Colorectal Cancer Prevention & Early Detection : Update 2008

1. Colorectal Cancer Prevention & Early Detection:Update 2008 Nadim G. Haddad,M.D. Associate Professor Director of GI Fellowship

2. Colorectal Cancer • The third most common cancer in U.S. • 148,810 new cases expected in 2008 • The second deadliest cancer • 49,960 deaths nationwide • More than 1 million Americans living with colorectal cancer

3. Colorectal Cancer Risk Factors • Age • 90% of cases occur in people 50 and older • Gender • slight male predominance, but common in both men and women • Race/Ethnicity • African Americans have highest incidence and mortality rate of all groups in U.S., Hispanics the lowest (with considerable variation depending on country of origin) • Increased rates also documented in Alaska Natives, some American Indian tribes, Ashkenazi Jews

4. Risk Factors (continued) • Increased risk with: • Personal history of inflammatory bowel disease, adenomatous polyps or colon ca • Family history of adenomatous polyps, colon cancer, other conditions *Individuals with these risk factors may require earlier and more intensive screening

5. CENTERS FOR DISEASE CONTROL AND PREVENTION Colorectal Cancer Sporadic (average risk) (65%–85%) Family history(10%–30%) Rare syndromes (<0.1%) Hereditary nonpolyposis colorectal cancer (HNPCC) (5%) Familial adenomatous polyposis (FAP) (1%)

6. Risk Factor - Polyps Different types: • Hyperplastic • minimal cancer potential • Adenomatous • approximately 90% of colon and rectal cancers arise from adenomas

7. Normal to Adenoma to Carcinoma Human colon carcinogenesis progresses by the dysplasia/adenoma to carcinoma pathway

8. Benefits of Screening • Cancer Prevention • Removal of pre-cancerous polyps prevent cancer (unique aspect of colon cancer screening) • Improved survival • Early detection markedly improves chances of long term survival

9. Benefits of Screening *1996 - 2003

10. Colorectal Screening Rates • Just 40% of colorectal cancers are detected at the earliest stage. • A little more than half* of Americans over age 50 report having had a recent colorectal cancer screening test • Slow but steady improvement in these numbers over the past decade (but all are not benefiting to the same degree) *varies based on data source

11. Colorectal Screening Rates Source: MMWR March 2008

12. Trends in Recent* Endoscopy Prevalence (%), by Educational Attainment and Health Insurance Status, Adults 50 Years and Older, US, 1997-2004 *A flexible sigmoidoscopy or colonoscopy within the past five years. Note: Data from participating states and the District of Columbia were aggregated to represent the United States. Source: Behavioral Risk Factor Surveillance System CD-ROM (1996-1997, 1999) and Public Use Data Tape (2001, 2002, 2004), National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention and Prevention, 1999, 2000, 2002, 2003, 2005.

13. Colorectal Screening Rates Low:Reasons (according to Patients) • Low awareness of CRC as a personal health threat • Lack of knowledge of screening benefits • Fear, embarrassment, discomfort • Time • Cost • Access • “My doctor never talked to me about it!”

14. Colorectal Cancer Screening 2008

15. ACS 2003 CRC Prevention and Early Detection Recommendations • Fecal Occult Blood Testing (FOBT) • Guaiac • Immunochemical • Flexible Sigmoidoscopy (FSIG) • FSIG + FOBT • Colonoscopy • Double Contrast Barium Enema (DCBE)

16. The 2008 CRC Guidelines Update was a Joint Effort of 5 Organizations • American Cancer Society • U. S. Multi-Society Task Force on Colorectal Cancer • American Gastroenterological Association • American College of Gastroenterology • American Society of Gastrointestinal Endoscopists • American College of Radiology

17. 2008 CRC Screening Guidelines: Process • Expert panel reviewed and deliberated on available evidence during two face-to-face meetings and a series of conference calls • Literature published between January 2002 and January 2008, as well as unpublished abstracts and manuscripts, were reviewed by panel

18. 2008 CRC Guidelines Update: Evidence Criteria and Limitations • Current evidence has a number of limitations: • Prospective studies are uncommon • Sample sizes tend to be small • Study participants often include higher risk, symptomatic patients and/or screening populations (magnitude of bias uncertain) • Priority placed on prospective studies of asymptomatic adults, with all subjects undergoing colonoscopy • Because adherence to regular screening is low, we are considering setting a test sensitivity threshold for test acceptance

19. CRC Screening Guidelines: What’s New? CRC screening tests are grouped into two categories: • Tests that detect cancer and precancerous polyps* • Tests that primarily detect cancer * It is the strong opinion of the consensus guidelines group that colon cancer prevention should be the primary goal of CRC screening. • Exams that are designed to detect both early cancer and precancerous polyps should be encouraged if resources are available and patients are willing to undergo an invasive test • If the full range of screening tests are not available, physicians should make every effort to offer at least one test from each category

20. CRC Screening Guidelines: What Else is New? • Two new tests recommended: • stool DNA (sDNA) and • computerized tomographic colonography (CTC) – sometimes referred to as virtual colonoscopy • The guidelines establish a sensitivity threshold for recommended tests • The guidelines delineate important quality-related factors for each form of testing The full article can be accessed at: http://caonline.amcancersoc.org/cgi/content/full/CA.2007.0018v1

21. 2008 CRC Screening GuidelinesBeginning at age 50, both men and women at average risk for developing colorectal cancer should use one of the screening tests below: * Colonoscopy should be done if test results are positive. ** For gFOBT or FIT used as a screening test, the take-home multiple sample method should be used. gFOBT or FIT done during a digital rectal exam in the doctor's office is not adequate for screening.

22. 2008 CRC Guidelines Continue to Emphasize Options Because: • Evidence does not yet support any single test as “best” • Uptake of screening remains disappointingly low • Individuals differ in their preferences for one test or another • Primary care physicians differ in their ability to offer, explain, or refer patients to all options equally • Access is uneven geographically, and in terms of test charges and insurance coverage • Uncertainty exists about performance of different screening methods with regard to benefits, harms, and costs (especially on programmatic basis)

23. Sensitivity of Take Home vs. In-Office FOBT FOBT Sensitivity: Take Home vs. In-Office Collins et al, Annals of Int Med Jan 2005

24. In-Office FOBT should be abandoned Conclusion • In-office FOBT is essentially worthless as a screening tool for CRC and must bestrongly discouraged However; • In a recent national survey, nearly 30%of physicians reported using single-sample, in-office FOBT as their primary method of screening for colorectal cancer. Nadel et al, Annals of Int Med Jan 2005

25. 2008 CRC Screening GuidelinesNew Tests

26. Stool DNA

27. Stool DNA Test (sDNA) • Rationale • Fecal occult blood tests detect blood in the stool – which is intermittent and non-specific • Colon cells are shed continuously • Polyps and cancer cells contain abnormal DNA • Stool DNA tests look for abnormal DNA from cells that are passed in the stool* *All positive tests should be followed with colonoscopy

28. Normal Epithelium LateAdenoma Early Cancer Late Cancer Adenoma Genetic Model of Colorectal Cancer Bat-26 (Sporadic) p53 Bat-26 (HNPCC) K-ras APC Mutation Dwell Time: Many decades 2-5 years 2-5 years Optimum phase for early detection Courtesy of Barry M. Berger. MD, FCAP EXACT Sciences

29. sDNA - Sample Collection

30. sDNA - Sample Collection Collection bucket inserted into bracket and installed under toilet seat Patient supplies whole stool sample; no diet or medication restrictions Patient seals sample in outer container and freezer pack Patient seals container and ships back to designated lab (all packing materials and labels supplied)

31. Performance Characteristics of Stool DNA in the Detection of CRC • Three versions of the previously marketed sDNA test have been evaluated • Version 1 (K-ras, APC, p53,BAT-26, DIA) was evaluated in the Imperiale trial • Version 1.1 (K-ras, APC, P53), PreGen-Plus is the currently marketed test • Version 2 (Vimentin only, or Vimentin + DIA) is currently under evaluation and is expected to enter the market in Fall 2008 • Earlier and more recent tests were evaluated in smaller, mixed populations

32. Performance Characteristics of Stool DNA in the Detection of CRC • Testing evaluates stool for the presence of altered DNA in the adenoma-carcinoma sequence • No dietary restrictions • No stool sampling (utilizes the entire stool) • Several studies suggesting strong patient acceptance • Testing interval uncertain • Uncertainty about the meaning of false positives

33. Stool DNA Limitations • Misses some cancers • Sensitivity for adenomas with current commercial version of test is low • Technology (and test versions) are in transition • Costs much more than other forms of stool testing (approximately $300 - $400 per test) • Not covered by most insurers

34. Stool DNA Limitations (cont.) • Appropriate re-screening interval is not known • Not clear how to manage positive stool DNA test if colonoscopy is negative • FDA issues • Test availability

35. CTC Image Optical Colonoscopy CT Colonography (CTC) Courtesy of Beth McFarland, MD

36. CT Colonography Rationale • Allows detailed evaluation of the entire colon • A number of studies have demonstrated a high level of sensitivity for cancer and large polyps • Minimally invasive (rectal tube for air insufflation) • No sedation required

37. 2-D view 3-D view Polyp CT Colonography Polyp Courtesy of Beth McFarland, MD

38. CTC Virtual “Fly Through” Courtesy of Beth McFarland, MD

39. CTC vs. Optical Colonoscopy: Meta-Analyses Halligan 2005, Mulhall 2005

40. CTC vs. Optical Colonoscopy: Sensitivities for All Polyps Pickhardt et al, NEJM 2003

41. The ACRIN CT Study • The ACRIN study is a multi-center study with each site using state-of-the-art technology • 15 participating sites • Patients underwent both CTC and colonoscopy • 2,531 asymptomatic patients studied • Findings published Sept 2008 in New England Journal

42. ACRIN Results Johnson et al, NEJM 2008

43. CTC - Extra-Colonic Findings Most have limited clinical impact, but some are important: • Asymptomatic cancers outside of colon and rectum • Aortic aneurysms • Renal and gall bladder calculi

44. CT Colonography Limitations • Requires full bowel prep (which most patients find to be the most distressing element of colonoscopy) • Colonoscopy is required if abnormalities detected, sometimes necessitating a second bowel prep • Steep learning curve for radiologists • Limited availability to high quality exams in many parts of the country • Most insurers do not currently cover CTC as a screening modality

45. CT Colonography Limitations • Extra-colonic findings can lead to additional testing (may have both positive and negative connotations) • Questions regarding: • Significance of radiation exposure • Management of small polyps

46. 2008 USPSTF Guidelines U.S. Preventive Services Task Force, Ann Intern Med 2008

47. Flat Lesions Background • Described in Japanese patients since 1980’s. Thought to be uncommon in the U.S. • Study published in March 2008 detected flat lesions at much higher rate than any previous U.S. reports • Colonoscopies in over 1800 veterans found: • Polyps in 37% • “Flat lesions” in 9.35% • 0.8% of flat lesions cancerous or pre-cancerous Soetikno, JAMA 2008

48. New Sources of Risk Colorectal cancer prevention largely focuses on finding polyps, but flatter, less visible lesions that are not polyps are also cancer risks. Polyp Elevated lesion Flat lesion Depressed lesion The New York Times: Illustrations by JAMA Flat Lesions Caveats • Most lesions not truly flat Soetikno, JAMA 2008

49. Flat Lesions Caveats (cont.) • Only about 1/3 of patients in Soetikno study were average risk screening population • 1/3 were high risk based on personal or family history • 1/3 were symptomatic • Flat lesions findings were different among average risk patients • Flat lesions found in only 6% • Cancer or pre-cancerous findings in only 0.3% Soetikno, JAMA 2008

50. Flat Lesions Implications for screening • JAMA study led some to question the ability of CTC to detect flat lesions (although CTC was not utilized in the study) • Results from early CTC/flat lesion studies were extremely variable (sensitivity 13% - 65%) • Recent study of experienced radiologist using advanced technology and protocols found 80% sensitivity for flat lesions