The design of a bone morphogenetic protein (BMP) clinical study
directly depends on the relationship of the proposed BMP treatment
to the standard of care. One must consider whether the intended
use of BMP will (a) replace or (b) augment the existing surgical
standard of care.
In (a), meaningful clinical studies can be designed to test whether
the proposed BMP treatment is equivalent or superior to the standard
of care. For example, recombinant human (rh) BMP-2 is being tested
as a complete replacement for the use of autograft in several spine
fusion procedures. These clinical studies compare the outcomes of patients
who have been randomized to receive rhBMP-2 or autograft. All patients
receive the same spine instrumentation. The control group (autograft)
was chosen because it represents the current standard of care. The most
conservative design for studies of this sort is one that will definitively
establish the statistical equivalence of the two treatments. The
size of each group should be sufficiently large to allow a conclusion
that the outcome of the two treatments differs by no more than a small
amount (e.g., ±10%) and should limit
the probability of a type II error (not detecting a difference that
actually exists) to less than 20%.
In (b), the only meaningful design for a clinical study is one that
tests whether the proposed BMP treatment is superior to the standard
of care. For example, rhBMP-2 is being tested as an adjunct to standard
surgical care in the acute treatment of open tibial fractures that
require surgical management. In a study of this sort, every patient
receives the standard of care (in this case, internal fixation with
an intramedullary nail), and patients are randomized to receive
either no additional treatment or rhBMP-2 treatment. The control group
does not offer an alternative to BMP, because there is no alternative
adjuvant treatment currently available.
The most conservative study design for studies of this sort is one
that will detect relatively small differences between the control
group and the BMP treatment group. However, statistically significant
differences are not sufficient for these studies to be meaningful.
The differences detected in these types of studies must also be
clinically meaningful and represent a true clinical benefit for
the patient.
In summary, the intended use of BMP treatment (and its relationship
to the existing surgical standard of care) directly affects the design
of clinical studies that evaluate that particular intended use.
Most importantly, this relationship determines the choice of the
appropriate control group, an essential requirement for any clinical
study to yield definitive safety and efficacy results.
Participation in Industry-Sponsored BMP Clinical
Trials
BMP clinical studies provide an excellent opportunity for collaboration
between academic institutions and companies. The financial resources
of companies make it possible to sponsor large, well controlled
clinical studies that may be beyond the financial resources of individual academic
institutions. When properly designed and executed, such studies
not only rigorously assess the safety and efficacy of a proposed
BMP treatment but also provide important information about the practice
of orthopaedic surgery. Surgeons interested in participating in
such studies should contact the companies developing AMP therapies
and carefully consider the following issues.
(a) Do I have the interest in participating and how does my standard
surgical management compare with that of the clinical study protocol?
Every clinical study requires compromises. To ensure comparability,
multicenter clinical studies require that all participating surgeons follow
the same surgical management principles. While allowing some flexibility,
this requirement usually results in some modification of surgeons’ standard
management practices for patients that will be enrolled in the study.
It is essential that surgeons are professionally comfortable and
technically proficient with the modifications that may be required
for the clinical study to have validity at their institutions. These
modifications are also reviewed and approved by surgeons’ Institutional Review
Boards, as part of the overall review of the BMP clinical trial.
(b) Are there adequate resources within my institution to enable me
to conduct the study? The conduct of a BMP clinical study involves
considerable preliminary activities, such as shepherding the protocol
through one’s 11(B), extensive record keeping, laboratory
tests (e.g., serum sampling) that are not usually administered,
follow-up visits that may be more frequent than typically performed
by one’s institution, etc. The coordination, and much of
the execution, of these activities is best left to full-time dedicated
research coordinators within an institution. Without such staff,
the obstacles to successfully participating in a study are substantial.
(c) Does my practice include a reasonable number of patients
in the population proposed for the study? It is especially frustrating for
a surgeon to invest considerable time and energy to start a study
and then find that patient enrollment is far slower than anticipated.
Thus, it is important that a surgeon objectively assess whether
his or her practice includes the patient population that will be
studied in the clinical trial. An up-to-date database of the patients
treated in the surgeon’s institution for the past 2 years
is a very effective assessment tool and one that we strongly recommend
for surgeons interested in performing any clinical study.
Moreover, the surgeon must be comfortable presenting and defending
all alternatives of a randomized clinical trial. If the surgeon
and his or her patients choose to participate in a study primarily
on the basis of receiving the new treatment, a patient’s randomized
assignment to the control group may result in that patient’s
refusal to participate or in an early withdrawal of the patient.
This in turn may significantly limit the surgeon’s ability to
enroll and participate in the trial.
(d) Do my practice and institutional responsibilities allow me to
commit the time necessary to successfully conduct and complete the
proposed trial? Every clinical study requires a relatively long-term
commitment from each participating surgeon. Preliminary activities
may take 6 months or more, patient enrollment may take several months
to several years, and required patient follow-up may continue for
several years following surgery, including periods after a patient
is discharged from a surgeon’s care. Before agreeing to participate
in a clinical trial, surgeons should carefully consider the duration
of the commitment that is required and their abilities to meet these
considerable demands on their time.