Statistics from Altmetric.com
- Hyperplastic polyposis syndrome
- colorectal carcinoma
- first-degree relatives
- colorectal cancer
- colonic polyps
- Helicobacter pylori
- acid-related diseases
- non-ulcer dyspepsia
- genetic polymorphisms
- gastric neoplasia
- family cancer
I read with great interest the two reports by Boparai and coworkers in Gut on the increased risk of colorectal carcinoma (CRC) in 77 hyperplastic/serrated polyposis syndrome (HPS) patients during follow-up1 and in 347 HPS first-degree relatives (FDRs).2 To date, follow-ups have been performed in only 43/308 (14.0%) HPS patients and HPS coexistence with CRC in FDRs has been analysed in only 29/308 (9.4%) HPS patients described up to the end of 2009 (table 1).
Nevertheless, there are some points to be explained:
The term ‘(sessile) serrated adenoma’ should be abbreviated as SSA, instead of SA (serrated adenoma). SSA, characterised by abnormal proliferation, is included in the non-dysplastic group contrary to SA, which belongs to the dysplastic serrated polyp category because of the associated epithelial dysplasia.3
Only 5/77 (6.5%) cases of ‘interval CRC’ developed in HPS patients during follow-up: two with cancer (T1 and T3 stage) in an unbelievably short time (4.3 and 7.7 months, respectively) and two with ‘carcinoma in situ=Tis’ CRC.
According to the modified Vienna classification and the comment in the recent WHO classification, the use of the term Tis is strongly discouraged. These lesions are encompassed within the term ‘high-grade dysplasia’ (HGD).4 5
It is particularly necessary to explain a confusing quotation about HPS FDRs in the abstract: ‘Based on the estimated HPS prevalence of 1:3000 in the general population’.2 This statement is an erroneous interpretation of an abstract published in 2001,6 for a study in which HPS prevalence was estimated by colonoscopy: ‘in patients in whom ≥20 distal HP were found previously during FS (flexible sigmoidoscopy) screening trial of 40,674 asymptomatic people’.6 In other words, the presented HPS prevalence was not for the general population but only for patients with distal hyperplastic polyps (HP).
Based on the incorrect interpretation of the HPS prevalence in the general population, statistical analysis concerning the projected RR of HPS in FDRs was performed; the result, which was as high as 39 (95% CI 13 to 121), was false and was estimated unnecessarily.2
During the period of 33 years (1977–2009), only 308 HPS patients were described (nine patients/year). The rarity of HPS was confirmed in a colonoscopy-based screening programme of 50 148 participants in which merely 28 HPS patients (0.056%) were identified.7 8 An acknowledged HPS population incidence was specified as 1:100 000.9
Another misleading quotation, ‘previously published case series report that up to 50% of HPS patients have FDR with CRC’, was probably inadvertently taken from the abstract of Lage et al: ‘Familial history of CRC/polyps was positive in 6/12 (50%) of cases’.10 However, a careful analysis of the text reveals a family CRC history in 5/14 patients: one with an unknown type of polyps and two with single HPs (table 1),10 leaving only 2/14 (14.3%) patients with CRC. Furthermore, screening colonoscopy was performed in 17 FDRs (table 2),10 with no FDR harbouring CRC and only two fulfilling the HPS criteria.10
Patients with HPS are still reported to be at >50% risk of CRC, but this risk is highly exaggerated. Admittedly, from 1977 to 2009, CRC was diagnosed in 122/308 (39.6%) HPS patients (table 1); nevertheless, its frequency seems to be dependent on the method of patient selection. In a preliminary analysis of 124 HPS patients described up to 2004, CRC was found in 1/15 (6.6%), 26/57 (45.6%) and 14/17 (82.3%) patients who were asymptomatic, had symptoms or had a colectomy because of CRC, respectively.7 This result was partly confirmed by the absence of CRC in 28 HPS patients selected from 50 148 asymptomatic participants in a colonoscopy-based screening,7 8 in contrast to the CRC prevalence of 22/27 (81.5%) in HPS patients diagnosed at the initial colonoscopy performed principally due to clinical symptoms suggestive of CRC.1
It is essential to judge the information included in abstracts critically and to verify the results presented in the main texts.
Competing interests None.
Provenance and peer review Not commissioned; internally peer reviewed.
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.