Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults

BACKGROUND

We evaluated the performance characteristics of computed tomographic (CT) virtual colonoscopy for the detection of colorectal neoplasia in an average-risk screening population.

METHODS

A total of 1233 asymptomatic adults (mean age, 57.8 years) underwent same-day virtual and optical colonoscopy. Radiologists used the three-dimensional endoluminal display for the initial detection of polyps on CT virtual colonoscopy. For the initial examination of each colonic segment, the colonoscopists were unaware of the findings on virtual colonoscopy, which were revealed to them before any subsequent reexamination. The sensitivity and specificity of virtual colonoscopy and the sensitivity of optical colonoscopy were calculated with the use of the findings of the final, unblinded optical colonoscopy as the reference standard.

RESULTS

The sensitivity of virtual colonoscopy for adenomatous polyps was 93.8 percent for polyps at least 10 mm in diameter, 93.9 percent for polyps at least 8 mm in diameter, and 88.7 percent for polyps at least 6 mm in diameter. The sensitivity of optical colonoscopy for adenomatous polyps was 87.5 percent, 91.5 percent, and 92.3 percent for the three sizes of polyps, respectively. The specificity of virtual colonoscopy for adenomatous polyps was 96.0 percent for polyps at least 10 mm in diameter, 92.2 percent for polyps at least 8 mm in diameter, and 79.6 percent for polyps at least 6 mm in diameter. Two polyps were malignant; both were detected on virtual colonoscopy, and one of them was missed on optical colonoscopy before the results on virtual colonoscopy were revealed.

CONCLUSIONS

CT virtual colonoscopy with the use of a three-dimensional approach is an accurate screening method for the detection of colorectal neoplasia in asymptomatic average-risk adults and compares favorably with optical colonoscopy in terms of the detection of clinically relevant lesions.

Use of colonoscopy to screen asymptomatic adults for colorectal cancer. Veterans Affairs Cooperative Study Group 380

BACKGROUND AND METHODS

The role of colonoscopy in screening for colorectal cancer is uncertain. At 13 Veterans Affairs Medical Centers, we performed colonoscopy to determine the prevalence and location of advanced colonic neoplasms and the risk of advanced proximal neoplasia in asymptomatic patients (age range, 50 to 75 years) with or without distal neoplasia. Advanced colonic neoplasia was defined as an adenoma that was 10 mm or more in diameter, a villous adenoma, an adenoma with high-grade dysplasia, or invasive cancer. In patients with more than one neoplastic lesion, classification was based on the most advanced lesion.

RESULTS

Of 17,732 patients screened for enrollment, 3196 were enrolled; 3121 of the enrolled patients (97.7 percent) underwent complete examination of the colon. The mean age of the patients was 62.9 years, and 96.8 percent were men. Colonoscopic examination showed one or more neoplastic lesions in 37.5 percent of the patients, an adenoma with a diameter of at least 10 mm or a villous adenoma in 7.9 percent, an adenoma with high-grade dysplasia in 1.6 percent, and invasive cancer in 1.0 percent. Of the 1765 patients with no polyps in the portion of the colon that was distal to the splenic flexure, 48 (2.7 percent) had advanced proximal neoplasms. Patients with large adenomas (> or = 10 mm) or small adenomas (< 10 mm) in the distal colon were more likely to have advanced proximal neoplasia than were patients with no distal adenomas (odds ratios, 3.4 [95 percent confidence interval, 1.8 to 6.5] and 2.6 (95 percent confidence interval, 1.7 to 4.1], respectively). However, 52 percent of the 128 patients with advanced proximal neoplasia had no distal adenomas.

CONCLUSIONS

Colonoscopic screening can detect advanced colonic neoplasms in asymptomatic adults. Many of these neoplasms would not be detected with sigmoidoscopy.

Long-term colorectal-cancer incidence and mortality after lower endoscopy

BACKGROUND

Colonoscopy and sigmoidoscopy provide protection against colorectal cancer, but the magnitude and duration of protection, particularly against cancer of the proximal colon, remain uncertain.

METHODS

We examined the association of the use of lower endoscopy (updated biennially from 1988 through 2008) with colorectal-cancer incidence (through June 2010) and colorectal-cancer mortality (through June 2012) among participants in the Nurses' Health Study and the Health Professionals Follow-up Study.

RESULTS

Among 88,902 participants followed over a period of 22 years, we documented 1815 incident colorectal cancers and 474 deaths from colorectal cancer. With endoscopy as compared with no endoscopy, multivariate hazard ratios for colorectal cancer were 0.57 (95% confidence interval [CI], 0.45 to 0.72) after polypectomy, 0.60 (95% CI, 0.53 to 0.68) after negative sigmoidoscopy, and 0.44 (95% CI, 0.38 to 0.52) after negative colonoscopy. Negative colonoscopy was associated with a reduced incidence of proximal colon cancer (multivariate hazard ratio, 0.73; 95% CI, 0.57 to 0.92). Multivariate hazard ratios for death from colorectal cancer were 0.59 (95% CI, 0.45 to 0.76) after screening sigmoidoscopy and 0.32 (95% CI, 0.24 to 0.45) after screening colonoscopy. Reduced mortality from proximal colon cancer was observed after screening colonoscopy (multivariate hazard ratio, 0.47; 95% CI, 0.29 to 0.76) but not after sigmoidoscopy. As compared with colorectal cancers diagnosed in patients more than 5 years after colonoscopy or without any prior endoscopy, those diagnosed in patients within 5 years after colonoscopy were more likely to be characterized by the CpG island methylator phenotype (CIMP) (multivariate odds ratio, 2.19; 95% CI, 1.14 to 4.21) and microsatellite instability (multivariate odds ratio, 2.10; 95% CI, 1.10 to 4.02).

CONCLUSIONS

Colonoscopy and sigmoidoscopy were associated with a reduced incidence of cancer of the distal colorectum; colonoscopy was also associated with a modest reduction in the incidence of proximal colon cancer. Screening colonoscopy and sigmoidoscopy were associated with reduced colorectal-cancer mortality; only colonoscopy was associated with reduced mortality from proximal colon cancer. Colorectal cancer diagnosed within 5 years after colonoscopy was more likely than cancer diagnosed after that period or without prior endoscopy to have CIMP and microsatellite instability. (Funded by the National Institutes of Health and others.).

Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies

BACKGROUND & AIMS

The miss rate of colonoscopy for neoplasms is poorly understood. The aim of this study was to determine the miss rate of colonoscopy by same day back-to-back colonoscopy.

METHODS

Two consecutive same day colonoscopies were performed in 183 patients. The patients were randomized to undergo the second colonoscopy by the same or a different endoscopist and in the same or different position.

RESULTS

The overall miss rate for adenomas was 24%, 27% for adenomas < or = 5 mm, 13% for adenomas 6-9 mm, and 6% for adenomas > or = 1 cm. Patients with two or more adenomas at the first examination were more likely than patients with no or one adenoma detected at the first examination to have one or more adenomas at the second examination (odds ratio, 3.3; 95% confidence interval, 1.69-6.46). Right colon adenomas were missed more often (27%) than left colon adenomas (21%), but the difference was not significant. There was evidence of variation in sensitivity between endoscopists, but significant miss rates for small adenomas were found among essentially all endoscopists.

CONCLUSIONS

Using current colonoscopic technology, there are significant miss rates for adenomas < 1 cm even with meticulous colonoscopy. Miss rates are low for adenomas > or = 1 cm. The results suggest the need for improvements in colonoscopic technology.

The LKB1 tumor suppressor negatively regulates mTOR signaling

Germline mutations in LKB1, TSC2, or PTEN tumor suppressor genes result in hamartomatous syndromes with shared tumor biological features. The recent observations of LKB1-mediated activation of AMP-activated protein kinase (AMPK) and AMPK inhibition of mTOR through TSC2 prompted us to examine the biochemical and biological relationship between LKB1 and mTOR regulation. Here, we report that LKB1 is required for repression of mTOR under low ATP conditions in cultured cells in an AMPK- and TSC2-dependent manner, and that Lkb1 null MEFs and the hamartomatous gastrointestinal polyps from Lkb1 mutant mice show elevated signaling downstream of mTOR. These findings position aberrant mTOR activation at the nexus of these germline neoplastic conditions and suggest the use of mTOR inhibitors in the treatment of Peutz-Jeghers syndrome.

Serrated lesions of the colorectum: review and recommendations from an expert panel

Serrated lesions of the colorectum are the precursors of perhaps one-third of colorectal cancers (CRCs). Cancers arising in serrated lesions are usually in the proximal colon, and account for a disproportionate fraction of cancer identified after colonoscopy. We sought to provide guidance for the clinical management of serrated colorectal lesions based on current evidence and expert opinion regarding definitions, classification, and significance of serrated lesions. A consensus conference was held over 2 days reviewing the topic of serrated lesions from the perspectives of histology, molecular biology, epidemiology, clinical aspects, and serrated polyposis. Serrated lesions should be classified pathologically according to the World Health Organization criteria as hyperplastic polyp, sessile serrated adenoma/polyp (SSA/P) with or without cytological dysplasia, or traditional serrated adenoma (TSA). SSA/P and TSA are premalignant lesions, but SSA/P is the principal serrated precursor of CRCs. Serrated lesions have a distinct endoscopic appearance, and several lines of evidence suggest that on average they are more difficult to detect than conventional adenomatous polyps. Effective colonoscopy requires an endoscopist trained in the endoscopic appearance of serrated lesions. We recommend that all serrated lesions proximal to the sigmoid colon and all serrated lesions in the rectosigmoid > 5 mm in size, be completely removed. Recommendations are made for post-polypectomy surveillance of serrated lesions and for surveillance of serrated polyposis patients and their relatives.

Accuracy of CT colonography for detection of large adenomas and cancers

BACKGROUND

Computed tomographic (CT) colonography is a noninvasive option in screening for colorectal cancer. However, its accuracy as a screening tool in asymptomatic adults has not been well defined.

METHODS

We recruited 2600 asymptomatic study participants, 50 years of age or older, at 15 study centers. CT colonographic images were acquired with the use of standard bowel preparation, stool and fluid tagging, mechanical insufflation, and multidetector-row CT scanners (with 16 or more rows). Radiologists trained in CT colonography reported all lesions measuring 5 mm or more in diameter. Optical colonoscopy and histologic review were performed according to established clinical protocols at each center and served as the reference standard. The primary end point was detection by CT colonography of histologically confirmed large adenomas and adenocarcinomas (10 mm in diameter or larger) that had been detected by colonoscopy; detection of smaller colorectal lesions (6 to 9 mm in diameter) was also evaluated.

RESULTS

Complete data were available for 2531 participants (97%). For large adenomas and cancers, the mean (+/-SE) per-patient estimates of the sensitivity, specificity, positive and negative predictive values, and area under the receiver-operating-characteristic curve for CT colonography were 0.90+/-0.03, 0.86+/-0.02, 0.23+/-0.02, 0.99+/-<0.01, and 0.89+/-0.02, respectively. The sensitivity of 0.90 (i.e., 90%) indicates that CT colonography failed to detect a lesion measuring 10 mm or more in diameter in 10% of patients. The per-polyp sensitivity for large adenomas or cancers was 0.84+/-0.04. The per-patient sensitivity for detecting adenomas that were 6 mm or more in diameter was 0.78.

CONCLUSIONS

In this study of asymptomatic adults, CT colonographic screening identified 90% of subjects with adenomas or cancers measuring 10 mm or more in diameter. These findings augment published data on the role of CT colonography in screening patients with an average risk of colorectal cancer. (ClinicalTrials.gov number, NCT00084929; American College of Radiology Imaging Network [ACRIN] number, 6664.)

Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings

BACKGROUND AND METHODS

The clinical significance of a distal colorectal polyp is uncertain. We determined the risk of advanced proximal neoplasia, defined as a polyp with villous features, a polyp with high-grade dysplasia, or cancer, among persons with distal hyperplastic or neoplastic polyps as compared with the risk among persons with no distal polyps. We analyzed data from 1994 consecutive asymptomatic adults (age, 50 years or older) who underwent colonoscopic screening for the first time between September 1995 and December 1998 as part of a program sponsored by an employer. The location and histologic features of all polyps were recorded. Colonoscopy to the level of the cecum was completed in 97.0 percent of the patients.

RESULTS

Sixty-one patients (3.1 percent) had advanced lesions in the distal colon, including 5 with cancer, and 50 (2.5 percent) had advanced proximal lesions, including 7 with cancer. Twenty-three patients with advanced proximal neoplasms (46 percent) had no distal polyps. The prevalence of advanced proximal neoplasia among patients with no distal polyps was 1.5 percent (23 cases among 1564 persons; 95 percent confidence interval, 0.9 to 2.1 percent). Among patients with distal hyperplastic polyps, those with distal tubular adenomas, and those with advanced distal polyps, the prevalence of advanced proximal neoplasia was 4.0 percent (8 cases among 201 patients), 7.1 percent (12 cases among 168 patients), and 11.5 percent (7 cases among 61 patients), respectively. The relative risk of advanced proximal neoplasia, adjusted for age and sex, was 2.6 for patients with distal hyperplastic polyps, 4.0 for those with distal tubular adenomas, and 6.7 for those with advanced distal polyps, as compared with patients who had no distal polyps. Older age and male sex were associated with an increased risk of advanced proximal neoplasia (relative risk, 1.3 for every five years of age and 3.3 for male sex).

CONCLUSIONS

Asymptomatic persons 50 years of age or older who have polyps in the distal colon are more likely to have advanced proximal neoplasia than are persons without distal polyps. However, if colonoscopic screening is performed only in persons with distal polyps, about half the cases of advanced proximal neoplasia will not be detected.

Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. The National Polyp Study Workgroup

BACKGROUND

The identification and removal of adenomatous polyps and post-polypectomy surveillance are considered to be important for the control of colorectal cancer. In current practice, the intervals between colonoscopies after polypectomy are variable, often a year long, and not based on data from randomized clinical trials. We sought to determine whether follow-up colonoscopy at three years would detect important colonic lesions as well as follow-up colonoscopy at both one and three years.

METHODS

Patients were eligible if they had one or more adenomas, no previous polypectomy, and a complete colonoscopy and all their polyps had been removed. They were randomly assigned to have follow-up colonoscopy at one and three years or at three years only. The two study end points were the detection of any adenoma, and the detection of adenomas with advanced pathological features (defined as those > 1 cm in diameter and those with high-grade dysplasia or invasive cancer).

RESULTS

Of 2632 eligible patients, 1418 were randomly assigned to the two follow-up groups, 699 to the two-examination group and 719 to the one-examination group. The percentage of patients with adenomas in the group examined at one and three years was 41.7 percent, as compared with 32.0 percent in the group examined at three years (P = 0.006). The percentage of patients with adenomas with advanced pathological features was the same in both groups (3.3 percent).

CONCLUSIONS

Colonoscopy performed three years after colonoscopic removal of adenomatous polyps detects important colonic lesions as effectively as follow-up colonoscopy after both one and three years. An interval of at least three years is recommended before follow-up colonoscopy after both one and three years. An interval of at least three years is recommended before follow-up examination after colonoscopic removal of newly diagnosed adenomatous polyps. Adoption of this recommendation nationally should reduce the cost of post-polypectomy surveillance and screening.

Natural history of untreated colonic polyps

The natural history of untreated colonic polyps is uncertain. A retrospective review of Mayo Clinic records from a 6-yr period just before the advent of colonoscopy identified 226 patients with colonic polyps greater than or equal to 10 mm in diameter in whom periodic radiographic examination of the colon was elected over excisional therapy. In all patients, follow-up of polyps spanned at least 12 mo (mean, 68 mo; range, 12-229 mo) and included at least two barium enema examinations (mean, 5.2; range, 2-17). During the follow-up period, 83 polyps (37%) enlarged. Twenty-one invasive carcinomas were identified at the site of the index polyp at a mean follow-up of 108 mo (range, 24-225 mo). Actuarial analysis revealed that the cumulative risk of diagnosis of cancer at the polyp site at 5, 10, and 20 yr was 2.5%, 8%, and 24%, respectively. In addition, 11 invasive cancers were found at a site remote from the index polyp during the same follow-up period. These data further support the recommendation for excision of all colonic polyps greater than or equal to 10 mm in diameter. Periodic examination of the entire colon is recommended in this group of patients to identify neoplasms arising at a site remote from the index polyp. Although this study has limitations inherent to any retrospective analysis, comparable prospective data are unlikely to be available in the future because of the current widespread availability of colonoscopy.

Poor-prognosis colon cancer is defined by a molecularly distinct subtype and develops from serrated precursor lesions

Colon cancer is a clinically diverse disease. This heterogeneity makes it difficult to determine which patients will benefit most from adjuvant therapy and impedes the development of new targeted agents. More insight into the biological diversity of colon cancers, especially in relation to clinical features, is therefore needed. We demonstrate, using an unsupervised classification strategy involving over 1,100 individuals with colon cancer, that three main molecularly distinct subtypes can be recognized. Two subtypes have been previously identified and are well characterized (chromosomal-instable and microsatellite-instable cancers). The third subtype is largely microsatellite stable and contains relatively more CpG island methylator phenotype-positive carcinomas but cannot be identified on the basis of characteristic mutations. We provide evidence that this subtype relates to sessile-serrated adenomas, which show highly similar gene expression profiles, including upregulation of genes involved in matrix remodeling and epithelial-mesenchymal transition. The identification of this subtype is crucial, as it has a very unfavorable prognosis and, moreover, is refractory to epidermal growth factor receptor-targeted therapy.

BRAF mutation is associated with DNA methylation in serrated polyps and cancers of the colorectum

BACKGROUND AND AIMS

Mutations in BRAF have been linked with colorectal cancers (CRC) showing high level microsatellite instability (MSI-H). However, the distribution of BRAF mutations in MSI-H cancers remains to be clarified with respect to precursor lesions and the CpG island methylator phenotype (CIMP).

METHODS

Forty three hyperplastic polyps (HP), nine mixed polyps (MP), five serrated adenomas (SA), 28 conventional adenomas (AD), 18 hereditary non-polyposis colorectal cancers (HNPCC), and 127 sporadic CRC (46 MSI-H and 81 non-MSI-H) were collected from patients undergoing colectomy for either CRC or hyperplastic polyposis. Twenty five of 57 serrated lesions were derived from four patients with hyperplastic polyposis. HP were further subdivided according to recently documented morphological criteria into 27 classical HP and 16 variant lesions described as "sessile serrated adenoma" (SSA). All tumours were screened for BRAF activating mutations.

RESULTS

The BRAF mutation was more frequent in SSA (75%) and MP (89%) than in classical HP (19%), SA (20%), and AD (0%) (p<0.0001), and also in sporadic MSI-H cancers (76%) compared with HNPCC (0%) and sporadic non-MSI-H cancers (9%) (p<0.0001). The BRAF mutation was identified more often in CIMP-high serrated polyps (72%) and CIMP-high CRC (77%) than in CIMP-low (30%) and CIMP-negative (13%) polyps (p = 0.002) as well as CIMP-low (18%) and CIMP-negative (0%) CRC (p<0.0001).

CONCLUSIONS

The BRAF mutation was frequently seen in SSA and in sporadic MSI-H CRC, both of which were associated with DNA methylation. Sporadic MSI-H cancers may originate in SSA and not adenomas, and BRAF mutation and DNA methylation are early events in this "serrated" pathway.

Gavage of Fecal Samples From Patients With Colorectal Cancer Promotes Intestinal Carcinogenesis in Germ-Free and Conventional Mice

BACKGROUND & AIMS

Altered gut microbiota is implicated in development of colorectal cancer (CRC). Some intestinal bacteria have been reported to potentiate intestinal carcinogenesis by producing genotoxins, altering the immune response and intestinal microenvironment, and activating oncogenic signaling pathways. We investigated whether stool from patients with CRC could directly induce colorectal carcinogenesis in mice.

METHODS

We obtained stored stool samples from participants in a metagenome study performed in Hong Kong. Conventional (male C57BL/6) mice were given azoxymethane to induce colon neoplasia after receiving a course of antibiotics in drinking water. Mice were gavaged twice weekly with stool from 5 patients with CRC or 5 healthy individuals (controls) for 5 weeks. Germ-free C57BL/6 mice were gavaged once with stool from 5 patients with CRC or 5 controls. We collected intestinal tissues from mice and performed histology, immunohistochemistry, expression microarray, quantitative polymerase chain reaction, immunoblot, and flow cytometry analyses. We performed 16S ribosomal RNA gene sequencing analysis of feces from mice.

RESULTS

Significantly higher proportions of conventional mice fed with stool from individuals with CRC than control stool developed high-grade dysplasia (P < .05) and macroscopic polyps (P < .01). We observed a higher proportion of proliferating (Ki-67-positive) cells in colons of germ-free mice fed with stool from patients with CRC vs those fed with stool from controls (P < .05). Feces from germ-free and conventional mice fed with stool from patients with CRC vs controls contained different microbial compositions, with lower richness in mice fed with stool from patients with CRC. Intestines collected from conventional and germ-free mice fed with stool from patients with CRC had increased expression of cytokines that modulate inflammation, including C-X-C motif chemokine receptor 1, C-X-C motif chemokine receptor 2, interleukin 17A (IL17A), IL22, and IL23A. Intestines from conventional and germ-free mice fed with stool from patients with CRC contained higher proportions of T-helper 1 (Th1) cells (2.25% vs 0.44%) and Th17 cells (2.08% vs 0.31%) (P < .05 for each) than mice fed with stool from controls. Real-time polymerase chain reaction arrays revealed up-regulation of genes involved in cell proliferation, stemness, apoptosis, angiogenesis, invasiveness, and metastasis in mice fed with stool from patients with CRC.

CONCLUSIONS

We fed stool samples from patients with CRC and heathy individuals to germ-free mice and conventional mice with azoxymethane. We found stool from patients with CRC to increase the numbers of polyps, levels of intestinal dysplasia and proliferation, markers of inflammation, and proportions of Th1 and Th17 cells in colon, compared with stool from individuals without CRC. This study provides evidence that the fecal microbiota from patients with CRC can promote tumorigenesis in germ-free mice and mice given a carcinogen.

Management of early invasive colorectal cancer. Risk of recurrence and clinical guidelines

PURPOSE

The purpose of this study was the evaluation of various factors in the formulation of guidelines for treatment of early invasive colorectal cancer, in which malignant cells extend through the muscularis mucosa into the submucosa but do not deeply invade the muscularis propria.

METHOD

A total of 182 patients were followed for at least five years or until death, with early invasive cancer diagnosed between 1982 and 1989. Patients were grouped according to the level of invasion, as follows: 64 patients with slight carcinoma invasion of the muscularis mucosa (200-300 microns; sm1), 82 with intermediate invasion (sm2), and 36 with carcinoma invasion extending to the inner surface of the muscularis propria (sm3).

RESULT

The configuration, diameter, and histologic grade of adenocarcinoma and lymphovascular invasion were correlated with level of invasion. After endoscopic polypectomy or local resection, 4 patients showed local recurrence and 13 patients showed lymph node metastasis. None of these 17 patients had sm1 disease. The level of invasion, configuration, and location were significant risk factors for development of lymph node metastasis or local recurrence (P < 0.05), but lymphovascular invasion, histologic grade, and diameter were not risk factors.

CONCLUSIONS

Preoperative assessment of the level of invasion using this classification, in which the submucosa is divided into three depths, may decrease the incidence of unnecessary surgery for sessile polyps. Assessment according to the level of invasion is useful in the formulation of appropriate guidelines for the treatment of early invasive cancer.

Endoscopic mucosal resection outcomes and prediction of submucosal cancer from advanced colonic mucosal neoplasia

BACKGROUND & AIMS

Large sessile colonic polyps usually are managed surgically, with significant morbidity and potential mortality. There have been few prospective, intention-to-treat, multicenter studies of endoscopic mucosal resection (EMR). We investigated whether endoscopic criteria can predict invasive disease and direct the optimal treatment strategy.

METHODS

The Australian Colonic Endoscopic (ACE) resection study group conducted a prospective, multicenter, observational study of all patients referred for EMR of sessile colorectal polyps that were 20 mm or greater in size (n=479, mean age, 68.5 y; mean lesion size, 35.6 mm). We analyzed data on lesion characteristics and procedural, clinical, and histologic outcomes. Multiple logistic regression analysis identified independent predictors of EMR efficacy and recurrence of adenoma, based on findings from follow-up colonoscopy examinations.

RESULTS

Risk factors for submucosal invasion were as follows: Paris classification 0-IIa+c morphology, nongranular surface, and Kudo pit pattern type V. The most commonly observed lesion (0-IIa granular) had a low rate of submucosal invasion (1.4%). EMR was effective at completely removing the polyp in a single session in 89.2% of patients; risk factors for lack of efficacy included a prior attempt at EMR (odds ratio [OR], 3.8; 95% confidence interval, 1.77-7.94; P=.001) and ileocecal valve involvement (OR, 3.4; 95% confidence interval, 1.20-9.52; P=.021). Independent predictors of recurrence after effective EMR were lesion size greater than 40 mm (OR, 4.37; 95% confidence interval, 2.43-7.88; P<.001) and use of argon plasma coagulation (OR, 3.51; 95% confidence interval, 1.69-7.27; P=.0017). There were no deaths from EMR; 83.7% of patients avoided surgery.

CONCLUSIONS

Large sessile colonic polyps can be managed safely and effectively by endoscopy. Endoscopic assessment identifies lesions at increased risk of containing submucosal cancer. The first EMR is an important determinant of patient outcome-a previous attempt is a significant risk factor for lack of efficacy.

Prognostic factors in colorectal carcinomas arising in adenomas: implications for lesions removed by endoscopic polypectomy

Endoscopic polypectomy has become the preferred technique for the removal of most colorectal adenomas. Whether polypectomy alone or segmental colectomy is the appropriate management of the patient whose adenoma contains carcinoma is a controversial issue. We studied 129 colorectal carcinomas that arose in adenomas and in which invasion was no deeper than the submucosa of the underlying colonic wall. The following factors were evaluated: location; gross appearance (sessile versus pedunculated); histologic type of adenoma (tubular, villous, mixed); grade of carcinoma; level of invasion (0--carcinoma confined to the mucosa, 1--head, 2--neck, 3--stalk, 4--submucosa of underlying colonic wall); vascular invasion; and adequacy of excisional margins. Patients were divided into two groups with respect to outcome: adverse (dead from colorectal carcinoma, alive with colorectal carcinoma or positive nodes on colectomy), and favorable (absence of above). Sixty-three patients were treated by polypectomy alone and 66 by colectomy (21 preceded by polypectomy); there were no operative deaths. Mean follow-up was 81 mo. None of 65 patients with carcinoma confined to the mucosa had an adverse outcome, but 8 of 64 patients with invasive carcinoma did. Level 4 invasion (p less than 0.001) and rectal location (p = 0.025) were the only statistically significant adverse prognostic factors. Seven of 28 level 4 lesions and six of 42 rectal lesions had an adverse outcome; level 4 lesions were overrepresented in the rectum (14 of 42; p = 0.032). We conclude that the level of invasion should be the major factor in determining prognosis for the management of carcinoma arising in an adenoma.

A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps

BACKGROUND

Virtual colonoscopy is a new method of imaging the colon in which thin-section, helical computed tomography (CT) is used to generate high-resolution, two-dimensional axial images. Three-dimensional images of the colon simulating those obtained with conventional colonoscopy are then reconstructed off-line. We compared the performance of virtual and conventional colonoscopy for the detection of colorectal polyps.

METHODS

We prospectively studied 100 patients at high risk for colorectal neoplasia (60 men and 40 women; mean age, 62 years). We performed virtual colonoscopy immediately before conventional colonoscopy. We inserted a rectal tube and insufflated the colon with air to the maximal level that the patient could tolerate. We administered 1 mg of glucagon intravenously immediately before CT scanning to minimize the degree of smooth-muscle spasm and peristalsis and to reduce the patient's discomfort.

RESULTS

The entire colon was clearly seen by virtual colonoscopy in 87 patients and by conventional colonoscopy in 89. Fifty-one patients had normal findings on conventional colonoscopy. In the other 49, we identified a total of 115 polyps and 3 carcinomas. Virtual colonoscopy identified all 3 cancers, 20 of 22 polyps that were 10 mm or more in diameter (91 percent), 33 of 40 that were 6 to 9 mm (82 percent), and 29 of 53 that were 5 mm or smaller (55 percent). There were 19 false positive findings of polyps and no false positive findings of cancer. Of the 69 adenomatous polyps, 46 of the 51 that were 6 mm or more in diameter (90 percent) and 12 of the 18 that were 5 mm or smaller (67 percent) were correctly identified by virtual colonoscopy. Although discomfort was not specifically recorded, none of the patients requested that virtual colonoscopy be stopped because of discomfort or pain.

CONCLUSIONS

In a group of patients at high risk for colorectal neoplasia, virtual and conventional colonoscopy had similar efficacy for the detection of polyps that were 6 mm or more in diameter.

Morphologic reappraisal of serrated colorectal polyps

The "hyperplastic polyp" is considered a benign lesion with no malignant potential, whereas "serrated adenoma" is a precursor of adenocarcinoma. The morphologic complexity of the serrated adenoma varies from being clearly adenomatous to being difficult to distinguish from hyperplastic polyp, which creates a need for more detailed morphologic analysis of all serrated polyps. We evaluated 24 morphologic variables in 289 serrated polyps from the colon and rectum. Cluster analysis and discriminant analysis were performed. A subset of polyps was immunostained for hMLH1 and hMSH2. Major differences were found between right-sided and left-sided polyps. A distinct group of serrated polyps with abnormal proliferation was identified throughout the colon and rectum. These polyps demonstrated decreased expression of hMHL1 and hMSH2 compared with polyps with normal proliferation. Left-sided serrated polyps with normal proliferation further clustered into three groups: vesicular cell-type, goblet cell-type, and mucin-poor-type. We recommend evaluation of the localization, size, and morphologic features when serrated polyps are included in colorectal carcinogenesis research. Polyps with abnormal proliferation are similar to the polyps in "hyperplastic polyposis" and, because of their decreased expression of hMLH1 and hMSH2, may be the subset of polyps associated with the development of colorectal carcinoma via the microsatellite instability pathway.

Deep Learning Localizes and Identifies Polyps in Real Time With 96% Accuracy in Screening Colonoscopy

BACKGROUND & AIMS

The benefit of colonoscopy for colorectal cancer prevention depends on the adenoma detection rate (ADR). The ADR should reflect the adenoma prevalence rate, which is estimated to be higher than 50% in the screening-age population. However, the ADR by colonoscopists varies from 7% to 53%. It is estimated that every 1% increase in ADR lowers the risk of interval colorectal cancers by 3%-6%. New strategies are needed to increase the ADR during colonoscopy. We tested the ability of computer-assisted image analysis using convolutional neural networks (CNNs; a deep learning model for image analysis) to improve polyp detection, a surrogate of ADR.

METHODS

We designed and trained deep CNNs to detect polyps using a diverse and representative set of 8,641 hand-labeled images from screening colonoscopies collected from more than 2000 patients. We tested the models on 20 colonoscopy videos with a total duration of 5 hours. Expert colonoscopists were asked to identify all polyps in 9 de-identified colonoscopy videos, which were selected from archived video studies, with or without benefit of the CNN overlay. Their findings were compared with those of the CNN using CNN-assisted expert review as the reference.

RESULTS

When tested on manually labeled images, the CNN identified polyps with an area under the receiver operating characteristic curve of 0.991 and an accuracy of 96.4%. In the analysis of colonoscopy videos in which 28 polyps were removed, 4 expert reviewers identified 8 additional polyps without CNN assistance that had not been removed and identified an additional 17 polyps with CNN assistance (45 in total). All polyps removed and identified by expert review were detected by the CNN. The CNN had a false-positive rate of 7%.

CONCLUSION

In a set of 8,641 colonoscopy images containing 4,088 unique polyps, the CNN identified polyps with a cross-validation accuracy of 96.4% and an area under the receiver operating characteristic curve of 0.991. The CNN system detected and localized polyps well within real-time constraints using an ordinary desktop machine with a contemporary graphics processing unit. This system could increase the ADR and decrease interval colorectal cancers but requires validation in large multicenter trials.

Mixed hyperplastic adenomatous polyps/serrated adenomas. A distinct form of colorectal neoplasia

We present the clinicopathologic characteristics of 110 colorectal mixed hyperplastic adenomatous polyps (MHAP) that exhibited the architectural but not the cytologic features of a hyperplastic polyp. They are compared with 60 traditional adenomas, 40 hyperplastic polyps, and five colonic polyps that contained admixed but well-defined hyperplastic and adenomatous glands (HP/AD). The patients with MHAP ranged in age from 15 to 88 years (mean, 63 years). Five patients had two or more (up to seven) lesions. MHAP measured 0.2-7.5 cm in diameter. They were distributed throughout the colorectum, but a slight preponderance of large lesions (more than 1.0 cm) occurred in the cecum and appendix. All MHAP were characterized by a serrated glandular pattern simulating that seen in hyperplasia (27% of MHAP were initially diagnosed as hyperplastic polyps). However, MHAP were distinguished by the presence of goblet cell immaturity, upper zone mitoses, prominence of nucleoli, and the absence of a thickened collagen table. Although surface mitotic activity, nuclear pseudostratification, and nuclear cytoplasmic ratio were greater in MHAP than in hyperplastic polyps, they were slightly less than in traditional adenomas. Thirty-seven percent of MHAP contained foci of significant dysplasia; 11% contained areas of intramucosal carcinoma. We conclude that these lesions reflect a morphologically unique variant of adenoma and suggest that they be termed "serrated adenoma" in order to emphasize their neoplastic nature. We further offer the hypothesis that MHAP may arise from the neoplastic transformation of a more differentiated cell in the crypt than the traditional adenoma.

Flat and depressed colonic neoplasms: a prospective study of 1000 colonoscopies in the UK

BACKGROUND

Flat and depressed colorectal tumours were originally thought to be unique to the Japanese population. Recently there have been reports of flat and depressed lesions in western countries but they have been thought to be uncommon.

METHODS

In this prospective study, 1000 consecutive patients attending for routine colonoscopy were examined for flat or depressed lesions. The examinations were done by one European colonoscopist using methods developed in Japan.

FINDINGS

321 adenomas were found: 202 (63%) were polypoid, 36% (117) were flat and 2 (0.6%) appeared depressed. Most adenomas contained areas of mild or moderate dysplasia but 10% (31) were severely dysplastic. Six Dukes' A adenocarcinomas were identified together with 25 more advanced adenocarcinomas. The likelihood of Dukes' A cancer or severe dysplasia increased from 4% (3/70) in small flat lesions, to 6% (9/154) in small polyps, 16% (8/50) in larger polyps, 29% (14/49) in large flat lesions, and 75% (3/4) in depressed lesions. 54% (20/37) lesions containing severe dysplasia or Dukes' A carcinoma were flat or depressed.

INTERPRETATION

The polyp-carcinoma hypothesis prompts colonoscopists to search only for polypoid lesions when screening for cancer, and many early colorectal neoplasms may therefore be missed. Colonoscopists require training in the recognition of flat and depressed lesions to detect colorectal tumours in the early stages.

High prevalence of sessile serrated adenomas with BRAF mutations: a prospective study of patients undergoing colonoscopy

BACKGROUND & AIMS

Sporadic colorectal cancers with a high degree of microsatellite instability are a clinically distinct subgroup with a high incidence of BRAF mutation and are widely considered to develop from serrated polyps. Previous studies of serrated polyps have been highly selected and largely retrospective. This prospective study examined the prevalence of sessile serrated adenomas and determined the incidence of BRAF and K-ras mutations in different types of polyps.

METHODS

An unselected consecutive series of 190 patients underwent magnifying chromoendoscopy. Polyp location, size, and histologic classification were recorded. All polyps were screened for BRAF V600E and K-ras codon 12 and 13 mutations.

RESULTS

Polyps were detected in 72% of patients. Most (60%) were adenomas (tubular adenomas, tubulovillous adenomas), followed by hyperplastic polyps (29%), sessile serrated adenomas (SSAs; 9%), traditional serrated adenomas (0.7%), and mixed polyps (1.7%). Adenomas were more prevalent in the proximal colon (73%), as were SSAs (75%), which tended to be large (64% >5 mm). The presence of at least one SSA was associated with increased polyp burden (5.0 vs 2.5; P < .0001) and female sex (P < .05). BRAF mutation was rare in adenomas (1/248 [0.4%]) but common in SSAs (78%), traditional serrated adenomas (66%), mixed polyps (57%), and microvesicular hyperplastic polyps (70%). K-ras mutations were significantly associated with goblet cell hyperplastic polyps and tubulovillous adenomas (P < .001).

CONCLUSIONS

The prevalence of SSAs is approximately 9% in patients undergoing colonoscopy. They are associated with BRAF mutation, proximal location, female sex, and presence of multiple polyps. These findings emphasize the importance of identifying and removing these lesions for endoscopic prevention of colorectal cancer.

Comparison of microsatellite instability, CpG island methylation phenotype, BRAF and KRAS status in serrated polyps and traditional adenomas indicates separate pathways to distinct colorectal carcinoma end points

The aim of this study was to compare BRAF and KRAS, CpG island methylator phenotype (CIMP), and microsatellite instability (MSI) status in each of the histologic categories, including end-point carcinomas with residual adenoma, of the serrated polyp neoplasia pathway and the traditional (nonserrated) adenoma-carcinoma sequence. Deoxyribonucleic acid (DNA) was extracted from the selected samples and assayed for BRAF, KRAS2 codon12, 13, CIMP using markers hMLH1, MGMT, MINT1, MINT2, p16, and MSI using an assay for BAT25 and BAT26. A BRAF mutation was present in 82% of serrated carcinomas (SCas), 62% of serrated adenomas (SAs), 83% of serrated polyps with abnormal proliferation (SPAPs-syn. sessile serrated adenoma [SSA]), 76% of microvesicular serrated polyps (MVSPs), and was not found in any of the histologic categories of the traditional adenoma-carcinoma sequence. KRAS2 mutations were found in 43% of the goblet cell serrated polyp (GCSP) category, 13% of MVSPs, 7% of SPAPs, and 24% of SAs; in 26% of large traditional adenoma (lTAs) compared with small traditional adenomas (sTAs) (0/30; P<0.005) and in 37.3% of traditional carcinomas (TCa). CIMP-H (>1 marker positive) was significantly more frequent in SPAP, SA, and SCa compared with MVSP (P<0.05); CIMP-H was present in 10% of sTAs but was found more frequently in lTA (44.4%; OR 7.2; P=0.007) and TCa (38.9%; OR 5.8; P=0.007). Higher CIMP levels (4 or more markers positive) were significantly more frequent in advanced categories of the serrated pathway (SAs [31%] and SCas [30%]) compared with lTAs [0%] and TCAs [3.4%] (OR 12.2; P=0.02). MSI-H was identified only in the adenocarcinoma component of SCas (9/11) or in the contiguous SAs (3/7). The findings indicate that a BRAF mutation is a specific marker for a serrated polyp pathway that has its origin in a hyperplastic polyp (MVSP) and a potential end point as MSI carcinoma. CIMP-High (CIMP-H) develops early in this sequence and MSI-H develops late. The data provided a less complete picture of a second serrated pathway, identified by a KRAS2 mutation in SAs, but showed that the progressive stages of both iterations of the serrated neoplasia pathway are separate and distinct from those of the traditional adenoma-carcinoma sequence.

Real-Time Use of Artificial Intelligence in Identification of Diminutive Polyps During Colonoscopy: A Prospective Study

BACKGROUND

Computer-aided diagnosis (CAD) for colonoscopy may help endoscopists distinguish neoplastic polyps (adenomas) requiring resection from nonneoplastic polyps not requiring resection, potentially reducing cost.

OBJECTIVE

To evaluate the performance of real-time CAD with endocytoscopes (×520 ultramagnifying colonoscopes providing microvascular and cellular visualization of colorectal polyps after application of the narrow-band imaging [NBI] and methylene blue staining modes, respectively).

DESIGN

Single-group, open-label, prospective study. (UMIN [University hospital Medical Information Network] Clinical Trial Registry: UMIN000027360).

SETTING

University hospital.

PARTICIPANTS

791 consecutive patients undergoing colonoscopy and 23 endoscopists.

INTERVENTION

Real-time use of CAD during colonoscopy.

MEASUREMENTS

CAD-predicted pathology (neoplastic or nonneoplastic) of detected diminutive polyps (≤5 mm) on the basis of real-time outputs compared with pathologic diagnosis of the resected specimen (gold standard). The primary end point was whether CAD with the stained mode produced a negative predictive value (NPV) of 90% or greater for identifying diminutive rectosigmoid adenomas, the threshold required to "diagnose-and-leave" nonneoplastic polyps. Best- and worst-case scenarios assumed that polyps lacking either CAD diagnosis or pathology were true- or false-positive or true- or false-negative, respectively.

RESULTS

Overall, 466 diminutive (including 250 rectosigmoid) polyps from 325 patients were assessed by CAD, with a pathologic prediction rate of 98.1% (457 of 466). The NPVs of CAD for diminutive rectosigmoid adenomas were 96.4% (95% CI, 91.8% to 98.8%) (best-case scenario) and 93.7% (CI, 88.3% to 97.1%) (worst-case scenario) with stained mode and 96.5% (CI, 92.1% to 98.9%) (best-case scenario) and 95.2% (CI, 90.3% to 98.0%) (worst-case scenario) with NBI.

LIMITATION

Two thirds of the colonoscopies were conducted by experts who had each experienced more than 200 endocytoscopies; 186 polyps not assessed by CAD were excluded.

CONCLUSION

Real-time CAD can achieve the performance level required for a diagnose-and-leave strategy for diminutive, nonneoplastic rectosigmoid polyps.

PRIMARY FUNDING SOURCE

Japan Society for the Promotion of Science.