Key points

Introduction

Cancer is fundamentally a genetic disease caused by mutational or epigenetic alterations in DNA. There has been a remarkable expansion of the molecular understanding of colonic carcinogenesis in the last 30 years and that understanding is changing many aspects of colorectal cancer (CRC) care. This article provides a general update on CRC and highlights how genetics is changing clinical care.

Introduction

Cancer is fundamentally a genetic disease caused by mutational or epigenetic alterations in DNA. There has been a remarkable expansion of the molecular understanding of colonic carcinogenesis in the last 30 years and that understanding is changing many aspects of colorectal cancer (CRC) care. This article provides a general update on CRC and highlights how genetics is changing clinical care.

Molecular pathways

The molecular understanding of colonic carcinogenesis continues to rapidly evolve ( Fig. 1 ). Both hereditary and sporadic CRCs are genetically driven diseases. Hereditary CRC syndromes are caused by germline mutations and sporadic CRC is driven by alterations in DNA structure (mutations) or function (epigenetics).

Molecular pathways to CRC. The chromosomal instability (CIN) pathway ( top ) is driven by sequential mutational events (activating mutations in oncogenes, such as KRAS , and inactivating mutations in tumor suppressor genes, such as APC or p53 ) and leads to CRCs that are typically aneuploid, microsatellite stable (MSS), and may have KRAS but not BRAF mutations. The microsatellite instability (MIN) pathway ( middle ) is caused by germline mutations in one of the DNA mismatch repair genes and leads to Lynch syndrome CRCs, which are hypermutated, have microsatellite instability (MSI), and may have KRAS but not BRAF mutations. The CpG island methylator phenotype (CIMP) or serrated pathway leads to CRCs that are MSI or MSS and often have BRAF but not KRAS mutations. These molecular pathways have different histology, prognosis, and response to therapy.

( Adapted from Janne PA, Mayer RJ. Chemoprevention of colorectal cancer. N Engl J Med 2000;342(26):1961.)

About 6% of all CRCs are caused by a hereditary syndrome for which the genetic basis has been identified. Germline mutations in more than 15 genes that are required for DNA repair and/or regulation of signaling pathways that affect DNA repair are now known to cause an increased risk of CRC. These syndromes can be nonpolyposis syndromes, such as Lynch syndrome, or can be colonic polyposis syndromes with development of scores to thousands of adenomatous, hamartomatous, and/or serrated polyps ( Table 1 ).

The three major pathways to CRC (chromosomal instability [CIN], microsatellite instability (MIN)/Lynch, and CpG island methylator phenotype [CIMP]/serrated) are described in Fig. 1 . Both the CIN and MIN pathways are thought to progress through the classic adenoma-carcinoma sequence, whereas the CIMP pathway is a serrated polyp-carcinoma sequence. As a first approximation, CRCs from these pathways are distinguished based on microsatellite stability (MSS) and the mutational status of KRAS and BRAF as shown in Fig. 1 but this is an oversimplification: the mutational/epigenetic composition of CRCs is highly varied with the average CRC containing around 90 different mutations. The Cancer Genome Atlas separated CRCs into hypermutated (16% of CRCs; median mutation number, 728) and nonhypermutated (median of 58 mutations) groups and found that the mutational profiles of the groups differ. Several more complex classification systems are being developed in hopes of more precisely predicting biologic behavior and response to therapy. It is becoming increasingly clear that there are genetic subsets of CRCs that have different risk factors, prognosis, and response to treatment. This article highlights how the genetic understanding of CRC is changing almost all facets of CRC care.

Public health burden

The global incidence of CRC was estimated to be about 1.4 million in 2012; it is the third most common cancer worldwide accounting for about 10% of the total cancer burden. There is up to a 10-fold difference in CRC incidence and mortality around the world, with higher rates generally found in high-income countries. Time trends show incidence and mortality are continuing to increase in low-income countries but have stabilized or are declining in many high-income countries.

There has been an impressive decline in the United States over the last 30 years in the incidence (36%) and mortality (50%) of CRC ( Fig. 2 ) that has been attributed largely to CRC screening. Nonetheless, CRC remains the fourth most common cancer (>134,000 cases) and second leading cause of cancer-related mortality (>49,000 deaths) in the United States with substantial demographic disparities in CRC outcomes.

Time trends of colorectal cancer incidence and mortality in the United States.

( From Siegel R, Desantis C, Jemal A. Colorectal cancer statistics, 2014. CA Cancer J Clin 2014;64(2):112; with permission.)

CRC rates are decreasing in all age groups except those younger than age 50 where an increase in rectal cancer has been seen. Reports from Surveillance, Epidemiology, and End Results CRC registry data initially identified concerning increased incidence of young adult CRCs with an annual percentage increase of about 1.5% to 2% for patients 20 to 49 years of age.

African Americans have the poorest CRC outcomes in the United States, are diagnosed at a later stage of disease, and have about a 10% lower 5-year survival than white persons. These disparities are thought to be largely caused by differences in access to care. In 2012, 70% of white persons compared with 66% of African Americans aged 50 and older ever had a screening colonoscopy or sigmoidoscopy. Similarly, screening rates are lower among Hispanics (56.8%) and persons with annual incomes less than $15,000 (53.8%). For uninsured individuals, screening rates are particularly low, with one study citing 33% of uninsured patients undergoing any test for CRC screening compared with 77% of insured patients. The identification of geographic “hotspots” of high CRC mortality within the United States by Siegel and colleagues may be caused by poverty and racially driven decreased access to care.

Risk and protective factors

Increasing age, male sex, African American race, and birth/residence in high-income countries are well established nonmodifiable risk factors for CRC. Although observational studies have consistently reported obesity, lack of physical activity, alcohol/tobacco use, and diets high in red/processed meats and low in fruits/vegetables/fiber are associated with an increased CRC risk, it is not clear to what extent changing these risk factors in an adult modulates risk. Nonetheless, prevention recommendations for CRC ( Box 1 ) are similar to those for cardiovascular disease prevention and seem prudent.

Screening and early detection

CRC screening is the standard of care in the United States. The US Preventive Services Task Force and all the gastrointestinal and cancer societies recommend colon cancer screening in the general population starting by age 50 (some recommend by age 45 in African Americans) ( Table 2 ). Screening for CRC can prevent CRC by identification and removal of colonic polyps in addition to early detection of CRCs. The concept of precancerous polyps eventually giving rise to CRC was suggested as early as 1927. The recognition by Dukes in 1932 that CRC survival was better in earlier stage disease birthed the notion that removal of precancerous polyps and detection of early stage CRC could reduce the impact of CRC. The history of CRC screening has been recently chronicled, and indeed much of the recent decline in CRC incidence and mortality in the United States (see Fig. 2 ) is thought to be caused by these efforts.

CRC screening is discussed in detail elsewhere in this issue (see Anderson BW, Ahlquist, DA: Molecular Detection of Gastrointestinal Neoplasia: Innovations in Early Detection and Screening , in this issue) but the current status of screening tests is summarized here. CRC screening tests recommended in the United States are divided into those that primarily detect cancer (fecal occult blood tests [FOBT], fecal immunochemical tests [FIT], stool DNA tests) and those that can also detect polyps (flexible sigmoidoscopy [FS], colonoscopy, computed tomography colonography). Multiple controlled trials have shown that guaiac-based FOBTs decrease CRC mortality by 12% to 33%. Although FITs have not been studied in controlled trials with cancer mortality as the end point, they are preferred over guaiac-based FOBTs based on their higher sensitivity for CRC. Stool DNA tests, in turn, have a higher sensitivity but lower specificity than FITs for CRC and advanced adenomas.

FS has been shown in controlled trials to reduce CRC mortality by about 25%. Despite a lack of controlled trials, case-control and cohort studies have indicated superiority of colonoscopy over FS, making colonoscopy the preferred intervention because of its higher sensitivity for CRC and adenoma detection compared with FS. Several controlled trials are currently underway comparing the effectiveness of colonoscopy to FOBTs including one in the United States (the CONFIRM trial in the United States funded by the Veterans Administration [ ClinicalTrials.gov NCT01239082 ]); the final results of these trials are still a decade away.

Although arguments can be made for which CRC screening test is superior, the best screening test is the one that gets done and done well. Determinants of which screening options are feasible in a country depends largely on health resource availability. The World Gastroenterology Organization has outlined a resource-based cascade of CRC screening options to provide a framework for health care systems to do the best with the available resources and the Organization is actively engaged in many countries to promote affordable screening options and provide training in endoscopic skills in low-resource countries.

Although there are quality issues relevant to all CRC screening tests, colonoscopy is thought to be the most operator-dependent. Thus, substantial efforts have been made to improve colonoscopy quality. Missed adenomas are thought to be the most common reason for interval CRCs (CRCs that occur before the next recommended screening examination). The adenoma detection rate (ADR), which is the fraction of patients undergoing screening colonoscopy who had one or more adenomas detected, inversely correlates with interval CRC risk and has been adopted by the American College of Gastroenterology and American Society for Gastrointestinal Endoscopy as perhaps the most important quality indicator in screening colonoscopy. The initial targets for ADR (25% for men; 15% for women) were set at levels slightly below the mean detection rates of adenomas in screening colonoscopy studies. ADRs have been increasing progressively with improvements in colonoscopic equipment, preparation quality, and endoscopic technique; recent studies report many endoscopists have ADRs exceeding 40% or even 50%. Based on these data, recommended ADR targets were recently revised to 30% for men and 20% for women.

Endoscopists with ADRs less than 25% should take steps to improve their ADR. One predictor of a low ADR is short withdrawal time (WT; time of mucosal inspection during withdrawal of the scope from the cecum). It is recommended that WTs be recorded as a colonoscopy quality measure and that average WT should be 6 minutes or more in screening colonoscopies. Endoscopists with low ADRs and short WTs should slow down and take more time to inspect the colonic mucosa and improve their technique (examining proximal portion of folds, suctioning pools of liquid, adequate distention of the colonic lumen). Although incomplete resection of polyps is also thought to be an important contributor to interval cancers, quality measures have not yet address this issue.