麻豆传媒

Breast cancer patients at enhanced clinical risk by classical parameters who omitted chemotherapy on the basis of a low 21-gene recurrence score (Oncotype DX) of <11 had an approximately 98% 3-year disease-free survival rate, according to results from the West German Study Group Phase III PlanB Trial.

Reporting in the , the investigators also found a predictive dichotomy between the traditional prognostic markers identified by pathology and recurrence score (RS). "We had expected less discordance between the different factors," lead author Oleg Gluz, MD, of Evangelical Hospital Bethesda, Moenchengladbach, Germany, commented to 麻豆传媒. "Our results support the use of RS together with other factors and show the need to be cautious when making a chemotherapy decision based on one factor alone -- for example, histologic grade 3 or positive lymph node."

This Oncotype genomic assay is already widely used for guiding adjuvant treatment decisions in hormone receptor (HR)-positive breast cancer for which the benefit of chemotherapy is unclear.

"This paper is a very important proof of concept for what many clinicians have been doing in practice for the past few years without completely understanding how this would play out down the road," said , of the Dana-Farber Cancer Institute, Boston, in an interview with 麻豆传媒. "The Oncotype is a very powerful test for teasing out which patients seem to benefit from chemotherapy, but these data provide prospective validation -- a piece that was missing."

Noting that previous studies have focused on node-negative breast cancer, Burstein added, "This study confirms those and extends RS to node-positive breast cancer. The nugget this paper adds is that if you look at patients who had up to three positive nodes, the outcome is still very good without chemotherapy. It's very reassuring."

Prognostic factors identified by routine pathology are not robust enough to achieve the same outcome on their own, said Burstein, who was not part of the study.

During 2009-2011, the PlanB study recruited 3,198 post-surgical patients, ages 18 to 76 (median 56) with node-positive or high-risk node-negative breast cancer. Of these, 85.9% were locally HR-positive; 41.1% had node-positive and 32.5% had grade 3 disease. Chemotherapy was omitted for 348 patients (15.3%) who had an RS of <11. The omission of chemotherapy in these patients was based on the assumption that the 5-year disease-free survival rate should be >90%.

After 35 months' median follow-up, 3-year disease-free survival (DFS) in patients with RS <11 and who had endocrine therapy alone was a promising 97.4% (95% CI 95.6-99.1). "Even with only a 3-year follow-up, such a high rate of DFS almost excludes possible benefit from adjuvant chemotherapy," wrote Gluz and his associates.

In other RS categories, DFS was 91.9% (95% CI 89.0-94.8) for a score of >25, and 97.8% (CI 96.6-98.8) for an intermediate score of 12-25 (P<0.001). It is not known if adding chemotherapy to endocrine therapy will benefit patients with an intermediate RS of 12-25, the investigators noted.

"We also found high 3-year DFS (98%) in chemotherapy-treated patients with intermediate RS (12-25) and poorer 3-year DFS (92%) in chemotherapy-treated patients with RS >25."

These findings confirm those of the (Trial Assigning IndividuaLized Options for Treatment) study in node-negative patients, as well as retrospective analyses of earlier prospective trials.

Nodal status, central and local grade, the Ki-67 protein encoded by the MKI67 gene, estrogen receptor (ER), progesterone receptor (PR), tumor size, and RS were univariate prognostic factors for disease-free survival. Only nodal status, both central and local grade, and RS were independent multivariate factors. Histologic grade was discordant between central and local laboratories in 44% of locally HR-positive tumors.

RS had only a weak-to-moderate positive correlation with Ki-67 protein and with central and local histologic grade. It had a moderate negative correlation with immunohistochemistry-determined progesterone receptor (PRO status and a weak negative correlation with ER status. "What the data speak to is there's enough vagary in how the pathologist calls the grade and how the measurements for Ki-67 and ER and PR are done that they can't fully substitute for the test," said Burstein.

He noted that this has become an important question because in most parts of the world, genomic profiling is prohibitively expensive (about $4,000) and there has been interest in whether a pathology-based "poor man's" genomic assay would be adequate for making treatment decisions.

This paper highlights the importance of breast cancer biology, commented , of Smith Breast Center, Baylor College of Medicine, Houston.

"Over the last decade, molecular testing has helped us identify which patients can be spared the toxicity and cost of chemotherapy," said Rimawi, who was not involved in the study. "This means we have more evidence to support a more tailored approach for each patient, and that we can de-escalate treatment in some patients without compromising their outcomes."

Limitations of this study include the availability of central Ki-67 assessment alone, the non-obligatory implementation of RS results obtained before randomization, and missing RS data in a small group of patients. In addition, these early survival results need to be confirmed by longer follow-up before definitive conclusions can be drawn about interactions between prognostic markers, the authors stressed.

Comment