Colon Cancer Screening

From Update in Gastroenterology and Hepatology: Applying Sound Principles in Daily Practice

Educational Objectives

The goal of this program is to improve colorectal cancer (CRC) screening. After hearing and assimilating this pro­gram, the clinician will be better able to:

1.   Discuss the similarities in and differences between the guidelines and guideline-making processes of the American Cancer Society and the United States Preventive Services Task Force.

2.   Describe the characteristics of tests used for screening of CRC.

3.   Recognize factors that increase the risk for CRC.

4.   Utilize the guidelines for surveillance in patients with adenomatous polyps and in those with positive family history.

5.   Describe the inherited syndromes associated with CRC.

Faculty Disclosure

In adherence to ACCME Standards for Commercial Support, Audio-Digest requires all faculty and members of the planning committee to disclose relevant financial relationships within the past 12 months that might create any per­sonal conflicts of interest. Any identified conflicts were resolved to ensure that this educational activity promotes quality in health care and not a proprietary business or commercial interest. For this program, the faculty and plan­ning committee reported nothing to disclose.

Acknowledgements

Drs. Ransohoff and Sandler were recorded at Update in Gastroenterology and Hepatology: Applying Sound Princi­ples in Daily Practice, held April 17-19, 2009, in Chapel Hill, NC, and sponsored by the University of North Carolina at Chapel Hill, School of Medicine. The Audio-Digest Foundation thanks the speakers and the University of North Carolina at Chapel Hill, School of Medicine for their cooperation in the production of this program.

Current Issues in Colon Cancer Screening

David F. Ransohoff, MD, Professor of Medicine, Clinical Professor of Epidemiology, and Director, Clinical Research Cur­riculum, University of North Carolina School of Medicine, Chapel Hill

Guidelines: for colorectal cancer (CRC) published in 2008 by American Cancer Society (ACS) with multisociety task force (American Gastroenterological Association, American Society for Gastrointestinal Endoscopy, and American College of Gastroenterology); ACS and United States Preventive Services Task Force (USPSTF) 2 major recommending organizations; similarities  —  colonoscopy every 10 yr and sigmoidoscopy endorsed by both; computed tomographic colonography (CTC), virtual colonoscopy, and stool DNA testing endorsed by ACS and multisociety task force (USPSTF found evidence insufficient); differences in outcome and processes present; process by USPSTF  —structured and explicit; uses rules of evidence, with focus on outcome main one; uses dif­ferent levels of strength of evidence; personnel neutral; conflict of interest clearly managed; expensive; process by ACS  —  not structured or explicit; new rules for CRC screening  —  50% rule (test accepted if able to detect 50% of CRC at one application, ie, if sensitivity ³50%); other rule that prevention better than early detection;

Problems with rules: 50% rule  —  false-positive rate and harms from virtual colonoscopy and extracolonic lesions not considered; outcome of mortality reduction, cost, and program of repeated testing not considered; also whether screening performed at one time or over time not considered; second rule  —  lacks detail, and does not consider what is necessary to achieve prevention (ie, harms and cost of prevention); implication that location and removal of  every polyp required, but polyps common in individuals >50 yr of age; seems to favor colonoscopy

Conflict of interest: biases in making guidelines important; specialty composition of guideline panel likely influ­ences development; specialty societies use guidelines to enlarge areas of expertise in competitive marketplace; fed­eral committees may focus on limiting cost (each has own agenda); as professional, consider organization making guidelines and process of forming guidelines; as clinician, determine which guidelines to follow; process by ACS reportedly politicized; formulating guidelines  —  requires consistent application of methods for gathering, interpret­ing, and rating evidence; promotes transparency and internal consistency; needs clear and sensible rules of evi­dence; if role of profession to serve public before profession, need to carefully manage politics

Screening tests: guaiac-based fecal occult blood test (gFOBT)  —looks at peroxidase-like activity of hemoglobin (Hb); does not look at Hb directly; not strongly sensitive or specific; peroxidase activity from vegetables can cause false-positive result; immunochemical-based (iFOBT)  —  looks directly at human Hb, so can be made more sensi­tive without incurring false-positive rates because of specificity; clinical trial data for guaiac-based tests show that CRC mortality reduced by 15% to 33%; sensitivity for CRC with one application of gFOBT 13% to 33% (60% for immunochemical-based FOBT with same specificity); if gFOBT (Hemoccult II) previously recommended by USP­STF and all recommending organizations as one alternative, case for iFOBT even stronger because of higher sensi­tivity with same specificity; possible role of iFOBT in future—   as primary test in program of repeated screening; as interval test between screening colonoscopies, possibly 5 yr after negative colonoscopy, or if preparation for colo­noscopy suboptimal; not readily available in United States

Virtual colonoscopy: study of CT screening for detection of large adenomas and cancers found high sensitivity (90%) for 1-cm lesions, 86% specificity for 0.5-cm lesions, and false-positive rate 14%; extensive training required for radiologists; issue of whether small lesions should be left in or whether important flat lesions missed; not sup­ported by Medicare due to cost and potential harm from work-up of incidental lesions; future role uncertain; possi­bility of laxative-free preparation

Stool DNA testing: detects CRC mutations shed in stool (mutations in adenomatous polyposis coli (APC), K-ras, and p53 genes); sensitivity 60% with one application; expensive; impractical and not supported by Centers for Medi­care and Medicaid Services (CMS)

Colonoscopy: improvement  possible in preparation of patient and in skill of operator; effective as primary test or work-up for other tests; reportedly 90% protective (from National Polyp Study [NPS]); NPS  —  cohort study; com­pared intervention groups to historical controls; intervention groups followed as to whether subjects developed pol­yps after polypectomy (no polypectomy in controls); results showed significant reduction in incidence of CRC, cumulative incidence of CRC, and years of follow-up; problem that subjects in intervention group had baseline colonoscopy, so individuals with CRC excluded; subjects in control group had no baseline colonoscopy, so CRC not excluded; in control group, CRC that appeared possibly present at baseline; in some cases appearing as failures, possible CRC already present; unable to adjust, unless dwell time of CRC known; therefore, CRC incidence rate exaggerated, and 90% reduction probably overestimate; case-control study by Baxter  —purpose to assess CRC mortality reduction from colonoscopy; administrative data used (not charts; unable to determine significant infor­mation, eg, quality of preparation); results showed reduction of CRC mortality on left by »60% (0% on right); at­tributed to bias, but unknown whether bias alone can explain difference between right and left; possible that colonoscopy not complete in cecum; no chart data and more nongastroenterologists performing colonoscopy at time; possible that different biologies exist for right and left; speaker believes 60% to 70% reduction in CRC mor­tality more fair estimate; CRC screening really postpolypectomy surveillance; when performing surveillance, con­sider individual’s risk for CRC in future and whether frequent colonoscopy incurs greater risk than benefit; with 9 colonoscopies from 50 to 80 yr of age, risk for serious complication 3% and risk for death 0.3%; more effective than most other kinds of cancer screening

How to Screen Individuals at High Risk for
Colon Cancer

Robert S. Sandler, MD, Professor of Medicine and Epidemiology, and Chief of Gastroenterology and Hepatology, Univer­sity of North Carolina School of Medicine, Chapel Hill

Introduction: CRC  —  second leading cause of cancer death in United States; more deaths than from breast cancer or prostate cancer; in 2008, »149,000 new cases and 50,000 deaths; 2 categories of guidelines (first used to detect pol­yps and CRC and another to detect CRC only); all tests in first category structural tests (endoscopy or x-rays); in­clude flexible sigmoidoscopy every 5 yr, colonoscopy every 10 yr, or CTC every 5 yr; other category tests feces; satisfy 50% rule (detect 50% of cancer); 50% low bar at present, since able to detect 70% to 80% of polyps; include guaiac-based tests, fecal immunochemical test (FIT), and stool DNA test; interval for tests unknown

Recommendations from USPSTF: guidelines evidence-based and conservative; only 3 tests recommended, ie, FOBT, sigmoidoscopy, and colonoscopy; for first time, age to stop testing recommended; 3 tests recommended for individuals 50 to 75 yr of age; adults between 76 to 85 yr of age no longer screened routinely; no screening for those >85 yr of age; insufficient evidence for CTC and fecal DNA testing

Individuals at high risk for CRC: factors that increase risk include age, black ethnicity, adenomas, cancer, family history, ulcerative colitis (UC), and familial syndromes

Age: risk increases at 50 yr of age, and screening should start then; for women at 30 yr of age, need to screen 59,000 to diagnose one case of cancer

Ethnicity: American College of Gastroenterology (ACG) recommends that screening in blacks be started at 45 yr of age; at 45 to 49 yr of age, blacks 33% more likely to develop CRC than whites; blacks have higher risk at every age and almost 50% more likely to develop CRC at 55 to 59 yr of age; number needed to screen to find one cancer  —  in blacks, at age 45, almost 3000 individuals (at age 55,900 individuals); blacks  —  less likely to be screened, compared to whites; detect 3 times as many cases if focus on people ³55 yr of age (6 times if focus on people 65 yr of age)

Adenomatous polyps: if present, 3 times higher risk for CRC; surveillance colonoscopy lowers risk; guidelines for follow-up  —  depends on number and size of polyps; if only 1 to 2 in number and small in size, wait 5 to 10 yr for next colonoscopy; if ³3 or large (³1 cm), wait 3 yr; if ³10, interval <3 yr; if unable to remove entire lesion, pa­tient to return in 2 to 6 mo; hyperplastic polyp  —  no malignant potential and requires no special follow-up; rec­ommended interval after negative colonoscopy 10 yr; guidelines for colonoscopy surveillance after CRC  —  from ACS and multisociety task force; clear colon at time of surgery; perform colonoscopy before surgery to ensure that no polyps present anywhere else; if CRC obstructing or patient has emergency surgery, should attempt to complete clearing operation within 3 mo after surgery; once colon cleared, perform colonoscopy at 1 yr (old guidelines recommended 3 yr); if normal, wait another 3 yr, and subsequent interval dependent on findings; if 1 to 3 adenomas found, stretch interval to 5 yr; if large adenoma or multiple adenomas found, interval 3 yr

Family history: increases risk 3-fold; recommendations  —  if first-degree relative >60 yr of age, same screening as for average-risk individuals, but start at 40 yr of age; if first-degree relative <60 yr of age, colonoscopy every 5 yr, starting at 40 yr of age; basis of recommendations data from large cohort study which graphed incidence of CRC in patients with no family history and in those with family history (curves same, but shifted over 10 yr); same recommendations if family history for adenoma rather than CRC

UC: increased risk seen in those with long-standing UC; risk starts 8 yr after onset of disease; risk greater if panco­litis present; in left-sided colitis, recommendation to begin screening 12 to 15 yr after onset of disease; conflict­ing information about whether prognosis worse in patients with CRC in setting of UC; frequency of screening also controversial issue; no clear evidence that surveillance improves survival; general recommendation  —  colonoscopy with biopsy every 1 to 2 yr, taking 4 biopsies every 10 cm (8 to 12 specimens in one bottle, marking location for each); if patient has low-grade dysplasia, perform colectomy or wait 3 to 6 mo; if high-grade dyspla­sia found and confirmed by second pathologist, perform colectomy

Familial adenomatous polyposis (FAP): autosomal dominant; 100% risk for CRC in individuals who carry gene; hundreds to thousands of polyps seen; caused by mutation in APC tumor suppressor gene; location of mutation may influence phenotype; if mutation at end of gene, patient may develop attenuated polyposis (smaller number of polyps); can develop fundic gland polyps (no malignant potential); duodenal adenomas  —  relatively common; risk for duodenal cancer in FAP greater than risk for CRC; removing colon prevents CRC, but difficult to remove duodenum; recommendations  —  in patients with FAP, annual sigmoidoscopy starting at 10 to 12 yr of age (rec­tum always involved); patients with duodenal adenomas have risk for CRC of 3% to 5%, and should have endos­copy every 1 to 3 yr, starting at age 20 yr; small risk for pancreatic cancer, some risk for thyroid cancer (perform thyroid examination), and slight risk for gastric cancer (0.6%); children may develop hepatoblastoma (not seen in adults); able to test for gene, and strategy begins with testing affected relative; if test negative in patient with florid polyposis, test result called indeterminate (unable to find mutation); if unable to find mutation, no need to test other family members; if affected relative positive, test patient; if both relative and patient positive, means gene inherited and must determine phenotype with sigmoidoscopy; if relative positive and patient negative, per­form routine CRC screening as in average-risk patient

MYH-associated polyposis (MAP): if patient has FAP phenotype, but tests negative for gene, test for MAP; MYH mutation another cause of attenuated FAP; defect in base excision repair gene; O₂ free radicals damage DNA bases; if normal mechanism to repair damage absent, polyposis results; accounts for 7% to 20% of APC muta­tion-negative cases; autosomal recessive; typically negative family history; siblings affected; gene not inherited unless also carried by spouse

Lynch syndrome: diagnosis by Amsterdam II criteria; 3-2-1 mnemonic (³3 family members, 2 successive genera­tions, and 1 individual <50 yr of age); difference between Amsterdam I and II criteria not just colon cancers in family, but also hereditary nonpolyposis colorectal cancer (HNPCC)-related cancers (stomach, endometrium, ovaries, and small intestine); risk calculators available; 5 different genes causing HNPCC located on 3 different chromosomes; of genes, most commonly mutated MLH1 and MSH2 (account for 90% of cancers); MSH6 ac­counts for 10%); defect in HNPCC mismatch repair (MMR) gene

Microsatellite instability (MSI): due to MMR failure; addition of short repeated sequences of DNA; 95% of HNPCC tumors have MSI, but 10% to 15% of sporadic cancers have MSI; testing  —  requires microdissecting tu­mor specimen, then performing polymerase chain reaction and comparing results; 5 markers classified as MSI markers, and number present determines whether MSI present; genetic testing  —start with affected relative; con­sider MSI testing or immunohistochemistry on tumor to target which gene to look for; testing of other relatives follows only if affected relative positive; if person known to carry gene, perform colonoscopy every 2 yr, starting at age 20 yr; recommendations for patients with HNPCC  —  annual endometrial and ovarian cancer screening starting at age 30 yr, and includes pelvic examination, endometrial aspirate, and possibly transvaginal ultraso­nography; consider prophylactic hysterectomy; risk for endometrial cancer 60% to 70%; also consider annual urinalysis and skin examination; other risk factors  —obesity; diabetes; previous cholecystectomy; alcohol; ciga­rette smoking; radiation; however, increased risk not significant and no special surveillance required

Suggested Reading

Baxter NN et al: Association of colonoscopy and death from colorectal cancer. Ann Intern Med 150:1, 2009; Bouzourene H et al: A cost-effective algorithm for hereditary nonpolyposis colorectal cancer detection. Am J Clin Pathol 125:823, 2006; Campbell PT et al: Mismatch repair polymorphisms and risk of colon cancer, tumour microsatellite instability and interactions with lifestyle factors. Gut 58:661, 2009; Castells A: MYH-associated polyposis: adenomas and hyperplastic polyps, partners in crime? Gastroenterology 135:1857, 2008; Chang EY et al: A prospective analysis of microsatellite in­stability as a molecular marker in colorectal cancer. Am J Surg 191:646, 2006; Graser A et al: Comparison of CT colonog­raphy, colonoscopy, sigmoidoscopy and faecal occult blood tests for the detection of advanced adenoma in an average risk population. Gut 58:241, 2009; Hundt S et al: Comparative evaluation of immunochemical fecal occult blood tests for colorectal adenoma detection. Ann Intern Med 150:162, 2009; Jenkins MA et al: Colon Cancer Family Registry. Pathology features in Bethesda guidelines predict colorectal cancer microsatellite instability: a population-based study. Gastroenterol­ogy 133:48, 2007; Levi Z et al: A quantitative immunochemical fecal occult blood test for colorectal neoplasia. Ann Intern Med 146:244, 2007; Maisonneuve P et al: Screening and surveillance for the early detection of colorectal cancer and ade­nomatous polyps. Gastroenterology 135:710, 2008; Sieber OM et al: Disease severity and genetic pathways in attenuated familial adenomatous polyposis vary greatly but depend on the site of the germline mutation. Gut 55:1440, 2006; van Ros­sum LG et al: Random comparison of guaiac and immunochemical fecal occult blood tests for colorectal cancer in a screening population. Gastroenterology 135:82, 2008; Whitlock EP et al: Screening for colorectal cancer: a targeted, up­dated systematic review for the U.S. Preventive Services Task Force. Ann Intern Med 149:638, 2008; Zarchy T: Colonos­copy screening in average-risk individuals ages 40-49 versus 50-59 years. Gastroenterology 135:1016, 2008.