Introduction
Colorectal cancer is a major cause of morbidity and mortality in British Columbia and the world.[1] Approximately two-thirds of all cases are operated on with curative intent and, in these cases, the literature cites a frequency of recurrences that varies from 30% to 50%.[2] For many years the standard of care for patients with these cancers has included regular follow-up after diagnosis and initial treatment. Every edition of the BC Cancer Agency’s Cancer Treatment Policy Manual[3] has provided detailed recommendations from the Gastrointestinal Tumor Group for follow-up. This article provides a review and update on this topic done by an ad hoc subcommittee of the Tumor Group. The recommendations at the end of the article have been reviewed and approved by the whole Tumor Group.

The primary purposes of follow-up
There are six primary reasons to follow up patients with colorectal cancer:

• To ensure that problems due to surgery or therapy are identified and managed. 
• To detect the development of neoplastic polyps or second primary neoplasms, which occur with increased frequency in patients already having one cancer diagnosed. 
• To identify, at an early stage, evidence of disease recurrence that may be amenable to salvage by secondary surgical procedure, again with curative intent. 
• To detect and offer palliative therapy to patients with symptomatic recurrences. However, the majority of recurrences are not detected during routine follow-up, but rather between routine visits, when the patient presents to their physician with new symptoms. 
• To provide support and counseling. 
• To assess the natural history of the disease and/or the results of therapy (outcome analysis).

This is not an exhaustive list.

Why follow up?
Follow-up of patients with resected colorectal cancers with physical examination, laboratory tests, or imaging procedures rarely identifies curable recurrences or metastases. If there is no survival benefit why should follow-up take place? Is regular follow-up of the 70 to 80 resected patients justified to pick up the two patients who are salvaged from recurrence for 5 years? Only one of the purposes of follow-up is related to salvaging patients with recurrences. The other five purposes are of equal value. Issues such as identifying treatable postoperative complications, detecting metachronous bowel cancers and premalignant polyps, providing psychosocial support and counseling, and gathering data about the natural history of the disease and the results of treatment need to be considered when making recommendations regarding the frequency and nature of follow-up. Also, although the value of rigorous follow-up remains controversial, we should not forget that a small proportion of patients with potentially curable recurrences or second primary cancers may be found. The following suggestions are intended to encourage standardized follow-up while at the same time maximizing its benefits without wasting resources.

Results of follow-up studies
Five randomized trials comparing intense to routine or no follow-up have been published.[4,5,18-20] Two meta-analyses have been done.[6,21] Table 1 [11] summarizes the results.

The Bruinvels meta-analysis[6] pooled seven non-randomized studies comparing intensive follow-up to groups of concurrent control patients on routine or no follow-up. In this study, groups of patients closely followed with carcinoembryonic antigen (CEA) determinations had a 9% better 5-year survival. No other tests were found to be useful. The methodology has been criticized, as individual patient data was not accessed and none of the studies were randomized.

A large randomized Swedish trial[7] comparing a group of patients followed with intensive follow-up with endoscopy but without CEA determinations to a control group is completed but not reported. A more detailed analysis of the usefulness of CEA determinations is presented further on in this article.

Based on the published studies, most authors[2,4-6,8,9,22] conclude that more intensive follow-up using routine imaging techniques and laboratory testing cannot be supported on either a clinical or cost/benefit basis.[10] Pietra[19] and Rosen[21] do not agree and recommend intensive follow-up.

Follow-up recommendations
The following recommendations are summarized in Table 2 [12].

Frequency
For patients who have had their primary tumors resected for cure (whether they have had adjuvant radiotherapy or chemotherapy or not), it is recommended that they be seen every 3 months for the first 2 years, every 6 months for the next 3 years, and annually thereafter.

History and physical examination
On each follow-up visit, a history and physical examination should be done. This is directed at evidence of recurrent disease and residual effects of surgery. The history should specifically focus on the GI tract by asking direct questions about such issues as weight loss, blood loss, abdominal pain, and so on. During the physical examination attention should be paid to the patient’s nutritional status, liver, and lymph nodes, particularly in the left supraclavicular fossa. Rectal examination or inspection and palpation of perineal scar after abdominal perineal resection is important.

Evaluation of the remaining colon and rectum
Three to eight percent of patients with resected colorectal cancer will develop a metachronous cancer,[2] and early detection will increase the chance of cure. Twenty-five percent to forty percent will develop a polyp,[10] the early detection and removal of which may prevent a second cancer from developing. The best method of surveillance is by colonoscopy.

Whenever possible a colonoscopy should be performed during the preoperative period to “clear” the colon of synchronous neoplasia, preferably before surgery so that all cancers and polyps may be included in the resection. If not done prior to surgery, a colonoscopy should be done at 6 months after surgery and, if negative, again at 3 years postoperatively and thereafter at intervals of every 5 years. If the patient is not suitable for colonoscopy, sigmoidoscopy combined with an air contrast barium enema is an alternative. These intervals have been chosen because it takes from 5 to 10 years for a polyp to transform to a malignant neoplasm.

Patients who have undergone sphincter sparing low anterior resection for rectal cancers or who have had local excision only should have serial digital examinations and proctoscopy (but not colonoscopy or barium enema) to look for anastomotic recurrence. These should be done at 3 months, 6 months, 1 year, and 2 years after surgery. In other types of colon resections above the rectum, anastomotic recurrences are extremely rare, so colonoscopy to detect local recurrence is not indicated.

Laboratory tests
There is no evidence that standard laboratory tests such as blood counts and liver enzymes help detect disease at a resectable stage.

CEA
If the condition of the patient is such that they can tolerate a major operation such as a liver resection, it is recommended that postoperative serum CEA testing be performed every 3 months in patients with Stage II or III disease for at least 2 years after diagnosis. An elevated CEA level, if confirmed by retesting, warrants further evaluation for metastatic disease but does not by itself justify the institution of systemic therapy for presumed metastatic disease. It should be noted that the question of the use of the carcinoembryonic antigen remains controversial. Its use should be limited to those patients who could benefit from having metastatic disease in the liver or lung resected.

The American Society of Clinical Oncology Tumor Marker Panel[11] recommended that CEA determinations be done every 3 months for the first 2 years if the patient is suitable for major salvage surgery. The Bruinvels meta-analysis[6] of non-randomized studies noted a 9% survival advantage for CEA monitored patients. A large (n = 1447) randomized British trial[12] comparing CEA monitoring with no CEA has completed entry. Preliminary results[13] demonstrate no survival difference. Moertel[15] believed CEA use is not justified, though his methodology has been criticized.

There are several issues to bear in mind when considering CEA testing:

• False negatives are common, and a negative CEA is not informative and does not exclude a relapse because approximately 30% of all colorectal cancer recurrences do not produce CEA. False negative CEA test results are more common in poorly differentiated tumors. 
• False positives unassociated with disease recurrence are not infrequent and may be associated with smoking, adjuvant chemotherapy, or various other medical conditions. CEA levels of below 10 that are stable on repeat determination are more likely to be due to non-malignant causes. CEA levels that are rising on repeat determinations done several weeks apart is suggestive of a more ominous situation. 
• If the test is included in the follow-up regimen, elevated levels should promptly lead to assessment for potentially curable disease, i.e., physical examination, chest X-ray, abdominal ultrasound, abdominal CT in patients with colon cancer, and combined pelvic and abdominal CT in those patients with rectal cancer. A barium enema or colonoscopy should be done to evaluate the status of the bowel. 
• The majority of patients who have elevated CEA levels are found to have extensive unresectable metastases, and there is probably minimal benefit in identifying such relapses before they become clinically detectable.

Diagnostic imaging
No imaging procedures have been demonstrated to be of value in improving survival.[6,15] 

Metastatic disease
For patients who present with non-resectable metastatic disease, follow-up will be dependent upon their physical condition, problems, and therapy.

Who should carry out the follow-up?
Follow-up may be done by the family physician, the local specialist, or a BC Cancer Agency physician, depending on the circumstances. This depends upon the location of the cancer, stage, treatment provided, age, and place of residence of the patient. For example, a patient who has had a local excision or an anterior resection of a rectal cancer should be followed by a physician who is experienced in proctoscopy.

Dr Klaassen is a clinical professor in the Department of Medicine at UBC and staff medical oncologist at the British Columbia Cancer Agency, Vancouver Centre. Dr Weinerman is an honorary professor in the Department of Medicine at UBC, an adjunct professor in the Department of Biology at the University of Victoria, and the regional vice-president and medical oncologist at the British Columbia Cancer Agency, Vancouver Island Cancer Centre. Dr Lee is a clinical instructor in the Department of Medicine at UBC and a medical oncologist at the British Columbia Cancer Agency, Fraser Valley Cancer Centre. Dr Taylor is a clinical associate professor in the Department of Surgery at UBC, a staff surgeon at Mount Saint Joseph’s Hospital, and a consulting surgeon at the British Columbia Cancer Agency.