Surveillance Strategies after Curative Treatment of Colorectal Cancer
http://www.100md.com
《新英格兰医药杂志》
This Journal feature begins with a case vignette highlighting a common clinical problem. Evidence supporting various strategies is then presented, followed by a review of formal guidelines, when they exist. The article ends with the authors' clinical recommendations.
A 63-year-old man underwent complete resection of a T3N0M0, stage II adenocarcinoma of the ascending colon. No adjuvant therapy is planned. There is no family history of colorectal cancer. How should he be followed?
The Clinical Problem
Colorectal adenocarcinoma is the second leading cause of death due to cancer in the United States, accounting for more than 50,000 deaths annually.1 Surgery is the cornerstone of therapy for such tumors, and 70 to 80 percent of patients have tumors that, at diagnosis, can be resected with curative intent. Adjuvant radiation therapy, chemotherapy, or both are useful in selected patients.2,3,4 Among patients who have undergone resection for localized disease, the five-year survival rate is 90 percent; the rate decreases to 65 percent when metastasis to regional lymph nodes is present.1 More than 90 percent of relapses occur by five years, the majority within three. The most common sites of recurrence are the liver, the local site, the abdomen, and the lung.5,6
After initial treatment, surveillance is widely practiced to address treatment complications and related medical issues; to provide advice and reassurance; and to identify and optimally manage recurrent disease, new primary cancers, and precancerous polyps.7,8 A central goal is to improve survival through the earlier identification of recurrences and metachronous colorectal cancers that are still amenable to surgery with curative intent. For example, five-year survival rates of approximately 30 percent are reported after complete resection of isolated liver metastases.9
Plans for surveillance after curative treatment of colorectal cancer most commonly include periodic history taking and physical examinations; some combination of laboratory tests (e.g., carcinoembryonic-antigen tests, liver-function tests, complete blood counts, and fecal occult-blood tests); diagnostic imaging studies (e.g., chest radiography, ultrasonography, computed tomography , magnetic resonance imaging, and barium enema); and endoscopic procedures (e.g., sigmoidoscopy and colonoscopy). However, this extensive list begs the questions of what the standard of care is and whether more intensive follow-up meaningfully improves survival or other outcomes.
Unfortunately, the data on the effectiveness of surveillance strategies after curative resection of colorectal cancer6,10,11,12,13,14,15,16,17 have limitations. Individual components of surveillance have not been well studied, and not all are necessarily important. Because many relapses are first signaled by new symptoms,2,11,13 additional surveillance testing will not detect disease any sooner. Outcomes other than survival or control of disease are evaluated only infrequently. Although many patients view follow-up as important even if recurrence is not detected earlier,18 testing that is poorly justified can lead to both psychological and physical harm to patients, as well as unnecessary costs.19,20
With a surveillance strategy, physicians seek to optimize outcomes and avoid potential liability.21 However, practice patterns with respect to surveillance demonstrate marked variability.22,23,24 When available practice guidelines are used as the gold standard,3,4,6,25,26,27,28,29 there is evidence that both overuse of tests (e.g., complete blood counts and liver-function tests)24 and underuse (e.g., colonoscopy)30 occur.
Since more than 500,000 patients annually undergo surveillance after treatment of their colorectal cancers,2 determining the appropriate surveillance plan has substantial financial ramifications. For example, among 11 separate surveillance strategies, a nearly 30-fold difference was noted in charges (range, $910 to $26,717, in 1992 dollars) for five years of follow-up.22
Strategies and Evidence
We identified six randomized trials reported since 1990 that have compared different low-intensity and high-intensity programs of surveillance for patients after curative surgery for adenocarcinoma of the colon or rectum.10,11,12,13,14,15 The studies also fulfilled methodologic criteria for further evaluation in subsequent meta-analyses.6,16,17
All six trials included patients with resected colon or rectal cancer and a range of disease stages from early to more advanced; the trials were conducted during an era when the indications for adjuvant treatment were less well defined than they are today. The frequency and content of screening in the intensive-surveillance and control groups varied greatly among trials. In the groups that received intensive surveillance, the initial interval between visits ranged from three to six months, which, in four of the six studies, was shorter than in the control groups.11,12,13,15 In one trial, the sole difference between the control group and the group receiving intensive surveillance was the frequency of visits and related testing12; in the remaining five trials, selected additional tests were performed only in the groups receiving intensive surveillance.
In only two studies was the surveillance content for the control group truly minimal, focusing mainly on the reporting of suspicious symptoms.13,15 In the other four studies, blood tests, bowel surveillance, and chest radiography were also part of the control strategy in some way. Indeed, some tests were performed with equal frequency in both study groups, as was the case with carcinoembryonic-antigen testing in three studies.10,12,14 In only one trial was the follow-up approach used for the intensive-surveillance group adjusted on the basis of the anticipated risk of relapse for each patient.15 Sample sizes ranged from 106 to 597 patients. The two studies that provided justification for their planned sample sizes (325 and 597 patients) were designed to detect a 15 to 20 percent reduction in mortality.10,12 In general, the first five years of follow-up were the principal focus. The majority of patients were followed for no less than five years, or until they died.
Overall, the number of recurrences was similar in the control and intensive-surveillance groups, but the recurrences were found earlier or were more likely to be asymptomatic in the groups receiving intensive surveillance. In five of six trials, surgery with curative intent was reported more frequently in the groups receiving intensive surveillance11,12,13,14,15; the overwhelming majority of these procedures were for metastatic or recurrent disease as opposed to new primary cancers. However, a statistically significant improvement in overall survival in the group receiving intensive surveillance was reported in only two of six studies.11,15 Limited information was available regarding the quality of life, costs, and potential harm associated with increased surveillance. Available data from one study suggested that the perceived benefits of more frequent surveillance offset the anxiety, inconvenience, and costs entailed by the extra visits and testing.12,31
Three groups of investigators6,16,17 have reported independent meta-analyses that include five or all six of these randomized studies and address, in part, the concern that negative results were due to inadequate sample sizes. A reduction in death from all causes with more intensive follow-up was more apparent in these pooled meta-analyses, with all three groups reporting a significant reduction in relative risk of approximately 20 percent. The associated difference in absolute risk was 7 percentage points. One of these groups of investigators performed a related analysis of cost effectiveness and reported that with intensive follow-up the cost per year of life gained was ￡3,402 ($6,096 in U.S. dollars today). This is much lower than the threshold for cost acceptability of ￡30,000 ($54,749 in U.S. dollars today) specified by the United Kingdom's National Health Service.32
Areas of Uncertainty
Specific Surveillance Tests
The particular aspects of intensive surveillance that account for the associated improvement in survival are uncertain. Since new symptoms are commonly the first sign of relapse, the identification of suspicious symptoms that warrant further assessment, such as new abdominal pain, blood in the stool, or a change in bowel habits, is important. Follow-up visits allow for medical assessment, reassurance, and counseling, including a discussion of the potentially increased risk of colorectal cancer in family members (e.g., the risk increases by a factor of two to three if the patient is a first-degree relative and by a factor of three to four if the patient is 50 years or younger at diagnosis) and the importance of family members' being advised about the need for screening.29,33 However, physical examination that is not guided by the history or symptoms5,13,34 and the use of many routine laboratory tests (e.g., complete blood counts, liver-function tests, and fecal occult-blood tests)2,5,12,13,14 appear to have a low yield as the first method of detecting the recurrence of resectable tumors.
Chest radiography may identify resectable pulmonary metastases. Its use in surveillance was advocated by the authors of one meta-analysis6 on the basis of its inclusion in the program of intensive surveillance in all six trials and the recognition that survival after surgical removal of localized lung metastases is similar to that after resection of liver metastases. However, the role of chest radiography remains controversial.
Among 167 patients who were randomly assigned to receive intensive surveillance, Schoemaker et al.10 attributed only one instance of long-term survival to the performance of annual chest radiography. A large cohort study identified 6 patients (0.5 percent) who survived for a median follow-up of 62 months (range, 41 to 75) among 1247 who were followed with chest radiographs every 3 to 6 months after resection of stage II or stage III colon cancer.2
Carcinoembryonic-antigen measurement is commonly the first test to identify relapse.2,11,13 It has been proposed as a useful2,16,34,35 and cost-effective tool for surveillance after colorectal cancer, even for patients in whom the antigen level was not elevated at diagnosis. The test was identified as potentially important in two meta-analyses.6,16 Yet when 216 patients who had undergone primary resection of colorectal cancer and had subsequent confirmed elevation of the carcinoembryonic antigen level were randomly assigned, because of this test, to either aggressive assessment for recurrent disease (possibly including laparotomy) or to no intervention, the study was stopped early when the former strategy appeared unlikely to provide a significant survival advantage (five-year survival rate, 22 percent vs. 20.4 percent, respectively; P=0.25).36 Some have suggested that features of the study design (e.g., overly conservative criteria for what constituted an elevated level of carcinoembryonic antigen) may have contributed to the negative results.6,24
One meta-analysis suggested a survival benefit associated with performing CT (every 3 to 12 months),16 and the other two meta-analyses reported a survival benefit specifically with liver imaging.6,17 However, others have reported that radiographic evaluation of the liver, at least when performed annually, did not increase the number of curative hepatectomies.10 The track record for detecting curable pelvic or local relapse is also disappointing,11,13,14 and when adjunctive radiation therapy is used, related post-treatment changes further complicate the interpretation of such imaging.
The possibility that synchronous colorectal cancers and polyps may be present at the time of diagnosis of the index primary colorectal tumor supports initial evaluation of the entire large bowel. Since, as compared with the general population, patients with a prior sporadic colorectal cancer have a risk of a second primary colorectal tumor that is increased by a factor of 1.5 to 3, serial colonoscopy remains important during follow-up37,38 and is preferred over other screening approaches in this setting.39 Otherwise, randomized trials suggest that the yield of more frequent bowel surveillance for detecting recurrent disease is low.6,10,11,13,14,16 Failure of treatment at the local site is a greater concern in patients with rectal cancer than in those with colon cancer. Yet even when more intensive surveillance was associated with improved surgical control of recurrent local disease,11 most local recurrences were detected on the basis of elevated carcinoembryonic antigen levels or the development of symptoms rather than by endoscopy.
Other Uncertainties
The optimal frequency of follow-up after primary treatment for colorectal cancer is also an unsettled question. Our growing understanding of the inherited risk of colorectal cancer29 33 is likely to affect recommendations regarding bowel-surveillance strategies,29,33,40 as may newer interventions, such as aspirin therapy,41 that are intended to decrease the risk of colorectal adenomas. The cost effectiveness of various screening strategies is also uncertain.42 The roles of newer forms of technology — such as endoscopic ultrasonography, virtual colonoscopy, immunoscintigraphy with the use of an anti–carcinoembryonic antigen antibody, and positron-emission tomography — are still being defined.39,43,44
Although most patients endorse the concept of follow-up, limited data are available that prove and quantify quality-of-life and psychosocial benefits.18,24 Similarly, the question of what type of practitioner is best qualified to perform surveillance and in what setting (e.g., hospital vs. general practice) is controversial and currently under study in a randomized trial.17,24
Colorectal cancers are heterogeneous. The tailoring of surveillance approaches on the basis of the initial site and stage of cancer and the likelihood and anticipated pattern of relapse deserves further investigation. The effects on surveillance strategies of newer prognostic markers,45 adjuvant therapy for primary and recurrent colorectal cancer,3,4,46 and newer approaches to isolated recurrences that cannot be removed by surgery are still being defined.
Available data from randomized trials suggest that in cases of recurrent disease, there are improvements in the rate of survival and quality of life among patients who receive immediate chemotherapy, as compared with those who receive delayed chemotherapy, and among patients who receive chemotherapy compared with those who receive the best supportive care alone.47,48 Chemotherapeutic strategies for colorectal cancer have improved.49 It is possible that earlier detection of relapse may enhance the potential benefits of chemotherapy.
Guidelines
Table 1 presents a summary of recommendations from selected, representative surveillance guidelines.3,4,6,25,26,28 All of these, as well as other guidelines not listed that focus primarily on the role of bowel surveillance,27,29 emphasize the importance of complete preoperative or perioperative colonoscopy, demonstration that the colon is free of other cancer and polyps, and subsequent evaluations every three to five years. Other testing is either not strongly or not uniformly recommended or is recommended only as indicated by symptoms or other findings, as well as the patient's medical fitness and willingness to undergo diagnostic evaluation and treatment of recurrence.
Table 1. Summary of Guidelines for Representative Colorectal Surveillance after Primary Surgery with Curative Intent.
Conclusions and Recommendations
Optimal strategies for surveillance after initial treatment with curative intent for colorectal cancer remain uncertain. Pending new information and given the limitations of available data, we recommend the following approach to surveillance in asymptomatic patients such as the one in the vignette.
Physician visits with a focus on the clinical and disease-related history, directed physical examination guided by this history, coordination of follow-up, and counseling are recommended every three to six months for the first three years after treatment, with decreased frequency thereafter for two years. An appropriate diagnostic evaluation is indicated if there are suspicious symptoms or findings, and patients should be encouraged to call between visits if new symptoms develop. Counseling topics should include the rationale for surveillance; the anticipated benefits, risks, and uncertainties of more intense follow-up; and the role of surgery for treatment of relapse. Patients should understand that family members are at increased risk for colorectal cancer, and they should be encouraged to inform their relatives of this risk and to urge them to discuss screening with their physicians.29,33
After it has been determined that the colon is free of cancer and polyps, we recommend colonoscopy every three to five years, depending on colonoscopic findings, primarily for the detection of polyps and second primary cancers.8,29,38,39 The risk of a second colorectal cancer continues beyond five years after the initial diagnosis, and so should colonoscopic surveillance. Selected guideline panels25,28 recommend more frequent endoscopic surveillance for local recurrence after treatment for rectal cancer, but its yield as the first indicator of such recurrence in asymptomatic patients who have received appropriate primary treatment appears to be low.11
The principal reason to perform further testing during surveillance is to identify disease that is still amenable to surgery with curative intent. In asymptomatic patients who are not medically fit or who would not pursue salvage surgery if it were offered, no further surveillance testing is recommended. For other asymptomatic patients, the best way to proceed is controversial. Certain guideline panels and recent meta-analyses recommend or suggest a role for carcinoembryonic-antigen testing, liver imaging, chest radiography, or all of these tests on the basis of observations of a small but significant survival benefit with strategies that incorporate them. However, other experts do not support the routine use of these tests, and their potential disadvantages (including patient anxiety and cost) also warrant consideration. Given the available information, we believe that carcinoembryonic-antigen testing every three to six months for up to five years is potentially the most useful of these options. However, in our judgment, pending further evaluation in prospective studies, the limitations of the available data, uncertainties in their interpretation, and the related lack of expert consensus preclude a recommendation for the routine use of these three tests for surveillance in all asymptomatic patients. In addition, the routine use of newer imaging techniques such as endoscopic ultrasonography and positron-emission tomography in surveillance is not currently recommended.
Available data regarding surveillance beyond five years are more limited. However, the risk of relapse from the initial colorectal cancer after five years is low. Accordingly, with the exception of ongoing colonoscopic follow-up as outlined above, we do not recommend other surveillance measures aside from customary general medical monitoring beyond the five-year point.
Often, physicians from many specialties participate in the care of patients who have had colorectal cancer. To ensure appropriate follow-up and avoid redundant evaluations, we recommend that one physician take primary responsibility for surveillance testing in coordination with other providers.
The opinions expressed in this article are those of the authors and do not necessarily represent the views of the American Society of Clinical Oncology.
We are indebted to Leonard Saltz, M.D., Deborah Schrag, M.D., M.P.H., and Katherine Virgo, Ph.D., for their constructive review of the manuscript.
Source Information
From the Division of Solid Tumor Oncology, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York (D.G.P.); the Division of Hematology–Oncology and the Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago (A.B.B.); and the Cancer Policy and Clinical Affairs Department, American Society of Clinical Oncology, Alexandria, Va. (M.R.S.).
Address reprint requests to Dr. Pfister at Memorial Hospital, Box 188, 1275 York Ave., New York, NY 10021, or at pfisterd@mskcc.org.
References
Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5-26.
Goldberg RM, Fleming TR, Tangen CM, et al. Surgery for recurrent colon cancer: strategies for identifying resectable recurrence and success rates after resection. Ann Intern Med 1998;129:27-35.
NCCN colon cancer clinical practice guidelines in oncology. J Natl Compr Cancer Network 2003;1:40-53.
NCCN rectal cancer clinical practice guidelines in oncology. J Natl Compr Cancer Network 2003;1:54-63.
Kjeldsen BJ, Kronborg O, Fenger C, Jorgensen OD. The pattern of recurrent colorectal cancer in a prospective randomised study and the characteristics of diagnostic tests. Int J Colorectal Dis 1997;12:329-334.
Figueredo A, Rumble RB, Maroun J, et al. Follow-up of patients with curatively resected colorectal cancer: a practice guideline. BMC Cancer 2003;3:26-26.
Kattlove H, Winn RJ. Ongoing care of patients after primary treatment for their cancer. CA Cancer J Clin 2003;53:172-196.
Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993;329:1977-1981.
Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999;230:309-318.
Schoemaker D, Black R, Giles L, Toouli J. Yearly colonoscopy, liver CT, and chest radiography do not influence 5-year survival of colorectal cancer patients. Gastroenterology 1998;114:7-14.
Pietra N, Sarli L, Costi R, Ouchemi C, Grattarola M, Peracchia A. Role of follow-up in management of local recurrences of colorectal cancer: a prospective randomized trial. Dis Colon Rectum 1998;41:1127-1133.
Kjeldsen BJ, Kronborg O, Fenger C, Jorgensen OD. A prospective randomized study of follow-up after radical surgery for colorectal cancer. Br J Surg 1997;84:666-669.
Ohlsson B, Breland U, Ekberg H, Graffner H, Tranberg K-G. Follow-up after curative surgery for colorectal carcinoma: randomized comparison with no follow-up. Dis Colon Rectum 1995;38:619-626.
Makela JT, Laitinen SO, Kairaluoma MI. Five-year follow-up after radical surgery for colorectal cancer: results of a prospective randomized trial. Arch Surg 1995;130:1062-1067.
Secco GB, Fardelli R, Gianquinto D, et al. Efficacy and cost of risk-adapted follow-up in patients after colorectal cancer surgery: a prospective, randomized and controlled trial. Eur J Surg Oncol 2002;28:418-423.
Renehan AG, Egger M, Saunders MP, O'Dwyer ST. Impact on survival of intensive follow up after curative resection for colorectal cancer: systematic review and meta-analysis of randomised trials. BMJ 2002;324:813-813.
Jeffery GM, Hickey BE, Hider P. Follow-up strategies for patients treated for non-metastatic colorectal cancer. Cochrane Database Syst Rev 2002;1:CD002200-CD002200.
Stiggelbout AM, de Haes JCJM, Vree R, et al. Follow-up of colorectal cancer patients: quality of life and attitudes towards follow-up. Br J Cancer 1997;75:914-920.
Loprinzi CL, Hayes D, Smith T. Doc, shouldn't we be getting more tests? J Clin Oncol 2000;18:2345-2348.
Deyo RA. Cascade effects of medical technology. Annu Rev Public Health 2002;23:23-44.
Kievit J. Colorectal cancer follow-up: a reassessment of empirical evidence on effectiveness. Eur J Surg Oncol 2000;26:322-328.
Virgo KS, Vernava AM, Long WE, McKirgan LW, Johnson FE. Cost of patient follow-up after potentially curative colorectal cancer treatment. JAMA 1995;273:1837-1841.
Cooper GS, Yuan Z, Chak A, Rimm AA. Geographic and patient variation among Medicare beneficiaries in the use of follow-up testing after surgery for nonmetastatic colorectal carcinoma. Cancer 1999;85:2124-2131.
Earle CC, Grunfeld E, Coyle D, Cripps MC, Stern HS. Cancer physicians' attitudes toward colorectal cancer follow-up. Ann Oncol 2003;14:400-405.
ESMO minimum clinical recommendations for diagnosis, adjuvant treatment and follow-up of colon cancer. Ann Oncol 2001;12:1053-1054.
Benson AB III, Desch CE, Flynn PJ, et al. 2000 Update of American Society of Clinical Oncology colorectal cancer surveillance guidelines. J Clin Oncol 2000;18:3586-3588.
Smith RA, Cokkinides V, von Eschenbach AC, et al. American Cancer Society guidelines for early detection of cancer. CA Cancer J Clin 2002;52:8-22.
Berman JM, Cheung RJ, Weinberg DS. Surveillance after colorectal cancer resection. Lancet 2000;355:395-399.
Winawer S, Fletcher R, Rex D, et al. Colorectal cancer screening and surveillance: clinical guidelines and rationale -- update based on new evidence. Gastroenterology 2003;124:544-560.
Knopf KB, Warren JL, Feuer EJ, Brown ML. Bowel surveillance patterns after a diagnosis of colorectal cancer in Medicare beneficiaries. Gastrointest Endosc 2001;54:563-571.
Kjeldsen BJ, Thorsen H, Whalley D, Kronborg O. Influence of follow-up on health-related quality of life after radical surgery for colorectal cancer. Scand J Gastroenterol 1999;34:509-515.
Renehan AG, O'Dwyer ST, Whynes DK. Cost effectiveness analysis of intensive versus conventional follow up after curative resection for colorectal cancer. BMJ 2004;328:81-81.
NCCN colorectal cancer screening clinical practice guidelines in oncology. J Natl Compr Cancer Network 2003;1:72-93.
Graham RA, Wang S, Catalano PJ, Haller DG. Postsurgical surveillance of colon cancer: preliminary cost analysis of physician examination, carcinoembryonic antigen testing, chest x-ray, and colonoscopy. Ann Surg 1998;228:59-63.
McCall JL, Black RB, Rich CA, et al. The value of serum carcinoembryonic antigen in predicting recurrent disease following curative resection of colorectal cancer. Dis Colon Rectum 1994;37:875-881.
Lennon T, Houghton J, Northover J. What is the value of clinical follow-up for colorectal cancer patients? The experience of the CRC/NIH CEA second-look trial. Nottingham, England: Nottingham International Colorectal Cancer Symposium, 1995. (Accessed May 10, 2004, at http://www.york.ac.uk/inst/crd/ehc36.pdf.)
Green RJ, Metlay JP, Propert K, et al. Surveillance for second primary colorectal cancer after adjuvant chemotherapy: an analysis of Intergroup 0089. Ann Intern Med 2002;136:261-269.
Winawer SJ, Zauber AG, O'Brien MJ, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. N Engl J Med 1993;328:901-906.
Helm J, Choi J, Sutphen R, Barthel JS, Albrecht TL, Chirikos TN. Current and evolving strategies for colorectal cancer screening. Cancer Control 2003;10:193-204.
Lynch HT, de la Chapelle A. Hereditary colorectal cancer. N Engl J Med 2003;348:919-932.
Sandler RS, Halabi S, Baron JA, et al. A randomized trial of aspirin to prevent colorectal adenomas in patients with previous colorectal cancer. N Engl J Med 2003;348:883-890.
Schrag D. Defining optimal treatment for stage II colon cancer: does decision analysis help? Gastroenterology 1999;117:1005-1007.
Rotondano G, Esposito P, Pellecchia L, Novi A, Romano G. Early detection of locally recurrent rectal cancer by endosonography. Br J Radiol 1997;70:567-571.
Willkomm P, Bender H, Bangard M, Decker P, Grunwald F, Biersack H-J. FDG PET and immunoscintigraphy with 99mTc-labeled antibody fragments for detection of the recurrence of colorectal carcinoma. J Nucl Med 2000;41:1657-1663.
Gryfe R, Kim H, Hsieh ETK, et al. Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 2000;342:69-77.
Kemeny N, Huang Y, Cohen AM, et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl J Med 1999;341:2039-2048.
Nordic Gastrointestinal Tumor Adjuvant Therapy Group. Expectancy or primary chemotherapy in patients with advanced asymptomatic colorectal cancer: a randomized trial. J Clin Oncol 1992;10:904-911.
Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure in patients with metastatic colorectal cancer. Lancet 1998;352:1413-1418.
Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000;343:905-914.
Winawer SJ, St John DJ, Bond JH, et al. Prevention of colorectal cancer: guidelines based on new data. Bull World Health Organ 1995;73:7-10.(David G. Pfister, M.D., A)
A 63-year-old man underwent complete resection of a T3N0M0, stage II adenocarcinoma of the ascending colon. No adjuvant therapy is planned. There is no family history of colorectal cancer. How should he be followed?
The Clinical Problem
Colorectal adenocarcinoma is the second leading cause of death due to cancer in the United States, accounting for more than 50,000 deaths annually.1 Surgery is the cornerstone of therapy for such tumors, and 70 to 80 percent of patients have tumors that, at diagnosis, can be resected with curative intent. Adjuvant radiation therapy, chemotherapy, or both are useful in selected patients.2,3,4 Among patients who have undergone resection for localized disease, the five-year survival rate is 90 percent; the rate decreases to 65 percent when metastasis to regional lymph nodes is present.1 More than 90 percent of relapses occur by five years, the majority within three. The most common sites of recurrence are the liver, the local site, the abdomen, and the lung.5,6
After initial treatment, surveillance is widely practiced to address treatment complications and related medical issues; to provide advice and reassurance; and to identify and optimally manage recurrent disease, new primary cancers, and precancerous polyps.7,8 A central goal is to improve survival through the earlier identification of recurrences and metachronous colorectal cancers that are still amenable to surgery with curative intent. For example, five-year survival rates of approximately 30 percent are reported after complete resection of isolated liver metastases.9
Plans for surveillance after curative treatment of colorectal cancer most commonly include periodic history taking and physical examinations; some combination of laboratory tests (e.g., carcinoembryonic-antigen tests, liver-function tests, complete blood counts, and fecal occult-blood tests); diagnostic imaging studies (e.g., chest radiography, ultrasonography, computed tomography , magnetic resonance imaging, and barium enema); and endoscopic procedures (e.g., sigmoidoscopy and colonoscopy). However, this extensive list begs the questions of what the standard of care is and whether more intensive follow-up meaningfully improves survival or other outcomes.
Unfortunately, the data on the effectiveness of surveillance strategies after curative resection of colorectal cancer6,10,11,12,13,14,15,16,17 have limitations. Individual components of surveillance have not been well studied, and not all are necessarily important. Because many relapses are first signaled by new symptoms,2,11,13 additional surveillance testing will not detect disease any sooner. Outcomes other than survival or control of disease are evaluated only infrequently. Although many patients view follow-up as important even if recurrence is not detected earlier,18 testing that is poorly justified can lead to both psychological and physical harm to patients, as well as unnecessary costs.19,20
With a surveillance strategy, physicians seek to optimize outcomes and avoid potential liability.21 However, practice patterns with respect to surveillance demonstrate marked variability.22,23,24 When available practice guidelines are used as the gold standard,3,4,6,25,26,27,28,29 there is evidence that both overuse of tests (e.g., complete blood counts and liver-function tests)24 and underuse (e.g., colonoscopy)30 occur.
Since more than 500,000 patients annually undergo surveillance after treatment of their colorectal cancers,2 determining the appropriate surveillance plan has substantial financial ramifications. For example, among 11 separate surveillance strategies, a nearly 30-fold difference was noted in charges (range, $910 to $26,717, in 1992 dollars) for five years of follow-up.22
Strategies and Evidence
We identified six randomized trials reported since 1990 that have compared different low-intensity and high-intensity programs of surveillance for patients after curative surgery for adenocarcinoma of the colon or rectum.10,11,12,13,14,15 The studies also fulfilled methodologic criteria for further evaluation in subsequent meta-analyses.6,16,17
All six trials included patients with resected colon or rectal cancer and a range of disease stages from early to more advanced; the trials were conducted during an era when the indications for adjuvant treatment were less well defined than they are today. The frequency and content of screening in the intensive-surveillance and control groups varied greatly among trials. In the groups that received intensive surveillance, the initial interval between visits ranged from three to six months, which, in four of the six studies, was shorter than in the control groups.11,12,13,15 In one trial, the sole difference between the control group and the group receiving intensive surveillance was the frequency of visits and related testing12; in the remaining five trials, selected additional tests were performed only in the groups receiving intensive surveillance.
In only two studies was the surveillance content for the control group truly minimal, focusing mainly on the reporting of suspicious symptoms.13,15 In the other four studies, blood tests, bowel surveillance, and chest radiography were also part of the control strategy in some way. Indeed, some tests were performed with equal frequency in both study groups, as was the case with carcinoembryonic-antigen testing in three studies.10,12,14 In only one trial was the follow-up approach used for the intensive-surveillance group adjusted on the basis of the anticipated risk of relapse for each patient.15 Sample sizes ranged from 106 to 597 patients. The two studies that provided justification for their planned sample sizes (325 and 597 patients) were designed to detect a 15 to 20 percent reduction in mortality.10,12 In general, the first five years of follow-up were the principal focus. The majority of patients were followed for no less than five years, or until they died.
Overall, the number of recurrences was similar in the control and intensive-surveillance groups, but the recurrences were found earlier or were more likely to be asymptomatic in the groups receiving intensive surveillance. In five of six trials, surgery with curative intent was reported more frequently in the groups receiving intensive surveillance11,12,13,14,15; the overwhelming majority of these procedures were for metastatic or recurrent disease as opposed to new primary cancers. However, a statistically significant improvement in overall survival in the group receiving intensive surveillance was reported in only two of six studies.11,15 Limited information was available regarding the quality of life, costs, and potential harm associated with increased surveillance. Available data from one study suggested that the perceived benefits of more frequent surveillance offset the anxiety, inconvenience, and costs entailed by the extra visits and testing.12,31
Three groups of investigators6,16,17 have reported independent meta-analyses that include five or all six of these randomized studies and address, in part, the concern that negative results were due to inadequate sample sizes. A reduction in death from all causes with more intensive follow-up was more apparent in these pooled meta-analyses, with all three groups reporting a significant reduction in relative risk of approximately 20 percent. The associated difference in absolute risk was 7 percentage points. One of these groups of investigators performed a related analysis of cost effectiveness and reported that with intensive follow-up the cost per year of life gained was ￡3,402 ($6,096 in U.S. dollars today). This is much lower than the threshold for cost acceptability of ￡30,000 ($54,749 in U.S. dollars today) specified by the United Kingdom's National Health Service.32
Areas of Uncertainty
Specific Surveillance Tests
The particular aspects of intensive surveillance that account for the associated improvement in survival are uncertain. Since new symptoms are commonly the first sign of relapse, the identification of suspicious symptoms that warrant further assessment, such as new abdominal pain, blood in the stool, or a change in bowel habits, is important. Follow-up visits allow for medical assessment, reassurance, and counseling, including a discussion of the potentially increased risk of colorectal cancer in family members (e.g., the risk increases by a factor of two to three if the patient is a first-degree relative and by a factor of three to four if the patient is 50 years or younger at diagnosis) and the importance of family members' being advised about the need for screening.29,33 However, physical examination that is not guided by the history or symptoms5,13,34 and the use of many routine laboratory tests (e.g., complete blood counts, liver-function tests, and fecal occult-blood tests)2,5,12,13,14 appear to have a low yield as the first method of detecting the recurrence of resectable tumors.
Chest radiography may identify resectable pulmonary metastases. Its use in surveillance was advocated by the authors of one meta-analysis6 on the basis of its inclusion in the program of intensive surveillance in all six trials and the recognition that survival after surgical removal of localized lung metastases is similar to that after resection of liver metastases. However, the role of chest radiography remains controversial.
Among 167 patients who were randomly assigned to receive intensive surveillance, Schoemaker et al.10 attributed only one instance of long-term survival to the performance of annual chest radiography. A large cohort study identified 6 patients (0.5 percent) who survived for a median follow-up of 62 months (range, 41 to 75) among 1247 who were followed with chest radiographs every 3 to 6 months after resection of stage II or stage III colon cancer.2
Carcinoembryonic-antigen measurement is commonly the first test to identify relapse.2,11,13 It has been proposed as a useful2,16,34,35 and cost-effective tool for surveillance after colorectal cancer, even for patients in whom the antigen level was not elevated at diagnosis. The test was identified as potentially important in two meta-analyses.6,16 Yet when 216 patients who had undergone primary resection of colorectal cancer and had subsequent confirmed elevation of the carcinoembryonic antigen level were randomly assigned, because of this test, to either aggressive assessment for recurrent disease (possibly including laparotomy) or to no intervention, the study was stopped early when the former strategy appeared unlikely to provide a significant survival advantage (five-year survival rate, 22 percent vs. 20.4 percent, respectively; P=0.25).36 Some have suggested that features of the study design (e.g., overly conservative criteria for what constituted an elevated level of carcinoembryonic antigen) may have contributed to the negative results.6,24
One meta-analysis suggested a survival benefit associated with performing CT (every 3 to 12 months),16 and the other two meta-analyses reported a survival benefit specifically with liver imaging.6,17 However, others have reported that radiographic evaluation of the liver, at least when performed annually, did not increase the number of curative hepatectomies.10 The track record for detecting curable pelvic or local relapse is also disappointing,11,13,14 and when adjunctive radiation therapy is used, related post-treatment changes further complicate the interpretation of such imaging.
The possibility that synchronous colorectal cancers and polyps may be present at the time of diagnosis of the index primary colorectal tumor supports initial evaluation of the entire large bowel. Since, as compared with the general population, patients with a prior sporadic colorectal cancer have a risk of a second primary colorectal tumor that is increased by a factor of 1.5 to 3, serial colonoscopy remains important during follow-up37,38 and is preferred over other screening approaches in this setting.39 Otherwise, randomized trials suggest that the yield of more frequent bowel surveillance for detecting recurrent disease is low.6,10,11,13,14,16 Failure of treatment at the local site is a greater concern in patients with rectal cancer than in those with colon cancer. Yet even when more intensive surveillance was associated with improved surgical control of recurrent local disease,11 most local recurrences were detected on the basis of elevated carcinoembryonic antigen levels or the development of symptoms rather than by endoscopy.
Other Uncertainties
The optimal frequency of follow-up after primary treatment for colorectal cancer is also an unsettled question. Our growing understanding of the inherited risk of colorectal cancer29 33 is likely to affect recommendations regarding bowel-surveillance strategies,29,33,40 as may newer interventions, such as aspirin therapy,41 that are intended to decrease the risk of colorectal adenomas. The cost effectiveness of various screening strategies is also uncertain.42 The roles of newer forms of technology — such as endoscopic ultrasonography, virtual colonoscopy, immunoscintigraphy with the use of an anti–carcinoembryonic antigen antibody, and positron-emission tomography — are still being defined.39,43,44
Although most patients endorse the concept of follow-up, limited data are available that prove and quantify quality-of-life and psychosocial benefits.18,24 Similarly, the question of what type of practitioner is best qualified to perform surveillance and in what setting (e.g., hospital vs. general practice) is controversial and currently under study in a randomized trial.17,24
Colorectal cancers are heterogeneous. The tailoring of surveillance approaches on the basis of the initial site and stage of cancer and the likelihood and anticipated pattern of relapse deserves further investigation. The effects on surveillance strategies of newer prognostic markers,45 adjuvant therapy for primary and recurrent colorectal cancer,3,4,46 and newer approaches to isolated recurrences that cannot be removed by surgery are still being defined.
Available data from randomized trials suggest that in cases of recurrent disease, there are improvements in the rate of survival and quality of life among patients who receive immediate chemotherapy, as compared with those who receive delayed chemotherapy, and among patients who receive chemotherapy compared with those who receive the best supportive care alone.47,48 Chemotherapeutic strategies for colorectal cancer have improved.49 It is possible that earlier detection of relapse may enhance the potential benefits of chemotherapy.
Guidelines
Table 1 presents a summary of recommendations from selected, representative surveillance guidelines.3,4,6,25,26,28 All of these, as well as other guidelines not listed that focus primarily on the role of bowel surveillance,27,29 emphasize the importance of complete preoperative or perioperative colonoscopy, demonstration that the colon is free of other cancer and polyps, and subsequent evaluations every three to five years. Other testing is either not strongly or not uniformly recommended or is recommended only as indicated by symptoms or other findings, as well as the patient's medical fitness and willingness to undergo diagnostic evaluation and treatment of recurrence.
Table 1. Summary of Guidelines for Representative Colorectal Surveillance after Primary Surgery with Curative Intent.
Conclusions and Recommendations
Optimal strategies for surveillance after initial treatment with curative intent for colorectal cancer remain uncertain. Pending new information and given the limitations of available data, we recommend the following approach to surveillance in asymptomatic patients such as the one in the vignette.
Physician visits with a focus on the clinical and disease-related history, directed physical examination guided by this history, coordination of follow-up, and counseling are recommended every three to six months for the first three years after treatment, with decreased frequency thereafter for two years. An appropriate diagnostic evaluation is indicated if there are suspicious symptoms or findings, and patients should be encouraged to call between visits if new symptoms develop. Counseling topics should include the rationale for surveillance; the anticipated benefits, risks, and uncertainties of more intense follow-up; and the role of surgery for treatment of relapse. Patients should understand that family members are at increased risk for colorectal cancer, and they should be encouraged to inform their relatives of this risk and to urge them to discuss screening with their physicians.29,33
After it has been determined that the colon is free of cancer and polyps, we recommend colonoscopy every three to five years, depending on colonoscopic findings, primarily for the detection of polyps and second primary cancers.8,29,38,39 The risk of a second colorectal cancer continues beyond five years after the initial diagnosis, and so should colonoscopic surveillance. Selected guideline panels25,28 recommend more frequent endoscopic surveillance for local recurrence after treatment for rectal cancer, but its yield as the first indicator of such recurrence in asymptomatic patients who have received appropriate primary treatment appears to be low.11
The principal reason to perform further testing during surveillance is to identify disease that is still amenable to surgery with curative intent. In asymptomatic patients who are not medically fit or who would not pursue salvage surgery if it were offered, no further surveillance testing is recommended. For other asymptomatic patients, the best way to proceed is controversial. Certain guideline panels and recent meta-analyses recommend or suggest a role for carcinoembryonic-antigen testing, liver imaging, chest radiography, or all of these tests on the basis of observations of a small but significant survival benefit with strategies that incorporate them. However, other experts do not support the routine use of these tests, and their potential disadvantages (including patient anxiety and cost) also warrant consideration. Given the available information, we believe that carcinoembryonic-antigen testing every three to six months for up to five years is potentially the most useful of these options. However, in our judgment, pending further evaluation in prospective studies, the limitations of the available data, uncertainties in their interpretation, and the related lack of expert consensus preclude a recommendation for the routine use of these three tests for surveillance in all asymptomatic patients. In addition, the routine use of newer imaging techniques such as endoscopic ultrasonography and positron-emission tomography in surveillance is not currently recommended.
Available data regarding surveillance beyond five years are more limited. However, the risk of relapse from the initial colorectal cancer after five years is low. Accordingly, with the exception of ongoing colonoscopic follow-up as outlined above, we do not recommend other surveillance measures aside from customary general medical monitoring beyond the five-year point.
Often, physicians from many specialties participate in the care of patients who have had colorectal cancer. To ensure appropriate follow-up and avoid redundant evaluations, we recommend that one physician take primary responsibility for surveillance testing in coordination with other providers.
The opinions expressed in this article are those of the authors and do not necessarily represent the views of the American Society of Clinical Oncology.
We are indebted to Leonard Saltz, M.D., Deborah Schrag, M.D., M.P.H., and Katherine Virgo, Ph.D., for their constructive review of the manuscript.
Source Information
From the Division of Solid Tumor Oncology, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York (D.G.P.); the Division of Hematology–Oncology and the Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago (A.B.B.); and the Cancer Policy and Clinical Affairs Department, American Society of Clinical Oncology, Alexandria, Va. (M.R.S.).
Address reprint requests to Dr. Pfister at Memorial Hospital, Box 188, 1275 York Ave., New York, NY 10021, or at pfisterd@mskcc.org.
References
Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5-26.
Goldberg RM, Fleming TR, Tangen CM, et al. Surgery for recurrent colon cancer: strategies for identifying resectable recurrence and success rates after resection. Ann Intern Med 1998;129:27-35.
NCCN colon cancer clinical practice guidelines in oncology. J Natl Compr Cancer Network 2003;1:40-53.
NCCN rectal cancer clinical practice guidelines in oncology. J Natl Compr Cancer Network 2003;1:54-63.
Kjeldsen BJ, Kronborg O, Fenger C, Jorgensen OD. The pattern of recurrent colorectal cancer in a prospective randomised study and the characteristics of diagnostic tests. Int J Colorectal Dis 1997;12:329-334.
Figueredo A, Rumble RB, Maroun J, et al. Follow-up of patients with curatively resected colorectal cancer: a practice guideline. BMC Cancer 2003;3:26-26.
Kattlove H, Winn RJ. Ongoing care of patients after primary treatment for their cancer. CA Cancer J Clin 2003;53:172-196.
Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993;329:1977-1981.
Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999;230:309-318.
Schoemaker D, Black R, Giles L, Toouli J. Yearly colonoscopy, liver CT, and chest radiography do not influence 5-year survival of colorectal cancer patients. Gastroenterology 1998;114:7-14.
Pietra N, Sarli L, Costi R, Ouchemi C, Grattarola M, Peracchia A. Role of follow-up in management of local recurrences of colorectal cancer: a prospective randomized trial. Dis Colon Rectum 1998;41:1127-1133.
Kjeldsen BJ, Kronborg O, Fenger C, Jorgensen OD. A prospective randomized study of follow-up after radical surgery for colorectal cancer. Br J Surg 1997;84:666-669.
Ohlsson B, Breland U, Ekberg H, Graffner H, Tranberg K-G. Follow-up after curative surgery for colorectal carcinoma: randomized comparison with no follow-up. Dis Colon Rectum 1995;38:619-626.
Makela JT, Laitinen SO, Kairaluoma MI. Five-year follow-up after radical surgery for colorectal cancer: results of a prospective randomized trial. Arch Surg 1995;130:1062-1067.
Secco GB, Fardelli R, Gianquinto D, et al. Efficacy and cost of risk-adapted follow-up in patients after colorectal cancer surgery: a prospective, randomized and controlled trial. Eur J Surg Oncol 2002;28:418-423.
Renehan AG, Egger M, Saunders MP, O'Dwyer ST. Impact on survival of intensive follow up after curative resection for colorectal cancer: systematic review and meta-analysis of randomised trials. BMJ 2002;324:813-813.
Jeffery GM, Hickey BE, Hider P. Follow-up strategies for patients treated for non-metastatic colorectal cancer. Cochrane Database Syst Rev 2002;1:CD002200-CD002200.
Stiggelbout AM, de Haes JCJM, Vree R, et al. Follow-up of colorectal cancer patients: quality of life and attitudes towards follow-up. Br J Cancer 1997;75:914-920.
Loprinzi CL, Hayes D, Smith T. Doc, shouldn't we be getting more tests? J Clin Oncol 2000;18:2345-2348.
Deyo RA. Cascade effects of medical technology. Annu Rev Public Health 2002;23:23-44.
Kievit J. Colorectal cancer follow-up: a reassessment of empirical evidence on effectiveness. Eur J Surg Oncol 2000;26:322-328.
Virgo KS, Vernava AM, Long WE, McKirgan LW, Johnson FE. Cost of patient follow-up after potentially curative colorectal cancer treatment. JAMA 1995;273:1837-1841.
Cooper GS, Yuan Z, Chak A, Rimm AA. Geographic and patient variation among Medicare beneficiaries in the use of follow-up testing after surgery for nonmetastatic colorectal carcinoma. Cancer 1999;85:2124-2131.
Earle CC, Grunfeld E, Coyle D, Cripps MC, Stern HS. Cancer physicians' attitudes toward colorectal cancer follow-up. Ann Oncol 2003;14:400-405.
ESMO minimum clinical recommendations for diagnosis, adjuvant treatment and follow-up of colon cancer. Ann Oncol 2001;12:1053-1054.
Benson AB III, Desch CE, Flynn PJ, et al. 2000 Update of American Society of Clinical Oncology colorectal cancer surveillance guidelines. J Clin Oncol 2000;18:3586-3588.
Smith RA, Cokkinides V, von Eschenbach AC, et al. American Cancer Society guidelines for early detection of cancer. CA Cancer J Clin 2002;52:8-22.
Berman JM, Cheung RJ, Weinberg DS. Surveillance after colorectal cancer resection. Lancet 2000;355:395-399.
Winawer S, Fletcher R, Rex D, et al. Colorectal cancer screening and surveillance: clinical guidelines and rationale -- update based on new evidence. Gastroenterology 2003;124:544-560.
Knopf KB, Warren JL, Feuer EJ, Brown ML. Bowel surveillance patterns after a diagnosis of colorectal cancer in Medicare beneficiaries. Gastrointest Endosc 2001;54:563-571.
Kjeldsen BJ, Thorsen H, Whalley D, Kronborg O. Influence of follow-up on health-related quality of life after radical surgery for colorectal cancer. Scand J Gastroenterol 1999;34:509-515.
Renehan AG, O'Dwyer ST, Whynes DK. Cost effectiveness analysis of intensive versus conventional follow up after curative resection for colorectal cancer. BMJ 2004;328:81-81.
NCCN colorectal cancer screening clinical practice guidelines in oncology. J Natl Compr Cancer Network 2003;1:72-93.
Graham RA, Wang S, Catalano PJ, Haller DG. Postsurgical surveillance of colon cancer: preliminary cost analysis of physician examination, carcinoembryonic antigen testing, chest x-ray, and colonoscopy. Ann Surg 1998;228:59-63.
McCall JL, Black RB, Rich CA, et al. The value of serum carcinoembryonic antigen in predicting recurrent disease following curative resection of colorectal cancer. Dis Colon Rectum 1994;37:875-881.
Lennon T, Houghton J, Northover J. What is the value of clinical follow-up for colorectal cancer patients? The experience of the CRC/NIH CEA second-look trial. Nottingham, England: Nottingham International Colorectal Cancer Symposium, 1995. (Accessed May 10, 2004, at http://www.york.ac.uk/inst/crd/ehc36.pdf.)
Green RJ, Metlay JP, Propert K, et al. Surveillance for second primary colorectal cancer after adjuvant chemotherapy: an analysis of Intergroup 0089. Ann Intern Med 2002;136:261-269.
Winawer SJ, Zauber AG, O'Brien MJ, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. N Engl J Med 1993;328:901-906.
Helm J, Choi J, Sutphen R, Barthel JS, Albrecht TL, Chirikos TN. Current and evolving strategies for colorectal cancer screening. Cancer Control 2003;10:193-204.
Lynch HT, de la Chapelle A. Hereditary colorectal cancer. N Engl J Med 2003;348:919-932.
Sandler RS, Halabi S, Baron JA, et al. A randomized trial of aspirin to prevent colorectal adenomas in patients with previous colorectal cancer. N Engl J Med 2003;348:883-890.
Schrag D. Defining optimal treatment for stage II colon cancer: does decision analysis help? Gastroenterology 1999;117:1005-1007.
Rotondano G, Esposito P, Pellecchia L, Novi A, Romano G. Early detection of locally recurrent rectal cancer by endosonography. Br J Radiol 1997;70:567-571.
Willkomm P, Bender H, Bangard M, Decker P, Grunwald F, Biersack H-J. FDG PET and immunoscintigraphy with 99mTc-labeled antibody fragments for detection of the recurrence of colorectal carcinoma. J Nucl Med 2000;41:1657-1663.
Gryfe R, Kim H, Hsieh ETK, et al. Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 2000;342:69-77.
Kemeny N, Huang Y, Cohen AM, et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl J Med 1999;341:2039-2048.
Nordic Gastrointestinal Tumor Adjuvant Therapy Group. Expectancy or primary chemotherapy in patients with advanced asymptomatic colorectal cancer: a randomized trial. J Clin Oncol 1992;10:904-911.
Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure in patients with metastatic colorectal cancer. Lancet 1998;352:1413-1418.
Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000;343:905-914.
Winawer SJ, St John DJ, Bond JH, et al. Prevention of colorectal cancer: guidelines based on new data. Bull World Health Organ 1995;73:7-10.(David G. Pfister, M.D., A)