Nine Members of the Oncology Community Speak Out about
Bevacizumab's Role in Metastatic Breast Cancer
The oncology community was recently stirred with potentially
practice-changing news: the 12-to-1 vote by the FDA's Oncologic
Drugs Advisory Committee (ODAC) to remove the advanced breast
cancer indication for bevacizumab (Avastin). The Committee's vote
does not affect the current availability of the popular drug, but
should the FDA follow ODAC's recommendation-as is its usual
practice-the management of many patients with breast cancer may
change.
Bevacizumab is currently approved in combination with paclitaxel
for the first-line treatment of advanced HER2-negative breast
cancer based on encouraging results of the E2100 trial.1
The FDA provisionally granted accelerated approval but requested
confirmatory data, which Genentech supplied in November 2009 with
the results of the AVADO and RIBBON-1 trials.2,3 On July
20, 2010, ODAC recommended against conversion to full approval.
Recent Studies Fail to Confirm Degree of Improvement in
PFS
The panel concluded that bevacizumab plus commonly used
chemotherapy (taxane or anthracycline-based, or capecitabine
[Xeloda]) increased progression-free survival (PFS) but not overall
survival (OS). They noted that the two recent studies failed to
confirm the magnitude of PFS improvement seen in E2100-5.5 months
(HR = 0.48; P < .0001). PFS was improved in AVADO
by 0.09 months in combination with docetaxel (HR = 0.62;
P = .0003); and in RIBBON-1 by 1.2 months in combination
with taxane/anthracycline (HR = 0.64; P < .0001)
and by 2.9 months in combination with capecitabine (HR = 0.69;
P = .0002). In addition, the committee expressed concern
about toxicity.
Mixed Response
"We are disappointed by the committee's recommendation and
believe [bevacizumab] should continue to be an option for women
with this incurable disease," said Sandra Horning, MD, Senior Vice
President and Global Head, Clinical Development Hematology/Oncology
at Genentech. "We will continue to discuss the data from the more
than 2,400 women who participated in three phase III studies with
the FDA. This recommendation does not impact [bevacizumab's]
approved uses for other cancer types."
In addition, the recommendation sparked mixed response from many
in the oncology community. In
interviews with ODAC voting members, breast cancer specialists,
oncologists in private practice, and a patient advocate (see
below), The ASCO Post captured initial reactions to the
recommendation and its potential consequences. ■
References
1. Miller K, Wang M, Gralow J, et al: Paclitaxel plus
bevacizumab versus paclitaxel for metastatic breast cancer. N Engl
J Med 357:2666-2676, 2007.
2. Miles DW, Chan A, Dirix LY, et al: Phase III study of
bevacizumab plus docetaxel compared with placebo plus docetaxel for
first-line treatment of human epidermal growth factor receptor
2-negative metastatic breast cancer. J Clin Oncol 28:3239-3247,
2010.
3. Robert NJ, Dieras V, Glaspy J, et al : RIBBON-1:
Randomized, double-blind, placebo-controlled, phase III trial of
chemotherapy with or without bevacizumab for first-line treatment
of HER2-negative locally recurrent or metastatic breast cancer. J
Clin Oncol 27(15S; abstract 1005), 2009.
"There has not been a single metric by
which the FDA has reviewed drugs for advanced breast
cancer."
Back in 2005, when we saw the E2100 data, we had great hopes
that we were entering an era when a new class of drugs would be
potent in advanced breast cancer. It's fair to say that over the
past 5 years, after five randomized controlled trials, the
collective experience has suggested this drug is less impressive
than we had hoped. That is a different statement than a regulatory
"yes or no" question, but the recognition is that there may be less
here than we first imagined. And though most breast cancer
specialists believe bevacizumab (Avastin) is reasonably well
tolerated, there are side effects and it is costly. The FDA is not
charged with assigning cost-benefit, but it is an unspoken though
real issue when we are thinking about bevacizumab.
On the other hand, we use lots of drugs that have not been shown
to improve survival, such as ixabepilone (Ixempra) and capecitabine
(Xeloda), which was approved based on response rate, and lapatinib
(Tykerb), which was approved based on an increase in
progression-free survival (PFS). Gemcitabine plus paclitaxel,
however, was approved based on a survival benefit. This
suggests there has not been a single metric by which the FDA has
reviewed drugs for advanced breast cancer.
In the wake of the recommendation from the Oncologic Drugs
Advisory Committee (ODAC), the breast cancer community of
researchers, patient advocates, pharmaceutical sponsors, NCI, and
the FDA should come together to articulate what should be the
accepted clinical trial endpoint, what should be the most important
treatment goal, and what predicts for overall survival (OS) and for
symptom relief or for a better experience for patients. For
bevacizumab, or any other drug, we should shift away from "good or
bad, yes or no." ■
-Harold J. Burstein, MD, Dana-Farber Cancer Institute
and
Associate Professor of Medicine, Harvard Medical School,
Boston
"It would be disappointing, though, to
lose this agent [bevacizumab] from our armamentarium."
The totality of evidence for first-line bevacizumab with
chemotherapy supports an impact on PFS and a hint of a survival
benefit, but the degree of benefit may be chemotherapy drug- and
schedule-dependent. The concern is that if the FDA removes
the current label, we may be throwing the baby out with the
bathwater. That is, bevacizumab may work best when combined with
metronomic or continuous, regular dosing of a chemotherapy
agent-weekly paclitaxel as was used in E2100 (a 5.5-month
prolongation of time to progression), or capecitabine (Xeloda), the
strongest signal in RIBBON-1. It is possible this form of
chemotherapy dosing is also antiangiogenic, so a combination with
bevacizumab would modulate angiogenesis even more effectively.
In contrast to E2100, the AVADO trial did not show a strong
signal for bevacizumab added to every-3-week docetaxel, with a much
shorter prolongation of response. So I choose a weekly taxane
instead when selecting a bevacizumab regimen. And RIBBON-1 raised
the concern that the placebo arm in the anthracycline/taxane
(non-capecitabine) comparison performed slightly better than the
bevacizumab arm, though not statistically significant. However, the
capecitabine results were more encouraging. Therefore, the
focus should be the weekly paclitaxel regimen as in E2100, which
gave an amount of time free of progression that is meaningful to
our patients.
When there is initial enthusiasm regarding benefit of a new
agent, the FDA has taken a pragmatic approach in recent years.
Accelerated approval is granted, with full approval pending rapid
confirmation with similarly robust data in additional trials. This
did not occur with AVADO and RIBBON-1, leading to a unanimous
negative ODAC vote. It would be disappointing, though, to lose this
agent from our armamentarium. An option could be to keep the label
restricted as it is now and rapidly conduct a trial that duplicates
E2100 (and possibly also the capecitabine arm of RIBBON-1), which
could then lead to full approval. At the same time, further
study is needed to demonstrate clinically meaningful efficacy of
bevacizumab in combination with other common chemotherapy agents
and regimens in the front-line setting. ■
-Kathy S. Albain, MD, FACP, Professor of Medicine, and
Director, Breast Research Program, Loyola University Chicago
Stritch School of Medicine
"I
believe we have to set the bar high for cancer drugs, even in the
metastatic setting."
I am as disappointed as anyone else to not see a survival
benefit with bevacizumab, but I am willing to be honest about the
fundamental implications of the data. I believe we have to set the
bar high for cancer drugs, even in the metastatic setting. Patients
expect survival advantages, and we should always try to produce
this. In my view, expensive drugs like bevacizumab that do
not produce a survival benefit should be put aside to allow us to
focus our energies on alternatives that may prove superior.
Otherwise, our health-care dollars will be consumed by drugs that
are not offering what our patients need. ■
-Matthew Ellis, MB, BChir, PhD, Professor of Medicine, Chief
of
Medical Oncology, Washington University School of Medicine, St.
Louis
"If we only accepted cancer treatments
that improve survival, we would severely limit options for our
patients."
I was surprised and disappointed over the advisory panel
recommendation since the three major clinical trials met their
prespecified primary endpoint, PFS. The arguably simplistic view is
that we want every new drug to give a statistically significant
improvement in OS, but the situation is more complicated than this,
which merits careful consideration. If we only accepted cancer
treatments that improve survival, we would severely limit options
for our patients. For instance, in pancreatic cancer no agent
improves survival, including gemcitabine, but this does not mean we
don't offer treatment to these patients. Turning to breast cancer,
there are two regimens that have statistically improved OS, with
data published in peer-reviewed journals and approved by the FDA
(paclitaxel/gemcitabine and capecitabine/docetaxel), but we
actually don't use them very much. There are many treatments that
have received regulatory agency approval for other endpoints
outside of OS. An example is lapatinib's approval with capecitabine
in the refractory HER2 breast setting or in combination with
letrozole as first-line treatment for patients with HER2-positive
and ER-positive advanced breast cancer based on PFS.
I feel that bevacizumab is perhaps being held to a higher
standard than many other agents in the setting of advanced cancers.
Many of the treatments currently used and reimbursed are lacking
formal phase III comparisons showing that they improve OS, or even
PFS or response rate. The concept of clinical benefit to the
individual patient or in the context of public health is sometimes
not easily quantifiable, although we use response, PFS, and even OS
as surrogates for that endpoint. Based on the data that have been
available for many months, the three first-line trials of
bevacizumab in breast cancer met the prespecified endpoints of
improving median PFS. I am concerned that if registration studies
are designed with a prespecified endpoint, the endpoint is met, and
then a committee recommends to regulatory agencies not to formally
approve the agent for physicians to consider as an option for their
patients, it would put into question how trials are designed. This
scenario could be a disturbing lesson for patients who volunteer
for studies, for nonprofit funding groups, and for pharmaceutical
companies who might invest years and millions of dollars to develop
a drug in an area of unmet need, satisfy the endpoint required for
registration, and then have the drug rejected by the FDA. I hope
the advisory panel recommendation will not be a setback for drug
discovery.
In addition, I respect the job that ODAC must perform, but I
have to say that I was and remain concerned about the recent vote
related to bevacizumab. I hope when the FDA meets September 17,
they will vote at least to allow bevacizumab in combination with
paclitaxel, and convert this regimen to full approval. In this
case, their decision will not appear arbitrary but will be
evidence-based. The FDA would come out a winner, and we would still
have this regimen as an option for our patients.
As a disclosure, I have received research grants from Genentech
and have served in the capacity of uncompensated advisor. ■
-Edith A. Perez, MD, Director Breast Cancer Program,
Serene M. and Frances C. Durling Professor of Medicine, Mayo
Clinic, Jacksonville
"If a drug is expensive, relative to its
benefits, then we need to develop a dialogue to address this
issue."
I found it ironic that a representative of Genentech was in my
office promoting bevacizumab for breast cancer the same day the FDA
committee voted against it. It's important for us as physicians to
try to resolve these different spins on information.
Right now, at Wilshire Oncology we use bevacizumab plus a taxane
as front-line treatment for metastatic disease in about 50% of
patients. What I've been hoping for is the development of
biomarkers of response to this drug, which would not only help us
select appropriate patients for treatment but would also provide a
subset of patients likely to demonstrate positive results in
clinical trials of bevacizumab. Any future bevacizumab research
should also focus on biomarkers.
At ASCO this year, we heard encouraging data in other tumor
sites. The addition of bevacizumab to standard regimens extended
PFS by 3.8 months in ovarian cancer, and in gastric cancer the
geographic subset of Pan-American patients experienced a 4.7-month
gain in overall survival (though the larger population including
European patients did not benefit). In the context of the
recommendation in breast cancer, will the FDA look favorably on
these findings? I would hope so, but this raises concern about the
use of bevacizumab in other tumors. We are looking for leadership
at the FDA to help sort out these differences.
As a community physician, I want access to bevacizumab and any
other drug that provides meaningful benefit. If it is expensive,
relative to its benefits, then we need to develop a dialogue to
address this issue. Perhaps manufacturers might be reimbursed
proportionally to the benefit the drug provides, ie, have different
payment schedules for bevacizumab in colon, breast, and lung
cancers, where the magnitude of benefits varies. It's a new
approach, but without such innovations we may wind up with evidence
of varying benefits and lack of clinical access to the drugs with
lesser benefits, and our ability to control disease will remain
more limited. ■
-Cary Presant, MD, FACP, Director of Medical Oncology,
Wilshire Oncology Medical Group, Los Angeles, and Past
President,
Association of Community Cancer Centers
"Bevacizumab may be a good drug for a
subset, but it should not be on the market for all HER2-negative
patients until this is known."
Since the ODAC vote, I've been monitoring the BCMets.org mailing
list of over 1,000 women living with metastatic breast cancer.
Although you would think their immediate reaction might be
negative, instead it is decidedly mixed, based on patients'
personal experiences with the drug's activity, toxicities-and, of
course, cost. There has been ambivalence about bevacizumab from the
beginning. In my experience, patients do not necessarily
argue just for more options, but for more beneficial options-drugs
that are better than what we have now or at least provide a unique
benefit.
Many advocates believe OS should be the sole standard for drug
approval, although OS is hard to prove due to crossovers and
subsequent treatments, especially in the first-line setting. I do
agree with ODAC that PFS is only a clinically meaningful endpoint
when it represents a significant period of time, and when there is
solid evidence that quality of life is not compromised. Some
patients do well on bevacizumab, and others have a host of minor
and major problems. Bevacizumab has significant toxicities, so it
had better provide real benefit, and there is no good evidence of
that.
Some women do appear to have strong responses, maintained even
when the chemotherapy agent is discontinued. Preliminary findings
from post hoc analysis of E2100 suggest that tumors with different
VEGF mutations respond differently. We hope that Genentech will
take the ODAC recommendation as an incentive to start looking at
this issue more urgently. Bevacizumab may turn out to be a good
drug for a subset of patients, but it should not be on the market
for all those with HER2-negative disease until this is known. And I
have real concerns about subjecting patients with early breast
cancer in adjuvant trials to a potentially toxic drug with minimal
benefit.
ODAC's recommendation does show that the accelerated approval
process works. Patient advocates were concerned that accelerated
approval of bevacizumab would lead to an automatic full approval,
despite problems with the drug. I was gratified to see that when
confirmatory studies do not show meaningful clinical benefit,
despite early promise, a drug can be removed from the market. ■
-Musa Mayer, Breast Cancer Survivor, Patient Advocate,
Author, Creator of AdvancedBC.org
"My fear is that pharmaceutical
companies will be discouraged from developing anticancer agents if
they feel the bar they must jump over is simply too high."
Personally, I was not expecting ODAC to recommend full approval
of bevacizumab, but I was surprised that it recommended withdrawal
of the indication. A number of drugs have been approved for breast
cancer in recent years on the basis of modest benefit. "Modest
benefit" does not mean that some patients don't derive a more
substantial benefit. Increasingly, however, many have wanted
new drugs to show more benefit than older drugs, especially when
treatment costs are high. The FDA is prohibited from considering
cost, but society looks through a different lens-and this lens
includes cost.
My fear is that pharmaceutical companies will be discouraged
from developing anticancer agents if they feel the bar they must
jump over is simply too high. On the other hand, we don't want to
be in a situation where all that is being developed are drugs that
add minimally to what we have. In truth, while PFS is probably a
meaningful endpoint for some patients, ultimately our goal is to
help patients live longer and better. If PFS doesn't help them live
longer, then the question becomes whether it helps them live
better. I suspect that bevacizumab improves quality of life for
some patients, but certainly not for all. Unfortunately, measuring
quality of life is notoriously difficult.
Although I believe there are subpopulations that get substantial
benefit and do particularly well with bevacizumab, I don't know how
to select those patients. Perhaps we will sort this question out in
the adjuvant trials. It does seem that the FDA is sending a message
to the oncology community, a message that drug approval in the
future may look different than it has in the past. We will have to
think carefully about drug development, and in this era of
increasingly targeted therapies, it is our responsibility to
identify populations that will benefit the most from new
agents. ■
-Eric P. Winer, MD, Director of the Breast Oncology
Center
at Dana-Farber Cancer Institute and Professor of
Medicine,
Harvard Medical School, Boston
"I think it unfortunate that insurance
companies (including Medicare) have taken FDA approval as an
indication of good medical practice."
The details upon which ODAC based its decision are obvious:
First and foremost, bevacizumab does not prolong OS. But clearly
there is biologic activity, perhaps a positive interaction with
weekly paclitaxel. I wish the system would let us give drugs the
way we think they should be given, which in my practice tends to be
with weekly paclitaxel. Maybe because of how I give bevacizumab, I
usually get results like those in E2100. The first patient I
treated had a 2-year remission and excellent quality of life.
Another had liver metastases disappear on MRI and has had a 7-month
remission. Another patient is a 40-year-old with terrible organ
disease, including liver and CNS metastases. I started her on the
drug in March 2009 and she is still doing great. Now she asks me,
'What will happen to me if the FDA takes [bevacizumab] away?' While
I could prescribe bevacizumab off-label, the price of the drug is
unconscionable.
I think it unfortunate that insurance companies (including
Medicare) have taken FDA approval as an indication of good medical
practice, especially since the FDA tends to weigh one thing above
the other in fairly arbitrary bureaucratic ways. The FDA review is
a legally mandated adversary proceeding in which the drug company
seeks approval and the FDA seeks evidence that the medication is
both safe and effective. The FDA is not charged to, nor is it
equipped to, set standards of medical care. Its job is to control
marketing claims and protect consumers against dangerous drugs.
Input from both patients and payers should be heard in open
proceedings independent of the FDA review, which is, of necessity,
very detailed and technical. Such sessions, in which patient
benefit is a major concern, are needed to help arrive at a fair
evaluation of medical practice and what should be covered. ■
-Steven E. Vogl, MD, Private Practice, Bronx, NY
"The crux of the issue is whether
overall survival should be the primary endpoint…though a survival
benefit is increasingly hard to show."
E2100 demonstrated several months' increase in PFS with the
addition of bevacizumab, while PFS benefit was smaller in
subsequent trials and an OS benefit was not shown. In some ways I
can understand, therefore, where the FDA advisory committee was
coming from in its recommendation. However, the primary endpoint of
the studies was PFS, and in all three studies, PFS was
statistically significantly improved. From this standpoint,
bevacizumab should be approved, if this was the agreement with the
FDA.
There is the more global picture, however, that we want our
patients to actually survive longer. These studies bring up a major
point; in fact, it is the crux of the issue, but it was not fully
addressed at the hearing: whether OS should be the primary endpoint
in metastatic breast cancer. Unfortunately, it is increasingly hard
to show a survival benefit in a population that is living a median
of at least 2 years now. One of the few positive trials in this
regard was with trastuzumab (Herceptin), which is an extremely
effective drug that has a definite target. If there is a survival
benefit with bevacizumab, obviously it is not that large, at least
in an unselected population.
We need better drugs, we need targeted drugs, and we need to
understand the subset for which they are effective. There are
definitely many patients who can benefit from bevacizumab-that has
been shown in the data-but we have to learn how to identify them. I
can see no reason to stop the adjuvant trials with bevacizumab. The
agent clearly has activity in breast cancer, as shown by the
increased response rate and PFS benefit. ■
-Sandra M. Swain, MD, Medical Director,
Washington Cancer Institute, Washington, DC
Compiled and reported by Caroline Helwick.
Editor's Note: FDA indicated it is the Agency's policy not to
comment on ODAC's recommendations. The Agency will make a decision
about whether to revoke the breast cancer indication later in the
year. See
here for perspectives from ODAC voting members Wyndham Wilson,
MD, and Joanne Mortimer, MD. Watch for further coverage in The ASCO
Post.
Disclosure of Potential Conflicts of
Interest
Harold J. Burstein, MD: Dr. Burstein has no
financial interest or other relationship with the manufacturers of
any products discussed in this report.
Kathy S. Albain, MD, FACP: Dr. Albain has
disclosed: Attending a one-time advisory board meeting for
Genentech (compensated) and a one-time advisory board meeting for
Roche (compensated), pertaining to their pipeline but not dealing
with bevacizumab.
Matthew Ellis, MB, BCHir, PhD: Dr. Ellis has
disclosed: Consultant, Genentech (<$10,000); Consultant, Roche
(<$10,000)
Edith A. Perez, MD: Dr. Perez has disclosed:
Research grants, Genentech; Advisor, Genentech
(uncompensated).
Cary Presant, MD, FACP: Dr. Presant has
disclosed: Research funding (Genentech, Roche, DiaTech Oncology);
Consultant/advisor (Genentech).
Musa Mayer: Ms. Mayer has disclosed: Consultant
for programs supported by Genentech (uncompensated).
Eric P. Winer, MD: Dr. Winer has disclosed:
Principal investigator for a Genentech-sponsored trial at
Dana-Farber Cancer Institute.
Steven E. Vogl, MD: Dr. Vogl has no financial
interest or other relationship with the manufacturers of any
products discussed in this report.
Sandra M. Swain, MD: Dr. Swain has disclosed:
Research funding, Genentech; Advisory Board, Genentech/Roche
(uncompensated).
