#90782: Colorectal Cancer

Excessive Alcohol Use

Solid evidence indicates that excessive alcohol use is associated with increased risk of colorectal cancer. Analysis of pooled data found that alcohol consumption greater than 45 g/day was associated with a 41% increase in risk of colorectal cancer [15,16]. The more pronounced association between current alcohol intake and larger adenomas suggests that alcohol may act during the promotional phase of the adenoma-carcinoma sequence [15,16].

Cigarette Smoking

Cigarette smoking is associated with an increased risk of colorectal cancer incidence and mortality, significantly increased risk of small and large adenomas, adenoma recurrence following polypectomy, and a long cancer induction period (35 years minimum). Rates of colorectal cancer mortality are highest in current smokers, intermediate in former smokers, and lowest in nonsmokers. Increased risk was observed after 20 years of smoking in men and women. Estimates from U.S. data attribute 12% of all colorectal cancer deaths to smoking [17,18]. Current smoking (vs. never smoking) increases the risk of developing colorectal cancer by 18% [19,20].

Obesity

Obesity, defined as a body mass index (BMI) ≥30, has been consistently associated with increased incidence and mortality from colorectal cancer, particularly in men. Compared with BMI <22, the risk of colorectal cancer increases with a BMI >28.5 by 60% in men and 30% in women. A BMI ≥30 increases the overall risk of colorectal cancer by 45%. The mechanism of increased vulnerability to colorectal cancer in obese patients is not known but may involve the elevated release and bioavailability of growth factors, insulin, and insulin-like growth factor 1. Heightened risk in obese patients appears to be mitigated by high levels of physical activity [21,22].

BMI is associated with risk of colorectal adenomas and colorectal cancer, but few studies have accrued large enough sample sizes to allow stratified analyses. Evaluation of pooled data from 8,213 participants in seven prospective studies found higher BMI was significantly associated with most histologic characteristics of metachronous adenomas in men, but not in women. The researchers concluded that body mass may affect colorectal carcinogenesis at comparatively early stages, particularly in men [23].

A study of 11,598 survivors of incident primary colorectal cancer examined the effect of obesity on risk of developing a second obesity-associated cancer (e.g., postmenopausal breast, kidney, pancreas, esophageal adenocarcinoma, endometrium). Compared with colorectal cancer survivors of normal prediagnostic BMI, the risk of developing a second obesity-associated cancer was increased 39% in overweight patients and 47% in obese patients [24]. This compares to the risk for developing a first primary obesity-associated cancer, which was increased by 18% in overweight persons and 61% in obese persons. The authors state that colorectal cancer survivors who were overweight or obese before diagnosis had an increased risk of second obesity-associated cancers relative to normal-weight survivors. Elevated risk of developing a second cancer in colorectal cancer survivors is more likely the result of the increased prevalence of overweight and obesity rather than increased susceptibility [24].

Researchers have associated a common mutation in colorectal cancer with elevated risk of metabolic disease. APC is a tumor-suppressor gene that indirectly regulates cell proliferation by encoding a protein called beta-catenin. APC inactivation by mutation leads to loss of beta-catenin function, which results in unchecked cellular replication and other processes that drive progression to malignant phenotype. Activation of the Wnt signaling pathway, normally mediated by beta-catenin, also occurs. Beta-catenin-Wnt signaling is involved in glucose metabolism and metabolic diseases such as obesity and type 2 diabetes. Using a molecular pathologic epidemiology database, researchers found that risk of beta-catenin-negative colorectal cancer was significantly higher with greater BMI and lower with increased physical activity level. Risk of beta-catenin-positive colorectal cancer was unrelated to BMI or physical activity level [25].

Polyp Removal

Removal of adenomatous polyps significantly reduces the risk of colorectal cancer. This will be discussed in detail later in this course.

Physical Activity

A sedentary lifestyle has been associated with an increased risk of colorectal cancer, although this finding has not been consistent [14]. More consistent is the association between regular physical activity and a decreased incidence of colon but not rectal cancer, with an estimated 22% to 27% risk reduction [26,27,28,29].

Diet Low in Fat and Meat

Colon cancer rates are high in populations with high total fat intakes and are lower in those consuming less fat [30]. On average, fat comprises 40% to 45% of total caloric intake in high-incidence Western countries; in low-risk populations, fat accounts for only 10% of dietary calories [31]. Several case-control studies have explored the association of colon cancer risk with meat or fat consumption as well as protein and energy intake [32]. Positive associations with meat consumption or fat intake have been found frequently but have not always achieved statistical significance [33]. One hypothesis is that heterocyclic amines formed when meat or fish are cooked at high temperatures may contribute to the increased risk of colorectal cancers associated with meat consumption observed in epidemiologic studies [34,35].

Diet High in Fiber

Despite evidence from case-control studies of a protective effect, results from a large prospective study found no difference in the risk of colorectal cancer between women in the highest quintile group compared with the lowest quintile group with respect to dietary fiber, after adjusting for age, known risk factors, and total energy intake [36]. One study evaluated the associations between dietary fiber, fat, and colorectal cancer risk in the Women's Health Initiative prospective cohort, which included 134,017 women [37]. During a mean 11.7 years follow-up (1993–2010), 1,952 incident cases of colorectal cancer were identified. When fiber and fat intake were assessed individually, the authors found a modest trend toward lower cancer risk with increased intakes of total fiber, suggesting a mild protective effect of higher fiber intake on risk of colorectal cancer, but not when combined with intake of dietary fats [37]. Results of a pooled analysis of 3,209 participants combined from two trials indicate that men may experience more benefit from dietary fiber than women [38].

Diet High in Fruits and Vegetables

Overall, results from more rigorously designed randomized controlled trials have washed out findings of significant correlation in earlier studies that linked higher fruit and vegetable consumption with lower rates of colorectal cancer. Diets low in fat and meat and high in fiber, fruits, and vegetables started as an adult do not appear to reduce the risk of colorectal cancer by a clinically important degree [37,39].

Cohort studies have demonstrated associations between specific diet or exercise regimens with improvements in disease-specific and/or overall survival in patients following treatment for colorectal cancer, but these results have not been replicated by prospective randomized trials. When verification by more rigorous studies is absent, cohort study data should be interpreted with caution, because numerous uncontrolled variables are present that may confound the observational findings [14].

Physical Activity

A meta-analysis of prospective cohort studies evaluating physical activity in patients found a 25% reduction in colorectal cancer-specific mortality associated with any amount of physical activity (vs. no activity) and a 30% reduction associated with a high amount of physical activity (vs. low amount). After colorectal cancer was diagnosed, a 26% reduction in colorectal cancer-specific mortality was associated with participation in any physical activity (vs. no activity), and a 35% reduction was associated with a high amount of physical activity (vs. a low amount) [40].

A 2022 analysis used data from the Global Burden of Disease 2019 study to analyze colorectal cancer deaths associated with low physical activity and high body mass index (BMI) [41]. The analysis included data from 1990 to 2019 at global, regional, and national levels. In 2019, colorectal cancer deaths attributed to low physical activity and high BMI were an estimated 58.7 and 85.9 per 100,000 population, respectively. Corresponding age-standardized mortality rates were 0.77 (low physical activity) and 1.07 (high BMI). Since 1990, age-adjusted mortality rates from colorectal cancer attributable to low physical activity and high BMI have increased in many geographic regions, particularly in low-middle and middle sociodemographic index regions. Countries with a higher baseline burden in 1990 and a higher sociodemographic index in 2019 had a faster decline in age-adjusted mortality rates of colorectal cancer attributed to high BMI and low physical activity [41].

Diet

Among the observational study findings, patients with stage III colon cancer who had the lowest Western dietary pattern post-treatment showed significantly greater rates of disease-free survival and overall survival versus patients with highest Western dietary pattern [42,43]. Also, patients with the highest dietary glycemic load showed significantly greater overall survival rates compared with those with the lowest dietary glycemic load. Another uncontrolled cohort study of patients diagnosed with colorectal cancer found the extent of red and processed meat ingestion was associated with a 29% greater risk of death before colorectal cancer diagnosis, but red meat ingestion after diagnosis had no effect on overall mortality [44].

Plasma Vitamin D Level

The National Comprehensive Cancer Network does not currently recommend routine screening for vitamin D deficiency or supplementation of vitamin D in patients with colorectal cancer.

(https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf Last Accessed: March 14, 2022)

Level of Evidence: 2a (Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate.)

There is evidence that vitamin D may be an important cofactor in immune protection against colorectal cancer risk. A large, population-based case-control study, derived from the Nurses' Health Study and Health Professionals Follow-Up Study, found a significant association between plasma vitamin D level and colorectal cancer risk according to the degree of local antitumor immune response. The study consisted of 318 colorectal cancer cases and 624 matched controls. Subjects were divided into three groups based on the median plasma vitamin D level (tertile I 19.0 ng/mL, tertile III 37.4 ng/mL) and analyzed according to the degree of lymphocytic immune reactivity within and surrounding the tumor. Subjects in the highest vitamin D tertile were seen to have a significantly lower risk of developing colorectal cancer subtype showing an intense intratumoral cellular immune reaction. This association was not found for tumor subtypes characterized by a poor intratumoral immune response. The authors discuss possible mechanisms and conclude that these observations support a role for vitamin D in cancer immunoprevention through tumor-host interaction [45].

Chemopreventive agents are often prescribed to healthy subjects at risk for colorectal cancer, who will take the agent for the rest of their lives to prevent a potential cancer. In addition to the preventive benefit, this raises the bar very high when defining acceptable safety and toxicity [46].

Practice guidelines and expert opinion have been hesitant to recommend chemoprevention of colorectal cancer. One reason is that very promising earlier findings have often washed out under rigorous evaluation. Epidemiologic and large cohort studies have found a number of agents with significant association to reduced colorectal cancer risk. Not infrequently, these findings were verified by other observational studies, followed by identification in pre-clinical research of plausible mechanisms for a cause-effect relationship. However, results from rigorous investigation using well-designed randomized controlled trials reveal new safety concerns or fail to confirm the significant relationships suggested by data from uncontrolled trials. Thus, guideline authors and experts are reluctant to suggest chemoprevention in the absence of large-scale, long-term, randomized controlled trials [47].

Use of surrogate endpoint markers in many chemoprevention trials may also dissuade recommendation. As the precursor of most colorectal cancers, adenomas have often been used as surrogate endpoints in efficacy evaluation of agents for prevention. Their use as surrogate markers of colorectal cancer in chemoprevention randomized controlled trials permits the reduction of the study observation period from roughly 10 years required for assessing colorectal cancer development to around 2 years. Despite the theoretical and pragmatic basis, preventive efficacy based on this surrogate endpoint may contribute to reluctance in recommending colorectal cancer chemoprevention [46].

The true benefit of chemoprevention is reliant on lifetime colorectal cancer risk in the patient population. Greatest potential benefit may come from use in patients diagnosed by colonoscopy with pre-malignant lesions, with family history of colorectal cancer, or genetically diagnosed and surgically resected for colorectal tumors. Chemoprevention will probably show modest benefit at best when used as prevention in average-risk patients [48,49].

Cyclooxygenase Inhibitors

A 2015 prospective observational study published the first-ever results of cyclooxygenase-2 (COX-2) inhibitor and aspirin use as adjuvant therapy following resection in patients with stage III colon cancer. All patients received standard adjuvant chemotherapy with fluorouracil (5-FU) plus leucovorin with or without irinotecan. In the aspirin arm of 799 patients, 75 (9.4%) used aspirin during and after chemotherapy. In the COX-2 inhibitor arm of 843 patients, 59 (7.5%) used celecoxib or rofecoxib after completing chemotherapy. Both groups had a median follow-up of 6.5 years [50]. Among patients taking aspirin (vs. no aspirin), recurrence-free survival (i.e., time period until tumor recurrence, death with recurrence, or development of a new invasive colon cancer) was increased by 49%, disease-free survival (i.e., time period until tumor recurrence, occurrence of a new colon cancer, or death from any cause) was increased by 32%, and overall survival (i.e., time period until death from any cause) was increased by 37%. Adjusted hazard ratios were censored at five years to minimize misclassification from non-cancer death and showed increases in disease-free survival by 39% and overall survival by 52% (vs. no aspirin). Patients taking a COX-2 inhibitor (vs. no COX-2 inhibitor) found increases in recurrence-free survival by 47%, disease-free survival by 40%, and overall survival by 50%. Censor of survival data at five years found disease-free survival increased by 53% and overall survival by 74% [50].

Although this study was not designed to identify the optimal dose and duration of aspirin or COX-2 inhibitors for protection against colorectal cancer, the data suggest a dose-response relationship in aspirin with increased frequency, while any dose of COX-2 inhibitors was associated with benefit. The statistically significant associations between aspirin and COX-2 inhibitor use and reduced colon cancer recurrence and mortality found in this study will continue to be evaluated [50].

Celecoxib, rofecoxib, and aspirin share a similar mechanism of action in colon (and presumably rectal) cancer involving COX-2 inhibition. COX synthesizes the conversion of arachidonic acid to prostaglandins. Prostaglandins mediate tumor growth by altering stem cell gene expression, hypermethylating genes involved in proliferation and differentiation, promoting angiogenesis and Wnt/CTNNB1 signaling, and inhibiting apoptosis. Thus, suppression of prostaglandin synthesis through COX inhibition interferes with the processes involved in tumor promotion and growth [50,51].

Long-term follow-up data from two large studies initiated in the 1980s found that ≥300 mg aspirin daily taken for five or more years was associated with a 37% overall reduction in colorectal cancer risk. In subjects who remained adherent to the protocol for 5 or more years, those randomized to aspirin were found to have a 40% risk reduction in colorectal cancer mortality after 20 years and absolute risk reduction from 3.1% to 1.9% relative to those receiving placebo. Mortality reduction was primarily from the effect of aspirin on proximal colon cancer. These findings were serendipitous, because the research was designed to examine the protective effects of aspirin against cardiovascular events [52,53].

Prospective studies have demonstrated significant reduction in colorectal cancer among regular aspirin users [54]. In a randomized controlled trial of 861 persons with Lynch syndrome, primary colorectal cancer developed in 4.2% of patients taking daily aspirin 600 mg, compared with 6.9% in those receiving daily placebo (mean follow-up: 55.7 months). Time to first colorectal cancer was increased 37% with aspirin versus placebo; with regression analysis incorporating multiple primary events, aspirin led to a 44% reduction in colorectal cancer incidence. In subjects completing at least two years of intervention, time to first colorectal cancer was increased 59% and incidence of colorectal cancer was reduced 63%. Adverse events did not differ between aspirin and placebo groups during the intervention [55]. A planned 10-year follow-up to this trial (the double-blind, randomised CAPP2 trial) included 861 patients with Lynch syndrome from 43 international centers worldwide. The participants were randomly assigned to receive either 600-mg aspirin daily (427 participants) or placebo (434 participants). Cancer outcomes were monitored for at least 10 years from recruitment; some of the participants (i.e., English, Finnish, and Welsh participants) were monitored for up to 20 years. The primary endpoint was development of colorectal cancer [56]. Forty (9%) of the aspirin group developed colorectal cancer compared with 58 (13%) of the placebo group. Noncolorectal Lynch syndrome cancers were reported in 36 participants who received aspirin and 36 participants who received placebo. Adverse events between the aspirin and placebo groups were similar [56]. Likewise, a randomized controlled trial of patients with a history of adenomas or colorectal cancer found a statistically significant 21% reduction in risk of adenoma recurrence in patients randomized to aspirin (versus placebo) [57].

A prospective cohort study examined the effects of aspirin in participants following a diagnosis of colorectal cancer. Regular use of aspirin after colorectal cancer diagnosis was associated with a 29% increase in colorectal cancer-specific survival and a 21% increase in overall survival [58]. In the long-term Nurses' Health Study and the Health Professional Follow-up Study, 964 patients diagnosed with rectal or colon cancers were evaluated. In those with PI3K-mutant colorectal cancer, regular use of aspirin was associated with a 46% increase in overall survival [59].

The benefit of aspirin in prevention of colorectal cancer is not apparent until 10 years after aspirin therapy is started and is most effective when started between 50 and 59 years of age. Because of the time required before a reduced incidence in colorectal cancer is realized, persons 60 years of age and older are less likely to realize a benefit [60]. Additionally, aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs) are associated with potentially serious adverse effects that should be considered when determining the risk-benefit ratio [57]. Aspirin use can result in excessive bleeding, gastrointestinal bleeds, and hemorrhagic stroke. The estimated average increased risk of upper gastrointestinal complications was 10 to 30 per 1,000 people over a 10-year period, with men on the higher end and women on the lower end. Risk increases with age [60].

While no studies have assessed adenoma or colorectal cancer risk reduction with use of NSAIDs in the general (and presumably average-risk) population, multiple lines of evidence from epidemiologic studies, observational cohort studies, and randomized controlled trials have consistently affirmed the association between NSAID use and a 30% to 50% reduction in adenomatous polyps, incident disease, and death from colorectal cancer [57,61,62,63]. In one study, patients with familial adenomatous polyposis (FAP) who were followed over four years of treatment with NSAIDs showed a trend in reduction in adenoma incidence and statistically significant reductions in polyp number and size. A 34% reduction in adenoma recurrence risk and a 55% reduction in advanced adenoma incidence were found in patients with a history of adenomas [57].

The NSAIDs sulindac and celecoxib have been shown in randomized controlled trials to induce adenoma regression in patients with FAP, which, together with supportive preclinical data, led the U.S. Food and Drug Administration (FDA) to approve celecoxib for patients with FAP in 1999. However, in 2011, the FDA requested Pfizer voluntarily withdraw the FAP indication for celecoxib, because the company never fulfilled a condition for approval requiring postmarketing evaluation to verify clinical benefit, which Pfizer did [64]. Despite the change of celecoxib use in FAP to off-label status and withdrawal of regulatory approval, several health insurance companies have codified the use of celecoxib in FAP as an authorized indication [65].

The consistently positive findings of NSAID benefit in suppressing the development of adenomas and improving recurrence-free, disease-free, and overall survival in patients with histories of adenomas and colon cancer has posed a dilemma for researchers and clinicians, given the known toxicity profile. NSAID-related morbidity is fairly common and potentially serious and includes upper gastrointestinal bleeding, renal dysfunction, and serious cardiovascular events such as myocardial infarction, heart failure, and hemorrhagic stroke. Among other findings, use of NSAIDS increases the risk of serious cardiovascular events by 50% to 60% [61,66].

Hormones (for Women Only)

The Women's Health Initiative (WHI) randomized participants to estrogen plus progestin or placebo. At a mean follow-up of 11.6 years, women receiving active hormone therapy had a 28% lower risk of colorectal cancers [67]. However, in the hormone therapy group, colorectal cancers that developed were significantly more likely to exhibit lymph node involvement and higher stages (regional and distant) compared with those in the placebo group. Deaths from colorectal cancers in the active group were somewhat higher, but the difference from placebo was not statistically significant [67]. A meta-analysis of cohort studies observed a 14% risk reduction for incidence of colorectal cancer associated with combined hormone therapy [68].

Conjugated equine estrogens do not improve incidence or survival in invasive colorectal cancer [67]. Definite harms have been established in using combined estrogen plus progestin hormone in postmenopausal women. The WHI trial found increased risks of invasive breast cancer, coronary heart disease events, and thromboembolic events [67,69].

Vitamin E

A prospective cohort study of 35,215 women found an inverse association between the risk of colon cancer and vitamin E intake [70]. However, a later cohort study found no relationship between every-other-day use of vitamin E 600 IU and colorectal cancer, and a meta-analysis of 14 randomized trials of supplemental antioxidant vitamins involving 170,025 individuals found no evidence for prevention of colorectal adenoma or colorectal cancer [71,72].

Vitamin D

A systematic review of published cohort studies found that daily intake of 1,000 IU of vitamin D and 25-hydroxyvitamin D serum concentration of 33 ng/mL were each associated with a 50% risk reduction of colorectal cancer [73]. A population-based case-control study found an inverse relationship between vitamin D intake and colorectal cancer risk [74]. More recent research is focused on the role of vitamin D as an adjunct treatment after a diagnosis of colorectal cancer. For example, two randomized, placebo-controlled trials of vitamin D in patients with metastatic colorectal cancer are underway to assess patient survival as a primary endpoint. The first study is a phase II trial comparing high-dose vitamin D3 (8,000 IU/day for two weeks followed by 4,000 IU/day) versus a standard dose (400 IU/day). The second study is a phase I–II trial comparing customized oral doses of vitamin D3 titrated to raise serum 25(OH)D levels to 80–100 ng/mL versus 2,000 IU/day. The results of these and subsequent phase III trials may provide more definitive answers about the role of vitamin D in the treatment of colorectal cancer [75].

Folate

An observational study of women with a family history of colon cancer found use of folic acid supplements for more than 15 years was associated with a 75% lower risk of colorectal cancer [76]. One hypothesis is that folate is required for DNA synthesis, and suboptimal amounts may cause abnormalities in DNA synthesis or repair [77]. However, a trial that randomized 1,021 men and women with recent colorectal adenoma history to daily folic acid 1 mg or placebo found folic acid was associated with greater risks of developing ≥1 advanced adenoma, ≥3 adenomas, and extra-colonic malignancy compared with placebo [78]. General population studies have not found benefit of folic acid on colorectal cancer risk, but outcomes obtained over relatively short duration may have missed detection of benefit from longer exposure and/or follow-up [57].

Calcium

Researchers have suggested that calcium's action of binding bile acids and fatty acids may lower colon cancer risks through reducing exposure to toxic intraluminal compounds [79]. To study the effects of calcium on adenoma recurrence, persons with a recent history of colorectal adenomas were randomized to daily 3 g calcium carbonate (1,200 mg elemental calcium) or placebo. At four-year follow-up, those receiving calcium (compared with placebo) showed a 19% reduction in developing one or more recurrent adenoma and the average number of adenomas was 24% lower. This reduced risk was likely to extend up to five years following cessation of calcium supplementation [80,81].

Calcium has not shown benefit in patients with FAP. In the general population, there was no significant effect of calcium on risk of colorectal cancer, although studies were of relatively short duration [57]. There is fair evidence that 1,000–1,200 mg/day oral calcium without vitamin D supplementation increases the risk of myocardial infarction. Calcium supplementation with vitamin D at doses less than 1,000 mg/day has few harmful effects [82,83].

Clinicians should perform an individualized assessment of colorectal cancer risk in all adults in order to understand patient risk level for colorectal cancer. Patient risk is assessed by a thorough personal and family history to identify factors associated with increased vulnerability to colorectal cancer. The colorectal cancer risk factors of smoking, obesity, coronary artery disease, diabetes, acromegaly, renal transplantation, and cholecystectomy have no bearing on the timing, frequency, and modality of colorectal cancer screening or intervention (in the absence of adenomatous polyps or colorectal cancer) [86].

Familial Colorectal Cancer Risk Factors

A targeted colorectal cancer family history should include a detailed family history of cancer and polyps, especially in first-degree (e.g., parent, sibling, child) and second-degree (e.g., grandparent, uncle/aunt, half sibling) relatives on both sides of the family [87,88]. Clinicians should ask about polyps in relatives, including [89]:

In addition to familial factors, patients' personal risk factors should also be assessed (Table 1).

Patients should be assessed for all cancer types. Cancer syndromes include risk for multiple types of malignancy; colorectal cancer is not always a presenting cancer. A three-generation pedigree is the gold standard. The minimum for colorectal cancer should include cancer and polyp history for the patient's generation and two previous generations. The patient's risk status can change over time with updated personal or family history.

Assessment Red Flags

Findings suggestive of heritable colorectal cancer risk are termed "red flags" and direct the healthcare provider to probe further. One red flag is a personal history of colon cancer diagnosed before 60 years of age or endometrial cancer diagnosed before 50 years of age [87,88]. Early age at diagnosis suggests that genetic factors are playing a strong role in the development of disease.

A family history of colon or endometrial cancer diagnosed before 50 years of age is another red flag. Early age at diagnosis of cancer in a closely related family member suggests that genetic factors are playing a role in the development of disease, and these factors can be passed on to other relatives. If multiple family members have been diagnosed with colorectal cancer, or other Lynch/hereditary nonpolyposis colorectal cancer (HNPCC)-related cancers, this strongly suggests genetic factors are increasing individual cancer risks, especially among first-degree relatives [87,88].

One to two polyps in a lifetime is common, but more than 10 in a lifetime is unusual and suggests genetic contribution. Polyposis is associated with increased colorectal cancer risk. In addition, diagnosis of two or more Lynch/HNPCC-associated cancers suggests an inherited mutation, increasing the overall risk for cancer in different organs.

Past diagnosis of Lynch/HNPCC, FAP, or other inherited cancer syndrome in a family member is another risk factor. Many of these conditions are inherited in a dominant pattern, but not everyone who inherits gene mutations for these conditions develops cancer. Therefore, a diagnosis of HNPCC in a grandparent may be relevant to the patient.

Patient Colorectal Cancer Risk Level

Of total colorectal cancer cases, 75% are due to sporadic disease without apparent inherited origin, 10% to 30% are due to familial risk factors, and 5% to 6% are due to heritable genetic mutations. The absolute risk of colorectal cancer by 79 years of age is [90,91]:

Family history of two or more relatives with colorectal cancer substantially increases the possibility of a genetic syndrome, and relative to older individuals, young patients reporting a positive colorectal cancer family history are more likely to represent a high-risk pedigree [92,93]. Patient risk level is categorized as high, increased (moderate), or average based on the presence of specific factors (Table 2) [94].

Heritable gene mutations that confer elevated risk of colorectal cancer broadly cluster into two groups: stability genes, including mutations in DNA mismatch repair (MMR) genes responsible for Lynch syndrome, and tumor suppressor genes, including APC gene mutations responsible for FAP. Lynch syndrome and FAP account for the vast majority of heritable colorectal cancer cases and 5% to 6% of all colorectal cancer cases [95]. The absolute risks for colorectal cancer in mutation carriers of hereditary colorectal cancer syndromes are [95]:

Individuals with single-gene disorders are at increased risk of developing colorectal cancer, and single-gene disorders related to known syndromes account for 10% to 15% of colorectal cancer cases. The hereditary syndromes and involved genes include Lynch syndrome, FAP, familial colorectal cancer, and rare genetic syndromes [89].

Lynch Syndrome

Lynch syndrome is the most prevalent form of hereditary colorectal cancer, accounting for 3% to 5% of all cases. It primarily involves defects in MMR genes, most commonly MSH2, MLH1, PMS1, PMS2, or MSH6. In affected families, 15% to 60% of family members possess MSH2 or MLH1 mutations [90,95,96].

Lynch syndrome is an autosomal dominant disorder in which families and patients possess a germline mutation in a DNA MMR gene or loss of expression of the MSH2 gene due to deletion in the EPCAM gene. These genes function to maintain DNA fidelity during replication and are inactivated in Lynch syndrome [97].

Genetic Testing. Genetic risk assessment of Lynch syndrome considers family cancer history and patient age if diagnosed with colorectal cancer or malignancies associated with Lynch syndrome. Mutation in MMR genes can be detected using immunohistochemistry techniques (IHCs) or DNA microsatellite instability (MSI) analysis. Several validated computer models predict MMR gene mutation probability (even when MSI or IHC information is absent) and also incorporate family history of endometrial cancer. Mutation detection rates are higher for patients with more striking family histories or informative tumor testing data [98,99].

Clinical Features. Colorectal cancer and extracolonic malignancies are the primary consequences of Lynch syndrome. Colorectal cancer associated with Lynch syndrome is characterized by early age of onset, excess synchronous and metachronous colorectal neoplasm, right-sided dominance (roughly 67%), and extracolonic tumors. The average age of colorectal cancer diagnosis in patients with Lynch syndrome is 44 to 52 years, versus 71 years in sporadic colorectal cancer. MLH1 and MSH2 account for close to 90% of gene mutations, and the lifetime risk of colorectal cancer in MLH1 and MSH2 mutation carriers is 68.7% in men and 52% in women [97].

Risk of extracolonic malignancy is greatest for endometrial cancer. At least one female member in about half of all Lynch syndrome pedigrees is affected, and 50% of women with an MMR gene mutation present with endometrial cancer as first malignancy. Patients with Lynch syndrome have an elevated risk of several other cancers. Risk of extracolonic tumor development by 70 years of age in Lynch syndrome is shown below, with prevalence rate ranges reflecting differences between specific MMR mutations [97]:

The adenoma-carcinoma sequence of polyp-to-cancer dwell time is an estimated mean 35 months, considerably more rapid than the 10- to 15-year average in sporadic colorectal cancer. This accelerated rate is likely the result of MMR gene dysfunction that creates frequent DNA mismatches in multiple genes to disrupt their normal function [97]. Until recently, Lynch syndrome was termed hereditary nonpolyposis colorectal cancer, a misnomer because polyps are usually present [100].

Diagnosis. Clinical criteria to identify patients with Lynch syndrome were published in 1990 and termed the Amsterdam criteria. These were revised and expanded with the 1999 Amsterdam II criteria, which included extracolonic cancers. The Amsterdam II defines minimum criteria for a clinical diagnosis of Lynch syndrome as at least three relatives with a Lynch-associated cancer (e.g., colorectal cancer, endometrial, small bowel, ureter, renal pelvis) and [95,101]:

The 2004 updated Bethesda Guidelines were developed to improve the false-negative rates with Amsterdam II and outline criteria to prompt MSI tumor testing to identify Lynch syndrome. Tumors meeting one or more of these criteria require testing for MSI [95,97,102]:

Although more sensitive than Amsterdam II in identifying families with Lynch syndrome, only 15% to 30% of families not meeting Amsterdam II but meeting Bethesda criteria exhibit MSI gene mutation. Thus, Amsterdam II or Bethesda criteria may be used to help identify patients who should receive genetic testing, but they should not be used as diagnostic instruments [95,103].

Surveillance. The differing surveillance approach in persons with Lynch syndrome relative to average-risk persons is dictated by the biologic behavior of Lynch syndrome [89]. Lynch syndrome develops earlier than sporadic colorectal cancer, which suggests screening should begin earlier in life. Most Lynch syndrome colorectal cancers occur in the right colon, making sigmoidoscopy alone insufficient. Annual colonoscopic surveillance is recommended [95]. The accelerated progression from normal mucosa to adenoma to cancer suggests a shorter colonoscopy screening interval (i.e., every one to two years). The substantially higher lifetime incidence of colorectal cancer suggests that surveillance should use the most sensitive test available.

Patients with Lynch syndrome are at an elevated risk of extracolonic cancers, especially endometrial and ovarian. While routine screening in women with Lynch syndrome is recommended due to substantially increased risk of endometrial cancer, routine transvaginal ultrasound screening for endometrial cancer is insensitive, nonspecific, and without benefit in the general population.

Interventions. A study randomized 861 Lynch mutation carriers to daily aspirin 600 mg or placebo. No difference was found at 24 months, but 56-month follow-up found somewhat lower adenoma rate and colorectal cancer risk in the aspirin group. Further analysis and a planned 10-year follow-up found decreased incidence of all Lynch-associated cancers in the aspirin group [55,56].

Prophylactic surgery is an alternative to annual colorectal cancer and endometrial cancer screening. The high risk of developing metachronous lesions is the basis for prophylactic surgery [103]. The incidence of metachronous colorectal cancers has been reported to be 16% at 10 years, 41% at 20 years, and 63% at 30 years following segmental colectomy [104]. With the increased incidence of synchronous and metachronous neoplasms, the treatment of choice for a patient with Lynch syndrome with neoplastic lesions in the colon is generally an extended colectomy. The results of a follow-up study help in the selection of surgical approach. In this trial, 382 MMR mutation carriers were followed over time after surgery. During follow-up, metachronous colorectal cancer developed in no patient receiving total or subtotal colectomy compared with 22% of patients receiving segmental colectomy [105,106]. An important factor in the decision to offer prophylactic surgery is the ability of the patient to comply with surveillance examinations.

Consideration of total or subtotal colectomy should be balanced with patient comorbidities, clinical stage of the disease, patient wishes, and surgical expertise. One retrospective study examined data collected on 242 patients with Lynch syndrome who underwent surgery for a first colon cancer between 1984 and 2009 [107]. Patients underwent either standard segmental colectomy or extended colectomy. Primary outcomes measured were risk of subsequent colorectal cancer, overall and disease-specific survival, and operative mortality. One patient died of postoperative septicemia within 30 days after segmental colectomy. Subtotal colectomy decreased the risk of subsequent colorectal cancer compared with segmental resection. The cumulative risk of subsequent colorectal cancer was 20% in 10 years and 47% within 25 years after standard resection, and 4% and 9% after extended surgery, respectively. However, disease-specific and overall survival within 25 years did not differ significantly between the standard and extended surgery groups (82.7% vs. 87.2%) [107]. Although no data have been published showing a survival advantage in extended versus segmental resection for patients with Lynch syndrome, clinicians might consider extensive colectomy to prevent subsequent colorectal cancer in patients with Lynch syndrome [108]. Also, subtotal or total colectomy does not eliminate rectal cancer risk, and the risk of developing rectal cancer following abdominal colectomy is estimated at 12% at 12 years post-surgery [104,109].

Familial Adenomatous Polyposis

FAP accounts for 1% of all colorectal cancers and involves germline mutations in the tumor suppressor gene APC [90,96]. Ashkenazi Jews have elevated risk of colorectal cancer due to APC gene mutation, which occurs in 6% to 7% of this population [110]. Other FAP disorder variants include [90,96]:

Genetic diagnosis of FAP in pre-symptomatic patients is performed with linkage or direct detection of APC mutations by analyzing lymphocyte DNA in a blood sample. Linkage analysis tests blood samples from multiple persons to identify gene carriers in close and ancillary family members [90,96,103].

Clinical Features. FAP is caused by parental transmission of mutation in the APC gene, a tumor suppressor or gatekeeper gene that controls cell proliferation. The most common FAP phenotype is development of hundreds to thousands of colorectal polyps, with usual onset during adolescence or early adulthood. Malignancy develops in one or more polyps as early as 20 years of age, and colorectal cancer develops in almost 100% of patients by 40 years of age if the colon is not removed for primary prevention. Other characteristics of FAP can include polyps in the upper gastrointestinal tract; extracolonic manifestations, such as congenital hypertrophy of retinal pigment epithelium, osteomas and epidermoid cysts, supernumerary teeth, and desmoid formation; and other malignancies, such as thyroid tumors, small bowel cancer, hepatoblastoma, and brain tumors (particularly medulloblastoma) [90,96,103]. The lifetime risk of extracolonic tumor development in FAP is [89]:

Diagnosis. The clinical diagnostic criteria of FAP is a patient with 10 to 99 adenomatous colon polyps diagnosed by 40 years of age, or more than 100 polyps diagnosed at an older age than expected [89].

Surveillance. The recommended age at which surveillance for polyposis should begin involves a trade-off. On one hand, a patient who waits until the late teens to begin surveillance faces a remote possibility that a cancer will have developed at an earlier age. Although it is rare, colorectal cancer can develop in a teenager who carries an APC mutation. On the other hand, it is preferable to allow people at risk to develop emotionally before they are faced with a major surgical decision regarding the timing of colectomy. Therefore, surveillance is usually begun in the early teenage years (10 to 15 years of age) [111]. Surveillance has consisted of either flexible sigmoidoscopy or colonoscopy (preferred) every year. If flexible sigmoidoscopy is utilized and polyps are found, colonoscopy should be performed. Historically, sigmoidoscopy may have been a reasonable approach at the time in identifying early adenomas in a majority of the patients [112]. However, colonoscopy should be considered the tool of choice in light of improved instrumentation for full colonoscopy, safer and deeper sedation (with propofol), recognition that malignancy is more common in the right colon with attenuated FAP, and the growing tendency to defer surgery for a number of years. Individuals testing negative for an otherwise known family mutation do not need FAP-oriented surveillance and can undergo average-risk population screening. In the case of families where no family mutation has been identified in an affected person, clinical surveillance is warranted [90,96,103].

Colon surveillance should not be stopped in carriers of an APC mutation who do not yet manifest polyps, because adenomas occasionally do not appear before the fourth and fifth decades of life. In some circumstances, full colonoscopy is preferred over the more limited sigmoidoscopy. Tolerability of endoscopic procedures among pediatric patients has improved with the use of deeper intravenous sedation [90,96,103].

Interventions. After an APC mutation is identified in a patient or member of their family, evaluation for polyposis by flexible sigmoidoscopy or colonoscopy begins promptly. In those showing polyps, the only effective management to prevent colorectal cancer is eventual colectomy. In patients with early-stage classic FAP, the surgeon, endoscopist, and patient/family may opt to delay surgery for several years in the interest of achieving social milestones. Carefully selected patients with attenuated FAP who show minimal polyp burden and are of advanced age may also defer decision-making about colectomy [111].

The timing of risk-reducing surgery is based on symptomatology and the number, size, and histology of polyps. Surveillance colonoscopy is not useful after numerous polyps have developed, because it is no longer possible to remove and biopsy all of them. It is appropriate for patients at this time to consult with a surgeon experienced with available options, including total colectomy and postcolectomy reconstruction techniques. Rectum-sparing surgery followed by sigmoidoscopic surveillance of the remaining rectum is an option for patients who wish to avoid total colectomy, provided they are able to understand the risks and consequences and to follow through with surveillance recommendations [111].

Familial Colorectal Cancer

Many families exhibit aggregation of colorectal cancer and/or adenomas in the absence of known or identifiable genetic susceptibility factors; this is termed familial colorectal cancer [113]. The presence of colorectal cancer in more than one family member may be caused by hereditary factors, shared environmental risk factors, or even chance. Familial colorectal cancer accounts for 20% of all colorectal cancer cases [114].

In the general population, 7% to 10% of individuals have a first-degree relative with colorectal cancer and 14% to 20% have either a first-degree or a second-degree relative with colorectal cancer [90,96,103]. A simple family history of colorectal cancer (i.e., colorectal cancer in one or more close relatives, known hereditary colon cancer absent) confers a two- to six-fold increase in risk, with degree of risk influenced by family member's age of colorectal cancer onset, the number of affected relatives, closeness of the genetic relationship, and whether colorectal cancer has occurred across generations. A positive family history of colorectal cancer appears to increase the risk of colorectal cancer earlier in life such that at 45 years of age, the annual incidence is more than three times higher than in average-risk people; at age 70 years, the risk is similar to that in average-risk individuals [115]. The incidence in individuals 35 to 40 years of age is about the same as that of an average-risk person at 50 years of age. There is no evidence to suggest that colorectal cancer in people with one affected first-degree relative is more likely to be proximal or more rapidly progressive [90,96,103].

Although controlled comparisons have not been performed of genetic screening in persons with modest family history of colorectal cancer, expert opinion is fairly consistent that colorectal cancer screening should begin earlier in life (35 to 40 years of age, when risk magnitude approximates that of an individual 50 years of age) [115]. Screening in persons with average risk of colorectal cancer should begin at 50 years of age with repeat screening every 10 years. Increased risk with greater extent of family history warrants room for clinical judgment in favor of even earlier screening based on family history, and shortening the frequency of screening interval to every five years. There is no empirical or logical support to initiate colorectal cancer screening 10 years younger in age than the family member with youngest age of colorectal cancer detection [90,116].

Other Genetic Factors

In addition to FAP and Lynch syndrome, several rare genetic syndromes confer an increased risk for colorectal cancer, including [90,96]:

Factors that Suggest Hereditary Colorectal Cancer Predisposition Syndrome

With the exception of autosomal recessive inheritance with MYH-associated polyposis, all gene mutations known to cause colorectal cancer predisposition are inherited in an autosomal dominant fashion [113]. Thus, family characteristics consistent with autosomal dominant inheritance of cancer predisposition are important to identify because they indicate high risk and possibly the presence of a cancer-predisposing mutation. Factors that suggest a hereditary colorectal cancer predisposition syndrome include [111,113]:

Oligopolyposis (i.e., polyp count greater than expected) can involve as few as 10 to 15 polyps, and the diverse pathology of polyps requires careful attention to polyp count and histology to determine whether genetic testing and/or further clinical evaluation is appropriate [111].

Genetic Testing

As discussed, many genes associated with inherited colorectal cancer syndromes have been identified, and genetic testing is available for diagnosis and is the accepted standard of clinical care. Genetic testing of asymptomatic persons without colorectal cancer symptoms or precursors (adenomatous polyps) is performed to identify increased probability of developing colorectal cancer. Positive findings should lead to diagnostic testing to investigate the presence of occult cancer, followed by treatment if cancer or precursors are found. The intent is to prevent the development of colorectal cancer or increase the likelihood of curative outcome afforded by early detection. Patients can also use this information for decisions related to family planning, work, or retirement.

Disease-causing mutations can be found in most families affected by one of the inherited syndromes, and once a mutation is found in an index case of the family, relatives can be tested for the presence or absence of that mutation with near-100% accuracy. Cancer screening and management is then based on the genetic testing results [117].

Clinical issues somewhat unique to genetic testing include genetic counseling and informed consent for genetic testing. Genetic screening for inherited colorectal cancer syndromes can be hampered by patient or proband resistance, but consent to testing is greatly improved with coordination between the pathologist, referring surgeon or oncologist, and a cancer genetics counselor [98,117].

Clinical criteria used to identify candidates for genetic testing to determine the presence of an inherited susceptibility to colorectal cancer include [90,96,103]:

Screening/Surveillance Recommendations for Hereditary Colorectal Cancer

Patients diagnosed with a hereditary colorectal cancer syndrome or with a highly suggestive family or personal history require a more intensive and frequent screening and surveillance protocol than patients with average risk because of their high risk for colorectal and extracolonic malignancies. Table 3 provides a summary of recommendations for patients with specific hereditary colorectal cancer syndromes [111]. For each hereditary colorectal cancer syndrome, the left column lists malignancies associated with the syndrome, and the corresponding right column describes screening or surveillance approach specific to the at-risk malignancy.

Patients with inflammatory bowel disease, which includes ulcerative colitis and Crohn disease, have an elevated risk of developing colorectal cancer. The extent that colorectal cancer risk is elevated depends on the extent and duration of disease, but earlier age at onset is not associated with greater risk. Older estimates of colorectal cancer risk in patients with ulcerative colitis indicated a 2% greater risk after 10 years, 7.7% to 8% after 20 years, and 15.8% to 18% after 30 years of disease [118]. More recent estimates are somewhat lower, the result of more widespread prescribing of chemoprotective aminosalicylates, earlier and more liberal use of colectomy for medically refractory disease, and higher rates of surveillance colonoscopy. Studies involving patients with either ulcerative colitis or Crohn disease have shown comparable risk in both diseases [118].

The extent of inflammatory bowel syndrome is defined as the point in time when histologically identified disease is most extensive. Most colorectal cancers develop in patients with pancolitis, and disease extent is a major risk factor for colorectal cancer in patients with inflammatory bowel syndrome [118]. Patients with left-sided disease (up to the splenic flexure) have an intermediate risk level, while proctitis, ulcerative proctosigmoiditis, and backwash ileitis have little to no influence on risk level. A family history of sporadic colorectal cancer in a first-degree relative doubles the risk of colorectal cancer, and risk increases nine-fold if the first-degree relative was younger than 50 years of age when first diagnosed with colorectal cancer.

The extent of macroscopic and histologic inflammation is associated with increased risk of colorectal cancer, which can develop in areas of endoscopically normal but histologically active colitis. Colorectal cancer can occur in areas where colitis has remitted or where histologic findings show inactive colitis such as crypt distortion in the absence of active inflammation. Lack of endoscopic inflammation at the time of neoplastic detection does not mean absence of inflammation in the area before neoplastic development, and risk of neoplasia is not increased in mucosa that has never been inflamed. Thus, histologic instead of macroscopic evidence of tissue changes from inflammatory bowel syndrome serves as a more accurate determinant for assessing colorectal cancer risk. In the context of surveillance, extent of disease should be defined histologically [118].

Practice recommendations for the diagnosis and treatment of colorectal cancer in inflammatory bowel syndrome patients were developed and published by the American Gastroenterology Association [118]. The guideline format presents a series of clinically relevant questions raised by an expert panel, followed by the response based on analysis of the published research.

Natural History of Dysplasia

Colorectal cancer in inflammatory bowel syndrome develops from dysplasia in most cases, and although imperfect, dysplasia is considered the best marker of colorectal cancer risk in inflammatory bowel syndrome. Predicting the natural history of dysplasia is more difficult, because dysplasia is present in 75% to 90% of patients with inflammatory bowel syndrome and colorectal cancer, but colorectal cancer can develop in the absence of previous history of dysplasia. Not all patients with low-grade dysplasia progress through a phase of detectable high-grade dysplasia before developing cancer. Importantly, interpretation of dysplasia in mucosal biopsy specimens is highly subject to observer subjectivity. Therefore, pathologists with particular expertise in gastrointestinal disorders should review all cases diagnosed as indefinite, low-grade dysplasia, or high-grade dysplasia.

Colectomy

Strong evidence indicates that patients with inflammatory bowel syndrome and a non-adenoma-like dysplasia-associated lesion or mass should receive a colectomy. Patients with inflammatory bowel syndrome and an adenoma-like dysplasia-associated lesion or mass, without evidence of flat dysplasia elsewhere in the colon, can be managed safely by polypectomy and continued surveillance.

There is also strong evidence that colectomy for flat high-grade dysplasia treats undiagnosed synchronous cancer and prevents metachronous cancer. However, current evidence is insufficient to assess the balance of benefits and harms.

Surveillance Colonoscopy

Surveillance colonoscopy is at least moderately effective in reducing colorectal cancer risk in patients with inflammatory bowel syndrome. It is recommended for patients with inflammatory bowel disease who are at an increased risk of colorectal cancer. Patients most likely to benefit are those with extensive ulcerative colitis or Crohn disease.

Surveillance colonoscopy in patients with inflammatory bowel syndrome should include extensive biopsies of all anatomic segments of colorectal mucosa. Definitive data are lacking to inform the optimal surveillance intervals, but one- to three-year intervals are suggested. Careful mucosa inspection and sufficient number of biopsy specimens should be obtained from all anatomic segments of the colon.

Newer Imaging Techniques

Chromoendoscopy is more sensitive in dysplasia detection than white-light endoscopy when used by endoscopists with expertise. However, the natural history of chromoendoscopically detected dysplasia is unknown. In addition, more research is needed to determine the utility of narrow band imaging and confocal endomicroscopy in detecting dysplasia.

Chemopreventive Agents

Ursodeoxycholic acid has demonstrated significant reductions in colorectal cancer in patients with ulcerative colitis who also have primary sclerosing cholangitis. Aminosalicylates are also considered chemopreventive against colorectal cancer. Oral or topical corticosteroids, while demonstrating antineoplastic effects in clinical trials, are associated with too many side effects for routine chemopreventive use. There is insufficient evidence to inform a recommendation for or against the use of azathioprine, 6-mercaptopurine, folic acid, calcium or multivitamin supplements, or statins.

With screening colonoscopy, a colonoscope (a thin tube with a light and video camera on one end connected to a display monitor) is inserted through the rectum and guided through the length of the colon for observation on the monitor screen. Instruments to remove polyps and obtain biopsy are inserted through the rectum as needed [131]. Colonoscopy allows direct visualization of the colonic mucosa, lesion biopsy, and polyp removal over the entire colon. The sensitivity and specificity for colorectal cancer and advanced adenomas are very high, and colonoscopy is the confirmatory test used with all other screening approaches when positive findings occur [120].

Potential Complications and Harms

Colonoscopy may fail to detect as many as 6% of colorectal malignancies, and the miss rate for adenomas smaller than 1 cm has ranged from 12% to 17% [132]. This is largely the result of high inter-operator variability in adenoma detection rate. Greater awareness of this hazard from inadequate colonoscopy performance has led to heightened emphasis on training and continuous quality assurance of endoscopists [120]. In addition, colonoscopy is an invasive procedure, requires an invasive bowel cleansing, is time-consuming and uncomfortable, and thus possesses several characteristics that negatively affect patient acceptance as a first-line screening test [120].

Clinically significant complications that require medical intervention are rare and include perforation, bleeding, and cardiovascular events. Complication rates may increase in older patients [133,134]. More than 85% of serious colonoscopy complications occur during polypectomy, and a study of 97,000 colonoscopies found polypectomy associated with a seven-fold increase in risk of bleeding or perforation [135]. Up to 33% of patients report one or more minor, transient gastrointestinal symptoms after colonoscopy, and a review of 12 studies involving 57,742 colorectal cancer screening colonoscopies in average-risk patients found the aggregate rate of serious complications was 2.8 per 1,000 procedures [134,136].

Recommendations to Optimize the Adequacy of Colonoscopy Bowel Preparation

The U.S. Multi-Society Task Force on Colorectal Cancer has published guidelines for adequate pre-colonoscopy bowel cleansing [137]. The goals of this consensus document are to provide expert, evidence-based recommendations for clinicians to optimize colonoscopy preparation quality and patient safety.

The adequacy of pre-procedure bowel cleansing merits special attention because this patient factor is strongly associated with colonoscopy success. Up to 20% to 25% of colonoscopies are attempted in patients with inadequate bowel preparation, leading to diminished adenoma detection rates, longer procedural time, lower cecal intubation rates, and increased electrocautery risk [138,139,140].

Patient risk factors for inadequate preparation include older age, male sex, higher BMI, history of inadequate preparation, history of constipation, and use of opioids or other constipating medications. Patients with complex past medical histories or current conditions, including previous gastric or colonic resection, spinal cord injury, Parkinson disease, and stroke, are generally more difficult to prepare adequately. Diabetes is associated with the highest prevalence of inadequate bowel preparation [137].

A preliminary assessment of preparation quality should be done in the recto-sigmoid colon. If the indication is screening or surveillance and the preparation is clearly inadequate for polyp detection greater than 5 mm, terminate and reschedule the procedure or attempt an additional bowel cleansing approach without canceling the procedure that day. If the colonoscopy is complete to cecum, and the preparation ultimately is deemed inadequate, the examination should be repeated, generally within one year; intervals shorter than one year are indicated when advanced neoplasia is detected and there is inadequate preparation.

Adequacy of bowel preparation should be assessed after completing appropriate efforts to clear residual bowel debris. The rate of adequate preparation should be routinely recorded, and adequate patient preparation should be achieved in at least 85% of all examinations per physician [137].

Split-dose bowel-cleansing regimens are strongly recommended for screening colonoscopy. A same-day regimen is an acceptable alternative to split dosing, especially for patients undergoing afternoon examination. The second dose of split preparation should ideally begin four to six hours before the time of colonoscopy, with completion of the last dose at least two hours before the procedure time. With split-dose bowel-cleansing regimens, diet recommendations include low-residue or full liquids until evening on the day before colonoscopy.

Healthcare professionals should give oral and written patient instructions for all components of colonoscopy preparation and emphasize the importance of compliance. The physician performing the colonoscopy should ensure that appropriate support and process measures are in place for patients to achieve adequate colonoscopy preparation quality.

Selection of a bowel-cleansing regimen should consider patient's medical history, medications, and, when available, previously reported bowel preparation adequacy. A split-dose regimen of a 4-L polyethylene glycol electrolyte lavage solution (PEG-ELS)-based cleansing agent provides high-quality bowel cleansing. In healthy, non-constipated individuals, a 4-L PEG-ELS formulation produces a bowel-cleansing quality comparable to lower-volume PEG formulations.

Over-the-counter bowel cleansing agents have variable efficacy depending on the agent, dose, timing of administration, and whether used alone or in combination. Regardless of the agent, efficacy and tolerability are enhanced with a split-dose regimen. Although over-the-counter purgatives are generally safe, caution is required in certain populations, such as strictly avoiding magnesium-based preparations in patients with chronic kidney disease. Routine use of adjunctive agents for bowel cleansing before colonoscopy is not recommended.

Split-dose bowel cleansing is associated with greater willingness to repeat the regimen compared with day-before regimens. In addition, low-volume bowel cleansing agents are associated with greater compliance in repeat colonoscopies.

There is insufficient evidence to recommend specific bowel preparation regimens for children, adolescents, and elderly persons, but sodium phosphate preparations should be avoided in the elderly, in children younger than 12 years of age, and in those with risk factors for complications from this medication, including known or suspected inflammatory bowel disease.

Additional bowel purgatives should be considered in patients with risk factors for inadequate preparation. Low-volume preparations or extended time delivery for high-volume preparations are recommended for patients after bariatric surgery. Tap water enemas should be used to prepare the colon for sigmoidoscopy in pregnant women. There is insufficient evidence to recommend specific regimens for persons with a history of spinal cord injury; additional bowel purgatives should be considered.

There is also insufficient evidence to recommend a single salvage strategy for patients whose poor preparation precludes effective colonoscopy completion. In these cases, large-volume enemas may be attempted in patients who present for colonoscopy and report brown effluent despite compliance with the colon-cleansing regimen. Through-the-scope enema with completion of colonoscopy the same day may also be considered, especially for patients receiving propofol sedation. Waking the patient from sedation and continuing with further oral ingestion of cathartic with same-day or next-day colonoscopy is associated with better outcomes than delayed colonoscopy.

Quality Indicators for Colonoscopy Performance

In 2015, an estimated 11 million outpatient colonoscopies were performed in the United States [141]. In addition to patient bowel preparation, optimal colonoscopy efficacy depends on operator performance. Inadequate colonoscopy performance demonstrably worsens the ability to prevent colorectal cancer diagnoses and deaths, and practice recommendations have been developed to better ensure quality colonoscopy performance [142].

Cecal Intubation. Cecal intubation involves advancing the colonoscope beyond the ileocecal valve, allowing the colonoscopist to visualize the medial wall of the cecum between the ileocecal valve and the appendiceal orifice. Cecal intubation is essential for optimal colonoscopy because many colorectal neoplasms are harbored in the proximal colon, including the cecum, and low cecal intubation rates are linked to higher rates of interval proximal colon cancer [143]. Colonoscopists should be able to intubate the cecum in ≥95% of screening colonoscopies in healthy adults. Photography of the cecum is mandated to verify intubation [142].

Adenoma Detection. Missed adenoma detection is strongly associated with failure to prevent colorectal cancer during multi-year follow-up colonoscopy trials, and most interval colorectal cancers are due to missed lesions and incomplete polypectomy. The marked variation in colonoscopist adenoma detection rates within practice groups, and the essential role of adenoma detection in colorectal cancer prevention led to adenoma detection as a performance target [144,145,146]. The examination is considered adequate if detection of polyps >5 mm is unimpeded.

In screening colonoscopies of asymptomatic, average-risk persons, a minimum adenoma detection target rate of 25% is recommended. Adenoma detection rates of less than 25% indicate that performance improvement steps should be initiated. Adenoma detection rate is considered the primary measure of mucosal inspection quality and is the single most important quality measure in colonoscopy. Colonoscopists with high adenoma detection rates clear colons better, and patients with precancerous lesions are brought back earlier for their next colonoscopy. Colonoscopists with low adenoma detection rates fail to identify patients with precancerous lesions and multiple lesions, placing these patients at elevated risk for cancer from inappropriately long intervals between colonoscopy [142].

Withdrawal Time. The time taken to remove the colonoscope after cecum intubation (excluding time for biopsies or polypectomy) is termed withdrawal time, and colonic mucosa should be carefully examined for polyps during scope withdrawal. The recommended colonoscope withdrawal time should be at least six minutes in colorectal cancer screening of patients without previous bowel surgery (when no biopsies or polypectomies are performed) [142]. Numerous studies have demonstrated increased detection of significant neoplastic lesions in colonoscopic examinations with an average withdrawal time of at least six minutes, and longer withdrawal time is associated with higher detection rates [147,148,149].

Correction of Poor Performance. The objective for measuring quality indicators is to improve patient care by identifying poor performers for retraining or removal of their privileges to perform colonoscopy if performance cannot be improved. Most quality indicators are amenable to improvement. An exception may be withdrawal time; despite overwhelming evidence that withdrawal time is positively associated with detection, imposing longer withdrawal times on colonoscopists has not been found effective [142].

CT colonography, also termed virtual colonoscopy, involves examination of computer-generated colorectal images constructed from abdominal CT imaging that simulate a conventional colonoscopy. Pre-procedure laxatives are required to clean the colon, and the colon is insufflated with air just prior to the CT examination, which may be uncomfortable [150]. The risk of complications is extremely low because the test is non-invasive. CT colonography is now in use to perform screening and diagnostic imaging in patients with incomplete colonoscopy or for whom colonoscopy is contraindicated. Randomized trials are in progress comparing CT colonography with immunochemical FOBT (iFOBT) and colonoscopy, and should produce valuable information concerning patient acceptance, diagnostic yield, and costs [120,151].

Potential Complications and Harms

Specificity for polyp detection is consistently high with CT colonography, but the broadly variable sensitivity requires confirmatory colonoscopy for findings suggestive of colorectal cancer. Another disadvantage with CT colonography is the inability to remove polyps [152]. Extracolonic abnormalities are common in CT colonography, most commonly renal, splenic, uterine, hepatic, ovarian, pancreatic, and gallbladder abnormalities. Very little information is available on the clinical value of their detection or the impact on patient anxiety and psychologic function [153,154]. One study found CT colonography to be a useful diagnostic tool in patients who previously underwent incomplete optical colonoscopy [155].

Flexible sigmoidoscopy involves anal insertion of a sigmoidoscope (similar to the colonoscope) to visualize the rectum and sigmoid colon—the lower one-third of the colon. The scope inflates the large bowel with air to improve imaging, and polyp removal or biopsy may be performed during the procedure [156]. A 60-cm flexible sigmoidoscope was introduced decades ago that is more tolerable to patients than the older, rigid sigmoidoscope. It allows a more complete distal colon examination and can discover up to 65% of polyps, compared with 25% using the older instrument [157].

Potential Complications and Harms

Sigmoidoscopy can be an uncomfortable or painful procedure. Women may have more pain during the procedure, which may discourage them from returning for future screening sigmoidoscopies. Sigmoidoscopy can also cause perforation of the colon, bleeding, severe abdominal pain, and death, although this is rare [84,156]. Bleeding and perforation are the most common complications. Most cases of bleeding occur in patients who have polyps removed [156].

Double-contrast barium enema (DCBE) consists of the patient receiving an enema with a barium solution. Air is then pumped into the colon, and a series of x-rays are performed to image the entire colon and rectum [158].

Potential Complications and Harms

DCBE is no longer recommended as an alternative test for colorectal cancer screening, and its use has declined dramatically. DCBE effectiveness for polyp detection is substantially lower than that of colonoscopy and CT colonography [115].

In FOBT testing, the patient collects stool samples that are analyzed for presence of blood. Different FOBT tests involve different collection approaches but commonly require collection of consecutive stool specimens for up to three days. The first FOBTs to enter clinical use were guaiac-based (gFOBT); more recent versions employ immunochemical tests (iFOBT) or markers of DNA mutation (stool DNA tests or sDNA) [1].

Colorectal lesions and adenomatous polyps tend to bleed, and the resulting presence of hemoglobin in stool that is detectable even with intermittent or minimal bleeding formed the basis for gFOBT use in colorectal cancer screening. Hemoglobin is used as a biomarker for detecting blood in stool with guaiac, which identifies peroxidase-like activity that characterizes hemoglobin. However, gFOBT cannot discriminate human from nonhuman or intact from partially digested hemoglobin and is being phased out of clinical use. This results in detection of blood from ingested meat and upper airway and gastrointestinal bleeding as well as colorectal lesions. The low specificity of gFOBT requires confirmatory colonoscopy to validate positive findings [159].

iFOBT was developed to detect intact human hemoglobin originating from colorectal tissue. Unlike gFOBT, it does not detect hemoglobin from nonhuman dietary sources or partly digested human hemoglobin originating from the upper respiratory or gastrointestinal tract [160]. The sDNA variation of FOBT incorporates markers of DNA mutation that detect molecular genetic changes associated with colorectal cancer gene mutations shed into the stool [161].

Potential Complications and Harms

The very low sensitivity of gFOBT leads to a high proportion of false-positive results when confirmed by colonoscopy or DCBE plus flexible sigmoidoscopy, which a systematic review of published clinical trials estimated at greater than 80% [162]. iFOBT is increasingly recognized as superior to gFOBT for sensitivity, accuracy, and compliance, and it shows greater ability in detecting advanced neoplasia. While iFOBT requires colonoscopy confirmation of positive results and cannot detect many precancerous polyps, higher participation in iFOBT than in colonoscopy screening may offset some of its comparative limitations [120].

DNA fecal testing is emerging as a potentially important addition to the stool-based tests for colorectal cancer screening. More research is needed to understand the role of sDNA testing in organized colorectal cancer screening and unaddressed factors, such as screening interval, patient adherence, and costs [120].

The American College of Physicians (ACP) published their practice recommendations for colorectal cancer screening based on the review and synthesis of guidelines for screening colorectal cancer produced by several other professional organizations. Several tests to detect adenomatous polyps and cancer were evaluated for colorectal cancer screening efficacy, including flexible sigmoidoscopy, colonoscopy, DCBE, and CT colonography. Tests to primarily detect cancer (e.g., gFOBT, iFOBT, and sDNA) were also assessed [84].

Screening Initiation

The ACP recommends that individualized assessment of colorectal cancer risk should be performed in all adults [84].

Average-risk patients should begin at 50 years of age with a stool-based test, flexible sigmoidoscopy, or optical colonoscopy [163]. High-risk patients should begin at 40 years of age or 10 years younger than age of colorectal cancer diagnosis in the youngest family member. Test selection should be based on the benefits and harms of the specific test, the availability of the test, and patient preference. Screening is not recommended in adults older than 75 years of age or with a life expectancy of less than 10 years [84,163].

Note: The National Cancer Institute disputes recommendations of initiating colorectal cancer screening 10 years before the age at diagnosis in the youngest family colorectal cancer case, stating that direct evidence or strong rational argument is absent for aggressive screening methods in patients with modest family history of colorectal cancer [1].

In response to rising rates of colorectal cancer among persons younger than 50 years of age, the U.S. Preventive Services Task Force (USPSTF) lowered its recommended age of initiation of screening to 45 years of age, though the strength of recommendation is slightly lower than for those 50 years of age and older [121].

Clinical Considerations and Best Practice Advice for Colorectal Cancer Screening

African American individuals with average risk should begin colorectal cancer screening at 40 years of age due to the higher colorectal cancer incidence and mortality rates. Healthcare professionals should consider patients' personal, cultural, and religious preferences in screening test selection. For example, annual FOBT is not a good strategy for patients who may be unwilling or unable to follow-up yearly. Some women prefer a female endoscopist, and colonoscopy by a male endoscopist should be recommended only after discussion and patient consent.

Recommended Colorectal Cancer Screening Intervals

The National Comprehensive Cancer Network recommends screening for persons at average risk for colorectal cancer begin at 45 years of age after available options have been discussed. Currently, recommended options include: colonoscopy every 10 years; annual high-sensitivity guaiac-based testing or fecal immunochemical test; multitarget-stool DNA-based testing (every 3 years); flexible sigmoidoscopy every 5 to 10 years; or CT colonography every 5 years.

(https://www.nccn.org/professionals/physician_gls/pdf/colorectal_screening.pdf Last Accessed: March 14, 2022)

Level of Evidence: 2a (Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate.)

Clinicians should select the screening test with the patient on the basis of a discussion of benefits, harms, costs, availability, frequency, and patient preferences. The ACP recommends that patients between 50 and 75 years of age with average risk should be screened [163]:

These recommended intervals, especially for colonoscopy, are based on the assumption of optimal patient preparation and operator performance in the initial screen, allowing removal and biopsy of all polyps and detection of any precancerous lesion. Inadequate colonoscopy performance and resultant failure to detect adenomas or precancerous lesions places the patient at much greater risk of developing colorectal cancer (referred to as interval colorectal cancer) and renders the recommended interval unsafe [132].

The timing of follow-up surveillance colonoscopy after initial colorectal cancer screening colonoscopy is an essential component of colorectal cancer prevention (Table 5). Adenomatous polyps are cancer precursor lesions and the most common neoplasm found during colorectal cancer screening. Their detection and removal reduces colorectal cancer incidence and mortality, but patients with adenomas have heightened risk of developing interval cancers (metachronous adenomas or colorectal cancer) within three to five years of colonoscopy and polypectomy [164].

The basis for recommended time intervals between screening and surveillance colonoscopy should involve evidence that examinations prevent interval cancers and cancer-related mortality. Interval diagnosis of advanced adenomas has been used as a surrogate marker for colorectal cancer incidence or mortality. The U.S. Multi-Society Task Force guidelines for post-polypectomy surveillance in average-risk patients emphasize use of baseline colonoscopy findings for risk stratification, which is clustered into two groups [165]:

The British Society of Gastroenterology surveillance guidelines categorizes patients into three risk groups [166]:

Surveillance at one year was recommended for high-risk patients over concerns of missed lesions at baseline, differing from U.S. guideline emphasis (and assumption) of high-quality baseline examination [164]. This update of surveillance recommendations was developed to address emerging issues in post-colonoscopy surveillance [164].

Limitations of Colonoscopic Surveillance

As discussed, interval colorectal cancers are advanced adenomas that develop after polypectomy or negative baseline colonoscopy and before the next screening colonoscopy, a 10-year period for most patients. Within five years of negative screening colonoscopy, the risk of developing advanced adenomas is 1.3% to 2.4%. The greatest risk of interval colorectal cancer is within five years of screening colonoscopy, usually resulting from missed lesions progressing to diagnosable colorectal cancer [167].

Studies suggest that most interval colorectal cancers result from missed lesions during baseline colonoscopy. Failure to detect lesions is directly associated with colonoscopy examination quality [146,168]. Residual neoplastic tissue from incomplete adenoma removal can also progress to malignancy. Interval colorectal cancers may differ from prevalent colorectal cancers by more frequent location in the proximal colon and by molecular/genetic properties that confer more aggressive growth. The relationship is established between inadequate colonoscopy quality and risk of interval cancer following colonoscopy.

Halting Surveillance

Colonoscopy risks increase with advancing age and at some point outweigh the benefits of surveillance and screening. The USPSTF recommends clinicians selectively offer screening for colorectal cancer in adults 76 to 85 years of age [121]. Evidence indicates that the net benefit of screening all persons in this age group is small. In determining whether this service is appropriate in individual cases, patients and clinicians should consider the patient's overall health, prior screening history, and preferences. Patients with high-risk adenoma may especially benefit from continued surveillance.

Data from more than 180,000 patients with colorectal cancer were entered into the Surveillance, Epidemiology, and End Results (SEER) cancer database from 1975–2015 and analyzed [10]. Histologic subtypes in the population were overwhelmingly adenocarcinoma (92.1%); others included neuroendocrine carcinoid (4.4%), unspecified carcinoma (0.8%), and squamous cell (0.7%). The relative five-year survival rates were highest for carcinoid tumors (90.1%) and lowest for neuroendocrine tumors (14.4%) [10]. The SEER cancer database for 1975–2018 does not include information on histologic subtypes of colorectal cancer [172].

Colorectal lesions present as a broad spectrum of neoplasms that range from benign growths to invasive tumors. Most colorectal cancers develop slowly over years, typically beginning as non-cancerous polyps on the inner lining of the colon or rectum. Some, but not all, polyps develop into cancer, and the risk of malignant progression is influenced by polyp type. Colorectal lesions are classed into three groups [173]:

Adenomas are the primary precursor lesion of colorectal cancer. These polyps are benign tumors that may transform into malignancy. Of all patients with adenomatous polyps discovered by screening colonoscopy, one-year follow-up colonoscopy reveals additional polyps in 29%. The risk of colorectal malignancy in patients with history of polyp removal is 2.7 to 7.7 times that of the general population [174,175].

Epithelial-derived adenoma or adenocarcinoma tumors represent the predominant colorectal cancer tumor type. More than 95% of colorectal cancers are carcinomas, and more than 95% of these adenocarcinomas. Other histologic types account for the remaining 2% to 5%. Adenomas are histologically classified, by order of increasing malignant potential, as tubular, tubulovillous, or villous adenomas. Characteristics of adenomas that highly predict malignant transformation include [1,176]:

Adenomas may reflect an innate or acquired tendency of the colon to form tumors. Benign and malignant tissue occurs within colorectal tumors, and 20-year follow-up of patients with adenomas has found a 25% malignancy rate in adenoma sites. Removal of adenomatous polyps is linked with reduced colorectal cancer incidence and represents the foundation of primary colorectal cancer prevention [170].

The transition from normal epithelium to adenoma to carcinoma is associated with acquired molecular events. The mucosa in the large intestine regenerates roughly every six days. Crypt cells migrate from the base of the crypt to the surface, where they undergo differentiation and maturation and ultimately lose the ability to replicate. As noted, most colorectal carcinomas are adenocarcinomas. Adenomas precede adenocarcinomas, with roughly 10% of adenomas eventually developing into adenocarcinomas during a process that occurs over up to 8 to 10 years with sporadic colorectal cancers. Dysplastic adenomas progress to colorectal malignancies through a multistep process involving inactivation of a variety of tumor-suppressor and DNA-repair genes and simultaneous activation of oncogenes. Colonic epithelial cells are selectively vulnerable to the transformation from normal colonic epithelium to adenomatous polyp to invasive carcinoma [171,177,178].

CIN is the most common form of genomic instability and is found in as many as 85% of colorectal cancers. The hallmark of the CIN phenotype is mutations that inactivate the APC gene, found in up to 70% of sporadic colorectal cancers, and that cause FAP. APC mutations occur during the earliest stages of neoplasia and are predominantly associated with the classic tubular adenoma pathway and CIN tumor [169]. Increasing size, increasing number, and worsening histology of polyps reflect the linear process of carcinogenesis along the CIN pathway [179].

As discussed, the APC gene is a tumor-suppressor gene that indirectly regulates the transcription of several critical cell proliferation genes by encoding transcription factor beta-catenin, a protein involved in cell adhesion, signal transduction, transcription regulation, cell cycle control, apoptosis, and maintenance of chromosomal segregation fidelity. APC inactivation produces loss of beta-catenin function, allowing unchecked cellular replication at the crypt surface, and activation of oncogenes c-myc and cyclin D1 that drive the progression to malignant phenotype [103,171,180].

MSI tumors are characterized by MMR system defects. DNA MMR genes correct nucleotide base miss-pairs and small insertions or deletions that occur during DNA replication. The MMR defect promotes adenoma development and accelerates the progression from adenoma to carcinoma. These colorectal malignancies are distinguished at the molecular level by alterations in repeating units of DNA that occur normally throughout the genome, termed DNA microsatellites. Microsatellite unstable tumors are generally considered mutually exclusive of CIN tumors [169,181].

The mechanisms that underlie MSI involve MMR gene inactivation by aberrant methylation or somatic mutation. Roughly 20% to 30% of colorectal cancers display a characteristic pattern of gene hypermethylation, termed the CpG island methylator phenotype (CIMP). Some CIMPs display MSI, and these account for roughly 90% of Lynch syndrome cases and 15% to 20% of sporadic colon and rectal cancers [106,171,182].

Transforming growth factor-beta signaling is a tumor-suppressor pathway in the colon. Deregulation in this pathway occurs by inactivating mutations in receptor genes, post-receptor signaling pathway genes, and transforming growth factor-beta superfamily members [169,181].

Functionally significant mutations in TGFBR2, a signaling receptor gene, are detected in up to 30% of all colorectal cancers. They are most common in MSI tumors but also occur in 15% of CIN tumors and are associated with transformation of late adenomas to malignancy.

Mutations of PI3K pathway genes occur in up to 40% of colorectal cancer cases and may promote the transition from adenoma to carcinoma. PTEN, a tumor suppressor gene that negatively regulates PI3K signaling, is mutated in up to 30% of MSI tumors and 9% of CIN tumors. The PI3K pathway is modulated by epidermal growth factor receptor (EGFR) signaling in part via KRAS activation [169,181].

The most clinically important oncogene in colorectal cancer, KRAS is a downstream effector of EGFR that signals (through BRAF) the activation of mitogen activated kinase (MAPK) pathways and promotion of cell growth and survival. KRAS mutations occur in roughly 40% of colorectal cancers, primarily in CIN tumors secondary to inactivating APC mutations [169,181].

Mutated in roughly 10% to 15% of colorectal cancers, BRAF encodes a protein kinase that acts as the downstream effector of KRAS in the RAS/RAF/MAPK signaling pathway. KRAS and BRAF mutations are mutually exclusive; activating mutation in either gene is sufficient to promote tumorigenesis via increased MAPK signaling. BRAF mutations are more frequent in MSI tumors (35%) than CIN tumors (5%) [169,181].

Patient history and physical examination are the foundations of assessment. A thorough disease history should be obtained by eliciting disease-specific symptoms, associated symptoms, and family history. A cancer-specific history helps direct the exploration of associated pathology or metastatic disease and any further workup. When possible, all patients should undergo a full colonic evaluation with histologic assessment of the colorectal lesion before treatment. Patients should also be assessed for their fitness to undergo surgery, including assessment of cardiac risk, and preoperative radiologic staging should be routinely performed [187,188].

The incidence of colorectal cancer increases with age. Patients younger than 44 years of age account for fewer than 5% of cases, and the mean age at diagnosis is 71 years. Men and women older than 50 years of age have similar rates of colorectal cancer. However, the colorectal cancer prevalence in men increases in tandem with age beyond 50 years [97].

With increasingly widespread and effective screening, colorectal cancer is frequently detected at an earlier, asymptomatic phase. Physical examination findings early in the disease course can be normal or nonspecific (e.g., fatigue, weight loss) [114,187]. With more advanced colon cancer, common clinical presentations include iron-deficiency anemia, rectal bleeding, abdominal pain and tenderness, change in bowel habits, intestinal obstruction or perforation, hepatomegaly, and ascites. Right-sided lesions are more likely to bleed and cause diarrhea, while left-sided tumors are usually detected later and may present as bowel obstruction [114,187].

In addition to these signs and symptoms in colon cancer, physical examination of patients with rectal cancer may reveal a palpable mass and bright red blood in the rectum. Adenopathy, hepatomegaly, or pulmonary signs may be present with metastatic rectal cancer. Proctosigmoidoscopy and digital rectal examination should be performed to determine tumor distance from the anal verge, mobility, and position relative to the sphincter complex.

Healthcare professionals should be attentive to both common and uncommon signs and symptoms during the history and physical exam that suggest colorectal cancer. More common diagnostic factors include increasing age, rectal bleeding, rectal mass, change in bowel habits, family history, abdominal mass or distension, and anemia [171,189,190,191].

Rectal Bleeding

Although patients presenting with rectal bleeding may have a benign condition, this is a common symptom in patients with colon and rectal cancer. A primary care study found a positive correlation between each new episode of rectal bleeding in patients older than 45 years of age and colorectal cancer [191].

Change in Bowel Habit

Especially with rectal bleeding present, an increased frequency or looser stools is common in left-sided colorectal cancer. Bowel habit changes with reduced frequency and hard stools have low predictive value for colorectal cancer.

Rectal Mass

Palpable rectal mass is present in 40% to 80% of patients with rectal cancer [192]. Assessment using digital rectal examination is useful to estimate tumor proximity to the sphincter but unreliable to determine tumor involvement of the pelvic wall and suitability for surgery. These latter investigations are more accurately assessed by magnetic resonance imaging (MRI) and transrectal endoscopic ultrasound.

Positive Family History

Although only 10% to 20% of patients with colorectal cancer have a positive family history of colorectal cancer, persons with one affected first-degree relative are more than twice as likely to develop colorectal cancer, while those with two affected first-degree relatives are four times more likely to develop colorectal cancer [87,88].

Abdominal Changes

The abdominal examination is typically unremarkable in patients with colorectal cancer, but the presence of a palpable tumor mass is common in advanced disease. Presence of abdominal distension indicates ascites or intestinal obstruction secondary to advanced disease. Patients are unlikely to have colorectal cancer when abdominal pain is present in the absence of other gastrointestinal symptoms, but those with colorectal cancer often have abdominal pain in addition to other symptoms.

Anemia

Anemia is present in close to 90% of patients with right-sided colon cancer at the time of diagnosis [190].

Other Signs and Symptoms

Weight loss and anorexia are more associated with advanced disease, as are palpable lymph nodes.

Patients with suspected colorectal cancer require a complete colon examination, and this is best performed with colonoscopy [193,194]. Flexible sigmoidoscopy may be appropriate for low-risk patients, such as those with isolated rectal bleeding or who are younger than 50 years of age. However, positive findings with flexible sigmoidoscopy require pre- or postoperative confirmation and additional visualization of the entire colon, because roughly 5% of patients also harbor synchronous tumors [171,195].

In the absence of intestinal obstruction contraindicating the administration of bowel preparation, colonoscopy is the first-line investigational choice because it demonstrates the highest sensitivity for colorectal cancer of any diagnostic modality, lacks the radiation exposure of CT, and enables the removal of incidental polyps and biopsy of suspicious lesions. The disadvantages of colonoscopy include a false-negative rate of 2% to 6% and accuracy that is highly operator-dependent and strongly influenced by patient adherence to proper preparatory bowel cleansing. Tumor localization is improved with administration of intraluminal ink or tattooing of the suspected cancer site [132,195].

CT colonography sensitivity in colorectal cancer detection is comparable to optical colonoscopy and has been used following incomplete colonoscopy assessment. DCBE has also been used in cases of poor colonoscopy visualization of the sigmoid colon (e.g., with severe diverticular disease), usually combined with flexible sigmoidoscopy. However, the superior sensitivity and specificity of CT colonography have led to the phasing out of DCBE for these indications [193,194].

Elderly or frail patients may have difficulties with immobility or an inability to tolerate bowel preparation, which can impede conventional colonoscopy. One alternative is colorectal imaging using plain CT scan. Plain abdominal CT scan with oral contrast (but without bowel preparation) of symptomatic patients has shown an 88% to 94% sensitivity for colon cancer detection at 12- to 30-month follow-up [196,197].

Serum concentrations of carcinoembryonic antigen (CEA) are elevated in about 80% of patients with colorectal cancer, but CEA lacks sufficient sensitivity or specificity for use in screening or diagnosis. Instead, its greatest value comes from detecting colorectal cancer recurrence in patients who have undergone surgical resection. Patients should have baseline CEA values measured for comparison during the surveillance period to monitor for signs of recurrence [188].

Routine complete blood count, liver biochemistry, bone mineral density profile, and renal function are recommended before treatment to establish patient baseline values, to assess for hepatic and renal metastases, and to identify anemia [188].

During the diagnostic workup, other conditions with similarity to colon or rectal cancer should be considered and ruled out. These include [114,171]:

Rare gastrointestinal tumors should also be ruled out, such as:

After colorectal cancer is diagnosed, additional imaging is required for disease staging. Liver and chest imaging, preferably using CT, is necessary to detect metastases. Rectal cancers should be staged using endorectal ultrasonography or MRI. Positron emission tomography (PET) imaging is increasingly used in colorectal cancer to detect extrahepatic metastases in patients considered for hepatic resection of presumed liver-only metastatic disease. PET is also used to localize disease in patients thought to have a recurrence, as reflected by emergent symptoms or rising CEA [171,198,199].

Practice guideline recommendations for imaging to stage colorectal cancer have been published by the American Society of Colon and Rectal Surgeons (ASCRS) and by Cancer Care Ontario [188,200]. They recommend contrast-enhanced CT of the chest, abdomen, and pelvis should be performed in all patients with colon cancer (unless contraindicated) to estimate disease stage and identify metastases. If local excision is considered for low rectal cancer (0–5 cm from the anal verge), transrectal ultrasonography is preferred over MRI to improve discrimination between T1 and T2 lesions. For upper rectal cancers (10–15 cm above the anal verge), whereby the mesorectal fascia is not threatened, MRI is not considered superior to pelvic CT.

MRI can stage the local rectum but is not adequate to assess regional disease at the level of the inferior mesenteric artery or distant disease. CT of the abdomen and pelvis should be used to assess for distant metastases and regional disease, including lymph node involvement along the inferior mesenteric artery. Pelvic CT and/or transrectal ultrasonography are recommended with contraindications to MRI. All patients with rectal cancer should have preoperative radiologic staging with contrast-enhanced CT to assess for metastatic disease [188,200].

Histologic confirmation of colon cancer is ideal, and for rectal cancer, it is essential [171]. Research has demonstrated an association between the number of lymph nodes examined in colon and rectal cancer surgery and oncologic outcomes [201]. In patients with colon or rectal cancer, the American Joint Committee on Cancer (AJCC) and National Cancer Institute jointly recommend examination of a minimum of 12 lymph nodes to rule out regional lymphatic node involvement [202].

The AJCC has developed the TNM classification system, and this approach is the universal standard in clinical cancer care [202]. The AJCC TNM classification system is identical for colon and rectal cancer. The 2018 update to the AJCC system uses the pathologic stage (also called the surgical stage), as this is likely to be more accurate than the clinical stage, which takes into account the results of the physical exam, biopsies, and imaging tests done prior to surgery (Table 6) [202,203,204]. The system was initially developed as a prognostic tool. While numerous studies have evaluated other clinical, pathologic, and molecular parameters for validity in outcome prediction, none have been validated in multi-institutional prospective trials, and the TNM system remains the only prognostic tool validated in multi-institutional prospective studies. With TNM [203]:

When the T, N, and M categories have been determined (usually after surgery), the information is combined for stage grouping, with stage I the least advanced and stage IV the most advanced (Table 7) [127,187,188].

In rectal cancer, AJCC staging does not apply to the following malignant histologies [205]:

Patient prognosis is most powerfully associated with clinical and histopathologic stage of colorectal cancer at diagnosis as reflected by the TNM classification and staging. The National Cancer Institute SEER database tracks five-year relative survival rates for colon and rectal cancer, based on how far the cancer has spread; it does not group cancers by AJCC TNM stages. Instead, it groups cancers into localized, regional, and distant stages (Table 8) [13].

Several other factors have shown prognostic significance, including the number of harvested and processed lymph nodes, histologic grade, and evidence of lymphovascular and perineural invasion. In patients with metastatic colorectal cancer, the level of circulating tumor cells measured at baseline after the initiation of new therapy was an independent predictor of survival. In patients with baseline CEA values ≥25 ng/mL, those with low baseline levels of circulating tumor cells (fewer than three) had longer survival, and measurements of both circulating tumor cell number and CEA level at 6 to 12 weeks independently predicted survival [215]. Additionally, an emerging focus in research and literature is the role of host immune-centered factors (e.g., anti-tumor cells in the liver) in the clinical outcomes of colorectal liver metastases [216,217].

Patients with chemotherapy-refractory colorectal cancer who initially respond and then become resistant to cetuximab or other monoclonal antibodies have essentially run out of therapeutic options. This emergence of secondary drug resistance within 9 to 18 months of initiation is a major limitation of anti-EGFR therapies. A substantial proportion of patients with colorectal cancer who initially respond to anti-EGFR therapies have, at the time of disease progression, tumors with focal amplification or somatic mutations in KRAS that were undetectable before initiation of anti-EGFR therapy. Drug-resistant KRAS alteration results from pre-existent KRAS mutant and amplified clones and from new mutations arising from ongoing mutagenesis [227]. A mechanism by which KRAS mutation nullifies anti-EGFR therapy involves bypassing the need for upstream EGFR signals to activate downstream oncogenic processes [169,181]. It is now established that patients with any KRAS or NRAS mutation should not be treated with cetuximab or panitumumab, as these mutations strongly predict resistance to EGFR inhibitor agents. In contrast, non-mutational KRAS, termed wild-type KRAS, responds to targeted therapy [111,201,217,222,228,229].

Treatment of localized and locally advanced colon cancer primarily involves surgical resection, and roughly 80% of patients with colon cancer exhibit localized disease amenable to resection with curative intent [188]. Aside from palliative resection (e.g., alleviating obstruction), the objective of surgery is curative resection based on clear macroscopic and histologic resection margins. Practice recommendations from the ASCRS were published to optimize surgical care of these patients (Table 9) [188].

For resectable non-metastatic colon cancer, the National Comprehensive Cancer Network preferred surgical procedure is colectomy with en bloc removal of the regional lymph nodes.

(https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf Last Accessed: March 14, 2022)

Level of Evidence: 2a (Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate.)

The primary treatment for localized resectable colon cancer is colectomy with en bloc removal of all associated regional lymph nodes and involved adjacent structures. The extent of a curative resection for colon cancer depends on the site of the primary lesion and lymphovascular drainage of the cancer site. The length of bowel resected is governed by the blood supply to that segment. In the absence of synchronous pathology, an anatomic colon resection for cancer should achieve at least a 5-cm negative margin on either side of the tumor. Colectomy with local excision is not adequate for curative resection, because it increases risks of tumor spillage into the peritoneal cavity and tumor progression from lack of lymphadenectomy [188,220].

Surgery is curative in 25% to 40% of highly selected patients who develop resectable metastases in the liver and lung. Refinements in surgical technique and preoperative imaging have improved patient selection and resection outcomes [201,232,233,234].

Before surgery, all patients should be given information about the likelihood of having a stoma, why it might be necessary, and how long it might be needed. The psychologic and emotional impacts of having a stoma should not be overlooked. Between 16% and 26% of patients with a stoma will experience negative psychologic symptoms immediately postoperatively, including anxiety, depression, and suicidal ideation [235,236]. Having a stoma also can potentially decrease patients' quality of life as they experience changes to body image, sexual function, social isolation, stigma, embarrassment, and decreased mood [237]. A trained stoma professional should provide specific information on the care and management of stomas to all patients considering surgery that might result in a stoma [238].

Given that tumor depth, nodal metastasis, and distant metastasis strongly predict post-surgical prognosis in colon cancer, staging should be performed following surgical resection using TNM staging, histologic grade of the tumor, and resection completeness [188].

Stage II

The value of adjuvant chemotherapy for stage II colon cancer is controversial. In one study, adjuvant 5-FU-based chemotherapy was evaluated in patients with high-risk stage II colon cancer following curative resection. Compared with surgery alone, adjuvant 5-FU showed inconsistent benefit; these and other results led to guidelines issued by the American Society of Clinical Oncology (ASCO) stating that evidence does not support the routine use of adjuvant chemotherapy for patients with stage II colon cancer [239,240].

The NCCN guideline also states there is no survival advantage by adding oxaliplatin to 5-FU/leucovorin, including in patients 70 years of age or older [220]. The combination of folic acid, 5-FU, and oxaliplatin (FOLFOX) is considered reasonable in high-risk cases, but it is not indicated in good-to-average-risk stage II cancers.

Stage III

Stage III colon cancer denotes lymph node involvement. Studies have shown that prognosis is related to the number of involved lymph nodes; patients with one to three involved nodes have a significantly better survival than those with four or more involved nodes. Before 2000, 5-FU was the only adjuvant chemotherapy with activity in stage III colon cancer. With patients in many earlier trials of adjuvant 5-FU not showing a survival benefit, modifications and additions to the core 5-FU therapy were investigated in stage III colon cancer. More recently, capecitabine was established as comparable to 5-FU/leucovorin. The addition of oxaliplatin to 5-FU/leucovorin (FLOX) improved overall survival compared with 5-FU/leucovorin alone and has become the reference standard for the future generation of clinical trials for stage III colon cancer [201,220,241].

For stage II/III colon cancer, the NCCN asserts that adjuvant bevacizumab, cetuximab, panitumumab, or irinotecan should not be used outside of clinical trials [220]. In stage III colon cancer, FOLFOX is superior to 5-FU/leucovorin, and capecitabine/oxaliplatin (CAPEOX) is superior to bolus 5-FU/leucovorin. FLOX is an alternative to FOLFOX or CAPEOX, but FOLFOX or CAPEOX are preferred [220].

Unlike in rectal cancer, the role of adjuvant radiation therapy is poorly defined in colon cancer treatment. Radiation therapy has no current adjuvant role following curative resection but may have a potential role in patients with residual disease [220]. If used, radiation fields should include the tumor bed, as defined by preoperative radioimaging or surgical clips. Radiation should be given in doses of 45–50 Gy in 25 to 28 fractions; the dose in the small bowel should be no greater than 45 Gy [220]. Neoadjuvant chemoradiotherapy that includes 5-FU should be delivered concurrently to aid resectability. Conformal external beam radiation is preferred; intensity-modulated radiation therapy should be limited to unique clinical situations. Intraoperative radiation therapy should be considered in T4 or recurrent cancer [220].

Rectal cancer surgery involves surgical resection of the primary tumor. Surgical approach is guided by tumor location, disease stage, and presence of high-risk features (e.g., positive margins, lymphovascular invasion, perineural invasion, poorly differentiated histology) [205].

Polypectomy alone may be sufficient when polyps with invasive cancer can be completely resected with clear margins and show favorable histologic features, generally select T1 cancers [248]. Approaches with minimal morbidity and mortality include transanal excision and transanal endoscopic microsurgery. Local excision is appropriate in selected T1 tumors, with mesorectal excision preferred for all other T1–T2/N0 tumors. Endoscopic microsurgery cannot perform excision and staging of mesorectal lymph nodes, a limitation because T1 lesions have a 6% to 11% risk of harboring nodal metastasis [249]. Local recurrence rates range from 7% to 21% for T1 lesions and 26% to 47% for T2 lesions [249,250,251].

Total mesorectal excision with autonomic nerve preservation via low-anterior resection is preferred, followed by colorectal anastomosis in advanced mid- to upper-rectal tumor. Low anterior rectal resection is associated with bowel urgency, increased bowel frequency, clustering, and fecal incontinence from loss of rectum reservoir function. The colonic J-pouch is the superior approach for improving postoperative bowel function [59,252]. In patients unsuitable for sphincter-preservation, total mesorectal excision via abdominoperineal resection is preferred, although this leaves patients with a permanent colostomy [253,254,255]. Total mesorectal excision has demonstrated reproducible reductions in local recurrence and improvement in disease-free and overall survival [256].

Despite the low rate of local relapse after meticulous mesorectal excision, the heightened tendency for first failure to solely occur in locoregional sites requires the ongoing routine use of adjuvant radiation therapy [187,257].

Multimodality therapy has been the standard of care for patients with locally advanced rectal cancer since 1990, when the National Cancer Institute recommended adjuvant therapy for stage II and III disease [205]. This was based on findings of 33% to 55% reduction in local recurrence and significant prolongation in disease-free survival. Although the National Cancer Institute recommended adjuvant therapy, subsequent findings have shown superior efficacy, lower toxicity, and better long-term outcomes with neoadjuvant therapy. The Institute now recommends neoadjuvant therapy as the preferred treatment option for patients with stages II or III disease [205,258,259,260].

Preoperative radiation therapy is more effective because well-oxygenated tissue responds better to irradiation; postoperative tissue is relatively hypoxic from surgery and may be more resistant to radiation therapy. Also, postoperative complications may delay initiating adjuvant therapy [261].

Neoadjuvant Chemoradiation Therapy

As stated, neoadjuvant chemoradiation therapy is the preferred treatment option for patients with stage II or III disease, although adjuvant chemoradiation therapy remains an acceptable option [205]. Preoperative chemoradiation therapy is the standard of care for patients with clinically staged T3–T4 or node-positive disease (stages II/III) with benefits found in multiple trials, including [205]:

The most common neoadjuvant regimens for locally advanced tumors of the mid and lower third of the rectum are [261,262,263]:

When followed by proper surgical approach and execution, both regimens provide excellent local control for locally advanced tumors. Combined neoadjuvant radiation therapy and surgery may result in substantial long-term morbidity, including chronic bowel, sphincter, and sexual dysfunction, making careful selection of patients with greatest potential benefit from radiation therapy essential [264,265]. Neoadjuvant radiation therapy or chemoradiation therapy should not be used in low-risk operable rectal cancer [238].

Adjuvant Therapy

Compared with adjuvant chemoradiation therapy, preoperative chemoradiation therapy is preferred because it decreases local recurrence and adverse effects. However, the evidence demonstrates that compared to observation alone or radiation therapy alone following surgery, adjuvant chemoradiation therapy improves survival and reduces local recurrence rates in patients with resected stage II or III rectal cancer who have not received preoperative radiation therapy [230,258].

Many patients do not benefit from conventional 5-FU therapy, and introduction of newer chemotherapy regimens and biologic agents in colon cancer have prompted efforts to enhance survival benefits by optimizing radiation sensitization and chemotherapeutic selection and delivery. The NCCN now recommends m(modified)FOLFOX, CAPEOX, FOLFIRINOX, or mFOLFIRINOX as adjuvant chemotherapy in stage II/III rectal cancer. This comes with the caveat that conclusive data in rectal cancer are lacking, with recommendation for use in rectal cancer based solely on extrapolation of colon cancer data [230]. The merit of adding oxaliplatin to adjuvant 5-FU/leucovorin in stage II/III rectal cancer is the subject of ongoing debate due to issues with acute toxicity [205].

Radiotherapy Toxicity

The greater toxicity concerns with pelvic irradiation of rectal cancer involve potential late-onset morbidity. Relative to patients receiving surgical resection alone, those with additional radiation therapy treatment have shown increased risks of chronic bowel problems, sphincter dysfunction, sexual dysfunction, and elevated risk of surgical morbidity [258].

The improved local tumor control with neoadjuvant radiation therapy should be weighed against greater risks for acute toxicity (e.g., pelvic or perineal wound infection) and chronic/late-onset toxicity (e.g., stool frequency and incontinence problems, pelvic fractures, worsening sexual function). The frequency of these adverse effects found in patients receiving radiation therapy plus surgery versus surgery-only includes fecal incontinence in 62% vs. 38%, and urinary incontinence requiring pad wearing in 56% vs. 33%, respectively [258].

As discussed, 5-FU has been the foundation of chemotherapy for colorectal cancer for more than four decades. As a single agent, it inhibits tumor cell growth through at least three different mechanisms that ultimately disrupt cellular viability or DNA synthesis, transcription, and replication.

Capecitabine is an oral fluoropyrimidine that undergoes a three-step enzymatic conversion to 5-FU, with the last step occurring in the tumor cell.

Leucovorin is a reduced form of folic acid that does not require enzymatic reduction reaction for activation. This agent allows for purine and pyrimidine synthesis, both of which are needed for normal erythropoiesis. Leucovorin counteracts the toxic effects of current standard combination chemotherapy for colorectal cancer and potentiates the effects of 5-FU and its derivatives by stabilizing the binding of the drug's metabolite to its target enzyme to prolong drug activity.

Irinotecan is inactive in its parent form and is converted by the carboxylesterase enzyme to its active metabolite form SN-38, which is 1,000 times more potent than its parent compound. SN-38 binds to and stabilizes the topoisomerase I-DNA complex and prevents the relegation of DNA after it has been cleaved by topoisomerase I, inhibiting DNA replication. Irinotecan is a current standard therapy for metastatic colon cancer as the combination 5-FU/leucovorin/irinotecan.

A third-generation platinum-based antineoplastic agent, oxaliplatin is used in combination with 5-FU/leucovorin for metastatic colorectal cancer. As with other platinum compounds, oxaliplatin destroys tumor cells through interaction with DNA to form intra-strand/inter-strand DNA cross-linking that interferes with DNA base pairing, replication, and gene transcription, resulting in cell death [270].

Cetuximab is a partially humanized monoclonal antibody against EGFR that specifically binds to the extracellular domain of EGFRs. The cetuximab-bound EGFR inhibits activation of receptor-associated kinases, which inhibit cell growth, induce apoptosis, and decrease production of matrix metalloproteinase and VEGF. Cetuximab is indicated for the treatment of KRAS mutation-negative (wild-type), EGFR-expressing metastatic colorectal cancer. Importantly, patients with mutant KRAS tumors may experience worse outcome when cetuximab is added to multiagent chemotherapy regimens containing bevacizumab.

Bevacizumab is a partially humanized monoclonal antibody that binds to VEGF to inhibit angiogenesis. The inhibition of new blood vessel formation denies blood, oxygen, and other nutrients needed for tumor growth.

Panitumumab is a fully humanized antibody that binds to EGFR. It is approved by the FDA for use in chemotherapy-refractory metastatic colorectal cancer and is indicated for wild-type KRAS metastatic colorectal cancer.

Ziv-aflibercept is a novel anti-VEGF molecule that acts as a decoy receptor for VEGF-A, VEGF-B, and placental growth factor. The antiangiogenic mechanism of ziv-aflibercept involves competition with VEGF in the blood and extravascular space to prevent VEGF from interacting with its receptors on endothelial cells. It is indicated for metastatic colorectal cancer that is resistant to or has progressed after an oxaliplatin regimen [271].

Regorafenib inhibits multiple tyrosine kinase pathways, including VEGF, and was approved in 2012 for the treatment of metastatic colorectal cancer in patients previously treated with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy; an anti-VEGF therapy (e.g., bevacizumab, ziv-aflibercept); and, if KRAS wild type, an anti-EGFR therapy (e.g., cetuximab, panitumumab).

The basis of chemotherapy for the treatment of colon and rectal cancer is combination therapy, with agents identified to work synergistically to manage unresectable lesions and minimize drug resistance. These combinations are generally known by their acronyms (Table 11).

When 5-FU was the only available chemotherapeutic option with colorectal cancer activity, trials in patients with locally advanced, unresectable, or metastatic disease showed partial response, prolonged time-to-progression of disease, and improved survival and quality of life compared with best supportive care only. Several trials analyzing the activity and toxicity of various 5-FU/leucovorin regimens found comparable results and median survival of roughly 12 months [280,281,282].

Randomized studies found capecitabine equivalent in efficacy to the 5-FU/leucovorin regimen [283,284]. Other studies in metastatic colorectal cancer found non-inferiority between capecitabine/oxaliplatin (CAPOX) and 5-FU/oxaliplatin regimens as first-line therapy [285,286].

In patients with previously untreated metastatic colorectal cancer, adding irinotecan or oxaliplatin to 5-FU/leucovorin has led to improved treatment response, progression-free survival, and overall survival [287,288,289].

A comparison of FOLFOX4 against irinotecan, 5-FU, and leucovorin (IFL) showed progression-free survival of 8.7 vs. 6.9 months and overall survival of 19.5 vs. 15.0 months [290]. Comparisons of FOLFOX and FOLFIRI found identical progression-free survival and overall survival, although patients were allowed to cross over after progression. The toxicity profiles of the regimens differed [291,292].

Patients randomized to FOLFIRI, modified IFL (mIFL), or capecitabine/irinotecan (CAPIRI) showed progression-free survival of 7.6 vs. 5.9 months with FOLFIRI vs. mIFL, and 7.6 vs. 5.8 months with FOLFIRI vs. CAPIRI. CAPIRI also led to the highest rates of grade 3 or greater nausea, vomiting, diarrhea, dehydration, and hand-foot syndrome [293].

FOLFOX and FOLFIRI are first-line treatments for patients with metastatic colorectal cancer, with FOLFIRI preferred when using irinotecan [293].

CAPOX was found comparable to 5-FU and oxaliplatin as an oxaliplatin-based regimen for first-line treatment of metastatic colorectal cancer [285,286]. As second-line treatment following progression on irinotecan and 5-FU/leucovorin, patients randomized to FOLFOX4 or infusional 5-FU/leucovorin showed a median time-to-progression of 4.6 versus 2.7 months [294].

Bevacizumab is effective when added to FOLFIRI or FOLFOX as first-line treatment of metastatic colorectal cancer. In a 2009 study of patients with metastatic colorectal cancer, patients randomized to FOLFIRI/bevacizumab showed an overall survival of 28.0 months compared with 19.2 months with mIFL/bevacizumab [295]. In a separate study, patients randomized to IFL/bevacizumab or IFL/placebo showed progression-free survival of 10.6 vs. 6.2 months and overall survival of 20.3 vs. 15.6 months [296].

A trial randomized 1,401 patients with stage IV colorectal cancer to CAPOX or FOLFOX4, and then to bevacizumab or placebo. Patients receiving bevacizumab versus placebo showed progression-free survival of 9.4 vs. 8.0 months and overall survival of 21.3 vs. 19.9 months. Patients in the pooled CAPOX versus FOLFOX4 arms had a progression-free survival of 8.0 vs. 8.5 months. Overall survival had less benefit from bevacizumab than previously reported [297].

In another study, patients who progressed on FOLFIRI were randomized to FOLFOX plus bevacizumab or placebo, and showed a progression-free survival of 7.43 vs. 4.7 months, and overall survival of 12.9 vs. 10.8 months [298]. Based on these studies, bevacizumab was deemed a reasonable addition to FOLFIRI or FOLFOX as first-line treatment of metastatic colorectal cancer.

In a 2012 study, patients progressing on a first-line regimen that included bevacizumab were randomized to a different chemotherapy regimen plus continued bevacizumab or placebo. Participants who continued bevacizumab showed an overall survival of 11.2 months and progression-free survival of 5.7 months, compared with 9.8 months and 4.1 months, respectively, with placebo [299]. These results led to FDA approval of bevacizumab continuation in patients with progression during first-line chemotherapy, allowing patients to continue bevacizumab after switching to a different regimen containing irinotecan or oxaliplatin that may improve the synergistic activity [300].

FOLFOXIRI plus bevacizumab was compared to FOLFIRI plus bevacizumab in patients with untreated metastatic colorectal cancer, who showed a progression-free survival of 12.1 vs. 9.7 months and overall survival of 31.0 vs. 25.8 months. FOLFOXIRI led to significantly more grade 3/4 toxicities, including neutropenia, stomatitis, and peripheral neuropathy [301].

As second-line therapy, 1,226 patients with metastatic colorectal cancer randomized to FOLFIRI plus ziv-aflibercept or placebo showed overall survival of 13.50 vs. 12.06 months and progression-free survival of 6.90 vs. 4.67 months. Both statistically significant outcomes favored ziv-aflibercept, and FOLFIRI plus ziv-aflibercept is an accepted second-line regimen for patients previously treated with FOLFOX [302].

Tumors with KRAS mutations are cetuximab-insensitive, but adding cetuximab to multiagent chemotherapy improves survival in patients with colorectal cancers lacking KRAS mutation (i.e., KRAS wild type). As discussed, patients with mutant KRAS tumors may experience worse outcomes when cetuximab is combined with bevacizumab. These differences are evident in the clinical trial data.

Patients who progressed on irinotecan regimens randomized to cetuximab plus irinotecan or placebo showed a time-to-progression of 4.2 vs. 1.5 months [303]. A trial of 1,198 patients with stage IV colorectal cancer randomized to FOLFIRI plus cetuximab or placebo found improved progression-free survival but not overall survival with cetuximab. With emerging evidence that cetuximab response is limited to patients with wild-type KRAS tumors, the results were re-analyzed by KRAS status. A significant interactive effect was found for KRAS mutation status and cetuximab treatment response but not progression-free survival, with KRAS wild-type outcomes favoring FOLFIRI and cetuximab [304].

In a 2009 study, patients were randomized to capecitabine/oxaliplatin/bevacizumab plus cetuximab or placebo for metastatic colorectal cancer. The median progression-free survival was 9.4 vs. 10.7 months, and patients with KRAS gene mutation (versus wild-type) receiving cetuximab had progression-free survival of 8.1 vs. 10.5 months. Patients with KRAS tumor mutation receiving cetuximab (as opposed to placebo) showed progression-free survival of 8.1 vs. 12.5 months and overall survival of 17.2 vs. 24.9 months [295].

The benefit of adding cetuximab to first-line combination chemotherapy was studied in patients with KRAS wild-type tumors. The 1,630 patients were randomized into three treatment groups and cetuximab or placebo:

In patients receiving chemotherapy plus placebo versus cetuximab, the overall survival was 17.9 vs. 17.0 months and progression-free survival was 8.6 vs. 8.6 months. In patients treated continuously (arm A) versus intermittently (arm C), median survival was 15.8 vs. 14.4 months [305,306]. None of these findings were statistically significant.

In a separate study, patients with EGFR-expressing metastatic colorectal cancer were randomized to first-line FOLFOX-4 plus cetuximab or placebo. The participants did not differ in response rate or progression-free survival. However, in patients with KRAS wild-type tumors, the response rate was 61% vs. 37% and progression-free survival was 7.7 vs. 7.2 months. In contrast, patients with KRAS mutant tumors showed progression-free survival of 5.5 vs. 8.6 months [307].

Panitumumab is approved for use in patients with chemotherapy-refractory metastatic colorectal cancer. In clinical trials, panitumumab as single agent or combination therapy demonstrated improvements in progression-free survival and overall survival comparable to cetuximab [308,309,310].

The safety and efficacy of regorafenib was evaluated by a single clinical trial of 760 patients with previously treated metastatic colorectal cancer. Participants were randomized to regorafenib or placebo plus best supportive care and showed a median overall survival of 6.4 vs. 5.0 months [311].

Second-line chemotherapy with irinotecan in patients treated with 5-FU/leucovorin as first-line therapy led to improved overall survival versus infusional 5-FU or supportive care [312]. Conversely, patients who progressed on irinotecan and 5-FU/leucovorin and then received FOLFOX4 or 5-FU/leucovorin showed a median time-to-progression of 4.6 vs. 2.7 months [294].

In general, the imaging appearances of liver metastases are nonspecific, requiring biopsy specimens for histologic diagnosis. CT is the imaging modality of choice for evaluating hepatic metastases. CT permits better evaluation of the involvement of extrahepatic tissues, including the bones, bowel, lymph nodes, and mesentery. MRI may be superior to CT and PET scan for detection and characterization of small lesions [313].

According to the National Compre-hensive Cancer Network, hepatic resection is the treatment of choice for resectable liver metastases from colorectal cancer. Complete resection must be feasible based on anatomic grounds and the extent of disease; maintenance of adequate hepatic function is required.

(https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf Last Accessed: March 14, 2022)

Level of Evidence: 2a (Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate.)

Advances in chemotherapy have steadily improved survival in patients with colorectal cancer liver metastases, with trials now reporting a median survival of 20 months. However, with chemotherapy alone, five-year survival has been poor historically—less than 1%. This has been modestly improved in trials using FOLFOX and/or FOLFOXIRI, with five-year survival rates of 5% to 10% [290,314]. Despite advances in chemotherapy, liver resection is the best option for achieving long-term survival and may be curative in stage IV disease confined to the liver [315,316]. Resection of liver metastases with clear margins is associated with a 5-year survival rate of 45% and 10-year overall survival rate of 25% [274,317,318,319].

Hepatic metastases are considered suitable for resection based on the following criteria [201]:

Cancer Care Ontario recommends that patients with extra-hepatic metastases limited to the lungs may be suitable for liver resection if all pulmonary metastases are eradicated [317]. Studies of patients with combined liver and lung resection found three-year survival of 36% to 59%, and five-year survival of 9% to 74% [320]. The study showing 74% survival at five years calculated survival from the first metastasectomy instead of the more common second metastasectomy (usually the lungs). Median survival was 42 months when calculated from last metastasectomy [321]. Pooled data from all studies showed five-year survival of 30% [320]. Routine liver resection is not recommended in patients with portal nodal disease or non-pulmonary extra-hepatic metastases [317].

Liver resection is recommended in patients with initially unresectable liver metastases sufficiently downstaged by neoadjuvant chemotherapy [201]. If complete resection has been achieved, adjuvant chemotherapy should be used; neoadjuvant chemotherapy in patients without extra-hepatic metastases led to complete resection in 15% to 36%, and the five-year survival in these patients (33% to 42%) is similar to survival in patients with liver metastases considered resectable without chemotherapy [201,320]. Consensus is lacking on the best regimen to convert isolated liver metastases from unresectable to resectable [201].

Resection of all lesions, including those with radiographic complete response, is recommended when technically feasible and an adequate functional liver remnant can remain. When a lesion with radiographic complete response is present in an unresectable portion of the liver, surgery may still be an option if all other visible disease can be resected. Adjuvant chemotherapy should also be considered. Closely follow the lesion to allow localized treatment or further resection for in-situ recurrence [317].

Cancer Care Ontario recommends perioperative chemotherapy for patients with resectable liver metastases and extra-hepatic metastases amenable to resection with clear margins [317]. However, the role of adjuvant chemotherapy in potentially curative liver metastases resection is uncertain [201]. Before FOLFOX and FOLFIRI were introduced, two trials randomized patients after resection of liver metastases to 5-FU/leucovorin or observation. Both studies closed early due to poor accrual, but some data were obtained. Patients randomized to 5-FU/leucovorin or observation had five-year disease-free survival of 33.5% vs. 26.7% and overall survival of 51.1% vs. 41.1% [322]. In patients randomized to post-surgery 5-FU/leucovorin, the progression-free survival was 27.9 months compared with 18.8 months in the observation group [323].

Since the introduction of FOLFOX and FOLFIRI, multiagent chemotherapy has been evaluated as adjuvant therapy following resection of colorectal cancer liver metastases. In one study, patients randomized to 5-FU/leucovorin or FOLFIRI showed disease-free survival of 21.6 vs. 24.7 months; disease-free survival and overall survival were statistically comparable [324].

In another study, patients with up to four resectable liver metastases received perioperative FOLFOX (six cycles before and after surgery) or surgery alone. The progression-free survival was 42.4% vs. 36.2%. Reversible postoperative complications were more frequent after chemotherapy than surgery alone (25% vs. 16%), and there was one fatality after chemotherapy versus two fatalities after surgery [325].

Based on these findings, some physicians feel perioperative therapy is reasonable [201]. However, improved overall survival from resection plus chemotherapy has not been found.

Hepatic intra-arterial chemotherapy with floxuridine for liver metastases has shown higher overall response rates but no consistent improvement in survival compared with systemic chemotherapy [201]. In one trial, patients receiving curative liver resection were randomized to combined hepatic intra-arterial floxuridine and dexamethasone plus systemic 5-FU/leucovorin or to systemic 5-FU/leucovorin alone. Combined therapy improved two-year progression-free survival (57% vs. 42%) and overall survival (86% vs. 72%) but not median survival (72.2 vs. 59.3 months) [326].

A meta-analysis of randomized trials of fluoropyrimidine systemic therapy found no survival advantage. Furthermore, hepatic intra-arterial therapy is associated with increased local toxic effects, including liver function abnormalities and fatal biliary sclerosis [327].

Radiofrequency ablation (RFA) has emerged as a safe technique (2% major morbidity and less than 1% mortality rate) that may provide for long-term tumor control [328]. With RFA, high-frequency alternating current is delivered through needle electrodes inserted into the hepatic tumor area. The generated heat induces localized coagulative necrosis and tissue destruction. RFA is performed under imaging guidance, and the patient receives local or general anesthesia [329].

With hepatic colorectal cancer metastases, RFA is indicated as primary treatment in patients medically unfit for surgery; when the number, location, and size of metastases contraindicate resection; for treatment of post-resection recurrence; and as resection adjunct to ablate small-volume colonies in the future remnant liver. The National Institute for Health and Clinical Excellence (NICE) concluded in 2009 that RFA safety and efficacy evidence was sufficient to support its use in patients unfit or unsuitable for hepatic resection and in patients with previous hepatic resection [329].

Cryosurgical ablation is an option for patients with tumors that cannot be resected and for patients who are not candidates for liver resection [330,331]. Other local ablative techniques include embolization and interstitial radiation therapy [332]. Patients with limited pulmonary metastases, or with both pulmonary and hepatic metastases, may also be considered for surgical resection, with five-year survival possible in select patients [333].

Neutropenia, thrombocytopenia, and anemia may develop with the chemotherapeutic agents used in colorectal cancer treatment. Management of these short-term complications is temporary drug cessation and supportive treatment until recovery of bone marrow function [171].

Elevations in serum liver enzymes are common during treatment with oxaliplatin. Rarely, there is evidence of a hepatic veno-occlusive disease that presents with evidence of portal hypertension or persistent abnormalities in liver biochemistry [171].

Diarrhea, nausea, vomiting, and/or abdominal pain commonly occur with chemotherapeutic agents. Management is symptomatic, with loperamide for diarrhea, antiemetics for nausea and vomiting, and analgesia for pain [171].

Alopecia is a short-term adverse effect of certain chemotherapies. This effect will resolve with cessation of treatment, but in the interim, management is largely cosmetic.

Acneiform rash is very common in patients being treated with cetuximab. It primarily occurs on the face and upper torso, often improves with continued treatment, and is reversible. This complication is associated with improved chance of treatment response independent of KRAS status [171].

Loose stool, urgency, and fecal incontinence are common after radiation therapy for rectal cancer [171]. Patients should be prepared for this long-term complication.

Bladder dysfunction can result from damage to the pelvic nerves during rectal cancer surgery. Symptoms can include urinary urgency, incontinence, and retention. Urinary catheterization may be required to relieve retention [171].

Erectile dysfunction can also occur due to pelvic nerve damage. In one study of 28 men treated for colorectal cancer, 24 reported experiencing erectile dysfunction after treatment (i.e., chemotherapy, radiation, and/or surgery) [334]. Almost none of the men in the study received adequate care and education related to this complication.

Pulmonary fibrosis occurs in less than 1% of patients being treated for colorectal cancer [171]. This generally presents as dry cough, dyspnea, basal crepitations, and pulmonary infiltrates on chest x-ray or CT.

Neurotoxicity is a common adverse effect of oxaliplatin, usually presenting as acute or chronic peripheral neuropathy. The acute form develops in more than 90% of patients, with usual onset during or shortly after the first few infusions. Symptoms include paresthesias and dysesthesias in the hands, feet, and perioral region, and may be exacerbated by cold. It is self-limiting [171].

The chronic form is a cumulative axonal sensory neuropathy and may be dose limiting. The neuropathy is reversible in most patients after halting treatment. No intervention has shown definitive prevention of neurotoxicity.

Anti-EGFR agents have a specific adverse effect profile primarily involving skin toxicities. Electrolyte abnormalities also occur with these agents, especially magnesium-wasting syndrome. Cetuximab is associated with an infusion reaction caused by the immunogenicity of the chimeric antibody. The most prominent adverse effects of anti-EGFR agents are skin lesions (e.g., acneiform eruption, paronychial inflammation) and hair abnormalities (including a marked increase in the length of eyelashes). These are sometimes dose-limiting complications that, while not fatal, can greatly interfere with patients' quality of life. The development of skin toxicities (particularly more intense reactions) has actually been associated with better outcomes of cetuximab and panitumumab. Preliminary evidence shows benefit with use of a pre-emptive prophylactic skin treatment regimen of skin moisturizers, sunscreen, topical steroids, and doxycycline [225].

Outcomes from several large clinical trials were pooled and analyzed and demonstrated that following resection of the primary tumor, 85% of colon cancer recurrences occur within three years and 95% occur within five years. These results underscored the importance of regular surveillance for a minimum of five years following the resection of stage II and III colon cancer [335]. Accordingly, several professional organizations have published updated practice recommendations for surveillance of patients with resected stage II and III colon cancer. The recommendations by the ASCO, the NCCN, and the joint European Society of Medical Oncology and Japanese Society of Medical Oncology (ESMO/JSMO) are broadly similar but differ on some parameters (Table 12) [335,338].

Guidelines for surveillance of patients following resection of stage II/III colon and rectal cancer have been produced by Cancer Care Ontario and endorsed by the ASCO. Many recommendations for patients with stage II/III rectal cancer are the same as those described for patients with colon cancer [339]. A medical history, physical examination, and CEA testing should be performed every six months for five years. In addition to abdominal and chest CT imaging, pelvic CT should be performed every 6 to 12 months for two to three years, then annually until five years from surgery.

Rectosigmoidoscopy should be performed every six months for two to five years in patients who did not receive pelvic radiation [339]. In the absence of complete pre-diagnosis colonoscopy, a colonoscopy should be done as soon as is reasonable after completing adjuvant therapy and within six months of completing primary treatment. New and persistent or worsening symptoms, such as pelvic pain, sciatica, and difficulty urinating or defecating, may indicate rectal cancer recurrence.

Measurement of the serum glycoprotein CEA as a tumor marker for colorectal cancer has been used to help guide patient management and follow-up. Serum CEA testing is not valuable in screening for colorectal cancer because of its low sensitivity and specificity [340]. Use of postoperative CEA testing is usually limited to patients who may benefit from further intervention, including [339]:

The NICE recommends offering follow-up for the first three years to all patients with primary colorectal cancer undergoing treatment with curative intent [238]. Follow-up should begin at an outpatient clinic visit four to six weeks after potentially curative treatment. Regular surveillance with colonoscopy, CEA testing, and CT of the chest, abdomen, and pelvis, should be provided as indicated by the treating oncology team. Any clinical, radiologic, or biochemical finding suspicious of recurrent disease should initiate further testing [238]. Regular follow-up may be halted when the patient and healthcare professional have discussed and agreed that likely benefits no longer outweigh risks of further tests or when the patient can no longer tolerate further treatments.

After any treatment, patients should receive specific information on managing the effects of treatment on their bowel function. This could include information on incontinence, diarrhea, difficulty emptying bowels, bloating, excess flatus, diet, and where to go for help in the event of symptoms. Verbal and written information should be clearly understood by the patient and free from jargon. Information about support organizations or Internet resources may be included [238].

As a result of the evolving demographics in the United States, interaction with patients for whom English is not a native language is inevitable. It is each practitioner's responsibility to ensure that information and instructions are explained in such a way that allows for patient understanding. In this multicultural landscape, interpreters are a valuable resource to help bridge the communication and cultural gap between clients/patients and practitioners. Interpreters are more than passive agents who translate and transmit information back and forth from party to party. When they are enlisted and treated as part of the interdisciplinary clinical team, they serve as cultural brokers, who ultimately enhance the clinical encounter. In any case in which information regarding diagnostic procedures, treatment options, and medication/treatment measures is being provided, the use of an interpreter should be considered.