Oliver Sartor1
1Department of Medicine and Urology, Tulane University, New Orleans, Louisiana 70115, USA
Correspondence: Prof. Oliver Sartor, Fax: +1-504-988-5059 E-mail: osartor@tulane.edu
Harnessing the body's immune system for the treatment of metastatic cancer has been a dream since the late 19th century 1.
Since that time progress has been intermittent and mostly
disappointing. Interferon and Interleukin-2 represented steps forward
but in retrospect the steps were not large and rarely are these agents
used in the clinic today except as the control arm of randomized studies
designed to demonstrate that a new drug is better.
Sipuleucel-T
represents an interesting and important conceptual step forward. Books
have been written about drug development but this one did not follow the
script (in fact, a new book is probably justified). Yes, scientific
advances translated into new therapeutic advance but the twists and
turns along the way contained multiple unanticipated events. Further,
the approval of this agent has led to fundamental questions about our
society and what we value.
If the primary endpoint is
not survival, but there is a survival benefit observed, how much does
that matter? How large should a randomized trial be to justify approval?
Which experts should evaluate new cancer therapies and comment on their
suitability for approval? Who has the responsibility of protecting men
and women who make regulatory decisions that are unpopular? How do we
measure conflicts of interests and exclude those with real or apparent
conflicts from the regulatory process? How much correlative science is
required to create a sense of trust in clinical results? Should response
rates and progression rates correlate with survival? Did the control
group do worse than expected? How much can (or should) our society pay
for incremental progress? Are we now reaching limits in terms of what we
can afford to pay for new therapies? What are the ramifications and
consequences if our society decides that some drugs are too expensive to
provide? Although the FDA's ability to mandate post-marketing trials is
unquestioned, who should be authorized to question the wisdom of their
trial designs? Is this drug development or drama? Actually it is a touch
of both. Some of these questions I will touch on and some I will not.
For now, space constraints require only a handful of these issues be
addressed in relationship to sipuleucel-T.
Let us start with the science and how scientific understanding led to a new medicine. Dranoff and colleagues demonstrated 2
in 1993 that granulocyte-monocyte colony stimulating factor (GM-CSF),
when expressed in tumor cells, can inhibit cancerous growth in certain
model systems. This fundamentally important observation led to
experiments that combined GM-CSF and prostatic acid phosphatase to form a
fusion protein that in turn is used to generate immune responses. This
concept was scaled-up and applied to humans with prostate cancer. Large
clinical trials have now led to a new FDA approval in the USA. This
approval, in March 2010, was the first FDA approval for advanced
prostate cancer since 2004. Sipuleucel-T was only the second agent in
history to be associated with a survival benefit for metastatic
castrate-resistant prostate cancer (mCRPC).
Randomized
trials in patients with mCRPC patients with minimal or no symptoms
demonstrated a 4.1 month survival benefit for sipuleucel-T over
'control' treated patients. This large randomized trial (IMPACT or
D9902B) with 512 patients was recently published in the New England Journal of Medicine 3.
The trial confirmed the survival benefit observed in smaller earlier
trials (D9901) and served as an appropriate basis for the FDA approval.
Is
the survival advantage in the IMPACT trial due to a poorly performing
control group? No, survival in the control arm for IMPACT (21.7 months)
was nearly identical to that previously observed 4 for asymptomatic mCRPC patients treated in the control group of the large atrasentan trial (20.3 months).
In
IMPACT, no response rate or progression-free survival (PFS) advantage
was seen for sipuleucel-T over the control group. Is our current process
of measuring response and progression truly flawed, or do
post-progression alterations in tumor growth kinetics account for the
interesting survival advantages observed? Clearly post-progression
delays in growth are a possible explanation but this remains a
hypothesis rather than fact (at this time). Does every trial with this
class of agents now depend on analysis of overall survival (OS)? If both
responses and PFS are unreliably measured, are there any other
intermediate endpoints that can serve as an index of sipuleucel-T
activity? The conundrum of PFS and survival being de-linked is now
becoming a familiar story in mCRPC. In a large randomized phase III
trial, satraplatin/prednisone had a PFS but not OS advantage over
prednisone alone 5.
Similar findings with regard to PFS and OS were reported for
bevacizumab/docetaxel/prednisone as compared to docetaxel/prednisone 6. Randomized phase II studies with OGX-011 7 and ZD4054 8
have been associated with OS but not PFS improvements. Have we been
deluding ourselves for years when evaluating PFS, or are the new agents
different to the old ones? There is clearly more we need to understand
about markers of progression in mCRPC.
Now that
survival advantages with sipuleucel-T are demonstrated, the FDA has
approved the drug, and patients are clamoring for this novel treatment
that has a bit of panache; what is the next step? Ordinarily the next
step would focus on new indications (such as earlier stage prostate
cancer), or new therapeutic combinations (e.g. ipilumimab plus
sipuleucel-T comes to mind). However this usual discussion has been
subverted by the price of therapy. The sponsor (Dendreon) has decreed
that centers administering the drug will be billed $93
000 for the three infusions required for therapy. Costs to the patient
and their insurance companies will be considerably more after
administration and other costs are calculated. How is such a price
justified? For that matter how is the cost of any new drug in the USA
determined and justified? Do we need to re-think the current process and
focus discussion on why medical care in the USA consumes a greater
percentage of the gross domestic product (GDP) as compared to any other
nation? Is now the time to do so?
The debate on costs
involves many stakeholders and the consequences are large. Investment
in new products is fueled by the expectation of profitable returns and
if poor returns on investment become an expectation, then new progress
against dreaded disease will not be forthcoming. The USA has marvelous
medical technology and breath-taking costs but the lifespan of our
citizens is not longer than our peers. How this debate is resolved will
be key to the future of both medicine and the global economy.
