Intensive follow-up was ineffective in breast cancer
ACP J Club. 1994 Nov-Dec;121:76. doi:10.7326/ACPJC-1994-121-3-076
The GIVIO Investigators. Impact of follow-up testing on survival and health-related quality of life in breast cancer patients.A multicenter randomized controlled trial. JAMA. 1994 May 25;271:1587-92.
To study whether intensive follow-up with scheduled diagnostic procedures for women with breast cancer improves survival and quality of life.
Randomized controlled trial with a median follow-up of 71 months.
26 general hospitals in Italy.
1320 consecutive women with histologically confirmed operable (T1 to T3, N0 to N1, or M0) noninflammatory, unilateral breast carcinoma who were < 70 years old. Exclusion criteria were in situ carcinoma and previous malignant disease other than squamous or basal cell carcinoma of the skin.
Randomization, done within 6 weeks of surgery, was stratified based on center and axillary node status. All women attended clinics every 3 months for 2 years and then every 6 months for 3 years. Treatment guidelines were provided for adjuvant treatment and metastatic disease. 655 women were allocated to usual care with diagnostic tests done based on clinical suspicion only. 665 women were allocated to usual care plus regularly scheduled 6-month chest roentgenography and annual bone scan, liver echography, and laboratory tests to detect distant metastases. Follow-up compliance was 92%.
Main outcome measures
Survival and quality of life were measured using a study-specific questionnaire based on the Functional Living Index-Cancer Scale, Sickness Impact Profile, Profile of Mood States, and Cancer Inventory of Problem Situation.
An intention-to-treat analysis was used. The groups did not differ for survival (132 deaths [20%] in women in the intensive care group vs 122 deaths [18%] in women in the usual care group) (Table) or recurrence-free survival (53.4 vs 54.1 mo). They also did not differ for quality of life (overall health and quality-of-life perception, emotional well-being, body image, social functioning, symptoms, and satisfaction with care) during all follow-up periods (6, 12, 24, and 60 mo).
For women who had surgery for breast cancer, intensive follow-up with scheduled diagnostic procedures did not improve survival or quality of life compared with women who received follow-up that included diagnostic tests based on clinical suspicion.
Sources of funding: Italian National Research Council, National Cancer Institute, and Zeneca Pharmaceuticals of Milan, Italy.
For article reprint: Dr. A. Liberati, GIVIO Coordinating Center, Laboratory of Clinical Epidemiology, Instituto di Ricerche Farmacologiche Mario Negri, Via Eritrea 62, 20157 Milan, Italy. FAX 39-2-332-00231.
Table. Intensive follow-up vs usual care for breast cancer*
|Outcome at a median follow-up of 71 mo||Intensive follow-up||Usual care||RRI (95% CI)||NNH|
|Death||20%||18%||10% (-12 to 37)||Not significant|
*Abbreviations defined in Glossary; RRI, NNH, and CI calculated from data in article.
These 2 studies, which examined the benefits of routine diagnostic testing in asymptomatic women who are being followed for nonmetastatic breast cancer, are important. Both were randomized trials in women with nonmetastatic breast cancer: 1 group had routine diagnostic testing (annual chest roentgenography, bone scan, liver ultrasound, alkaline phosphatase and gamma-glutamyl transpeptidase measurements in 1 study and chest roentgenography and bone scan every 6 months in the other study), whereas the comparison group was followed by physician assessment and mammography with diagnostic tests done only when there was clinical suspicion of metastases.
The results of these well-designed clinical trials are remarkably similar. First, and most important, neither study showed a survival benefit in women who had routine diagnostic testing. This indicates that even though routine testing may detect asymptomatic metastases early, this does not translate into a survival advantage if patients are followed clinically and widely accepted treatment protocols are used when metastatic disease is identified. That these observations were made in randomized clinical trials overcomes the effects of lead-time or length-time biases that may have been present in previous nonrandomized studies, thereby strengthening the clinical relevance and validity of the results. These results should allay concerns that physicians are jeopardizing their patients' survival if they do not do routine diagnostic testing during follow-up of primary breast cancer.
The GIVIO investigators extended these observations to show that routine diagnostic testing did not improve health-related quality of life, at least when quality of life was measured annually. It is possible that the investigators missed short-term differences in quality of life caused by the anxiety associated with routine testing (e.g., waiting for test results, reacting to positive test results) or an unsatisfied desire for testing as a means of reassurance that a recurrence has not occurred. These possibilities should be explored in future investigations. This is particularly important because the GIVIO investigators identified a clear wish on the part of women in both of their study groups to have routine diagnostic tests done even if they were asymptomatic.
Although 1 of the studies showed that recurrences were diagnosed earlier in women who had routine diagnostic testing, neither study showed a survival benefit, the most clinically relevant outcome in this group of patients. Early diagnosis of recurrence that cannot be translated into an increase in survival simply lengthens the time during which a woman must live with the knowledge that she has an incurable disease, a situation that is not desirable. Given these results, routine diagnostic testing in patients with breast cancer should not be done during follow-up of women who are asymptomatic, although a role for these tests in certain research situations may exist. All women in these studies had routine physician assessment (history and physical examination) and annual mammography. Because these procedures may detect locoregional recurrences before systemic spread, it is quite possible they may help improve survival. History and physical examination should remain the cornerstone of the follow-up for women with nonrecurrent breast cancer until further information becomes available.
Pamela J. Goodwin, MD, MSc
Mount Sinai HospitalToronto, Ontario, Canada