document
Transcript of document
immunochemical test (FIT) on same stool samples. Results: There was significant differencein expression of COX-2, MMP-7, B2M, and CKB mRNAs among cancer, AA, and healthysubjects. The 99 percentile of copies of each gene in healthy subjects was used as cutoffvalue. The specificity of fecal multi-markers RNA panel was 97.3%, while that of FIT was98.2%. The sensitivity of each assay for CRC was 75.7% for COX-2, 55.9% for MMP-7,24.3% for B2M, and 28.8% for CKB, respectively, while that of FIT was 66.7%. The sensitivityof fecal B2M and CKB mRNA assay was low, however, fecal B2M plus CKB mRNA assaysdetected five CRC subjects and five AA subjects with negative results of fecal COX-2 plusMMP-7 mRNA assays. The sensitivity of each assay for CRC was not significantly higherthan that of FIT, however, fecal multi-markers RNA panel was significantly higher than thatof FIT (84.7% vs. 66.7%, P= 0.003). Among subjects with AA, stage 0, I, or II, who areoften cured by endoscopic or surgical resection, fecal multi-markers RNA panel had signific-antly higher sensitivity than FIT (78.1% vs. 51.0%, P < 0.001). Conclusions: Fecal B2Mplus CKB mRNA assays detected ten colorectal advanced neoplasia subjects with negativeresults of fecal COX-2 plus MMP-7 mRNA assays. These results strongly suggest that fecalmulti-markers RNA panel would be a promising approach for CRC screening.
Su1144
Relationship Among Colorectal Cancer, Advanced Histology and Polyp Size inPatients Undergoing Screening Colonoscopy in North Eastern ItalyMarco Marino, Elena Zucchi, Giovanni Terrosu, Debora Berretti, Ilva Lodolo, RobertoMaieron, Maurizio Zilli
The University Hospital of Udine began the Bowel Cancer Screening Program(BCSP)in2008.Subjects aged 50-69 are being invited to complete one single immunochemical fecaltest(FIT)test every 2 years. At the beginning of our program screening was decided to givepriority to subjects aged 60-69(about 70% of all FIT invitations).Those positive were referredfor colonoscopy.We report the preliminary experience of 691 screening participants referredin our Center following positive FIT. METHODS AND AIMS:we review data on 691 patientsreferred to screening colonoscopy in our centre from Oct 2008 to May 2010. RESULTS:be-tween Oct 2008 to May 2010 34119 FIT kits dispatched had been returned by 38.4%of men and 42.3% of women invited. Of the 13477 returning tests 835 had a positivetest(6.4%).Responders to colonoscopy were 82.75% (691 colonoscopies,430 male,261female,medium age 62.39).All colonoscopies were performed in the morning with a comple-tion rate of 96%(caecal or terminal ileum intubation).Bowel preparation score(Boston BowelPreparation Scale-BBPS)ranges between 2 and 3 in 94.5% of subjects.Only low bowelpreparation quality and stenosing neoplasia impacted the completion rate.Colorectal can-cer(CRC)was diagnosed in 68 subjects (9.8%)of whom 2 with synchronous CRC(0.3%).Ona total of 1295 polyps, ≥ 1 colonic adenomas were present in 399 patients(57.74%)witha total of 873 adenomas.Most of the polyps were removed in the same session in 88.4% ofsubjects.Advanced adenomas were 374 in 265 patients (38.35%).Table 1 and 2 respectivelyshow CRC and advanced adenomas compared to polyp sizes. Complications were recordedonly during operative colonoscopy:early bleeding in 5.07% and late bleeding in 0.58% allcontrolled endoscopically or conservatively.Only in 1 subject was necessary blood transfu-sions.We had 1 post-polypectomy syndrome treated conservatively and 1 perforation success-fully treated by surgery. CONCLUSION:over the period reviewed, 691 colonoscopies wereperformed with excellent completion rates.Complications were within acceptable range-s.Significant pathology (cancer and adenoma) was found in over 2/3 of patients with positiveFIT.The extremely high rate of cancer/adenoma detected can partly be explained by: initialBCSP in our local population, higher incidence of colon cancer/adenomas in our region,relative initial older age of the involved population (most of them older than 60 y.o.),theperformance of FIT and the good bowel preparation in almost all the subjects.The previoushypothesis, however, do not completely explain why in our cohort we found a so highpercentage of malignant (0.19%) and advance lesions (11.79%) in polyps ≤ than 9 mm.In this case a further hypothesis could include an over reading of advanced adenoma.Toconfirm these latest findings a second opinion from another pathologist could be advisable.Table 1
Table 2
Su1145
Prevalence of Missed Polyps in Patients With Poor Bowel Preparation onScreening ColonoscopyReena V. Chokshi, Christine E. Hovis, Dayna S. Early, Jean S. Wang
Background: No clear guidelines exist for the management of patients with poor bowelpreparation on screening colonoscopy. The appropriate timing of subsequent colonoscopyis uncertain, and the prevalence of missed polyps in the setting of poor bowel preparation
S-415 AGA Abstracts
is unknown. Aims: To assess the prevalence of significant gastrointestinal pathology onrepeat colonoscopy in patients with poor bowel preparation on initial screening colonoscopy.Methods: Using our endoscopy database, we identified all patients undergoing average-riskscreening colonoscopy from 2004-2009 with poor, inadequate, or unsatisfactory bowelpreparation on initial screening colonoscopy. We only included colonoscopies that werecompleted to the cecum. Endoscopy and pathology reports were examined to determinethe number, size, and type of polyps detected during colonoscopy. Data from repeat colonos-copies were collected through 2010 for each patient. Results: A total of 373 patients hadpoor, inadequate, or unsatisfactory bowel preparation during initial screening colonoscopy,with an adenoma detection rate of 26.0%. Any polyps detected during the initial colonoscopywere resected. A repeat colonoscopy was completed in 35.7% of the initial cohort. Amongthe 133 patients undergoing repeat colonoscopy, 33.8% had at least one adenoma detectedand 18.0% had multiple or advanced adenomas detected (>= 3 adenomas, one adenoma>= 1 cm, or any adenoma with villous features or high-grade dysplasia). Among patientswith at least one adenoma detected on repeat colonoscopy, 30.4% had no polyps seen oninitial colonoscopy, and the mean time between colonoscopies was 340 days. Among patientswith multiple or advanced adenomas detected, 25% had no polyps seen on initial colonos-copy, and the mean time between colonoscopies was 271 days. Conclusions: Adenomas,including high-risk lesions, were frequently detected on repeat colonoscopy in patients withpoor bowel preparation on initial screening colonoscopy. The mean time between initialand repeat colonoscopies was less than one year, suggesting that these lesions were likelymissed on initial colonoscopy. This study highlights the importance of repeating colonoscopyin patients with poor bowel preparation given a high prevalence of missed adenomas andhigh-risk lesions.
Su1146
Detectability of Colorectal Neoplasia With Fluorine-18-2-Fluoro-2-Deoxy-D-Glucose Positron Emission Tomography/Computed Tomography (FDG-PET/CT)Tomoko Hirakawa, Jun Kato, Yoshihiro Okumura, Keisuke Hori, Sakuma Takahashi,Hideyuki Suzuki, Mitsuhiro Akita, Reiji Higashi, Shunsuke Saito, Eisuke Kaji, ToshioUraoka, Sakiko Hiraoka, Kazuhide Yamamoto
BACKGROUND: The detectability of colorectal neoplasia with fluorine-18-2-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography (FDG-PET/CT) has notfully examined. The purpose of this study was to analyze the detectability of colorectalneoplasia with PET/CT, using colonoscopic findings as controls. METHODS: This study isa retrospective review of medical charts and radiologic and endoscopic images of patientswho underwent both PET/CT and colonoscopy within one year interval. A total of 492patients (306 men and 186 women) with the median age of 66 years old were analyzed.The colon was divided into six segments on PET/CT images: cecum, ascending colon,transverse colon, descending colon, sigmoid colon, and rectum. All tumors suspected ofmalignancy with magnifying colonoscopy were pathologically examined by using specimenscollected with biopsy, polypectomy, or surgical resection, while the pathology of the remain-ders were not always confirmed and they were considered as nonmalignant. After determiningfindings of PET/CT and colonoscopy independently, the results were compared in each ofthe six colonic sites of all examined patients. The detectability of PET/CT was determinedusing colonoscopic examination as a gold standard. The analysis was performed for tumorswith the size of 5 mm or more at colonoscopy. RESULTS: In all, 270 colorectal lesions 5mm or more in size, including 70 pathologically confirmed malignant lesions, were foundin 172 patients with colonoscopy. Among 200 nonmalignant lesions, 164 (82%) were 10mm or smaller, 30 (15 %) were 11-20 mm, and 6 (3 %) were 21 mm or larger. Malignantlesions consisted of 57 colorectal cancers, 10 lymphomas, 2 cancers of adjacent organspenetrating to colorectal lumen, and 1 metastatic tumor. The average size of the malignantlesion was 31 mm (range 6-80). The sensitivity and specificity of PET/CT for detecting anyof colorectal lesions were 36% and 98%, respectively. For detecting lesions of 11 mm orlarger, the sensitivity greatly increased up to 85%, with the specificity not so changed (97%).Moreover, the sensitivity for tumors 21 mm or larger was 96% (48/50), suggesting that thesensitivity of PET/CT for colorectal lesions is size-dependent, and that large part of clinicallyrelevant polyps were detectable with PET/CT. Moreover, only for malignant lesions of 11mmor larger, the sensitivity was high with 97%. Multivariable analysis on tumors 11 mm orlarger revealed that the size 20 mm or smaller (p = 0.0023), localization in the right colon(p = 0.048), and women (p = 0.019) were the factors predictive of false-negative with PET/CT. CONCLUSION: Large part of relevant colonic lesions were detectable with PET/CT,but it should be noted that the modality are relatively likely to miss tumors in the right-colon and those in women.
Su1147
Pathology of Cancers Diagnosed in Participants That Tested Negative in theFirst Round of a FOBT Based Screening PilotKaram S. Boparai, Maaike Denters, Patrick M. Bossuyt, Paul Fockens, Carel V. Noesel,Evelien Dekker
Background and aims: Mass-screening for colorectal cancer (CRC) using fecal occult bloodtests (FOBT) is aimed at detecting cancer in an early stage. FOBT has an imperfect sensitivityfor the presence of CRC, thus leaving a proportion of CRCs undetected. This may eitherbe caused by a relatively decreased vascularisation of these CRCs undetected by FOBTor by an increased progression rate to CRC in these patients. We aimed to analyze thehistopathological and molecular characteristics of CRCs that were not detected by FOBT.Methods: In 2006 and 2008, a cohort of around 10,000 average risk persons aged 50 to74 were invited to participate in a Dutch FOBT based CRC screening pilot randomizingguaiac and FIT. All persons that tested negative in the first round but who were diagnosedwith CRC within the following 3 years were identified through a cross-linkage between thestudy database and the database of the National Cancer Registry. Of all identified cases,clinical information was obtained from the hospital where the patient is treated. After revision,microsatellite instability- (MSI) and somatic mutation analysis was performed in the APC(mutation cluster region), KRAS (exon 2) and BRAF (exon 15) genes of all specimens.
AG
AA
bst
ract
s