Peutz–Jeghers syndrome polyps are polyclonal with expanded progenitor cell compartment

Peutz–Jeghers syndrome (PJS) is an autosomal dominant cancer susceptibility syndrome characterised by mucocutaneous melanin pigmentation, hamartomatous polyps, and an 18‐fold increase in intestinal and extraintestinal cancer risk.¹ PJS is caused by a germline mutation in LKB1, a gene that plays a role in cellular polarity.² Proper cellular polarity is critical for accurate asymmetrical stem cell division.³ The pathogenesis and neoplastic risk, if any, of hamartomatous Peutz–Jeghers polyps remain unclear.

Based on rare observations of neoplastic changes in PJS polyps and the finding of biallelic inactivation of the gene involved in PJS, the existence of a unique hamartoma–carcinoma sequence has been proposed in this disorder.⁴,⁵ This concept suggests that PJS polyps are clonal premalignant lesions responsible, at least in part, for the high rates of gastrointestinal cancer in these patients. However, dysplastic changes have been found rarely. Clarification of the risk of neoplastic transformation in PJS polyps would assist in the design of polyp surveillance strategies for PJS patients.

We studied the clonality of laser microdissected epithelium from two fresh frozen PJS polyps obtained from each of two female PJS patients from different well characterised PJS families, employing X chromosome inactivation of the HUMARA gene.⁶ Diagnosis of PJS in these two patients was confirmed by identifying the LKB1 germline mutations—two new mutations resulting in premature truncations (c.829‐830insGGGCG, p.Asp277Glyfs12 and c.718C→A, p.Ser240X). The clonality assay showed no significant change in the shift of ratios between the small and large HUMARA allele for all polyps after digestion of the unmethylated allele, indicative of a polyclonal nature of these polyps. Furthermore, a loss of heterozygosity (LOH) analysis of four markers surrounding LKB1 showed no LOH in any of the polyps. As LOH would be indicative of monoclonal cell expansion, its absence is consistent with the polyclonal nature of the polyps as found in the clonality assay.

The polyclonal nature and lack of LKB1 LOH is compatible with our previous hypothesis that elongation of the progenitor zone, caused by a germline defect in asymmetrical stem cell division and subsequent mucosal prolapse, may play a role in polyp formation.³,⁷ To investigate whether the length of the progenitor zone in PJS patients is altered compared with the length of the progenitor zone in normal intestinal mucosal controls, immunohistochemical Ki67 analysis was undertaken to obtain proliferation indices. Six PJS polyps from the PJS patients studied in the clonality assay were divided into three areas: (1) normal flat intestinal mucosa adjacent to the stalk of the polyp; (2) the area that forms the transition between flat intestinal mucosa and hamartomatous mucosa; and (3) the head of the polyp consisting of hamartomatous mucosa. The length of the proliferative zone in PJS flat intestinal, transitional, and polyp head mucosa combined was elongated compared with the crypts in the normal intestinal mucosa of the controls (p<0.001) (table 1, fig 1). However, the labelling index—defined as the percentage of positive Ki67 cells within the proliferative compartment—was not statistically different, showing that the percentage of proliferating cells in the crypt remains unchanged, but that in PJS mucosa the length of the proliferative zone is expanded.

Table 1 The length of the progenitor zone in cell positions and the labelling index for normal intestine and Peutz–Jeghers syndrome (PJS) polyps and statistical analysis (t test).

	n	Mean length ofprogenitor zone	SD	Labelling index (,%)	p Values
Normal	60	30	3	73.30
PJS
total	110	46	7	74.10	Normal vs PJS intestine:	<0.001
flat	30	43	5	73.30	PJS flat vs transition:	0.023
					PJS flat vs head:	0.012
transition	40	48	9	75.50	PJS transition vs head:	0.655
head	30	47	7	73.50

Flat, normal flat intestinal mucosa adjacent to polyp stalk; head, head of polyp; transition, transition zone between flat mucosa and hamartomatous mucosa.

Figure 1 Ki67 staining pattern in intestinal crypts of normal and polyp tissue from patients with Peutz–Jeghers syndrome (PJS) and controls. In all samples the number and location of Ki67 positive cells was scored in 10 completely visible and longitudinally sectioned crypts. The proliferative compartment was defined as the cellular compartment extending between the cell positions of the lowermost and uppermost Ki67 positive labelled cell using the bottom of the crypt as the reference cell position 0.⁸ (A) Schematic overview of Ki67 staining, black  =  positive Ki67 staining. (B) Ki67 staining on normal control intestine. (C) Ki67 staining on PJS normal intestine. (D) Ki67 staining on PJS polyp.

In conclusion, we have shown that PJS polyps are polyclonal expansions, arguing against the presence of a hamartoma–carcinoma sequence in PJS. Furthermore, the progenitor zone in intestinal crypts in PJS patients is expanded compared with normal individuals. We postulate that this is caused by disruption of the delicate balance between symmetrical and asymmetrical stem cell divisions, possibly leading to altered stem cell lineage turnover rates and protracted clonal evolution, thereby accelerating progression to cancer.⁹,¹⁰ These findings have practical consequences for surveillance strategies in PJS patients. Removal of the gastrointestinal hamartomatous polyps may not necessarily reduce cancer risk.

Acknowledgements

This work was supported by the Netherlands Digestive Disease Foundation (WS01‐03).