Progress in the adjuvant treatment of colon cancer: has it influenced clinical practice?
E. Van Cutsem,F. Costa
DOI: https://doi.org/10.1001/JAMA.294.21.2758
2005-12-07
Abstract:COLON CANCER IS A COMMON MALIGNANCY IN ECOnomically developed countries, particularly in North America, Europe, and Australia, and is one of the leading causes of cancer-related deaths in the Western world. Every year, colorectal cancer is responsible for an estimated 400 000 deaths worldwide. Among patients with colon cancer, 70% present with apparently localized disease. In these patients, surgery can be curative, but relapses after complete resection are frequent. Colon cancer is not uniformly fatal and there are large differences in survival depending on the stage of the disease at diagnosis. The pathological stage is currently the most important determinant of prognosis. The classification system described by Dukes in 1930 is still widely used, but it no longer fulfills the requirements of modern tumor staging because it does not take into account distant metastases, number of lymph nodes involved, and carcinomas limited to the submucosa. Therefore, the TNM classification of the American Joint Committee on Cancer (AJCC) is recommended both for daily clinical practice and for clinical trials. The AJCC Cancer Staging Manual is now used to stage colon cancer and all other cancers, and the latest edition stratifies colon cancer stages II and III further by use of T stage (ie, tumor depth of penetration) and N stage. According to Surveillance, Epidemiology, and End Results (SEER) US national cancer registry data from January 1, 1991, through December 31, 2000, for 119 363 patients staged according to the new AJCC sixth edition criteria, the overall 5-year colon cancer–specific survival for this entire cohort was 65.2%. Five-year colon cancer–specific survival by stage was 93.2% for stage I, 84.7% for stage IIA, 72.2% for stage IIB, 83.4% for stage IIIA, 64.1% for stage IIIB, 44.3% for stage IIIC, and 8.1% for stage IV cancer. Data derived from 50 042 patients in the US National Cancer Data Base (from 1987 through 1993) showed a 5-year survival rate of 59.8% for stage IIIA, 42.0% for stage IIIB, and 27.3% for stage IIIC colon cancer. Over the last 15 years major advances have been made in the adjuvant treatment of colon cancer. Since the mid 1990s it is generally accepted that adjuvant treatment of stage III colon cancer (ie, lymph node positive) decreases relapse and improves survival. Demonstration of a survival advantage with a year of adjuvant 5-fluorouracil and levamisole has dramatically changed the outcome for patients with stage III colon cancer. The pivotal study showed a 15% absolute reduction (40% relative reduction) in the risk of recurrence and a 16% absolute reduction (33% relative reduction) in the overall death rate with a combination of surgery plus 5-fluorouracil and levamisole in patients with stage III colon cancer. Subsequently, it was demonstrated that 5-fluorouracil and folinic acid for 6 months was at least as active as 12 months of 5-fluorouracil and levamisole and was not more toxic. In this issue of JAMA, Jessup et al report the influence of adjuvant chemotherapy recommendations for patients with stage III colon cancer in the community. Using data from the National Cancer Data Base between 1990 and 2002, 85 934 selected cases from 560 US hospitals were analyzed; this covers the time when both 5-fluorouracil and levamisole and 5-fluorouracil and folinic acid were the only proven options in the adjuvant setting. The analysis shows significant changes with respect to both pattern of care and survival benefit: adjuvant treatment use increased from 39% in 1991 to 64% in 2002 and, compared with surgery alone, the survival increased from 8% improvement in the first 1991 subgroup of patients receiving adjuvant therapy to 16% in the 1997 subgroup that received adjuvant therapy. Because the analyzed data were not obtained from prospective randomized clinical trials, conclusions about treatment effectiveness in different patient subgroups are limited and should be interpreted with caution. The