CAF Produces Small But Significant Survival Advantage Over CMF in High-Risk Node-Negative Breast Cancer

SAN ANTONIO--The first results of a large Intergroup study show that adjuvant chemotherapy with CAF is superior to CMF in high-risk node-negative patients, Laura Hutchins, MD, of the University of Arkansas, reported at the ASCO plenary session for SWOG. The study also suggests that S phase fraction can be used to stratify patients into risk groups.

Patients with T1-3a node-negative invasive breast cancer were assigned to risk group, based on prognostic groups identified in a prior Intergroup trial, according to tumor size and hormone-receptor status. S phase fraction was used to categorize patients in the "uncertain" risk group (those with hormone-receptor-positive tumors under 2 cm in size).

Of 3,977 eligible patients, 1,208 were eventually assigned to the low-risk category and were followed with no additional adjuvant therapy, while 2,163 high-risk patients were randomized to receive CMF (cyclophosphamide, methotrexate, fluorouracil) or CAF (cyclophosphamide, Adriamycin, fluorouracil) for 6 cycles, followed by no further therapy or tamoxifen (Nolvadex) for 5 years.

Results for all patients show a "modest though significant" benefit in estimated 5-year disease-free survival for CAF (86%) vs CMF (84%) (P = .03). "This advantage for CAF is seen in overall survival as well (92 vs 91%)," Dr. Hutchins said. CAF also had a survival advantage for subgroup analyses by menopausal and estrogen-receptor (ER) status (see Table).

Toxicity was greater in the CAF arms, with 11 patients having grade 3 or higher cardiotoxicity vs 3 for CMF. Grade 4 granulocytopenia, grade 2 or higher nausea/vomiting, stomatitis, and alopecia were also significantly greater with CAF.

Effects of Tamoxifen

In ER-positive patients, adding tamoxifen to chemotherapy significantly increased 5-year survival, both disease-free (88% vs 82%) and overall (94% vs 91%). However, not only did tamoxifen not improve survival in ER-negative patients, there was a trend toward decreased disease-free survival in this group with tamoxifen (83% vs 86% for chemotherapy alone).

Similar results were seen when tamoxifen use was analyzed according to menopausal status. "Although these subset analyses should not be interpreted as conclusive in showing a negative effect for tamoxifen in ER-negative patients, there clearly does not seem to be a benefit in this subset," she said.

Use of S Phase Fraction

Looking at results by treatment assignment according to S phase fraction, Dr. Hutchins said, "disease-free survival was very similar for those patients in the initial low-risk group identified by small tumor size and those patients assigned to low risk on the basis of S phase fraction, despite these patients having a larger tumor size."

The 5-year disease-free survival rates were 89% and 88%, respectively. Thus, she said, "S phase fraction identified a group of node-negative patients with 1 to 2 cm tumors that did well without adjunctive therapy."

In her discussion, Dr. Nancy Davidson, of Johns Hopkins, emphasized that the differences in survival between CAF and CMF were small but came without major differences in toxicity. "The only caveat is that we don’t know if doxorubicin use would have an unpredicted long-term cardiotoxicity in these women with reasonably good prognosis," she said.

She believes either CMF or CAF is a reasonable choice in the setting of node-negative breast cancer. "Presumably, patients and physicians who are intent on maximizing anticancer effects will select CAF, whereas those who are most concerned about toxicity will choose CMF."

Dr. Davidson called the findings on S phase fraction "exciting," but warned that its use may not be practical outside of clinical trials due to the problem of lack of laboratory standardization.