SEARCH

SEARCH BY CITATION

Keywords:

  • colorectal cancer;
  • endoscopy;
  • large bowel;
  • prevention;
  • screening

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

Abstract.  Bretthauer M (Oslo University Hospital, Oslo, Norway). Colorectal cancer screening (Review). J Intern Med 2011; 270: 87–98.

Colorectal cancer (CRC) is the third most common cause of cancer death worldwide and a major health problem. In this review, the different approaches for CRC screening will be outlined with emphasis on evidence-based medicine. Evidence from randomized trials on the effectiveness of CRC screening is summarized. Several screening tools for CRC are available. They can be categorized according to their mode of action: early detection tools such as the faecal occult blood test (FOBT) and cancer prevention tools such as flexible sigmoidoscopy and colonoscopy. Meta-analyses of randomized trials show that FOBT screening reduces CRC mortality by 16% (risk ratio 0.84; 95% confidence interval (CI) 0.78–0.9) compared with 30% (risk ratio 0.7; 95% CI 0.6–0.81) for flexible sigmoidoscopy screening. FOBT screening is cheap and noninvasive, but results in large numbers of false-positive tests and needs to be repeated frequently. Flexible sigmoidoscopy is more invasive, but is effective for once-only screening. Although colonoscopy screening is used in some countries, no randomized trials have been conducted to estimate its benefit, and therefore, it should not be recommended at the present time. Faecal occult blood test and flexible sigmoidoscopy are the two CRC screening tools that can be recommended as they have been proven to reduce CRC mortality. Colonoscopy has the potential to be superior to FOBT and flexible sigmoidoscopy, but needs to be evaluated in randomized trials before any recommendation can be provided.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

With more than 550 000 annual deaths, colorectal cancer (CRC) is the third most common cause of cancer death worldwide and a major health problem in industrialized countries [1]. The incidence of CRC has been increasing during the past decades in most countries. In the Nordic countries, the yearly incidence of CRC has increased from 22 cases per 100 000 individuals in 1960 to 34 per 100 000 in 2007 [2] (see Fig. 1). To date, the lifetime risk of CRC for individuals of average risk in industrialized countries is about 5%. CRC is a disease of the elderly; the incidence of CRC rapidly increases after the age of 50 years, and most cases of CRC are diagnosed in individuals who are 60 years of age or older [1].

image

Figure 1. Time trends in colorectal cancer incidence for men (panel A) and women (panel B) in Nordic countries (adapted from 1). Smoothed 5-year incidence rates are displayed from the start of registration in the different countries until 2005.

Download figure to PowerPoint

Five-year survival of CRC exceeds 90% if the disease is diagnosed at an early stage (no tumour extension beyond the bowel wall), but survival is only about 60% for patients with tumours with lymph node involvement and under 10% if metastases are present [3]. However, as clinical symptoms develop late in the course of the disease, early detection is often not achieved in symptomatic patients.

Genetic syndromes associated with a high risk of CRC have been identified. Familial adenomatous polyposis (FAP) is an autosomal dominant condition, characterized by the presence of multiple adenomatous polyps in the colon and a lifetime risk of CRC of more than 80%. Hereditary nonpolyposis colon cancer is an autosomal dominant condition, characterized by early onset of CRC but, in contrast to FAP, without polyposis [4]. These genetic disorders account for less than 5% of all CRC cases [5]. A strong family history of CRC, without known mutations characterizing the above-mentioned genetic syndromes, is also recognized as a risk factor for CRC. Individuals with multiple first-degree relatives with CRC, or any first-degree relatives with CRC diagnosed before the age of 60, have an increased risk of developing CRC, compared with the general population [6]. Most CRC cases (at least 80%), however, develop in so-called average-risk individuals, i.e. in individuals without any known CRC risk factor.

It is believed that the vast majority of CRCs develop from benign precursor lesions, the so-called adenomatous polyps or adenomas, through a series of genetic changes over a long-time period (i.e. the adenoma–carcinoma sequence; Fig. 2). It is estimated that it takes at least 10 years for a small adenoma to develop into cancer. Adenomas can be detected by various methods and removed by gastrointestinal endoscopy, such as colonoscopy or flexible sigmoidoscopy.

image

Figure 2. Adenoma–carcinoma sequence.

Download figure to PowerPoint

Here, an overview of CRC screening in average-risk individuals and an outline of the various screening methods available today will be provided, as well as an algorithm for general practitioners and clinicians to guide patients through the various choices for CRC screening.

Approaches to cancer screening

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

Early detection

The term screening is used for both preventive and early detection methods. The most commonly used screening methods such as mammography for breast cancer screening or prostate-specific antigen testing for prostate cancer screening are examples of screening tools that are intended to detect early-stage cancer. The goal in using these tools is to reduce site-specific cancer mortality because of early detection of the disease. Screening with methods of early detection is not able to reduce the incidence of the target disease. On the contrary, screening with these tools may result in an increase in incidence because of overdiagnosis.

Prevention

Preventive screening is distinctly different from early detection screening. Preventive screening methods target the disease before it becomes malignant, by detecting and removing precursor lesions. Examples are cervical cancer screening that detects preinvasive cervical neoplasia and CRC screening by endoscopy. The aim of these methods is primarily to prevent invasive disease, thereby reducing the incidence of the disease. Reduced incidence will, as a consequence, result in reduced mortality (fewer people will get the disease, so fewer can die as a result of it).

Prerequisites for screening

Population screening is intended for an apparently healthy population. Therefore, screening should only be recommended after careful consideration of both the benefits and harms. In 1968, the World Health Organisation defined a set of prerequisites to be fulfilled before recommending or implementing screening [7]. The key issues are summarized in Box 1. CRC is a very good target for screening and early detection because of (i) the presence of a precursor, (ii) the long time interval from precursor to development of invasive disease, (iii) the ability of various screening tests to detect and remove precursor lesions and (iv) the high incidence of the disease.

Screening evaluation

The evaluation of screening interventions demands a thorough scrutiny of the evidence, a proper control group and knowledge of the principles of cancer screening epidemiology. Lead time is defined as the time interval from the diagnosis of a screening-detected cancer to the time-point at which the cancer would have been detected clinically. Even if screening is not prolonging survival (i.e. the patient dies at the same point in time no matter whether the cancer was detected late clinically or early by screening), because of lead time bias, survival estimates would still favour screening because cancer is detected earlier than without screening. However, in this situation, screening would not provide any benefit. Therefore, careful analyses taking into account lead time bias are essential to prevent the communication of effects of screening when there are not any. Further, screening can only be effective when early-stage cancer has better treatment options compared with late-stage disease. As treatment of symptomatic cancer becomes more successful and more patients survive the disease, the effect of early detection through screening will be reduced. This may be the case in breast cancer screening; because of the dramatic improvements in breast cancer treatment and awareness, screening mammography may not be as effective today as it was 20 years ago [8].

Box 1 The major principles of screening (adapted from 7)

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References
  •  Screening should be directed towards an important health problem.
  •  There should be a simple, safe, precise and validated screening test.
  •  Treatment started at an early stage should be of more benefit than treatment initiated later.
  •  There should be evidence that the screening test is effective in reducing mortality and morbidity.
  •  The benefit of screening should outweigh the physical and psychological harm caused by the test, diagnostic procedures and treatment.
  •  The opportunity cost of the screening programme should be economically balanced in relation to expenditure on medical care as a whole.
  •  There should be a plan for managing and monitoring the screening programme and an agreed set of quality assurance standards.
  •  Potential screening participants should receive adequate information about benefits and disadvantages of participation.

CRC screening tools

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

Faecal occult blood testing

The rationale for the faecal occult blood test (FOBT) in the clinical diagnosis of CRC and its use as a CRC screening tool is based on the observation that small, macroscopically invisible traces of blood (occult blood) are released into the bowel lumen by CRCs. The FOBT analyses stool samples for the presence of this occult blood.

FOBT has been in use in CRC screening for several decades and is still the most common CRC screening test in Europe [9]. Its effect on CRC mortality has been validated in randomized trials, and it is simple to perform, noninvasive and cheap. However, FOBT has a relatively poor sensitivity for CRC, because it is a nonspecific test for gastrointestinal bleeding, resulting in false positives in individuals with gastrointestinal bleeding due to causes other than CRC, such as erosions or ulcers, inflammatory bowel disease, or medication with antiplatelet agents or anticoagulants. Moreover, FOBT is not designed to detect precursor lesions of CRC. Thus, an effect on CRC incidence cannot be expected with FOBT screening.

Currently, two principle types of FOBT screening are available: guaiac-based tests (gFOBTs) and immunochemical tests (iFOBTs). Whereas the former detects the presence of any blood in the stool (including blood of dietary origin, such as raw meat), the latter is more specific for human blood. Furthermore, gFOBTs can be divided into rehydrated and nonrehydrated tests, according to their mechanism of hydration of stool samples. Rehydration increases the sensitivity for CRC, but decreases specificity (leading to more false-positive test results). This is a typical compromise for diagnostic tests; the more sensitive a test is, the lower its specificity and vice versa.

FOBT screening must be repeated at short time intervals to increase sensitivity. Thus, the concept of FOBT screening is a repetitive approach, with either annual or biennial screening of individuals in the target population. The FOBT screening programme is typically organized in a repetitive fashion, in which individuals have to deliver stool samples for the analysis of occult blood at a laboratory, or FOBT test cards can be exposed to stools in individuals’ homes. The effectiveness of an FOBT screening programme is achieved by repetitive FOBT screening rounds.

Test performance

There are a large variety of FOBTs from different manufacturers on the market. Most countries currently lack legal regulation for medical diagnostic tools to prove clinical efficacy before marketing. Therefore, many FOBTs that are currently marketed around the world lack data on test performance, such as sensitivity, specificity and storage and transport durability. Generally, the sensitivity for CRC is relatively low (between 30% and 80%) [10–13]. Test sensitivity can be increased through repetitive screening in FOBT programmes. In these programmes with several screening rounds, the overall ‘programme sensitivity’ increases to about 50–60% for nonrehydrated tests and 80–90% for the rehydrated tests [11]. Specificity has been reported to range between 91% and 98% [14, 15]. Table 1 provides an overview of the performance of FOBTs and other screening modalities.

Table 1. Estimated performance characteristics for the whole colon for different colorectal cancer (CRC) screening modalities [9–15]
Screening testSensitivity for CRC (%)Sensitivity for advanced adenomas (%)Specificity for CRC (%)Specificity for advanced adenomas (%)
  1. Flex sig, flexible sigmoidoscopy; advanced adenomas, adenomas with high-grade dysplasia, villous components or size <9 mm; n.a., not applicable; FOBT, faecal occult blood test.

FOBT
 gFOBT11–6411–4191–98n.a.
 iFOBT56–8927–5691–97n.a.
 Flex sig60–7050–8160–7050–80
 Colonoscopy959595–9990–95

The gFOBT positivity rate ranges from 0.8% to 15%, depending on the particular test used, the screening history of the participants and the prevalence of CRC in the screening population. Accordingly, positive predictive values for gFOBT screening vary considerably between studies and have been estimated to be between 0.9% and 19% [15]. Thus, the majority of positive gFOBT results are false positives.

Immunochemical faecal occult blood tests have better performance characteristics when compared to gFOBTs [10]. Although iFOBTs have been reported to be able to detect advanced adenomas in addition to CRC, most adenomas are in fact missed by iFOBTs (the sensitivity of iFOBTs for advanced adenomas is below 60%, Table 1). iFOBTs are more expensive than gFOBTs. To date, no randomized trials have investigated the effect of repetitive iFOBTs on CRC mortality or incidence. The necessity of performing randomized trials for iFOBTs is currently under debate [16]. As a general approach, new screening tests should always be investigated in randomized trials compared with the current gold standard, according to the principles of comparative effectiveness research.

Efficacy and effectiveness

Four randomized controlled trials have investigated the effect of gFOBT screening on CRC mortality, and the results have been presented in nine reports ([17–25]). Three of the trials [17–23, 25] used repetitive, multiple screening rounds with FOBT (either annually or biennially), whereas the Swedish trial only offered two screening rounds (at the start of the trial period) with a 2-year screening interval [24].

Figure 3 shows a meta-analysis of data from the four randomized trials of the effect of repetitive FOBT screening on CRC mortality. Combining the results from these studies demonstrates that FOBT screening reduces CRC mortality by 16% as compared to no screening in an intention-to-screen analysis, taking into account noncompliance in the screening group. A 25% risk reduction was observed for those individuals who attended at least one screening round (risk ratio 0.75, 95% confidence interval (CI) 0.66–0.84) [10].

image

Figure 3. Meta-analysis (random effects model) of randomized controlled trials of the effect of faecal occult blood test screening on colorectal cancer mortality (adapted from Fretheim and Bretthauer [47]).

Download figure to PowerPoint

Compliance with the screening intervention varied across the studies and between screening rounds (from 53% in the UK study to 78% in one arm of the US study). The three European trials (Denmark, Sweden and United Kingdom) were population based in design, allocating individuals directly from general practice records or population registers to invitation to screening or the control group, whereas the US trial enroled subjects only after they had agreed to participate in the trial. Thus, the compliance rate in the US trial is not directly comparable with the rates observed in the European trials. The Danish and UK trials used nonrehydrated FOBTs, resulting in a low test positivity rate (0.8–3.8%) compared with the US and Swedish trials, which used rehydrated tests (test positivity 1.7–15.4%).

The high test positivity rate in the US trial resulted in a high rate of colonoscopies performed in the screening group. Consequently, after 18 years of follow-up, the incidence of CRC was reduced significantly in the screening group compared with the control group: incidence rate ratio 0.80 (95% CI 0.70–0.90) and 0.83 (95% CI 0.73–0.94) for the annual screening and biennial screening groups, respectively [25]. This can be largely attributed to colonoscopy screening rather than any effect of FOBT screening on CRC incidence. In the other FOBT trials, the rate of colonoscopy was much lower, and no significant effect on CRC incidence was observed.

The effect of iFOBT screening on CRC mortality has been investigated in one randomized trial. This study included 94 000 individuals from Jiashan County, China [26], randomly assigned to a single round of iFOBT screening or no screening. After 8 years of follow-up, there was no difference in CRC mortality between the two groups. However, rectal cancer mortality was reduced in the screening group, and there was a reduction also for total CRC mortality during the early phases of follow-up, which was not evident during the final years of follow-up. The reason for this may be because of the fact that screening was not performed repetitively, as in the other randomized controlled trials (and as generally recommended). The effect of a single screen is only expected to last for a limited duration after the screening intervention; after that, there will again be no difference between the mortality rates in the screening and control groups [25]. Thus, it is difficult to interpret the results of the Chinese trial.

Complications

The complication rate of FOBTs is negligible [24]. However, complications do occur during follow-up of screen-positive individuals (most of whom are disease free). The rate of serious complications (mainly bleeding and bowel perforation) because of colonoscopy is between 0.2% and 0.8% [27].

Flexible sigmoidoscopy

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

Flexible sigmoidoscopy is an endoscopic examination of the distal colon and rectum with a flexible endoscope: either a sigmoidoscope (60 cm long) or a colonoscope (130/160 cm long). Flexible sigmoidoscopy is performed after cleansing the distal colorectum using an enema, usually administered about 30–60 min before the examination. The procedure is restricted by the length of the endoscope, the extent of bowel cleansing and patient tolerance and is often regarded as successful if the rectum and sigmoid colon have been adequately examined. Flexible sigmoidoscopy is usually performed without sedation.

Test performance

The use of flexible sigmoidoscopy in CRC screening is based on the observation that a majority of adenomas and cancers are located in the distal colon (within reach of flexible sigmoidoscopy) [28, 29]. Flexible sigmoidoscopy has a high sensitivity and specificity for both adenomas and CRC within this area of examination. The sensitivity for advanced adenomas (those with a high malignant potential) and CRC in the whole colon is about 60–70% of that achieved by colonoscopy, depending on the distribution of adenomas and CRC in the colon, the prevalence of lesions in the distal colon as precursors for proximal lesions and the threshold for follow-up colonoscopy after flexible sigmoidoscopy screening [30] (Table 1). For the distal colon, sensitivity and specificity of the tool are comparable to those achieved by colonoscopy provided adequate bowel cleansing and performance of the technique by an experienced endoscopist.

Considerable variation in the quality of flexible sigmoidoscopy has been reported between endoscopists with regard to detection of colorectal polyps and adenomas [31–33]. Therefore, as for other operator-dependent CRC screening tools, a high level of competency and rigorous training and quality assurance policies are necessary to achieve good-quality CRC screening programmes involving flexible sigmoidoscopy [29].

Efficacy and effectiveness

Three randomized controlled trials have been conducted to evaluate the effect of flexible sigmoidoscopy screening on CRC incidence and mortality [34–36]. The most recent, the UK flexible sigmoidoscopy screening trial, is the largest of these studies and provides the most comprehensive data [35]. The trial enroled more than 170 000 people aged 55–64 who were randomly assigned to flexible sigmoidoscopy screening or no screening. Screening was performed between 1994 and 1999, with an attendance rate of 71%. After 11 years of follow-up, CRC incidence and mortality were significantly lower in individuals invited to screening compared with the control group: absolute difference in incidence 35 CRC cases, risk ratio 0.77, 95% CI 0.70–0.84; absolute difference in mortality 14 CRC deaths, risk ratio 0.69, 95% CI 0.59–0.82 [35]. In people who actually attended the screening, CRC incidence was reduced by 33% (risk ratio 0.67, 95% CI 0.60–0.76) and CRC mortality by 43% (risk ratio 0.57, 95% CI 0.45–0.72). The effects were similar in men and women.

The Norwegian Colorectal Cancer Prevention (NORCCAP) trial has reported interim results after a significantly shorter follow-up time (6 years for CRC mortality, 7 years for CRC incidence) compared with the UK trial (11 years for both end-points) [36]. In the intention-to-screen analysis, CRC mortality was reduced by 27% (risk ratio 0.73, 95% CI 0.47–1.13). No apparent effect of screening was seen for CRC incidence, with similar incidence rates in the screening and control groups. The most obvious reason for this is the short follow-up time in this study. A follow-up time of at least 10 years is required before any effect of flexible sigmoidoscopy screening can be expected, as shown in previous statistical modelling studies [37]. This is as a result of the findings of screen-detected cancers at the screening intervention, which lead to an early peak of incidence of CRC in screened individuals and the long time interval for the progression from adenoma to CRC. For individuals who attended screening, CRC mortality was significantly reduced by 59% (hazard ratio 0.41, 95% CI 0.21–0.82) [35]. CRC incidence of rectosigmoid CRC was reduced from 79 individuals per 100 000 person-years in the control group to 58 per 100 000 person-years in screening attendees (P = 0.1) [35].

A small-scale, single-centre, randomized study from Telemark County, Norway, was the first randomized controlled trial to evaluate flexible sigmoidiscopy. These trials started in 1983, when 400 individuals were randomly assigned to screening and 399 to a no-screening control group [34]. After 13 years of follow-up, the trial reported an 80% risk reduction in CRC incidence in the screening group compared with the control group in an intention-to-screen analysis. There are two possible explanations for the good efficacy results of the intervention in this study compared with the results of later studies. First, the attendance rate for screening was very high (81%), and secondly, all individuals with polyps of any kind at screening were offered colonoscopy both after screening and again after 2 and 6 years [34].

The two Norwegian studies are population based, as they recruited individuals directly from the population registry by random allocation to either the screening or control group. Thus, observations regarding efficacy are directly transferable to effectiveness. By contrast, the UK trial used a two-step invitation procedure, in which only people who actively expressed an interest in participating were enroled in the trial. Although CRC incidence in the trial control group was similar to the incidence expected in the general population [33], it may not be straightforward to apply the results of the UK trial to clinical practice. The UK trial provides the most reliable and valid estimates of the efficacy of flexible sigmoidoscopy screening, but the effectiveness of flexible sigmoidoscopy screening in the general population is still uncertain [38].

Two other large randomized trials are currently ongoing. The Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial in the USA is the only trial that includes a repetitive flexible sigmoidoscopy screening intervention. In total, 77 465 men and women aged 55–74 were randomly assigned to screening and 77 477 individuals to the no-screening control group [39]. The Italian Screening COlon REtto (SCORE) trial largely follows the UK flexible sigmoidoscopy screening trial design but is smaller in size, with 34 292 individuals randomly assigned to either the screening or control group [40]. The results of these trials will enable the efficacy of flexible sigmoidoscopy screening to be further quantified.

Complications

Major complications because of flexible sigmoidiscopy screening such as bleeding and bowel perforation are very rare during the screening itself. However, follow-up colonoscopy is associated with a small, but considerable risk of adverse events (mainly bleeding or bowel perforation because of polyp removal). The observed adverse event rates are similar to the rates observed for follow-up of screen-positive cases with other CRC screening tools, as colonoscopy is the universal follow-up method.

Comparison of FOBT and flexible sigmoidoscopy

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

A recent feasibility study from the Netherlands reported very interesting data on compliance and yield with gFOBT, iFOBT and flexible sigmoidoscopy screening [41]. Although compliance with screening was higher in the FOBT groups (50% for gFOBT and 62% for iFOBT compared with 32% for flexible sigmoidoscopy), the yield of advanced neoplasia (a combination of high-risk adenomas and CRC) per 100 invited individuals was significantly higher in the flexible sigmoidoscopy arm (2.4), compared with iFOBT (1.5) and gFOBT (0.6). This indicates that flexible sigmoidoscopy may be a more effective screening tool compared with FOBT.

Evaluation of the effectiveness of screening interventions should be based on absolute effects of the interventions on CRC incidence and mortality on an intention-to-treat basis. Tables 2 and 3 show the effectiveness of gFOBT and flexible sigmoidoscopy screening on CRC mortality (intention-to-screen analyses) as derived from randomized trials. As shown in Figs 3 and 4, FOBT screening reduces CRC mortality by 16% (risk ratio 0.84; 95% CI 0.78–0.9), and flexible sigmoidoscopy screening reduces mortality by 30% (risk ratio 0.7; 95% CI 0.6–0.81). CI values are overlapping, and thus, the possible superiority of one test over the other cannot be determined. No head-to-head clinical trials comparing the effects of FOBT and flexible sigmoidoscopy have been conducted.

Table 2. Effects of gFOBT and flexible sigmoidoscopy screening on CRC mortality (intention-to-screen analyses) in randomized trials (adapted from Bretthauer [46])
StudyNo. of individuals in screening/control groupsMean follow-up time (years)Absolute risk reduction (CRC deaths per 100 000 person-years)Relative risk ratio (95% CI)
  1. CI, confidence interval; CRC, colorectal cancer; FOBT, faecal occult blood test.

  2. aFlexible sigmoidoscopy. bEstimate from data given in the paper. cCI not reported.

FOBT
 UK [20]76 466/76 3841111/100 0000.87 (0.78–0.97)
 Denmark [48]30 967/30 9661716/100 0000.84 (0.71–0.99)
 USA [17]31 157/15 3941827/100 0000.75 (0.62–0.91)
 Sweden [21]34 144/31 16415.511/100 0000.84 (0.71–0.99)
Flex Siga
 UK [32]57 099/112 9391114/100 0000.69 (0.59–0.82)
 Norway [33]13 823/41 913610/100 000b0.73 (0.47–1.13)
 Norway [31]   400/3991146/100 000b0.33c
Table 3. Effects of gFOBT and flexible sigmoidoscopy screening on CRC incidence (intention-to-screen analyses) in randomized trials (adapted from Bretthauer [49])
StudyNo. of individuals in screening/control groupsMean follow-up time (years)Absolute risk reduction (CRC cases per 100 000 person-years)Relative risk ratio for CRC incidence (95% CI)
  1. CI, confidence interval; CRC, colorectal cancer; FOBT, faecal occult blood test.

  2. aFlexible sigmoidoscopy. bEstimate from data given in the paper.

FOBT
 UK [22]76 466/76 38411+0.2/100 0001.04 (0.95–1.14)
 Denmark [48]30 967/30 96617+0.4/100 0001.02 (0.93–1.12)
 USA [48]31 157/15 394180.6/100 0000.83 (0.73–0.94)
 Sweden [21]34 144/31 16415.50.7/100 0000.96 (0.86–1.06)
Flex Siga
 UK [32]57 099/112 93911  35/100 0000.77 (0.70–0.84)
 Norway [33]13 823/41 9137None
 Norway [31]   400/39913 149/100 000b0.2 (0.03–0.95)
image

Figure 4. Meta-analysis (random effects model) of randomized controlled trials of the effect of flexible sigmoidoscopy screening on colorectal cancer mortality (panel A) and incidence (panel B) (adapted from Fretheim and Bretthauer [47]).

Download figure to PowerPoint

Colonoscopy

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

Colonoscopy is the gold standard procedure for clinical investigation of the colon and rectum. It has a high sensitivity for both adenomas and cancer (Table 1). Colonoscopy is used as a follow-up intervention for individuals with positive results with another CRC screening tool. Colonoscopy is also increasingly used as a primary CRC screening method. Several European countries such as Poland and Germany have introduced colonoscopy screening programmes, and colonoscopy is the most commonly used method for CRC screening in the USA [42]. However, despite its popularity, colonoscopy has never been investigated in a randomized trial with regard to its ability to reduce the incidence and mortality of CRC. Therefore, despite intriguing theoretical assumptions of a high degree of efficacy, in fact the efficacy of colonoscopy on CRC incidence and mortality is currently unknown, and thus, recommendations for the use of colonoscopy have been questioned for good reason [12, 42, 43]. Colonoscopy is not recommended as a primary CRC screening tool in the European Union.

Two large-scale randomized trials investigating colonoscopy for screening are currently in progress. In the Spanish trial, 55 000 individuals between 50 and 69 years of age are being randomly assigned to either iFOBT or colonoscopy screening [44]. The trial started to recruit in 2008, screening is currently in progress at eight centres in Spain, and the final results are expected in 2021 after 10 years of follow-up. The Nordic–European Initiative on Colorectal Cancer (NordICC) is a multicentre, multinational randomisied trial, in which 66 000 individuals are randomly assigned to either colonoscopy or no screening. Screening started in 2009, and a 15-year follow-up period after screening is planned, with an interim analysis after 10 years due around 2022 [45]. Further, a randomized trial in the USA comparing FOBT and colonoscopy screening is currently in the planning stage (personal communication, Jason Dominitz and Doug Robertson, October 2010).

Clinical significance of CRC screening

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

As stated above, the lifetime risk of CRC in Western countries is about 5%. This means that 95% of individuals will never develop the disease, irrespective of screening or other preventive measures. Screening can only address the 5% of the population who will suffer from CRC and reduce their risk, whilst the remaining 95% will have no personal gain from screening. However, as few clear risk factors for CRC have been identified, it is difficult to predict who will develop the disease and who will not.

Tables 2 and 3 show both relative and absolute risk reductions for CRC as a result of screening. Relative risk reductions are important to determine the effects of interventions. Absolute risk reductions, however, are the most important variables both for healthcare providers when determining priorities in healthcare services and for individuals who want to estimate their potential gain from participating in a screening programme or from any other medical intervention.

The absolute risk reductions shown for the various CRC screening methods (see Tables 2 and 3) are larger than those recently reported for mammography screening [45] and may thus be regarded as clinically significant. However, the decision to undergo screening should be a personal one. The aim of the medical community is simply to provide the individual with unbiased, informative and comprehensive information about the advantages and disadvantages of screening [46].

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References

This article is in part based on recent reports of the Norwegian Council for Quality Improvement and Priority Setting in Health Care [47] and a published report of evidence regarding CRC screening by the author [46].

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to cancer screening
  5. Box 1 The major principles of screening (adapted from )
  6. CRC screening tools
  7. Flexible sigmoidoscopy
  8. Comparison of FOBT and flexible sigmoidoscopy
  9. Colonoscopy
  10. Clinical significance of CRC screening
  11. Acknowledgements
  12. Conflict of interest statement
  13. References
  • 1
    Globocan 2002 database. http://www-dep.iarc.fr (accessed 22.02.10).
  • 2
    Engholm G, Ferlay F, Christensen N et al. NORDCAN: cancer incidence, mortality, prevalence and prediction in the Nordic Countries, Version 3.5. Association of the Nordic Cancer Registries. 2009 Danish Cancer Society (http://www.ancr.nu).
  • 3
    Ries L, Melbert D, Krapcho M et al. (eds). SEER Cancer Statistics Review, 1975–2004. Bethesda, MD: National Cancer Institute, 2007.
  • 4
    Lindgren G, Liljegren A, Jaramillo E et al. Adenoma prevalence and cancer risk in familial non-polyposis colorectal cancer. Gut 2002; 50: 22834.
  • 5
    Burt RW, Bishop DT, Lynch HT et al. Risk and surveillance of individuals with heritable factors of colorectal cancer. WHO Bulletin OMS 1990; 88: 65565.
  • 6
    Rex D, Johnson DA, Lieberman DA et al. Colorectal cancer prevention 2000: screening recommendations of the American College of Gastroenterology. Am J Gastroenterol 2000; 95: 86877.
  • 7
    Wilson JMG, Jungner YG. Principles and Practice of Mass Screening for Disease. Geneva: WHO, 1968.
  • 8
    Kalager M, Zelen M, Langmark F et al. Effect of screening mammography on breast-cancer mortality in Norway. N Engl J Med 2010; 363: 120310.
  • 9
    rsa Lv, Anttila A, Ronco G et al. Cancer screening in the European Union. Report on the implementation of the council recommendation on cancer screening – First Report. European Commission, Luxembourg 2008
  • 10
    Hewitson P, Glasziou P, Watson E et al. Cochrane systematic review of colorectal cancer screening using the fecal occult blood test (Hemoccult): an update. Am J Gastroenterol 2008; 103: 15419.
    Direct Link:
  • 11
    Whitlock EP, Lin JS, Liles E, Beil TL, Fu R. Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force. Ann Intern Med 2008; 149: 63858.
  • 12
    Malila N, Oivanen T, Malminiemi O, Hakama M. Test, episode, and programme sensitivities of screening for colorectal cancer as a public health policy in Finland: experimental design. BMJ 2008; 337: a2261.
  • 13
    van Dam L, Kuipers EJ, van Leerdam ME. Performance improvements of stool-based screening tests. Best Practice Res Clin Gastroenterol 2010; 24: 47992.
  • 14
    van Rossum LG, van Rijn AF, Laheij RJ et al. Random comparison of guaiac and immunochemical fecal occult blood tests for colorectal cancer in a screening population. Gastroenterology 2008; 135: 8290.
  • 15
    Hundt S, Haug U, Brenner H. Comparative evaluation of immunochemical fecal occult blood tests for colorectal adenoma detection. Ann Intern Med 2009; 150: 1629.
  • 16
    Lansdorp-Vogelaar I, van Ballegoijen M, Zauber AG et al. Effect of rising chemotherapy costs on the cost savings of colorectal cancer screening. J Natl Cancer Inst 2009; 101: 141222.
  • 17
    Kronborg O, Fenger C, Olsen J et al. Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996; 348: 146771.
  • 18
    Jorgensen OD, Kronborg O, Fenger C. A randomised study of screening for colorectal cancer using faecal occult blood testing: results after 13 years and seven biennial screening rounds. Gut 2002; 50: 2932.
  • 19
    Kronborg O, Jorgensen OD, Fenger C et al. Randomized study of biennial screening with a faecal occult blood test: results after nine screening rounds. Scand J Gastroenterol 2004; 39: 84651.
  • 20
    Mandel JS, Bond JH, Church TR et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. N Engl J Med 1993; 328: 136571.
  • 21
    Mandel JS, Church TR, Ederer F et al. Colorectal cancer mortality: effectiveness of biennial screening for fecal occult blood. J Natl Cancer Inst 1999; 91: 4347.
  • 22
    Hardcastle JD, Chamberlain JO, Robinson MH et al. Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Lancet 1996; 348: 14727.
  • 23
    Robinson MH, Hardcastle JD, Moss SM et al. The risks of screening: data from the Nottingham randomised controlled trial of faecal occult blood screening for colorectal cancer. Gut 1999; 45: 58892.
  • 24
    Lindholm E, Brevinge H, Haglin E. Survival benefit in a randomized clinical trial of faecal occult blood screening for colorectal cancer. Br J Surg 2008; 95: 102936.
  • 25
    Mandel JS, Church TR, Bond JH et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Engl J Med 2000; 30: 343.
  • 26
    Zheng S, Chen K, Liu X et al. Cluster randomization trial of sequence mass screening for colorectal cancer. Dis Colon Rectum 2003; 46: 518.
  • 27
    Hanley JA. Analysis of mortality data from cancer screening studies. Looking in the right window. Epidemiology 2005; 16: 78690.
  • 28
    Levin B, Liebrman DA, McFarland B et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on colorectal cancer, and the American College of Radiology. Gastroenterology 2008; 134: 157095.
  • 29
    Levin TR, Farraye FA, Schoen RE et al. Quality in the technical performance of screening flexible sigmoidoscopy: recommendations of an international multi-society task group. Gut 2005; 54: 80713.
  • 30
    Imperiale TF, Wagner DR, Lin CY et al. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings. N Engl J Med 2000; 343: 16974.
  • 31
    Atkin W, Rogers P, Cardwell C et al. Wide variation in adenoma detection rates at screening flexible sigmoidoscopy. Gastroenterology 2004; 126: 124756.
  • 32
    Bretthauer M, Skovlund E, Grotmol T et al. Inter-endoscopist variation in polyp and neoplasia pick-up rates in flexible sigmoidoscopy for colorectal cancer screening. Scand J Gastroenterol 2003; 38: 63542.
  • 33
    Fracchia M, Senore C, Armaroli P et al. Assessment of the multiple components of the variability in the adenoma detection rate in sigmoidoscopy screening, and lessons for training. Endoscopy 2010; 42: 44855.
  • 34
    Thiis-Evensen E, Hoff GS, Sauar J, Langmark F, Majak BM, Vatn MH. Population-based surveillance by colonoscopy: effect on the incidence of colorectal cancer. Telemark Polyp Study I. Scand J Gastroenterol 1999; 34: 41420.
  • 35
    Atkin WS, Edwards R, Kralj-Hans I et al. Once-only flexible sigmoidoscopy screening in prevention of colorectal cancer: a multicentre randomised controlled trial. Lancet 2010; 375: 162433.
  • 36
    Hoff G, Grotmol T, Skovlund E, Bretthauer M, for the Norwegian Colorectal Cancer Prevention Group. Risk of colorectal cancer seven years after flexible sigmoidoscopy screening: randomised trial. BMJ 2009; 338: B1846.
  • 37
    Cuzick J, Cafferty FH, Edwards R, Møller H, Duffy SW. Surrogate endpoints for cancer screening trials: general principles and an illustration using the UK Flexible Sigmoidoscopy Screening Trial. J Med Screen 2007; 14: 17885.
  • 38
    Bretthauer M. Which tool is best for colorectal cancer screening? Flexible sigmoidoscopy shows promise, but randomised data are needed BMJ 2010; 340: c2831.
  • 39
    Weissfeld JL, Schoen RE, Pinsky PF et al. Flexible sigmoidoscopy in the PLCO cancer screening trial: results from the baseline screening examination of a randomized trial. J Natl Cancer Inst 2005; 97: 98997.
  • 40
    Segnan N, Senore C, Andreoni B et al. Baseline findings of the Italian multicenter randomized controlled trial of “once-only sigmoidoscopy”—SCORE. J Natl Cancer Inst 2002; 94: 176372.
  • 41
    Hol L, van Leerdam ME, van Ballegooijen M et al. Screening for colorectal cancer: randomised trial comparing guaiac-based and immunochemical faecal occult blood testing and flexible sigmoidoscopy. Gut 2010; 59: 628.
  • 42
    Hoff G, Dominitz JA. Contrasting US and European approaches to colorectal cancer screening: which is best? Gut 2010; 59: 40714.
  • 43
    Ransohoff DF. Have we oversold colonoscopy? Gastroenterology 2005; 129: 1815.
  • 44
    Colorectal cancer screening in average-risk population: immunochemical 11 fecal occult blood testing versus colonoscopy. ClinicalTrials.gov. http://www.clinicaltrials.gov/ct2/show/NCT00906997
  • 45
    Northern-European Initiative on Colorectal Cancer (NordICC). ClinicalTrials.10 gov. http://clinicaltrials.gov/ct2/show/NCT00883792.
  • 46
    Fretheim A, Bretthauer M. Results from clinical effectiveness evaluations of screening for colorectal cancer. The Norwegian Knowledge Centre for the Health Services on behalf of The Norwegian Council for Quality Improvement and Priority Setting in Health Care, Oslo 2010.
  • 47
    The Lancet. The trouble with screening. Lancet 2009; 373: 1223.
  • 48
    Allison J. Colorectal cancer screening guidelines: the importance of evidence and transparency. Gastroenterology 2010; 138: 164852.
  • 49
    Bretthauer M. Evidence for colorectal cancer screening. Best Pract Res Clin Gastroenterol 2010; 24: 41725.