January 25, 2012 — Just months after the US Food and Drug Administration (FDA) revoked the metastatic breast cancer indication for bevacizumab (
Avastin, Roche/Genentech), new data show a benefit from this drug in nonmetastatic disease in a neoadjuvant setting.
Two large studies, published in the January 25 issue of the
New England Journal of Medicine, demonstrate that the addition of bevacizumab to docetaxel-based chemotherapy regimens significantly improves the rate of pathological complete response before surgery in women with nonmetastatic HER2-negative breast cancer.
In the
National Surgical Adjuvant Breast and Bowel Project (NSABP) B-40 trial, conducted in the United States, the neoadjuvant addition of bevacizumab to a variety of chemotherapy regimens significantly increased the rate of pathological complete response, compared with no bevacizumab (34.5% vs 28.2%;
P = .02); patients received the targeted therapy for the first 6 cycles of chemotherapy.
In the
GeparQuinto (GBG44) trial, conducted in Germany, the rates of pathological complete response were also better with than without bevacizumab in patients undergoing 2 different chemotherapy regimens (18.4% vs 14.9%; odds ratio with bevacizumab, 1.29; 95% confidence interval, 1.02 to 1.65;
P = .04); all patients received 8 cycles of bevacizumab.
The primary end point in both studies was pathological complete response. However, the GBG44 trial used a more stringent definition and had more patients with more advanced disease, which resulted in fewer patients achieving the end point.
trials do not resolve existing controversies.
These results stir up old concerns, suggests an
editorial accompanying the study. "These 2 well-performed trials do not resolve existing controversies surrounding bevacizumab therapy," write Alberto J. Montero, MD, and Charles Vogel, MD, from the division of hematology/oncology at the University of Miami Sylvester Comprehensive Cancer Center in Florida.
In brief, those controversies are about the "ever-increasing cost" of new cancer treatments and the predictive value of surrogate markers. What remains unclear is whether bevacizumab will ever ultimately be shown to improve survival in breast cancer.
In the metastatic setting, the improvement in progression-free survival with the use of bevacizumab never panned out in terms of overall survival, which precipitated the recent FDA decision.
Now, in the neoadjuvant setting, the surrogate marker in question is an even earlier measure than progression-free survival — pathological complete response.
But the German investigators are quite hopeful that the results from their 1948-patient study will eventually translate into a survival benefit.
"Given that pathological complete response has been shown to be highly correlated with outcome, particularly in patients with triple-negative disease, we speculate that the beneficial effect will be sustained," write the authors, led by Gunter von Minckwitz, MD, from the German Breast Group in Neu-Isenburg. They were referring to their finding that the rates of pathological complete response were 27.9% without bevacizumab and 39.3% with bevacizumab in a subset of 663 patients with triple-negative tumors (
P = .003).
The Americans are more sanguine about any would-be survival benefit in their 1206-patient trial.
"The effect of adding bevacizumab in the NSABP B-40 trial was less dramatic than was the effect of adding docetaxel in the NSABP B-27 trial, so it is not clear whether the neoadjuvant effect of bevacizumab would translate into a substantial benefit to patients," write the authors, led by Harry Bear, MD, from Virginia Commonwealth University in Richmond. Notably, unlike the GBG44 trial, the NSABP B-40 trial did not find that patients with triple-negative disease received a significant benefit from the addition of bevacizumab; however, there was a trend, Dr. Bear and colleagues report.
Even if a survival benefit is eventually found in these trials, Drs. Montero and Vogel express concern in their editorial that the whole enterprise of targeted therapies for cancer is too costly.
"In the context of unsustainable expenditures for cancer care in the United States, any survival benefit of bevacizumab, or other molecularly targeted drugs, will be balanced against the considerable development costs of modern molecularly targeted oncology drugs," they write.
The addition of bevacizumab also increased the rates of a variety of adverse events in both trials. In the NSABP B-40 trial, those events included hypertension, left ventricular systolic dysfunction, the hand–foot syndrome, and mucositis.
Definition of Response Is Critical
It is important to examine and understand the different end points used in the 2 trials, explain Drs. Montero and Vogel in their editorial.
In the German GBG44 trial, pathological complete response was defined as the absence of residual tumor in the breast and nodes. In the American NSABP B-40 trial, the less stringent definition of the absence of residual tumor in the breast was used.
In the GBG44 study, the overall rate of pathological complete response with the addition of bevacizumab was 3.5 percentage points higher than the rate without bevacizumab.
In the NSABP B-40 study, the rate with bevacizumab was 6.3 percentage points higher than the rate without bevacizumab. But, if the GBG44 definition of pathological complete response is applied to the NSABP B-40 results, the affect largely disappears, note the editorialists.
"When the more stringent definition of pathological complete response was used, the differences noted in the NSABP B-40 were no longer significant," they explain.
Study Details
These 2 trials are among a number of new trials examining the use of bevacizumab in combination with chemotherapy in the neoadjuvant setting that were hatched by investigators after the FDA granted bevacizumab accelerated approval in 2008 for the first-line treatment of HER2-negative metastatic breast cancer.
In NSABP B-40, patients were randomized to receive neoadjuvant therapy consisting of docetaxel, docetaxel plus capecitabine, or docetaxel plus gemcitabine for 4 cycles, with all regimens followed by treatment with doxorubicin–cyclophosphamide for 4 cycles.
Patients were also randomly assigned to receive or not to receive bevacizumab for the first 6 cycles of chemotherapy.
Patients in NSABP B-40 were required to have a primary tumor 20 mm or larger and tumor stage T1c to T3, nodal stage N0 to N2a, and no distant metastases.
In GBG44, patients were randomized to receive neoadjuvant epirubicin and cyclophosphamide followed by docetaxel, with or without concomitant bevacizumab. Patients with untreated HER2-negative breast cancer were eligible if they had large tumors (median, 40 mm), hormone-receptor-negative disease, or hormone-receptor-positive disease with palpable nodes or positive findings on sentinel-node biopsy, and no increased cardiovascular or bleeding risk.
NSABP B-40 was funded by the National Cancer Institute, F. Hoffmann-La Roche, Genentech USA, and Eli Lilly. The NSABP B-40 authors have disclosed no relevant financial relationships. GBG44 was funded by sanofi-aventis and Roche, Germany. Some of the GBG44 authors report financial relationships with sanofi-aventis, Roche, and other companies.
N Engl J Med. 2012; 366:299-309, 310-320, 374-375.
GBG44 Abstract,
NSABP B-40 Abstract,
Editorial 
