Colorectal Cancer Screening Optimizing Mortality Reductions Agenda - PDF document

Colorectal Cancer Screening Optimizing Mortality Reductions Agenda through Colorectal Cancer Screening • Which test is best? Asian Health Symposium San Francisco. October 2017 • How to increase screening among Asians? • How do you prepare for the future? John M. Inadomi Cyrus E. Rubin Chair and Head Division of Gastroenterology University of Washington CRC Screening CRC Screening U.S. Preventive Services Task Force U.S. Preventive Services Task Force 2016 Updated Guidelines • Available Strategies: • Start and Stop: Ages 50‐75 years – 76‐85 years: do not screen routinely – Fecal occult blood testing annually* – Older than 85 years: do not screen – Sigmoidoscopy • Starting at age 45 years • Every 5 years • Every 10 years with annual FOBT * – Could reduce cancer and increase life‐years – Colonoscopy every 10 years – Could increase harms of screening (colonoscopy) – FIT‐DNA testing every 1 or 3 years • Individuals >75 years of age who have not – CT colonography every 5 years previously undergone screening may still benefit – Depending on comorbid illness, could benefit past age – Methylated Septin9 DNA (unknown interval) 80 years *FIT or high‐sensitivity gFOBT (Hemoccult Sensa) 1

Fecal Occult Blood Test FIT Detection of CRC: Meta‐Analysis • Benefits – High quality evidence supporting mortality reduction with FOBT (multiple RCTs) – FIT easier: single sample, no dietary exclusions – Highly cost‐effective, may be cost‐saving • Deficits – Insensitive for cancer precursor (polyps) – Annual testing needed Lee. Ann Intern Med 2014 Multitarget Stool (FIT‐DNA) Multitarget Stool vs. FIT Colonoscopy Stool DNA sensitivity FIT • KRAS mutations sensitivity Colorectal Cancer 65 92.3 (83.0‐97.5) 73.8 (61.5‐84.0) • NDRG4 and BMP3 methylation Advanced adenoma 757 42.4 (38.9‐46.0) 23.8 (20.8‐27.0) • β‐actin / SSP≥ 1cm • Immunochemical test for human hemoglobin Colonoscopy Stool DNA specificity FIT • FDA approved specificity • CMS reimbursed Non‐advanced 9167 86.6 (85.9‐87.2) 94.9 (94.4‐95.3) adenomas and negative colon Negative colon 4457 89.8 (88.9‐90.7) 96.4 (95.8‐96.9) Imperiale TF et al. N Engl J Med 2014 2

Multitarget Stool (FIT‐DNA) Virtual Colonoscopy (CT Colonography) • Benefits – Sensitivity higher than FIT (92.3% vs. 73.8%) – FDA approved for primary screening of CRC • Deficits – Specificity lower than FIT (86.6% vs. 94.9%) – Insensitive for cancer precursors (polyps) – Unknown screening intervals – Cost ($649 retail) – not cost‐effective Pickhardt. NEJM 2003 Sessile Serrated Polyps Colonoscopy vs. CT Colonography • Sessile serrated polyps – RCT colonoscopy vs. CT colonography • 8,844 participants – Diagnosis of high‐risk SSP • ≥1cm or any grade of dysplasia – OR 5.5 (95% CI 2.6‐11.6) – CT inferior to colonoscopy for detection of SSP IJSpeert Am J Gastro 2016 J Gastroenterol 2013 3

Adherence to CRC Screening Adherence to CRC Screening Participation Rate (%) Participation Rate (%) Colonoscopy 14.8* Colonoscopy 14.8* CTC 25.2* Full prep CTC 28.1 (NS) Reduced prep Sali. J Natl Cancer Inst 2015 Sali. J Natl Cancer Inst 2015 Adherence to CRC Screening Colonoscopy vs. CT Colonography Scenario Results Participation Rate (%) Base Case (CT $558) Colon dominates CT Colonoscopy 14.8* CTC 25.2* Cost of CT $100‐480 Colon cost‐effective Full prep Cost of CT <$100 CT preferred CTC 28.1 (NS) Reduced prep Dominates = more effective and less expensive FIT 50.4* Cost‐effective = <$50,000 per life year saved Preferred = ICER for Colon vs. CT >$50,000 per life year saved Am J Gastroenterol 2007;132:809‐811 Sali. J Natl Cancer Inst 2015 4

Blood Based Test Virtual Colonoscopy Septin 9 • Benefits – Cool name • Gene codes for Guanosine triphosphate (GTP)‐ – May have greater adherence than colonoscopy binding protein • Deficits – Cytoskeleton formation and filamentous structure – Inferior detection of sessile serrated polyps • Oncogene or tumor suppressor gene – Extracolonic findings in 16% – Methylated Septin 9 is a biomarker for CRC – Radiation exposure • FDA approved for CRC screening – Not cost‐effective compared with FIT or – Individuals who refuse other tests colonoscopy How To Increase Screening Blood Based Test Septin 9 Sensitivity 95% CI Specificity 95% CI Colorectal 74.8% 67.0‐81.6% 87.4% 83.5‐90.6% cancer Advanced 27.4% 18.7‐37.6% adenomas Adenomas 20.7% 15.1‐27.3% Jin. J Gastroenterol Hepatol 2015 5

Adherence to CRC Screening Adherence to CRC Screening Methods • Overall adherence 60% (stable since 2010) • Design – 55% colonoscopy within 10 years – Prospective quasi‐experimental study – 5% FOBT within the previous year – Interventions: – <1% sigmoidoscopy within 5 years • Recommend FOBT • Large variations between racial/ethnic • Recommend Colonoscopy groups • Choice of FOBT or colonoscopy – Disparities vs. Differences • Setting – Access vs. Utilization – Urban, diverse underserved population Methods Methods Study Design – Clinic Randomization • Subjects – Average risk for development of CRC FOBT Choice Colonoscopy • No family history of CRC • No personal history of CRC or adenomas, IBD General Medicine Clinic – Not up‐to‐date with CRC screening • Outcomes – Completion of screening strategy within one year Colonoscopy Choice FOBT • FOBT plus colonoscopy if positive • Colonoscopy Family Health Center and Positive Health Program 6

Methods Methods • Goal of study: Identify patient factors associated with adherence • Research personnel enroll subjects • Requirement: reduce systems/access barriers – One encounter – Study survey • Open access colonoscopy (factors associated with screening adherence) – Language • PCP counsels patient about CRC screening • Spanish, Cantonese, Mandarin, English and specific test(s) available to the clinic – Capacity • < 2 week wait for colonoscopy – Follow‐up to determine adherence – Cost • Healthy San Francisco – Support • Rides to / from hospital if necessary Results Recommendation for Colonoscopy: Subjects: 997 enrolled Lower Adherence 100% 3% 90% 15% 80% 68.8% 67.2% 70% 60% 34% 18% 50% 38.2% 40% 30% 20% 30% 10% 0% FOBT Colonoscopy Choice (regardless of White (NH) Black (NH) Asian Hispanic Other intent) Arch Intern Med 2012 Arch Intern Med 2012 7

Recommendation for Colonoscopy: Overall Adherence Varies by Race Lower Adherence 100% 100% 90% RR: 1.5‐3.2 90% * * 80% RR: 1.7‐3.6 80% 68.8% 67.2% 72.6% 70% 69.5% NS 70% 58.1% 62.0% 60% black 60% 50% 51.4% white 50% 40% Asian 40% 30% Latino 30% 20% 20% 10% 10% 0% 0% FOBT Colonoscopy Choice (regardless of Adherent (colonoscopy or FOBT) intent) *p = 0.001 Arch Intern Med 2012 Arch Intern Med 2012 Factors Associated with Adherence Variable no./total no. (%) Univariate Analysis OR 95% C.I. Study Group Colonoscopy 193/332 (58) 1 – FOBT 231/344 (67) 1.50 1.08 – 2.06 Choice 221/321 (69) 1.61 1.16 – 2.24 Race/Ethnicity Black 90/177 (51) 1 – White 92/149 (62) 1.55 0.99 – 2.42 Asian 214/298 (72) 2.46 1.66 – 3.63 Latino 234/337 (69) 2.19 1.50 – 3.20 Language English 315/556 (57) 1 – Spanish 190/260 (73) 2.12 1.52 – 2.95 Cantonese or Mandarin 136/175 (78) 2.72 1.82 – 4.08 8

Factors Associated with Adherence Race/Ethnicity and Language Variable no./total no. (%) Multivariable Study Group OR 95% C.I. • Latino and Asian subjects adhered more often Colonoscopy 193/332 (58) 1 – than white and black subjects FOBT 231/344 (67) 1.44 1.03 – 2.02 Choice 221/321 (69) 1.66 1.18 – 2.35 • Effect disappeared when language introduced – Increased adherence driven by those who preferred Race/Ethnicity to speak Cantonese, Mandarin or Spanish Black 90/177 (51) 1 – White 92/149 (62) 1.43 0.89 – 2.30 – Within Asian and Latino participants Asian 214/298 (72) 1.43 0.86 – 2.38 • Non‐English speakers adhered at higher rate Latino 234/337 (69) 1.31 0.75 – 2.31 • What? Language English 315/556 (57) 1 – Spanish 190/260 (73) 1.67 0.94 – 2.98 Cantonese or Mandarin 136/175 (78) 2.13 1.23 – 3.70 Adherence to CRC Screening: Race/Ethnicity and Language Study Summary • Language may be a surrogate for: • The best test is the one that gets done – Immigration status: The “healthy immigrant” – Providing choice of CRC screening test – Health belief system increases adherence to screening • Are disparities in screening due to differences • Variation by race/ethnicity in health beliefs? – Whites adhere more often to colonoscopy – Impact – Non‐whites adhere more often to FOBT – Severity • Race/ethnicity and language – Self‐efficacy – Family, friends, social network – Surrogates for immigrant status, health beliefs, trust in physicians 9