دليل فحص سرطان القولون والمستقيم

ia_gov_sa · guide · processed · 90 chunks

458
CLINICAL GUIDELINES
CME
ACG Clinical Guidelines: Colorectal Cancer Screening 2021
Aasma Shaukat, MD, MPH, FACG \( ^{1,2} \) , Charles J. Kahi, MD, MSc, FACG \( ^{3-7} \) , Carol A. Burke, MD, FACG \( ^{4} \) , Linda Rabeneck, MD, MPH, MACG \( ^{5} \) , Bryan G. Sauer, MD, MSc, FACG (GRADE Methodologist) \( ^{6} \) and Douglas K. Rex, MD, MACG \( ^{3} \)
Colorectal cancer (CRC) is the third most common cancer in men and women in the United States. CRC screening efforts are directed toward removal of adenomas and sessile serrated lesions and detection of early-stage CRC. The purpose of this article is to update the 2009 American College of Gastroenterology CRC screening guidelines. The guideline is framed around several key questions. We conducted a comprehensive literature search to include studies through October 2020. The inclusion criteria were studies of any design with men and women age 40 years and older. Detailed recommendations for CRC screening in average-risk individuals and those with a family history of CRC are discussed. We also provide recommendations on the role of aspirin for chemoprevention, quality indicators for colonoscopy, approaches to organized CRC screening and improving adherence to CRC screening. CRC screening must be optimized to allow effective and sustained reduction of CRC incidence and mortality. This can be accomplished by achieving high rates of adherence, quality monitoring and improvement, following evidence-based guidelines, and removing barriers through the spectrum of care from noninvasive screening tests to screening and diagnostic colonoscopy. The development of cost-effective, highly accurate, noninvasive modalities associated with improved overall adherence to the screening process is also a desirable goal.
SUPPLEMENTARY MATERIAL accompanies this paper at http://links.lww.com/AJG/B890 and http://links.lww.com/AJG/B891
Am J Gastroenterol 2021;116:458–479. https://doi.org/10.14309/ajg.0000000000001122
INTRODUCTION
In the United States, colorectal cancer (CRC) ranks second to lung cancer as a cause of cancer mortality and is the third most commonly occurring cancer in both men and women. A study estimated that in 2020 approximately 147,950 new CRC cases would have been diagnosed and 53,200 individuals would have died of the disease (1). Between 2011 and 2015, the average annual incidence rates per 100,000 population were 45.9 and 34.6 for men and women respectively (2). CRC incidence and mortality rates have shown a steady decline of approximately 1.7% and 3.2%, respectively per year. The decline began in the mid 1980s and has accelerated since the early 2000s. It is believed to be driven by changes in risk factors, early detection of cancer through CRC screening, and removal of precancerous polyps with colonoscopy, in addition to advances in surgical and treatment approaches.
Most CRCs develop through the adenoma-carcinoma sequence, presenting opportunities to prevent cancer by removing its precursor lesions, in addition to identifying CRC in its earliest, curable stages (3). Approximately 70% of sporadic CRCs develop from adenomatous polyps and 25%–30% arise from sessile serrated lesions (SSLs) through the SSL-to-carcinoma pathway (4). CRC screening efforts are directed toward removal of adenomas, SSLs and detection of early-stage CRC. Certain screening modalities such as colonoscopy, sigmoidoscopy, CT colonography and to a
lesser extent stool-based testing, will detect advanced adenomatous polyps, whereas colonoscopy is optimal for the detection of SSLs. Endoscopic removal of polyps reduces CRC incidence and CRC mortality (5,6). Given new evidence regarding enhancing screening adherence, newer methods for CRC screening, and evidence to support the efficacy of screening, the purpose of this article is to update the 2009 American College of Gastroenterology (ACG) CRC screening guideline (7).
METHODS
The guideline is framed around several key questions which are outlined below. The key questions were developed by the authors and vetted through the ACG leadership. We placed emphasis on having practical recommendations that would be helpful for practicing providers in the United States. We conducted a focused literature search and used existing guidelines and technical reviews on CRC screening by key organizations. We used a modified Grading of Recommendations, Assessment, Development and Evaluation methodology (8) to evaluate the quality of the evidence and strength of recommendation. We used “we recommend” for strong recommendations and “we suggest” for conditional recommendations. Two Grading of Recommendations, Assessment, Development
\( ^{1} \) Division of Gastroenterology, Minneapolis Veterans Affairs Medical Center, University of Minnesota, Minneapolis, Minnesota, USA; \( ^{2} \) Division of Gastroenterology, Department of Medicine, University of Minnesota, Minneapolis, Minnesota, USA; \( ^{3} \) Division of Gastroenterology, Indiana University School of Medicine, Indianapolis, Indiana, USA; \( ^{4} \) Division of Gastroenterology, Cleveland Clinic, Cleveland, Ohio, USA; \( ^{5} \) Department of Medicine, University of Toronto, Toronto, Ontario, Canada; \( ^{6} \) Department of Medicine, University of Virginia, Charlottesville, Virginia, USA; \( ^{7} \) Department of Medicine, Richard L. Roudebush Veterans Affairs Medical Center, Indianapolis, Indiana, USA. Correspondence: Aasma Shaukat, MD, MPH, FACG. E-mail: shaukat@umn.edu. Received March 17, 2020; accepted December 2, 2020
The American Journal of GASTROENTEROLOGY
VOLUME 116 | MARCH 2021 www.amjgastro.com
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

ACG Clinical Guidelines
459
and Evaluation-trained methodologists assisted in evidence synthesis and grading of the evidence.
Literature search
We conducted a comprehensive literature search with the help of a librarian from the University of Minnesota on the key questions using Ovid (MEDLINE), EMBASE, and the Cochrane databases from 1980 to October 2020. Emphasis was placed on studies from 2008 onward, since publication of the last guideline. The references for review articles were also searched. A detailed search strategy is provided in Supplementary Appendix 1 (see Supplementary Digital Content 1, http://links.lww.com/AJG/B890). The inclusion criteria were observational studies and randomized controlled trials (RCTs) with men and women age 40 years and older. Exclusion criteria were patients/populations with familial cancer syndromes (hereditary non-polyposis colorectal cancer and polyposis syndromes) and special populations such as patients with human immunodeficiency virus or previous transplant. Outcomes included were CRC incidence, CRC mortality, incidence of colorectal advanced neoplasia defined as adenomas or SSL ≥10 mm, ≥3 adenomas/SSL, any villous histology, high-grade dysplasia or submucosal cancer in a colonic polyp or a traditional serrated adenoma, and harms of screening (complications, anesthesia-related complications, deaths, and overdiagnosis through additional testing).
Key questions
KQ1a. In average-risk individuals, what are the effectiveness and harms of CRC screening in reducing the incidence of advanced neoplasia and CRC, and CRC mortality?
KQ1b. How does the effectiveness vary by modality, age, and race?
KQ2. In average-risk individuals, how does the effectiveness of CRC screening vary by screening interval in reducing colorectal advanced neoplasia incidence, CRC incidence, and CRC mortality?
KQ3. In individuals with a family history of CRC or adenomatous polyps, what is the effectiveness of CRC screening in reducing CRC incidence and CRC mortality?
KQ4. In individuals with a family history of CRC or adenomatous polyps, how does the effectiveness of CRC screening vary by screening interval in reducing colorectal advanced neoplasia incidence, CRC incidence, and CRC mortality?
KQ5. In individuals with a family history of CRC or adenomatous polyps, how does the effectiveness of CRC screening vary by screening modality in reducing colorectal advanced neoplasia incidence, CRC incidence, and CRC mortality?
KQ6. What are the quality indicators for different modalities of CRC screening associated with diagnostic performance of the screening test and incidence of postcolonoscopy colorectal cancer?
KQ7. What are the effectiveness and harms of aspirin chemoprevention for the endpoints of reduction in the incidence of CRC or mortality of CRC?
KQ8. What interventions improve adherence to CRC screening and to each modality of screening?
KQ9. What interventions improve adherence to follow-up of a positive CRC screening test, such as fecal immunochemical testing (FIT)?
RESULTS
See Table 1 for summary and Supplementary Appendix 2 (see Supplementary Digital Content 2, http://links.lww.com/AJG/B891) for updates from the 2009 guideline. Results for individual questions are provided below.
KQ1a. In average-risk individuals, what are the effectiveness and harms of CRC screening in reducing incidence of advanced neoplasia and CRC, and CRC mortality?
KQ1b. How does the effectiveness vary by screening modality, age, and race?
Recommendations
1. We recommend CRC screening in average-risk individuals between ages 50 and 75 years to reduce incidence of advanced adenoma, CRC, and mortality from CRC.
Strong recommendation; moderate-quality evidence
2. We suggest CRC screening in average-risk individuals between ages 45 and 49 years to reduce incidence of advanced adenoma, CRC, and mortality from CRC.
Conditional recommendation; very low-quality evidence
3. We suggest that a decision to continue screening beyond age 75 years be individualized.
Conditional recommendation; very low-quality evidence
4. We recommend colonoscopy and FIT as the primary screening modalities for CRC screening.
Strong recommendation; low-quality evidence
5. We suggest consideration of the following screening tests for individuals unable or unwilling to undergo colonoscopy or FIT: flexible sigmoidoscopy, multitarget stool DNA test, CT colonography or colon capsule.
Conditional recommendation; very low-quality evidence
6. We suggest against Septin 9 for CRC screening.
Conditional recommendation, very low-quality of evidence
DISCUSSION
The “ideal” screening test should be noninvasive, have high sensitivity and specificity, be safe, readily available, convenient, and inexpensive. For CRC screening, there are multiple approved tests and strategies, each with its strengths and weaknesses. In some instances the “best” screening test can be considered the one that is acceptable to the patient and gets completed. One approach to CRC screening tests is to divide them as 1-step (direct) tests (i.e., colonoscopy, which is diagnostic and therapeutic) or 2-step tests that require colonoscopy if positive, to complete the screening process. All screening tests other than colonoscopy are 2-step tests. A major limitation of non-colonoscopy-based CRC screening tests (eg, stool-based, flexible sigmoidoscopy, CT colonography [CTC], or colon capsule [CC]) is that a positive test requires a follow-up colonoscopy. This 2-step testing approach represents a continuum of screening, requires strong systems-based support to complete the screening cascade, and is more effectively applied in organized screening (9). In the United States, there are few select health care systems with organized, programmatic screening, and most screening is accomplished with a 1-step opportunistic approach. Because the focus of the guideline is on providers practicing in the United States, the review highlights options for CRC screening currently in use, which mainly include colonoscopy, and in an
© 2021 by The American College of Gastroenterology
The American Journal of GASTROENTEROLOGY
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

460
Shaukat et al.
Table 1. Summary and strength of GRADE recommendations
<table><tr><td colspan="2">Summary</td><td>Recommendation strength</td><td>GRADE quality of evidence</td></tr><tr><td>1</td><td>We recommend colorectal cancer (CRC) screening in average-risk individuals between ages 50 and 75 yr to reduce incidence of advanced adenoma, CRC, and mortality from CRC</td><td>Strong</td><td>Moderate</td></tr><tr><td>2</td><td>We suggest CRC screening in average-risk individuals between ages 45 and 49 yr to reduce incidence of advanced adenoma, CRC, and mortality from CRC</td><td>Conditional</td><td>Very low</td></tr><tr><td>3</td><td>We suggest that a decision to continue screening beyond age 75 yr be individualized</td><td>Conditional</td><td>Very low</td></tr><tr><td>4</td><td>We recommend colonoscopy and fecal immunochemical testing (FIT) as the primary screening modalities for CRC screening</td><td>Strong</td><td>Low</td></tr><tr><td>5</td><td>We suggest consideration of the following screening tests for individuals unable or unwilling to undergo a colonoscopy or FIT: flexible sigmoidoscopy, multitarget stool DNA test, CT colonography, or colon capsule</td><td>Conditional</td><td>Very low</td></tr><tr><td>6</td><td>We suggest against Septin 9 for CRC screening</td><td>Conditional</td><td>Very low</td></tr><tr><td>7</td><td>We recommend that the following intervals should be followed for screening modalities: FIT every 1 yr; colonoscopy every 10 yr</td><td>Strong</td><td>Low</td></tr><tr><td>8</td><td>We suggest that the following intervals should be followed for screening modalities: multitarget stool DNA test every 3 yr; flexible sigmoidoscopy every 5–10 yr; CT colonography every 5 yr; colon capsule every 5 yr</td><td>Conditional</td><td>Very low</td></tr><tr><td>9</td><td>We suggest initiating CRC screening with a colonoscopy at age 40 or 10 yr before the youngest affected relative, whichever is earlier, for individuals with CRC or advanced polyp in 1 first-degree relative (FDR) at age &lt;60 yr, or CRC or advanced polyp in ≥2 FDR at any age. We suggest interval colonoscopy every 5 yr</td><td>Conditional</td><td>Very low</td></tr><tr><td>10</td><td>We suggest consideration of genetic evaluation with higher familial CRC burden (higher number and/or younger age of affected relatives)</td><td>Conditional</td><td>Very low</td></tr><tr><td>11</td><td>We suggest initiating CRC screening at age 40 or 10 yr before the youngest affected relative and then resuming average-risk screening recommendations for individuals with CRC or advanced polyp in 1 FDR at age ≥60 yr.</td><td>Conditional</td><td>Very low</td></tr><tr><td>12</td><td>In individuals with 1 second-degree relative (SDR) with CRC or advanced polyp, we suggest following average-risk CRC screening recommendations</td><td>Conditional</td><td>Low</td></tr><tr><td>13</td><td>We recommend that all endoscopists performing screening colonoscopy should measure their individual cecal intubation rates (CIRs), adenoma detection rates (ADRs), and withdrawal times (WTs)</td><td>Strong</td><td>Moderate-quality evidence for ADR; low-quality evidence for WT, CIR</td></tr><tr><td>14</td><td>We suggest that colonoscopists with ADRs below the recommended minimum thresholds (&lt;25%) should undertake remedial training</td><td>Conditional</td><td>Very low</td></tr><tr><td>15</td><td>We recommend that colonoscopists spend at least 6 min inspecting the mucosa during withdrawal</td><td>Strong</td><td>Low</td></tr><tr><td>16</td><td>We recommend that colonoscopists achieve a CIR of at least 95% in screening subjects</td><td>Strong</td><td>Low</td></tr><tr><td>17</td><td>We suggest low-dose aspirin in individuals between ages 50–69 yr with a cardiovascular disease risk of ≥10% over the next 10 yr, who are not at an increased risk for bleeding and willing to take aspirin for at least 10 yr to reduce the risk of CRC</td><td>Conditional</td><td>Low</td></tr><tr><td>18</td><td>We recommend against the use of aspirin as a substitute for CRC screening</td><td>Strong</td><td>Low</td></tr></table>
The American Journal of GASTROENTEROLOGY
VOLUME 116 | MARCH 2021 www.amjgastro.com
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

ACG Clinical Guidelines
461
Table 1. (continued)
<table><tr><td></td><td>Summary</td><td>Recommendation strength</td><td>GRADE quality of evidence</td></tr><tr><td>19</td><td>We recommend organized screening programs to improve adherence to CRC screening compared with opportunistic screening</td><td>Strong</td><td>Low</td></tr><tr><td>20</td><td>We suggest the following strategies to improve adherence to screening: patient navigation, patient reminders, clinician interventions, provider recommendations and clinical decision support tools</td><td>Conditional</td><td>Very low</td></tr><tr><td>21</td><td>We suggest the following strategies to improve adherence to follow-up of a positive screening test: mail and phone reminders, patient navigation, and provider interventions</td><td>Conditional</td><td>Very low</td></tr></table>
organized setting, fecal immunochemical test (FIT). Other 2-step tests such as flexible sigmoidoscopy, multitarget stool DNA test, CTC, and CC are reserved for individuals unwilling or unable to undergo colonoscopy or FIT, or those with incomplete colonoscopy (CTC or CC). Comparative effectiveness studies are lacking. The options for screening are discussed below and summarized in Table 2.
ONE-STEP SCREENING
Colonoscopy
Colonoscopy is the most commonly performed gastrointestinal procedure in the United States. It allows for not only the detection of early-stage cancers but also the detection and removal of polyps and confers a long-term protection from CRC incidence and mortality (5,10,11). A systematic review of 6 observational studies reported a pooled reduction of 69% (95% confidence interval [CI] 13%–78%) in overall CRC incidence and reduction of 68% (95% CI 57%–77%) in CRC mortality associated with screening colonoscopy (12). In the Nurses' Health Study and Health Professionals Follow-up Study, Nishihara et al. (13) reported a reduced mortality from CRC after screening colonoscopy (hazard ratio [HR] 0.32; 95% CI 0.24–0.45) overall, and a reduction in death from proximal colon cancer (HR 0.47; 95% CI 0.29–0.76). In a case-control study among US veterans, Kahi and Pohl et al. (14) reported reduction in CRC mortality of 70% with screening colonoscopy (odds ratio [OR] 0.30; 95% CI 0.24–0.38) including a 52% reduction in proximal CRC mortality (OR 0.48; 95% CI 0.35–0.66) in veterans exposed to screening colonoscopy. A nested case-control study of members of Kaiser Permanente reported a 67% reduction in the risk of death from CRC (OR 0.33; 95% CI 0.21–0.52), with a 65% reduction in proximal CRC (OR 0.35, 95% CI 0.18–0.65) (15). Brenner et al. (16) reported a reduction of 91% (95% CI 87%–93%) in CRC incidence, including a 78% reduction in proximal CRC incidence (95% CI 67%–86%) in a German population-based study of screening colonoscopy. These and other select studies are summarized in Table 3 (17–20).
TWO-STEP APPROACH SCREENING TESTS
Stool-based tests
Three large RCTs with 11–30 years of follow-up were conducted in Europe and the United States (21–23). These trials randomized average-risk individuals between ages 45 and 80 years to annual or biennial screening using guaiac fecal occult blood testing (gFOBT) compared with usual care. With biennial screening, after 13, 20, and 30 years of follow-up, there was a corresponding 18% reduction in CRC mortality (21,24–30). With annual FOBT screening, there was a sustained 33% reduction in CRC mortality
over 30 years (23). The Minnesota FOBT trial also reported a reduction in CRC incidence of 20% after 18 years of follow-up (31).
There have been many advances in stool-based tests for CRC screening. gFOBT has been largely replaced by FIT, which has higher sensitivity for CRC (32–34). The FIT sampling technique is easier as many tests require a single fecal sample and it has higher adherence than gFOBT because no dietary modifications or medication restrictions are required. Both quantitative and qualitative FITs are commercially available, with the option to set thresholds for the detection for quantitative tests based on population risk. However, in the United States, the Food and Drug Administration (FDA) requires that FIT be reported as positive or negative and does not allow reporting of the actual amount of hemoglobin per gram of stool. Studies have shown that FIT has greater sensitivity for detecting CRC and advanced adenomas than both standard and high sensitivity FOBT with comparable specificity (35,36). In a meta-analysis of 19 studies of asymptomatic average-risk adults, the pooled sensitivity of a 1-time FIT was 79% (95% CI 0.69–0.86), and pooled specificity was 94% (95% CI 0.92–0.95) for CRC at a cutoff of 20 \( \mu \) g of hemoglobin/gm of stool, the most commonly used cutoff in the United States (37). A more recent systematic review and meta-analysis including 19 studies reported 91% sensitivity and 90% specificity for FIT for detection of CRC at a quantitative cutoff of 10 \( \mu \) g/g (38). The sensitivity of a 1-time FIT for detection of advanced adenoma ranges from 6% to 56% (34) and between 5% and 16% for SSLs (39–41), based on the underlying population and FIT cutoff used. Although there are no randomized clinical trials on long-term outcomes such as reduction in CRC mortality with FIT, the programmatic effectiveness of annual FIT over multiple rounds of screening showed an overall CRC detection of 80% (42). In a cost-effectiveness modeling study, Knudsen et al. (43) found that a comparable number of life-years was gained with a screening strategy of the annual fecal-based test (such as FIT) and colonoscopy every 10 years. It must be emphasized that a positive FIT requires a follow-up colonoscopy, and quality assurance programs to ensure that colonoscopy is performed. Rates of diagnostic colonoscopy for the evaluation of a positive stool test are suboptimal (44) and require improvement in both programmatic and particularly opportunistic screening (45) (See KQ8 and KQ9 below). Implementation of an organized CRC screening program by Kaiser Permanente in Northern California based on primarily FIT outreach reported an increase in CRC screening rates from 38% to 82% among members, with a corresponding decrease in CRC incidence of 25% and CRC mortality of 52% over the same period (46). Biennial FIT is a strategy comparable to annual FIT, with similar
© 2021 by The American College of Gastroenterology
The American Journal of GASTROENTEROLOGY
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

462
Shaukat et al.
Table 2. Summary of performance characteristics for CRC screening tests
<table><tr><td></td><td>Performance characteristics</td><td>Pros</td><td>Cons</td></tr><tr><td colspan="4">Stool- and blood-based tests</td></tr><tr><td>\( FIT^a \)</td><td>79% sensitivity and 94% specificity for CRC</td><td>NoninvasiveNo risk of complicationsCan be done at homeProgrammatic screening possible</td><td>Positive results require colonoscopyNeeds to be repeated annuallyLow sensitivity for advanced adenomasDoes not detect serrated lesions</td></tr><tr><td>mtsDNA stool test</td><td>92% sensitivity and 87% specificity for CRCLong-term reduction in CRC incidence and mortality is unknown</td><td>NoninvasiveNo risk of complicationsCan be done at homeBetter sensitivity for advanced adenomas and large serrated lesions than FIT alone</td><td>Positive results require colonoscopyRepeat interval unknown but 3 years proposedMore expensive than FIT aloneConcern for overtesting and harms from a positive test and negative colonoscopy</td></tr><tr><td>Septin 9</td><td>48% sensitivity and 91% specificity for CRCLong-term reduction in CRC incidence and mortality is unknown</td><td>Minimally invasiveNo risk of complicationsCan be added to routine blood draw</td><td>Low sensitivity for CRCRepeat interval unknownPositive results require colonoscopy</td></tr><tr><td colspan="4">Direct visualization tests</td></tr><tr><td>Colonoscopy</td><td>100% detection rate for CRC. Reported incidence of PCCRC 3%-9%Long-term reduction in CRC incidence31%-71% and CRC mortality 65%-88% from observational studies</td><td>Diagnostic and therapeuticCan detect cancers and precursor polypsInfrequent repeat interval (q10 years) possible</td><td>Operator dependentRequires bowel preparation and sedationRisk of complications 4-8 in 10,000</td></tr><tr><td>Flexible sigmoidoscopy</td><td>90%-100% sensitivity for distal colon CRCLong-term reduction in CRC incidence21%; reduction in CRC mortality 26%</td><td>Less invasive than colonoscopyLow risk of complications</td><td>Positive results require colonoscopyNeeds to be repeated every 5-10 yearsRequires enema preparation</td></tr><tr><td>CT colonography</td><td>90%-100% for CRCVariable sensitivity for polyps, poor sensitivity for flat lesions and sessile serrated lesions</td><td>Less invasive than colonoscopyDoes not require sedationLower risk of complications than colonoscopy</td><td>Positive results require colonoscopyRequires bowel preparationFollowup may be required for extracolonic findingsLimited availability of trained radiologists across the United States</td></tr><tr><td>colon capsule</td><td>81% sensitivity and 93% specificity for polyps ≥6 mm</td><td>Minimally invasiveDoes not require sedationNewer generation tests can be done at home</td><td>Requires bowel preparationPositive examinations require colonoscopyRepeat interval unknown</td></tr><tr><td colspan="4">CRC, colorectal cancer; FIT, fecal immunochemical test; RCT, randomized controlled trial.\( ^{a} \)At FIT cutoff of 20 μg/g of stool.</td></tr></table>
yield of colorectal neoplasia (47) and lower burden of screening (48). Recent systematic reviews and meta-analyses have reported a relative reduction in CRC mortality of 12% over 15 years with biennial fecal-based testing (highly sensitive gFOBT or FIT) compared with no screening (49,50). In a recent guidance, the American College of Physicians also suggested biennial fecal-based stool testing as an option comparable to colonoscopy every 10 years (51). However, given the larger effect estimate of annual FOBT in clinical trials, and the effectiveness of annual FIT CRC screening programs, annual testing is the preferred interval.
The multitarget stool DNA (mtsDNA) test is an FDA-approved stool test that comprises an assay for mutant KRAS, methylated BMP3, methylated NDRG4, and a FIT for hemoglobin (39). Cutoff values are calculated by an analytic regression algorithm. In a study with 9,989 average-risk individuals undergoing colonoscopy comparing the diagnostic accuracy of the mtsDNA test with FIT alone, there was higher sensitivity for detection of CRC (92% vs 74%),
advanced adenoma (42% vs 24%), and SSLs ≥10 mm (42% vs 5%) but lower specificity for detection of CRC or advanced lesions (87% vs 95%). The specificity of the mtsDNA test decreases with advancing patient age. In 2 recent modeling studies, annual FIT and colonoscopy every 10 years were found be more effective and less costly than mtsDNA testing every 3 years (52,53). There is active interest from patients and providers in the optimal follow-up of individuals with a positive mtsDNA test and a negative colonoscopy. This is discussed in the section on special considerations below. Based on the current available data (54,55) we recommend that asymptomatic individuals with a positive mtsDNA test and a negative high-quality colonoscopy not undergo additional testing, such as upper endoscopy, CT of the abdomen, or repeat colonoscopy at an interval shorter than the recommended repeat screening interval (unless indicated by other symptoms or laboratory testing). If the mtsDNA test is negative, the interval for a repeat mtsDNA test or transition to another screening test is 3 years as per manufacturer
The American Journal of GASTROENTEROLOGY
VOLUME 116 | MARCH 2021 www.amjgastro.com
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

ACG Clinical Guidelines
463
Table 3. Select studies summarizing effectiveness of screening colonoscopy in reducing CRC incidence and mortality
<table><tr><td></td><td>US 2009</td><td>Canada 2005</td><td>Germany 2014</td><td>Switzerland 2012</td><td>US (NHS) 2013</td><td>US \(2013^b\)</td><td>US (VHA) 2018</td><td>US (Kaiser Permanente) 2018</td></tr><tr><td>Design</td><td>Cohort</td><td>Case control</td><td>Case control</td><td>Cohort</td><td>Cohort</td><td>Case control</td><td>Case control</td><td>Case Control</td></tr><tr><td>\(N^a\)</td><td>715</td><td>\(2,915^a\)</td><td>\(6,332^a\)</td><td>22,686</td><td>\(88,902^a\)</td><td>980</td><td>\(24,820^a\)</td><td>5,207</td></tr><tr><td>CRC incidence</td><td>0.52(0.22–0.82)</td><td>0.69(0.44–1.07)</td><td>0.09(0.07–0.13)</td><td>0.31(0.16–0.59)</td><td>NR</td><td>0.29(0.15–0.58)</td><td>NR</td><td>NR</td></tr><tr><td>Relative risk reduction in CRC incidence</td><td>48%</td><td>31%</td><td>91%</td><td>69%</td><td>NR</td><td>71%</td><td>NR</td><td>NR</td></tr><tr><td>Proximal CRC incidence</td><td>NR</td><td>1.02(0.72–1.45)</td><td>0.22(0.14–0.33)</td><td>NR</td><td>NR</td><td>0.36(0.16–0.80)</td><td>NR</td><td>NR</td></tr><tr><td>Relative risk reduction in proximal CRC incidence</td><td>NR</td><td>2% increase</td><td>78%</td><td>NR</td><td>NR</td><td>64%</td><td>NR</td><td>NR</td></tr><tr><td>CRC mortality</td><td>0.35(0.0–1.06)</td><td>NR</td><td>NR</td><td>0.12(0.01–0.93)</td><td>0.32(0.24–0.45)</td><td>NR</td><td>0.30(0.24–0.38)</td><td>0.33(0.21–0.52)</td></tr><tr><td>Relative risk reduction in CRC mortality</td><td>65%</td><td>NR</td><td>NR</td><td>88%</td><td>68%</td><td>NR</td><td>70%</td><td>67%</td></tr><tr><td>Proximal CRC mortality</td><td>NR</td><td>NR</td><td>NR</td><td>NR</td><td>0.47(0.29–0.76)</td><td>NR</td><td>0.48(0.35–0.66)</td><td>0.35(0.18–0.65)</td></tr><tr><td>Relative risk reduction in proximal CRC mortality</td><td>NR</td><td>NR</td><td>NR</td><td>NR</td><td>53%</td><td>NR</td><td>52%</td><td>65%</td></tr></table>
CRC, colorectal cancer; NHS, Nurses Health study; NR, not reported; VHA, Veterans Health Administration.
\( ^{a} \) N is total for study and the screening colonoscopy cohort was a subgroup.
\( ^{b} \) Included late stage cancers (stage IIB and higher only).
recommendations. Longitudinal studies on outcomes after the mtsDNA test and optimal repeat interval are awaited.
Flexible sigmoidoscopy
Flexible sigmoidoscopy allows direct evaluation of the left side of the colon and if adenomas are found, referral for a colonoscopy is required. Four large randomized trials of flexible sigmoidoscopy screening with comparable 10–13 years of follow-up have been published (56–59). Two trials, from the United Kingdom (Flexi Scope trial) and Italy (SCORE), offered once-only flexible sigmoidoscopy examination to participants age 55–64 years and reported a reduction in CRC incidence by 23% and 18% and CRC mortality by 31% and 22%, respectively (56,57). The US trial (Prostate, Lung, Colorectal and Ovarian [PLCO]), which included participants age 55–74 years and offered flexible sigmoidoscopy screening every 3–5 years, reported that CRC incidence was reduced by 21% and CRC mortality by 26% (59). One trial from Norway compared once-only flexible sigmoidoscopy plus FOBT or once-only flexible sigmoidoscopy with no screening and found comparable reduction in CRC incidence and mortality with the 2 strategies (60). In this study, flexible sigmoidoscopy screening reduced CRC incidence by 20% and CRC mortality by 27% after 11 years of follow-up (60). A systematic review reported that the largest reduction in CRC mortality from flexible sigmoidoscopy screening was seen for distal CRC, of about 37% (61). A recent study of pooled flexible sigmoidoscopy trials reported no reduction in CRC incidence or mortality in women 60 or older (58). In an updated follow-up of the Norwegian NORCCAP trial, authors also reported no reduction in CRC incidence or mortality with flexible sigmoidoscopy screening in women (62). Rates of screening flexible
sigmoidoscopy have declined in the United States for the following reasons: the infrastructure need is similar to that of colonoscopy, it does not examine the entire colon it requires colonoscopy for those with adenoma findings and the lack of sedation makes the procedure uncomfortable. Given recent evidence of lower effectiveness in women and practical issues with scheduling and availability, flexible sigmoidoscopy should be considered a screening test for individuals unwilling to undergo colonoscopy or FIT.
Alternate imaging tests for individuals who refuse colonoscopy or FIT or are not candidates for colonoscopy
CTC and CC are 2 nonendoscopic screening tests in this category. The diagnostic accuracy of CTC in average-risk screenees has been assessed in multiple studies. The sensitivity ranges between 68% and 98% for lesions ≥6 mm and 67%–94% for lesions ≥10 mm, whereas specificity ranges from 80% to 93% for lesions ≥6 mm and 86%–98% for lesions ≥10 mm (63,64). However, the diagnostic accuracy of CTC for SSLs is significantly lower than that of colonoscopy (3.1% vs 0.8% for colonoscopy and CTC, respectively) (65). There also remain concerns for detection of right-sided and flat polyps and operator dependence (66). The CC is FDA approved for imaging the colon in patients with previously incomplete colonoscopy or in patients with lower gastrointestinal bleeding who are at too high risk to undergo a colonoscopy. The test characteristics of CC have been improving because of enhancements to software and hardware. In a prospective study of 884 average-risk individuals due for screening, 79% of CCs could be completed, and sensitivity to detect adenomas ≥6 mm was 81% and specificity was 93%. For polyps ≥10 mm, the sensitivity and specificity were 80% and 97%,
© 2021 by The American College of Gastroenterology
The American Journal of GASTROENTEROLOGY
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

464
Shaukat et al.
respectively (67). Another recent study with 253 participants who underwent back-to-back colonoscopy and CC reported a per-patient sensitivity for >9 mm polyps of 87% (68). A recent comparative efficacy study of CTC and CC enrolled 321 individuals at 14 medical centers and reported that the sensitivity of CTC and CC for polyps ≥6 mm was 32% and 84%, respectively, and that for polyps ≥10 mm was 53% and 84%, respectively (69). The reasons why CTC had low diagnostic yield are unknown, and further analyses are awaited. At this time, CTC and CC are options for individuals unable to undergo colonoscopy or FIT, provided that the tests are locally available and reimbursed for the indication of screening. It is important to note that both tests require a follow-up diagnostic colonoscopy if the result is positive.
Blood-based tests
A blood-based test assessing methylated Septin 9 is FDA approved for CRC screening in average-risk individuals age 50 years or older who refuse other CRC screening methods. A screening study reported a sensitivity of 48% for CRC detection and 11% for advanced adenoma detection (70). With enhancements in the test assay, a small case-control study reported a sensitivity of 90% and a specificity of 88% for CRC (71). A recent systematic review aggregated 39 eligible studies and reported a pooled sensitivity for CRC of 62% and specificity of 90% (72). Given the low sensitivity and the lack of longitudinal and comparative data on test performance, the test is not considered an optimal screening modality at this time.
CRC screening in average-risk individuals starting at age 45 years
Recent studies have highlighted a rising incidence of CRC in individuals younger than 50 years in the United States. Although CRC incidence has continued to decline in those age 50 years and older, the incidence rates have doubled in 20- to 49-year-olds (73–75). In 2018, the American Cancer Society published guidelines with a qualified recommendation to lower the starting age for CRC screening from 50 to 45 years of age in the average-risk adult population (76), even though current recommendations of the US Preventive Services Task Force (USPSTF) and the Multi-Specialty Task Force (MSTF) are to begin screening at age 50 years (77–79). These qualified recommendations were based on predictive modeling analyses and age-cohort epidemiological evidence of a relative increase of 51% in CRC incidence among individuals younger than 50 years between 1974 and 2013 (75,80). Modeling studies using 2 of the 3 Cancer Intervention and Surveillance Network models found that initiating screening at age 45 years was on the efficiency spectrum (76) and that initiating screening at age 45 years instead of age 50 years would result in approximately 25 more life-years gained per 1,000 individuals screened.
Recent epidemiological data show alarming trend reversals for CRC incidence in the United States. Notably, CRC incidence among individuals age 50–64 years, which had historically declined by 2%–3% per year, has increased by 1% annually between 2011 and 2016 (1, 81). A similar increase in the incidence rate is observed in individuals younger than 50 years and has been partly ascribed to a birth cohort effect, in which increased CRC risk in individuals born after 1950 is carried forward as they age. It has been estimated that persons born around 1990 have twice the risk of colon cancer and 4 times the risk of rectal cancer compared with those born around 1950 (80). Although the reasons for these observations are complex and multifactorial, the fact that other developed countries are
reporting similar increases in early-onset CRC and birth-cohort effects suggests that the Western lifestyle (especially exemplified by the obesity epidemic) is a significant contributor (82).
An advantage of initiating screening at age 45 years instead of 50 years includes reduced CRC risk due to early detection of CRC in this age group. Over time, detection and removal of polyps in individuals age 45–49 years would reduce the incidence of CRC in those age 50 years and older. In addition, preventing young CRC is a desirable goal because the societal impact of CRC death at an early age is particularly devastating.
The disadvantages of such an approach are the resources required to screen an additional 21 million individuals between ages 45 and 49 years and detracting from efforts to screen individuals age 50 years and older to reach the target goal of 80% set by the National Colorectal Cancer Roundtable. Given that current rates of screening uptake are close to 60% (57.9% ages 50–64 and 62.4% ages 50–75) (83), expanding the population to be screened may reduce these rates as emphasis shifts to screening 45- to 49-year-olds at the expense of efforts to screen the unscreened 50- to 75-year-olds. A recent Markov analysis evaluated the population-level impact of lowering the screening age to 45 years and found that although it may be cost-effective to begin younger, elevating the screening rate in persons 50–75 to the target of 80% would prevent 3 times as many deaths attributed to CRC for approximately 66% less cost (84).
There are few empirical data regarding the effectiveness of screening in younger average-risk individuals, and the most appropriate screening modality in this age group is not known.
CRC screening in elderly individuals
With gains in life expectancy, and increases in the geriatric population, CRC screening in the elderly has substantial public health consequences. Because few detailed subgroup analyses of screening trials have been reported, few empirical data exist on when best to stop offering screening (85). There are several reasons why the elderly may not derive the same benefit or may even be harmed by screening. The first reason is diminished life expectancy. The benefits of polypectomy are delayed by 7–10 years after screening has occurred (86), and, thus, screening is of limited benefit for those not expected to live for at least a further 7–10 years. Second, there is an increase in competing causes of death in the elderly. The value of screening decreases as the risk of dying of other causes increases; hence, for the elderly, the benefit may become small enough to be negligible or even negatively impact their life expectancy. Third, elderly individuals may be more susceptible to risks associated with undergoing screening compared with their younger counterparts (87). These risks vary from anxiety, false-positive results, and unnecessary treatments to complications from procedures related to screening, such as dehydration, electrolyte disturbance, impact on renal function with the preparation, alteration of anticoagulation or antiplatelet agents, risk of perforation and hemorrhage during colonoscopy, and cardiovascular events periprocedure. In older patients, the benefits of early detection and prevention of CRC may be offset by higher risk of procedure-related harm and diminished health and life expectancy. Several authors have addressed this issue using different approaches (88). Lin et al. (89) reported that elderly individuals undergoing screening derive 15% or less benefit from screening compared with their younger counterparts in terms of gains in life expectancy. In their analysis of elderly individuals age between 70 and 94 years, Ko et al. (90) found that the risk of screening-related complications was higher than the estimated benefit from screening in some subgroups. A more recent simulation study found that the optimal age
The American Journal of GASTROENTEROLOGY
VOLUME 116 | MARCH 2021 www.amjgastro.com
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

ACG Clinical Guidelines
465
to forego FIT-based screening varied considerably based on sex, comorbidity status, and screening history, ranging from 66 years old for individuals with poor health and adequate previous screening to 90 years in case of individuals in good health and no previous screening (91). Therefore, the point where benefits of screening become negligible or are outweighed by potential harm likely varies significantly between individuals. The decision to continue or discontinue screening in the elderly should not be solely based on chronological age but should also take into account health status, screening history, benefits and harms of screening, and values and preferences of the patient (92–94). The most recent guideline on CRC screening from the USPSTF concluded that in adults age 76–85 years, the decision to screen for CRC should be an individualized, taking into account the patient's overall health and screening history. The guideline specifies that screening would be most appropriate for those not previously screened, those healthy enough to undergo treatment if CRC is detected, and those without substantially limited life expectancy. In adults age 86 years and above, screening is not recommended because of competing causes of mortality. The guideline identified when to stop screening as an important area of future research (85). Although further studies are awaited, providers should consider life expectancy, patient risk, values, and preferences and participate in shared decision making for screening individuals older than 75 years.
Boosting CRC screening rates in African Americans
African Americans have among the highest rates of CRC of any racial/ethnic group in the United States (1). Compared with whites, incidence rates are 24% higher in African American men and 19% higher in African American women (95). Stage-adjusted CRC mortality is also disproportionately higher in African Americans, with rates being 47% higher in African American men and 34% higher in African American women compared with whites (96). The reasons for these differences are not entirely clear but disparities in care, such as lower rates of screening, diagnostic follow-up, and treatment are postulated. One study estimates that 19% of the racial disparity in CRC mortality rates can be attributed to lower screening rates and 36% to lower stage-specific survival among African Americans (97). Health systems with equal access to the screening-diagnosis-treatment care continuum do not show such disparities (98,99). Nationally, screening rates in blacks are lower than in whites, suggesting an unmet need for efforts to improve screening in this group. Recent trends from Surveillance Epidemiology and End Results (SEER) show a decline in CRC incidence and mortality for black men and women (74). Based on recent SEER data, modeling studies (70) show similar benefit of CRC screening in African Americans and whites starting at age 45 years. Special efforts and outreach programs are needed to boost screening in African Americans to reduce the disparities in screening rates and reduce incidence rates.
Harms of screening
Harms of stool-based tests include anxiety about false-positive results and harms related to colonoscopy. The main harms of colonoscopy are bleeding (pooled event rate of 8 per 10,000) and perforation (pooled event rate of 4 in 10,000) (61). The risk of complications is greater with polypectomy and in older age groups (100). Other harms include the risk of electrolyte imbalance and nephropathy from bowel preparations or cardiopulmonary events from moderate or deep sedation and splenic injury (101,102). With colonoscopy, there is also concern for post-colonoscopy CRC (PCCRC), defined as CRC diagnosed after a
colonoscopy which did not detect cancer (103). The rate of PCCRC is estimated at 1 per 3,174 colonoscopies (95% CI 1 per 2,710 to 1 per 3,875) (103,104). A strong quality monitoring and improvement program is key to reducing PCCRC. This is discussed further in KQ6.
The harms of CTC include concerns for radiation exposure and extracolonic findings. With improved protocols requiring less radiation (1–5 mSv), the risk of radiation exposure may not be an issue. Extracolonic findings are reported in 27%–69% of studies with a wide range of work-up (105). The downstream sequelae of these incidental findings have not been adequately quantified.
The harms of CC come from the potential side effects of the preparation required before the examination (e.g., electrolyte imbalances) and the possibility of capsule retention in the small bowel. However, in the trial evaluating screening CC, no serious harms were reported (67).
KQ2. In average-risk individuals, how does the effectiveness of CRC screening vary by screening interval in reducing colorectal advanced neoplasia incidence, CRC incidence, and CRC mortality?
Recommendations
7. We recommend that the following intervals should be followed for screening modalities:
FIT every 1 year
Colonoscopy every 10 years
Strong recommendation; low-quality evidence
8. We suggest that the following intervals should be followed for screening modalities:
Multitarget stool DNA test every 3 years
Flexible sigmoidoscopy every 5–10 years
CTC every 5 years
CC every 5 years
Conditional recommendation; very low-quality evidence
DISCUSSION
There are no RCTs comparing various screening intervals. The optimal interval to repeat FOBT/FIT is not known. In the long-term follow-up of the Minnesota FOBT trial, CRC mortality was reduced by 33% with annual screening and by 18% with biennial screening (23). The European RCTs found biennial FOBT to be effective in reducing CRC mortality. One modeling study (106) favored annual FIT. In a cost-effectiveness analysis, annual FIT and colonoscopy performed every 10 years yielded similar life years gained. Ongoing RCTs that compare annual FIT and biennial FIT with colonoscopy for CRC incidence and mortality reduction will further address the question. For now, annual FIT screening is recommended. Several population-based studies have reported a low risk of CRC after a negative screening colonoscopy for at least 10 years and up to 20 years (107–109). Lee et al. reported a 46% and 88% reduced risk of CRC and CRC related deaths, respectively, up to 12 years after a negative colonoscopy (HR 0.54; 95% CI 0.31–0.94 and 0.12; 95% CI 0.02–0.82, respectively). Pilonis et al. reported the effect of colonoscopy in the Polish population compared with the general population and found the standardized incidence and mortality ratios to be reduced 10 years or 15 years after a negative
© 2021 by The American College of Gastroenterology
The American Journal of GASTROENTEROLOGY
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

466
Shaukat et al.
colonoscopy, compared with the general population (standardized mortality ratio of 0.13; 95% CI 0.0–0.17 and 0.15; 95% CI 0.06–0.31 for the 10- and 15-year interval, respectively). There are no longitudinal studies of repeat testing or interval on the mtsDNA test or Septin 9. Modeling studies performed by the Cancer Intervention and Surveillance Network have compared various screening strategies and found comparable life-years gained with the following intervals: annual FIT, colonoscopy every 10 years, flexible sigmoidoscopy every 10 years with annual FIT, CTC every 5 years, and mtsDNA test every 3 years (61). In another modeling study, rescreening 10 years after a negative screening colonoscopy at age 50 years reduced CRC compared with no further screening, and using high sensitivity FOBT or FIT annually or CTC every 5 years was less costly than continued colonoscopy (110).
KQ3. In individuals with a family history of CRC or adenomatous polyps, what is the effectiveness of CRC screening in reducing CRC incidence and CRC mortality?
KQ4. In individuals with a family history of CRC or adenomatous polyps, how does the effectiveness of CRC screening vary by screening interval in reducing colorectal advanced neoplasia incidence, CRC incidence and CRC mortality?
KQ5. In individuals with a family history of CRC or adenomatous polyps, how does the effectiveness of CRC screening vary by screening modality in reducing colorectal advanced neoplasia incidence, CRC incidence and CRC mortality?
Recommendations
9. We suggest initiating CRC screening with a colonoscopy at age 40 or 10 years before the youngest affected relative, whichever is earlier, for individuals with CRC or advanced polyp in 1 first-degree relative (FDR) at age <60 years or CRC or advanced polyp in ≥2 FDR at any age. We suggest interval colonoscopy every 5 years.
Conditional recommendation; very low-quality evidence
10. We suggest consideration of genetic evaluation with higher familial CRC burden (higher number and/or younger age of affected relatives).
Conditional recommendation; very low-quality evidence
11. We suggest initiating CRC screening at age 40 or 10 years before the youngest affected relative and then resuming average-risk screening recommendations for individuals with CRC or advanced polyp in 1 FDR at age \( \geq \) 60 years.
Conditional recommendation; very low-quality evidence
12. In individuals with 1 second-degree relative (SDR) with CRC or advanced polyp, we suggest following average-risk CRC screening recommendations.
Conditional recommendation; low-quality evidence
DISCUSSION
It is estimated that between 3% and 10% of US adults have an FDR with CRC, and higher proportions have either an FDR or SDR with CRC (111,112). On average, a family history of CRC is believed to be associated with a 2-fold increase in CRC risk, but the magnitude depends on the age of the individual at risk, age at
diagnosis of the relative(s), degree of familial relation between the individual and relative(s), and number of affected relative(s). Professional organizations (7,113) have published CRC screening recommendations, which primarily take into account the age and number of affected relatives. In all situations, it is critical for clinicians to obtain a 3-generation family history and remain alert for features suggestive of an inherited CRC syndrome, such as clustering of cancer cases within one side of a family, younger age at diagnosis, or the presence of synchronous and metachronous cancers. Maintaining a high index of suspicion for an inherited CRC syndrome and using appropriate screening tools (114) is critical because a family history of CRC or advanced adenomas can guide clinicians to appropriately modify the starting age for screening and the interval of subsequent surveillance.
Studies assessing CRC risk based on family history have varied considerably with regard to setting, patient population, degree of CRC risk, study design, and methodology. One important issue has been inconsistent distinction between the increased risk for the individual based on their family history vs risk to family members of an individual with cancer (115,116). The former scenario is the focus of the following sections because most patients present or are referred for screening because of concerns about their family history of CRC or advanced polyps.
Age of the individual at risk
Several studies and meta-analyses (115,117–119) have reported that the risk of CRC decreases with increasing age of the individual at risk. An analysis (119) of the Nurses' Health Study and Health Professionals Follow-up Study reported that the relative risk of subjects with an FDR with CRC decreased from 5.37 at ages 30–44 years to nearly 1 after the age of 65 years. A recent systematic review and meta-analysis (120) analyzed nearly 9.3 million subjects from 63 studies, and found that, overall, a family history of CRC in an FDR was associated with a higher risk of CRC (risk ratio [RR] 1.76; 95% CI 1.57–1.97). The increased risk was more pronounced in younger individuals (RR 3.29 [95% CI 1.67–6.49] for <40 years vs 1.42 [95% CI 1.24–1.62] for ≥40 years). Compelling information can also be derived from a secondary analysis (121) of the PLCO randomized CRC screening trial, which included nearly 145,000 individuals. Overall, a family history of CRC was associated with modestly increased CRC incidence (HR 1.30; 95% CI 1.10–1.50) and CRC mortality (HR 1.31; 95% CI 1.02–1.69). The HR for incident CRC in subjects with only 1 FDR with CRC was 1.23 (95% CI 1.07–1.42). The study supports the notion that the relevance of a family history of CRC was the individual at risk ages (121,122). A cost-effectiveness analysis by Naber et al. (123) estimated that the risk of developing CRC in a person with 1 affected FDR decreased with age, from 5 fold for ages 30–44 years to no difference at ≥70 years.
Age of the affected relative(s)
The age of the affected relative is inversely associated with an individual's CRC risk. Guidelines have traditionally used a dichotomous categorization with age 60 years as the threshold to designate the risk category (based on a 2-fold CRC risk cutoff), with more intensive surveillance recommended for those with an FDR <60 years old at the time of their CRC diagnosis (7,113). The recommendation to start screening at age 40 years or 10 years
The American Journal of GASTROENTEROLOGY
VOLUME 116 | MARCH 2021 www.amjgastro.com
Copyright © 2021 by The American College of Gastroenterology. Unauthorized reproduction of this article is prohibited.

ACG Clinical Guidelines
467
before the youngest affected relative is based on the seminal study by Fuchs et al. (119), showing that cumulative incidence of CRC was similar between those age 40 years with a family history and those age 50 years without a family history. A population-based study from Utah (124) reported that the risk of colorectal neoplasia among FDRs of patients diagnosed with CRC was greater when the index case was diagnosed at age <60 years (HR 2.11; 95% CI 1.70–2.63 vs 1.77; 95% CI 1.58–1.99 for ≥60 years old). Another study from Utah (125) found that the risk of CRC was increased among FDRs of index cases with CRC regardless of the age of diagnosis but was highest when both index case and FDR were younger. The HR ranged from 1.6 to 7.0 for FDR <50 years old (overall HR 2.28, 95% CI 1.86–2.80) and 1.7 to 2.3 for FDR ≥50 years old (overall HR 1.81, 95% CI 1.71–1.92). A systematic review and meta-analysis (117) estimated that the pooled risk of CRC was 3.55 (95% CI 1.84–6.83) with ≥1 FDR <50 years old compared with 2.18 (95% CI 1.56–3.04) with ≥1 FDR ≥50 years old, although the difference did not reach statistical significance. The meta-analysis by Taylor et al. (126) reported RRs of 3.31 (95% CI 2.79–3.89), 2.53 (95% CI 2.24–2.85), 2.22 (95% CI 2.04–2.40), and 1.97 (95% CI 1.83–2.12) with ≥1 FDR diagnosed at <50 years, between 50 and 59 years, between 60 and 69 years, and between 70 and 79 years, respectively. Conversely, the secondary analysis of the PLCO trial (121) reported that individuals (who were all at least 55 years old) with 1 FDR with CRC had a comparable risk of CRC regardless of age at diagnosis in the affected FDR: The HRs were 1.27 (95% CI 0.97–1.63), 1.33 (95% CI 1.06–1.62), and 1.14 (95% CI 0.93–1.45) if the FDR was <60, 60–70, and >70 years old, respectively (P trend = 0.59) (121).
Degree of familial relation
The closer the familial relation between a person and an affected relative with CRC, the higher the risk for that person. This is illustrated in an analysis of the Nordic Twin Study of Cancer (127), which showed that monozygotic twins of affected cotwins had a 3-fold increased risk of CRC compared with the general population, whereas dizygotic twins had a 2-fold increased risk. In assessing different CRC family history scenarios, the most common situation is that of 1 FDR with CRC (>90%) (126). Recent evidence suggests that the risk of CRC in an individual with an affected FDR does not depend on the identity of the relative. In a large colonoscopy study involving 16 Asia-Pacific regions (128), the risk of CRC (adjusted OR [AOR] 0.90, 95% CI 0.34–2.35, P = 0.830), advanced colorectal neoplasia (AOR 1.07, 95% CI 0.75–1.52, P = 0.714), and colorectal adenoma (AOR 0.96, 95% CI 0.78–1.19, P = 0.718) in subjects with either parent affected was similar to that of subjects with affected siblings.
The Utah population-based study (124) found that all relatives of an index person with CRC were at increased risk, including FDR, SDRs, and first cousins. However, the magnitude of the risk associated with an affected SDR is generally more modest than observed for FDR. In an analysis by Taylor et al. (126), the RRs associated with SDR alone (without concomitant FDR) ranged from 1.05 (95% CI 0.99–1.11) to 1.48 (1.11–1.93) with increasing numbers of SDRs affected, whereas the RRs associated with concomitant 1 FDR and at least 1 SDR ranged from 2.12 (95% CI 1.90–2.35) to 3.37 (95% CI 2.20–4.93).
Number of affected FDRs
The literature consistently shows that the higher the number of affected relatives with CRC, the greater the relative and absolute
risks for an individual to develop CRC (112,117,118,126). The cost-effectiveness analysis by Naber et al. (123) used the Microsimulation Screening Analysis (MISCAN) model to estimate costs and effects of colonoscopy screening strategies with different age ranges and intervals, based on lifetime CRC risk estimate inputs derived from the studies by Taylor et al (126) and Fuchs et al (119). It is important to note that the study did not take into account the age of the affected relative with CRC, and that high numbers of FDRs or FDRs and SDRs ( \( \geq \) 2) with CRC or Lynch syndrome-related cancers should trigger suspicion and workup for an underlying inherited CRC syndrome. Nevertheless, the findings show that increasing numbers of FDRs are associated with increased lif

…(92,736 more chars)