Follow up after colorectal cancer surgery is controversial with regards to its benefits. The main reasons for this are: (1) conflicting evidence from different studies, (2) inability to compare results from various studies due to the variation in follow up practices, (3) lack of sufficient randomised trials and (4) inadequate number of patients in the study group.

With lack of clear evidence either way it was impossible to say whether following up patients after their primary resection was worthwhile or not. However the publication of two fairly recent meta-analyses, have given some clear evidence supporting the role of intensive follow up after colorectal cancer surgery. This has to be taken positively, but with some caution until additional evidence can be gathered from further study.

A significant consideration has been the cost involved in follow up and a concern for the individual health trusts when allocating budgets. With a combination of insufficient data supporting follow up and high costs involved, the governmental based reports had presented a pessimistic attitude and recommended that follow up programmes are not worthwhile. Funds could better spent. These conclusions may not be valid in light of the recent meta-analysis reports.

As this entire subject is still controversial, I decided to review the evidence published so far. The initial part of this presentation is concentrated on the facts regarding colorectal follow up. The latter part shows the evidence and conclusions from major studies and reports.

One of the main reasons for follow up is to detect recurrent disease following a "curative" resection. The purpose of follow up is to detect recurrence at an early stage so that patients could have effective treatment either by means of further surgery or chemotherapy.

The overall recurrence rates quoted in the literature are between 30-50%. If we go by the Dukes stage the recurrence rates are higher with Dukes C (70%), than Dukes A (15%). It is also important to appreciate that majority of the recurrences (80%) occurs within the first two years after primary surgery and is extremely rare after five years. This has a major influence especially in the design of follow-up protocols.

There are two major sites of recurrences with colorectal cancer (Loco-regional and distant liver recurrences). Loco-regional recurrence include those at the anastomotic site and in the tissues adjacent to the anastomotic site especially pelvic organs. Among the distant metastatic sites, the most common organ involved is the liver, followed by the lung.

It has been estimated that up to 50% of the patients who undergo curative resection will develop recurrences at some stage. The incidence of loco-regional recurrences as shown by Barillari et al, are around 10% and it has been shown that 50% of these patients have symptomatic recurrences at presentation. Evidence also has shown that 60-90% of these patients have evidence of distant spread in addition to the loco-regional spread. This would make the chance of a further curative resection less likely. Studies have shown that curative resections are possible in only around 28% of patients and even among them there is 50% chance of a further recurrence developing.

Liver metastasis develops in nearly 40% of patients after "curative" resection for a colorectal cancer. A small subset of patients however appears to develop solitary or small number of metastases restricted to a single organ. Extirpation of metastatic disease in carefully selected patients can result in cure. Of the 12% to 20% of patients with colorectal liver metastasis and no evidence of extrahepatic disease, 45% to 55% have disease that is potentially resectable for cure. For patients who undergo liver resection the
5 year survival rates has improved and is around 25-35%.

Lung metastasis develops subsequently in 10% to 20% of patients after their curative resection. Only 1.3% to 4% of the entire group of patients who develop pulmonary metastasis are considered candidates for pulmonary resection.

The second main reason for follow-up of colorectal cancer patients is to identify patients who develop metachronous lesions. The published annual incidence of anyone developing a metachronous cancer is 0.35%. Cali et al, identified that the cumulative incidence at 18 years following the primary surgery is 6%. It has been shown that the incidence of developing these cancers is highest between 5 and 12 years after the first cancer.

Metachronous adenomas develop in a significant number of patients (up to 50% in some series) and can develop on multiple occasions; they have the potential to become malignant. There is evidence that removing adenomas at colonoscopy reduces the likelihood of carcinomas developing.

There have been some studies on the patient's attitudes towards attending follow-up clinics. Though some papers have suggested that there are higher anxiety levels among patients for a few days prior to their clinic visit, the majority of them felt reassured after attending the clinic and being told that they do not have evidence of any recurrent disease. It has also been shown that patients would prefer follow up even if it did not lead to earlier detection of recurrences.

When looking at the follow-up methods adopted in the different studies, especially to the late 1990s there was no uniform protocol and there was much variation among what constituted a minimal and intensive follow-up. In 1997, the EORTC-GITCCG group clearly defined a minimal and an intensive follow-up schedule. This has been helpful as it is now creates a common platform for the design of a follow-up programme. It has become easier to compare and merge results from different studies, especially when conducting a meta-analysis.

Clinical examination is usually carried out at each clinic visit and this entails an appropriate history taking, abdominal & rectal examination. In cases of a previous rectal cancer, rigid sigmoidoscopy is carried out at most centres. This is to mainly check for any recurrences at the anastomotic site (in cases of relatively low anastomosis. Though the main purpose of designing a follow-up schedule was to try and identify patients with asymptomatic recurrences, it has been shown from various studies that by the time recurrences are identified, majority of patients would have developed symptoms, which would have prompted appropriate investigations.

CEA has been found to be the most cost-effective approach in identifying potentially resectable metastases from colonic cancer. It is the only available tumour marker and it gives the first indication of a tumour recurrence in nearly 60% of patients. The sensitivity is around 80% though it can be variable. As identified by Schofield, it is important to recognise that 30% of patients with colorectal cancers do not express the CEA antigen. Evidence also suggests that CEA is a better marker for picking up distant recurrence rather than loco-regional recurrences. McCall et al, reported that time interval between an elevation of CEA level and the occurrence of symptoms can be variable (1-30 months), but the average time appears to be 6 months.

The main reasons for performing a colonoscopy is to (1) identify patients with metachronous polyps, (2) to detect metachronous cancers and (3) to check the anastomotic site for recurrences. Colonoscopy was performed on an annual basis but there is now clear evidence that there is no role for annual colonoscopy. This is based on the better understanding of the polyp-carcinoma sequence. The current evidence suggests that colonoscopy should be carried out at 3-5 yearly intervals after making sure that the residual colon is free of synchronous polyps / tumours either around the time of the operation or 6 months later. Patients who did not undergo complete colonoscopy or barium enema before surgery should be offered a colonoscopy within six months of the primary operation.

Sigmoidoscopy or proctoscopy should be done in addition to rectal examinations during the clinic visits, mainly in patients who had a previous anterior or sigmoid resection to check the anastomotic site for recurrences.

Imaging of the liver, by means of an Ultrasound / CT scan is carried out at 3-4 monthly intervals during the first 2-3 years, as the chance of a recurrence developing is highest during this time period. Following this, the scans can be performed at 6-12 monthly intervals until the 5th year. The current evidence does not support for a regular chest X-ray, but if patients have an elevated CEA levels, then a CXR should be performed to exclude pulmonary metastasis.

Though the main reason for following up patients after a curative resection is to identify asymptomatic recurrences, but studies have shown that despite follow-up only 45-55% have asymptomatic recurrences when detected at routine follow up. A significant proportion of patients will develop symptoms between clinic visits prompting further investigations.

The role of CEA has been the most debatable issue. The opinion seems to be divided on its benefit. The problem is that although CEA can detect recurrences (not always) before symptoms have developed, it raises the possibility of second-look exploratory surgery in cases where the investigations do not pick-up the site of recurrences. This has been shown not to be worthwhile as there is a significant morbidity attached to this type of surgery. This approach has been abandoned from clinical practice in general. The current evidence suggests that if CEA is elevated in an asymptomatic patient, efforts should be made to detect the site of recurrent disease by means of investigations rather than second-look surgery.

The most recent meta-analysis published by Renehan et al in the BMJ 2002 (April issue), which included 1342 patients from five trials, reported that intensive follow up was associated with a reduction in all cause mortality (combined RR 0.81, 95% confidence interval 0.70 to 0.94, P=0.002). The effect was most pronounced in the four extramural detection trials that used CT and frequent measurements of serum CEA (RR 0.73, 0.60 to 0.89, P=0.002). They also identified that intensive follow up was associated with significantly earlier detection of all recurrences (difference in means 8.5 months, 7.6 to 9.4 months, P<0.001) and an increased detection rate for isolated local recurrences (RR 1.61, 1.12 to 2.32, P=0.011). Their analysis showed that using modern follow up regimens (including CT or frequent measurements of serum CEA, or both) there was an absolute reduction in mortality of 9-13%. This would compare favourably with the 5% benefit observed for adjuvant chemotherapy in Dukes C patients.

The meta-analysis published by Rosen et al in 1998, which included 2005 patients (from two randomized and three comparative-cohort studies), found that in the intensive follow-up group:

· Cumulative 5-year survival was 1.16 times higher (P=0.003)
· Two and a half times more curative re-resections were performed for recurrent cancer (P=0.0001)
· Survival rates for patients with recurrent cancer was 3.62 times higher than the control (P=0.0004)

However they were not able to demonstrate significant benefit in survival rates among patients who underwent curative resection for recurrent cancers.

Follow-up of patients with colorectal cancer represents a significant expenditure in the health budget. With limited resources available and the cost-effective measures in place for each individual trusts, this is to be seriously considered. The Norwegian group have estimated the costs of an intensive follow up programme as shown above.

The conclusions of various reports and important studies are summarised below.

Though the NHS centre for review in their report in 1997 suggested that there was insufficient data to justify routine follow up, there seems to be clear evidence since then (based on two meta-analyses and a formidable review article by Kievet) that intensive follow up is useful and does improve the survival rates.

The meta-analysis by Brunivels et al included patients from non-randomised trials of varying intensities of follow up. Though they noticed a significant increase (9%) in the 5-year survival rates, they accepted that no definite conclusions could be drawn because of the weaknesses in the evidence.

There seems to a wide variation between government-based reports and the findings from the meta-analyses. What the meta-analyses have shown is that when they pooled data from various randomized studies and other important studies they have been able to show better survival rates in the patients subjected to intensive follow up. The latest meta-analysis showed an absolute reduction in mortality of 9-13%, which translates to improved survival rates. This is more than comparable to the 5% benefit with chemotherapy for colorectal cancer patients. With the introduction of multi-disciplinary meetings for these patients, it is very likely that the quality of care would increase and these may help towards further improvement in the curative resections and survival rates.

In conclusion, follow up of colorectal cancer patients after a curative resection still appears to be controversial, but with the recent evidence emerging especially from meta-analyses, there seems to be some substantial benefit for the patients in terms of improved survival rates, at least equal to benefits from adjuvant chemotherapy.

In order to determine the best follow-up strategy for different Dukes stages, extremely large trials would be required and none of the individual, previous studies have recruited sufficient numbers, though recently planned studies will, if recruitment targets are met. The large number of patients required, the length of time the study must run, variations in follow-up practices in different countries, and the problems of managing patients on different follow-up strategies within the same centre all pose problems for the successful outcome of a randomized trial. These problems are are not insurmountable. Although such a trial will require considerable international cooperation, its conclusions will have wide ranging implications and potential benefits. Recent evidence is promising and shows significant benefits with intensive follow up.

References:
Kings Fund Forum. Cancer of the colon and rectum: Consensus statement. Br J Surg 1990; 77: 1063-1065

Rosen M, Chan L, Beart RW Jr, Vukasin P, Anthone G. Follow-up of colorectal cancer: a meta-analysis. Dis Colon Rectum 1998 Sep; 41(9): 1116-26

Scottish Health Purchasing Information Centre. Follow up in colorectal cancers. 1999

Barillari P, Ramacciato G, Manetti et al. Surveillance of colorectal cancer. Effectiveness of early detection of intramural recurrences on prognosis and survival of patients treated fro cure. Dis Colon Rectum 1996; 38:288-93.

Cali RL, Pitsch RM, Thorson AG, Watson P, Tapia P, Blatchford GJ et al. Cumulative incidence of metachronous colorectal cancer. Dis Colon Rectum 1993; 36: 388-393

Safi F, Link KH, Beger HG. Is follow-up of colorectal cancer patients worthwhile? Dis Colon Rectum 1993; 36: 636-642

Stiggelbout AM; de Haes JC; Vree R; van de Velde CJ; Bruijninckx CM; van Groningen K; Kievit J. Follow-up of colorectal cancer patients: quality of life and attitudes towards follow-up. Br J Cancer 1997; 75(6): 914-20

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(5): 509-15

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; 37:875-81.

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 Jan; 114(1): 7-14

Tornqvist A, Ekelund G, Leandoer L. The value of intensive follow up after curative resection for colorectal carcinoma. Br J Surg 1982; 69:725-8.

Northover J, Centre for Reviews & Dissemination, Effective Health Care Bulletin 1997

Tobaruela E; Enriquez JM; Diez M; Camunas J; Muguerza J; Granell J. Evaluation of serum carcinoembryonic antigen monitoring in the follow-up of colorectal cancer patients with metastatic lymph nodes and a normal preoperative serum level. Int J Biol Markers 1997 Jan-Mar; 12(1): 18-21

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 met-analysis of randomised trials. BMJ 2002; 324(7341): 813

Norwegian Gastro Intestinal Cancer Group (NCICG), 1997

NHS Centre for Reviews and Dissemination. The management of colorectal cancer. Effective health care bulletin 1997 Vol 3 (6).

SIGN. Colorectal cancer. Guideline no 16, 1997

Bruinvels DJ. Stiggelbout AM, Kievet J et al. Follow up of patients with colorectal cancer: a meta analysis. Annals of Surgery 1994; 219(2): 174-82

Kievit J. Colorectal cancer follow-up: a reassessment of empirical evidence on effectiveness. Eur J Surg Oncol 2000 Jun;26(4):322-8.