Correspondence
Adjuvant Systemic Therapy and Survival after Breast Cancer
N Engl J Med 1994; 331:402-404August 11, 1994
- Article
To the Editor:
Olivotto and colleagues (March 24 issue)1 attempt to isolate the effects of policies regarding treatment by choosing single years in which other factors that might influence survival, such as screening activity, are thought to be approximately constant. This approach can never be entirely satisfactory, since survival rates are the products of complex processes involving the quality of data (the completeness of registration, the accuracy of recorded diagnoses, changes in the classification of tumors, and the completeness of follow-up), statistical artifacts (lead-time bias), host factors (improved general health and greater public awareness of cancer), and health care factors (screening, diagnosis and early detection, innovations in treatment, and the organization and availability of treatment).
In Scotland, a more or less continuous improvement in rates of survival after breast cancer was observed from 1968 through 1987 (Figure 1Figure 1
Survival of Patients with Breast Cancer in Scotland, According to Age and Period of Registration.), particularly among women 35 to 44 years of age2. This period predated national screening for breast cancer. In the age groups from 30 to 74 years old, the incidence rate in the same period rose by about 5 percent every five years, whereas mortality rose by only 2 percent3. Among women 30 to 44 years of age, the incidence rose by an average of 2 percent every five years, but mortality fell by 7 percent. Adjuvant chemotherapy may have contributed to this improvement, but the continuous increase in survival in all age groups over a 20-year period underscores the multifactorial nature of population-based survival rates.3 ReferencesR. Sankaranarayanan, M.D.
Roger J. Black, M.A.
International Agency for Research on Cancer, 69372 Lyon, France1
Olivotto IA ,Bajdik CD ,Plenderleith IH , et al. Adjuvant systemic therapy and survival after breast cancer. N Engl J Med 1994;330:805-810
Full Text | Web of Science | Medline2
Black RJ, Sharp L, Kendrick SW, eds. Trends in cancer survival in Scotland, 1968-1990. Edinburgh, Scotland: National Health Service, 1993.
3
Coleman MP, Esteve J, Damiecki P, Arslan A, Renard H, eds. Trends in cancer incidence and mortality. Lyon, France: International Agency for Research on Cancer, 1993:411-32. (IARC scientific publications no. 121.)
To the Editor:
Olivotto et al. observed that in British Columbia the survival of women with breast cancer improved significantly from 1974 through 1984, the period when adjuvant chemotherapy became widely used. They suggested that this improvement was a result of adjuvant chemotherapy. In 1987 the General Accounting Office reported an increase in survival after breast cancer from 1950 through 19821. In the opinion of that panel, earlier detection of the disease was the chief factor in the increased survival. The Commission on Cancer of the American College of Surgeons reported that in 1990, as compared with 1983, patients were treated at an earlier stage of the disease, probably because of the use of mammography2. Similar observations have been made by the American Cancer Society3.
The study by Olivotto et al. controlled for neither tumor size nor stage of disease, and the results are clearly subject to lead-time bias. It did not even compare treated patients with untreated patients. It is ridiculous to suppose that chemotherapy has improved the disease-specific seven-year survival of all patients with breast cancer in British Columbia from 65.2 to 76.3 percent. Using the results of randomized trials, oncologists are still debating whether adjuvant chemotherapy has any overall survival benefit. Mueller of McMaster University in Hamilton, Ontario, argues that about 6 percent of treated women have their lives prolonged by about 14 months4.
A study of this quality that supported conservative surgery for breast cancer or for any other neoplasm would never have been published by the Journal. This paper is evidence of a strong bias among the authors and reviewers in favor of adjuvant chemotherapy. Chemotherapy for breast cancer will demonstrate its efficacy when it affects the age-adjusted mortality rate from breast cancer.
4 ReferencesRichard A. Evans, M.D.
1011 Augusta Dr., Houston, TX 77057-20151
Cancer patient survival: what progress has been made? Washington, D.C.: General Accounting Office, 1987:41.
2
Osteen RT ,Cady B ,Chmiel JS , et al. 1991 National survey of carcinoma of the breast by the Commission on Cancer. J Am Coll Surg 1994;178:213-219
Web of Science | Medline3
Remington PL ,Lantz PM . Using a population-based cancer reporting system to evaluate a breast cancer detection and awareness program. CA Cancer J Clin 1992;42:367-371
CrossRef | Web of Science | Medline4
Mueller CB . The case against the use of adjuvant chemotherapy in breast cancer. Am Coll Surg Bull 1993;78:25-31
Medline
Author/Editor Response
The authors reply:
To the Editor: Contrary to Dr. Evans's assertion, randomized studies have demonstrated that adjuvant chemotherapy and adjuvant therapy with tamoxifen have improved the overall rate of survival and that the benefit persists for 10 years after a diagnosis of breast cancer1. The purpose of our study was to determine whether the benefits seen in randomized studies had been realized when these therapies were introduced in a large geographic region2. We were not testing a particular regimen, so it would have been meaningless to compare treated with untreated women, because the indications for adjuvant treatment were based on the stage of disease at diagnosis, which would confound any such comparison. The data available to us did not include the stage of disease for all women with diagnosed breast cancer.
We agree with Sankaranarayanan and Black that population-based survival rates are multifactorial, especially when one considers an interval of 20 or 30 years. Diagnostic approaches and the degree of completeness in reporting cases to cancer registries have changed with time. In our province2 and the United States,3 the increased use of screening mammography since the mid-1980s has been associated with an increased incidence and earlier diagnosis of breast cancer. The raw data on five-year survival reported from Scotland by Sankaranarayanan and Black are difficult to interpret without information on other aspects of the incidence of breast cancer that may have been changing in the population they studied from 1968 through 1987. The average five-year survival rate they report is approximately 10 percent lower than that reported for a comparable period in the United States,4 a fact that emphasizes the difficulty of making direct international comparisons. Over the comparatively short period from 1974 to 1984, the incidence of breast cancer in our population was stable2. Furthermore, for women 50 years old or older in the 1974 and 1980 cohorts, before the use of adjuvant tamoxifen, the two survival curves could be superimposed on each other, suggesting that factors such as earlier diagnosis and changes in registry reporting were not likely to have been at work.
As we acknowledged in our article, although we cannot exclude a small contribution of lead-time bias to our results, the observation that the improvements in survival occurred at different times, correlating with the introduction of adjuvant systemic therapies for each age group, supports the view that treatment has had an effect on survival after breast cancer in our population.
4 ReferencesI.A. Olivotto, M.D.
C.D. Bajdik, M.Math.
C.M. Coppin, B.M., D.Phil.
British Columbia Cancer Agency, Vancouver, BC V5Z 4E6, Canada1
Early Breast Cancer Trialists' Collaborative Group
Web of Science | Medline2
Olivotto IA ,Bajdik CD ,Plenderleith IH , et al. Adjuvant systemic therapy and survival after breast cancer. N Engl J Med 1994;330:805-810
Full Text | Web of Science | Medline3
Miller BA ,Feuer EJ ,Hankey BF . The increasing incidence of breast cancer since 1982: relevance of early detection. Cancer Causes Control 1991;2:67-74
CrossRef | Web of Science | Medline4
Boring CC ,Squires TS ,Tong T ,Montgomery S . Cancer statistics, 1994. CA Cancer J Clin 1994;44:7-26
CrossRef | Web of Science | Medline
