Hyperplastic polyposis syndrome and risk of colorectal cancer

Abstract

Hyperplastic polyposis syndrome is a widely accepted, but poorly understood, risk factor for colorectal cancer. A recent report has laid the foundations for improving the management of patients with this enigmatic disorder by identifying the features associated with colorectal cancer risk, as well as estimating the magnitude of this risk.

For at least the past decade, hyperplastic polyposis syndrome (HPS)—a relatively rare condition characterized by multiple colonic polyps with serrated glandular architecture—has been widely accepted as a risk factor for colorectal cancer (CRC). Guidelines for the surveillance and prevention of CRC in patients with HPS however, are still evolving.¹ Boparai et al.,² in a large, retrospective, multicenter study of endoscopy findings in 77 patients with HPS, have confirmed the increased risk of CRC and highlighted some of the difficulties encountered in planning colonoscopic surveillance for patients with this condition. The authors report that just under 30% of patients with HPS presented at baseline with CRC, and a further 7% developed a colorectal malignancy while under surveillance during 5 subsequent years.

To inform and guide clinical practice, there is a need to first understand the factors that contribute to the risk of CRC in HPS. In the study by boparai and colleagues, the authors have shown that the risk rises as the number of serrated polyps rises. In addition, the observation that CRCs discovered during surveillance were not confined to polyps with a diameter exceeding 1 cm, or to those recognizable at colonoscopy as malignant, and not to the right side of the colon, are particularly important; these findings dispel some widely held opinions regarding the development of CRC in HPS. However, during the time span of this retrospective study (26 years) new insights might have influenced the management of patients with HPS and, hence, the reported CRC risk.

First, recognition of the inherent phenotypic heterogeneity of HPS—with polyp counts varying from 5 to ≥100 and a range of site, size and histological subtype distributions within the colon—indicates that the risk of CRC is not likely to be uniform in patients with this disorder.^3,4 the cohort studied in the current report seems to represent a subtype with small numbers (median polyp count 15) of large polyps (61% had hyperplastic polyps ≥10 mm), and they might have a different CRC risk than a cohort in which the median polyp count is considerably higher.

Second, the presence of traditional adenomas has recently been reported to be associated with a fourfold increased risk of developing CRC.^5,6 Approximately two-thirds of the patients with HPS in the current study had at least one traditional adenoma. This information alone might have altered management of patients with HPS (for example, more aggressive polyp removal or earlier referral for surgery), as four of the five CRCs were diagnosed within a year of a previous colonoscopy at which many polyps, although biopsied, had been left in situ.

Third, although histological observation of malignant changes in occasional serrated polyps first appeared in the literature in the late 1970s, acceptance of the malignant potential of a subset of serrated polyps has been slow. Beliefs that all serrated polyps are innocuous continue to the present day, and might have influenced management of the patients in this cohort. Furthermore, the difficulties associated with the detection and removal of large flat lesions that may be located on mucosal folds could have contributed to the development of some interval cancers. Consequently, boparai et al.² make the appropriate suggestion that patients with HPS might benefit from surveillance at specialist centers that perform advanced colonoscopic imaging and polypectomy.

HPS was defined by burt and Jass in 2000 as follows: at least five histologically diagnosed hyperplastic polyps proximal to the sigmoid colon, two of which are >10 mm in diameter; or any number of hyperplastic polyps occurring proximal to the sigmoid colon in an individual who has a first-degree relative with hyperplastic polyposis; or >30 hyperplastic polyps of any size but distributed throughout the colon.⁷ The arbitrary nature of the criteria has prompted several investigators, including boparai et al.,² to suggest that the number of hyperplastic polyps indicative of HPS be reduced to 20. In addition, boparai and colleagues² have sensibly combined the sessile and traditional serrated polyp categories, as inconsistent and confusing terminology related to the subsets of serrated polyps with atypical histologic features is common, as are the differences of opinion that arise even between specialist gastrointestinal pathologists when classifying serrated polyps.

Larger studies, with patient numbers ranging from 32 to 151, reporting the features of HPS began to appear in 2006,^5,6,8,9 and have consistently reported a baseline CRC incidence of 25–37%, similar to the current study. Nonetheless, the problem still remains that the reports in the literature are retrospective studies. Boparai et al.² have also had to resort to a retrospective study in order to gain the critical mass of person-years of follow-up needed to assess the risk of CRC in patients with HPS; in this case seven institutions collectively amassed 295 person-years of follow-up between them. A prospective, population-based study of this rare condition, utilizing modern classification and management of polyps, would require the cooperation of multiple centers to obtain adequate follow-up within a reasonable time period.

A genetic contribution to HPS is becoming more generally accepted, but until genetic predispositions and CRC risks for each of the subtypes are definitively identified, management of patients with HPS will need to proceed on a case-by-case basis.¹⁰ The general approach emerging is that if polyps cannot be controlled endoscopically, surgical options should be considered—particularly if adenomatous features are present. If only large polyps are removed, a CRC might be missed as suggested by boparai et al.² In addition, to focus on removing only the proximal polyps jeopardizes the effectiveness of surveillance not just for the young-onset patient for whom CRC is more likely to be distal,⁵ but also for others—like the two patients in this study who were found to have distal CRCs. However, since prophylactic options for these patients (repeated endoscopy and/or surgery) are neither trivial nor without risk, there remains the need to identify those patients with HPS who are at the greatest risk of developing CRC (box 1). There is the potential to develop an immunohistochemistry panel capable of assessing serrated polyps for increased malignant potential, which would be particularly important in this syndrome given that there are many patients with HPS (60–70%),^2,5,6,9 including those with adenomas, who will never develop a CRC.

Box 1 |. Developments that would contribute to more effective management of HPS.

• A uniform classification scheme for serrated polyps, and identification of specific morphological features that increase the risk of malignancy

• An immunohistochemistry panel that transcends morphology to confirm the presence of serrated polyps that have malignant potential

• Identification of genetic predispositions for subsets of hyperplastic polyposis syndrome and assigned risks

• Identification of genetic and environmental modifiers associated with an increased risk of colorectal cancer

• Standardized, evidence-based recommendations for management