Here's a C&P of the argument FOR approval for Avastin:
Another
breast cancer expert approached by Medscape Medical News, Gabriel
Hortobagyi, MD, FACP, professor of breast medical oncology at the
University of Texas M.D. Anderson Cancer Center, who was present at the
ODAC meeting, said: "This is an interesting development."
"Bevacizumab
is clearly active in breast cancer, and there is more experience with
this agent in metastatic breast cancer than with many, if not most,
other agents evaluated for the metastatic breast cancer indication," he
said. In 3 separate randomized trials, adding bevacizumab to various
commonly used chemotherapy regimens resulted in significantly improved
overall response rate and time to progression. These beneficial effects
were consistently found in all 3 trials and in all subgroups examined,
he pointed out.
"Anecdotally, bevacizumab combinations can produce
dramatic responses in individual patients, and it is my clinical
impression that responses of such quality are seldom seen with the same
chemotherapy programs without bevacizumab. As a frequent user of this
agent, I have also found that it is very well tolerated by the great
majority of patients; in very few do I need to modify dose or schedule,
and in even fewer do I need to discontinue treatment because of
toxicities."
However, a "puzzling observation" in all 3 studies is
the total absence of a survival benefit - "the survival curves of the
control and bevacizumab arms were entirely superimposable," Dr.
Hortobagyi said. It is difficult to demonstrate a significant survival
benefit in first-line therapy of metastatic breast cancer, because such
benefit is often diluted by multiple subsequent treatments and crossover
designs, he explained, but "we often see some separation of the curves,
even if the difference doesn't reach statistical significance."
"In
the trials in question, and as you would expect simply by chance, some
parts of the bevacizumab curves were slightly above and others slightly
below the control-arm curves. While none of these differences (whether
positive or negative) were statistically significant, the FDA reviewer
elected to present them in an incredibly biased manner, where positive
(but nonsignificant) survival differences were labeled 'no difference,'
and negative (also nonsignificant) differences were labeled as favoring
the placebo arm," Dr. Hortobagyi explained.
"This either denotes
ignorance of common statistical methods or an underlying agenda to place
the agent in an unfavorable perspective. There were other surprising
statements made by the FDA reviewer attributing certain toxicities to
bevacizumab, when in fact those toxicities are well recognized as being
associated with the underlying chemotherapy. For instance, bowel
perforations are a well-recognized complication of taxanes, especially
docetaxel; angina and myocardial infarction are reported complications
of capecitabine; and congestive heart failure is a well-known
complication of anthracyclines (doxorubicin and epirubicin)," he
continued.
"The FDA reviewer continuously referred to these
complications as bevacizumab-related, creating an artificially high
bevacizumab-related mortality rate in the minds of ODAC members," Dr.
Hortobagyi said.
He summarized the outcome of the meeting as
follows: "ODAC members found the benefit of bevacizumab insufficient in
view of the toxicities listed by both the sponsor and the FDA reviewer,
and based on that belief, they felt that the 2 new trials failed to
confirm the encouraging results of E2100, and that there was
insufficient efficacy to expand or even maintain the metastatic breast
cancer indication of bevacizumab."
However, he added that "during
the discussion, I got the strong impression that several ODAC members
had a poor understanding of clinical trials and statistics (based on
some of the questions that were somewhat bizarre!), and that very few
had any experience using bevacizumab in the clinic. Otherwise, they
would have known that the frequent toxicities reported on paper caused
few symptoms and actually did not affect quality of life, and perhaps
they would have seen the response and progression-free survival benefits
in a different light."
"I thought that the discussion brought to
light some issues in the process. The sponsor complied entirely with the
FDA's requests when accelerated approval was granted; however, the FDA
instructions were vague, and the 'magnitude of benefit' was never
defined. Despite the fact that the FDA never requested prolongation of
survival, much of the discussion by ODAC members was based on this lack
of survival benefit. Therefore, there was a major disconnect between the
FDA and ODAC," Dr. Hortobagyi explained.
"Ultimately, ODAC
members acted on their understanding of the risk/benefit ratio of the
drug, and recommended removal of the metastatic breast cancer
indication. I think this was the wrong recommendation, although
witnessing the process I can see how inadequate instruction by the FDA,
unfamiliarity with the drug, and a biased presentation by the FDA
reviewer led to this result," he said.
"I am concerned, in part,
because this recommendation might remove an important treatment option
from patients with metastatic breast cancer. I am also concerned because
'clinical benefit' remains poorly defined. This precedent will set many
drugs currently under development up for failure. It would be important
for the FDA to clearly define expectations from sponsors and provide
clear guidance to ODAC members, so that disconnects such as those
observed during this discussion stop happening," Dr. Hortobagyi told
Medscape Medical News.
Dr. Perez reports serving on steering
committees for Bayer HealthCare and Genentech, and on an independent
monitoring committee for Novartis. Dr. Hortobagyi was at the meeting as a
consultant to Genentech, for which he will receive an honorarium, and
also had travel and accommodation paid for by the company; he has
disclosed no other relevant financial relationships.