Dose-Dense Adjuvant Chemo Validated in Breast Cancer

SAN ANTONIO—Adjuvant dose-dense chemotherapy for breast cancer was validated by the updated results of Intergroup C9741, most notably in women with estrogen-receptor (ER)-negative disease. Clifford Hudis, MD, of Memorial Sloan-Kettering Cancer Center, presented the results, with 6.5 years median follow-up, at the 28th San Antonio Breast Cancer Symposium (abstract 41). The 3-year results were presented at the San Antonio meeting in 2002.

The study evaluated standard chemotherapy given every 2 weeks vs every 3, and concurrent administration of the agents vs sequential administration. The study's 2 2 factorial design allowed investigators to answer these two questions.

Dose-dense chemotherapy aims to decrease the time for tumor regrowth, enable subsequent treatment of smaller tumor volumes, and produce greater killing of cancer cells, Dr. Hudis explained.

The study randomized 2,005 node-positive breast cancer patients to either four cycles each of sequential doxorubicin, paclitaxel, and cyclophosphamide or four cycles of doxorubicin and cyclophosphamide given concurrently, followed by four cycles of paclitaxel. Within each group, patients were randomized to receive treatment every 2 weeks or every 3 weeks. Treatment duration for the sequential regimen was 22 weeks when given every 2 weeks and 33 weeks when given every 3 weeks. In the concurrent arm, treatment duration was 14 and 21 weeks, respectively. The updated results were based on 508 disease-free survival events (vs 315 at 3 years) and 370 deaths (vs 182) in 1,972 evaluable patients.

The analysis found similar outcomes regardless of whether the regimens were given concurrently or sequentially, Dr. Hudis reported. The dose-dense approach, however, proved more beneficial regardless of whether the drugs were given sequentially or concurrently. Patients treated every 2 weeks had significantly better disease-free and overall survival than those on the every-3-week schedule. After a median follow-up of 6.5 years, 230 relapses occurred in the 2-week arm, compared with 278 in the 3-week arm, for a hazard ratio of 1.25 favoring the dose-dense approach (P = .012). Similarly, deaths numbered 168 in the 2-week arm vs 202 in the 3-week arm (P = .049; no hazard ratio reported).

The absolute differences in event numbers for the dose-dense vs standard-density chemotherapy at 6.5 years was 48 fewer events and 34 fewer deaths for dose-dense therapy, a change in difference of only 5 events and 2 deaths, compared with the 3-year numbers (43 fewer events and 32 fewer deaths). "These stable data suggest that additional analyses are unlikely to be informative," he said, and no additional data analyses are planned.

"An exploratory analysis focused on ER-receptor status with the critical caveat that ER testing in this trial was not performed or reviewed centrally," Dr. Hudis said. He reported "a very small advantage for ER-positive patients treated every 2 weeks vs every 3 weeks, with 7 fewer events (126 vs 133), whereas in the ER-negative subset there is a significant advantage for every-2-week administration (99 vs 127 events, P = .014)." He added that this effect is mirrored in overall survival. For every-2-week vs every-3-week schedules, deaths among ER-positive patients numbered 81 and 92, respectively, vs 83 and 105, respectively, for ER-negative patients (P = .039). "The separation of the curves is early and consistent for ER-negative patients for both disease-free and overall survival. In the ER-positive patients, it is less clear . . . In terms of the cumulative hazard ratios over time, the effect in ER-positive patients is less robust but still favorable," he said.

The updated analysis found no association of acute myeloid leukemia or myelodysplastic syndrome with any particular schedule or regimen (range, 0.4% to 1%). Similarly, grade 3 post-therapy cardiac events were rare (1% to 2.5%) and not associated with schedule or regimen.

"Long-term follow-up of C9741 reveals no change in our initial conclusions. Disease-free and overall survival are the same with sequential or concurrent doxorubicin/cyclophosphamide but superior for dose-dense administration of these agents with paclitaxel," Dr. Hudis said. "There is also no change in the initial conclusions regarding toxicity. Every-2-week therapy is tolerable and faster by one-third, and there is no evidence of increased late risks."

In the discussion, Peter Ravdin, MD, PhD, of the University of Texas Health Science Center, San Antonio, noted that the benefit for dose-dense therapy "continues to be significant, but the risk reductions, both for relapse and mortality, are less impressive than in the first report, which found a 30% benefit." Dr. Hudis replied that at 6.5 years, the overall reduction in risk of relapse was 25%, compared with 26% at 3 years.