Scandinavian Journal of Gastroenterology
ISSN: 0036-5521 (Print) 1502-7708 (Online) Journal homepage: http://www.tandfonline.com/loi/igas20
Screening Strategies in Gastrointestinal Cancer R. H. Riddell To cite this article: R. H. Riddell (1990) Screening Strategies in Gastrointestinal Cancer, Scandinavian Journal of Gastroenterology, 25:sup175, 177-184, DOI: 10.3109/00365529009093141 To link to this article: http://dx.doi.org/10.3109/00365529009093141
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 5
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=igas20 Download by: [RMIT University Library]
Date: 22 March 2016, At: 15:10
Screening Strategies in Gastrointestinal Cancer R. H. RIDDELL Dept. of Pathology, McMaster University Medical Centre, Hamilton. Canada
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Riddell RH. Screening strategies in gastrointestinal cancer. Scand J Gastroenterol 1990, SUPPI PI 175), 177-184 The concept of s c r e e ~ n gis based on the notion that regular examination can reduce the mortality from gastrointestinal cancer, but there is as yet little evidence that this is the case in either Barrett’s esophagus or ulcerative colitis. Screening is carried out by regular endoscopy with multiple biopsies in an attempt to detect dysplasia, which has traditionally been regarded as the end point of surveillance and the time when resection should be considered. Considerable departure from these guidelines has occurred. Attempts are increasingly being made to select patients in both of these groups who might be at particular high risk of developing dysplasia because of the presence of nuclear aneuploidy, the assumption being that these patients are at particular risk of developing dysplasia and carcinoma. However, data to support stratification at this level are unavailable. In addition, there is not good agreement as to whether surveillance is indicated at all in some groups of patients such as those with less than total colitis or short-segment Barrett’s esophagus. The end point of surveillance has also been questioned, with reluctance to recommend colectomy in ulcerative colitis with the finding of low-grade dysplasia, possibly because the sampling problem is so great that it may be impossible to confirm that diagnosis on repeated endoscopy and biopsy. While better markers than dysplasia may be required to predict patients at highest risk, there are currently no marker other than the development of invasive carcinoma on which the decision to resect can be based. In Barrett’s esophagus the current trend is to delay esophagectomy until at least intramucosal carcinoma is demonstrated in view of the mortality from esophagectomy, which is considered in many centers too high to justify a prophylactic resection. Less conventional endoscopic methods of destroying foci of dysplasia are also being tried, but data on their utility are lacking. The whole concept of surveillance remains beset with numerous problems, including cost, causing the practicality of surveillance to be questioned, but alternative constructive suggestions in these patients have not emerged.
Key words: Aneuploidy; Barrett’s esophagus; cancer; dysplasia; screening strategies; surveillance; ulcerative colitis Robert H . Riddell, Ph. D.,Dept. of Pathology, McMaster University Medical Centre, Hamilton, Canada
The aim of screening is to reduce the mortality, and, possibly, also the morbidity, of gastrointestinal cancer, by detection and intervention either in an early ‘curable’ stage or even before the invasive stage of the tumor has been reached. The subject of screening strategies in gastrointestinal cancer is extensive and varies geographically with regard to the type and site of tumor. It may also vary with local interest and expertise, both of which influence referral patterns.
Geographically, there are well-recognized areas in which some tumors dominate with regard to both prevalence and mortality, and these include hepatoma and esophageal, gastric, and colorectal cancer. It is therefore possible to approach screening on two levels. The first concerns those associated with what is regarded locally as a major problem with the population at large well aware of the morbidity and mortality of the disease. This may be reflected by some form of local ‘screening program’ aimed at either
Downloaded by [RMIT University Library] at 15:10 22 March 2016
118
R. H.Riddell
the entire population, such as cytologic screening in parts of China for esophageal cancer or endoscopic or radiological imaging in Japan for the early detection of gastric cancer. It may also involve the early detection of a well-defined subpopulation at risk who can subsequently be carefully monitored. The identification of a high-risk marker for this subpopulation varies with the disease and might include hepatitis B for hepatoma, smoking/alcohol for esophageal squamous carcinoma, or a family history, the presence of occult blood in the stool, and adenomas for colorectal carcinoma. At a second level are diseases that occur less frequently on a national or geographic basis but which are sufficiently prevalent and the problem sufficiently complex to still require expertise not generally readily available; this results in patients at risk being referred. This level might include patients with Barrett’s esophagus at risk of developing adenocarcinoma, postgastrectomy patients at risk of developing gastric cancer, or ulcerative colitis and familial adenomatous or juvenile polyposis patients at risk of developing colorectal cancer. It might also include a local high-risk population such as that of the Veterans Administration hospitals in the USA for both adenocarcinoma and squamous carcinoma of the esophagus and, possibly, hepatocellular carcinoma. There is a dynamic element in the evolution of new diseases or risk groups, which remains a problem whose seriousness is unknown. These include the increased risk of patients with celiac disease developing small-intestinal lymphoma, carninoma, and possibly squamous carcinoma of the esophagus; that of HIV-positive patients developing anal cancer; and the apparent increase in adenocarcinoma of the gastroesophageal junction. This paper examines and compares the problems of screening for carcinoma arising in Barrett’s epithelium or in chronic ulcerative colitis. FACTORS GOVERNING SCREENING Screening is dependent on various factors for it to be successful, the most obvious being that the increased risk of developing carcinoma in a defined population and that when disease presents
clinically it is frequently so advanced that the prognosis is poor and unacceptable. The corollary is that surveillance will result in tumors being diagnosed at a less advanced clinicopathologic stage, and which, if treated, will result in a lower mortality from the tumor. Screening also depends on the presence of a readily detectable marker that can be utilized as an end point for screening, such as the presence of dysplasia of a specific grade, or even of invasive carcinoma if there is a reasonable chance that it is at an early pathologic stage. In the gastrointestinal tract this effectively means invasive carcinoma not extending beyond the muscularis propria and ideally confined to the mucosa or submucosa. Further, the effect of a screening program should not be the result of an artefact such as lead time bias, in which the diagnosis is made producing artificially long survival times. This implies that tumors must remain localized, whether in an invasive or non( pre)invasive state, until diagnosed so that local therapy is effective. In the case of breast cancer the question of whether early detection achieves this goal is still debated vigorously. In contrast, esophageal and large-bowel cancer show a good relationship between stage and prognosis, but virtually all disease-related deaths occur in the first few years after excision, so that deaths from disease are rare after 5 years, and even 2- or 3-year survival curves give a very good prediction of survival.
SELECTION OF PATIENTS AT RISK It is correct but ultimately simplistic to state that patients with extensive or total ulcerative colitis, or Barrett’s esophagus, are at increased risk of developing colorectal carcinoma. Ulcerative colitis In ulcerative colitis the subgroup of patients at highest risk are those with ‘extensive’ of total involvement of the large bowel who have had the disease for at least 7 years. Patients in this group have a lifetime risk of the order of 25-50% of developing colorectal carcinoma (1-11). However, there are various subgroups whose risk of
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Screening in Gastrointestinal Cancer
colorectal cancer is unknown, including those with less extensive colitis, such as patients with proctosigmoidits, left-sided colitis, or disease extending into the transverse colon. This poses the question of whether the increased risk is proportional to the extent of disease. In any of these groups it is unclear whether microscopic disease implies the same risk as endoscopically or radiologically defined disease. Also, are patients with proctosigmoiditis whose disease over a 15-year period extends proximally at the same risk, and at what point should surveillance begin? Although there are currently no answers to these questions, 5-6% of the North American population will develop colorectal cancer, and the rectosigmoid will be affected in about three-quarters. There is therefore a need to demonstrate that in patients with proctosigmoiditis or left-sided colitis there is an excess risk over and above that expected. Barrett’s esophagus In Barrett’s esophagus the main question centers on the basis for diagnosis. The definition of the length of glandular mucosa extending 3 cm or more proximally up the tubular esophagus (12) suffers from the major disadvantage that between 1and 3 cm of Barrett’s epithelium may be present but remain undiagnosed. This has raised the question of what constitutes so-called short-segment Barrett’s esophagus and whether this is also at increased risk of developing carcinoma. Furthermore, there are no data by which to judge whether the risk of cancer in Barrett’s esophagus starts at day 1, or whether that is a latent period in which the risk from carcinoma increases. The major problem is that, unlike ulcerative colitis, the onset of Barrett’s esophagus has no clinical features by which it can be gauged. There is also no indication as to whether the risk of cancer is proportional to the anatomic extent of Barrett’s esophagus. Fortunately, from a diagnostic viewpoint there is an increasing trend to consider the presence of incomplete or complete intestinal metaplasia as the sole criterion for the diagnosis, provided that it can be established by biopsy and that there is no concurrent metaplasia at the gastric cardia. Surprisingly, this seems virtually never to be the case (13).
179
SCREENING Once a population at risk has been defined, it is necessary to institute screening and, if possible, define whether all patients should receive similar surveillance procedures or whether some form of stratification can be introduced because of different risk factors. NEED FOR A REPRODUCIBLE MARKER The use of dysplasia as a marker in both Barrett’s esophagus and ulcerative colitis has many problems. The first is that of sampling, particularly in ulcerative colitis; the second that of reproducibility; the third its use as the end point of screening. In Barrett’s esophagus, very early invasive carcinoma may be the most justifiable and rational indication for resection.
THE SAMPLING PROBLEM IN BARRETT’S ESOPHAGUS AND ULCERATIVE COLITIS There is a marked contrast when the area of mucosa under surveillance in Barrett’s esophagus is compared with that in ulcerative colitis. In Barrett’s esophagus the surface area of mucosa being screened is only a fraction of that in ulcerative colitis and rarely exceeds 100 cm2. Numerous biopsy specimens can be taken within a very small distance, so that sampling problems are minimal; indeed, one standard is to take fourquadrant specimens every 1-2 cm. In ulcerative colitis the standard practice is to take one or two specimens every lOcm of bowel, so that if the circumference of the bowel is 10cm. only 510mm2 of mucosa would be sampled for every 100 cm2 of mucosa, a tremendous sampling problem. Indeed, it would take 25 uniformly spaced specimens within a 10-cm length of large bowel to guarantee detection of a 2-cm diameter patch of dysplasia. INTERPRETATION O F DYSPLASIA The morphologic diagnosis of dysplasia still has problems. Despite the best efforts of an international panel of pathologists there is still inter-
180
R . H . Riddell
(and intra-) observer variability. However, these are relatively uncommon at each end of the dysplasia spectrum (negative for dysplasia and high-grade dysplasia), but like any spectrum arbitrarily and subjectively divided, areas of overlap inevitably occur in the mid-spectrum. There is also no clinical agreement on the criteria for the timing of colectomy both between institutions and also within them.
Downloaded by [RMIT University Library] at 15:10 22 March 2016
THE END POINTS OF SURVEILLANCE Barrett’s esophagus The major limiting factor for recommending resection of Barrett’s esophagus is that few data are available on the likelihood of invasive carcinoma being present when either low- or highgrade dysplasia has been demonstrated on biopsy. The same criteria probably apply to Barrett’s esophagus as to ulcerative colitis-namely that invasive carcinoma can arise directly from dysplasia of any grade and that it is only by histologic examination of the resected esophagus that invasion can be excluded. In practice, it is relatively uncommon for invasion to be present without a recognizable endoscopic abnormality that proves to be at least dysplastic on biopsy. The second major problem is the operative mortality for esophagectomy, which varies between 5% and 20%, with most centers reporting results in the 10-15% range. While this may be an acceptable operative mortality in patients with proven invasive adenocarcinoma, particularly if there is a reasonable chance for cure, it is unacceptable for a prophylactic operation. In one combined study from two major institutions in the USA, seven patients with high-grade dysplasia underwent resection. None had invasive carcinoma in the resected specimen, but two died in the postoperative period (14). The trend in Barrett’s esophagus is therefore to study the patient closely with endoscopy and biopsy until evidence of or an extremely strong clinical suspicion of invasion is found. This might be the presence of invasive carcinoma, at least into the lamina propria (intramucosal carcinoma). The operation then becomes one for invasive carcinoma rather than a prophylactic one, and an
increased mortality becomes more acceptable, particularly because the bulk of these carcinomas are intramucosal or submucosal histologically, rarely have nodal metastases, and consequently tend to have a very good prognosis once the patient is home (15). However, attempts to eradicate endoscopically areas of invasive adenocarcinoma or dysplasia, primarily of high grade, are occasionally attempted both to provide local control and even cure and to circumvent the mortality associated with reactions. Patients who are at poor operative risk stand to gain most from such unconventional treatment. Ulcerative colitis Two decades ago the ‘treatment’ for patients with total ulcerative colitis was to undergo panproctocolectomy as a preventative measure to protect against the subsequent development of colorectal cancer. In 1967 the finding of the morphologic change of dysplasia was described in patients with colitic cancer, a change that appeared invariably to affect the rectum and could therefore be detected on sigmoidoscopy and biopsy (16). Prophylactic proctocolectomy was rapidly abandoned in favor of follow-up using rectal biopsy specimens in an attempt to detect dysplasia. The past decade has seen a series of developments aimed at fine-tuning ‘surveillance’, the object of which was initially to detect dysplasia so that proctocolectomy could selectively be offered to patients presumed to be at highest risk. It rapidly became apparent that despite surveillance patients with ulcerative colitis were still developing, and sometimes dying from, colorectal cancer. Furthermore, patients under surveillance were sometimes found on a subsequent colonoscopy to have developed invasive carcinoma, from which they sometimes died, and there is as yet no study that has shown a survival advantage for patients in a surveillance program (17). Currently, ‘surveillance’ implies looking for both colorectal cancer and dysplasia at the time of colonoscopy, which is usually carried out annually or biannually, but to date there is no predetermined ‘best’ time interval between colonoscopies. Indeed, a more appropriate question to examine would be the mortality from
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Screening in Gastrointestinal Cancer
cancer if surveillance were carried out at different time intervals, patients being randomized to groups requiring endoscopy at 6 months or at 1-, 2-, or 3-year intervals. At colonoscopy a careful search is made of the large bowel for velvety mucosal irregularities, pale plaque-like lesions, lesions that may be the endoscopic counterpart of either dysplasia or carcinoma (dysplasia-associated lesions or masses (DALM) (18)), in contrast to the usual con-colitic cancers. The presence of invasive carcinoma in these lesions may only be apparent after histologic examination of the resected specimen. In the absence of such lesions multiple biopsy specimens of the large bowel are taken-between 7 and 20-in a random search for dysplasia (see above). The end point of surveillance in ulcerative colitis is therefore a defined grade of dysplasia or a reasonable suspicion or or the presence of invasive carcinoma. Dysplasia is a reasonable end point, since the operative mortality for colectomy in the target group (who are usually less than 70 years of age) is of the order of 1% and considerably lower in those under 45 in whom elective colectomy is carried out. ALTERNATIVE APPROACHES FOR SCREENING A second approach is to determine the correlation between other factors and the subsequent development of dysplasia or carcinoma, the implication being that patients not having these factors may be able to have less frequent endoscopies, while those at increased risk may be followed as at present or offered early resection, precluding numerous subsequent endoscopies. Although there has been investigation into various potential markers, including changes in mucin, lectins, and oncogenes and their products, the most exciting potential marker in both fields is that of aneuploidy as a marker for both dysplasia and carcinoma; preliminary data suggest that this holds for both Barrett’s esophagus and ulcerative colitis. ANEUPLOIDY Aneuploidy is the presence of a population of
181
cells that have an excess quantity of DNA (greater than 2N) when measured, usually by flow cytometry. However, this is an extremely crude index of DNA content and in most studies must be at least 5% greater than normal to create a second peak in a DNA histogram. Other parameters, such as the fraction of cells in S-phase or an increase of cells in the normal mitotic (4N/tetraploid) range, are easily demonstrated. There is growing evidence that in ulcerative colitis (19-23) and Barrett’s esophagus (24-27) an aneuploid population of cells is present in both carcinomatous and dysplastic epithelium in the vast majority of patients. Aneuploidy is also found in patients with both diseases in the absence of dysplasia or carcinoma. The interpretation of this finding is critical, suggesting either that aneuploidy is a relatively non-specific event and therefore has a low positive predictive accuracy for dysplasia and carcinoma, or that it may precede these changes and therefore be a marker for patients most likely to develop dysplasia and carcinoma. Longitudinal studies are clearly required to answer these questions. There are also preliminary data to suggest that areas of dysplasia develop within considerably larger areas of aneuploidy (McHugh, 1990, personal communication). If this proves reliable, it would greatly reduce concerns about dysplasia or carcinoma being missed endoscopically, and additional attention could be given to patients with aneuploid mucosa for additional sampling, potentially reducing some of the sampling problems that occur, particularly in ulcerative colitis.
OTHER FACTORS AFFECTING THE DECISION FOR RESECTION The finding of an endoscopic lesion or dysplasia at the first colonoscopy is much more likely to be associated with an underlying invasive carcinoma (up to 67%) than if found in a patient who has had several previous endoscopies. However, some forms of colitic carcinoma are not accompanied by conventional dysplasia, at least as currently defined (28), and there are few data regarding this finding in Barrett’s esophagus,
182
R. H . Riddell
although those available support this contention (12).
Downloaded by [RMIT University Library] at 15:10 22 March 2016
REMAINING PROBLEMS WITH REGARD TO SURVEILLANCE Barrett’s esophagus Preliminary date with regard to the significance of dysplasia in Barrett’s esophagus may soon be available as the result of the Veterans Administration study in the USA, in which the detection and management of dysplasia have been carried out sufficiently frequently to generate meaningful data. A potential problem, however, is that the mean age of patients with dysplasia is quite high, increasing the expected operative mortality and tending to preclude operative intervention. These important results are eagerly awaited.
7. Dysplasia and carcinoma may be accompanied or preceded by loss of some markers of cell maturation, extension of markers of cell immaturity up to large-bowel crypts, changes in cell proliferation markers, and by changes in expression of some oncogenes or their products. The real questions are therefore whether advances can be made in the determination of which patients are really at risk and whether the endoscopic detection of dysplasia and colitic cancer can be increased to the point that patients derive obvious benefit from a surveillance program as measured by increased survival or a decreased numbers of cancers. The notion of managing each patient on an individual basis becomes a euphemism for ensuring permanent lack of reproducible criteria that can be subjected to analysis.
Ulcerative colitis DATA REQUIRED FROM FOLLOW-UP At an international workshop held to review STUDIES the knowledge of dysplasia and cancer in ulcerative colitis sponsored by the National Foun- There is an obvious lack of data from surveillance dation for Ileitis and Colitis (New York), the studies to answer the basic question of whether Meyerhoff Foundation (Johns Hopkins Hos- surveillance reduces mortality. Important seconpital), and the NIH (USA), held in the USA in dary questions also require answers in order that the risks and benefits of resection can be weighed June 1989, several points became apparent: 1. Enormous resources are spent on sur- against those of doing nothing under appropriate clinical circumstances if a patient presents with: veillance with little obvious benefit. 2. The extent of colitis has no standard method a) low-grade dysplasia (LGD) or b) high grade of definition, variously being established by dysplasia (HGD) either c) with or d) without an barium enema (single or air contrast), endoscopy, endoscopic abnormality from which the biopsy or biopsy, so that other than those in whom total has been taken both at e) first (diagnostic) and disease has been defined radiologically, the pre- at f ) subsequent (surveillance) colonoscopy or upper endoscopy. cise population at risk is unknown. Although the major variables include the grade 3. Dysplasia may be focal, so that random of dysplasia, the presence of an endoscopic abnorbiopsies present major sampling problems. 4.There has been no formal study of the endo- mality, and the detection of dysplasia at first or scopic appearance of dysplasia or of colitic car- subsequent colonoscopy, there are other variables that will ultimately be clarified. The risk of carcinoma. 5. Dysplasia is far from the ultimate marker of cinoma will also need to be subdivided into carcinoma and is not detected in up to a third patients with carcinoma g) at the time of presof patients under surveillance in whom clinical entation or h) at follow-up study of elective surgery is not carried out at presentation-that is, invasive carcinoma subsequently develops. 6. Dysplasia and carcinoma in some patients rate of inmeaselyear over and above that in are preceded by the development of an aneuploid group g. Some form of a control group has to be concell population in the absence of any histologic sidered in the final analysis, because the objective abnormality .
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Screening in Gastrointestinal Cancer
183
of surveillance is to prevent untimely deaths from ensure that the screening program has the desired carcinoma. Some thought will need to be given effectiveness and efficacy. These strategies are examined in the selected to the mortality, and possibly morbidity, of cancer when it found as a result of surveillance as areas of esophageal carcinoma arising in Barrett’s opposed to that in patients not undergoing sur- epithelium and carcinoma in ulcerative colitis. veillance. Years of life ‘saved’ from early detec- There are marked similarities between these tion when compared with the loss from operative groups including: problems in accurately defining mortality must be compared. Essentially, patients the population at risk; the screening method of undergoing surveillance in whom invasive car- choice, which appears to be endoscopy with multicinoma is found incidentally at resection but who ple biopsies; the presence of an aneuploid cell do not die of the disease represent a successful population may select patients at risk, which may use of surveillance, but only if their overall mor- allow stratification of the timing of endoscopy; tality is better than that of patients presenting there is considerable controversy regarding the symptomatically with cancer or than a control timing of excision; there are no data to support a population. Such controls are unethical unless reduction in mortality by screening; debate patients are medically unfit for surgery or refuse regarding whether the cost is justifiable; there is screening or resection, irrespective of what is no guarantee that surveillance will prevent the development of carcinoma or detect it at a curable found. Data generated will be more credible if com- stage of disease; and the risks and the benefits of parisons between subgroups can be carried out to screening need to be weighed against the risks document whether statistically significant dif- and the benefits of no surveillance. However, there are also marked dissimilarities ferences can be demonstrated between these subbetween the two: the area of mucosa at risk in groups. Barrett’s esophagus is far smaller than that in ulcerative colitis, making detection of areas of COST dysplasia and possible foci comparatively easy; Although screening provides a major problem at these may therefore be far easier to keep under the national level, the real questions are not the surveillance. There is a much higher mortality for overall cost but the cost to any individual, and esophagectomy than colectomy, which currently not how much does it cost potentially for each changes the end point for resection. Conversely, year a life saved; how can one estimate whether there is a much higher morbidity associated with this is worthwhile? Is it justified if patients bear pouch operations if they are carried out after this cost themselves despite the limitations of the colectomy. techniques? REFERENCES
SUMMARY Screening is a four-stage process with strategies that invariably include a) the selection or detection of a population at high risk for a specific disease; b) the institution of a screening procedure within that high-risk population, which will detect patients at either a pre-invasive or early invasive stage of the disease, when it is remediable to therapy, usually surgical excision; c) the carrying out of the appropriate therapeutic maneuvre when the marker becomes positive at an appropriate time; and d) follow-up and data analysis to
1 . Rosenstock E, Farmer RG, Petras R, et al. Surveillance for colonic carcinoma in ulcerative colitis. Gastroenterology 1985, 89, 1342-1346 2. Nugent FW, Haggitt RC, Colcher H, Kutteruf GC. Malignant potential of chronic ulcerative colitis. Preliminary report. Gastroenterology 1979.26, 1-5 3. Nugent FW, Haggitt RC. Cancer in ulcerative colitis: results of a 13 year prospective biopsy surveillance program. Gastroenterology 1988, 94, A327 4. Granqvist S, Gabrielsson N, Sundelin P, Thor-
geirsson T. Precancerous lesions in the mucosa in ulcerative colitis. Scand J Gastroenterol 1980, 15, 289-296 5. Dickinson RJ,Dixon HF, Axon ATR. Colonoscopy and the detection of dysplasia in patients with longstanding ulcerative colitis. Lancet 1980, 2, 620-622
Downloaded by [RMIT University Library] at 15:10 22 March 2016
184
R. H . Riddell
6. Lennard-Jones JE, Morson BC, Ritchie JK, Williams CB. Cancer surveillance in ulcerative colitis. Lancet 1983, 2, 149-152 7. Manning AP, Bulgim OR, Dixon MF, Axon ATR. Screening by colonoscopy for colonic epithelial dysplasia in inflammatory bowel disease. Gut 1987, 28, 1489-1494 8. Lashner BA, Hanauer SB, Silverstein MC. Optimal timing for colonoscopy to screen for cancer in ulcerative colitis. Ann Intern Med 1988, 108, 274-278 9. Melville DM, Lennard-Jones JE, Morson BC, Ritchie JK, Williams CB. Surveillance in extensive ulcerative colitis improves prognosis without always preventing cancer. Gut 1988, 29, A1494 10. Rutegard J, Ahsgren L, Sterling R, Janunger KG. Ulcerative colitis. Cancer surveillance in an unselected population. Scand J Gastroenterol 1988, 23, 139-145 11. Gyde SN, Prior P, Allan RN, et al. Colorectal cancer in ulcerative colitis: A cohort study of primary referrals from three centres. Gut 1988, 29, 2 0 6 2 17 12. Skinner DB, Walther BC, Riddell RH, et al. Barrett’s esophagus. Comparison of benign and malignant cases. Ann Surg 1983, 198,554-566 13. DeNardi F, Riddell RH. The normal esophagus. Am J Surg Pathol (in press) 14. Reid BJ, Lewin KJ, Haggitt RC, et al. Endoscopic biopsies can detect high grade dysplasia or early adenocarcinoma in Barrett’s esophagus without grossly recognizable neoplastic lesions. Gastroenterology 1988, 94, 81-90 15. Skinner DB, Walter BC, Riddell RH, Schmidt H, Iascone C, De Meester T. Barrett’s esophagus: comparison of benign and malignant cases. Ann Surg 1983, 198, 554-566 16. Morson BC, Pang LSC. Rectal biopsy as an aid to cancer control in ulcerative colitis. Gut 1967, 8, 423-434 17. Collins RH, Feldman M, Fordtran JS. Colon cancer, dysplasia, and surveillance in patients with ulcerative colitis. N Engl J Med 1987, 316, 16541658 18. Blackstone MO, Riddell RH, Rogers BHG, Levin B. Dysplasia-associated lesion or mass (DALM)
detected by colonoscopy in long-standing ulcerative colitis: an indication for colectomy. Gastroenterology 1981, 80, 366-374 19. Hammarberg C, Slezak P, Tribukait B. Early detection of malignancy in ulcerative colitis-a flow cytometric DNA study. Cancer 1984, 53, 291-295 20. Fozard JBJ, Quirke P, Dixon MF, et al. DNA aneuploidy in ulcerative colitis. Gut 1986,27, 14141418 21. Lofberg R, Tribukait B, Ost A, et al. Flow cytometric DNA analysis in longstanding ulcerative colitis: a method of prediction of dysplasia and carcinoma development? Gut 1987, 28, 1100-1106 22. Melville DM, Jass RJ, Shepherd NA, et al. Dysplasia and deoxyribonucleic acid aneuploidy in the assessment of precancerous changes in chronic ulcerative colitis. Gastroenterology 1988, 95, 668675 23. Rutegard J, Ahsgren L, Stenling R, Roos G. DNA content in ulcerative colitis: flow cytometric in a patient series from a defined area. Dis Colon Rectum 1988, 31, 710-715 24. McKinely MJ, Budsman DR, Gruenberg D, Bronzo RL, Weissman GS, Kahn E. DNA content in Barrett’s esophagus and esophageal malignancy. Am J Gastroenterol 1987, 82, 1012-1015 25. Reid BJ, Haggitt RC, Rubin CE, Rabinovitch PS. Barrett’s esophagus: Correlation between flow cytometry and histology in detection of patients at risk for adenocarcinoma. Gastroenterology 1987,93, 111 26. Rabinovitch PS, Reid BJ, Haggitt RC, Norwood TH, Rubin CE. Progression to cancer in Barrett’s esophagus is associated with genomic instability. Lab Invest 1988, 60, 65-71 27. Fennerty MB, Sampliner RE, Way D, Riddell R, Steinbronn K, Garewal HS. Discordance between flow cytometric abnormalities and dysplasia in Barrett’s esophagus. Gastroenterology 1989, 97, 815820 28. Riddell RH, Goldman H, Ransohoff DF, et al. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol 1483, 14, 931-968
ISSN: 0036-5521 (Print) 1502-7708 (Online) Journal homepage: http://www.tandfonline.com/loi/igas20
Screening Strategies in Gastrointestinal Cancer R. H. Riddell To cite this article: R. H. Riddell (1990) Screening Strategies in Gastrointestinal Cancer, Scandinavian Journal of Gastroenterology, 25:sup175, 177-184, DOI: 10.3109/00365529009093141 To link to this article: http://dx.doi.org/10.3109/00365529009093141
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 5
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=igas20 Download by: [RMIT University Library]
Date: 22 March 2016, At: 15:10
Screening Strategies in Gastrointestinal Cancer R. H. RIDDELL Dept. of Pathology, McMaster University Medical Centre, Hamilton. Canada
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Riddell RH. Screening strategies in gastrointestinal cancer. Scand J Gastroenterol 1990, SUPPI PI 175), 177-184 The concept of s c r e e ~ n gis based on the notion that regular examination can reduce the mortality from gastrointestinal cancer, but there is as yet little evidence that this is the case in either Barrett’s esophagus or ulcerative colitis. Screening is carried out by regular endoscopy with multiple biopsies in an attempt to detect dysplasia, which has traditionally been regarded as the end point of surveillance and the time when resection should be considered. Considerable departure from these guidelines has occurred. Attempts are increasingly being made to select patients in both of these groups who might be at particular high risk of developing dysplasia because of the presence of nuclear aneuploidy, the assumption being that these patients are at particular risk of developing dysplasia and carcinoma. However, data to support stratification at this level are unavailable. In addition, there is not good agreement as to whether surveillance is indicated at all in some groups of patients such as those with less than total colitis or short-segment Barrett’s esophagus. The end point of surveillance has also been questioned, with reluctance to recommend colectomy in ulcerative colitis with the finding of low-grade dysplasia, possibly because the sampling problem is so great that it may be impossible to confirm that diagnosis on repeated endoscopy and biopsy. While better markers than dysplasia may be required to predict patients at highest risk, there are currently no marker other than the development of invasive carcinoma on which the decision to resect can be based. In Barrett’s esophagus the current trend is to delay esophagectomy until at least intramucosal carcinoma is demonstrated in view of the mortality from esophagectomy, which is considered in many centers too high to justify a prophylactic resection. Less conventional endoscopic methods of destroying foci of dysplasia are also being tried, but data on their utility are lacking. The whole concept of surveillance remains beset with numerous problems, including cost, causing the practicality of surveillance to be questioned, but alternative constructive suggestions in these patients have not emerged.
Key words: Aneuploidy; Barrett’s esophagus; cancer; dysplasia; screening strategies; surveillance; ulcerative colitis Robert H . Riddell, Ph. D.,Dept. of Pathology, McMaster University Medical Centre, Hamilton, Canada
The aim of screening is to reduce the mortality, and, possibly, also the morbidity, of gastrointestinal cancer, by detection and intervention either in an early ‘curable’ stage or even before the invasive stage of the tumor has been reached. The subject of screening strategies in gastrointestinal cancer is extensive and varies geographically with regard to the type and site of tumor. It may also vary with local interest and expertise, both of which influence referral patterns.
Geographically, there are well-recognized areas in which some tumors dominate with regard to both prevalence and mortality, and these include hepatoma and esophageal, gastric, and colorectal cancer. It is therefore possible to approach screening on two levels. The first concerns those associated with what is regarded locally as a major problem with the population at large well aware of the morbidity and mortality of the disease. This may be reflected by some form of local ‘screening program’ aimed at either
Downloaded by [RMIT University Library] at 15:10 22 March 2016
118
R. H.Riddell
the entire population, such as cytologic screening in parts of China for esophageal cancer or endoscopic or radiological imaging in Japan for the early detection of gastric cancer. It may also involve the early detection of a well-defined subpopulation at risk who can subsequently be carefully monitored. The identification of a high-risk marker for this subpopulation varies with the disease and might include hepatitis B for hepatoma, smoking/alcohol for esophageal squamous carcinoma, or a family history, the presence of occult blood in the stool, and adenomas for colorectal carcinoma. At a second level are diseases that occur less frequently on a national or geographic basis but which are sufficiently prevalent and the problem sufficiently complex to still require expertise not generally readily available; this results in patients at risk being referred. This level might include patients with Barrett’s esophagus at risk of developing adenocarcinoma, postgastrectomy patients at risk of developing gastric cancer, or ulcerative colitis and familial adenomatous or juvenile polyposis patients at risk of developing colorectal cancer. It might also include a local high-risk population such as that of the Veterans Administration hospitals in the USA for both adenocarcinoma and squamous carcinoma of the esophagus and, possibly, hepatocellular carcinoma. There is a dynamic element in the evolution of new diseases or risk groups, which remains a problem whose seriousness is unknown. These include the increased risk of patients with celiac disease developing small-intestinal lymphoma, carninoma, and possibly squamous carcinoma of the esophagus; that of HIV-positive patients developing anal cancer; and the apparent increase in adenocarcinoma of the gastroesophageal junction. This paper examines and compares the problems of screening for carcinoma arising in Barrett’s epithelium or in chronic ulcerative colitis. FACTORS GOVERNING SCREENING Screening is dependent on various factors for it to be successful, the most obvious being that the increased risk of developing carcinoma in a defined population and that when disease presents
clinically it is frequently so advanced that the prognosis is poor and unacceptable. The corollary is that surveillance will result in tumors being diagnosed at a less advanced clinicopathologic stage, and which, if treated, will result in a lower mortality from the tumor. Screening also depends on the presence of a readily detectable marker that can be utilized as an end point for screening, such as the presence of dysplasia of a specific grade, or even of invasive carcinoma if there is a reasonable chance that it is at an early pathologic stage. In the gastrointestinal tract this effectively means invasive carcinoma not extending beyond the muscularis propria and ideally confined to the mucosa or submucosa. Further, the effect of a screening program should not be the result of an artefact such as lead time bias, in which the diagnosis is made producing artificially long survival times. This implies that tumors must remain localized, whether in an invasive or non( pre)invasive state, until diagnosed so that local therapy is effective. In the case of breast cancer the question of whether early detection achieves this goal is still debated vigorously. In contrast, esophageal and large-bowel cancer show a good relationship between stage and prognosis, but virtually all disease-related deaths occur in the first few years after excision, so that deaths from disease are rare after 5 years, and even 2- or 3-year survival curves give a very good prediction of survival.
SELECTION OF PATIENTS AT RISK It is correct but ultimately simplistic to state that patients with extensive or total ulcerative colitis, or Barrett’s esophagus, are at increased risk of developing colorectal carcinoma. Ulcerative colitis In ulcerative colitis the subgroup of patients at highest risk are those with ‘extensive’ of total involvement of the large bowel who have had the disease for at least 7 years. Patients in this group have a lifetime risk of the order of 25-50% of developing colorectal carcinoma (1-11). However, there are various subgroups whose risk of
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Screening in Gastrointestinal Cancer
colorectal cancer is unknown, including those with less extensive colitis, such as patients with proctosigmoidits, left-sided colitis, or disease extending into the transverse colon. This poses the question of whether the increased risk is proportional to the extent of disease. In any of these groups it is unclear whether microscopic disease implies the same risk as endoscopically or radiologically defined disease. Also, are patients with proctosigmoiditis whose disease over a 15-year period extends proximally at the same risk, and at what point should surveillance begin? Although there are currently no answers to these questions, 5-6% of the North American population will develop colorectal cancer, and the rectosigmoid will be affected in about three-quarters. There is therefore a need to demonstrate that in patients with proctosigmoiditis or left-sided colitis there is an excess risk over and above that expected. Barrett’s esophagus In Barrett’s esophagus the main question centers on the basis for diagnosis. The definition of the length of glandular mucosa extending 3 cm or more proximally up the tubular esophagus (12) suffers from the major disadvantage that between 1and 3 cm of Barrett’s epithelium may be present but remain undiagnosed. This has raised the question of what constitutes so-called short-segment Barrett’s esophagus and whether this is also at increased risk of developing carcinoma. Furthermore, there are no data by which to judge whether the risk of cancer in Barrett’s esophagus starts at day 1, or whether that is a latent period in which the risk from carcinoma increases. The major problem is that, unlike ulcerative colitis, the onset of Barrett’s esophagus has no clinical features by which it can be gauged. There is also no indication as to whether the risk of cancer is proportional to the anatomic extent of Barrett’s esophagus. Fortunately, from a diagnostic viewpoint there is an increasing trend to consider the presence of incomplete or complete intestinal metaplasia as the sole criterion for the diagnosis, provided that it can be established by biopsy and that there is no concurrent metaplasia at the gastric cardia. Surprisingly, this seems virtually never to be the case (13).
179
SCREENING Once a population at risk has been defined, it is necessary to institute screening and, if possible, define whether all patients should receive similar surveillance procedures or whether some form of stratification can be introduced because of different risk factors. NEED FOR A REPRODUCIBLE MARKER The use of dysplasia as a marker in both Barrett’s esophagus and ulcerative colitis has many problems. The first is that of sampling, particularly in ulcerative colitis; the second that of reproducibility; the third its use as the end point of screening. In Barrett’s esophagus, very early invasive carcinoma may be the most justifiable and rational indication for resection.
THE SAMPLING PROBLEM IN BARRETT’S ESOPHAGUS AND ULCERATIVE COLITIS There is a marked contrast when the area of mucosa under surveillance in Barrett’s esophagus is compared with that in ulcerative colitis. In Barrett’s esophagus the surface area of mucosa being screened is only a fraction of that in ulcerative colitis and rarely exceeds 100 cm2. Numerous biopsy specimens can be taken within a very small distance, so that sampling problems are minimal; indeed, one standard is to take fourquadrant specimens every 1-2 cm. In ulcerative colitis the standard practice is to take one or two specimens every lOcm of bowel, so that if the circumference of the bowel is 10cm. only 510mm2 of mucosa would be sampled for every 100 cm2 of mucosa, a tremendous sampling problem. Indeed, it would take 25 uniformly spaced specimens within a 10-cm length of large bowel to guarantee detection of a 2-cm diameter patch of dysplasia. INTERPRETATION O F DYSPLASIA The morphologic diagnosis of dysplasia still has problems. Despite the best efforts of an international panel of pathologists there is still inter-
180
R . H . Riddell
(and intra-) observer variability. However, these are relatively uncommon at each end of the dysplasia spectrum (negative for dysplasia and high-grade dysplasia), but like any spectrum arbitrarily and subjectively divided, areas of overlap inevitably occur in the mid-spectrum. There is also no clinical agreement on the criteria for the timing of colectomy both between institutions and also within them.
Downloaded by [RMIT University Library] at 15:10 22 March 2016
THE END POINTS OF SURVEILLANCE Barrett’s esophagus The major limiting factor for recommending resection of Barrett’s esophagus is that few data are available on the likelihood of invasive carcinoma being present when either low- or highgrade dysplasia has been demonstrated on biopsy. The same criteria probably apply to Barrett’s esophagus as to ulcerative colitis-namely that invasive carcinoma can arise directly from dysplasia of any grade and that it is only by histologic examination of the resected esophagus that invasion can be excluded. In practice, it is relatively uncommon for invasion to be present without a recognizable endoscopic abnormality that proves to be at least dysplastic on biopsy. The second major problem is the operative mortality for esophagectomy, which varies between 5% and 20%, with most centers reporting results in the 10-15% range. While this may be an acceptable operative mortality in patients with proven invasive adenocarcinoma, particularly if there is a reasonable chance for cure, it is unacceptable for a prophylactic operation. In one combined study from two major institutions in the USA, seven patients with high-grade dysplasia underwent resection. None had invasive carcinoma in the resected specimen, but two died in the postoperative period (14). The trend in Barrett’s esophagus is therefore to study the patient closely with endoscopy and biopsy until evidence of or an extremely strong clinical suspicion of invasion is found. This might be the presence of invasive carcinoma, at least into the lamina propria (intramucosal carcinoma). The operation then becomes one for invasive carcinoma rather than a prophylactic one, and an
increased mortality becomes more acceptable, particularly because the bulk of these carcinomas are intramucosal or submucosal histologically, rarely have nodal metastases, and consequently tend to have a very good prognosis once the patient is home (15). However, attempts to eradicate endoscopically areas of invasive adenocarcinoma or dysplasia, primarily of high grade, are occasionally attempted both to provide local control and even cure and to circumvent the mortality associated with reactions. Patients who are at poor operative risk stand to gain most from such unconventional treatment. Ulcerative colitis Two decades ago the ‘treatment’ for patients with total ulcerative colitis was to undergo panproctocolectomy as a preventative measure to protect against the subsequent development of colorectal cancer. In 1967 the finding of the morphologic change of dysplasia was described in patients with colitic cancer, a change that appeared invariably to affect the rectum and could therefore be detected on sigmoidoscopy and biopsy (16). Prophylactic proctocolectomy was rapidly abandoned in favor of follow-up using rectal biopsy specimens in an attempt to detect dysplasia. The past decade has seen a series of developments aimed at fine-tuning ‘surveillance’, the object of which was initially to detect dysplasia so that proctocolectomy could selectively be offered to patients presumed to be at highest risk. It rapidly became apparent that despite surveillance patients with ulcerative colitis were still developing, and sometimes dying from, colorectal cancer. Furthermore, patients under surveillance were sometimes found on a subsequent colonoscopy to have developed invasive carcinoma, from which they sometimes died, and there is as yet no study that has shown a survival advantage for patients in a surveillance program (17). Currently, ‘surveillance’ implies looking for both colorectal cancer and dysplasia at the time of colonoscopy, which is usually carried out annually or biannually, but to date there is no predetermined ‘best’ time interval between colonoscopies. Indeed, a more appropriate question to examine would be the mortality from
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Screening in Gastrointestinal Cancer
cancer if surveillance were carried out at different time intervals, patients being randomized to groups requiring endoscopy at 6 months or at 1-, 2-, or 3-year intervals. At colonoscopy a careful search is made of the large bowel for velvety mucosal irregularities, pale plaque-like lesions, lesions that may be the endoscopic counterpart of either dysplasia or carcinoma (dysplasia-associated lesions or masses (DALM) (18)), in contrast to the usual con-colitic cancers. The presence of invasive carcinoma in these lesions may only be apparent after histologic examination of the resected specimen. In the absence of such lesions multiple biopsy specimens of the large bowel are taken-between 7 and 20-in a random search for dysplasia (see above). The end point of surveillance in ulcerative colitis is therefore a defined grade of dysplasia or a reasonable suspicion or or the presence of invasive carcinoma. Dysplasia is a reasonable end point, since the operative mortality for colectomy in the target group (who are usually less than 70 years of age) is of the order of 1% and considerably lower in those under 45 in whom elective colectomy is carried out. ALTERNATIVE APPROACHES FOR SCREENING A second approach is to determine the correlation between other factors and the subsequent development of dysplasia or carcinoma, the implication being that patients not having these factors may be able to have less frequent endoscopies, while those at increased risk may be followed as at present or offered early resection, precluding numerous subsequent endoscopies. Although there has been investigation into various potential markers, including changes in mucin, lectins, and oncogenes and their products, the most exciting potential marker in both fields is that of aneuploidy as a marker for both dysplasia and carcinoma; preliminary data suggest that this holds for both Barrett’s esophagus and ulcerative colitis. ANEUPLOIDY Aneuploidy is the presence of a population of
181
cells that have an excess quantity of DNA (greater than 2N) when measured, usually by flow cytometry. However, this is an extremely crude index of DNA content and in most studies must be at least 5% greater than normal to create a second peak in a DNA histogram. Other parameters, such as the fraction of cells in S-phase or an increase of cells in the normal mitotic (4N/tetraploid) range, are easily demonstrated. There is growing evidence that in ulcerative colitis (19-23) and Barrett’s esophagus (24-27) an aneuploid population of cells is present in both carcinomatous and dysplastic epithelium in the vast majority of patients. Aneuploidy is also found in patients with both diseases in the absence of dysplasia or carcinoma. The interpretation of this finding is critical, suggesting either that aneuploidy is a relatively non-specific event and therefore has a low positive predictive accuracy for dysplasia and carcinoma, or that it may precede these changes and therefore be a marker for patients most likely to develop dysplasia and carcinoma. Longitudinal studies are clearly required to answer these questions. There are also preliminary data to suggest that areas of dysplasia develop within considerably larger areas of aneuploidy (McHugh, 1990, personal communication). If this proves reliable, it would greatly reduce concerns about dysplasia or carcinoma being missed endoscopically, and additional attention could be given to patients with aneuploid mucosa for additional sampling, potentially reducing some of the sampling problems that occur, particularly in ulcerative colitis.
OTHER FACTORS AFFECTING THE DECISION FOR RESECTION The finding of an endoscopic lesion or dysplasia at the first colonoscopy is much more likely to be associated with an underlying invasive carcinoma (up to 67%) than if found in a patient who has had several previous endoscopies. However, some forms of colitic carcinoma are not accompanied by conventional dysplasia, at least as currently defined (28), and there are few data regarding this finding in Barrett’s esophagus,
182
R. H . Riddell
although those available support this contention (12).
Downloaded by [RMIT University Library] at 15:10 22 March 2016
REMAINING PROBLEMS WITH REGARD TO SURVEILLANCE Barrett’s esophagus Preliminary date with regard to the significance of dysplasia in Barrett’s esophagus may soon be available as the result of the Veterans Administration study in the USA, in which the detection and management of dysplasia have been carried out sufficiently frequently to generate meaningful data. A potential problem, however, is that the mean age of patients with dysplasia is quite high, increasing the expected operative mortality and tending to preclude operative intervention. These important results are eagerly awaited.
7. Dysplasia and carcinoma may be accompanied or preceded by loss of some markers of cell maturation, extension of markers of cell immaturity up to large-bowel crypts, changes in cell proliferation markers, and by changes in expression of some oncogenes or their products. The real questions are therefore whether advances can be made in the determination of which patients are really at risk and whether the endoscopic detection of dysplasia and colitic cancer can be increased to the point that patients derive obvious benefit from a surveillance program as measured by increased survival or a decreased numbers of cancers. The notion of managing each patient on an individual basis becomes a euphemism for ensuring permanent lack of reproducible criteria that can be subjected to analysis.
Ulcerative colitis DATA REQUIRED FROM FOLLOW-UP At an international workshop held to review STUDIES the knowledge of dysplasia and cancer in ulcerative colitis sponsored by the National Foun- There is an obvious lack of data from surveillance dation for Ileitis and Colitis (New York), the studies to answer the basic question of whether Meyerhoff Foundation (Johns Hopkins Hos- surveillance reduces mortality. Important seconpital), and the NIH (USA), held in the USA in dary questions also require answers in order that the risks and benefits of resection can be weighed June 1989, several points became apparent: 1. Enormous resources are spent on sur- against those of doing nothing under appropriate clinical circumstances if a patient presents with: veillance with little obvious benefit. 2. The extent of colitis has no standard method a) low-grade dysplasia (LGD) or b) high grade of definition, variously being established by dysplasia (HGD) either c) with or d) without an barium enema (single or air contrast), endoscopy, endoscopic abnormality from which the biopsy or biopsy, so that other than those in whom total has been taken both at e) first (diagnostic) and disease has been defined radiologically, the pre- at f ) subsequent (surveillance) colonoscopy or upper endoscopy. cise population at risk is unknown. Although the major variables include the grade 3. Dysplasia may be focal, so that random of dysplasia, the presence of an endoscopic abnorbiopsies present major sampling problems. 4.There has been no formal study of the endo- mality, and the detection of dysplasia at first or scopic appearance of dysplasia or of colitic car- subsequent colonoscopy, there are other variables that will ultimately be clarified. The risk of carcinoma. 5. Dysplasia is far from the ultimate marker of cinoma will also need to be subdivided into carcinoma and is not detected in up to a third patients with carcinoma g) at the time of presof patients under surveillance in whom clinical entation or h) at follow-up study of elective surgery is not carried out at presentation-that is, invasive carcinoma subsequently develops. 6. Dysplasia and carcinoma in some patients rate of inmeaselyear over and above that in are preceded by the development of an aneuploid group g. Some form of a control group has to be concell population in the absence of any histologic sidered in the final analysis, because the objective abnormality .
Downloaded by [RMIT University Library] at 15:10 22 March 2016
Screening in Gastrointestinal Cancer
183
of surveillance is to prevent untimely deaths from ensure that the screening program has the desired carcinoma. Some thought will need to be given effectiveness and efficacy. These strategies are examined in the selected to the mortality, and possibly morbidity, of cancer when it found as a result of surveillance as areas of esophageal carcinoma arising in Barrett’s opposed to that in patients not undergoing sur- epithelium and carcinoma in ulcerative colitis. veillance. Years of life ‘saved’ from early detec- There are marked similarities between these tion when compared with the loss from operative groups including: problems in accurately defining mortality must be compared. Essentially, patients the population at risk; the screening method of undergoing surveillance in whom invasive car- choice, which appears to be endoscopy with multicinoma is found incidentally at resection but who ple biopsies; the presence of an aneuploid cell do not die of the disease represent a successful population may select patients at risk, which may use of surveillance, but only if their overall mor- allow stratification of the timing of endoscopy; tality is better than that of patients presenting there is considerable controversy regarding the symptomatically with cancer or than a control timing of excision; there are no data to support a population. Such controls are unethical unless reduction in mortality by screening; debate patients are medically unfit for surgery or refuse regarding whether the cost is justifiable; there is screening or resection, irrespective of what is no guarantee that surveillance will prevent the development of carcinoma or detect it at a curable found. Data generated will be more credible if com- stage of disease; and the risks and the benefits of parisons between subgroups can be carried out to screening need to be weighed against the risks document whether statistically significant dif- and the benefits of no surveillance. However, there are also marked dissimilarities ferences can be demonstrated between these subbetween the two: the area of mucosa at risk in groups. Barrett’s esophagus is far smaller than that in ulcerative colitis, making detection of areas of COST dysplasia and possible foci comparatively easy; Although screening provides a major problem at these may therefore be far easier to keep under the national level, the real questions are not the surveillance. There is a much higher mortality for overall cost but the cost to any individual, and esophagectomy than colectomy, which currently not how much does it cost potentially for each changes the end point for resection. Conversely, year a life saved; how can one estimate whether there is a much higher morbidity associated with this is worthwhile? Is it justified if patients bear pouch operations if they are carried out after this cost themselves despite the limitations of the colectomy. techniques? REFERENCES
SUMMARY Screening is a four-stage process with strategies that invariably include a) the selection or detection of a population at high risk for a specific disease; b) the institution of a screening procedure within that high-risk population, which will detect patients at either a pre-invasive or early invasive stage of the disease, when it is remediable to therapy, usually surgical excision; c) the carrying out of the appropriate therapeutic maneuvre when the marker becomes positive at an appropriate time; and d) follow-up and data analysis to
1 . Rosenstock E, Farmer RG, Petras R, et al. Surveillance for colonic carcinoma in ulcerative colitis. Gastroenterology 1985, 89, 1342-1346 2. Nugent FW, Haggitt RC, Colcher H, Kutteruf GC. Malignant potential of chronic ulcerative colitis. Preliminary report. Gastroenterology 1979.26, 1-5 3. Nugent FW, Haggitt RC. Cancer in ulcerative colitis: results of a 13 year prospective biopsy surveillance program. Gastroenterology 1988, 94, A327 4. Granqvist S, Gabrielsson N, Sundelin P, Thor-
geirsson T. Precancerous lesions in the mucosa in ulcerative colitis. Scand J Gastroenterol 1980, 15, 289-296 5. Dickinson RJ,Dixon HF, Axon ATR. Colonoscopy and the detection of dysplasia in patients with longstanding ulcerative colitis. Lancet 1980, 2, 620-622
Downloaded by [RMIT University Library] at 15:10 22 March 2016
184
R. H . Riddell
6. Lennard-Jones JE, Morson BC, Ritchie JK, Williams CB. Cancer surveillance in ulcerative colitis. Lancet 1983, 2, 149-152 7. Manning AP, Bulgim OR, Dixon MF, Axon ATR. Screening by colonoscopy for colonic epithelial dysplasia in inflammatory bowel disease. Gut 1987, 28, 1489-1494 8. Lashner BA, Hanauer SB, Silverstein MC. Optimal timing for colonoscopy to screen for cancer in ulcerative colitis. Ann Intern Med 1988, 108, 274-278 9. Melville DM, Lennard-Jones JE, Morson BC, Ritchie JK, Williams CB. Surveillance in extensive ulcerative colitis improves prognosis without always preventing cancer. Gut 1988, 29, A1494 10. Rutegard J, Ahsgren L, Sterling R, Janunger KG. Ulcerative colitis. Cancer surveillance in an unselected population. Scand J Gastroenterol 1988, 23, 139-145 11. Gyde SN, Prior P, Allan RN, et al. Colorectal cancer in ulcerative colitis: A cohort study of primary referrals from three centres. Gut 1988, 29, 2 0 6 2 17 12. Skinner DB, Walther BC, Riddell RH, et al. Barrett’s esophagus. Comparison of benign and malignant cases. Ann Surg 1983, 198,554-566 13. DeNardi F, Riddell RH. The normal esophagus. Am J Surg Pathol (in press) 14. Reid BJ, Lewin KJ, Haggitt RC, et al. Endoscopic biopsies can detect high grade dysplasia or early adenocarcinoma in Barrett’s esophagus without grossly recognizable neoplastic lesions. Gastroenterology 1988, 94, 81-90 15. Skinner DB, Walter BC, Riddell RH, Schmidt H, Iascone C, De Meester T. Barrett’s esophagus: comparison of benign and malignant cases. Ann Surg 1983, 198, 554-566 16. Morson BC, Pang LSC. Rectal biopsy as an aid to cancer control in ulcerative colitis. Gut 1967, 8, 423-434 17. Collins RH, Feldman M, Fordtran JS. Colon cancer, dysplasia, and surveillance in patients with ulcerative colitis. N Engl J Med 1987, 316, 16541658 18. Blackstone MO, Riddell RH, Rogers BHG, Levin B. Dysplasia-associated lesion or mass (DALM)
detected by colonoscopy in long-standing ulcerative colitis: an indication for colectomy. Gastroenterology 1981, 80, 366-374 19. Hammarberg C, Slezak P, Tribukait B. Early detection of malignancy in ulcerative colitis-a flow cytometric DNA study. Cancer 1984, 53, 291-295 20. Fozard JBJ, Quirke P, Dixon MF, et al. DNA aneuploidy in ulcerative colitis. Gut 1986,27, 14141418 21. Lofberg R, Tribukait B, Ost A, et al. Flow cytometric DNA analysis in longstanding ulcerative colitis: a method of prediction of dysplasia and carcinoma development? Gut 1987, 28, 1100-1106 22. Melville DM, Jass RJ, Shepherd NA, et al. Dysplasia and deoxyribonucleic acid aneuploidy in the assessment of precancerous changes in chronic ulcerative colitis. Gastroenterology 1988, 95, 668675 23. Rutegard J, Ahsgren L, Stenling R, Roos G. DNA content in ulcerative colitis: flow cytometric in a patient series from a defined area. Dis Colon Rectum 1988, 31, 710-715 24. McKinely MJ, Budsman DR, Gruenberg D, Bronzo RL, Weissman GS, Kahn E. DNA content in Barrett’s esophagus and esophageal malignancy. Am J Gastroenterol 1987, 82, 1012-1015 25. Reid BJ, Haggitt RC, Rubin CE, Rabinovitch PS. Barrett’s esophagus: Correlation between flow cytometry and histology in detection of patients at risk for adenocarcinoma. Gastroenterology 1987,93, 111 26. Rabinovitch PS, Reid BJ, Haggitt RC, Norwood TH, Rubin CE. Progression to cancer in Barrett’s esophagus is associated with genomic instability. Lab Invest 1988, 60, 65-71 27. Fennerty MB, Sampliner RE, Way D, Riddell R, Steinbronn K, Garewal HS. Discordance between flow cytometric abnormalities and dysplasia in Barrett’s esophagus. Gastroenterology 1989, 97, 815820 28. Riddell RH, Goldman H, Ransohoff DF, et al. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol 1483, 14, 931-968
