麻豆传媒

SAN ANTONIO -- Women with larger, genomically high-risk breast cancers had significantly lower risk of distant metastasis with chemotherapy that included an anthracycline, a new analysis of a randomized trial showed.

Patients with a 21-gene assay score ≥31 had a 68% reduction in the risk of distant metastasis at 5 years when their chemotherapy included an anthracycline (such as doxorubicin). Distant relapse-free survival (DRFS) and overall survival (OS) also favored anthracycline-containing chemotherapy but did reach statistical significance. The magnitude of benefit increased with risk scores out to 50.

A subgroup analysis showed that the benefit in high-risk patients was limited to those with tumors >2 cm, reported Nan Chen, MD, of the University of Chicago, at the San Antonio Breast Cancer Symposium (SABCS).

"Anthracyclines should be considered in patients with high genomic-risk, hormone receptor-positive [HR+], node-negative disease," she concluded.

The study added to an ongoing debate about the role of anthracyclines in early breast cancer, particularly given their recognized toxicities that include second malignancy and adverse heart effects. The authors of a recent Journal of Clinical Oncology editorial suggested "the era of anthracycline-based chemotherapy is waning" in the absence of well-identified subgroups of patients who benefit from the drugs. Chen's report seemingly provided some direction for decision making.

Chen reported findings from an analysis of the landmark TAILORx trial, which showed that the 21-gene OncoType DX genomic assay could stratify patients with early breast cancer into risk groups for recurrence and help identify those who might avoid chemotherapy. The primary analysis showed a benefit of chemotherapy for patients with HR+/HER2-, node-negative breast cancer associated with a risk score ≥26. Since then, other studies have evaluated risk-score cutoffs of 26 to 31, she said.

Ongoing Debate

During a discussion that followed Chen's presentation, Matthew Goetz, MD, of the Mayo Clinic in Rochester, Minnesota, noted that a showed that a taxane-cyclophosphamide (TC) combination overall was non-inferior to TC plus an anthracycline. If the addition of an anthracycline benefits high-risk patients and TC is given to low-risk patients, when would TC be preferred in the event of uncertainty?

"We also performed an analysis of patients with a recurrence score of 26-30 and demonstrated that those patients do not benefit from the addition of an anthracycline," said Chen. "So certainly, I think that's a population where TC would be very reasonable. We also noted that the benefit was primarily in tumors that were greater than 2 cm in size, so small tumors may still be very reasonable to treat with TC."

Harold Burstein, MD, PhD, of Dana-Farber Cancer Institute in Boston, asked whether anthracycline-containing chemotherapy offered a benefit to patients with T1 tumors. "I just want to make sure we're not suddenly dumping a lot of anthracycline on 1-cm cancers that have an OncoType DX score of 31, if we don't need to."

"I agree with you that I would caution providers in interpreting this data," said Chen. "I agree that not all stage 1 tumors would benefit from an anthracycline and that the benefit was primarily in tumors that were 2 cm or greater in size."

During an SABCS press briefing, Chen was asked about the study's relevance within the context of the recent evolution of breast cancer therapy since the TAILORx trial, particularly the emergence of CDK4/6 inhibitors for HR+ breast cancer.

"I don't think there is any definitive, prospective data to identify a subset of patients who may be spared anthracyclines, or chemotherapy overall, with CDK4/6 inhibitors, but that remains to be seen and is certainly something we're interested in," said Chen. "The goal of our analysis was to select a group of patients who we think may be at higher risk and may still benefit from anthracyclines, and I think we demonstrate that the recurrence score may be a good biomarker for that."

Key Findings

TAILORx involved more than 10,000 patients with early node-negative, HR+/HER2- breast cancer. The new analysis included 2,549 patients who had a risk score ≥11 and had received TC alone (n=2,111) or with an anthracycline (TAC, n=438). The primary survival outcome was 5-year distant relapse-free interval (DRFI). Investigators chose 31 as the cutoff for high risk, a subset that represented about 5% of patients.

The study population had a median age of about 55, about 60% of the patients were postmenopausal, median tumor size was 18-19 mm, fewer than 20% of the tumors were low grade, and more than 80% of the tumors were progesterone-receptor positive. Patients treated with TAC had a higher mean recurrence score (29.6 vs 22.3), including more than twice as many with a risk score of 31-100 (39.5% vs 14.5%).

Among patients with a risk score <31, the addition of an anthracycline had no effect on 5-year DRFI, which was 97.0% with TAC and 97.6% with TC. In the subgroup with a risk score ≥31, the 5-year DRFI rate was 96.1% with TAC and 91.0% with TC, a statistically significant difference (P=0.009). The addition of an anthracycline also was associated with significant improvement in DRFS at 5 years in the high-risk subgroup (95.4% vs 89.8%, HR 0.47, P=0.031), and a trend toward improved 5-year OS (97.3% vs 93.5%, HR 0546, P=0.167).

The advantages of adding an anthracycline remained consistent across a subgroup analysis with the exception of patients with tumors ≤2 cm, who had a worse DRFS with an anthracycline (HR 1.32).

The 5-year DRFI advantage with anthracycline therapy was seen in premenopausal and postmenopausal patients (96.9% vs 84.4%, P=0.032; 95.6% vs 93.4%, P=0.028, respectively).

As the risk score increased beyond 31, so too did the magnitude of the reduction in the hazard ratio with the addition of an anthracycline:

• 35: HR 0.69
• 40: HR 0.60
• 45: HR 0.52
• 50: HR 0.45

On the other hand, the benefits of TC or alternative chemotherapy regimens decreased as the risk score increased.

Chen acknowledged limitations of the analysis: post-hoc and not designed to evaluate the endpoint, nonrandomized chemotherapy choice, and late toxicities associated with anthracyclines.

"Despite the bias of higher-risk patients receiving anthracyclines, this analysis still noted a benefit in high genomic-risk patients," said Chen.

Comment