOH positive markers surrounding the p53 locus. Based on chromosome 5q, 8p and 18q loci; in the _BRAF_wt/MSS cohort, 38/70 (54%) of CIN cancers had a concurrent p53 mutation, compared to only 4/20 (20%) of CIN-negative cancers (p = 0.01). The BRAF mutant/MSS cancers demonstrated a similar but non-significant trend, with 16/37 (43%) CIN cancers having a p53 mutation compared to only 4/17 (24%) CIN-negative cancers (Table 2). Within the _BRAF_wt/MSS cohort, LOH at all individual chromosomal regions significantly correlated with a higher incidence of p53 mutation (Table 2). Average FAL per cohort demonstrated a higher rate of loss coinciding with a p53 mutation compared to those wild type for p53, significantly so for _BRAF_wt/MSS cancers (p = 0<0.001) (Table 3).

KRAS Mutation Rate and Presence of CIN in BRAFwt/MSS Cancers

BRAF and KRAS mutations were confirmed to be mutually exclusive. The _BRAF_wt/MSS cancers had a 46% (41 of 90) KRAS mutation rate (Table 1). There was no significant association between a KRAS mutation and CIN in these cancers, with a KRAS mutation present in 43% (32 of 74) CIN cancers, and in 56% (9 of 16) CIN-negative cancers (Table 2). Interestingly of the CIN cancers, the average FAL was significantly lower in KRAS mutant compared to KRAS wild type cancers (p = 0.045) (Table 3).

CpG Island Methylator Phenotype (CIMP) and CIN in BRAFmut/MSS Cancers

CIMP was present in 57% (27 of 47) _BRAF_mut/MSS cancers, whilst only 4% (3 of 79) _BRAF_wt/MSS cancers displayed CIMP (Table 1). Interestingly, a substantial proportion of _BRAF_mut/MSS cancers demonstrated both CIMP and CIN, where 72% of the CIMP high _BRAF_mut/MSS cancers showed CIN (Table 2).

CIN and Clinicopathological Data

The presence of CIN did not affect age of onset and gender distributions (Table 2 in Supplementary data S1). The rate of CIN did not vary regardless of whether the cancer originated from the proximal or distal colon in either cohort, with frequency of allelic loss (FAL) in the _BRAF_mut/MSS cancers at 0.33 and 0.32; and in _BRAF_wt/MSS at 0.33 and 0.42 for proximal and distal cancers respectively (Table 3; Table 2 in Supplementary data S1).

In _BRAF_mut/MSS cancers, CIN significantly associated with advanced stage, occurring in 15 of 17 (88%) stage III/IV cancers compared to 6 of 13 (46%) stage I/II cancers (p = 0.02) (Table 4). Average FAL scores were consistent with this finding, with increasing FAL correlating with advancing stage in _BRAF_mut/MSS CIN cancers (p = 0.03) (Table 3). By contrast, the _BRAF_wt/MSS group had high levels of CIN at both late and early stages of presentation with 27/33 (82%) stage III/IV, and 31/37 (84%) stage I/II CIN cancers (Table 4). This was also demonstrated by similar FAL rates found in the _BRAF_wt/MSS late compared to early stage cancers (Table 3).

Table 4. CIN Positivity and Clinicopathological AJCC Staging of Cancer Cohorts.

Between cohorts, a significantly lower number of _BRAF_mut/MSS CIN positive cancers presented at early stages I/II compared to the _BRAF_wt/MSS cancers with CIN. This was evident overall (p = 0.02) and at individual regions 5q, 17p and 18q (p = 0.02, p = 0.04, p = 0.01 respectively). The rate of CIN positive cancers presenting at advanced stages (III/IV) was similar in both cohorts (Table 4).

Of the tumours with no lymph node involvement, 6/13 (46%) of _BRAF_mut/MSS cancers showed CIN compared to 36/42 (86%) of _BRAF_wt/MSS cancers (p = 0.007) (Table 4). _BRAF_mut/MSS cancers showed a trend for increasing rates of CIN coinciding with spread to at least one lymph node (p = 0.07). Although not significant, these CIN cancers also had a higher rate of metastases at 100%, compared to the number of CIN cancers without metastases at 65%. The _BRAF_wt/MSS cancers showed similar levels of CIN either with or without lymph node spread or metastases (Table 4).

Of the _BRAF_mut/MSS cancers showing co-occurrence of CIMP and CIN, a greater majority presented at late stages (III/IV) (10/16, 63%), compared to those presenting at early stages (I/II) (4/13, 31%), although this did not reach significance (Table 3 in Supplementary data S1).

Cox proportional hazard models were used to investigate overall cancer specific survival of patients with CIN cancers. To account for any confounding effects the variables of BRAF status (_BRAF_mut/MSS and _BRAF_wt/MSS) and stage (I/II and III/IV), along with all statistically significant higher order interactions, were included in the model. No significant relationships between presence of overall CIN and survival were demonstrated in either cohort (_BRAF_mut/MSS p = 0.81, _BRAF_wt/MSS p = 0.22). However, analysis at individual loci revealed the _BRAF_mut/MSS cancers with CIN at 18q and 8p to have significantly worse survival once adjusted for stage (p = 0.02, p<0.05 respectively) (Table 5). _BRAF_wt/MSS cancers showed no difference in survival rates regardless of CIN status at any individual loci (Table 5).

Table 5. Cox Proportional Hazard Models showing Hazard Ratio (HR) of Overall Survival for the interaction between BRAF and CIN for overall and individual genomic regions, with 95% Confidence Interval (CI) adjusted for stage (I/II and III/IV).

Patients with _BRAF_mut/MSS cancers had worse survival rates compared to those with _BRAF_wt/MSS cancers once adjusted for stage, regardless of presence or not of CIN (HR = 3.4, 95% CI [1.65, 6.99], p = 0.001).

Discussion

Colorectal cancers arising via the serrated neoplastic pathway are characterised by a high frequency of BRAF mutation, MSI and CIMP. MSI cancers have been extensively studied, are diploid and confer an excellent prognosis. In the absence of MSI, a BRAF mutation correlates with a particularly poor patient outcome. This study reports that a high level of CIN is evident in BRAF mutant/MSS cancers and this demonstrates at least one mechanism of genomic instability by which these cancers form and progress. CIN was assessed with loss of heterozygosity (LOH) analysis with CIN assigned if at least one of the twelve markers was positive for LOH. The average fractional allelic loss (FAL) [38] which considers the number of markers showing LOH relative to the number of informative markers, was also calculated to further provide a quantifiable measure of the rate of CIN. Both methods found similarly high frequencies of CIN occurring in MSS cancers regardless of BRAF mutation status. Many of the CIN positive BRAF mutant/MSS cancers showed CIMP and thus this study challenges the belief that CIN and CIMP are mutually exclusive. It provides additional evidence that BRAF mutant/MSS cancers are fundamentally different to their MSI counterparts although they both potentially arise from BRAF mutant serrated polyps.

BRAF mutant/MSS cancers are an important subgroup of colorectal cancer due to their association with a particularly poor patient prognosis. However, due to their relatively low prevalence of approximately 5–7% of all colorectal cancers, the molecular mechanisms underlying this poor outcome have not been well studied. By identifying 60 BRAF mutant/MSS cancers from a large series of 1052 patients, the necessary power was attained to examine frequency of CIN in relation to other clinical and molecular variables. It is well established that BRAF mutant cancers predominantly occur in the proximal colon [12], [18], [37]. Certainly, the majority of _BRAF_mut/MSS cancers in this study were from this region (69%), however there was no difference in the rate of CIN found in proximal compared to distal cancers (Table 3; Table 2 in Supplementary data S1).

We have previously shown that BRAF mutant/MSS cancers have a comparable rate of p53 mutation to BRAF wild type/MSS cancers, and this was significantly greater than the p53 mutation rate found in BRAF mutant/MSI cancers. Furthermore, clinicopathological data between the BRAF mutant/MSS cancers and BRAF wild type/MSS cancers, also correlated in terms of a more advanced stage of presentation, comparable ages of onset, and equal gender distribution [37]. The extensive degree of CIN found in BRAF wild type/MSS cancers has been well documented. As BRAF mutant cancers are thought to derive from a serrated polyp, and BRAF mutant cancers that are microsatellite unstable are diploid; it may be postulated that BRAF mutant/MSS cancers would also be CIN-negative. However, based on our previous finding of multiple molecular and clinical similarities between the two MSS subgroups [37], we hypothesised that chromosomal instability would contribute to progression of a substantial proportion of BRAF mutant/MSS cancers. The data presented here support this hypothesis. We further hypothesise that in BRAF mutant cells, the acquisition of CIN is consistent with survival only in the absence of microsatellite instability. To find such similarly high rates of CIN occurring in a large cohort of BRAF mutant/MSS cancers provides further evidence of molecular similarities occurring between these two MSS subgroups although one may progress from serrated polyps, and the other from conventional adenomas.

The presence of CIN was examined over twelve markers spanning four chromosomal regions commonly deleted in colorectal cancer (chromosomes 5q, 8p, 17p and 18q). Although the overall rate of CIN was high in both BRAF mutant and wild type cohorts, examination of individual loci revealed differences between cohorts in the frequency of CIN events per loci. The BRAF mutant/MSS cancers had the greatest degree of CIN on the 8p chromosomal arm, and the least at the 18q region. In contrast, the BRAF wild type/MSS cancers had the highest frequency of instability on the 17p arm, and the least at 8p and 5q. A significant difference in the rate of CIN between the two groups was evident at the 18q chromosomal region with the BRAF wild type/MSS cancers having significantly greater loss at the 18q chromosomal region. Chromosome 18q21.1 harbours the tumour suppressor genes SMAD4, SMAD2 and deleted in colorectal cancer (DCC) genes. The SMAD4/2 gene complex is involved in signal transduction of the TGF-beta pathway and particularly in early stages of tumourigenesis can regulate the expression of target genes resulting in arrested growth and apoptosis [39], [40]. Within the MAPK kinase pathway, aberrant activation of downstream effectors of BRAF have been found to modulate TGF-beta mediated signalling of _SMAD_s [39], [41], where ERK has been implicated in inhibiting SMAD nuclear translocation [42]. As BRAF mutant/MSS cancers have a constitutively active MAPK pathway, loss of the SMAD loci may be redundant and could help to explain the significantly lower rate of deletion found at the 18q region in these compared to BRAF wild type/MSS cancers. DCC is subject to epigenetic silencing [43] and may be targeted by this mechanism rather than deletion in BRAF mutant cancers. Overall, differences in the rate of CIN found at individual regions, indicate that certain loci are preferentially deleted to inactivate the residing tumour suppressor gene according to the genetic background of the cancer.

A loss event of one allele is typically preceded or followed by a mutational event of the second allele at the same region to inactivate the target tumour suppressor gene. This was evident in the present study where in both BRAF mutant and wild type cancers, p53 mutation correlated with CIN at 17p. Average FAL scores were recalculated based on chromosomes 5q, 8p and 18q in order to remove the influence of 17p CIN on p53 mutation rate. Mutation of p53 significantly correlated with a higher FAL score for CIN in BRAF wild type cancers. A similar but non-significant trend was observed in BRAF mutant cancers. These data support our hypothesis that BRAF mutant cancers are not entirely distinct from those arising via the traditional adenoma-carcinoma pathway, but may share multiple molecular features required for tumour progression.

It has been postulated that CIN and CIMP are mutually exclusive [29], [30]. Interestingly, this study clearly demonstrates a high frequency of CIN and CIMP co-occurrence in the BRAF mutant/MSS cancers (72%), which suggests these forms of genetic and epigenetic instabilities can coexist within this molecular background. These findings highlight the necessity of stratification to identify molecular features of important cancer subgroups. An inverse relationship between CIN and CIMP has previously been observed in a predominantly BRAF wild type cancer cohort where only 9 BRAF mutant/MSS cancers were investigated [29]. A low level of CIN has been found in CIMP positive cancers using a genome wide array approach, however BRAF mutational status was not assessed [44]. The lack of association reported in these studies likely reflects the low frequency of CIMP in BRAF wild type cancers. In the present study, CIMP occurred in only 4% of BRAF wild type cancers, precluding the meaningful analysis of CIMP correlating with CIN in this subgroup.

In BRAF mutant/MSS cancers, significantly increasing levels of overall CIN were found with advanced stages of presentation (Tables 3 and 4), suggesting that CIN may contribute to progression of this type of cancer. This pattern was also observed at the chromosomal regions of 18q and 17p (Table 4). A considerable rate of CIN at the 8p region occurring at both early and late stages, may account for this region showing the greatest rate of CIN within this cancer type, and furthermore, may implicate a role for loss of an 8p tumour suppressor gene early in disease development.

In contrast, for overall CIN and at all individual chromosomal regions, the BRAF wild type/MSS cancers showed comparable rates of CIN at early and late stages, and this suggests that CIN is important for initiation or early progression within this cancer subgroup. The higher average fractional allelic loss (FAL) score observed for BRAF wild type versus mutant MSS cancers may reflect the earlier onset of CIN and therefore greater accumulation of CIN events (Table 3). In BRAF mutant/MSS cancers, the rate of CIN increased with increasing lymph node involvement and metastases although this did not reach statistical significance (Table 4). This pattern was not observed in the BRAF wild type cancers. A potentially synergistic relationship between BRAF mutation, CIN and tumour progression requires further investigation.

The mechanism underlying the association of CIN with advanced stage in BRAF mutant/MSS cancers requires elucidation. It is possible that early activation of the MAPK pathway is sufficient to initiate tumourigenesis. Tumour progression may then be promoted by factors underlying CIN such as disruption of mitotic spindle checkpoints and/or centrosome regulators, which may contribute to the particularly aggressive phenotype of this cancer subgroup. For example, overexpression of the Aurora kinases (AURKA and AURKB) have been associated with centrosome amplification and chromosomal mis-segregation [45]. In CRC, overexpression of AURKA is associated with CIN [46], while AURKB correlates with advanced stages [47] and a worse patient outcome in many cancer types [48], [49]. Increased levels of AURKB have also been associated with activation of the MAPK pathway in melanoma, and furthermore application of the BRAF inhibitor vemurafenib, abrogated AURKB expression [50]. This suggests that Aurora kinase family members, particularly AURKB, may play a role in mediating CIN in advanced BRAF mutant/MSS cancers. Additionally, in BRAF mutant/MSS cancers there was a greater rate of co-occurrence of CIMP and CIN in late compared to early stages (63% and 31% respectively) (Table 3 in Supplementary data S1). This may suggest that the compounding effects of both chromosomal and epigenetic instabilities could contribute to the aggressive phenotype and unfavourable outcomes observed in patients with a BRAF mutant/MSS cancer. As expected there were too few CIMP high cancers in the BRAF wild type/MSS cohort to allow for comparisons between cohorts.

Overall patient survival assessed with Cox proportional hazard models, adjusted for stage, indicated that BRAF mutant/MSS cancers with CIN at the 18q and 5q loci, conferred significantly poorer outcomes compared to those BRAF mutant/MSS cancers without CIN at these loci (Table 5). This may imply that CIN at certain genomic regions contributes to the worse survival of BRAF mutant/MSS cancers. CIN at 18q and 8p has previously been associated with a poor prognosis and worse overall survival [31], [51], [52], [53], [54]. However, loss at 18q relating to poor survival has been disputed in other studies [32], [55]. The data presented here supports the hypothesis that a poorer prognosis may be driven by CIN at defined chromosomal locations in MSS cancers bearing a BRAF mutation.

This discussion has eluded to possible molecular interactions that could facilitate the acquisition of CIN in these cancers. In accordance with previous findings [20], [22], the BRAF mutant/MSS cancers conferred poorer survival rates compared to the BRAF wild type cancers, irrespective of the presence or not of CIN (p = 0.001). The molecular mechanisms of how a BRAF mutation on a background of MSS confers such a detrimental patient outcome is currently unclear. BRAF mutant/MSS cancers are an aggressive cancer type and this may be related to the findings that a BRAF mutation correlates with overexpression of the angiogenic factor, VEGFA, in colorectal cancer [56]. Inducible Raf has been found to cause an invasive phenotype dependent on activation of the TGF-beta pathway [57] which develops a more proliferative role in advanced tumourigenesis [39], [40]. Furthermore, activated Raf can contribute to the epithelial-mesenchymal transition [57] which hallmarks the initiation of metastasis and may help to explain the aggressive nature of the BRAF mutant/MSS cancers. In contrast, the BRAF mutation on a background of MSI relates to a comparably favourable patient outcome. This may be due in part to the lower rate of VEGFA found in MSI compared to MSS CRCs [58] which may reduce the level of metastatic spread in MSI cancers. MSI cancers commonly have mutant TGFBR2 [59] which may negate the proliferative effects of the TGF-beta pathway and impair the epithelial to mesenchymal transition in MSI compared to MSS cancers [60].

Recent studies utilizing array based methodologies have found novel regions of loss and/or gain events associating with worse prognoses and disease progression [26], [61]. Furthermore, recent array based investigations have found MSI cancers with DNA copy number changes in unique regions of the genome [61], [62], [63], as well as copy neutral LOH events [64], [65]. By furthering this type of experimental approach, the relationship between MSI and CIN will be better understood.

This study, for the first time, reports a high frequency of CIN in the aggressive BRAF mutant/MSS cancers thought to progress via the serrated neoplastic pathway. Although similarly high rates of CIN were observed in BRAF mutant and wild type MSS cancers, differences existed between the two cohorts regarding the chromosomal regions most commonly targeted, and the stages of presentation at which CIN was most frequent. Additionally, we report on the novel finding of both CIN and CIMP co-occurring in BRAF mutant/MSS cancers, and potentially the combined impact of such may contribute to the detrimental outcome for patients with this cancer type.

Findings from this and our previous study [37], suggest that BRAF mutant/MSS cancers form an aggressive cancer subgroup with distinct clinical and molecular features. Extended investigations into the histology and molecular characteristics of these BRAF mutant/MSS cancers will help to reveal further potential causes of the worse prognosis observed in this cancer subgroup. Importantly, this could ultimately aid the identification of therapeutic targets to treat this aggressive form of colorectal cancer.

Supporting Information

Supplementary data S1

(DOCX)

Funding Statement

This research was supported by a National Health and Medical Research Council grant (NHMRC 442965); Pathology Queensland, Clinical and Statewide Services, Queensland, Australia; and an Australian Postgraduate Award (APA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

• 1.Fearon ER, Vogelstein B (1990) A genetic model for colorectal tumorigenesis. Cell 61: 759–767. [DOI] [PubMed] [Google Scholar]
• 2.Jass JR, Do KA, Simms LA, Iino H, Wynter C, et al. (1998) Morphology of sporadic colorectal cancer with DNA replication errors. Gut 42: 673–679. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Iino H, Jass JR, Simms LA, Young J, Leggett B, et al. (1999) DNA microsatellite instability in hyperplastic polyps, serrated adenomas, and mixed polyps: a mild mutator pathway for colorectal cancer? J Clin Pathol 52: 5–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 4.Jass JR, Biden KG, Cummings MC, Simms LA, Walsh M, et al. (1999) Characterisation of a subtype of colorectal cancer combining features of the suppressor and mild mutator pathways. J Clin Pathol 52: 455–460. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 5.Leggett B, Whitehall V (2010) Role of the serrated pathway in colorectal cancer pathogenesis. Gastroenterology 138: 2088–2100. [DOI] [PubMed] [Google Scholar]
• 6.Rosenberg DW, Yang S, Pleau DC, Greenspan EJ, Stevens RG, et al. (2007) Mutations in BRAF and KRAS differentially distinguish serrated versus non-serrated hyperplastic aberrant crypt foci in humans. Cancer Res 67: 3551–3554. [DOI] [PubMed] [Google Scholar]
• 7.Spring KJ, Zhao ZZ, Karamatic R, Walsh MD, Whitehall VL, et al. (2006) High prevalence of sessile serrated adenomas with BRAF mutations: a prospective study of patients undergoing colonoscopy. Gastroenterology 131: 1400–1407. [DOI] [PubMed] [Google Scholar]
• 8.Toyota M, Ho C, Ahuja N, Jair KW, Li Q, et al. (1999) Identification of differentially methylated sequences in colorectal cancer by methylated CpG island amplification. Cancer Res 59: 2307–2312. [PubMed] [Google Scholar]
• 9.Kambara T, Simms LA, Whitehall VL, Spring KJ, Wynter CV, et al. (2004) BRAF mutation is associated with DNA methylation in serrated polyps and cancers of the colorectum. Gut 53: 1137–1144. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 10.Peyssonnaux C, Eychene A (2001) The Raf/MEK/ERK pathway: new concepts of activation. Biol Cell 93: 53–62. [DOI] [PubMed] [Google Scholar]
• 11.Rajagopalan H, Bardelli A, Lengauer C, Kinzler KW, Vogelstein B, et al. (2002) Tumorigenesis: RAF/RAS oncogenes and mismatch-repair status. Nature 418: 934. [DOI] [PubMed] [Google Scholar]
• 12.Koinuma K, Shitoh K, Miyakura Y, Furukawa T, Yamashita Y, et al. (2004) Mutations of BRAF are associated with extensive hMLH1 promoter methylation in sporadic colorectal carcinomas. Int J Cancer 108: 237–242. [DOI] [PubMed] [Google Scholar]
• 13.Kane MF, Loda M, Gaida GM, Lipman J, Mishra R, et al. (1997) Methylation of the hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res 57: 808–811. [PubMed] [Google Scholar]
• 14.Ionov Y, Peinado MA, Malkhosyan S, Shibata D, Perucho M (1993) Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 363: 558–561. [DOI] [PubMed] [Google Scholar]
• 15.Lothe RA, Peltomaki P, Meling GI, Aaltonen LA, Nystrom-Lahti M, et al. (1993) Genomic instability in colorectal cancer: relationship to clinicopathological variables and family history. Cancer Res 53: 5849–5852. [PubMed] [Google Scholar]
• 16.Sinicrope FA, Rego RL, Halling KC, Foster N, Sargent DJ, et al. (2006) Prognostic impact of microsatellite instability and DNA ploidy in human colon carcinoma patients. Gastroenterology 131: 729–737. [DOI] [PubMed] [Google Scholar]
• 17.Popat S, Hubner R, Houlston RS (2005) Systematic review of microsatellite instability and colorectal cancer prognosis. J Clin Oncol 23: 609–618. [DOI] [PubMed] [Google Scholar]
• 18.Jass JR (2007) Classification of colorectal cancer based on correlation of clinical, morphological and molecular features. Histopathology 50: 113–130. [DOI] [PubMed] [Google Scholar]
• 19.Ogino S, Nosho K, Kirkner GJ, Kawasaki T, Meyerhardt JA, et al. (2009) CpG island methylator phenotype, microsatellite instability, BRAF mutation and clinical outcome in colon cancer. Gut 58: 90–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Samowitz WS, Sweeney C, Herrick J, Albertsen H, Levin TR, et al. (2005) Poor survival associated with the BRAF V600E mutation in microsatellite-stable colon cancers. Cancer Res 65: 6063–6069. [DOI] [PubMed] [Google Scholar]
• 21.Kakar S, Deng G, Sahai V, Matsuzaki K, Tanaka H, et al. (2008) Clinicopathologic characteristics, CpG island methylator phenotype, and BRAF mutations in microsatellite-stable colorectal cancers without chromosomal instability. Arch Pathol Lab Med 132: 958–964. [DOI] [PubMed] [Google Scholar]
• 22.Ogino S, Shima K, Meyerhardt J, McCleary N, Ng K, et al. (2012) Predictive and Prognostic Roles of BRAF Mutation in Stage III Colon Cancer: Results from Intergroup Trial CALGB 89803. Clin Cancer Res 18(3): 890–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 23.Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, et al. (1988) Genetic alterations during colorectal-tumor development. N Engl J Med 319: 525–532. [DOI] [PubMed] [Google Scholar]
• 24.Goel A, Arnold CN, Niedzwiecki D, Chang DK, Ricciardiello L, et al. (2003) Characterization of sporadic colon cancer by patterns of genomic instability. Cancer Res 63: 1608–1614. [PubMed] [Google Scholar]
• 25.O’Brien MJ, Yang S, Mack C, Xu H, Huang CS, et al. (2006) Comparison of microsatellite instability, CpG island methylation phenotype, BRAF and KRAS status in serrated polyps and traditional adenomas indicates separate pathways to distinct colorectal carcinoma end points. Am J Surg Pathol 30: 1491–1501. [DOI] [PubMed] [Google Scholar]
• 26.Watanabe T, Kobunai T, Yamamoto Y, Matsuda K, Ishihara S, et al. (2012) Chromosomal Instability (CIN) Phenotype, CIN High or CIN Low, Predicts Survival for Colorectal Cancer. J Clin Oncol 30(18): 2256–2264. [DOI] [PubMed] [Google Scholar]
• 27.Chang SC, Lin JK, Yang SH, Wang HS, Li AF, et al. (2006) Relationship between genetic alterations and prognosis in sporadic colorectal cancer. Int J Cancer 118: 1721–1727. [DOI] [PubMed] [Google Scholar]
• 28.Walther A, Houlston R, Tomlinson I (2008) Association between chromosomal instability and prognosis in colorectal cancer: a meta-analysis. Gut 57: 941–950. [DOI] [PubMed] [Google Scholar]
• 29.Goel A, Nagasaka T, Arnold CN, Inoue T, Hamilton C, et al. (2007) The CpG island methylator phenotype and chromosomal instability are inversely correlated in sporadic colorectal cancer. Gastroenterology 132: 127–138. [DOI] [PubMed] [Google Scholar]
• 30.Ogino S, Kawasaki T, Kirkner GJ, Ohnishi M, Fuchs CS (2007) 18q loss of heterozygosity in microsatellite stable colorectal cancer is correlated with CpG island methylator phenotype-negative (CIMP-0) and inversely with CIMP-low and CIMP-high. BMC Cancer 7: 72. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 31.Halling KC, French AJ, McDonnell SK, Burgart LJ, Schaid DJ, et al. (1999) Microsatellite instability and 8p allelic imbalance in stage B2 and C colorectal cancers. J Natl Cancer Inst 91: 1295–1303. [DOI] [PubMed] [Google Scholar]
• 32.Ogino S, Nosho K, Irahara N, Shima K, Baba Y, et al. (2009) Prognostic significance and molecular associations of 18q loss of heterozygosity: a cohort study of microsatellite stable colorectal cancers. J Clin Oncol 27: 4591–4598. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 33.American Joint Committee on Cancer (2010) AJCC Cancer Staging Handbook. Edge S, Byrd D, Compton C, Fritz A, Greene F, et al., editors. New York: Springer-Verlag.
• 34.Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, et al. (1998) A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 58: 5248–5257. [PubMed] [Google Scholar]
• 35.Nagasaka T, Koi M, Kloor M, Gebert J, Vilkin A, et al.. (2008) Mutations in both KRAS and BRAF may contribute to the methylator phenotype in colon cancer. Gastroenterology 134: 1950–1960, 1960 e1951. [DOI] [PMC free article] [PubMed]
• 36.Weisenberger DJ, Siegmund KD, Campan M, Young J, Long TI, et al. (2006) CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat Genet 38: 787–793. [DOI] [PubMed] [Google Scholar]
• 37.Bond CE, Umapathy A, Ramsnes I, Greco SA, Zhao ZZ, et al. (2012) P53 mutation is common in microsatellite stable, BRAF mutant colorectal cancers. Int J Cancer 130(7): 1567–1576. [DOI] [PubMed] [Google Scholar]
• 38.Vogelstein B, Fearon ER, Kern SE, Hamilton SR, Preisinger AC, et al. (1989) Allelotype of colorectal carcinomas. Science 244: 207–211. [DOI] [PubMed] [Google Scholar]
• 39.Derynck R, Akhurst RJ, Balmain A (2001) TGF-beta signaling in tumor suppression and cancer progression. Nat Genet 29: 117–129. [DOI] [PubMed] [Google Scholar]
• 40.Siegel PM, Massague J (2003) Cytostatic and apoptotic actions of TGF-beta in homeostasis and cancer. Nat Rev Cancer 3: 807–821. [DOI] [PubMed] [Google Scholar]
• 41.Javelaud D, Mauviel A (2005) Crosstalk mechanisms between the mitogen-activated protein kinase pathways and Smad signaling downstream of TGF-beta: implications for carcinogenesis. Oncogene 24: 5742–5750. [DOI] [PubMed] [Google Scholar]
• 42.Kretzschmar M, Doody J, Timokhina I, Massague J (1999) A mechanism of repression of TGFbeta/Smad signaling by oncogenic Ras. Genes Dev 13: 804–816. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 43.Derks S, Bosch LJ, Niessen HE, Moerkerk PT, van den Bosch SM, et al. (2009) Promoter CpG island hypermethylation- and H3K9me3 and H3K27me3-mediated epigenetic silencing targets the deleted in colon cancer (DCC) gene in colorectal carcinogenesis without affecting neighboring genes on chromosomal region 18q21. Carcinogenesis 30: 1041–1048. [DOI] [PubMed] [Google Scholar]
• 44.Cheng YW, Pincas H, Bacolod MD, Schemmann G, Giardina SF, et al. (2008) CpG island methylator phenotype associates with low-degree chromosomal abnormalities in colorectal cancer. Clin Cancer Res 14: 6005–6013. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Dar AA, Goff LW, Majid S, Berlin J, El-Rifai W (2010) Aurora kinase inhibitors–rising stars in cancer therapeutics? Mol Cancer Ther 9: 268–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46.Baba Y, Nosho K, Shima K, Irahara N, Kure S, et al. (2009) Aurora-A expression is independently associated with chromosomal instability in colorectal cancer. Neoplasia 11: 418–425. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 47.Katayama H, Ota T, Jisaki F, Ueda Y, Tanaka T, et al. (1999) Mitotic kinase expression and colorectal cancer progression. J Natl Cancer Inst 91: 1160–1162. [DOI] [PubMed] [Google Scholar]
• 48.Lin ZZ, Jeng YM, Hu FC, Pan HW, Tsao HW, et al. (2010) Significance of Aurora B overexpression in hepatocellular carcinoma. Aurora B Overexpression in HCC. BMC Cancer 10: 461. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 49.Kurai M, Shiozawa T, Shih HC, Miyamoto T, Feng YZ, et al. (2005) Expression of Aurora kinases A and B in normal, hyperplastic, and malignant human endometrium: Aurora B as a predictor for poor prognosis in endometrial carcinoma. Hum Pathol 36: 1281–1288. [DOI] [PubMed] [Google Scholar]
• 50.Bonet C, Giuliano S, Ohanna M, Bille K, Allegra M, et al. (2012) Aurora B is regulated by the Mitogen-activated Protein Kinase/Extracellular Signal-regulated Kinase MAPK/ERK signaling pathway and is a valuable potential target in melanoma cells. J Biol Chem 287(35): 29887–29898. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 51.Diep CB, Thorstensen L, Meling GI, Skovlund E, Rognum TO, et al. (2003) Genetic tumor markers with prognostic impact in Dukes’ stages B and C colorectal cancer patients. J Clin Oncol 21: 820–829. [DOI] [PubMed] [Google Scholar]
• 52.Lanza G, Matteuzzi M, Gafa R, Orvieto E, Maestri I, et al. (1998) Chromosome 18q allelic loss and prognosis in stage II and III colon cancer. Int J Cancer 79: 390–395. [DOI] [PubMed] [Google Scholar]
• 53.Watanabe T, Wu TT, Catalano PJ, Ueki T, Satriano R, et al. (2001) Molecular predictors of survival after adjuvant chemotherapy for colon cancer. N Engl J Med 344: 1196–1206. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 54.Tanaka T, Watanabe T, Kitayama J, Kanazawa T, Kazama Y, et al. (2009) Chromosome 18q deletion as a novel molecular predictor for colorectal cancer with simultaneous hepatic metastasis. Diagn Mol Pathol 18: 219–225. [DOI] [PubMed] [Google Scholar]
• 55.Bertagnolli MM, Redston M, Compton CC, Niedzwiecki D, Mayer RJ, et al. (2011) Microsatellite Instability and Loss of Heterozygosity at Chromosomal Location 18q: Prospective Evaluation of Biomarkers for Stages II and III Colon Cancer–A Study of CALGB 9581 and 89803. J Clin Oncol 29: 3153–3162. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 56.Vlajnic T, Andreozzi MC, Schneider S, Tornillo L, Karamitopoulou E, et al. (2011) VEGFA gene locus (6p12) amplification identifies a small but highly aggressive subgroup of colorectal cancer [corrected] patients. Mod Pathol 24: 1404–1412. [DOI] [PubMed] [Google Scholar]
• 57.Lehmann K, Janda E, Pierreux CE, Rytomaa M, Schulze A, et al. (2000) Raf induces TGFbeta production while blocking its apoptotic but not invasive responses: a mechanism leading to increased malignancy in epithelial cells. Genes Dev 14: 2610–2622. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 58.Hollingsworth SJ, Drye ER, Tou SI, Boulos PB (2005) Expression of angiogenic VEGF-A (soluble isoforms 121, 165) and lymphangiogenic VEGF-C in colorectal cancers with micro-satellite instability. J Surg Oncol 92: 317–325. [DOI] [PubMed] [Google Scholar]
• 59.Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, et al. (1995) Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instability. Science 268: 1336–1338. [DOI] [PubMed] [Google Scholar]
• 60.Pino MS, Kikuchi H, Zeng M, Herraiz MT, Sperduti I, et al. (2010) Epithelial to mesenchymal transition is impaired in colon cancer cells with microsatellite instability. Gastroenterology 138: 1406–1417. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 61.Andersen CL, Lamy P, Thorsen K, Kjeldsen E, Wikman F, et al.. (2011) Frequent genomic loss at chr16p13.2 is associated with poor prognosis in colorectal cancer. Int J Cancer 129(*): 1848–1858. [DOI] [PubMed]
• 62.Lassmann S, Weis R, Makowiec F, Roth J, Danciu M, et al. (2007) Array CGH identifies distinct DNA copy number profiles of oncogenes and tumor suppressor genes in chromosomal- and microsatellite-unstable sporadic colorectal carcinomas. J Mol Med 85: 293–304. [DOI] [PubMed] [Google Scholar]
• 63.Trautmann K, Terdiman JP, French AJ, Roydasgupta R, Sein N, et al. (2006) Chromosomal instability in microsatellite-unstable and stable colon cancer. Clin Cancer Res 12: 6379–6385. [DOI] [PubMed] [Google Scholar]
• 64.Andersen CL, Wiuf C, Kruhoffer M, Korsgaard M, Laurberg S, et al. (2007) Frequent occurrence of uniparental disomy in colorectal cancer. Carcinogenesis 28: 38–48. [DOI] [PubMed] [Google Scholar]
• 65.Melcher R, Hartmann E, Zopf W, Herterich S, Wilke P, et al. (2011) LOH and copy neutral LOH (cnLOH) act as alternative mechanism in sporadic colorectal cancers with chromosomal and microsatellite instability. Carcinogenesis 32: 636–642. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary data S1

(DOCX)