Interview Richard Pazdur, MD
program, biorepositories, and electronic
medical record systems – this informa-
tion will play a greater role in developing
treatments for hematologic malignan-
cies, as well as selecting the appropriate
patient population for those agents. The
growing evidence about the role genetic
mutations (particularly driver muta-
tions) play in cancer risk and prognosis
has also provided easily identifiable tar-
gets for new product development – for
instance, the approvals of imatinib,
which targets BCR-ABL, and the re-
cent approval of midostaurin, which
targets FLT3.
How do you see drug labels/
indications changing within
hematology as more targeted
approaches become available?
Targeted approaches have certainly
progressed and, as precision medi-
cine moves forward, the targets of
therapy have evolved from a
tissue- or organ-based approach to
a gene- or pathway-based targeting
paradigm.
Product indications and labeling
will have to evolve accordingly. As
the OCE director, I hope to be able
to integrate FDA resources to sup-
port regulators and researchers in
this transition.
For instance, approval decisions
will increasingly describe the use
of complementary and companion
diagnostics – tests that identify
patients whose cancers harbor the
targeted abnormalities – for the safe
and effective use of these drugs.
Companion diagnostics also will
provide information about how
a drug might be used or which
patients should receive a particular
drug. The FDA has typically ap-
proved companion diagnostics
under a “one drug, one diagnostic”
paradigm, but advances in molecular
sequencing mean that we are likely
to shift to a “one drug, a panel of
diagnostic markers” paradigm.
How could the development
and approval of targeted
therapies impact registration
trial designs?
For decades, the clinical development
of cancer treatments has followed
the conventional phase I, phase II,
and phase III paradigm. One of the
OCE’s ongoing initiatives is an effort
to modernize eligibility criteria and
patient selection for oncology clinical
trials. 1 It has become clear to us that,
far too often, the eligibility criteria
for enrolling patients in cancer
clinical trials appear to be simply cut-
and-pasted from one trial to another,
without a scientific or justifiable
38
ASH Clinical News
basis. We need a more thoughtful
approach.
We encourage clinical trial designs
that allow for enrollment of patients
who reflect the real-world popula-
tion likely to receive the therapy under
investigation, including patients with
HIV or a history of prior cancers and
older patients who, traditionally, might
be excluded from clinical trials. 2
Recently, researchers have discovered
early biomarkers of disease prognosis,
which has facilitated better patient
selection for drugs. These discoveries
have prompted a move away from the
sequential, three-phase design to a more
seamless approach – adding cohorts to a
first-in-human trial to investigate doses
and activity in a variety of cancers.
This type of approach proved suc-
cessful with the investigation of the PD-1
inhibitor pembrolizumab, as we described
in a perspective article in the New England
Journal of Medicine. 3 Pembrolizumab was
initially studied in a first-in-human trial
to determine the safety and recommended
dose of pembrolizumab. When investiga-
tors observed early impressive response
rates and durations of response, par-
ticularly among patients with metastatic
melanoma or non-small cell lung cancer,
cohorts were added to assess efficacy in
these two patient populations, as well as to