Article Text

Download PDFPDF

Proximal and progressive: adenomas in HNPCC
  1. D T Bishop
  1. ICRF Genetic Epidemiology Division, ICRF Clinical Centre in Leeds, Cancer Genetics Building, St James's University Hospital, Beckett Street, Leeds LS9 7TF, UK; t.bishop{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Characteristics of hereditary non-polyposis colorectal cancer adenomas

In this issue of Gut, Rijcken and colleagues1 compare adenomas found in hereditary non-polyposis colorectal cancer (HNPCC) patients with those found in patients without a family history of colorectal cancer [see 382]. HNPCC is an autosomal dominant syndrome associated with an increased risk of cancer at a number of anatomical sites but most noticeably of the bowel and endometrium. Some of the HNPCC adenomas examined in this study were derived from patients in families with known germline mutations in a DNA mismatch repair (MMR) gene (this topic has been reviewed recently in Gut by Wheeler and colleagues2 ). Colorectal carcinoma in patients with germline MMR mutations exhibit failure of DNA MMR as a result of loss of expression of both copies of the MMR gene. Failure of DNA repair implies that particular DNA sequences, if mutated somatically, are less likely to be corrected.2

There are three notable features of carcinomas arising as a result of HNPCC: (i) an excess of proximal lesions, (ii) an excess of mucinous and undifferentiated lesions, and (iii) evidence of rapid progression rates for some lesions.3, 4 Studies of HNPCC adenomas have been limited and therefore the study reported here is particularly welcome. The study focuses on the comparison of HNPCC adenomas and adenomas derived from subjects without a family history of colorectal cancer (termed “sporadic” in this publication). The authors reported an increased proportion of proximal HNPCC adenomas and that larger proximal HNPCC adenomas were more likely to be highly dysplastic than “sporadic” adenomas. Furthermore, HNPCC proximal adenomas were more likely to be dysplastic than HNPCC distal adenomas. These observations are therefore novel and are consistent with previous reports of increased numbers of proximal cancers in HNPCC by suggesting that the adenomas are more likely to be right sided and also that proximal HNPCC adenomas are more likely to progress than their distal counterparts.

Studies such as that of Rijcken et al are important for two distinct reasons. The first is that our information on the natural history of HNPCC is so limited, and in terms of evidence based approaches to clinical management and/or intervention for HNPCC patients, we need to improve our overall knowledge. There are still insignificant numbers of predisposed individuals identified (and those that are known are distributed over many centres) and under regular surveillance to allow collection of such information and material. The second reason is that by making comparisons between HNPCC and “sporadic” neoplasia, we might learn more about the adenoma-carcinoma process. It is now over 10 years since Fearon and Vogelstein put forward a hypothesis on the development of colorectal neoplasia in terms of the known genetic and epigenetic changes.5 While there is broad agreement about the general principles behind the model, relatively little is known of the details of tumour development within or outside the context of germline MMR mutations.6 Research into the basis of the adenoma-carcinoma sequence is limited by the inability to examine the adenoma, other than at the time at which it is removed from the bowel, at some stage in its evolution. Inferences about longitudinal development must be made by examining the set of genetic changes and/or pathological characteristics present in different adenomas removed at different points along their development (and with differing and unknown potentials to become carcinomas) and attempting to interpret the differing patterns observed at those points in time. If all adenomas had the same sequence of events this would not present a problem. Inter-individual variation therefore restricts interpretation. Consistent differences between HNPCC and sporadic neoplasia may suggest fundamental differences along the pathway.

Adenomas arising in an individual with a germline MMR mutation and in a subject without such a mutation could differ because (i) the specific somatic mutations in the critical genes along the development pathway may be different, (ii) the order in which the critical genes are impacted may differ because some genes may be inherently more mutable than others because of their DNA sequence, or (iii) the critical genes may differ in the two processes. However, Homfray et al showed that the pattern of APC mutations did not differ depending on the MMR mutation status of the patient in whom the adenoma arose, suggesting the same role for APC in HNPCC adenomas as in sporadic adenomas.7

This study provides further evidence that loss of mismatch gene expression is an early step in the progression of adenomas arising in patients with an MMR germline mutation. Loukola and colleagues8 found that analysis of microsatellite markers of tumour derived DNA was a useful screen for HNPCC. In this study, small HNPCC adenomas arising in patients with an MMR germline mutation did not show loss of expression of the MMR protein while larger adenomas had lost expression.

While the comparison of HNPCC and “sporadic” adenomas is the appropriate comparison to make, there are logistical constraints which limit the interpretation of some of the observations. For instance, screening practices differ for patients with germline mutations (regular colonoscopy) and those undergoing regular general population screening (sigmoidoscopy followed by colonoscopy if a positive finding in the distal colon). Such differences in screening methodology might well produce effects such as smaller adenomas in those undergoing regular screening and an excess of distal adenomas in the sporadic group, as indeed observed in this study. Such difficulties are inherent to this type of research and must be borne in mind.

This research is an important step towards furthering our understanding of the development of colorectal neoplasia. Comparisons of genetic changes and pathological characteristics of adenomas arising in those predisposed with those without overt predisposition may be interpretable because of our knowledge of MMR


Linked Articles