Meeting Highlights From the 2011 Breast Cancer Symposium

The following reports are based on presentations at the 2011 Breast Cancer Symposium, held September 8–10 in San Francisco. The symposium was sponsored in part by the American Society of Clinical Oncology.

Taxane-induced neuropathy gives no clue to breast cancer outcome
The development of taxaneinduced peripheral neuropathy does not appear to be a marker for better outcomes in women with early breast cancer, new data demonstrate.

In a secondary analysis of the Eastern Cooperative Oncology Group E1199 trial of 4,554 women, 13%–22% developed peripheral neuropathy, depending on which of four taxane regimens they received. Women who developed peripheral neuropathy did not have better recurrence- free, disease-free, or overall survival than did their counterparts who escaped this toxicity, lead investigator Dr. Bryan P. Schneider reported on behalf of several North American breast cancer groups.

Results were the same regardless of tumor type (estrogen receptor positive, HER2 positive, or triple negative) and the taxane regimen received. The findings are important given recent identification of six single-nucleotide polymorphisms (SNPs) predicting the likelihood of peripheral neuropathy in women receiving paclitaxel-containing chemotherapy (ASCO 2011, abstract 1000).

“Neuropathy did not correlate with outcome in E1199. The relevance is that SNPs or biomarkers that predict a subgroup protected from neuropathy are thus unlikely to enrich for inferior outcome,” he explained.

Discussant Dr. William J. Gradishar, director of the center for women’s cancer care at Northwestern University in Chicago, said that “at the end of the day, at least with the present data we have, it doesn’t appear that the presence of neuropathy has any bearing on overall clinical outcome from those end points.”

Trials thus far have failed to identify effective agents for preventing peripheral neuropathy in this setting, but there are some new agents on the horizon, he noted. “The hope will be, because these [chemotherapy drugs] are such a mainstay of our therapies in both the adjuvant and metastatic disease setting, that one or a few of [the new agents] may ultimately prove to be fruitful for our patients,” concluded Dr. Gradishar.

Women participating in the E1199 trial all received four cycles of doxorubicin and cyclophosphamide. They then received one of four taxane regimens: paclitaxel every 3 weeks for 4 cycles (control), paclitaxel weekly for 12 cycles, docetaxel every 3 weeks for 4 cycles, or docetaxel weekly for 12 cycles. Median follow-up was 95.5 months.

Peripheral neuropathy (grades 2–4, either sensory or motor) was the most common nonhematologic toxicity seen in the trial, according to Dr. Schneider of Indiana University in Indianapolis. It occurred in 22.0% of women treated with weekly paclitaxel, in 17.5% of women treated with paclitaxel every 3 weeks, in 14.7% of women treated with docetaxel every 3 weeks, and in 13.4% of women treated with weekly docetaxel.

In a multivariate model, women were significantly more likely to develop peripheral neuropathy if they had treatment-associated hyperglycemia (hazard ratio [HR], 1.47), and were marginally more likely if they were black, obese, or postmenopausal. Risk also varied by taxane regimen. It was higher for women who received weekly paclitaxel versus paclitaxel every 3 weeks (HR, 1.52), paclitaxel every 3 weeks versus weekly docetaxel (HR, 3.00), or paclitaxel every 3 weeks versus docetaxel every 3 weeks (HR, 1.91). In several types of analyses including multivariate models, the development of peripheral neuropathy was not significantly associated with recurrence- free, disease-free, or overall survival.

The findings contrast with an earlier study in which women receiving paclitaxel-containing therapy without trastuzumab did have better outcomes if they developed peripheral neuropathy (SABCS 2009, abstract A-2100).

Dr. Schneider reported that he had no relevant conflicts of interest. Dr. Gradishar had relationships with several companies that make taxanes.

Chemotherapy timing does not affect breast cancer recurrence

Women with early breast cancer who are treated with breast-conserving therapy have similar rates of freedom from locoregional recurrence whether chemotherapy is given before or after surgery, based on the experience of The University of Texas MD Anderson Cancer Center in Houston.

Investigators led by Dr. Elizabeth A. Mittendorf compared outcomes with these two approaches among nearly 3,000 women who had a segmental mastectomy plus whole-breast irradiation at the center in 1987–2005. Results showed that with a median follow-up of 7–8 years, the rate of freedom from locoregional recurrence was high (≥ 90% at 10 years) and did not differ significantly between the neoadjuvant and adjuvant chemotherapy groups after multiple potential confounders were taken into account.

“For appropriately selected patients, we can achieve high rates of locoregional control with breast-conserving therapy,” Dr. Mittendorf, a surgical oncologist, commented, adding, “It is important to obtain a negative margin at the time of surgery.”

“It appears that locoregional recurrence after breast-conserving therapy is driven primarily by [biological] factors and not by the timing of chemotherapy,” she said. However, discussant Dr. Barbara Fowble, a professor of clinical radiation oncology at the University of California San Francisco, said that the study was an example of “oversimplification of [a] complex issue … especially in the era of personalized therapy.”

She noted that factors such as breast cancer subtype (luminal A, luminal B, HER2 positive, or basal) can influence locoregional control after neoadjuvant chemotherapy with breast-conserving therapy.

“The identification of patients receiving neoadjuvant chemotherapy for whom mastectomy with or without radiation may result in a decreased risk of locoregional failure [when compared with breast-conserving therapy] is very important,” commented Dr. Fowble. “Initial stage, molecular subtype, and response to therapy will impact on outcome.”

Of the 2,984 women studied, 22% received neoadjuvant chemotherapy and 78% received adjuvant chemotherapy. Those in the former group were significantly more likely to have higher clinical stage and tumor grade, to have estrogen receptor-negative disease, and to have multifocal disease.

Median durations of follow-up were 7.2 and 7.9 years. In the neoadjuvant group, one-fifth of women had a pathological complete response, and there was a significant difference between the proportion having clinical stage II or III disease (93%) and the proportion having pathological stage II or III disease (46%; P < 0.001).

“Neoadjuvant chemotherapy downstages a significant number of patients with clinical stage II or III disease, presumably facilitating breast-conserving therapy,” Dr. Mittendorf said.

In an unadjusted analysis, the rate of freedom from locoregional recurrence was lower with neoadjuvant chemotherapy than with adjuvant chemotherapy (P < 0.001). For example, the 10-year rate was 90% with the former and 94% with the latter.

“Although the difference is statistically significant, I would suggest that these are excellent rates of locoregional control in our neoadjuvant chemotherapy patients, particularly considering that they presented with higher-stage disease, higher grade, and [more] ER-negative tumors,” she maintained.

In a multivariate analysis that took into account clinical and disease factors, the timing of chemotherapy was no longer significantly associated with the risk of locoregional recurrence. Eight factors conferred an elevated risk of such recurrence in this analysis: age younger than 50 years (HR, 1.9), clinical stage III disease (HR, 2.5), a grade 3 tumor (HR, 1.9), estrogen receptor-negative disease (HR, 2.4), multifocal disease (HR, 1.9), lymphovascular invasion (HR, 1.5), close or positive margins (HR, 2.5), and failure to receive hormone therapy in the context of estrogen receptor-positive disease (HR, 2.8).

When women were stratified according to the number of factors they had, there was still no significant difference in the rate of locoregional, recurrence- free survival between neoadjuvant and adjuvant chemotherapy, noted Dr. Mittendorf.

Dr. Mittendorf and Dr. Fowble reported no relevant conflicts of interest.

In the pipeline: entinostat may overcome AI resistance

Entinostat, a novel oral histone deacetylase inhibitor taken once weekly, may overcome resistance to hormonal therapy in breast cancer, the results of a randomized phase II trial suggest.

Investigators studied 130 women with estrogen receptor-positive advanced breast cancer progressing on aromatase inhibitor (AI) therapy. Results showed that, compared with exemestane plus placebo, exemestane plus entinostat reduced the risk of progression-free survival events by 27% and the risk of death by 44%. It also was well tolerated.

“The combination…is delaying the time to disease progression, allowing patients to maintain hormonal therapy longer, and delaying the need for change of therapy and subsequent considerations of chemotherapy,” principal investigator Dr. Denise A. Yardley reported.

“These results support our plans for a global, pivotal phase III study due to begin enrollment in 2012,” added Dr. Yardley, an oncologist with the Sarah Cannon Research Institute and Tennessee Oncology PLLC, both in Nashville.

In an additional finding, median progression-free survival was greatest, 8.5 months, in patients treated with the combination who had hyperacetylation of proteins in blood cells during the first cycle of treatment, suggesting that this could be a biomarker for efficacy.

These data reveal “for the first time in a matched controlled trial, evidence of hyperacetylation from an HDAC [histone deacetylase] inhibitor correlating with improved outcomes,” she noted.

Session chair Dr. Joyce O’Shaughnessy, an oncologist with the Baylor Sammons Cancer Center in Dallas, Texas Oncology, and US Oncology, characterized the trial’s findings as “very, very promising,” saying, “I agree that it’s time for a phase III” trial. To her knowledge, this is the first randomized trial of an HDAC inhibitor in breast cancer.

The randomized phase II trial, called ENCORE 301 (Entinostat Combinations Overcoming Resistance), was conducted among women in North America, the European Union, and Russia who were postmenopausal and had locally advanced or metastatic estrogen receptor-positive breast cancer that was progressing on a nonsteroidal AI (anastrozole or letrozole). Those with metastatic disease could have received up to one prior chemotherapy for metastases.

The patients were randomized in balanced fashion to double-blind treatment with exemestane plus placebo or exemestane plus entinostat. By study design, P values < 0.1 were considered statistically significant. Demographically, the patients were 62 years old on average. Eighty-five percent had experienced progression on an AI in the metastatic setting.

With a median follow-up of 18 months, entinostat was associated with longer median progression-free survival (4.28 vs 2.27 months; HR, 0.73; P = 0.06), with similar findings in subgroups stratified by locally advanced vs. metastatic disease, age, and site of metastasis, Dr. Yardley reported.

In 49 patients, blood samples collected before and during the first cycle of therapy were analyzed to determine the percent change of lysines in peripheral blood mononuclear cell proteins that were acetylated. Results showed that progression-free survival was greatest, at 8.54 months, among entinostat-treated patients who had hyperacetylation, defined as an abovemedian percent change in lysines that were acetylated during the first cycle of treatment.

These findings “suggest that entinostat-induced hyperacetylation appeared to track with improved outcomes in these patients and suggest quite provocatively within the very first cycle of treatment that this may be a potential marker to assess the benefit from this treatment,” said Dr. Yardley.

In an analysis that was exploratory because follow-up is still short, entinostat was also associated with better overall survival (26.9 vs 20.3 months; HR, 0.56; P = 0.027). Additional data with longer follow-up will be reported at the San Antonio Breast Cancer Symposium later this year.

The combination of entinostat and exemestane was well tolerated, with no unexpected safety signals. The most common grade 3/4 adverse events in the entinostat group were fatigue (13%), neutropenia (13%), nausea (5%), and vomiting (5%). The rate of discontinuation because of adverse events was 1% in the placebo group and 11% in the entinostat group.

Dr. Yardley reported no relevant conflicts of interest. Syndax Pharmaceuticals, developer of entinostat, was not among Dr. O’Shaughnessy’s disclosures.

Measure of bone metastases predicts breast cancer survival

The more metabolically active a woman’s bone metastases of breast cancer are on imaging, the greater her risk of death, researchers reported at the symposium.

In a retrospective cohort study of 269 women with newly diagnosed metastatic breast cancer, those whose bone metastases were “hotter” on a PET/CT—as assessed from maximum standardized uptake value (SUV-max)— had poorer overall survival. Compared with their counterparts with values in the bottom tertile of SUV-max values, women with values in the middle and top tertiles were roughly two to three times more likely to die after other prognostic factors were taken into account.

“To our understanding, this is the first large retrospective series to correlate SUV-max at metastatic breast cancer diagnosis with overall survival,” said lead author Dr. Komal Jhaveri of Memorial Sloan-Kettering Cancer Center, New York.

The risk of death also rose with increasing tertile of SUV-max in liver, lung, and lymph node metastases, but those associations were not signifycant. Their nonsignificance may have been due to the smaller numbers of patients having metastases in those sites, she said.

Session chair Dr. Robert R. Kuske, a radiation oncologist with Arizona Breast Cancer Specialists in Scottsdale, noted that SUV-max is a surrogate for how rapidly the tumor cells are proliferating and, hence, the tumor’s aggressiveness.

“So there should be a strong correlation between this SUV on the PET and the Ki-67 [histologic proliferation index],” he said. “This gives us another tool to evaluate how patients are going to do,” he commented. “Now [oncologists] can look at a metastasis on PET, measure the SUV, and get a handle on what the future of that patient is going to be—is it going to be a rapid slide to death, or is it going to be a very slow, indolent progression maybe over 10 or 15 years. And they can tailor the therapy based on the aggressiveness of the disease.” The PET/CT is usually done to determine whether the patient has metastases, so this additional prognostic information “is a little something extra you get from the test.”

Will these new findings be practice changing?

“I’ll start looking at SUV values in my patients with newly diagnosed metastasis,” said Dr. Kuske. “Yes, it will change my practice.”

In the study, Dr. Jhaveri and her coinvestigators reviewed records for women with metastatic breast cancer diagnosed between 2001 and 2008 who had a PET/CT scan performed within 60 days of the detection of metastases; had at least one avid metastasis in bone, liver, lung, or lymph node; and had not received chemotherapy in the month before the scan. Prior or current hormonal therapy was permitted.

The investigators determined the SUV-max for individual metastatic sites. They compared values only within a given site because of known variation across sites such as liver and lung, Dr. Jhaveri explained. The women had a median age of 58 years. The median time elapsed between primary breast cancer diagnosis and PET/CT was 2.3 years. A fifth each had triple-negative disease and HER2-positive disease. The median duration of followup was 40 months.

In a multivariate analysis that included other prognostic factors (grade, tumor phenotype, and visceral metastases), overall survival differed across tertiles of SUV-max for bone metastases (P = 0.006). Women with values in the middle and top tertiles were 1.87 and 2.67 times more likely to die, respectively, than their counterparts with values in the bottom tertile.

“Given the variation in SUV and the differential impact on survival by site, it is possible that SUV-max may not be the most optimal PET variable,” Dr. Jhaveri commented. She proposed that an alternate variable, called total lesion glycolysis (TLG), which incorporates tumor lesion size, might perform better and be more informative.

“Ultimately, prospective studies are required to further delineate the role of PET/CT as a prognostic tool in metastatic breast cancer,” she concluded.

Dr. Jhaveri and Dr. Kuske reported having no relevant conflicts of interest.

Score predicts late recurrence in ER-positive breast cancer

A new gene-based biomarker assay may help to estimate prognosis and make decisions about extended endocrine therapy in women with estrogen receptor-positive early breast cancer, new data suggest.

Researchers tested the biomarker, the Breast Cancer Index (BCI), among 249 women with estrogen receptor-positive early breast cancer from the MA.17 trial, in which women received 5 years of adjuvant tamoxifen therapy and, if still disease free, were then randomized to an additional 5 years of either letrozole or placebo.

Study results, reported at the meeting, showed that with each 5-unit increase in the 10-unit BCI score, women’s odds of late recurrence nearly tripled. The index as a whole was not helpful in predicting the benefit of added letrozole therapy in reducing recurrence risk. But one of its components, called H/I, was helpful: Women having a high H/I were about half as likely to have a recurrence if they received letrozole instead of placebo.

In the future, these findings might be used to develop a management algorithm for women with estrogen receptor-positive breast cancer, according to lead investigator Dr. Dennis C. Sgroi, director of breast pathology at the Massachusetts General Hospital in Boston.

“If patients are disease free after 5 years of adjuvant endocrine therapy, we might be able to then test them with the BCI. If this assay identifies these patients to be at low risk, there will be no further therapy for these patients. However, if they are at high risk by BCI, we can then explore the H/I component,” he explained.

“If they have high H/I, that indicates that these patients are likely to benefit from extended adjuvant therapy,” he continued. “For patients who have low H/I, one might consider using extended adjuvant therapy, or these patients might be considered as a future focus for research in clinical trials.

The investigators identified 83 women from the MA.17 trial who had had a recurrence and had primary tumor tissue available. They then matched the women by age, N stage, T stage, and prior receipt of chemotherapy in a 1:2 ratio with 166 women who had not had a recurrence and had primary tumor tissue available. Tissue was analyzed by reverse transcriptase- polymerase chain reaction to determine the BCI.

The BCI has two components, Dr. Sgroi explained. The HOXB13/IL- 17BR (H/I) component is based on expression of two genes regulated by estradiol. The molecular grade index (MGI) component is based on five genes related to pathological grade.

“We have shown in the past that these two biomarkers are complementary, as the combination of the two biomarkers outperforms each individually,” he noted. “In addition, we have shown that the combination biomarker outperforms standard clinicopathological parameters currently used to predict disease recurrence.”

In a multivariate analysis, with each 5-unit increase in the BCI score, women’s odds of recurrence increased 2.91-fold (P = 0.014). The findings were similar when categories were used: Women with high or intermediate scores had 2.21-fold higher odds of recurrence than did their counterparts with low scores (P = 0.019).

The BCI as a whole did not predict the benefit of extended therapy with letrozole, but the H/I component did, Dr. Sgroi reported.

In a multivariate analysis, women who had a high H/I had a 58% reduction in the odds of recurrence if they received letrozole instead of placebo (P = 0.037).

“Most notably, there was a statistically significant interaction between the H/I biomarker and treatment (P = 0.02), indicating that the magnitude of letrozole benefit depends on the expression level of H/I,” he said. “These results suggest that patients with tumors expressing high H/I will benefit from extended endocrine therapy.”

Dr. Sgroi reported receiving research funding from bioTheranostics Inc. The study was supported in part by Novartis, the manufacturer of letrozole.

The following reports are based on presentations at the 2011 Breast Cancer Symposium, held September 8–10 in San Francisco. The symposium was sponsored in part by the American Society of Clinical Oncology.

Taxane-induced neuropathy gives no clue to breast cancer outcome
The development of taxaneinduced peripheral neuropathy does not appear to be a marker for better outcomes in women with early breast cancer, new data demonstrate.

In a secondary analysis of the Eastern Cooperative Oncology Group E1199 trial of 4,554 women, 13%–22% developed peripheral neuropathy, depending on which of four taxane regimens they received. Women who developed peripheral neuropathy did not have better recurrence- free, disease-free, or overall survival than did their counterparts who escaped this toxicity, lead investigator Dr. Bryan P. Schneider reported on behalf of several North American breast cancer groups.

Results were the same regardless of tumor type (estrogen receptor positive, HER2 positive, or triple negative) and the taxane regimen received. The findings are important given recent identification of six single-nucleotide polymorphisms (SNPs) predicting the likelihood of peripheral neuropathy in women receiving paclitaxel-containing chemotherapy (ASCO 2011, abstract 1000).

“Neuropathy did not correlate with outcome in E1199. The relevance is that SNPs or biomarkers that predict a subgroup protected from neuropathy are thus unlikely to enrich for inferior outcome,” he explained.

Discussant Dr. William J. Gradishar, director of the center for women’s cancer care at Northwestern University in Chicago, said that “at the end of the day, at least with the present data we have, it doesn’t appear that the presence of neuropathy has any bearing on overall clinical outcome from those end points.”

Trials thus far have failed to identify effective agents for preventing peripheral neuropathy in this setting, but there are some new agents on the horizon, he noted. “The hope will be, because these [chemotherapy drugs] are such a mainstay of our therapies in both the adjuvant and metastatic disease setting, that one or a few of [the new agents] may ultimately prove to be fruitful for our patients,” concluded Dr. Gradishar.

Women participating in the E1199 trial all received four cycles of doxorubicin and cyclophosphamide. They then received one of four taxane regimens: paclitaxel every 3 weeks for 4 cycles (control), paclitaxel weekly for 12 cycles, docetaxel every 3 weeks for 4 cycles, or docetaxel weekly for 12 cycles. Median follow-up was 95.5 months.

Peripheral neuropathy (grades 2–4, either sensory or motor) was the most common nonhematologic toxicity seen in the trial, according to Dr. Schneider of Indiana University in Indianapolis. It occurred in 22.0% of women treated with weekly paclitaxel, in 17.5% of women treated with paclitaxel every 3 weeks, in 14.7% of women treated with docetaxel every 3 weeks, and in 13.4% of women treated with weekly docetaxel.

In a multivariate model, women were significantly more likely to develop peripheral neuropathy if they had treatment-associated hyperglycemia (hazard ratio [HR], 1.47), and were marginally more likely if they were black, obese, or postmenopausal. Risk also varied by taxane regimen. It was higher for women who received weekly paclitaxel versus paclitaxel every 3 weeks (HR, 1.52), paclitaxel every 3 weeks versus weekly docetaxel (HR, 3.00), or paclitaxel every 3 weeks versus docetaxel every 3 weeks (HR, 1.91). In several types of analyses including multivariate models, the development of peripheral neuropathy was not significantly associated with recurrence- free, disease-free, or overall survival.

The findings contrast with an earlier study in which women receiving paclitaxel-containing therapy without trastuzumab did have better outcomes if they developed peripheral neuropathy (SABCS 2009, abstract A-2100).

Dr. Schneider reported that he had no relevant conflicts of interest. Dr. Gradishar had relationships with several companies that make taxanes.

Chemotherapy timing does not affect breast cancer recurrence

Women with early breast cancer who are treated with breast-conserving therapy have similar rates of freedom from locoregional recurrence whether chemotherapy is given before or after surgery, based on the experience of The University of Texas MD Anderson Cancer Center in Houston.

Investigators led by Dr. Elizabeth A. Mittendorf compared outcomes with these two approaches among nearly 3,000 women who had a segmental mastectomy plus whole-breast irradiation at the center in 1987–2005. Results showed that with a median follow-up of 7–8 years, the rate of freedom from locoregional recurrence was high (≥ 90% at 10 years) and did not differ significantly between the neoadjuvant and adjuvant chemotherapy groups after multiple potential confounders were taken into account.

“For appropriately selected patients, we can achieve high rates of locoregional control with breast-conserving therapy,” Dr. Mittendorf, a surgical oncologist, commented, adding, “It is important to obtain a negative margin at the time of surgery.”

“It appears that locoregional recurrence after breast-conserving therapy is driven primarily by [biological] factors and not by the timing of chemotherapy,” she said. However, discussant Dr. Barbara Fowble, a professor of clinical radiation oncology at the University of California San Francisco, said that the study was an example of “oversimplification of [a] complex issue … especially in the era of personalized therapy.”

She noted that factors such as breast cancer subtype (luminal A, luminal B, HER2 positive, or basal) can influence locoregional control after neoadjuvant chemotherapy with breast-conserving therapy.

“The identification of patients receiving neoadjuvant chemotherapy for whom mastectomy with or without radiation may result in a decreased risk of locoregional failure [when compared with breast-conserving therapy] is very important,” commented Dr. Fowble. “Initial stage, molecular subtype, and response to therapy will impact on outcome.”

Of the 2,984 women studied, 22% received neoadjuvant chemotherapy and 78% received adjuvant chemotherapy. Those in the former group were significantly more likely to have higher clinical stage and tumor grade, to have estrogen receptor-negative disease, and to have multifocal disease.

Median durations of follow-up were 7.2 and 7.9 years. In the neoadjuvant group, one-fifth of women had a pathological complete response, and there was a significant difference between the proportion having clinical stage II or III disease (93%) and the proportion having pathological stage II or III disease (46%; P < 0.001).

“Neoadjuvant chemotherapy downstages a significant number of patients with clinical stage II or III disease, presumably facilitating breast-conserving therapy,” Dr. Mittendorf said.

In an unadjusted analysis, the rate of freedom from locoregional recurrence was lower with neoadjuvant chemotherapy than with adjuvant chemotherapy (P < 0.001). For example, the 10-year rate was 90% with the former and 94% with the latter.

“Although the difference is statistically significant, I would suggest that these are excellent rates of locoregional control in our neoadjuvant chemotherapy patients, particularly considering that they presented with higher-stage disease, higher grade, and [more] ER-negative tumors,” she maintained.

In a multivariate analysis that took into account clinical and disease factors, the timing of chemotherapy was no longer significantly associated with the risk of locoregional recurrence. Eight factors conferred an elevated risk of such recurrence in this analysis: age younger than 50 years (HR, 1.9), clinical stage III disease (HR, 2.5), a grade 3 tumor (HR, 1.9), estrogen receptor-negative disease (HR, 2.4), multifocal disease (HR, 1.9), lymphovascular invasion (HR, 1.5), close or positive margins (HR, 2.5), and failure to receive hormone therapy in the context of estrogen receptor-positive disease (HR, 2.8).

When women were stratified according to the number of factors they had, there was still no significant difference in the rate of locoregional, recurrence- free survival between neoadjuvant and adjuvant chemotherapy, noted Dr. Mittendorf.

Dr. Mittendorf and Dr. Fowble reported no relevant conflicts of interest.

In the pipeline: entinostat may overcome AI resistance

Entinostat, a novel oral histone deacetylase inhibitor taken once weekly, may overcome resistance to hormonal therapy in breast cancer, the results of a randomized phase II trial suggest.

Investigators studied 130 women with estrogen receptor-positive advanced breast cancer progressing on aromatase inhibitor (AI) therapy. Results showed that, compared with exemestane plus placebo, exemestane plus entinostat reduced the risk of progression-free survival events by 27% and the risk of death by 44%. It also was well tolerated.

“The combination…is delaying the time to disease progression, allowing patients to maintain hormonal therapy longer, and delaying the need for change of therapy and subsequent considerations of chemotherapy,” principal investigator Dr. Denise A. Yardley reported.

“These results support our plans for a global, pivotal phase III study due to begin enrollment in 2012,” added Dr. Yardley, an oncologist with the Sarah Cannon Research Institute and Tennessee Oncology PLLC, both in Nashville.

In an additional finding, median progression-free survival was greatest, 8.5 months, in patients treated with the combination who had hyperacetylation of proteins in blood cells during the first cycle of treatment, suggesting that this could be a biomarker for efficacy.

These data reveal “for the first time in a matched controlled trial, evidence of hyperacetylation from an HDAC [histone deacetylase] inhibitor correlating with improved outcomes,” she noted.

Session chair Dr. Joyce O’Shaughnessy, an oncologist with the Baylor Sammons Cancer Center in Dallas, Texas Oncology, and US Oncology, characterized the trial’s findings as “very, very promising,” saying, “I agree that it’s time for a phase III” trial. To her knowledge, this is the first randomized trial of an HDAC inhibitor in breast cancer.

The randomized phase II trial, called ENCORE 301 (Entinostat Combinations Overcoming Resistance), was conducted among women in North America, the European Union, and Russia who were postmenopausal and had locally advanced or metastatic estrogen receptor-positive breast cancer that was progressing on a nonsteroidal AI (anastrozole or letrozole). Those with metastatic disease could have received up to one prior chemotherapy for metastases.

The patients were randomized in balanced fashion to double-blind treatment with exemestane plus placebo or exemestane plus entinostat. By study design, P values < 0.1 were considered statistically significant. Demographically, the patients were 62 years old on average. Eighty-five percent had experienced progression on an AI in the metastatic setting.

With a median follow-up of 18 months, entinostat was associated with longer median progression-free survival (4.28 vs 2.27 months; HR, 0.73; P = 0.06), with similar findings in subgroups stratified by locally advanced vs. metastatic disease, age, and site of metastasis, Dr. Yardley reported.

In 49 patients, blood samples collected before and during the first cycle of therapy were analyzed to determine the percent change of lysines in peripheral blood mononuclear cell proteins that were acetylated. Results showed that progression-free survival was greatest, at 8.54 months, among entinostat-treated patients who had hyperacetylation, defined as an abovemedian percent change in lysines that were acetylated during the first cycle of treatment.

These findings “suggest that entinostat-induced hyperacetylation appeared to track with improved outcomes in these patients and suggest quite provocatively within the very first cycle of treatment that this may be a potential marker to assess the benefit from this treatment,” said Dr. Yardley.

In an analysis that was exploratory because follow-up is still short, entinostat was also associated with better overall survival (26.9 vs 20.3 months; HR, 0.56; P = 0.027). Additional data with longer follow-up will be reported at the San Antonio Breast Cancer Symposium later this year.

The combination of entinostat and exemestane was well tolerated, with no unexpected safety signals. The most common grade 3/4 adverse events in the entinostat group were fatigue (13%), neutropenia (13%), nausea (5%), and vomiting (5%). The rate of discontinuation because of adverse events was 1% in the placebo group and 11% in the entinostat group.

Dr. Yardley reported no relevant conflicts of interest. Syndax Pharmaceuticals, developer of entinostat, was not among Dr. O’Shaughnessy’s disclosures.

Measure of bone metastases predicts breast cancer survival

The more metabolically active a woman’s bone metastases of breast cancer are on imaging, the greater her risk of death, researchers reported at the symposium.

In a retrospective cohort study of 269 women with newly diagnosed metastatic breast cancer, those whose bone metastases were “hotter” on a PET/CT—as assessed from maximum standardized uptake value (SUV-max)— had poorer overall survival. Compared with their counterparts with values in the bottom tertile of SUV-max values, women with values in the middle and top tertiles were roughly two to three times more likely to die after other prognostic factors were taken into account.

“To our understanding, this is the first large retrospective series to correlate SUV-max at metastatic breast cancer diagnosis with overall survival,” said lead author Dr. Komal Jhaveri of Memorial Sloan-Kettering Cancer Center, New York.

The risk of death also rose with increasing tertile of SUV-max in liver, lung, and lymph node metastases, but those associations were not signifycant. Their nonsignificance may have been due to the smaller numbers of patients having metastases in those sites, she said.

Session chair Dr. Robert R. Kuske, a radiation oncologist with Arizona Breast Cancer Specialists in Scottsdale, noted that SUV-max is a surrogate for how rapidly the tumor cells are proliferating and, hence, the tumor’s aggressiveness.

“So there should be a strong correlation between this SUV on the PET and the Ki-67 [histologic proliferation index],” he said. “This gives us another tool to evaluate how patients are going to do,” he commented. “Now [oncologists] can look at a metastasis on PET, measure the SUV, and get a handle on what the future of that patient is going to be—is it going to be a rapid slide to death, or is it going to be a very slow, indolent progression maybe over 10 or 15 years. And they can tailor the therapy based on the aggressiveness of the disease.” The PET/CT is usually done to determine whether the patient has metastases, so this additional prognostic information “is a little something extra you get from the test.”

Will these new findings be practice changing?

“I’ll start looking at SUV values in my patients with newly diagnosed metastasis,” said Dr. Kuske. “Yes, it will change my practice.”

In the study, Dr. Jhaveri and her coinvestigators reviewed records for women with metastatic breast cancer diagnosed between 2001 and 2008 who had a PET/CT scan performed within 60 days of the detection of metastases; had at least one avid metastasis in bone, liver, lung, or lymph node; and had not received chemotherapy in the month before the scan. Prior or current hormonal therapy was permitted.

The investigators determined the SUV-max for individual metastatic sites. They compared values only within a given site because of known variation across sites such as liver and lung, Dr. Jhaveri explained. The women had a median age of 58 years. The median time elapsed between primary breast cancer diagnosis and PET/CT was 2.3 years. A fifth each had triple-negative disease and HER2-positive disease. The median duration of followup was 40 months.

In a multivariate analysis that included other prognostic factors (grade, tumor phenotype, and visceral metastases), overall survival differed across tertiles of SUV-max for bone metastases (P = 0.006). Women with values in the middle and top tertiles were 1.87 and 2.67 times more likely to die, respectively, than their counterparts with values in the bottom tertile.

“Given the variation in SUV and the differential impact on survival by site, it is possible that SUV-max may not be the most optimal PET variable,” Dr. Jhaveri commented. She proposed that an alternate variable, called total lesion glycolysis (TLG), which incorporates tumor lesion size, might perform better and be more informative.

“Ultimately, prospective studies are required to further delineate the role of PET/CT as a prognostic tool in metastatic breast cancer,” she concluded.

Dr. Jhaveri and Dr. Kuske reported having no relevant conflicts of interest.

Score predicts late recurrence in ER-positive breast cancer

A new gene-based biomarker assay may help to estimate prognosis and make decisions about extended endocrine therapy in women with estrogen receptor-positive early breast cancer, new data suggest.

Researchers tested the biomarker, the Breast Cancer Index (BCI), among 249 women with estrogen receptor-positive early breast cancer from the MA.17 trial, in which women received 5 years of adjuvant tamoxifen therapy and, if still disease free, were then randomized to an additional 5 years of either letrozole or placebo.

Study results, reported at the meeting, showed that with each 5-unit increase in the 10-unit BCI score, women’s odds of late recurrence nearly tripled. The index as a whole was not helpful in predicting the benefit of added letrozole therapy in reducing recurrence risk. But one of its components, called H/I, was helpful: Women having a high H/I were about half as likely to have a recurrence if they received letrozole instead of placebo.

In the future, these findings might be used to develop a management algorithm for women with estrogen receptor-positive breast cancer, according to lead investigator Dr. Dennis C. Sgroi, director of breast pathology at the Massachusetts General Hospital in Boston.

“If patients are disease free after 5 years of adjuvant endocrine therapy, we might be able to then test them with the BCI. If this assay identifies these patients to be at low risk, there will be no further therapy for these patients. However, if they are at high risk by BCI, we can then explore the H/I component,” he explained.

“If they have high H/I, that indicates that these patients are likely to benefit from extended adjuvant therapy,” he continued. “For patients who have low H/I, one might consider using extended adjuvant therapy, or these patients might be considered as a future focus for research in clinical trials.

The investigators identified 83 women from the MA.17 trial who had had a recurrence and had primary tumor tissue available. They then matched the women by age, N stage, T stage, and prior receipt of chemotherapy in a 1:2 ratio with 166 women who had not had a recurrence and had primary tumor tissue available. Tissue was analyzed by reverse transcriptase- polymerase chain reaction to determine the BCI.

The BCI has two components, Dr. Sgroi explained. The HOXB13/IL- 17BR (H/I) component is based on expression of two genes regulated by estradiol. The molecular grade index (MGI) component is based on five genes related to pathological grade.

“We have shown in the past that these two biomarkers are complementary, as the combination of the two biomarkers outperforms each individually,” he noted. “In addition, we have shown that the combination biomarker outperforms standard clinicopathological parameters currently used to predict disease recurrence.”

In a multivariate analysis, with each 5-unit increase in the BCI score, women’s odds of recurrence increased 2.91-fold (P = 0.014). The findings were similar when categories were used: Women with high or intermediate scores had 2.21-fold higher odds of recurrence than did their counterparts with low scores (P = 0.019).

The BCI as a whole did not predict the benefit of extended therapy with letrozole, but the H/I component did, Dr. Sgroi reported.

In a multivariate analysis, women who had a high H/I had a 58% reduction in the odds of recurrence if they received letrozole instead of placebo (P = 0.037).

“Most notably, there was a statistically significant interaction between the H/I biomarker and treatment (P = 0.02), indicating that the magnitude of letrozole benefit depends on the expression level of H/I,” he said. “These results suggest that patients with tumors expressing high H/I will benefit from extended endocrine therapy.”

Dr. Sgroi reported receiving research funding from bioTheranostics Inc. The study was supported in part by Novartis, the manufacturer of letrozole.

The following reports are based on presentations at the 2011 Breast Cancer Symposium, held September 8–10 in San Francisco. The symposium was sponsored in part by the American Society of Clinical Oncology.

Taxane-induced neuropathy gives no clue to breast cancer outcome
The development of taxaneinduced peripheral neuropathy does not appear to be a marker for better outcomes in women with early breast cancer, new data demonstrate.

In a secondary analysis of the Eastern Cooperative Oncology Group E1199 trial of 4,554 women, 13%–22% developed peripheral neuropathy, depending on which of four taxane regimens they received. Women who developed peripheral neuropathy did not have better recurrence- free, disease-free, or overall survival than did their counterparts who escaped this toxicity, lead investigator Dr. Bryan P. Schneider reported on behalf of several North American breast cancer groups.

Results were the same regardless of tumor type (estrogen receptor positive, HER2 positive, or triple negative) and the taxane regimen received. The findings are important given recent identification of six single-nucleotide polymorphisms (SNPs) predicting the likelihood of peripheral neuropathy in women receiving paclitaxel-containing chemotherapy (ASCO 2011, abstract 1000).

“Neuropathy did not correlate with outcome in E1199. The relevance is that SNPs or biomarkers that predict a subgroup protected from neuropathy are thus unlikely to enrich for inferior outcome,” he explained.

Discussant Dr. William J. Gradishar, director of the center for women’s cancer care at Northwestern University in Chicago, said that “at the end of the day, at least with the present data we have, it doesn’t appear that the presence of neuropathy has any bearing on overall clinical outcome from those end points.”

Trials thus far have failed to identify effective agents for preventing peripheral neuropathy in this setting, but there are some new agents on the horizon, he noted. “The hope will be, because these [chemotherapy drugs] are such a mainstay of our therapies in both the adjuvant and metastatic disease setting, that one or a few of [the new agents] may ultimately prove to be fruitful for our patients,” concluded Dr. Gradishar.

Women participating in the E1199 trial all received four cycles of doxorubicin and cyclophosphamide. They then received one of four taxane regimens: paclitaxel every 3 weeks for 4 cycles (control), paclitaxel weekly for 12 cycles, docetaxel every 3 weeks for 4 cycles, or docetaxel weekly for 12 cycles. Median follow-up was 95.5 months.

Peripheral neuropathy (grades 2–4, either sensory or motor) was the most common nonhematologic toxicity seen in the trial, according to Dr. Schneider of Indiana University in Indianapolis. It occurred in 22.0% of women treated with weekly paclitaxel, in 17.5% of women treated with paclitaxel every 3 weeks, in 14.7% of women treated with docetaxel every 3 weeks, and in 13.4% of women treated with weekly docetaxel.

In a multivariate model, women were significantly more likely to develop peripheral neuropathy if they had treatment-associated hyperglycemia (hazard ratio [HR], 1.47), and were marginally more likely if they were black, obese, or postmenopausal. Risk also varied by taxane regimen. It was higher for women who received weekly paclitaxel versus paclitaxel every 3 weeks (HR, 1.52), paclitaxel every 3 weeks versus weekly docetaxel (HR, 3.00), or paclitaxel every 3 weeks versus docetaxel every 3 weeks (HR, 1.91). In several types of analyses including multivariate models, the development of peripheral neuropathy was not significantly associated with recurrence- free, disease-free, or overall survival.

The findings contrast with an earlier study in which women receiving paclitaxel-containing therapy without trastuzumab did have better outcomes if they developed peripheral neuropathy (SABCS 2009, abstract A-2100).

Dr. Schneider reported that he had no relevant conflicts of interest. Dr. Gradishar had relationships with several companies that make taxanes.

Chemotherapy timing does not affect breast cancer recurrence

Women with early breast cancer who are treated with breast-conserving therapy have similar rates of freedom from locoregional recurrence whether chemotherapy is given before or after surgery, based on the experience of The University of Texas MD Anderson Cancer Center in Houston.

Investigators led by Dr. Elizabeth A. Mittendorf compared outcomes with these two approaches among nearly 3,000 women who had a segmental mastectomy plus whole-breast irradiation at the center in 1987–2005. Results showed that with a median follow-up of 7–8 years, the rate of freedom from locoregional recurrence was high (≥ 90% at 10 years) and did not differ significantly between the neoadjuvant and adjuvant chemotherapy groups after multiple potential confounders were taken into account.

“For appropriately selected patients, we can achieve high rates of locoregional control with breast-conserving therapy,” Dr. Mittendorf, a surgical oncologist, commented, adding, “It is important to obtain a negative margin at the time of surgery.”

“It appears that locoregional recurrence after breast-conserving therapy is driven primarily by [biological] factors and not by the timing of chemotherapy,” she said. However, discussant Dr. Barbara Fowble, a professor of clinical radiation oncology at the University of California San Francisco, said that the study was an example of “oversimplification of [a] complex issue … especially in the era of personalized therapy.”

She noted that factors such as breast cancer subtype (luminal A, luminal B, HER2 positive, or basal) can influence locoregional control after neoadjuvant chemotherapy with breast-conserving therapy.

“The identification of patients receiving neoadjuvant chemotherapy for whom mastectomy with or without radiation may result in a decreased risk of locoregional failure [when compared with breast-conserving therapy] is very important,” commented Dr. Fowble. “Initial stage, molecular subtype, and response to therapy will impact on outcome.”

Of the 2,984 women studied, 22% received neoadjuvant chemotherapy and 78% received adjuvant chemotherapy. Those in the former group were significantly more likely to have higher clinical stage and tumor grade, to have estrogen receptor-negative disease, and to have multifocal disease.

Median durations of follow-up were 7.2 and 7.9 years. In the neoadjuvant group, one-fifth of women had a pathological complete response, and there was a significant difference between the proportion having clinical stage II or III disease (93%) and the proportion having pathological stage II or III disease (46%; P < 0.001).

“Neoadjuvant chemotherapy downstages a significant number of patients with clinical stage II or III disease, presumably facilitating breast-conserving therapy,” Dr. Mittendorf said.

In an unadjusted analysis, the rate of freedom from locoregional recurrence was lower with neoadjuvant chemotherapy than with adjuvant chemotherapy (P < 0.001). For example, the 10-year rate was 90% with the former and 94% with the latter.

“Although the difference is statistically significant, I would suggest that these are excellent rates of locoregional control in our neoadjuvant chemotherapy patients, particularly considering that they presented with higher-stage disease, higher grade, and [more] ER-negative tumors,” she maintained.

In a multivariate analysis that took into account clinical and disease factors, the timing of chemotherapy was no longer significantly associated with the risk of locoregional recurrence. Eight factors conferred an elevated risk of such recurrence in this analysis: age younger than 50 years (HR, 1.9), clinical stage III disease (HR, 2.5), a grade 3 tumor (HR, 1.9), estrogen receptor-negative disease (HR, 2.4), multifocal disease (HR, 1.9), lymphovascular invasion (HR, 1.5), close or positive margins (HR, 2.5), and failure to receive hormone therapy in the context of estrogen receptor-positive disease (HR, 2.8).

When women were stratified according to the number of factors they had, there was still no significant difference in the rate of locoregional, recurrence- free survival between neoadjuvant and adjuvant chemotherapy, noted Dr. Mittendorf.

Dr. Mittendorf and Dr. Fowble reported no relevant conflicts of interest.

In the pipeline: entinostat may overcome AI resistance

Entinostat, a novel oral histone deacetylase inhibitor taken once weekly, may overcome resistance to hormonal therapy in breast cancer, the results of a randomized phase II trial suggest.

Investigators studied 130 women with estrogen receptor-positive advanced breast cancer progressing on aromatase inhibitor (AI) therapy. Results showed that, compared with exemestane plus placebo, exemestane plus entinostat reduced the risk of progression-free survival events by 27% and the risk of death by 44%. It also was well tolerated.

“The combination…is delaying the time to disease progression, allowing patients to maintain hormonal therapy longer, and delaying the need for change of therapy and subsequent considerations of chemotherapy,” principal investigator Dr. Denise A. Yardley reported.

“These results support our plans for a global, pivotal phase III study due to begin enrollment in 2012,” added Dr. Yardley, an oncologist with the Sarah Cannon Research Institute and Tennessee Oncology PLLC, both in Nashville.

In an additional finding, median progression-free survival was greatest, 8.5 months, in patients treated with the combination who had hyperacetylation of proteins in blood cells during the first cycle of treatment, suggesting that this could be a biomarker for efficacy.

These data reveal “for the first time in a matched controlled trial, evidence of hyperacetylation from an HDAC [histone deacetylase] inhibitor correlating with improved outcomes,” she noted.

Session chair Dr. Joyce O’Shaughnessy, an oncologist with the Baylor Sammons Cancer Center in Dallas, Texas Oncology, and US Oncology, characterized the trial’s findings as “very, very promising,” saying, “I agree that it’s time for a phase III” trial. To her knowledge, this is the first randomized trial of an HDAC inhibitor in breast cancer.

The randomized phase II trial, called ENCORE 301 (Entinostat Combinations Overcoming Resistance), was conducted among women in North America, the European Union, and Russia who were postmenopausal and had locally advanced or metastatic estrogen receptor-positive breast cancer that was progressing on a nonsteroidal AI (anastrozole or letrozole). Those with metastatic disease could have received up to one prior chemotherapy for metastases.

The patients were randomized in balanced fashion to double-blind treatment with exemestane plus placebo or exemestane plus entinostat. By study design, P values < 0.1 were considered statistically significant. Demographically, the patients were 62 years old on average. Eighty-five percent had experienced progression on an AI in the metastatic setting.

With a median follow-up of 18 months, entinostat was associated with longer median progression-free survival (4.28 vs 2.27 months; HR, 0.73; P = 0.06), with similar findings in subgroups stratified by locally advanced vs. metastatic disease, age, and site of metastasis, Dr. Yardley reported.

In 49 patients, blood samples collected before and during the first cycle of therapy were analyzed to determine the percent change of lysines in peripheral blood mononuclear cell proteins that were acetylated. Results showed that progression-free survival was greatest, at 8.54 months, among entinostat-treated patients who had hyperacetylation, defined as an abovemedian percent change in lysines that were acetylated during the first cycle of treatment.

These findings “suggest that entinostat-induced hyperacetylation appeared to track with improved outcomes in these patients and suggest quite provocatively within the very first cycle of treatment that this may be a potential marker to assess the benefit from this treatment,” said Dr. Yardley.

In an analysis that was exploratory because follow-up is still short, entinostat was also associated with better overall survival (26.9 vs 20.3 months; HR, 0.56; P = 0.027). Additional data with longer follow-up will be reported at the San Antonio Breast Cancer Symposium later this year.

The combination of entinostat and exemestane was well tolerated, with no unexpected safety signals. The most common grade 3/4 adverse events in the entinostat group were fatigue (13%), neutropenia (13%), nausea (5%), and vomiting (5%). The rate of discontinuation because of adverse events was 1% in the placebo group and 11% in the entinostat group.

Dr. Yardley reported no relevant conflicts of interest. Syndax Pharmaceuticals, developer of entinostat, was not among Dr. O’Shaughnessy’s disclosures.

Measure of bone metastases predicts breast cancer survival

The more metabolically active a woman’s bone metastases of breast cancer are on imaging, the greater her risk of death, researchers reported at the symposium.

In a retrospective cohort study of 269 women with newly diagnosed metastatic breast cancer, those whose bone metastases were “hotter” on a PET/CT—as assessed from maximum standardized uptake value (SUV-max)— had poorer overall survival. Compared with their counterparts with values in the bottom tertile of SUV-max values, women with values in the middle and top tertiles were roughly two to three times more likely to die after other prognostic factors were taken into account.

“To our understanding, this is the first large retrospective series to correlate SUV-max at metastatic breast cancer diagnosis with overall survival,” said lead author Dr. Komal Jhaveri of Memorial Sloan-Kettering Cancer Center, New York.

The risk of death also rose with increasing tertile of SUV-max in liver, lung, and lymph node metastases, but those associations were not signifycant. Their nonsignificance may have been due to the smaller numbers of patients having metastases in those sites, she said.

Session chair Dr. Robert R. Kuske, a radiation oncologist with Arizona Breast Cancer Specialists in Scottsdale, noted that SUV-max is a surrogate for how rapidly the tumor cells are proliferating and, hence, the tumor’s aggressiveness.

“So there should be a strong correlation between this SUV on the PET and the Ki-67 [histologic proliferation index],” he said. “This gives us another tool to evaluate how patients are going to do,” he commented. “Now [oncologists] can look at a metastasis on PET, measure the SUV, and get a handle on what the future of that patient is going to be—is it going to be a rapid slide to death, or is it going to be a very slow, indolent progression maybe over 10 or 15 years. And they can tailor the therapy based on the aggressiveness of the disease.” The PET/CT is usually done to determine whether the patient has metastases, so this additional prognostic information “is a little something extra you get from the test.”

Will these new findings be practice changing?

“I’ll start looking at SUV values in my patients with newly diagnosed metastasis,” said Dr. Kuske. “Yes, it will change my practice.”

In the study, Dr. Jhaveri and her coinvestigators reviewed records for women with metastatic breast cancer diagnosed between 2001 and 2008 who had a PET/CT scan performed within 60 days of the detection of metastases; had at least one avid metastasis in bone, liver, lung, or lymph node; and had not received chemotherapy in the month before the scan. Prior or current hormonal therapy was permitted.

The investigators determined the SUV-max for individual metastatic sites. They compared values only within a given site because of known variation across sites such as liver and lung, Dr. Jhaveri explained. The women had a median age of 58 years. The median time elapsed between primary breast cancer diagnosis and PET/CT was 2.3 years. A fifth each had triple-negative disease and HER2-positive disease. The median duration of followup was 40 months.

In a multivariate analysis that included other prognostic factors (grade, tumor phenotype, and visceral metastases), overall survival differed across tertiles of SUV-max for bone metastases (P = 0.006). Women with values in the middle and top tertiles were 1.87 and 2.67 times more likely to die, respectively, than their counterparts with values in the bottom tertile.

“Given the variation in SUV and the differential impact on survival by site, it is possible that SUV-max may not be the most optimal PET variable,” Dr. Jhaveri commented. She proposed that an alternate variable, called total lesion glycolysis (TLG), which incorporates tumor lesion size, might perform better and be more informative.

“Ultimately, prospective studies are required to further delineate the role of PET/CT as a prognostic tool in metastatic breast cancer,” she concluded.

Dr. Jhaveri and Dr. Kuske reported having no relevant conflicts of interest.

Score predicts late recurrence in ER-positive breast cancer

A new gene-based biomarker assay may help to estimate prognosis and make decisions about extended endocrine therapy in women with estrogen receptor-positive early breast cancer, new data suggest.

Researchers tested the biomarker, the Breast Cancer Index (BCI), among 249 women with estrogen receptor-positive early breast cancer from the MA.17 trial, in which women received 5 years of adjuvant tamoxifen therapy and, if still disease free, were then randomized to an additional 5 years of either letrozole or placebo.

Study results, reported at the meeting, showed that with each 5-unit increase in the 10-unit BCI score, women’s odds of late recurrence nearly tripled. The index as a whole was not helpful in predicting the benefit of added letrozole therapy in reducing recurrence risk. But one of its components, called H/I, was helpful: Women having a high H/I were about half as likely to have a recurrence if they received letrozole instead of placebo.

In the future, these findings might be used to develop a management algorithm for women with estrogen receptor-positive breast cancer, according to lead investigator Dr. Dennis C. Sgroi, director of breast pathology at the Massachusetts General Hospital in Boston.

“If patients are disease free after 5 years of adjuvant endocrine therapy, we might be able to then test them with the BCI. If this assay identifies these patients to be at low risk, there will be no further therapy for these patients. However, if they are at high risk by BCI, we can then explore the H/I component,” he explained.

“If they have high H/I, that indicates that these patients are likely to benefit from extended adjuvant therapy,” he continued. “For patients who have low H/I, one might consider using extended adjuvant therapy, or these patients might be considered as a future focus for research in clinical trials.

The investigators identified 83 women from the MA.17 trial who had had a recurrence and had primary tumor tissue available. They then matched the women by age, N stage, T stage, and prior receipt of chemotherapy in a 1:2 ratio with 166 women who had not had a recurrence and had primary tumor tissue available. Tissue was analyzed by reverse transcriptase- polymerase chain reaction to determine the BCI.

The BCI has two components, Dr. Sgroi explained. The HOXB13/IL- 17BR (H/I) component is based on expression of two genes regulated by estradiol. The molecular grade index (MGI) component is based on five genes related to pathological grade.

“We have shown in the past that these two biomarkers are complementary, as the combination of the two biomarkers outperforms each individually,” he noted. “In addition, we have shown that the combination biomarker outperforms standard clinicopathological parameters currently used to predict disease recurrence.”

In a multivariate analysis, with each 5-unit increase in the BCI score, women’s odds of recurrence increased 2.91-fold (P = 0.014). The findings were similar when categories were used: Women with high or intermediate scores had 2.21-fold higher odds of recurrence than did their counterparts with low scores (P = 0.019).

The BCI as a whole did not predict the benefit of extended therapy with letrozole, but the H/I component did, Dr. Sgroi reported.

In a multivariate analysis, women who had a high H/I had a 58% reduction in the odds of recurrence if they received letrozole instead of placebo (P = 0.037).

“Most notably, there was a statistically significant interaction between the H/I biomarker and treatment (P = 0.02), indicating that the magnitude of letrozole benefit depends on the expression level of H/I,” he said. “These results suggest that patients with tumors expressing high H/I will benefit from extended endocrine therapy.”

Dr. Sgroi reported receiving research funding from bioTheranostics Inc. The study was supported in part by Novartis, the manufacturer of letrozole.