Anthracyclines and Trastuzumab Bests Taxane/Carboplatin/Trastuzumab

San Antonio, TX—The long-awaited final results of the BCIRG 006 trial of adjuvant chemotherapy show that a regimen of concurrent anthracycline chemotherapy plus trastuzumab (Herceptin) outperformed anthracycline-containing chemotherapy alone and performed similarly to a non-anthracycline-containing regimen in patients with early HER2-amplified breast cancer. However, the non-anthracycline regimen had a much better adverse-effects profile, especially with respect to longer-term life-threatening adverse effects.

“Our interpretation of the study results is that trastuzumab provides similar disease-free survival and overall survival with an anthracycline and a non-anthracycline regimen in low- and high-risk patients.Acute and chronic toxicity favors TCH, the non-anthracycline regimen. The advantage of slightly fewer events with the anthracycline regimen plus trastuzumab comes with a cost of cardiotoxicity and leukemia,” said Dennis Slamon, MD, UCLA, CA, a pioneer in the development of trastuzumab for the treatment of HER2-positive breast cancer.

Although Dr Slamon said that, based on results of the trial, the benefit of anthracyclines appears to be restricted to higher-risk patients (those with topoisomerase II amplification), other experts said they were not ready to abandon anthracyclines for patients with breast cancer.Anthracyclines are used worldwide because they improve disease-free survival (DFS) by 3.5% and overall survival (OS) by 4% according to the Oxford overview, the “bible” for oncologists; furthermore, they are much cheaper than newer drugs, which is important for counties with nationalized health coverage.

BCIRG 006 randomized 3222 patients with documented HER2-positive early breast cancer to one of three treatment arms:4 cycles ofanthracycline (doxorubicin) plus cyclophosphamide followed by docetaxel (AC › T); 4 cycles of AC › T plus trastuzumab (AC > TH); or 6 cycles of docetaxel and carboplatin plus trastuzumab (TCH).Patients were enrolled from April 2001 to March 2004. Median age was around 53 years; about 60% had a mastectomy, about 68% had radiotherapy, and about 51% had hormone therapy. Twenty-nine percent of all patients were node-negative. About 38% had 1 to 3 nodes involved, about 23% had 4 to 10 nodes involved, and about 10% had more than 10 nodes involved.

Forty percent had tumors ≤2 cm; 53% had tumors between 2 and 5 cm in size; and 6% had tumors >5 cm. Patients were stratified for hormone receptor—positive and negative status and for tumor size. After April 2005, 23 patients (2.3% of those randomized to the control arm, crossed over to receive trastuzumab, leaving 97.5% of the control arm intact for comparisons of DFS, OS, and safety.

At the San Antonio Breast Cancer Symposium, Dr Slamon presented results of the third planned analysis of the trial. Median follow-up was 65 months; 656 DFS events (breast cancer relapse, second primary cancer, or death) had occurred; 348 deaths occurred. He said that in 2005, at the first planned analysis, DFS was 84% for AC-TH (P <.0001 compared to control arm), 80% for TCH (P = .0002), and 73% for the control arm. The updated DFS was 84% for AC › TH (P <.001), 81% for TCH (P =.04), and 75% for the control arm. The total number of DFS events in each arm was 257, 185, and 214, respectively. The difference between the two study arms was not statistically different, Dr. Slamon noted, although it was slightly numerically superior for AC › TH.

The third planned analysis of OS showed that 92%, 91%, and 87% of patients, respectively, were alive at 5 years. The number of deaths due to any cause was 141 in the control arm, 94 in the AC › TH arm, and 113 in the TCH arm. Again, the difference in OS between the two experimental arms was not statistically significant.

For node-negative disease, 5-year DFS in the three arms was 93%, 90%, and 85%, respectively. The difference was statistically significant favoring anthracycline-containing arm over the control arm (P =.02), with a trend favoring TCH. OS in node-negative patients was 97%, 96%, and 93%, respectively.

Looking at higher risk, node-positive disease, 5-year DFS was 80%, 78%, and 71%, respectively. Both experimental arms were significantly superior to control (P = .0003 for the anthracycline arm and P = .013 for TCH). In high-risk patients with ≥4 positive nodes, 5-year DFS was 73%, 72%, and 61%, respectively, which was highly significant for both arms versus control (P = .002 for both arms). OS in those with ≥4 positive nodes was 84%, 83%, and 76%, respectively. Only the anthracycline-containing arm was significantly superior to control (P = .02).

TCH won out when toxicity was analyzed. Both non-hematological and hematological toxicities occurred significantly less frequently with TCH, including Grade 3 and 4 arthralgia, myalgia, hand-foot syndrome, stomatitis, vomiting,neutropenia and leucopenia.. Acute leukemias were reported in 6, 1, and 1 patients, respectively. “All patients who developed leukemia has prior anthracycline exposure” Dr Slamon told listeners.

As has been shown previously, cardiac toxicity was more frequent in patients receiving an anthracycline, with a four- to five-fold increase in clinical congestive heart failure in the two arms that contained doxorubicin. “No surprise here,” Dr Slamon said. Although mean left ventricular ejection fraction (LVEF) declined at the beginning of treatment, patients treated with TCH recovered to normal at 5 years, while LVEF in patients randomized to the other two arms has not yet returned to normal. SABCS Abstract 62.