CRC Screening GuidelinesA challenge to the survival of the FIT test

James E. Allison, MD, FACPJames E. Allison, MD, FACPClinical Professor of Medicine EmeritusClinical Professor of Medicine EmeritusUniversity of California San FranciscoUniversity of California San FranciscoAdjunct Investigator Adjunct Investigator Kaiser Division of ResearchKaiser Division of Research

Lecture Outline

• The U.S. screening test preference evolution 1995 - 2009

• The influence of “thought leaders” and the media on preference

• An elephant emerges in the screening test room

• The 2008 ACS/Multisociety Taskforce Guidelines

– A challenge to fecal tests

• Evidence accumulates 2000-2009 and answers the challenge

• Innovative technology and strategies emerge leading to a FIT future

• Summary and Conclusions

The Opinion Leaders Speak 2000

“Colonoscopy every 10 years is the preferred screening strategy for average-risk persons age 50 and older if they have no risk factors for colorectal cancer other than age.” (2000)

Recommendation by a panel of American College of Gastroenterology(ACG) members with expertise in colorectal cancer screening and approved by the Board of Trustees and the Publication Committee of the ACG 2000

The Opinion Leaders Speak

“Stool screening has historically relied on detection of occult blood, which has been proven to be an inherently insensitive and nonspecific marker for screen relevant neoplasia.”

Osborn NK and Ahlquist DH, Gastroenterology 2005;128:192-206.

The Media Speaks

It's considered the most effective test for detecting colon cancer, and as Katie Couric says in her special report, "It really didn't hurt." Katie’s first colonoscopy.

Cram P, Fendrick MA, Inadomi J, et al. Arch Intern Med 2003;1601-1605.

The Katie Couric Effect

Flex Sigmo

FOBT

CT/MRI Colonography

Optical Colonoscopy

Cecal StampedeThe Headlong Rush for Screening Colonoscopy

Lawson MJ, Tobi M Dig Dis Sci 2008;53(4):871-4

CRC Screening Test Trends2000 - 2008

31

16

34

16

39

17

41

18

45

19

0

5

10

15

20

25

30

35

40

45

50

Total No health insurance

Pre

vale

nce

(%

)

1997 1999 2001 2002 2004

Trends in Endoscopy: US 1997-2004Increasing for Everyone but the Uninsured

*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.

Menu of recommended screening tests:

ACS/MSTF and ACR Guidelines

Levin B, Lieberman D, McFarland B, Smith RA et al. CA Cancer J Clin 2008

Average Risk Individuals

• Stool Tests

– Fecal occult blood testing (sensitive GT or FIT)– Stool DNA test (sDNA)

• Structural Exams

– Double-contrast barium enema – Flexible sigmoidoscopy– CT Colonography (CTC)– Colonoscopy

Throwing down the gauntlet

ACS/USMSTF and ACR GuidelinesPrecautions Re Menu of Options

• If fecal tests are used the “opportunity for prevention is both limited and incidental and not the primary goal of CRC screening with these tests.”

• “It is the strong opinion of this expert panel that colon cancer prevention should be the primary goal of CRC screening and that providers and patients should understand that noninvasive tests are less likely to prevent cancer compared with the invasive tests.”

The Rap Sheet on FOBT

• If used the opportunity for CRC prevention is limited and incidental.

• Occult blood has been proven to be an inherently insensitive and nonspecific marker for screen relevant neoplasia.

• Repeated testing required and unlikely to be done.

• It is not the most effective screening strategy.

Comparison Test Card FOBT/FIT

FIT Performance CharacteristicsEvidence Answers the Challenge

90 (89-90)91 (89-91)

97 (97-98)91 (89-91)

64 (36-86)41 (33-50)

82 (48-97)30 (21-40)

Hemoccult SensaCarcinomaAdvanced Adenoma

FlexSure OBTCarcinomaAdvanced Adenoma

SpecificityPercent (95% CI)

SensitivityPercent (95% CI)

Test

Allison JE, Sakoda LC, Levin TR, et al. J Natl Cancer Inst 2007;99: 1-9.

Park, D, Ryu S, Kim Y et al Am J Gastro 2010 in press

FIT Performance CharacteristicsEvidence Answers the Challenge

34Advanced Adenoma

92Cancer

Specificity(95% CI) %

Sensitivity(95% CI) %

Eiken/Polymedco FIT Fecal Hgb threshold

< 100 ng/ml

91

91

Levi Z, Rozen P, Hazazi R, et al. Ann Intern Med 2007;146:244-255.

Quantitative FITFecal Hgb level of Lesions Found at Colonoscopy

Normal

Characteristics

Colon site

Patients (n)(%)

Lesion size (SD){95%CI}, mm

Mean FIT result (SD)(95%CI), ng/ml

Advanced Adenoma

739 (73.9)

74 (7.4)

499 (774) {227-772}

485 (744) {315-654}

31 (12.7)

32 (17.2)

35 (143) {25-45}

12.6 (6.4) {11.2-14.1}

Cancer Stages

Colon Site

Proximal

Distal

12.4 (6.8) {10.1-14.7}

12.9 (6.2) {11.0-14.7}

30.7 (9.3) {26.0-35.4}

50.0 (7.10) {40.2-59.8}

501 (737) {229-724}

1399 (1452) {614-3411}

15 (88.2)

1637 (720) {1104-2171}

1045 (777) {652-1439}

701 (672) {285-1118}

Dukes A & B

Dukes C or D

Proximal

Distal

2 (11.8)

10 (58.8)

7 (41.2)

33.8 (10.30) {27.4-40.2}

31.7 (12.5) {22.4-41.0}

Stool DNA Test Versus FITEvidence Answers the Challenge

Test SensitivityPolyp1cm

(%)

SpecificityPolyp1cm

(%)

SensitivityCRCA

(%)

SpecificityCRCA

(%)

Stool DNA Test

Pre Gen VI (NEJM) 52 15 94

Pre Gen VI (Mayo) 25 20

FIT

Magstream 66 20 95 95

Hemoccult ICT 82 30 97 97

Patient Adherence to FOBTEvidence Answers the Challenge

Steele RJ, McClements P, Libby G, Black R1: Gut. 2008 Nov 26. Epub ahead of print

FLU-FOBT Program in Primary CareEvidence Answers the Challenge

Potter M, Phengrasamy La , et al Ann Fam Med 2009;7:17-23.

“But al thing which that shineth as the goldNis nat gold, as that I have herd it told”

- Geoffrey Chaucer Canterbury Tales (1387)

Lieberman D, Ann Intern Med 2004; 141:401-403.

“All that glitters is not gold;Often have you heard that told”

- William Shakespeare Merchant of Venice (1596 –1597)

Colonoscopy: As Good as Gold?

Conclusion: Even under ideal conditions, colonoscopy is not perfect.

Evidence Accumulates 2000 - 2009

• Data suggests that the protection against cancer afforded by having a negative colonoscopy is quite small in the proximal colon (1-33%) but quite large in the distal colon (80%).

• Complete colonoscopy strongly associated with fewer deaths from left-sided CRC but not from right-sided CRC.

• Distal CRC in the U.S. have been steadily decreasing since 1985 while rates for proximal colon cancers have remained largely unchanged.

Lakoff J, Paszat LF, Saskin R, et al Clinical Gastro and Hepatology 2008;6:1117-1121Baxter NN, Goldwasser MA, Paszat LF, et al Ann Intern Med. 2009 Jan 6;150(1):1-8.

Brenner H, Hoffmeister M, Arndt V, et al. J Natl Cancer Inst 2010;102:89–95Baxter NN, Rabeneck L J Natl Cancer Inst. 2010 Jan 20; 102(2):70-130.

Evidence Answers the Challenge• The USPSTF found that five multiple-strategy analyses

did not uniformly find that colonoscopic screening was the most effective or cost effective strategy and similar conclusions were reached by members of the Institute of Medicine and the National Research Council.

• In 2008, a decision analysis using 2 colorectal cancer microsimulation models from the Cancer Intervention and Surveillance Modeling Network supported colorectal cancer screening with annual screening using a sensitive FOBT (sensitive guaiac or fecal immunochemical test) from age 50 to 75 years.

Pignone M, Saha S, Hoerger T, et. Ann Intern Med 2002; 137:96–104National Research Council on the Economic Models of Colorectal Cancer Screening in

Average-Risk Adults. (http://www.nap.edu/catalog/11228.html)Zauber AG, Lansdorp-Vogelaar I. Knudsen AB, et al Ann Intern Med. 2008;149:659-669

Pohl H, Robertson DJ Gastroenterology 2009;136:1149-1151

Colonoscopy: The Facts

• Findings of screening colonoscopy in asymptomatic average risk populations reveal the fraction of persons with no colonic neoplasm is consistent ranging from 75% to 83%.

• The number of persons required to undergo screening colonoscopy on average is 23 to detect an advanced adenoma and 143 for cancer.

• The majority of screening colonoscopies will show no adenomas or cancers and highlight the need to identify a way to estimate absolute risk for individual persons so that screening colonoscopy may be more efficiently targeted to those with advanced neoplasia.

Kahi CJ, Rex DK, Imperiale TF Gastroenterology 2008:135:380-399

Screening Colonoscopy: The Case for Caution

• The risk (2.8 in 1000) of serious complications (perforations, hemorrhage, diverticulitis, CV events, severe abdominal pain and death) detracts from any benefit colonoscopy may have over other less invasive screening options

• Evidence suggests the manpower necessary to provide a skilled colonoscopic examination for all eligible U.S. citizens is inadequate.

• Since Medicare’s decision to reimburse for screening colonoscopy, some gastroenterologists are spending up to 50% of their practice time simply performing colonoscopy

Seef LC, Manninen DL, et al. Gastroenterology 2004; 127:1661-1669.Levin TR, Editorial Gastroenterology 2004; 127:1841-1849.Lieberman, DA, et al. N Engl J Med 2000; 343:162-8.

The Opinion Leaders Speak 2009

• The American College of Gastroenterology Screening Guidelines– “Colonoscopy every 10 years, beginning

at age 50, remains the preferred CRC screening strategy.”

• “It is impractical for a PCP to discuss 6 different options for CRC screening with each patient. Recommending one preferred strategy simplifies the discussion. Colonoscopy is the preferred strategy because it is the best test.”

Rex D, Johnson DA, Anderson JC et al Am J Gastroenterol. 2009;104:739-750 Rex D Medscape Medical News March 10,2009

Menu of recommended screening tests:

Screening for colorectal cancer: US Preventive Services Task Force recommendation statement. Ann Intern Med 2008;149:627-37

Average Risk Individuals Age 50 - 75

• Stool Tests

– Fecal occult blood testing Fecal Immunochemical Test (FIT) or

Sensitive Guaiac Test (Hemoccult Sensa)• Structural Exams

– Flexible sigmoidoscopy + sensitive GT or FIT– Colonoscopy

“Colonoscopy is the most common screening technique for colon cancer, but a better option might be the fecal

immunochemical tests (FIT), which could be easy, non-invasive, effective,

low-risk and inexpensive.”

Douglas K. Rex, MD, FACP, FACG, professor of medicine University of Indiana

DDW Plenary Session Chicago, IL May, 2009 as quoted in DDW News

Dr. Carrie Klabunde, Health Services and Economics Branch NCI

Division of Cancer Control and Population Sciences

“USMSTF/ACS CRC guidelines state that tests providing a full structural examination of the colon are preferred over other tests. Not only do we lack randomized, controlled trial data to warrant such

a preference, but there is evidence that other screening options that use colonoscopy as a

diagnostic follow-up test can play a role in systems that achieve high screening rates.

Summary

• FIT/FOBT is a CRC screening test with proven effectiveness for both early detection and prevention.

• FOBT/FIT is a cheap, effective and currently available way to estimate absolute risk for individual persons so that colonoscopy may be more efficiently targeted to those with advanced neoplasia.

• Effective repeated screenings can be achieved in large average risk populations.

• As of 2010, fecal blood remains the best stool marker for advanced neoplasms and cancer.

Conclusions

• At a time when all countries must confront the realities of limited resources, decisions on how to population screen for colon cancer should take into consideration upfront costs, patient preferences, and the potential risks of screening tests for otherwise healthy people.

• As screening has been shown to save lives, any test with proven benefit in the target population should be acceptable at this time. The screening test(s) selected should be at the discretion of the physician and of his/her patient.

• In 2010 a menu of screening test options remains evidence based and relevant.

Can we, with good conscience, recommend screening tests other than colonoscopy to our average risk patients ?

20

8

21

9

12

22

9

19

9

24

0

5

10

15

20

25

30

Total No health insurance

Pre

va

len

ce

(%

)

1997 1999 2001 2002 2004

Trends in Fecal Occult Blood Testing: US, 1997-2004 Declining since 2001 in all populations

*A fecal occult blood test within the past year. 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.

CRC Screening Test Trends2000 - 2008

FIT Performance CharacteristicsEvidence Answers the Challenge

Test

Hemoccult IICarcinomaAdvanced Adenoma

Hemoccult SensaCarcinomaAdvanced Adenoma

HemeSelectCarcinomaAdvanced Adenoma

SensitivityPercent (95% CI)

37 (22-55)32 (22-40)

79 (64-95)69 (59-78)

69 (51-86)67 (57-76)

SpecificityPercent (95% CI)

98 (97-98)98 (97.7-98.4)

87 (86-87)88 (86.7-88.2)

94 (94-95)95 (95-96)

Allison JE, Tekawa IS, Ransom LJ, Adrain AL. N Engl J Med 1996; 334:155-9.

FIT Performance CharacteristicsEvidence Answers the Challenge

Morikawa T, Katao J, Yamafi Y et al Gastroenterology 2005;125:422-428

33 (24-41)High grade dysplasia

20 (17-23)Advanced Adenoma

95 (94-95)66 (55-76)Carcinoma

Percent (95% CI)Percent (95% CI)Magstream 1000 HP

SpecificitySensitivityTest

Cancer

Variable

>50ng/ml

Specificity(95%CI)

Fecal Hemoglobin Threshold

84 (82-87)

All Advanced neoplasia

>50ng/ml

>75ng/ml

73 (63-82)

88 (85-90)

89 (87-91)

62 (52-72)

90 (88-92)

67 (57-77)

>100ng/ml

>75ng/ml

>100ng/ml

Sensitivity(95%CI)

100 (100-100)

94 (83-100)

88 (73-100)

91 (90-93)

93 (92-95)

Levi Z, Rozen P, Hazazi R, et al. Ann Intern Med 2007;146:244-255.

Fecal Hemoglobin Threshold

Quantitative FITFecal Hgb level of Lesions Found at Colonoscopy

Colonoscopy Clouds

• Interval cancers

• Harms

• Quality

Improved Screening Rates Using FIT

• 2008 Experience KPNC:

– Outreach Mailing: 419,000 tests

– Response Rate: 52%

– Positivity Rate: 5.4%

– PPV: 3.4% for cancer

– CRCs Detected by Outreach: 403