Non-dysplastic Areas Research Articles

Abstract 3600: Early detection of esophageal adenocarcinoma using near-infrared lectin-based imaging

Abstract Background: Long procedure times and sampling error frequently limit the efficacy of endoscopic surveillance for Barrett's esophagus (BE). Near-infrared (NIR) fluorescence imaging minimizes optical scattering and tissue autofluorescence. We assessed, in an ex vivo setting, the feasibility of using a topically applied NIR dye-labeled lectin for the detection of early neoplasia in BE. Methods: We recruited consecutive patients undergoing endoscopic mucosal resection (EMR) for BE-related early neoplasia. EMR specimens were freshly collected and sprayed at the bedside with a fluorescent lectin and then imaged ex vivo. A maximum of two punch biopsies were collected from each EMR under NIR light guidance. Fluorescence intensity from dysplastic and nondysplastic areas within EMRs were compared and from punch biopsies with different histological grades. Results: We analyzed 29 EMR specimens from 17 patients. Dysplastic regions showed significantly lower fluorescence across whole EMR specimens (P < 0.001). A 41% reduction in fluorescence was found for dysplastic compared to nondysplastic punch biopsies (P < 0.001), with a sensitivity and specificity for dysplasia detection of 80% and 82.9%, respectively. Conclusion: NIR imaging using fluorescently-labeled lectins can distinguish dysplastic from nondysplastic Barrett's mucosa ex vivo. We are currently planning a first-in-human study to validate this imaging platform. Citation Format: Andre A. Neves, Massimiliano Di Pietro, Maria O'Donovan, Dale J. Waterhouse, Sarah E. Bohndiek, Kevin M. Brindle, Rebecca C. Fitzgerald. Early detection of esophageal adenocarcinoma using near-infrared lectin-based imaging [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3600.

Detection of early neoplasia in Barrett's esophagus using lectin-based near-infrared imaging: an ex vivo study on human tissue.

Background and study aims Endoscopic surveillance for Barrett’s esophagus (BE) is limited by long procedure times and sampling error. Near-infrared (NIR) fluorescence imaging minimizes tissue autofluorescence and optical scattering. We assessed the feasibility of a topically applied NIR dye-labeled lectin for the detection of early neoplasia in BE in an ex vivo setting. Methods Consecutive patients undergoing endoscopic mucosal resection (EMR) for BE-related early neoplasia were recruited. Freshly collected EMR specimens were sprayed at the bedside with fluorescent lectin and then imaged. Punch biopsies were collected from each EMR under NIR light guidance. We compared the fluorescence intensity from dysplastic and nondysplastic areas within EMRs and from punch biopsies with different histological grades. Results 29 EMR specimens were included from 17 patients. A significantly lower fluorescence was found for dysplastic regions across whole EMR specimens ( P < 0.001). We found a 41 % reduction in the fluorescence of dysplastic compared to nondysplastic punch biopsies ( P < 0.001), with a sensitivity and specificity for dysplasia detection of 80 % and 82.9 %, respectively. Conclusion Lectin-based NIR imaging can differentiate dysplastic from nondysplastic Barrett’s mucosa ex vivo.

OC-053 Lectin-Based Near infra-red Molecular Imaging for Dysplasia Detection in Barrett’s Oesophagus: An ex-vivo Study on Human Tissue

Introduction Detection of early neoplasia in Barrett’s oesophagus (BO) by white-light endoscopy is challenging due to the inconspicuous nature of dysplasia. Molecular imaging using fluorescently labelled wheat-germ agglutinin (WGA) is a promising tool as this topically applied imaging agent shows lower binding to dysplastic versus non-dysplastic BO. 1 However in an endoscopy setting, the detection of fluorescence in the blue/green range is limited by high-levels of tissue autofluorescence. This limitation can be overcome by using near infra-red (NIR) imaging. We aimed to assess in an ex-vivo model the feasibility of WGA-based NIR imaging for detection of dysplasia in BO. Methods we recruited patients with early BO-related neoplasia undergoing endoscopic mucosal resection (EMR). Freshly collected EMR specimens were sprayed with WGA-IR800CW and then imaged with a high-sensitivity NIR camera. Fluorescence images were captured and up to two punch biopsies were collected from each EMR under fluorescence guidance. The EMRs were paraffin embedded, cut every 2 mm and processed for histopathological assessment. Each section was scored by an expert GI pathologist every 1 mm to construct a pathology grid, which was manually co-registered with the fluorescence image. The mean fluorescence intensity (MFI) of cells in dysplastic and non-dysplastic areas was compared by the Wilcoxon matched-pairs signed rank test. Only EMR specimens with at least one dysplastic gland were included in the analysis. In addition, the MFI of punch biopsies taken from dysplastic and non-dysplastic areas was also compared by Mann-Whitney test. Results Ten patients were recruited at a single centre. A total of 18 EMR specimens and 33 punch biopsies were collected, of which 10 were dysplastic. In the whole EMR analysis, we found a significantly lower MFI for dysplastic versus non-dysplastic areas (P = 0.0012), in accordance with the reported reduced binding of WGA to neoplastic BO epithelium. Similarly, we found a nearly 2 fold reduction in the MFI of punch biopsies taken from dysplastic as compared to non-dysplastic (ND) areas (P = 0.0002) (Figure 1). Conclusion WGA-based NIR imaging is an effective method for differentiating dysplastic from non-dysplastic BO mucosa ex vivo , which reduces the effects of tissue autofluorescence. In-vivo studies are now required to test the efficacy of this method for detecting dysplasia during endoscopic surveillance. Reference 1 Bird-Lieberman, et al . Nat Med 2012. Disclosure of Interest None Declared

Neoplasms arising in large gastric hyperplastic polyps: endoscopic and pathologic features

Little is known about gastric neoplasms arising from hyperplastic polyps (HPs). To investigate the risk factors associated with neoplasms within HPs and to evaluate the role of alterations of the p16-cyclin D1-pRb pathway in the malignant transformation of HPs. Retrospective, case-control study. Tertiary-care center. Between May 1995 and January 2011, a total of 809 HPs >1 cm were investigated. Associated neoplasms were present in 30 HPs (case group); 30 HPs without neoplasms were selected as a control group. Gastric polypectomy. The risk factors associated with neoplasms within HPs and immunohistochemical expression of p16, cyclin D1, p53, and Ki-67 between case and control groups. Of the 809 HPs, 15 had associated dysplasia, and 15 had carcinoma. Multivariate analysis showed that neoplasm was associated with patient age (odds ratio [OR] 1.159; 95% confidence interval [CI], 1.243-2.044; P < .001), polyp size (OR 1.103; 95% CI, 1.055-1.152; P < .001), and polyp lobulation (OR 4.549; 95% CI, 1.759-11.0766; P < .001) but not with location, multiplicity, intestinal metaplasia, growth pattern, or Helicobacter pylori infection. Loss of p16 expression and high Ki-67 expression were observed in dysplastic areas of HPs compared with the control group (p16 = 14.3% vs 60%; P = .001, Ki-67 = 60.7% vs 36.7%; P < .001). However, no significant differences were found in nondysplastic areas in both groups. Single-center, retrospective study. HPs >1 cm may indicate the presence of neoplasms. Loss of p16 and high Ki-67 expression may be markers of HP-associated dysplasia.

OC-017 Autofluorescence-targeted Optical Biopsy Accurately Diagnoses Dysplasia In Barrett’s Oesophagus And Can Detect The Field Of Molecular Change

IntroductionProbe-based confocal laser endomicroscopy (pCLE) allows optical biopsies in Barrett’s oesophagus (BO) to predict histological outcome but it is subject to sampling error if performed in a random fashion. We...

Sa1852 Autofluorescence-Targeted Optical Biopsy Accurately Diagnoses Dysplasia in Barrett's Esophagus and Can Detect the Field of Molecular Change

Introduction Probe-based confocal laser endomicroscopy (pCLE) allows optical biopsies in Barrett’s oesophagus (BO) to predict histological outcome but it is subject to sampling error if performed in a random fashion. We used autofluorescence imaging (AFI) to direct pCLE and added molecular biomarkers to the histopathological diagnosis. The aims of this study were to assess the diagnostic accuracy for dysplasia of AFI-targeted optical biopsies and to investigate the correlation between pCLE patterns and field of molecular change. Methods 46 patients with BO (non-dysplastic BE n = 20, indefinite for dysplasia n = 4, low grade dysplasia n = 10, high grade dysplasia (HGD) or intramucosal cancer (IMC) n = 12) were recruited at a single centre. Patients underwent high-resolution endoscopy followed by AFI and then pCLE was performed on AFI positive (AFI+) areas. Targeted biopsies were taken from AFI+ areas, followed by random biopsies as per Seattle protocol. pCLE sequences were graded according to published criteria. Cyclin A and p53 expression were assessed by immunohistochemistry and aneuploidy by flow-cytometry on AFI-targeted biopsies. Statistical analyses were performed using chi-square test. Results AFI-targeted pCLE correctly classified all the HGD/EC patients and had a sensitivity and specificity for any grade of dysplasia of 93 and 83%, respectively. The Seattle protocol had similar sensitivity for HGD/IMC and any grade of dysplasia (83 and 89%, respectively). For the per-location analysis, a total of 155 endoscopic areas were analysed with pCLE and molecular biomarkers. pCLE had a sensitivity and a specificity for HGD/IMC and any grade of dysplasia of 100/64% and 78/75%, respectively. Overall, 40% of pCLE irregular sequences corresponded to non-dysplastic areas (false positive). We found a statistically significant enrichment (p Conclusion AFI-targeted pCLE has a high diagnostic accuracy for dysplasia in BO. Tissue biomarkers are a useful adjunct to characterise the field of molecular abnormality associated with optical dysplasia. These results suggest that the presence of pCLE irregularity, even in the absence of histological dysplasia, relates to molecular changes and may warrant close follow up. Disclosure of Interest None Declared.

TLR4 activates the β-catenin pathway to cause intestinal neoplasia.

Colonic bacteria have been implicated in the development of colon cancer. We have previously demonstrated that toll-like receptor 4 (TLR4), the receptor for bacterial lipopolysaccharide (LPS), is over-expressed in humans with colitis-associated cancer. Genetic epidemiologic data support a role for TLR4 in sporadic colorectal cancer (CRC) as well, with over-expression favoring more aggressive disease. The goal of our study was to determine whether TLR4 played a role as a tumor promoter in sporadic colon cancer. Using immunofluorescence directed to TLR4, we found that a third of sporadic human colorectal cancers over-express this marker. To mechanistically investigate this observation, we used a mouse model that over-expresses TLR4 in the intestinal epithelium (villin-TLR4 mice). We found that these transgenic mice had increased epithelial proliferation as measured by BrdU labeling, longer colonic crypts and an expansion of Lgr5+ crypt cells at baseline. In addition, villin-TLR4 mice developed spontaneous duodenal dysplasia with age, a feature that is not seen in any wild-type (WT) mice. To model human sporadic CRC, we administered the genotoxic agent azoxymethane (AOM) to villin-TLR4 and WT mice. We found that villin-TLR4 mice showed an increased number of colonic tumors compared to WT mice as well as increased β-catenin activation in non-dysplastic areas. Biochemical studies in colonic epithelial cell lines revealed that TLR4 activates β-catenin in a PI3K-dependent manner, increasing phosphorylation of β-cateninSer552, a phenomenon associated with activation of the canonical Wnt pathway. Our results suggest that TLR4 can trigger a neoplastic program through activation of the Wnt/β-catenin pathway. Our studies highlight a previously unexplored link between innate immune signaling and activation of oncogenic pathways, which may be targeted to prevent or treat CRC.

Identification of molecular alterations in gastrointestinal carcinomas and dysplastic hamartomas in Peutz-Jeghers syndrome

Peutz-Jeghers syndrome (PJS) is caused by mutations in the LKB1 gene. It is characterized by gastrointestinal polyposis and an increased cancer risk, mainly in the gastrointestinal tract. Mechanisms of PJS-associated carcinogenesis are unclear. We investigated the involvement of candidate genes and molecular pathways in PJS-associated gastrointestinal cancers and dysplastic hamartomas. Cases were selected from the Dutch PJS cohort. Available tissue was immunostained for phospho-S6, β-catenin, P53 and SMAD4. DNA was isolated from carcinoma tissue and dysplastic and non-dysplastic areas of hamartomas specifically. Mutation analyses were done for BRAF, KRAS and P53, and loss of heterozygosity (LOH) analyses for LKB1 and P53. Twenty-four of 144 patients (17%) developed 26 gastrointestinal malignancies at a median age of 49 years (interquartile range: 35-60). Eleven of 792 hamartomas (1.4%) of 9 patients were classified as dysplastic. LOH of LKB1 was detected in three of six (50%) carcinomas and in the dysplastic part of three of five (60%) hamartomas. Aberrant P53 expression was observed in 8 of 15 (53%) carcinomas. Six carcinomas with P53 overexpression harboured a P53 mutation, with loss of the remaining wild-type allele in four. Two hamartomas showing P53 overexpression in high-grade dysplastic foci harboured a P53 mutation with LOH. Loss of nuclear SMAD4 was observed in high-grade dysplastic foci of two of four (50%) hamartomas, in contrast to low-grade dysplastic foci (0/4) and non-dysplastic epithelium. Our findings suggest a role for mutant P53 in PJS-associated gastrointestinal carcinogenesis. Inactivation of transforming growth factor-β/bone morphogenetic protein signalling and complete loss of LKB1 might be involved in dysplastic transformation of gastrointestinal hamartomas specifically.

Matrix Metalloproteinase-9 Expression in the Normal Mucosa–Adenoma–Dysplasia–Adenocarcinoma Sequence of the Colon

It has been proposed that matrix metalloproteinases (MMPs) play a role in tumor invasion. We determined protein expression of matrix metalloproteinase-9 (MMP-9) in colorectal cancer (CRC), corresponding normal mucosa and colorectal adenomas. For confirmation of immunohistochemical results MMP-9 TaqMan RT-PCR analysis was performed. Expression of MMP-9 was determined on paraffin embedded biopsy sections by immunohistochemistry in 31 CRC patients (from cancer tissue and corresponding normal mucosa) and in 30 patients with adenoma (nine adenomas with high grade of dysplasia). MMP-9 immunostaining was determined semi-quantitatively. For Taqman RT-PCR analyses normal mucosa (n = 5), adenoma without (n = 6) and with high grade dysplasia (n = 7) and CRC (n = 10) were investigated. Statistical analysis with ANOVA, LSD test and correlation analysis were performed. P value of <0.05 was considered significant. The MMP-9 expression in CRC was significantly higher compared to adenomas or the normal mucosa (P = 0.001). Significantly higher expression of MMP-9 has been observed in adenomas with high grade dysplasia compared to other adenomas or normal colon (P < 0.001). Diffuse strong MMP-9 expression was present in tumor as well as in stromal cells. In adenoma samples, dysplastic epithelial cells showed moderate intensive cytoplasmic MMP-9 expression, with a clear-cut differentiation between dysplastic and non-dysplastic areas. Staining intensity correlated with the grade of CRC. We demonstrate a significantly higher expression of MMP-9 in adenoma with high grade dysplasia-CRC sequence as compared to normal tissue. The over-expression of MMP-9 strongly suggests its association with colorectal carcinogenesis.

Evaluation of an Endocytoscopy System in the Surveillance of Patients with Barrett Esophagus

Background: Biopsy surveillance for Barrett patients with increased risk for neoplasia is time consuming and subject to sampling error. Guided biopsy acquisition from neoplastic areas invisible to conventional endoscopy may increase effectiveness of surveillance examination. We assessed the potential role of Endocytoscopy using large scale magnification comparable to cytologic analysis in identifying neoplasia in Barrett esophagus. Methods: 82 biopsy specimens from 7 consecutive patients (6 men, mean age 58 years) who presented for Barrett surveillance without visible neoplastic lesions were enrolled into this prospective study. Markings along the Barrett segment were placed using Argon plasma coagulation. Endocytoscopy images with 1100 magnification (Olympus Corp. Hamburg, Germany) were collected immediately right to the marking. A biopsy was then carefully taken from that same spot to ensure that identical areas were scanned and biopsied. Biopsies were assessed according to WHO criteria of neoplasia. The major outcome included image quality parameters of endocytoscopy and identification of characteristic changes of dysplasia based on previously published criteria for Barrett cytology (Geisinger et al. Cancer 1992). Endocytoscopy images were randomized and individually reviewed by a pathologist and a gastroenterologist blinded to final diagnosis. Results: Adenocarcinoma was found in 4 (4.9%) of the biopsies, high-grade intraepithelial neoplasia in 14 (18.3%) and low-grade intraepithelial neoplasia in 5 (6.1%). The remaining biopsies showed either normal Barrett mucosa in 51 biopsies (62.2%) or no Barrett epithelium in 7 biopsies (8.5%). On average 9 images were reviewed per site. A general evaluation of images was not possible or poor in 22% and of moderate quality in 78%. None of the image sequences per site was graded as good or excellent. Staining was sufficient in 71%; however, image resolution was mostly graded as insufficient. Interpretable images to diagnose dysplasia were seen in only 4 to 41% depending on various cytologic criteria. Using only interpretable images staining was graded significantly better in nondysplastic areas than in dysplastic areas (p < 0.02). However, none of the characteristic changes of dysplasia were different when blindly comparing nondysplastic with dysplastic sites. Conclusion: Endocytoscopy lacks sufficient image quality to be currently of assistance in identifying areas of dysplasia in Barrett esophagus when using large scale magnification equivalent to cytology. Lower magnification might be superior and comparable to histologic architecture and has therefore to be tested using the same methodology.

Dysplasia in Fundic Gland Polyps is Associated with Nuclear β-Catenin Expression and Relatively High Cell Turnover Rates

Background: Fundic gland polyps (FGPs) occur in both syndromic and sporadic form. Syndromic FGPs arise through mutations in the adenomatous polyposis coli (APC) gene, whereas sporadic FGPs are caused by $ -catenin gene mutations. Dysplasia in sporadic FGPs, found less often than in syndromic FGPs, was recently associated with APC rather than $ -catenin mutations. These data suggest different functional consequences of APC and $ -catenin mutations. To investigate this hypothesis, we examined proliferative activity, degree of apoptosis, # -catenin expression and p53 expression in syndromic and sporadic FGPs. Methods: Syndromic FGPs ( n r = r 9) from familial adenomatous polyposis (FAP) patients and sporadic FGPs ( n r = r 18) were studied. Proliferative activity, apoptotic cell death and expression of # -catenin and p53 were examined by immunohistochemistry. In FGPs containing dysplasia, areas with and without dysplasia were compared. Results: Syndromic and sporadic FGPs without dysplasia exhibited similar proliferative activity, degree of apoptosis, # -catenin and p53 expression. Dysplasia was observed more often in syndromic (4/9) than in sporadic FGPs (1/18). Within FGPs containing dysplasia, dysplastic areas showed abnormal nuclear # -catenin staining in 3/5 cases and higher rates of cell proliferation and apoptosis than non-dysplastic areas. Overexpression of p53 was not observed. Conclusion: The finding of similar rates of proliferation and apoptosis in syndromic and sporadic FGPs does not support the hypothesis that APC and $ -catenin gene mutations have different functional consequences in FGPs. The association of dysplasia with relatively high cell turnover rates and nuclear expression of $ - catenin indicates activation of the Wnt-APC- # -catenin pathway in dysplasia. The finding of dysplasia in some but not all syndromic FGPs suggests the involvement of other genes in addition to the APC gene in the development of dysplasia in FGPs.

Glut-1 expression in dysplastic and regenerative lesions of the colon.

Monosaccaride transporter proteins are responsible for transmembrane transport of monosaccarides into cells. Glucose transporter protein 1 (Glut-1) is most prevalent in the cell membranes of erythrocytes and facilitates transport of glucose in tissues with barrier functions, i.e. blood brain barrier. Expression of Glut-1 in malignant tumors is increased due to increased metabolic need of the proliferating cell populations. In colorectal adenomas and carcinomas, membranous expression of Glut-1 has been associated with higher grade of tumors and decreased survival time. We studied the expression of Glut-1 in dysplastic proliferations of the colon which included sporadic adenomas and dysplasia associated lesions (DALM) in patients with ulcerative colitis and reactive/regenerative proliferations of the colon, including non-dysplastic chronic colitis, acute colitis and ischemia. Two patterns of Glut-1 expression were detected. Most adenomas and DALMs showed at least focal membranous expression of Glut-1. In addition a second staining pattern was recognized which consisted of prominent supranuclear dots. This pattern of staining was not only seen in adenomas and DALM but also in non-dysplastic areas immediately surrounding sporadic adenomas, in regenerative chronic colitis and in areas surrounding acute inflammation. Areas away from dysplasia did not show any positive staining for Glut-1. We conclude that two distinct patterns of Glut-1 expression may be found in colonic epithelial proliferation: membranous staining, associated with dysplasia, and, heretofore not described, supranuclear staining which may be related to Glut-1 expression secondary to expression of specific growth factors and not necessarily related to dysplasia.

Endoscopic Mapping and Surrogate Markers for Better Surveillance in Barrett Esophagus

Surveillance methods in Barrett esophagus (BE) using light microscopic examination of random biopsy specimens may miss focal dysplasia. In addition, dysplastic foci identified initially may not be relocated subsequently, making chemoprevention studies difficult. By using a special gastroscope, systematic mapping (4-quadrant biopsy specimens at 1-cm intervals) was performed in 22 patients (33 total mappings yielding 700 biopsy specimens). H&E, immunohistochemistry, and DNA ploidy analysis were performed. c-erbB-2 and positive Ki-67 were detected only in dysplastic sites; thus, their detection did not precede morphologically identifiable dysplasia. On the other hand, aneuploidy and p53 were detected in dysplastic and nondysplastic areas. p53 was correlated with dysplasia, and S-phase narrowly missed correlation, while aneuploidy was not correlated. PCNA and bcl-2 were ubiquitous, limiting their usefulness. On second maps, epithelial type was reidentified with 81% accuracy. A significant correlation was found between p53 and dysplasia. Sites of dysplasia and abnormal biomarkers could be relocated accurately by using endoscopic mapping. Therefore, mapping combined with biomarker studies may provide better surveillance and serve as a useful technique in chemoprevention studies.

Inverted Sinonasal Papilloma: A Molecular Genetic Appraisal of Its Putative Status as a Precursor to Squamous Cell Carcinoma

Inverted papilloma (IP) is a proliferative lesion of the epithelium lining the sinonasal tract. Although IP often recurs after surgical excision and is sometimes associated with squamous cell carcinoma of the sinonasal cavity (SNSCC), its presumed neoplastic nature and putative role as a precursor to squamous cell carcinoma have not been confirmed at the molecular genetic level. We analyzed the pattern of X chromosome inactivation in IPs from nine female patients. Inactivation of a single allele is seen in monoclonal proliferations and may be indicative of a neoplastic process. We also analyzed 28 IPs and 6 concurrent SNSCCs for loss of heterozygosity (LOH) on chromosomal arms 3p, 9p21, 11q13, 13q11, and 17p13. Losses at these loci occur frequently during neoplastic transformation of the upper respiratory tract and can be detected in squamous cell carcinomas and the progenitor lesions from which they arise. X chromosome analysis was informative in four of the nine IPs. All four lesions demonstrated a monoclonal pattern of inactivation. LOH was not detected in any nondysplastic areas from the 28 IPs, but LOH at one or more chromosomal loci was present in all six of the concurrent SNSCCs. We conclude that IPs are monoclonal proliferations, yet they do not fit the profile of a prototypic precursor lesion. Unlike squamous epithelial dysplasia, IPs do not routinely harbor several of the key genetic alterations that are associated with malignant transformation of the upper respiratory tract.

RER(+) phenotype in prostate intra-epithelial neoplasia associated with human prostate-carcinoma development.

The mutator (RER(+)) phenotype has been shown to be a mutational mechanism for tumour-suppressor-gene inactivation in colorectal cancer. A group of 60 prostate-carcinoma patients was studied to determine the frequency, intratumour distribution and timing of mutator phenotype in this cancer. Ten microsatellite loci were analyzed in 172 carcinoma foci (CF) and in 57 associated non-cancerous prostate tissues, including 31 areas of prostate intra-epithelial neoplasia (PIN) and 26 non-dysplastic areas with glandular hyperplasia (HP). We detected lesions with the RER(+) phenotype in 42% (25/60) of the prostate tumours. Clonal foci with RER(+) phenotype were detected at similar frequencies in pre-cancereous PIN (16%, 5/31) as in associated carcinoma foci (22%, 37/172), but were detected in only one of the 26 non-dysplastic prostate tissues studied (4%). Thus, clonal RER(+) foci were significantly more frequent in CF than in HP (p < 0.05). MI itself was significantly more frequent in CF (53%, p < 0.0001) and in PIN (35%, p < 0.05) than in HP (12%). Furthermore, 5 PIN harboured microsatellite mutations also detected in the associated cancer. Our overall results therefore strongly suggest that the mutator phenotype may occur as an early event in prostate tumorigenesis.