The 1998 St. Gallen's Consensus Conference: an Assessment

A decrease in breast cancer mortality in the United States was reported for the first time in 1994 ( 1 ). This decrease has been attributed, in part, to better adjuvant therapies that have become standard after carefully conducted, randomized, controlled trials have established a clear benefit. In reality, however, physicians and patients must also make treatment decisions in situations where optimal trial data are lacking. Appropriately, leaders in the field have met at regular intervals to provide a consensus as to how new information can be incorporated into practice. The results of the 6th International Consensus Panel on the Treatment of Primary Breast Cancer are presented in this issue of the Journal ( 2 ). There are several noteworthy differences between the recommendations from this conference as compared with those from the 1995 consensus panel that reflect both our expanding knowledge base and our changing societal attitudes toward cancer and cancer treatment. Some of these changes stem from the growing confidence that adjuvant chemotherapy can improve survival in most cases of breast cancer, even though these differences may be small in absolute terms. To begin with, the 1998 consensus report redefined the threshold for systemic therapy: Women with less than a 10% chance of relapse at 10 years are in the “low-risk” group and therefore would not routinely be considered candidates for systemic therapy. This change recognizes avoidance of relapse (as compared with survival) as a treatment goal and continues the trend of considering systemic therapy for women in increasingly better prognostic groups. Initially, adjuvant therapies were offered to lymph node-positive women who had a 30%–l50% overall survival at 10 years ( 3 ). In 1988, the National Cancer Institute clinical alert extended the recommendation for adjuvant therapy to women with lymph node-negative breast cancer ( 4 ), a group with a 70% overall survival at 10 years. In 1995, the international consensus panel recommended that patients with a greater than 10% mortality at 10 years would be candidates for routine adjuvant systemic therapy ( 5 ). The current consensus panel lowered the threshold even further by suggesting systemic adjuvant therapy for patients with a greater than 10% risk of relapse ( 2 ). This shift toward a more inclusive treatment strategy may be resisted by some. However, it may be helpful to consider the scientific and social basis for these recommendations. Freedom from disease is becoming a recognized treatment goal. This was evident during the deliberations by the Oncology Drug Advisory Committee (ODAC) to the U.S. Food and Drug Administration as its members considered tamoxifen as a potential chemopreventive. In a large U.S. study of 13 388 women, tamoxifen decreased the risk of invasive breast cancer by 49% at a median follow-up of 55 months ( 6 ). At this time point, however, there was no difference in mortality between the women on tamoxifen and those on placebo. Despite the absence of a benefit in mortality reduction, the ODAC voted to approve tamoxifen for reduction in breast cancer incidence. Although there is a tendency to consider the relapse rate in the “good” and “intermediate” lymph node-negative prognostic groups to be small, the long-term survival impact may, in fact, be medically important. Examination of long-term follow-up data reveals that even the breast cancers in the best prognostic groups are associated with sizable disease-specific mortality. The National Surgical Adjuvant Breast and Bowel Project (NSABP) has examined the disease-free survival of lymph nodenegative patients who did not receive adjuvant systemic therapy according to tumor size ( 7 ). In their estrogen receptor (ER)-negative trial (B-13), the 5-year disease-free survival for untreated patients with tumors less than or equal to 1 cm in size was only 77%. In an ER-positive trial (B-14), the 5-year disease-free survival for patients with tumors of 1 cm or less in size was 82%. (It should be noted that the NSABP trials did not include patients with tumors of 1 cm or less in size that were detected only by mammography.) Rosen et al. ( 8 ) have also examined the long-term survival of women with lymph node-negative, T1 breast cancer who were treated by mastectomy from 1964 through 1970. Of 474 patients, 171 had a tumor size of less than or equal to 1 cm, and 303 patients had a tumor size of 1.1–2.0 cm. In women with a tumor size of 1 cm or less at 10 years of follow-up, 7% of the patients had died of breast cancer and 2% were alive with disease. By the end of 20 years, 10% of the patients had died of breast cancer and 2% were alive with disease. In women with tumors 1.1–2.0 cm at 10 years of follow-up, 18% of the patients had died of breast cancer and 4% were alive with disease. By the end of 20 years, 24% of the patients had died of breast cancer and 2% were alive with disease. Of interest, patients with tumor size of 1 cm or less tended to manifest more late recurrences than those with a tumor size of 1.1–2 cm. Twenty-five percent of the patients with tumor size 1 cm or less who developed a recurrence did so after the first 10 years of follow-up (years 10–21) compared with 16% of the patients with tumor size of 1.1–2.0 cm. Thus, in good- and intermediate-risk lymph node-negative breast cancers, the 20-year survival statistics are of concern. These long-term (20-year) survival statistics question the designation of good risk or intermediate risk in these lymph node-negative patients. Considered with the availability of effective therapies ( see below ), it is understandable that some physicians may offer such treatments to this subgroup of patients. A second change from the 1995 conference is the recommendation concerning hormonal therapy. The addition of tamoxifen to chemotherapy is now considered standard for premenopausal women with ER-positive, invasive tumors. Women with ER-negative tumors (even if >50 years of age), however, are unlikely to benefit from hormonal therapy, and, in this consensus statement, hormonal therapy is not recommended for postmenopausal women with hormone receptor-negative breast cancer. The 1995 Early Breast Cancer Trialists' Working Group metaanalysis ( 9 ) provides the basis for this recommendation. In this study, 18 000 women with ER-positive tumors and 12 000 women with untested tumors (of whom an estimated 8000 would have been ER positive) treated with tamoxifen therapy demonstrated a 10.9% absolute survival benefit for women with lymph node-positive tumors and 5.6% absolute survival benefit for women with lymph node-negative breast cancer. The proportional reductions were similar for lymph node-positive and lymph node-negative breast cancer (47% reduction in recurrence and a 26% survival advantage in women treated with 5 years of tamoxifen at 10 years of follow-up). Benefits were observed irrespective of age, menopausal status, or chemotherapy. A third change since the previous conference is the recommendation that the majority of patients be considered candidates for chemotherapy. This, perhaps, is potentially the most controversial of the recommendations. In 1995, chemotherapy was still under investigation for the largest subset of patients, postmenopausal, ER-positive women. In the 1998 consensus conference, chemotherapy is now an accepted practice option (Table 1). This change was due, in large part, to two important publications. In the 1995 Lancet overview of 18 000 women ( 10 ), polychemotherapy resulted in a statistically significant proportional reduction in recurrence (35% in women <50 years old and 20% in women ages 50–69 years). The corresponding reductions in overall mortality were 27% for women less than 50 years old and 11% for women 50–69 years of age. Polychemotherapy produced an absolute 10-year survival benefit of 7% for women with lymph node-negative disease and an 11% overall survival advantage for women with lymph node-positive breast cancer. The corresponding number for women ages 50–69 years are an absolute survival advantage of 2% in patients with lymph nodenegative disease and 3% in women with lymph node-positive disease. Although women under 50 years of age derived more benefit from polychemotherapy than women ages 50–69 years, both groups of women experienced statistically significant benefit. Furthermore, the effects of chemotherapy were in addition to those derived from tamoxifen therapy. These overview results are also consistent with the results of the NSABP-B20 trial ( 11 ) in which women with lymph node-negative, ER-positive breast cancer who were treated with chemotherapy plus tamoxifen had a 5-year disease-free and overall survival advantage compared with women treated with tamoxifen alone. Thus, the 1998 St. Gallen's consensus recommendations suggest that for the majority of women, chemotherapy would be considered an accepted treatment option. For patients with hormone receptorpositive disease, a combination of chemotherapy and hormonal therapy would be considered appropriate (Table 2). Despite the absolute improvement in survival afforded by chemotherapy, there may be concerns that the 2%–3% absolute increment in survival for the ER-positive, lymph node-negative population is too small to be worth the discomfort of adjuvant treatment. Understanding this concern, the St. Gallen conference participants acknowledged the need to individualize therapy based on physician judgment and patient preferences. Although physicians may be inherently conservative in the application of chemotherapy, many patients show a remarkable willingness to take potentially toxic treatments for small decrements in recurrence rate. According to a recent survey of 318 women who have received chemotherapy, the median improvement in the risk of breast cancer recurrence where adjuvant therapy would be acceptable was 0.5%–1% ( 12 ). Thus, as recommended by the consensus panel, patient choice and clinical judgement will need to be considered in deciding chemotherapy for patients with lymph node-negative, ER-positive tumors. The effect of the 1998 St. Gallen consensus recommendations is to extend the acceptability of systemic chemotherapy to a wider range of breast cancer patients—to those with better prognosis, and specifically to those with postmenopausal, ERpositive disease. As the criteria for treatment become progressively more inclusive, the larger questions become the optimal selection of systemic therapies for individual subsets and the identification of a subset of patients where therapy could safely be withheld. For both questions, molecular markers for disease biology may play an increasingly important role in finding the solutions. Already, the conference participants have considered known clinical prognostic factors, such as hormonal status, pathologic grade, and young age of patient (<35 years old), in their stratification of lymph node-negative patients into good-, intermediate-, or high-risk groups. As noted in the St. Gallen consensus, the ER status is a major deciding factor in therapeutic selection. However, other biomarkers, such as proliferative fraction or HER-2/neu status, may have importance as well. ( 10 ) highlighted the modest difference favoring anthracyclinebased chemotherapy compared with cyclophosphamide, methotrexate, and 5-fluorouracil (CMF)-like therapies. However, it is likely that only some patients benefit from the addition of anthracyclines to the adjuvant regimen. Emerging data suggest that the subset of lymph node-positive patients with tumors that overexpress HER-2 is likely to benefit from dose-intense anthracycline- based regimens. Retrospective analyses of two large randomized trials in women with lymph node-positive breast cancer ( 13 , 14 ) have suggested that patients with HER-2—positive tumors have an improved relapse-free and overall survival following dose-intensive anthracycline-based regimens compared with lower dose anthracyclines or non-anthracycline-containing regimens. These results are restricted to lymph node-positive patients and cannot be extrapolated to the lymph node-negative subsets. However, this possibility is intriguing and should be formally examined in the lymph node-negative setting. Breast cancer is a disease with a long natural history, and the full impact of today's research will not be fully realized for many years. Thus, the patterns of practice should be expected to change. The 1998 St. Gallen consensus conference produced a very reasonable set of recommendations based on clinical evidence and on the best judgment of experts in the field. Nevertheless, the scope of the changes, even when compared with 3 years ago, represents a substantial shift in philosophy. Given the speed of research today, we should expect no less in the next consensus meeting.