Colorectal cancer has been described in terms of genetic instability selectively affecting either microsatellite sequences (MIN) or chromosome number and structure (CIN). A subgroup with apparently stable, near-diploid chromosomes and stable microsatellites (MACS) also exists. These distinctions are important, partly because of their value in highlighting different pathways of carcinogenesis, and partly because of their direct relevance to prognosis. Study of early-onset cancer has often proved a fruitful resource for the identification of the nature and function of cancer susceptibility genes. In a study of colorectal cancer with stable microsatellite DNA, we describe 22 early-onset tumours (mean age=33), compared with 16 late-onset tumours (mean age=68). Both groups contained carcinomas with the MACS phenotype, characterized by near diploid DNA content, as defined by flow cytometry, and minimal chromosome arm deletion or amplification (six or less events per genome), determined by comparative genomic hybridization (CGH). Minimal chromosome imbalance correlated strongly with diploid DNA content (P<0.001). The proportion of MACS cancers was significantly greater in early-onset as compared to late-onset tumours (64 vs 13%, P=0.005). Of the chromosome arm imbalances commonly observed in late-onset tumours, only 18q− was observed more than twice amongst the 14 early-onset MACS tumours. Seventy-nine per cent of these MACS tumours were located in the distal colon, and 69% were at advanced clinico-pathological stages (with lymph node or distant metastasis). A positive family history of colorectal or other cancers was elicited in seven patients in the MACS early-onset group, and one additional patient in this group had a metachronous ovarian cancer. The results suggest that MACS cancer may have a genetic basis different from either MIN or CIN, and further studies of these cancers may lead to discovery of new mechanisms of colorectal carcinogenesis and cancer susceptibility.
Descriptive studies of the molecular pathology of human colorectal cancer indicate the existence of several discrete patterns of genomic alteration (Lengauer et al., 1998). The majority of these tumours have aneuploid karyotypes, usually associated with deficiency of p53 function (Carder et al., 1993; Ried et al., 1996; Cottu et al., 1996; De Angelis et al., 1999). The cells of tumours in this group show persistent chromosome instability (CIN) in culture, and clonal divergence of chromosome structure and function during growth in vivo (Lengauer et al., 1997). A second group of colorectal cancers are near diploid, but show substantial microsatellite instability (MIN), due invariably to deficiency in mismatch repair (Aaltonen et al., 1993; Ionov et al., 1993; Thibodeau et al., 1993; Peltomaki et al., 1993; Liu et al., 1996). Such tumours often but not always express normal p53 and display relatively few chromosome imbalances as detected by comparative genomic hybridization (CGH) (Ionov et al., 1993; Remvikos et al., 1995; Schlegel et al., 1995; Cottu et al., 1996; Konishi et al., 1996; Olschwang et al., 1997; Leung et al., 2000). Recently, we have drawn attention to a further group associated with near-diploid DNA content, few examples of chromosome imbalances, and stable microsatellite DNA (Georgiades et al., 1999), here called the microsatellite and chromosome stable (MACS) group. These differences in patterns of genomic change are significantly associated with divergent biological and behavioural features. MIN tumours arise predominantly in the proximal colon, are histologically mucinous or poorly differentiated, and are associated with a better than average prognosis (Lothe et al., 1993; Bubb et al., 1996; Ho et al., 2000; Gryfe et al., 2000). In contrast, CIN and MACS tumours, taken together, tend to be commoner in the distal colon and rectum, are moderately well differentiated and carry a poorer prognosis than MIN tumours, even after adjustment for stage of progression.
It is probable that these differences reflect the different routes whereby the malignant phenotype is developed and maintained in these tumours. Carcinogenesis requires a source of somatic mutation, coupled with acquisition of the ability of the mutant cells to survive and proliferate. Inactivation of the mismatch repair genes hMSH2 or hMLH1 permits cells sustaining nucleotide mismatch or abnormal alkylation events to survive at apoptosis checkpoints, despite their inability to effect accurate repair, and hence is critical for carcinogenesis by this route (Hawn et al., 1995; Toft et al., 1999). Similarly, cells without functional p53 may avoid apoptosis following double-strand DNA breaks that would otherwise be lethal (Merritt et al., 1994; Clarke et al., 1994). In both situations, the result is survival of clones of cells in which the mutation rate may be enhanced many-fold, either spontaneously or in response to extraneous genotoxic stimuli of the appropriate type. Very little is known, however, of the identity of other genes, defective function of which might account for the characteristic phenotypes of CIN and MACS cancers. One attractive candidate was the spindle checkpoint gene BUB1, deficiency of which might produce the aneuploid karyotypes of CIN tumours (Cahill et al., 1998). BUB1 mutations are seldom found in CIN tumours, however, and there is at present no information on the genes or even the types or existence of genomic defects that might characterize MACS cancers.
In this paper we compare the frequency of the MACS phenotype in patients in whom onset of colorectal cancer was recognized early (less than 45 years) or later in life (above 50 years). There is substantial precedent from the study of MIN tumours that early-onset colorectal cancer is likely to be associated with heterozygous germline inactivation of the cancer susceptibility genes hMSH2, hMLH1 and hMSH6 (Liu et al., 1995; Farrington et al., 1998; Chan et al., 1999). We argued that, despite their relative rarity, mismatch-repair competent cancers of early onset might prove of value in the search for as yet unidentified genetic lesions responsible for the initiation of CIN and MACS tumours. The upper age limit of 45 for definition of early-onset cancer was selected because our previous studies had shown that below this age there are clear differences in germline incidence of cancer-related mutations (Liu et al., 1995) and in overall colon cancer incidence between world populations (Yuen et al., 1997) relative to the (much commoner) late-onset tumours.
Fresh tumour samples were collected in a collaborative study of colorectal cancer at Hong Kong and Edinburgh between 1991 and 1997. All had been tested for microsatellite instability using methods and criteria described previously (Farrington et al., 1998; Chan et al., 1999). We identified for study 12 colorectal cancers from young Chinese patients (mean age 32.9; range 26–40) from Hong Kong and 10 from young Caucasian patients (mean age 33.0; range 22–44) from Edinburgh on the basis solely of their absence of microsatellite instability. For comparison, we also randomly selected 16 microsatellite-stable tumours from old Chinese patients (mean age 67.5; range 52–86) and five early-onset microsatellite-unstable colorectal cancers from Hong Kong. Late-onset microsatellite-stable tumours from the Edinburgh series have already been reported (Georgiades et al., 1999).
Taken together there were 22 early-onset microsatellite stable colorectal cancers. Seven were from male patients and 15 from female. Six tumours were from sites proximal to the splenic flexure and 16 were distal. Overall seven were Dukes' B, six were Dukes' C and eight were Dukes' D (Dukes' stage in one case was unavailable). Two were well-differentiated, 15 were moderately differentiated and five were poorly differentiated.
For the 16 late-onset tumours, 10 were from male patients and six were from female patients. The age of the patients ranged from 52 to 86 years. Two tumours were from proximal and 14 from distal sites. Overall, two were Dukes' A, six were Dukes' B, five were Dukes' C and three were Dukes' D. Histologically, one was well differentiated, 14 were moderately differentiated and one was poorly differentiated.
Consistent with reports by others (Remvikos et al., 1995; Schlegel et al., 1995), all five MIN colorectal cancers showed few chromosomal imbalances, as detected by comparative genomic hybridization (CGH). The results are listed in Figure 1a. The range and detail of chromosomal aberrations in the late-onset tumours are also closely similar to those described in previous reports (Ried et al., 1996; De Angelis et al., 1999; Georgiades et al., 1999; Curtis et al., 2000) and are listed in Figure 1c. Most showed widespread gross chromosomal aberrations with the number of chromosome arm gains or losses ranging from 0 to 22 with a mean of 10.7. The most frequent changes were 1p−, 7p+, 7q+, 8p−, 8q+, 13q+, 18q−, 20p+ and 20q+ (Table 1).
The chromosomal aberrations in the microsatellite-stable early-onset tumours are listed in Figure 1b. The overall numbers of chromosome arm gains and losses were much lower than in the late-onset group, ranging from 0 to 9 with a mean of 3.8. This difference is statistically significant (P<0.001, Mann–Whitney test). Certain chromosome arm imbalances have been observed repeatedly in earlier studies on late-onset colorectal cancer (Ried et al., 1996; De Angelis et al., 1999; Georgiades et al., 1999; Curtis et al., 2000), and have been confirmed in the late-onset Chinese cases reported above. We compared the early and late onset cancers at each of these chromosome arms. 1p−, 7p+, 7q+, 13q+, 20p+ and 20q+ were found at significantly lower frequency in the early-onset tumours (Table 1). The trend towards lower frequency of 8p− imbalance in early-onset tumours did not reach statistical significance. 8q+ and 18q− in the early-onset tumours occurred at frequencies that did not differ significantly from that in late-onset cancers.
We then analysed the patterns of chromosomal aberrations seen in individual tumours from the young and the old patients. Our previous analysis of late-onset colorectal tumours from Caucasian patients (Georgiades et al., 1999) and the present analysis of MIN colorectal cancers served as a reference point. Overall, the microsatellite stable tumours in the present series can be classified into two distinct subgroups: those with widespread (eight or more) gross chromosomal, and those with relatively few (six or less) chromosomal aberrations. On the basis of the CGH pattern alone, this latter group cannot be distinguished from MIN tumours. Using these criteria, four out of 16 (25%) microsatellite-stable late-onset tumours were chromosome stable; in contrast, 15 out of 22 (68%) microsatellite-stable early-onset tumours were chromosome stable (Table 2). This difference in the proportion of chromosome stable tumours between the young and the old patients was statistically significant (P=0.021). There are no significant correlations between chromosomal stability and patient gender or tumour location, stage or differentiation.
DNA ploidy as determined by flow cytometry is shown in Table 2. The majority of the late-onset, microsatellite-stable colorectal cancers (14/16, 88%) were DNA aneuploid. In contrast, more than two-thirds of the early-onset microsatellite stable tumours (16/22, 73%) showed a diploid or near diploid DNA content, a statistically significant difference (P<0.001). As expected, few tumours (only three of 19) classified as chromosome stable by CGH proved to have aneuploid DNA content by flow cytometry, in contrast to almost all (17 out of 19) that were classified as chromosome unstable (P<0.001). There were no significant correlations between DNA ploidy and patient gender or tumour location, stage or differentiation.
Early-onset colorectal cancer occurs more often in southern Chinese populations than in the West but in both it is much rarer than the late-onset disease (Yuen et al., 1997). Despite this, in previous studies of both young and old patient groups, we identified no difference between Chinese and Scottish patients in the incidence of MIN cancers (Liu et al., 1995; Farrington et al., 1998; Chan et al., 1999). Similarly, in the microsatellite-stable, early onset cancers that are the objects of this study, there are no significant differences between Chinese and Scottish populations in the percentages of tumours with chromosomal instability. This is the case whether such instability is defined by DNA ploidy or further refined by the inclusion of CGH data. The single outstanding distinction observed here is the high proportion of MACS tumours amongst early as opposed to late-onset cancers.
The data presented here confirm the distinctive properties of MACS tumours, described earlier in a series of late-onset Caucasian patients (Georgiades et al., 1999). Unlike MIN tumours, these cancers are predominantly distal in location, and the majority have the moderately well differentiated histology typical of adenocarcinomas at this site. Microsatellite instability is consistently absent at sensitive reporter sequences. Unlike CIN tumours, however, MACS carcinomas are usually DNA diploid and on CGH show few examples of chromosome arm imbalances (here six or less per genome). In this series, diploid DNA content and near-normal CGH patterns are highly concordant, as reported by others (Miyazaki et al., 1999), but there were some exceptions. Three tumours were classified as aneuploid on the basis of a small non-diploid peak evident by flow cytometry, but CGH patterns were near-normal, presumably because the number of unbalanced chromosomes was insufficient to be detected by this method. Two further tumours were excluded from the MACS group despite near-diploid DNA content, because their CGH pattern included many imbalances, with gains and losses approximately balanced. When MACS tumours were defined on the basis of both near-diploid DNA content and minimal CGH changes, however, a significant difference persisted between early and late-onset tumours in the proportion of cases found amongst microsatellite-stable cancers (Table 2). Similarly, a significant difference between early and late-onset tumours was sustained (P<0.01) when these relatively small numbers were augmented by addition of our previously reported data on the incidence of late-onset MACS tumours in Scottish patients (four out of 17) studied by identical methods (Georgiades et al., 1999).
It remains possible that MACS cancers evolve into CIN cancers, rather than representing a biologically separate group. In an analysis of the karyotypic abnormalities in colorectal cancer, Muleris et al. (1988, 1990) identified evidence for progression from a monosomic type, in which single copies of 17p and 18q were frequently lost, to a more complex pattern, apparently attained through polypoidization, with multiple chromosome arm gains and losses. In this regard, it is of interest that 18q− was the single chromosome arm abnormality that was observed more than twice amongst the 16 MACS tumours described here, being found four times in all. However, if evolution from MACS to CIN carcinomas does occur, it is clearly either rare or late, as even this small series includes several examples of MACS tumours at advanced clinicopathological stages: more than half had local lymph node or blood borne metastasis.
Of the 14 early-onset MACS cancer patients described here, seven had a family history of cancer, including colorectal cancer in five, two of them satisfying Bethesda criterion 3 (Figure 1 and Table 3). Another patient had metachronous colorectal and ovarian cancer, thus conforming to Bethesda criterion 2. This strongly suggests germ-line transmission of genetic susceptibility for MACS tumours amongst the early-onset patients. There is at present no indication of what the transmitted germ-line abnormalities might be. One theoretical possibility is that these tumours arise from cells with enhanced carcinogenic potential but no fundamental genomic instability. It appears more probable, by analogy with MIN and CIN, that MACS tumours also possess a form of genetic instability, albeit undetectable by the methods used. Potential mechanisms might include altered patterns of methylation, point mutations, balanced translocations, or small deletions affecting non-microsatellite sequences. Further study of families of this sort may therefore help elucidate the nature of genetic defects that confer susceptibility to development of this subclass of colorectal cancer.
Aaltonen LA, Peltomaki P, Leach FS, Sistonen P, Pylkkanen L, Mecklin JP, Jarvinen H, Powell SM, Jen J, Hamilton SR, Petersen GM, Kinzler KW, Vogelstein B, de la Chapelle A . 1993 Science 260: 812–816
Bubb VJ, Curtis LJ, Cunningham C, Dunlop MG, Carothers AD, Morris RG, White S, Bird CC, Wyllie AH . 1996 Oncogene 12: 2641–2649
Cahill DP, Lengauer C, Yu J, Riggins GJ, Willson JK, Markowitz SD, Kinzler KW, Vogelstein B . 1998 Nature 392: 300–303
Carder P, Wyllie AH, Purdie CA, Morris RG, White S, Piris J, Bird CC . 1993 Oncogene 8: 1397–1401
Chan TL, Yuen ST, Chung LP, Ho JW, Kwan KY, Chan AS, Ho JC, Leung SY, Wyllie AH . 1999 J. Natl. Cancer Inst. 91: 1221–1226
Clarke AR, Gledhill S, Hooper ML, Bird CC, Wyllie AH . 1994 Oncogene 9: 1767–1773
Cottu PH, Muzeau F, Estreicher A, Flejou JF, Iggo R, Thomas G, Hamelin R . 1996 Oncogene 13: 2727–2730
Curtis LJ, Georgiades IB, White S, Bird CC, Harrison DJ, Wyllie AH . 2000 J. Pathol. 192: 440–445
De Angelis PM, Clausen OP, Schjolberg A, Stokke T . 1999 Br. J. Cancer 80: 526–535
Farrington SM, Lin-Goerke J, Ling J, Wang Y, Burczak JD, Robbins DJ, Dunlop MG . 1998 Am. J. Hum. Genet. 63: 749–759
Georgiades IB, Curtis LJ, Morris RM, Bird CC, Wyllie AH . 1999 Oncogene 18: 7933–7940
Gryfe R, Kim H, Hsieh ET, Aronson MD, Holowaty EJ, Bull SB, Redston M, Gallinger S . 2000 N. Engl. J. Med. 342: 69–77
Hawn MT, Umar A, Carethers JM, Marra G, Kunkel TA, Boland CR, Koi M . 1995 Cancer Res. 55: 3721–3725
Ho JW, Yuen ST, Chung LP, Kwan KY, Chan TL, Leung SY, Chan AS, Tse C, Lam PW, Luk IS . 2000 Int. J. Cancer 89: 356–360
Ionov Y, Peinado MA, Malkhosyan S, Shibata D, Perucho M . 1993 Nature 363: 558–561
Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D . 1992 Science 258: 818–821
Konishi M, Kikuchi-Yanoshita R, Tanaka K, Muraoka M, Onda A, Okumura Y, Kishi N, Iwama T, Mori T, Koike M, Ushio K, Chiba M, Nomizu S, Konishi F, Utsunomiya J, Miyaki M . 1996 Gastroenterology 111: 307–317
Lengauer C, Kinzler KW, Vogelstein B . 1997 Nature 386: 623–627
Lengauer C, Kinzler KW, Vogelstein B . 1998 Nature 396: 643–649
Leung SY, Yuen ST, Chan TL, Chan AS, Ho JW, Kwan K, Fan YW, Hung KN, Chung LP, Wyllie AH . 2000 Oncogene 19: 4079–4083
Liu B, Farrington SM, Petersen GM, Hamilton SR, Parsons R, Papadopoulos N, Fujiwara T, Jen J, Kinzler KW, Wyllie AH, Vogelstein B, Dunlop MG . 1995 Nat. Med. 1: 348–352
Liu B, Parsons R, Papadopoulos N, Nicolaides NC, Lynch HT, Watson P, Jass JR, Dunlop M, Wyllie A, Peltomaki P, de la Chapelle A, Hamilton SR, Vogelstein B, Kinzler KW . 1996 Nat. Med. 2: 169–174
Lothe RA, Peltomaki P, Meling GI, Aaltonen LA, Nystrom-Lahti M, Pylkkanen L, Heimdal K, Andersen TI, Moller P, Rognum TO, Fossa SD, Haldorsen T, Langmark F, Brogger A, de la Chapelle A, Borresen AL . 1993 Cancer Res. 53: 5849–5852
Merritt AJ, Potten CS, Kemp CJ, Hickman JA, Balmain A, Lane DP, Hall PA . 1994 Cancer Res. 54: 614–617
Miyazaki M, Furuya T, Shiraki A, Sato T, Oga A, Sasaki K . 1999 Cancer Res. 59: 5283–5285
Muleris M, Salmon RJ, Dutrillaux B . 1988 Cancer Genet. Cytogenet. 32: 43–50
Muleris M, Salmon RJ, Dutrillaux B . 1990 Cancer Genet. Cytogenet. 46: 143–156
Olschwang S, Hamelin R, Laurent-Puig P, Thuille B, De Rycke Y, Li YJ, Muzeau F, Girodet J, Salmon RJ, Thomas G . 1997 Proc. Natl. Acad. Sci. USA 94: 12122–12127
Peltomaki P, Lothe RA, Aaltonen LA, Pylkkanen L, Nystrom-Lahti M, Seruca R, David L, Holm R, Ryberg D, Haugen A, Brogger A, Borresen AL, de la Chapelle A . 1993 Cancer Res. 53: 5853–5855
Remvikos Y, Vogt N, Muleris M, Salmon RJ, Malfoy B, Dutrillaux B . 1995 Genes Chromosomes Cancer 12: 272–276
Ried T, Knutzen R, Steinbeck R, Blegen H, Schrock E, Heselmeyer K, du Manoir S, Auer G . 1996 Genes Chromosomes Cancer 15: 234–245
Schlegel J, Stumm G, Scherthan H, Bocker T, Zirngibl H, Ruschoff J, Hofstadter F . 1995 Cancer Res. 55: 6002–6005
Thibodeau SN, Bren G, Schaid D . 1993 Science 260: 816–819
Toft NJ, Winton DJ, Kelly J, Howard LA, Dekker M, te Riele H, Arends MJ, Wyllie AH, Margison GP, Clarke AR . 1999 Proc. Natl. Acad. Sci. USA 96: 3911–3915
Yuen ST, Chung LP, Leung SY, Luk IS, Chan SY, Ho JC, Ho JW, Wyllie AH . 1997 Br. J. Cancer 76: 1610–1616
This work was supported by the Research Grants Council of the Hong Kong Special Administrative Region (HKU 7330/00M), the Committee on Research and Conference Grant from the University of Hong Kong (10202796), a donation from the Hong Kong Society of Gastroenterology, grants from the Cancer Research Campaign (SP2326/0101) and Scottish Health Department (K/MRS/50/C2417, C2723), and the Scottish Hospitals Endowments Research Trust (#1263).
About this article
A Summary of the Fight Colorectal Cancer Working Meeting: Exploring Risk Factors and Etiology of Sporadic Early-Age Onset Colorectal Cancer
Molecular Oncology (2019)
Gastrointestinal Endoscopy (2018)
Current Gastroenterology Reports (2018)
Reply to ‘Comment on ‘Distinct clinical outcomes of two CIMP-positive colorectal cancer subtypes based on a revised CIMP classification system”
British Journal of Cancer (2018)