| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | COVER ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
||||||||
|
||||||||||
Dr. Smith is Director of Cancer Screening in the Department of Cancer Control, American Cancer Society, Atlanta, GA.
Dr. von Eschenbach is Director of the Program Center for Genitourinary Cancers at The University of Texas M.D. Anderson Cancer Center, Houston, TX.
Dr. Wender is Clinical Professor Vice Chairman in the Department of Family Medicine at Thomas Jefferson University in Philadelphia PA. Dr. Wender contributed to both the prostate colorectal sections of this article.
Dr. Levin is Vice President of Cancer Prevention for the University of Texas M.D. Anderson Cancer Center in Houston, TX.
Dr. Byers is Professor of Preventive Medicine in the Department of Preventive Medicine Biometrics at the University of Colorado School of Medicine in Denver, CO.
Dr. Rothenberger is Professor Chief Divisions of Colon Rectal Surgery Surgical Oncology, Associate Director for Clinical Research Programs, University of Minnesota Cancer Center, Minneapolis, MN.
Dr. Brooks is Program Director for Prostate and Colorectal Cancers in the Department of Cancer Control at the American Cancer Society in Atlanta, GA.
Dr. Creasman is J. Marion Sims Professor of Obstetrics and Gynecology for the Medical University of South Carolina in Charleston, SC.
Dr. Cohen is a Gynecologic Oncologist and Professor of OB/GYN and Reproductive Science at Mt. Sinai Medical Center in New York, NY.
Dr. Runowicz is Professor and Director in the Division of Gynecologic Oncology in the Department of Obstetrics & Gynecology at the Albert Einstein College of Medicine Montefiore Medical Center in the Bronx, NY.
Dr. Saslow is Director of Breast and Cervical Cancers in the Department of Cancer Control at the American Cancer Society in Atlanta, GA.
Dr. Cokkinides is the Program Director for Risk Factor Surveillance in the Department of Epidemiology and Research Surveillance at the American Cancer Society in Atlanta, GA.
Dr. Eyre is Executive Vice President for Research and Medical Affairs, American Cancer Society, Atlanta, GA, and Editor in Chief of CA.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
Although it is likely that current screening protocols will be supplanted in the future by newer, more effective technologies, the establishment of an organized and systematic approach to early cancer detection would lead to greater utilization of existing technology and greater progress in cancer control.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
During 2000, the ACS reviewed and revised early detecion guidelines for prostate cancer, colorectal cancer, and endometrial cancer, and updated the narrative related to the use of early detection tests for lung cancer. This report includes the recom-mendations and rationale for the updated guidelines, as well as a summary table of all ACS early cancer detection guidelines (Table 1
).
|
|
Recommendations from the American Cancer Society Workshop on Early Prostate Cancer Detection, May 4-6, 2000 |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
The recent trend in prostate cancer incidence is marked by a dramatic increase in incidence rates beginning in the mid-1980s and peaking in 1993, mostly due to the intro-duction of prostate specific antigen (PSA) testing for early prostate cancer detection.6 During that period, prostate cancer incidence rates increased 108%.7 Incidence rates declined just as dramatically between 1992 and 1995, with an estimated annual percentage drop of 4.2%, and since 1995, rates have been relatively stable.6) Overall, prostate cancer mortality rates declined between 1994 and 1997 at an annual average rate of 4.4%, with death rates declining in white men at an average annual rate of 4.5%, and in black men at a average annual rate of 2.3%.6
Prostate cancer five-year survival is nearly 100% when the disease is diagnosed at a local or regional stage, but poor when diagnosed with distant metastases (32.6%).6 Most men with prostate cancer are diagnosed with local or regional disease (80%), but overall, both stage distribution and survival is poorer in African-American men compared with white men.6,7
Survival is strongly influ-enced by the histologic grade of tumors, with five-year survival nearly 100% among men with well-differentiated disease at the time of diag-nosis compared with 73.5% for men with poorly-dif-ferentiated disease.7 Five-year survival tends to be poorer in men younger than 50 probably because the disease tends to be more advanced at the time of diagnosis.7
Guidelines Development
In 1992, the ACS issued a guideline for average-risk men, recommending annual screening for prostate cancer with digital rectal examination (DRE) and serum PSA measurement beginning at age 50.8 In 1997, the ACS convened a workshop to review the recommendations for testing for early prostate cancer detection, and, based on the review of recent data, the guidelines were revised in a manner that placed greater emphasis on informed decision-making.9 Thus, the revised guideline recommended "that both the PSA test and the DRE be offered annually, beginning at age 50, to men who have a life expectancy of at least 10 years, and to younger men who are at high risk." Also, the revised guideline stressed that information should be provided to patients about the risks and benefits of testing.
In May 2000, the ACS convened a workshop to review data accumulated since 1997, and update guidelines for prostate cancer testing. The workshop was co-chaired by Drs. Gerald Woolam, Andrew von Eschenbach, and Richard Wender. The planners of the work-shop identified four relevant issues that warranted a review based on new evidence: (1) the efficacy of testing for early prostate cancer detection; (2) methodologies for PSA measurements; (3) the informed decision process; and (4) testing in high-risk groups.
Workshop participants revisited the historical evidence, but focused especially on research findings since 1997 and studies now underway. During the current guideline review, published articles related to prostate cancer detection, risk, and risk factors were identified using MEDLINE (National Library of Medicine) for the years 1995 through 2000, bibliographies of identified articles, and from the personal files of the advisory group and expert panel members. Key reports were distributed to workshop attendees prior to the May 2000 meeting. Following the workshop, a writing committee assembled by the workshop chairs discussed recent evidence and recommendations from the three workgroups for guideline modification, which were then decided by consensus. Each member reviewed the draft of this manuscript.
|
RECENT DATA ON PROSTATE CANCER TESTING FOR EARLY DETECTION |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
In 1993, PSA testing was made available to men ages 45 to 75 in Tyrol, Austria (population: 331,410).10 Some testing with DRE and PSA had already taken place between 1990 and 1993. Of 65,123 men in the county ages 45 to 75, 32.3% were tested in the first year, and 62% of men were tested at least once during the first four years of the study. Registry data from Tyrol indicate a significant shift towards more favorable stage at diagnosis, and an increase in the proportion of cases with organ-confined disease. In Tyrol, mortality remained constant between 1970 and 1993, but has declined 42% since 1994 (most recent update is 1998).11 In other parts of Austria, prostate cancer mortality rates are unchanged. The decrease in observed versus expected mortality is statistically significant.
Recent analysis of the National Cancer Institute’s (NCI) Surveillance Epidemiology, and End Results (SEER) data shows that prostate cancer mortality in white men younger than age 85 has declined to levels below those that existed prior to the PSA era, which began about 1986.6 In fact, for men ages 60 to 79, mortality rates in 1997 were lower than in any year since 1950.12
In the absence of evidence of mortality reduction from a prospective randomized trial [the Prostate Lung Colorectal and Ovary (PLCO) trial is still in progress],13 Gann proposed that a reduction in prostate cancer mortality below that which existed prior to the introduction of PSA testing would be compelling evidence of the efficacy of the screening test.14 Following Gann’s lead, Tarone, Chu, and Brawley evaluated recent age-specific trends in stage-specific survival rates among white men from the SEER database. Since it is distant-stage disease that is significantly more likely to be fatal in the near term compared with regional disease, the obser-vation that incidence rates of distant disease were declining while local and regional disease incidence rates were increasing is highly suggestive of a screening effect. They observed that the recent decline in mortality is associated with a decline in the incidence rate of advanced disease, and especially with an increase in the detection of organ-confined (i.e., non-metastatic), high-grade disease.12
Trend data from Olmsted County, MN, suggest a testing effect on mortality trends. Mortality rates increased from 1980 to 1992, and subsequently have declined 22% between 1992 and 1997. The decline in mortality was preceded by a decline in the incidence rate of advanced disease.15 Additional data supporting an effect of PSA testing on reducing prostate cancer mortality from a Department of Defense prospective population study (2,042 prostate cancer patients registered at the Center for Prostate Cancer Research, Walter Reed Army Medical Center) showed a decline in the rate of advanced disease and a corresponding improvement in disease-specific survival.16
While organizations differ in their recommendations about testing for early prostate cancer detection, there is greater agreement that physicians should be prepared to identify men who potentially would benefit from testing and provide information about prostate cancer risk, benefits, limitations, and potential harms associated with testing.17 However, evidence suggests that there are differences among physicians in what information is regarded as important for making informed decisions, and differences between what physicians and patients regard as important information for making informed decisions.18 Wolf et al have shown that patients differ in their decisions about testing, as well as individual circumstances (age, socioeconomic status, perceived risk, family history, etc.).19,20 To the extent that many men are uninformed about prostate cancer in general, and more specifically about testing for early prostate cancer detection, evidence suggests that recommendations for informed decision-making require supportive strategies for men and their health care providers.21
|
ACS GUIDELINE ON TESTING FOR EARLY PROSTATE CANCER DETECTION: UPDATE 2001 |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
High-risk groups include men of African descent (specifically, sub-Saharan African descent) and men with a first-degree relative diagnosed at a young age.22,23 Risk increases with the number of first-degree relatives affected by prostate cancer. The workgroup recommended that these men begin testing for early prostate cancer detection at age 45. Among men of African descent, age-specific risk increases steadily beginning at age 45. Men at appreciably higher risk of prostate cancer due to multiple first-degree relatives who were diagnosed with prostate cancer at an early age could begin testing at age 40. However, if PSA is less than 1.0 ng/dl, no additional testing is needed until age 45. If PSA is greater than 1.0 ng/dl but less than 2.5 ng/dl, annual testing is recommended. If PSA is 2.5 ng/dl or greater, further evaluation with biopsy should be considered. Men at high risk also should be informed about the benefits, limitations, and uncertainties associated with testing for early prostate cancer detection.
Prostate Cancer Early Detection Tests
Measurement of serum PSA level is the most accurate method for the detection of prostate cancer and is superior to DRE. Nevertheless, DRE should be included in testing whenever appropriate. The positive predictive value of an abnormal DRE in patients with low PSA levels (i.e., 1.0 ng/dl) is very low and does not warrant further evaluation. In men for whom DRE is an obstacle to testing, PSA alone is an acceptable alternative.
Since PSA is prostate-tissue specific and not prostate-cancer specific, there is no absolute value that is applicable to all men. The range of "normal" PSA levels has conventionally been considered to be between zero and 4.0 ng/dl. A lower cut-off value of 2.5 ng/dl has been shown to improve the early detection of organ-confined prostate cancers; however, this also increases the number of men undergoing biopsy in whom no cancer is detected.
Age-specific reference ranges and PSA density (amount/vol-ume) have been employed to improve specificity. Because PSA is prostate-tissue specific and not prostate-cancer specific, elevations of PSA into the "abnormal" range may occur due to benign prostatic hyperplasia or prostatitis. Benign prostate tissue produces a higher percentage of free PSA than does cancerous tissue. This biologic observation can be used to improve the predictive value of the test in men with elevated total PSA levels. For men with PSA results between 4.0 and 10.0 ng/dl, restricting transrectal ultrasound-guided biop-sy to men with less than 20% free-PSA improves testing accuracy. Applying this strategy to men with PSA levels between 2.5 and 10.0 ng/dl may lead to detection of early disease in a larger number of men and may result in a lower biopsy rate compared with older strategies.
Conclusion
The new guideline represents a stronger recommendation than was issued in the 1997 update. Because the potential benefits of early detection must be balanced against potential risks, the new guideline is consistent with the 1997 guideline insofar as it is not a recommendation for mass screening for prostate cancer in average-risk men. Rather, it is an endorsement that men should have an opportunity to be tested and should actively participate in the testing decision. Following a discussion of benefits and limitations of prostate cancer testing, some men may choose not to be tested at that time. However, by including "should" in the recommendation, the ACS is more clearly stating that asymptomatic men age 50 and older ought to have an annual opportunity to make an informed decision about testing for early prostate cancer detection. For this reason, the Advisory Group felt that it is improper to discourage testing, or not to offer testing.
The advisory group recommends that clinicians devote adequate time to inform patients about the potential benefits and risks of testing for early prostate cancer detection. Utilizing educational tools as a way to efficiently inform patients is encouraged. Asking questions to assess the patient’s beliefs about prostate cancer and the treatments for this disease may help the patient to arrive at a clear preference.
Despite ideal efforts to share the decision-making process with patients, some individuals will be unable or unwilling to state a clear preference and will prefer to follow the clinician’s advise to be tested or to forgo testing. For these individuals, the advisory group believes that the potential to prevent a premature prostate cancer death is high enough to justify testing this informed individual. This wording has been added to the new guideline.
The new guideline continues to emphasize the importance of informed decision-making as opposed to screening individuals without the opportunity to participate in the decision-making process. Meeting this standard will usually require that testing take place within the context of the clinician-patient relationship with the opportunity to discuss potential benefits and limitations of testing. Screening individuals outside of the clinical arena, for example, in community settings or health fairs, is only warranted if patients have the opportunity to participate in an educational process and to discuss their decision with a clinician. The advisory group believes that individuals are more likely to discuss these issues freely with their own clinicians and recommends that testing for early prostate cancer detection should occur within the context of the patient’s usual clinical care.
|
American Cancer Society Guidelines on Screening and Surveillance for the Early Detection of Adenomatous Polyps and Cancer—Update 2001 |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
Without preventive interventions, about 5.6% of Americans will develop colorectal cancer at some point during their lives.24 When colorectal cancer is diagnosed at an early, localized stage, five-year survival is 90%, yet only 37% of incident cases are diagnosed while still localized.6 The same methods used to detect colorectal cancers at early, curable stages can also identify and remove adenomas, which give rise to colorectal cancer.25,26 Methods for early detection can therefore actually prevent colorectal cancer.
Guidelines Development
Since 1980, the ACS has recommended screening for colorectal cancer,27 and guidelines for colorectal cancer screening were last updated in 1997.3 Late in 2000, the ACS Colorectal Cancer Advisory Committee convened a workshop in Atlanta to review recent scientific evidence for colorectal cancer screening. The workshop re-visited the historical evidence, but focused especially on research findings from the past five years and studies now underway.
During the current guideline review, published articles related to colorectal cancer, screening, risk, and risk factors were identified using MEDLINE (National Library of Medicine) for the years 1995 through 2000, bibliographies of identified articles, and through the personal files of the advisory group and expert panel members. Key reports were distributed to workshop attendees prior to the October 2000 meeting. Following the workshop, ACS Colorectal Cancer Advisory Group members discussed new evidence and deliberated over guidelines modifications, which were then decided by consensus. Each advisory group member and workshop atten-dee reviewed the draft of this manuscript.
|
DEFINITIONS: EARLY DETECTION METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
A single test of a stool sample in the clinical setting (as, for instance, is often done with the stool sample collected on the fingertip during a DRE) is not an adequate substitute for the recommended procedure of collecting two samples from three consecutive specimens. This is because colonic neoplasms often bleed intermittently and/or because blood is often not present throughout the entire stool. Further, several studies have shown that the yield of the three-sample protocol is higher than that for the first sample.28
FOBT or flexible sigmoidoscopy alone could be reason-able choices for routine cancer screening. Either test alone is better than no screening at all...
The most commonly used FOBTs are guaiac-based tests, such as Hemoccult II® and Hemo-ccultSENSA® (Smith, Kline, and French, Sunnyvale, CA), which are uncomplicated and can be pro-cessed in the physician’s office. There are immunochemical-based FOBTs that may be processed in a laboratory, or in a doctor’s office, but they are much less commonly used.29
Flexible Sigmoidoscopy
Flexible sigmoidoscopy refers to a procedure that allows direct visual examination of the distal portion of the colorectum by a trained examiner using a flexible 60-cm endoscope following a satisfactory cleansing of the descending and sigmoid colon and rectum. Even though flexible sigmoidoscopy visualizes only the distal portion of the colorectum, when the scope reaches the splenic flexure, it can nonetheless identify approximately 80% of those individuals with a significant neoplasm anywhere in the colorectum.30 This is because people often have more than one adenoma within the colorectum, and adenomas in the distal colorectum can be a marker of risk for lesions in the proximal colorectum.
Colonoscopy
Colonoscopy refers to a procedure that allows direct visual examination of the entire colorectum using a colonoscope. Ideally, the colonoscopic examination should be extended to the cecum. Even though some benign and malignant neoplasms can be missed by colonoscopy, it is considered the most sensitive of the screening methods. Thus, most other detection methods have been compared with colonoscopy, which is considered the standard.31
Double Contrast Barium Enema
Double contrast barium enema (DCBE) refers to a procedure that allows radiologic examination of the entire colorectum by instilling both barium and air to define the contours of the colorectal mucosa.32 DCBE is less sensitive than colonoscopy for visualizing smaller neoplasms, especially ade-nomas smaller than 1 cm.25
Digital Rectal Examination
DRE refers to the palpation of the anus and lower rectum by the practitioner using a gloved finger. Although DRE is a useful method for identifying masses in the anal canal or lower rectum, it has very poor sensitivity for detecting colorectal cancer due to limited reach. While DRE is often included as part of a routine physical examination, it is not recommended as a stand-alone screening test for colorectal cancer. However, DRE should be performed prior to insertion of a sigmoidoscope or colonoscope.
New Screening Methods
Several new methods being developed may be clinically useful for colorectal screening in the future. Computed tomography colonography (CT colonography, also sometimes called virtual colonoscopy) is a procedure that allows imaging the colorectum by CT radiographs.33 Molecular screening refers to a method of detecting mutations associated with colorectal neoplasia in the DNA of stool samples. This methodology is now being evaluated as a potential screening tool.34,35 Neither of these methods is currently ready for widespread clinical application outside the context of controlled clinical trials.
|
ACS COLORECTAL CANCER SCREENING GUIDELINES—UPDATE 2001 |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
The ACS guidelines for the early detection of colorectal adenomas and colorectal cancer are summarized in Tables 2 and 3. These guidelines offer a set of screening choices for different levels of colorectal cancer risk. Each of the choices has inherent characteristics related to accuracy, prevention potential, costs, and risks. Following is an explanation of the main features of these updated ACS recommendations.
|
|
Adults at average risk should begin colorectal cancer screening by age 50. Although the incidence of invasive disease is low at age 50, about 25% of adults at age 50 will have adenomatous polyps.26 Thus, the underlying logic for beginning screening at age 50 is related more to the potential to detect and remove precursor lesions than to the potential for detection of invasive disease.
The 2001 guideline provides five options for screening of average-risk individuals as sum-marized in Table 2
. The screening options are:
Annual FOBT alone has been shown to reduce risk of death from colorectal cancer by about one third when repeated annually.38 A positive FOBT should be followed by a colonoscopy because of the possibility that an important lesion can be visualized and biopsied during the examination. If a clinically relevant source of bleeding is not identified, then a source outside the large bowel should be investigated. An alternative to colonoscopy is DCBE, or flexible sigmoidoscopy followed by DCBE. The addition of flexible sigmoidoscopy to the DCBE is based on the likely greater sensitivity of endoscopy for smaller lesions, although the incremental benefit of flexible sigmoidoscopy when added to the DCBE has not been well evaluated. Because of the impor-tance of accuracy in a diagnostic workup, these examinations should be performed by experienced practitioners or centers.
Flexible sigmoidoscopy every five years also is regarded as an effective screening test. Evidence from case control studies has shown significant mortality reductions from cancers within reach of the sigmoidoscope, as well as a lower incidence of colorectal cancer associated with a prior history of screening with sig-moidoscopy.39–41,42 The periodicity for regular screening with flexible sigmoidoscopy is more frequent than for colonoscopy because of a range of factors that may affect the overall sensitivity of flexible sigmoidoscopy: (1) flexible sigmoidoscopy generally is not done by specialists and thus the range of practitioner experience and expertise is variable; (2) bowel preparation usually is less complete than for colonoscopy, meaning that important lesions may be obscured by stool; and (3) as no sedation is given with flexible sigmoidoscopy, patient discomfort or spasm may interfere with the completeness of the test. Two randomized, controlled trials are now underway to measure the benefits of sig-moidoscopic screening, but results will not be known for several years.43,44
Because combining flexible sigmoidoscopy with FOBT can increase the benefits of either test alone, the ACS regards annual FOBT accompanied by flexible sigmoidoscopy every five years as a better choice than either FOBT or flexible sigmoidoscopy alone.39,40,42 FOBT offers the potential to detect occult blood from a lesion anywhere in the colorectum, while flexible sigmoidoscopy is superior at detecting distal polyps as well as non-bleeding advanced neoplasias. Both direct and indirect evidence suggest combined testing with FOBT and flexible sigmoidoscopy offers advantages over either FOBT or flexible sigmoidoscopy alone.26,45
Examining the entire colorectum, either by colonoscopy every 10 years30 or by DCBE every five years,32 is also a good choice. A supplementary DCBE may be needed if a colonoscopic exam fails to reach the cecum, and a supplementary colonoscopy may be needed if a DCBE identifies a possible lesion, or does not adequately visualize the entire colorectum. DCBE is usually less expensive than colonoscopy, but colonoscopy appears to have greater sensitivity and provides direct visualization of lesions, which can be biopsied or excised during the procedure.
The choice of colonoscopy or DCBE for screening can be made on an individual basis, depending on factors such as personal pre-ference, cost, feasibility, tolerance of potential complications, and the local availability of trained clinicians able to offer a high-quality examination. For those who elect either colonoscopy or DCBE for screening, there is no need for annual FOBT.
Recommendations for Individuals at Increased risk
Approximately 15% to 20% of colorectal cancers occur among people at "increased risk" (approximately twice average risk).26 People who have been diagnosed as having adenomatous polyps should have a colonoscopy to remove all polyps from the colorectum, after which a colonoscopic exam should be repeated at an interval to be determined on the basis of the size, multiplicity, and histologic appearance of the adenoma(s) (Table 3).26 If a colonoscopic examination is not available, not feasible, or not desired by the patient, a DCBE, or flexible sigmoid-oscopy followed by a DCBE can be used for surveillance. A similar recommendation applies to individuals with a personal history of curative-intent resection of colorectal cancer (Table 2).
Evidence demonstrates that when a screening recommendation comes directly from the clinician, compliance with colorectal cancer screening can be quite high.
A family history of either colorectal cancer or colorectal adenomas increases one’s risk of developing colorectal cancer.46,47 Risk is increased for individuals with a family history involving first-degree relatives, and is even higher if a first-degree relative (parent, sibling, or offspring) has had a colorectal cancer or an adenomatous polyp diagnosed before age 60 years, and/or if more than one first-degree relative has been affected at any age. On the other hand, individuals with a single first-degree relative diagnosed with colorectal cancer or an adenomatous polyp after age 60, or with affected relatives who are more distant than first degree, can be considered to be at "average risk."
People with a family history of either colorectal cancer or colorectal adenomas that occurred in a first-degree relative before age 60, or in multiple first-degree relatives of any age (if not a hereditary syndrome), should have a colonoscopy, which should be repeated at intervals to be determined on the basis of the initial examination (Table 3). If a colonoscopy is not available, not feasible, or not desired by the patient, a DCBE, or flexible sigmoidoscopy followed by a DCBE can be used.
Recommendations for Individuals at High Risk
Approximately 5% to 10% of all colorectal cancers occur among people at "high risk," defined as much greater than twice average risk.26 Along a continuum of "high risk" are those individuals with inflammatory bowel disease and those individuals with one of two hereditary syndromes that place them at very high risk for colorectal cancer.
When either ulcerative colitis or Crohn’s disease affects a significant extent of the colorectum, the risk of colorectal cancer is greatly increased beginning eight years after the onset of colorectal symptoms.48 Individuals with extensive inflammatory bowel disease affecting the colon should begin endoscopic surveillance with biopsy for dysplasia every one to two years after eight years of symptoms. Prophylactic colectomy should be considered in the presence of persistent dysplasia.
Familial adenomatous polyposis syndrome (FAP) is a genetic condition affecting one in 10,000 people that is caused by a mutation in the APC gene on chromosome 5.47 People with FAP develop hundreds of colorectal polyps and will almost certainly develop colorectal cancer unless the colon is removed.
Hereditary non-polyposis colorectal cancer (HNPCC) syndrome is an inherited genetic condition that can also cause colorectal cancer among many people in a family, even though multiple polyps are not present. HNPCC is caused by mutations in mismatch repair genes located on chromosomes 2, 3, or 7.47 HNPCC has been classically defined as colorectal cancer in three or more family members, two of whom are first-degree relatives of the third, involving people in at least two generations, and with at least one person diagnosed with colorectal cancer under age 50. However, other variants of this classic syndrome exist. As genetic tests are now available to detect the mutations that lead to FAP and HNPCC, genetic counseling should be offered to individuals with family histories suggestive of these hereditary syndromes (Table 3).
|
CHANGES FROM THE 1997 RECOMMENDATION |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
The interdisciplinary task force was more specific, recommending either annual FOBT, flexible sigmoidoscopy every five years, or combined FOBT and flexible sigmoidoscopy beginning at age 50 and repeated at intervals recommended for each test.26 The ACS recommended, at a minimum, annual FOBT and flexible sigmoidoscopy every five years.3 The USPSTF did not recommend DCBE or colonoscopy for colorectal cancer screening, whereas both the interdisciplinary task force and the ACS endorsed these two tests as reasonable options.
The 1997 ACS recommendation for annual FOBT and flexible sigmoidoscopy every five years at a minimum represents an important difference between the Society’s guidelines and those of the USPSTF and the AHCPR Task Force. The ACS made this recommendation on the basis of three data sets that showed: the benefits of FOBT;50–52 substantial risk reduc-tion from sigmoidoscopic exams;39,40,42,45,53,54 and the safety, reasonable cost, and frequent availability of flexible sigmoidoscopy in many clinical systems.
Guidelines for average-risk individuals have been modified since the 1997 ACS recommendation3 to allow for greater flex-ibility in achieving screening goals, due to evidence showing little progress in colorectal screening rates.55 Based on data from the Behavioral Risk Factor Surveillance System (BRFSS), for example, fewer than one in five adults reported having had an FOBT in the previous year, and only 9.5% of adults reported having had both an FOBT test and flexible sigmoidoscopy during an interval recom-mended by the ACS.55
Additional evidence gathered from focus groups with primary care providers and medical directors of managed care organizations indicates an overall lack of preparedness to offer FOBT and flexible sigmoidoscopy in the primary care setting.56,57 Moreover, a recent report indicated that medical directors were more likely to regard flexible sigmoidoscopy as an unreasonable expectation in a capitated plan compared with other screening tests (i.e., mammography, Pap testing, etc.).58
At this time, therefore, economic and health care system disincentives to screening are impinging on the choices available to phy-sicians and patients. For these reasons, the ACS Colorectal Cancer Advisory Committee determined that FOBT or flexible sig-moidoscopy alone could be reasonable choices for routine colorectal cancer screening.
Either test alone is better than no screening at all, although both annual FOBT and flexible sigmoidoscopy every five years are preferred compared with either test alone. Nevertheless, the Advisory Committee strongly asserts that if FOBT alone is chosen, individuals should be tested annually using the recommended take-home multiple sample method.59 It is important to realize that it is the repeated use of this screening method in a properly implemented screening program that has proven effectiveness.
Annual screening is better than biennial screening,50–52 and all positive tests must be properly followed up with a complete examination of the colon.31,59 Specifically, there is no justification for repeating FOBT after an initial positive finding. Providers who are not yet prepared to offer tests other than FOBT in their clinics or by referral should immediately begin to find ways to make these tests available for their patients. Finally, the ACS asserts that third-party payers and health plans should not limit reimbursement for colorectal screening to FOBT alone, nor should providers suffer any economic disincentive to offering screening tests in addition to FOBT.
Recently, there has been increased attention to colonoscopy in the media and by some professional groups as the preferred method of screening for colorectal cancer. The inclusion of colonoscopy as a primary screening option has been an important feature of the ACS guidelines since 1997.3 Colonoscopy can directly examine the entire colorectum, while flexible sigmoidoscopy can examine only the distal portion of the colon. Nevertheless, because adenomas are often multiple, flexible sigmoidoscopy can often identify about 80% of people who have a significant neoplasm anywhere in their colorectum.30
Randomized trials of colonoscopic screening are just now getting underway, so the effectiveness of this method will not be precisely known for several years. However, the ACS Colorectal Cancer Advisory Committee believes that there is sufficient indirect evidence of benefit from colonoscopy to justify an individual’s selection of this test for colorectal cancer screening, provided the individual understands the potential risks associated with the procedure. The cost-effectiveness of colonoscopy will likely improve with declining procedure costs, increasing availability, and targeted age/risk algorithms. Nevertheless, the Advisory Committee concluded that there are remaining questions about the efficacy and effectiveness of colonoscopy, as well as practical issues related to cost, access, and availability that need to be answered before colonoscopy is likely to be widely available as a screening test for average-risk adults.
An alternative to colonoscopy for vis-ualization of the colorectum is a DCBE, which, when performed by an experienced radiologist, can be a cost-effective method for finding larger lesions.26,32 However, because DCBE may be less sensitive for detecting smaller neoplasms,60 it probably should be repeated more frequently (every five years) than colonoscopy.
A potential advantage of presenting a menu of appropriate screening options is that it makes it possible for any clinician to screen all of his or her eligible patients. Rather than recommending a specific option for everyone, the ACS suggests that, whenever possible, patients participate in a shared decision-making process. In this model, patients are presented with information about each of the screening options, such as accuracy, cost, potential for prevention, discomfort, and risk. Individuals then select the option that reflects their personal values and preferences.
An additional issue that is beginning to gain attention is when to stop screening. Although it is clear that individuals with significant co-morbidity that would preclude treatment should not be screened, there is little evidence or experience regarding an age at which screening should be discontinued. At this time, as long as an individual is in good health, he or she will likely benefit from screening.
Unfortunately, there are several obstacles to this shared decision-making approach. Logistic barriers, availability of well trained personnel, the time it takes to explain options to patients, cost, and insurance coverage frequently restrict options. Because of these circumstances, clinicians may be able to successfully implement only one or two of the screening modalities, limiting options for patients. As familiarity and screening skills grow in the broader medical community, and as insurance and cost obstacles are removed, a greater range of options will be available for more patients. Pignone and colleagues, for example, have shown that decision aids consisting of educational videos, brochures, and chart markers increased the ordering and per-formance of colorectal cancer screening tests.60a
Of primary importance at this time is that clinicians recommend at least one of the appropriate screening options for all of their eligible patients. Evidence demonstrates that when a screening recommendation comes directly from the clinician, compliance with colorectal cancer screening can be quite high.61 The ACS and other organizations are preparing materials to assist clinicians in informing patients about colorectal cancer screening options, and detailed information about the characteristics of colorectal cancer screening tests—and testing in the primary care setting—will be published in CA later this year.
Future developments will ex-pand the methods now available for the early detection of colorectal neoplasia. Lower dose, higher-resolution CT imaging of the bowel, for instance, is continuing in development.33,62,63 Genetic testing for mutations present in neoplastic cells excreted in feces is now also being studied as a screening method.34,35,59,64 Although these newer methods hold promise, it will likely be several years before there is adequate accumulation of evidence to determine their clinical effectiveness. Thus, although we anticipate the introduction of new screening modalities, it is not advisable to postpone screening using currently available methods.
We hope that these new ACS guidelines on screening and surveillance for the early detection of colorectal adenomas and cancer—which will be updated in the future, as new scientific data become available—will be useful for both clinicians and the general public.
Although we do not yet have randomized, controlled trial data to precisely compare the costs, risks, and benefits of the various methods for screening, a growing consensus indicates that there is a compelling case for screening using any of several methods or combinations of methods now available. By applying the knowledge we already have, it is likely that the majority of deaths from colorectal cancer in the US can be prevented.
|
American Cancer Society Guidelines on Testing for Early Endometrial Cancer Detection—Update 2001 |
|---|
|
TOPABSTRACTINTRODUCTIONRecommendations from the...RECENT DATA ON PROSTATE...ACS GUIDELINE ON TESTING...American Cancer Society...DEFINITIONS: EARLY DETECTION...ACS COLORECTAL CANCER SCREENING...CHANGES FROM THE 1997...American Cancer Society...ACS GUIDELINES ON TESTING...SUMMARYTesting for Early Lung...INFORMED DECISION-MAKING ABOUT...CANCER SCREENING IN THE...PARTICIPATION IN SCREENING--...CONCLUSIONREFERENCES |
|---|
Endometrial cancer is uncommon before age 40; the incidence increases with increasing age, peaking between ages 75 and 79, with a median age at diagnosis of 66.6 When endometrial cancer is diagnosed while still localized, five-year survival is 96%, compared with 77% for regional disease, and 44% for disease with distant metastasis. A majority of cases (77%) are diagnosed while still localized.6
Guidelines Development
The ACS guidelines for the early detection of endometrial cancer were last reviewed in October 1992, and recommended revisions were approved by the ACS Board of Directors in November 1992.65 At that time, it was recommended that all women at increased risk for endometrial cancer consider tissue sampling of the endometrium, beginning at menopause. Factors associated with increased endometrial cancer risk included infertility, failure to ovulate, obesity, abnormal uterine bleeding, unopposed estrogen therapy, and tamoxifen therapy. Endometrial tissue sampling at various intervals after menopause was indicated depending on the degree of risk and other factors as determined by the physician.
In 2000, the ACS convened an expert panel to review the existing guideline for the early detection of endometrial cancer in asymp-tomatic women in the context of evidence that has accumulated since the last revision.
During the current guideline review, published articles related to endometrial cancer screening, risk, and risk factors were identified using MEDLINE (National Library of Medicine) for the years 1991 through 2000, bibliographies of identified articles, and from the personal files of the panel members. The database was searched using key words including neoplasia, endometrium, early detection, as well as words related to each of the individual risk factors. Panel members reviewed the papers related to one or more assigned risk factors based on their area of expertise. Members met via conference call from May through October 2000 to discuss the data and recommendations. Each member reviewed the draft manuscript and provided suggestions for revisions.
Definitions: Early Detection Methods
Endometrial cancer may be detected through a variety of means. While screening for endometrial cancer has been evaluated in prospective studies, the efficacy of endometrial screening has never been evaluated in a large prospective randomized controlled trial.
Today, the endometrial biopsy is the most common technique used to obtain endometrial tissue, and evaluation of endometrial histology has been the gold standard for determining the status of the endometrium.66
Cooper and colleagues have argued that hysteroscopy-directed biopsy is superior to endometrial biopsy in the collection of endometrial tissue samples, but for individuals not trained or without access to equipment, in-office endometrial biopsy without visual control is regarded as a reasonable approach for diagnostic sampling.67 Endometrial biopsy is easily performed as an office procedure and has good sensitivity, with the small number of false negatives most likely a result of sampling error. The observation that the positive predictive value of endometrial biopsy in average-risk women and in asymptomatic women is low, while higher yields are seen in women in specific risk groups, has led to the suggestion that screening may be a reasonable strategy in women at increased risk.68 More recently, transvaginal ultrasound (TVU) has been used as a noninvasive screening test to evaluate the endometrium. However, as yet, there is no agreed-upon criterion for endometrial thickness that has both a high sensitivity and specificity; a high rate of false-positive results is also a limiting factor.69 Endometrial thickness can vary depending upon whether the woman is pre- or postmenopausal, is on replacement therapy (estrogen replacement therapy versus hormone replacement therapy), or is taking tamoxifen. TVU may be too non-specific to be an effective screening tool.70–72
Two recent studies reported disappointing results for endometrial cancer screening with endometrial biopsy and TVU for asymptomatic women on tamoxifen therapy,70,73 and each commented on the harms associated with diagnostic assessments, including unnecessary biopsies and, in some cases, uterine perforations. In an accom-panying editorial, Runowicz concluded that the current data do not warrant routine screening for endometrial cancer in asymp-tomatic women on tamoxifen therapy.74
Historically, the Pap test has led to the early diagnosis of some endometrial cancers, although detection is primarily incidental. Evaluations of the potential of the Pap test to be a test for both cervical cancer and endometrial cancer have shown both sensitivity and positive predictive value to be too low for the Pap test to be a practical screening test for endometrial cancer.66 However, abnormal endometrial cells on the Pap test may be markers for increased risk, especially when they are present in the secretory phase of the menstrual period,75 or when the patient is postmenopausal.76
It must be remembered that screening refers to the evaluation of the asymptomatic patient. When bleeding occurs, evaluation becomes diagnostic rather than screening. A history of bleeding or demonstrated radiographic evidence of endometrial pathology removes a patient from the realm of screening and demands investigation, i.e., an endometrial biopsy.
Risk Factors
The expert panel reviewed the factors generally believed to place women at increased risk of endometrial cancer: Increasing age, unopposed estrogen therapy, late menopause, tamoxifen therapy, nulliparity, infertility or failure to ovulate, obesity, hypertension, diabetes, and HNPCC. In addition, the potential benefit of screening average-risk women for endometrial cancer was examined.
Hormone Replacement Therapy
Unopposed estrogen therapy in women with an intact uterus increases the risk of endometrial cancer two- to 10-fold, with an average relative risk of 4 to 5, and risk increases with duration of use.77–82 Unopposed estrogen therapy is not recommended for women with an intact uterus for this reason. The addition of progestin to estrogen, either cyclic or continuous, reduces the risk of endometrial cancer to either the expected risk of women not taking HRT, or in some studies, to lower than expected.78–84 The length of cyclic progestin therapy appears important, with at least 10 days (preferably 14 days) necessary to decrease the risk associated with unopposed estrogen therapy.80,84 Cyclic therapy promotes regular withdrawal bleeding much like a menstrual period. Any other bleeding should be considered abnormal and evaluated accordingly. Continuous estrogen and pro-gestin therapy should lead to amenorrhea in most women.
Late Menopause
Late menopause, defined as menopause occurring after age 55, increases risk of endometrial cancer two-fold.77,85–88 Of greater concern is that peri- or post-menopausal women may report as menses what is actually abnormal bleeding. Bleeding after age 55 should be treated as symptomatic bleeding and not as menstruation.
Tamoxifen Therapy
In various studies, the relative risk of endometrial cancer in women taking tamoxifen ranged from 1.0 to 7.5.89–97 Although some case control studies have not noted an effect on risk,94,97,98 prospective data have shown an elevated risk for endometrial cancer in women who had or who were at high risk for breast cancer and were taking tamoxifen, with annual risk increasing by two in 1,000.90
Nulliparity & Infertility
Overall, nulliparity appears to confer an approximately two-fold risk of developing endometrial cancer compared with parity of one or more.88,99–108 Several studies have shown a marked decrease in risk associated with a greater number of full-term pregnancies.88,100,102–104,107 The literature does not support a significant risk of endometrial cancer associated with infertility, although some studies have shown an elevated risk.85,86,101,109
Polycystic Ovary Syndrome
Only 5% of endometrial cancers are diagnosed in women younger than age 40. Case reports have suggested that polycystic ovary syndrome (PCOS), which affects approximately 5% of reproductive aged women, may be a risk factor for endometrial cancer in younger women.110–112 Coulam et al observed a three-fold relative risk of developing endometrial cancer after a diagnosis of chronic anovulation.113 However, according to Dahlgren et al, patients with PCOS are also more likely to have other risk factors for endometrial cancer, such as histo-ries of nulliparity, than population-based controls.114 Although the data suggest that young patients with a history of PCOS may be at increased risk for endometrial cancer, the existing evidence is limited and is poorly controlled for known confounding risk factors.
Obesity & Hypertension
Studies show that very obese women have a two-to-four-fold relative risk of endometrial cancer.115–122 Obesity is defined as body weight over 200
