Clinical Evaluation of BMPs and Bone Marrow Stromal Cells in Orthopedic Surgery - Engineered Bone: Tissue Engineering Intelligence Unit - page 196

process and have shown overwhelming results in pre-clinical models. The BMPs have demon-
strated their ability to promote bone healing in large defects in rabbits, sheep, and dogs, 8,9 and
in spinal fusion models in dogs and monkeys, 10,11 and to accelerate bone healing in sheep. 12
Despite extensive evaluation of BMPs in preclinical models, confirming their substantial effi-
cacy in the treatment of bone defects in animals, some important issues, regarding efficacy in
human and safety, needed to be addressed before these proteins could be considered for routine
medical use. It took almost 15 years before the Food and Drug Administration (FDA) in the
US and the European Agency for the Evaluation of Medicinal Products (EAEMP) approved
rh-BMP2 and OP-1 (rh-BMP7) for spinal fusions and the treatment of nonunion of long
bones.
An alternative strategy to the use of BMPs is the combination of osteocompetent cells such
as bone marrow stromal cells (BMSCs) and different scaffolding materials. 13 These osteogenic
materials are currently under evaluation for the treatment of large bone defects in human in
several countries.
The purpose of this study was to review the clinical data available on the use of BMPs and
BMSCs, and to discuss the critical concerns that remain.
Clinical Experience with BMPs
Bone Morphogenetic Proteins can be used in clinical practice to accelerate the natural pro-
cess of bone healing, to reduce the incidence of nonunions (higher in the case of comminuted
open fractures), or to replace the current standard of care, namely bone autografting, for the
treatment of several diseases: nonunion of long bones, bone defects, spinal disorders requiring
spinal fusion. The treatment of these disorders is usually difficult and time-consuming and the
use of BMPs is expected to reduce significantly both the failure rate and the morbidity of bone
grafting.
Segmental Bone Loss and Nonunions
Most of the clinical studies reporting on the use of BMPs have focused on the treatment of
nonunions and segmental bone defects. Johnson and Urist described their experience with the
use of h-BMP, extracted from human bone, on five occasions. 14-18 In their latest report, 30
patients were treated with a composite made of an allograft, and a high concentration of h-BMP.
When the femoral defect was greater than 2 cm, it was supplemented with bone autograft.
Healing was achieved in 24 patients at an average of 6 months after BMP implantation. Al-
though effective, the use of allogeneic h-BMP does not appear a suitable option for medical
purposes: hundreds of kilograms of bone would be needed each time to obtain few milligrams
of h-BMP (40 kg for 1 mg); and the implantation of an allogeneic product would involve
potential risks of disease transmission and immunological reactions from host. The fabrication
of recombinant forms of naturally occurring BMPs (BMP 2 and 7) in the '90s has hastened the
clinical research in the field of bone delayed unions and nonunions. The application of OP-1
to heal clinical bone defects was first reported by Geesink in 1999 in a prospective randomized
double-blind study. 19 This study evaluated the use of OP-1 (3.5 mg) delivered on bovine type
1 collagen implants to treat 15mm-length fibular defects performed during an opening wedge
high tibial osteotomy. Control groups were filled with the matrix alone. Five of the six patients
treated with OP-1 healed compared to none of the control group. Consequently, OP-1 was
compared to autologous iliac crest bone grafting in the treatment of tibial nonunions. This
multicenter prospective randomized trial started in 1992 and involved 17 medical centers. 20
Under a US FDA approved Investigational Device Exemption, the study enrolled 122 patients
(124 tibial nonunions) with an established nonunion at least 9 months after injury. Additional
inclusion criteria included the need for an internal fixation and bone autografting (in the judge-
ment of their surgeon), and no surgery performed within 3 months prior to implantation. Half
of the patients were treated with OP-1 (3.5 mg/bovine type 1 collagen) and half of them
with autologous iliac crest bone grafting. A new rod was inserted in 90% of the cases. The