A Pooled Analysis of Adjuvant Chemotherapy for Resected Colon Cancer in Elderly Patients
A Pooled Analysis of Adjuvant Chemotherapy for Resected Colon Cancer in Elderly Patients
Abstract & Commentary
Synopsis: The balance of chemotherapy efficacy with side effects among elderly colon carcinoma patients continues to spur debate. Results from 7 randomized trials comparing surgery alone to postoperative fluorouracil-based regimens for stage II and III colon carcinoma were pooled to investigate the presence of interactions between chemotherapy use and age. Adjuvant therapy provided a significant overall and recurrence-free survival benefit that persisted in each age category, with no reduction in efficacy among older age groups. While leukopenia was significantly increased among patients older than 70 years of age (P = 0.001 with levamisole; P = 0.05 with leucovorin), no other interaction between age and toxicity was observed. These data support the conclusion that selected elderly colon cancer patients at high risk for recurrence may experience adjuvant treatment benefits equivalent to those enjoyed by younger patients with similar disease.
Source: Sargent DJ, et al. N Engl J Med. 2001;345: 1091-1097.
As the post-world war ii generation ages, swelling the ranks of Americans older than age 65 by 57% between 2000 and 2025, physicians can expect to treat substantially increasing numbers of elderly patients.1 Moreover, according to the Surveillance, Epidemiology, and End Results database of the National Cancer Institute, the largest increases in cancer incidence are occurring among older age groups. Between 1973 and 1991, for example, overall cancer incidence rates for age groups 65-74 and > 75 increased by 35% and 28%, respectively. Given these changing demographics and evidence of ongoing questions regarding age-related use of adjuvant treatment,2 Sargent and colleagues undertook a pooled analysis of randomized trials comparing postoperative fluorouracil-based therapy for Stage II/III colon carcinoma to treatment with surgery alone, and assessing benefits and toxicities among age categories.
A systematic search of Medline and bibliographies was performed, followed by discussions with primary investigators of identified appropriate trials. Seven studies met inclusion criteria, yielding 3351 total eligible patients.3-7 Five studies randomized patients to receive postoperative fluorouracil plus leucovorin for 1 year (4) or 6 months (1) vs. surgery alone, while 2 randomized to fluorouracil plus levamisole for 1 year. Doses of fluorouracil ranged from 370-450 mg/mL2. Leucovorin dose was 200 mg/mL2 in 4 studies and 20 mg/mL2 in one study. Levamisole dose was 50 mg/mL2 in the remaining 2 studies. All studies identified overall and recurrence-free survival as primary end points, recorded toxicities per World Health Organization or National Cancer Institute grading scale, and followed patients for more than 5 years. Patients were initially divided into age groups of equivalent size. After statistical analysis revealed identical results when divided into age groups of 10-year increments, results were reported by decade (< 50 yr, 51-60 yr, 61-70 yr, > 70 yr).
Not surprisingly, while the probability of death with recurrence of cancer was roughly equivalent among age groups (29-33%), probability of death without known recurrence monotonically increased with older age group. Among patients < 50 years of age, death rate without recurrence was 2%, contrasted to a 13% death rate in the > 70 year age group.
Studies were examined for differences in treatment effect and found to have no heterogeneity for either primary outcome. Although 4 studies did not reach statistical significance, the pooled analysis revealed improved 5-year survival rates for subjects undergoing adjuvant treatment compared with surgery alone (71% vs 64%, respectively; P = 0.001). Similarly, although 2 studies did not reach statistical significance, pooled analysis of patients in the adjuvant arm experienced improved 5-year recurrence free survival rates (69% vs 54%; P = 0.001). When stratified by age, no significant interaction between age group and primary outcome could be identified.
Because studies did show significant heterogeneity for toxicities, analyses of grade 3 or higher toxicities were performed separately for levamisole and leucovorin regimens. The only severe toxicity associated with increased age was leukopenia among patients > 70 years who were treated with either fluorouracil plus levamisole (P = 0.001) or fluorouracil plus leucovorin (P = 0.05).
Sargent et al assert that a perception of greater toxicity or poor treatment tolerance may mitigate use of chemotherapy among the elderly. Based on the data presented, they conclude that selected elderly patients at high risk for recurrence of colon carcinoma will derive the same survival benefit as younger patients, without a significant increase in treatment-associated toxicity.
Comment by Arden Morris, MD
Fluorouracil-based adjuvant therapy has become the standard of care for patients with Stage III colon cancer. However, practitioners (and possibly patients themselves) have been reluctant to treat elderly patients with the same regimens as younger patients,2 presumably due to concerns regarding decreased benefit and/or high toxicity. Furthermore, physicians may be reluctant to risk severely toxic side effects in a patient who may have a shorter prognosis simply due to advanced age. While these concerns are certainly valid in the setting of selected malignancies and regimens,8 no evidence exists that colon tumor biology differs within different age groups. Age-related toxicity data for adjuvant treatment among colon cancer patients remains sparse. Additionally, in the face of increasing life expectancy and population growth of elderly Americans, concerns that adequately aggressive treatment may not be provided deserve attention.
Using pooled data from rigorously conducted phase III trials, Sargent et al have provided important information regarding appropriate treatment for elderly colorectal cancer patients. Overall, patients experienced no significant difference in survival benefit with increasing age. Type II error, or inability to demonstrate a significant effect where one truly exists, is a risk among underpowered studies. Sargent et al addressed this potential issue with power calculations demonstrating that the pooled data achieved 80% power to discern a significant interaction at a hazard ratio of 1.4, or a risk reduction of 40% in younger patients and no risk reduction in elderly patients. It should be noted that fluorouracil-based chemotherapy use in high-risk Stage II and Stage III colon cancer patients conferred an overall risk reduction of one third that of surgery alone, or a hazard ratio of 1.5 for those undergoing surgery alone.5 Patient results were not stratified by stage, possibly due to issues of power loss due to decreased sample size.
Among the limitations of pooled data are potential bias due to differences in accrual, application of regimen, and follow-up techniques. Sargent et al attempted to control for these issues by communicating closely with investigators in each study included and by testing for and finding no significant heterogeneity in treatment effects. Moreover, results among the studies themselves justified pooling despite some of the protocol differences. For example, 6- vs. 12-month duration of treatment was shown to have no association with survival time.7
Because toxicities were significantly different among different studies, protocols using leucovorin were analyzed for toxicity separately from those using levamisole. Contrary to other chemotherapeutic regimens,8 little significant increase in toxicity manifested as age increased. A highly significant increase in leukopenia was seen with levamisole use among patients of age > 70 years, however.
Lastly, pooled data certainly inherit limitations of the original studies. Although phase III trials are the criterion standard for controlling confounding, pragmatic issues cannot be ignored. As Sargent et al acknowledge, patients selected for these protocols met inclusion criteria that required a minimum performance status and support system. Especially among the elderly, potentially subject to higher rates of mental illness, poverty, and debilitation, these results may not be entirely generalizable but are important to consider among normally or highly functioning patients.
References
1. US Bureau of the Census, International Database. 2000, U.S. Bureau of the Census.
2. Mor V, et al. J Am Geriatr Soc. 1985;33(9):585-589.
3. International Multicentre Pooled Analysis of Colon Cancer Trials (IMPACT) investigators. Lancet. 1995;345(8955):939-944.
4. Francini G, et al. Gastroenterology. 1994;106(4):899-906.
5. Laurie JA, et al. J Clin Oncol. 1989;7(10):1447-1456.
6. Moertel CG, et al. N Engl J Med. 1990;322(6):352-358.
7. O’Connell MJ, et al. J Clin Oncol. 1998;16(1):295-300.
8. Balducci L, Extermann M. Cancer. 1997;80(7):
1317-1322.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.