to the editor: I read with interest the article “Care of Cancer Survivors”¹ in the February 15, 2005, issue of American Family Physician. I was surprised to see such a firm recommendation for aggressive follow-up of fully resected colon cancer with inaccurately documented references. The authors recommend carcinoembryonic antigen monitoring every three months for the first two years following treatment, and then every six months for the next three years; this is controversial and not a fully endorsed recommendation. According to my investigation, the National Comprehensive Cancer Network is the only medical group recommending routine carcinoembryonic antigen monitoring testing, and no major medical organization recommends routine computed tomography (CT) scanning as listed in the article.¹ Also, the references listed in the article¹ far from fully endorse routine aggressive follow-up with carcinoembryonic antigen monitoring testing and/or CT scanning.

The latest Cochrane article I could find is not from 2004, as noted in the article,¹ but was last updated in 2002² and in no way firmly recommends aggressive follow-up. In its summary recommendations, it states: “Because of the wide variation in the follow-up programmes used in the included studies, it is not possible to infer from the data the best combination and frequency of clinic (or family practice) visits, blood tests, endoscopic procedures and radiological investigations to maximi[z]e the outcomes for these patients. Nor is it possible to estimate the potential harms or costs of intensifying follow-up for these patients.”²

The other reference³ listed by the authors does not recommend aggressive follow-up with carcinoembryonic antigen monitoring and CT scanning either but discusses the limitations of present evidence in decision making.

A 2004 review article⁴ discusses the controversy around aggressive follow-up and compares the recommendations of individual organizations.

The primary listed reference from Cochrane presents a more balanced overall recommendation: “The results of this review support the general principle of clinical follow-up for patients with CRC [colorectal cancer] after curative treatment. The exact details of the optimal follow-up regimen still need clarification.”²

IN REPLY: We would like to thank Dr. Lawson for his remarks. Several of his comments are already addressed in our review.¹ Specifically, with regard to surveillance following treatment for colorectal cancer, we stated “it is not possible to infer an optimal combination of tests or frequency of clinical follow-up for intensive colorectal cancer surveillance.”¹

We have cited general recommendations for follow-up of patients with colorectal cancer endorsed by the National Comprehensive Cancer Network (NCCN). The NCCN guidelines² are in line with recommendations by the American Society of Clinical Oncology (ASCO) and the American Society of Colon and Rectal Surgeons (ASCRS). NCCN and ASCO recommend carcinoembryonic antigen (CEA) testing every three months for at least two years (ASCO recommends testing for at least two years; NCCN for two years, then every six months for three more years), whereas ASCRS recommends monitoring CEA levels a minimum of three times per year during the first two years of follow-up. A recent pooled analysis³ from 17 adjuvant randomized trials showed that more than 25 percent of recurrences occur beyond three years following surgery. Thus, we feel that a five-year follow-up, as recommended by NCCN, is favored.

Computed tomography (CT) remains a controversial modality of colorectal cancer surveillance and is not recommended routinely by any organization at this point. However, recent studies^4,5 suggest that early asymptomatic recurrences can be detected by CT in the absence of CEA elevations. A significant number of these recurrences are amenable for curative resection, leading to an improved survival in comparison with unresectable patients.^4,5 At this time, CT cannot be recommended routinely as a surveillance modality for colorectal cancer. CT scans should be obtained in the work-up of symptomatic patients or in the presence of elevated CEA and can beconsidered in patients at high risk of recurrence, such as those with stage III disease.

The importance of rigorous surveillance in colorectal cancer is based on the prospect of salvage surgery at the time of recurrence. Resection of isolated recurrences can lead to a five-year survival of about 30 percent,⁵ whereas five-year survivorship following chemotherapy in patients with unresectable recurrent disease is well below 10 percent. The ability to achieve a successful curative resection is highly compromised in the setting of symptomatic detection. This is supported by a study⁴ of 154 relapses seen in a randomized phase III study of adjuvant fluorouracil/leucovorin in patients with stage II and III disease. At the time of publication of this study, none of the patients with resected symptomatic recurrences were alive at five years, whereas 18.6 and 25.9 percent of those with resected CEA recurrences and resected CT recurrences, respectively, remained alive. Whether the implementation of an intensive surveillance schedule that includes frequent CT scans is affordable to society is yet to be determined and should be the subject of future cost-effectiveness analysis studies.

Finally, the publication year listed for the Cochrane review⁶ reference in the article was inadvertently changed; the correct publication year for this work is 2002.