CHICAGO—More tantalizing evidence has been found linking African-American race and also high body mass index (BMI) to worse outcomes in women with breast cancer, but still nothing definitively practice changing was gleaned from two retrospective studies presented here at the ASCO Annual Meeting.
Researchers from the Breast Cancer Intergroup of North America took an exploratory look at a genomic profiling subset of postmenopausal African-American women with early-stage estrogen receptor (ER)-positive breast cancer in clinical trials, and suggested that worse survival was due in large part to tumors with higher proliferation rates of five genes leading to more aggressive disease, and not to differences in hormone levels or degree of endocrine responsiveness.
From Austria, a study showed that BMI has a significant impact on the efficacy of anastrozole but not tamoxifen, in adjuvant endocrine treatment of premenopausal breast cancer patients. The researchers speculated this was probably due to the influence of aromatase availability in fat tissue.
BMI & Hormone Manipulation
Georg Pfeiler, MD, a staff assistant physician in the Department of Obstetrics and Gynecology at Medical University of Vienna, reported that overweight premenopausal patients with breast cancer and endocrine-responsive disease do worse when treated with anastrozole compared with tamoxifen.
He described the hypothesis for this retrospective study, that overweight patients have higher levels of aromatase enzyme availability so BMI might have an impact on the efficacy of aromatase inhibitors.
The subset data were from on the Austrian Breast and Colorectal Cancer Group (ABCSG) 12 trial, which studied the effect of adding zoledronic acid to a combination of either goserelin and tamoxifen or goserelin and anastrozole in premenopausal women with endocrine-responsive early breast cancer. That trial—reported in May 2009—showed that the addition of zoledronic acid increased disease-free survival (NEJM 2009;360:2367-2370).
In this analysis, overweight patients treated with anastrozole were found to have had a significantly shorter disease-free survival (hazard ratio 1.60) and overall survival (hazard ratio 2.284) at five years compared with normal-weight patients receiving anastrozole, and compared with overweight women receiving tamoxifen.
And no differences were seen in disease-free or overall survival between overweight and normal-weight patients treated with tamoxifen, Dr. Pfeiler said.
The session Discussant, Pamela Jean Goodwin, MD, Chair in Breast Research and Professor of Medicine at the University of Toronto, said that the biologic plausibility of a BMI-survival connection is strong and that the topic is worthy of further study.
Dr. Goodwin said she had reservations, though, about the strength of the study's conclusions, since it was a retrospective, post-hoc analysis, and because of an imbalance in age: overweight subjects taking anastrozole were more likely to be under age 40.
In addition, the BMI-related effect was much less for disease-free survival than for overall survival, “which raises the question of whether BMI contributed to mortality in ways independent of breast cancer,” she said, also noting there were only two non-breast cancer related deaths in this study.
Dr. Pfeiler acknowledged the limitations.
“It's the right thing to do, to step a bit back [for a closer look at the data],” he said. “But I think we and others are going to confirm the data, and then perhaps put it into clinical practice.
In the meantime, Dr. Pfeiler recommended that clinicians keep the effect of BMI in mind when treating premenopausal breast cancer patients with anastrozole or tamoxifen.
Dr. Goodwin, in her discussion, said she did not consider this a practice-changing report. But after reviewing several important studies on aromatase inhibition, she said that the data raise the question of whether “aromatase inhibitors, ‘given at one-size-fits-all’ doses in overweight and obese women, are associated with optimal breast cancer outcomes.”
African-American Women, Highly Proliferative Tumors
A correlative science study based on a large Southwest Oncology Group (SWOG) trial of postmenopausal women with breast cancer found that reduced survival among African-American patients was due in large part to highly proliferative tumors, and not to differences in the degree of endocrine responsiveness.
“If confirmed, this finding might impact the choice of adjuvant therapy” in African American women with early ER-positive breast cancer, said first author Kathy S. Albain, MD, Professor of Medicine and Hematology/Oncology at Loyola University Chicago Stritch School of Medicine, speaking for the Breast Cancer Intergroup of North America.
Dr. Albain said an earlier analysis of SWOG adjuvant clinical breast cancer trials, with 6,676 patients, found that the 662 African-American women in the study had significantly worse survival rates than did others. (JNCI 2009; 101:984).
This finding persisted after data were adjusted for treatment, tumor factors, BMI, and socioeconomic status, and was true for both ER-positive and ER-negative cohorts, she said.
As triple negativity could not fully explain the disparity, the Intergroup researchers conducted this new analysis based on the ER-positive subset of patients in the SWOG Phase III 8814 study to determine whether the level of ER-positivity or expression of single genes within the 21-gene Recurrence Score assay might explain the disparity in outcomes.
S8814 compared outcomes of women receiving tamoxifen versus the CAF regimen plus tamoxifen. Among 1,477 patients, five-year disease-free survival for patients in all study arms was 63% for 139 African-American women, vs 74% for all others. Ten-year survival rates were 43% and 56%, respectively (Lancet 2009; 374:2055).
In the subset of women in S8814 available for Recurrence Score (RS) analysis of prognosis and prediction, the researchers assessed overall outcome by ER level, composite RS, and each of the genes analyzed by reverse transcriptase-polymerase chain reaction (RT-PCR).
African-American patients were found to have higher quantitative gene expressions by RT-PCR compared with other patients in all five genes in the proliferation group—CCNB1, MK167, MYBL2, BIRC5, and AURKA.
On the other hand, genes for estrogen-receptor, invasion, and HER-2 did not have a significant trend in African-Americans versus others, Dr. Albain said, “so it appears that the axis of genes involved in proliferation is important to look at.
“This is a new observation in a clinical trial population, suggesting there is something going on in the interactions of the host, environment of the host, and the actual tumor's biology that translates into a worse outcome. And this is despite their being enrolled on a state-of-the-art adjuvant trial.”
Discussant: Despite Limitations, Results Intriguing
In her discussion of this paper, Dr. Goodwin again noted that this was a post-hoc, unplanned analysis, and also that there were RS data on only one-quarter of women in the trial.
“These were small numbers, and the representativeness of the recurrence score to the full trial population is unclear,” she said. In addition, the contribution of socioeconomic and demographic attributes and BMI were not adequately examined in the study. “Nonetheless, the results are intriguing,” Dr. Goodwin said.