Colonoscopy Screening: Issues and Controversies Symposium on GI - PowerPoint PPT Presentation

Colonoscopy Screening: Issues and Controversies Symposium on GI Cancers St. Louis 9-12-09 David Lieberman MD Chief, Division of Gastroenterology and Hepatology Oregon Health Sciences University Portland VAMC

Risk Factors for CRC HNPCC Family 3% History FAP-1% 15-20% IBD-1% Colitis World wide: - 1M new cases/yr - > 500,000 deaths/yr - 2 nd leading cause of cancer death in USA Sporadic/ Average Risk 75%

Colorectal Cancer Screening Genetic Environmental Lifestyle Normal Colon Advanced Adenoma 10-20% 5-6% Lifetime Risk Lifetime Risk

Raising the bar Early Colon Colon Cancer Cancer Prevention Detection MD

CRC Screening Guidelines 2008 Tests which “primarily” detect early cancer: Screening Test ACS-MSTF- USPSTF* ACG ACR* gFOBT YES; requires YES; Yes; only “sensitive” test >50% sensitivity sensitive test for CRC only FIT YES; requires YES Preferred as >50% sensitivity cancer detection for CRC test Stool DNA YES; requires NO YES; every 3 >50% sensitivity years for CRC • ACS-MSFT-ACR = American Cancer Society; Multi-Society Task Force CRC; American College of Radiology • USPSTF = United States Preventive Services Task Force • ACG = Am College of Gastroenterology

CRC Screening Guidelines 2008 Tests which can detect early cancer and cancer precursor lesions Screening Test ACS-MSTF- USPSTF* ACG ACR* YES – every 5 NO – CT YES; every 5 Colonography years insufficient years evidence YES – every 5 YES – every 5 Flex Sig YES; every 5-10 years years with years sensitive FOBT YES – every 10 YES – every 10 YES - Colonoscopy years years preferred ACS-MSFT-SCR = American Cancer Society; Multi-Society Task Force CRC; American College of Radiology USPSTF = United States Preventive Services Task Force; ACG = Am College of Gastroenterology

Stool-Based Tests: Sensitivity Test Cancer Advanced Adenoma Standard gFOBT 33-50% 11% Sensitive gFOBT 50-75% 20-25% FIT 60-85% 20-50% Stool DNA (old version) 51% 18% Stool DNA (new version) 80+% 40% Key new references: sDNA: Ahlquist et al; Ann Intern Med 2008; 149;441-50 FIT: Hundt et al; Ann Intern Med 2009; 150: 162-9

Fecal Blood Tests • Complicated in real life • Requires programmatic adherence with (+) and (-) tests • Programmatic performance: UNKNOWN • Unlikely to result in much cancer prevention

Structural Colon Exams: Sensitivity of one-time test Test Cancer Advanced Adenoma CT Colonography Uncertain; likely >90% 90% if >10mm (CTC) 78% for all polyps >6mm Sigmoidoscopy >95% distal colon 70% 60-70% Proximal colon Colonoscopy >95% 88-98% Key new references: CTC: Johnson et al; NEJM 2008; 359; 1207-17

USA – Colorectal Screening 60 50 40 30 20 FOBT 10 Sigmoid 0 2000 2001 2002 2003 2004 2005 National Health Interview Survey

Colonoscopy Screening Studies (n > 1000) • Studies: 2000-2004 – Lieberman et al; VA Cooperative Study ;NEJM: 2000; 343: 162-8 (n = 3121) – Indiana Study; NEJM 2000; 343: 169-74 (n = 1994) – CT Colonography studies (n = 2447) (Pickhardt, Rockey, Cotton) – Fecal DNA Study; NEJM 2004; 351: 2704-14 (n = 4404) Results: – Spain, Am J Gastroenterol 2003; 98: 2648-54 (n = 2210) 5-10% with Advanced Neoplasia • Studies: 2005-2006 – Women: (Schoenfeld) NEJM 2005; 352: 2061-8 (n = 1463) 0.5-1.0% with Cancer – Taiwan; Gastrointest Endosc 2005; 61: 547-53 (n = 1708) – Japan, Gastroenterology 2005; 129: 422-8 (n = 21,805 with iFOBT) – Seattle, JAMA 2006; 295: 2357-65 (N = 1244) – Poland, NEJM 2006; 355: 1863-72 (n = 50,148) • Experience: 2007-2008 – Germany (n = > 2 million )

Colonoscopy • Evidence: – Cohort Studies: feasibility and safety – Case-Control Studies Study Mortality Incidence Muller/Sonnenberg; ---- OR = 0.47 1995 Case- Control Singh; 2006 ---- After negative CSP: 5 yrs: SIR 0.55 10 yrs: SIR 0.28 Baxter; 2009 Left colon: OR 0.33 Right colon: OR 0.99 Kahi; 2009 SMR 0.35 SIR 0.33 Cohort

Colonoscopy • Quality in practice: unknown • Program performance: unknown • Potential for cancer prevention: High

Interval Cancer after Colonoscopy 0.3-0.9% • New, fast growing lesions – Sawhney et al; Gastroenterology 2006; 131: 1700-5 • Incomplete removal – Pabby et al; Gastrointest Endosc 2005; 61: 385-91 • Missed lesions – 2-12% of polyps >1cm are missed

Colon Screening Issues • Flat polyps Soetikno et al; JAMA 2008; Lieberman; JAMA 2008 Submucosal cancer Indigocarmine

Baxter et al; Ann Intern Med 2009; 150: 1-8 Colon Screening Issues • Case control Study: 10,292 cases – CRC (any) : OR 0.63 (0.57-0.69) – Left colon: OR 0.33 (0.278-0.39) – Proximal colon: OR 0.99 (0.86-1.14) • Why? – Biology – Quality

Colonoscopy: Serious Adverse Events /1000 Setting Year n Bleed Perforation Other* VACSP 380-13 centers 2002 3196 1.9 0 1.0 Screening only; 30d f/u * hospitalization Teaching Hosp 2003 23,508 2 1 ----- Australia 0.3-3.2 0-1.3 Amb Surg Centers 2003 116,000 ---- 0.3 ----- USA per 1000 per 1000 Community-UK 2004 9223 ----- 1.3 ----- Medicare Sample 2003 39,286 ---- 2 ----- Age >65 yrs Community 2006 12,407 2 0.16 ----- USA-Wisconsin HMO-USA 2006 16,318 3.2 0.9 ----- Community-Poland 2006 50148 0.26 0.1 0.7 Screening only * hospitalization Outpatient-Canada 2008 97,091 1.6 0.85 ----- Claims data California Medicaid 2009 277,434 ----- 0.8 ----- Claims data Manitoba; Claims data 2009 21,191 0.86 1.18 30d f/u Medicare; Claims data 2009 53,220* 6.4 0.6 CV events with 30d f/u polypectomy

Rex, 2006 Lieberman et al; Gastrointest Endosc 2007; 65:757-66 Lieberman et al; Gastrointest Endosc 2009; 65: 757-66 Colon Cancer Screening: Colonoscopy Effectiveness depends on: • Appropriate utilization • High-quality exam to cecum QUALITY • Low rate of missed lesions • Low rate of incompletely removed lesions • Low rate of adverse events

Colon Cancer: Incidence and Mortality per 100,000 600 Male 500 Female Incidence 400 Does it make sense to Mortality 300 screen women at the 200 same age as men? 100 0 45-50 55-60 65-70 75-80 AGE

Lieberman et al, CGH 2005: 3: 798-805 CRC Screening: what is found? % with Polyp(s) > 9mm

Number needed to endoscope: NNE To identify one patient with polyp > 9mm Asymptomatic Screening 45 40 35 30 <50 50-59 25 50 60-69 20 59 70-79 50 15 >79 59 10 5 0 Male Female

CRC and Race/Ethnicity Incidence Mortality Male Female Male Female White 58.9 43.2 22.1 15.3 Black 71.2 54.5 31.8 22.4 Asian 48.0 35.4 14.4 10.2 Hispanic 47.3 32.8 16.5 10.8 Native 46.0 41.2 20.5 14.2 ALL 59.2 43.8 22.7 15.9 Stage White Black Local 40% 35% Regional 37% 35% Distant 19% 24%

Colorectal Cancer: Incidence Trends by Race Why? Source: American Cancer Society; 2008

Black vs White Differences • Socio-economic – Lack of access to health care – Lower insurance • MD-Patient – Failure of MD to recommend screening – Failure of patient to accept screening • Cultural barriers • Biology

Lieberman et al; JAMA 2008; 300: 1417-22 Black vs White % with polyp (s) > 9mm Men : OR 1.16 1.01,1.34 Women : OR 1.62 (1.39-1.89) 10 18 White White 9 16 Black Black 8 14 7 12 6 10 5 8 4 6 3 2 4 1 2 0 0 <50 50-59 60-69 70-79 >79 ALL < 50 50-59 60-69 70-79 >79 ALL When barriers to screening are not an issue, asymptomatic black men and women are more likely to have polyp(s) >9mm compared to whites

Colon Screening Issues • Are we ready for customized screening based on : – Sex – consider screening women later – Race – consider screening Blacks with colonoscopy, since they represent a high-risk group

USPSTF; Ann Intern Med 2008; 149: 627-37 Lieberman et al, CGH 2005: 3: 798-805 Colon Screening Issues • Stopping screening based on age per 100,000 600 Male – No routine screening for adults 75-85 years 500 Female – Against any screening after age 85 years Polyp(s) >9mm 400 Incidence 300 200 Mortality 100 0 45-50 55-60 65-70 75-80

10-Year Interval: Is it safe? • Missed lesions • Epidemiology • Interval cancers • Case-control studies • Lawsuits

Lessons from USA Experience • Colon Cancer Screening is effective • Prevention should be a primary goal • Obstacles to screening can be overcome: • Effectiveness of screening programs depends on QUALITY and Adherence