Article Text

AFAP: variety is the spice of life
  1. H T LYNCH,
  1. Department of Preventive Medicine and Public Health,
  2. Creighton University School of Medicine,
  3. 2500 California Plaza,
  4. Omaha, Nebraska 68178, USA

    Statistics from

    See article on page 548

    Familial adenomatous polyposis (FAP) has been known for over 100 years1; Gardner’s syndrome, the familial association of multiple colonic adenomas and early onset colorectal cancer (CRC) with osteomas and multiple cutaneous fibromas or epidermoid cysts has been known for almost 50 years.2 The discovery of the APC gene3 has changed the diagnosis and clinical management of FAP radically. Firstly, it became clear that FAP and Gardner’s syndrome have the same genetic basis. The expression of osteomas and of the cutaneous signs varies greatly among people with the same mutation. Other phenotypic variations have been found to have a genetic basis. For example, congenital hypertrophy of the retinal pigment epithelium is associated with mutations distal to exon 9,4 whereas susceptibility to desmoid tumours is associated with mutations at codons 1444 and 15785 and a profuse number of colonic adenomas, often more than 5000 (compared with the more typical 1000–2000), is associated with mutations between codons 1250 and 1464.6

    A variant called attenuated FAP (AFAP) has been associated with mutations in the 5′ portion of the APC gene.7 AFAP is characterised by the occurrence of often fewer than 100 colonic adenomas (sometimes as few as five or 10), with proximal colonic predominance, a later age of adenoma development (c. 35–40 years), and a still later age of CRC, often in the mid-50s. These features require screening by colonoscopy rather than flexible sigmoidoscopy. Given the later age of colonic adenoma and CRC onset, we initiate colonoscopy at age 20–25 years. This recommendation would be modified by the occurrence of earlier colonic neoplasia in the family. Unlike classic FAP, there is no age where surveillance can be discontinued on account of a normal phenotype. The adenomas are also often flat. Fundic gland polyps and duodenal adenomas may also occur, hence the need for upper endoscopy with a wide angle endoscope for all patients with FAP regardless of location of mutation.

    Brensinger and colleagues (see page 548) have studied 31 at-risk or affected members from four families with a mutation in the opposite end of the APC gene (codon 1979 or 2644). The phenotype was remarkably similar to AFAP. Among mutation carriers, some examined at an older age had fewer than 100 colorectal adenomas whereas other family members may have presented with classic (profuse) polyposis at a young age. CRC occurred at a mean age of 50 years, which was much older than that found in classic FAP pedigrees. Extracolonic phenotypic findings known to associate with classic FAP were variably expressed in these families. These authors reported a disassociation of severity of extracolonic manifestations from the number of colorectal polyps. We have observed the same thing in AFAP. In our limited experience, we have observed fundic gland polyps, duodenal adenomas and periampullary carcinoma, but there seems to be a paucity of other extracolonic manifestations relative to classic FAP. However, more research is needed on this topic.

    The diagnostician must be aware of the phenotypic importance of mutation location within the APC gene. However, do we know enough about these phenotypic correlations to use this information for genetic counselling and, ultimately, for surveillance and management of these patients at high risk of cancer? Are these phenotypic correlations clinically important enough to say that all patients with FAP should be tested to identify the mutation? We believe so. There is no question that colonoscopy, as opposed to flexible sigmoidoscopy, is the favoured screening approach in AFAP (given its proximal predilection to colonic adenomas) or in any extreme 3′ or 5′ mutation in APC. When too many colonic adenomas are present for polypectomy, particularly when they are clustered in the caecum, we have recommended prophylactic subtotal colectomy. However, in cases with small numbers of adenomas, management by polypectomy is an option. Having said this, the efficacy of polypectomy to prevent CRC in these patients has not been established.

    Because of the late onset of both colonic adenomas and CRC, its right-sided adenoma predilection, and the fact that the colonic adenomas may be so few (<10), the importance of AFAP may be completely missed unless a comprehensive family history is obtained and carefully examined. This history must account for colonic adenomas, their age of onset, location within the colon, and the occurrence of CRC. Indeed, there is a dire need for education of physicians about AFAP, with particular attention given to how to diagnose and manage it, since in our experience, in spite of the fact that the disorder was described a decade ago,8 9 a surprisingly small number of physicians express knowledge of the syndrome and/or confidence in its diagnosis and management.

    When do we order a test on a young patient with multiple colonic adenomas and a negative or unremarkable family history, and how do we interpret such test results? This is a vexing problem for which the answer remains elusive. We recommend APC gene testing for any patient with more than five colonic adenomas at an early age (<25 years). Unfortunately, a negative molecular genetic test for mutation in the APC gene will not rule out a genetic cancer risk to the patient, siblings or progeny. He or she might harbour a germline mutation that has not yet been identified. Roughly 20% of frank FAP cases have negative mutation tests.10 The patient’s early adenomas could be a manifestation of another hereditary CRC syndrome such as hereditary non-polyposis colorectal cancer (HNPCC). Conversely, should an APC mutation or, for that matter, a mutation for other hereditary forms of CRC such as HNPCC, be identified, then one would be in a position to do DNA testing on that individual’s parents, siblings, and progeny. This may be especially valuable in AFAP because of the absence of a florid phenotype in many cases.

    No solid data exist on the true population frequency of AFAP. Such estimates will be difficult, due to its subtle phenotypic manifestations coupled with the fact that the detailed family histories necessary for the diagnosis of AFAP are notoriously ignored in the clinical practice setting.

    See article on page 548


    View Abstract

    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.

    Linked Articles