Mature Data Confirm Advantage of CAFT Versus Tamoxifen in Older, Node-Positive, Receptor-Positive Women

SAN FRANCISCO—Mature survival analyses from the North American Breast Intergroup Trial (INT-0100; SWOG-8814) confirm that chemotherapy plus tamoxifen (Nolvadex) produces better overall and disease-free survival than tamoxifen alone in postmenopausal women with node-positive, receptor-positive breast cancer.

Kathy S. Albain, MD, professor of medicine, Cardinal Bernardin Cancer Center of Loyola University Medical Center, Maywood, Illinois, presented the updated results at the 37th Annual Meeting of the American Society of Clinical Oncology (ASCO abstract 94).

"The overall survival curves do not diverge in favor of cyclophosphamide, doxorubicin [Adriamycin], 5-fluorouracil, and tamoxifen (CAFT) until after year 4, but the advantage persists after that," Dr. Albain said. "The CAFT advantage also persists after adjustment for all of the significant prognostic factors."

The objectives of INT-0100 were to determine (1) if CAFT was superior to the 1988 standard of tamoxifen alone in postmenopausal women with node-positive, receptor-positive disease; and (2) if CAF followed by tamoxifen was better than concurrent CAFT. Dr. Albain reported final data on the CAFT vs tamoxifen comparison. She said that data on the concurrent vs sequential comparison are not yet mature.

The trial enrolled 1,477 eligible postmenopausal women with estrogen-receptor (ER)-positive, node-positive breast cancer. Patients were randomized to tamoxifen alone (n = 361), to CAF followed by tamoxifen (n = 566), or to concurrent CAF plus tamoxifen (n = 550).

Tamoxifen was given for 5 years, and CAF was given at standard doses for six cycles using oral cyclophosphamide for 14 days and doxorubicin and 5-FU on days 1 and 8. Radiotherapy was given either at the start or after the completion of CAF, or at the start of tamoxifen alone.

The main CAFT vs tamoxifen comparison met early reporting criteria for disease-free survival in 1996. A disease-free survival advantage for CAFT at 4 years was reported at the 1997 ASCO annual meeting. Median follow-up is now 7.3 years.

Dr. Albain reported a 9% absolute improvement in 5-year disease-free survival with CAFT vs tamoxifen alone (76% vs 67%, P = .002). The benefit in 5-year overall survival was 5% greater with CAFT (84% vs 79%, P = .006). In the Cox multivariate model for survival, patients were at greater risk for a poor outcome if they had four or more positive nodes, had tumors larger than 2 cm, had PR-negative tumors, or were black.

Short-term toxicities and treatment-related mortality were reported at ASCO in 1997. With regard to long-term toxicities, Dr. Albain reported a 2.1% incidence of congestive heart failure in patients on CAFT alive and disease-free at 1 year vs 0.3% for tamoxifen alone.

Endometrial cancers were reported in 4 (1.1%) patients on tamoxifen and in 12 (1.1%) on CAFT. Overall, there were six cases of myelodysplastic syndrome (MDS) and three of acute myelogenous leukemia (AML), all on the CAFT arms.

Dr. Albain said that there was no apparent correlation between age and CAFT benefit. "One third of the women in this trial were over age 65," she pointed out. The interaction term for age and treatment was not significant.

Also, during the discussion period, Dr. Albain said that new studies were underway with FISH analyses for HER-2 and histologic grade to determine if they can predict optimal candidates for CAFT.

CAFT vs the Future

In the discussion, Matthew Ellis, MD, PhD, associate professor of medicine, Duke University Medical Center, said that the results of this study in postmenopausal women are particularly important because 50% of cases of breast cancer are in women age 65 or older, "a population expected to increase substantially over the next 20 years."

Dr. Ellis also said that the 5% absolute survival benefit of CAFT demonstrated in Dr. Albain’s study should be weighed against the risks, which include a 2.1% risk of congestive heart failure, a 0.8% risk of AML or MDS, and a 0.27% risk of chemotherapy-related death.

"There are likely to be subpopulations of vulnerable patients in whom the risks outweigh the benefits, so clinical discretion should be exercised when using the CAF regimen," Dr. Ellis said.

The regimens used in this study reflect best practice when the protocol was designed in 1988. Dr. Ellis said that since then, taxanes, trastuzumab (Herceptin), and aromatase inhibitors have all entered adjuvant clinical trials that include postmenopausal women with ER-positive disease. "We will be examining the results of these studies carefully to determine if CAFT will remain the standard of care in the future," he said.

From the breast cancer treatment standpoint, he said, breast cancer can currently be considered to be four distinct diseases: ER positive and HER-2 negative, ER positive and HER-2 positive, ER negative and HER-2 positive, and ER negative and HER-2 negative.

ER-positive/HER-2-positive disease is an interesting tumor subtype that represents about 20% of the ER-positive population (based on immunohistochemistry), he said.

"These tumors may be relatively resistant to adjuvant tamoxifen, but sensitive to adjuvant doxorubicin and perhaps adjuvant trastuzumab," Dr. Ellis commented. Furthermore, he said, "there is evidence from a neoadjuvant endocrine therapy study that an aromatase inhibitor may be much more effective than tamoxifen against ER-positive/HER-2-positive tumors." (See J Clin Oncol 19:3808-3816.)

Dr. Ellis said that the Cancer and Leukemia Group B (CALGB) is initiating an adjuvant study to determine the best endocrine therapy for ER-positive/HER-2-positive tumors.

"Overall," he said, " these findings emphasize that we are beginning to recognize that, like lymphoma and leukemia, breast cancer is actually a number of different tumor subtypes that will be further defined by genetic analysis in the next 10 years or so. The challenge will be to find the right therapeutic approach for each breast cancer subtype."