AUGMENT BONE GRAFT

P100006 · Biomimetic Therapeutics, LLC · NOX · Sep 1, 2015 · Orthopedic

Device Facts

• Record ID: P100006
• Device Name: AUGMENT BONE GRAFT
• Applicant: Biomimetic Therapeutics, LLC
• Product Code: NOX · Orthopedic
• Decision Date: Sep 1, 2015
• Decision: APPR
• Device Class: Class 3
• Attributes: Therapeutic

Indications for Use

Augment® Bone Graft is indicated for use as an alternative to autograft in arthrodesis (i.e., surgical fusion procedures) of the ankle (tibiotalar joint) and/or hindfoot (including subtalar, talonavicular, and calcaneocuboid joints, alone or in combination), due to osteoarthritis, post-traumatic arthritis, rheumatoid arthritis, psoriatic arthritis, avascular necrosis, joint instability, joint deformity, congenital defect, or joint arthropathy in patients with preoperative or intraoperative evidence indicating the need for supplemental graft material.

Device Story

Augment® Bone Graft is a combination device/drug consisting of recombinant human platelet-derived growth factor B homodimer (rhPDGF-BB) and a bioresorbable beta-tricalcium phosphate (β-TCP) matrix. The device is used in ankle and hindfoot arthrodesis procedures as an alternative to autograft, eliminating the need for a second surgery to harvest bone. The surgeon mixes the rhPDGF-BB solution with the β-TCP granules immediately prior to implantation. The β-TCP acts as an osteoconductive scaffold for bone growth and a void filler, while rhPDGF-BB acts as a chemo-attractant and mitogen for cells involved in wound healing and bone formation. The device is implanted by orthopedic surgeons in an OR setting. By avoiding autograft harvest, the device reduces donor site morbidity, such as pain and infection. Clinical success is evaluated by the surgeon through patient pain scores (VAS), functional indices (FFI, AOFAS), and radiographic assessment of fusion.

Clinical Evidence

Pivotal prospective, randomized, controlled, non-inferiority trial (IDE #G050118) of 434 subjects (414 treated) comparing Augment® Bone Graft to autograft in ankle/hindfoot arthrodesis. Primary radiographic endpoint (CT-based bridging bone at 24 weeks) was inconclusive due to high attenuation of β-TCP. Effectiveness was established via clinical/functional endpoints (VAS weight-bearing pain, FFI, AOFAS) at 24 and 52 weeks, showing equivalence to autograft. Safety profile was similar to autograft; no significant differences in TEAEs, SAEs, or infection rates. 14.5% of Augment patients developed transient anti-rhPDGF-BB antibodies; 2.1% had transient neutralizing activity with no clinical correlation.

Technological Characteristics

Combination product: rhPDGF-BB (0.3 mg/mL in sodium acetate buffer) and synthetic β-TCP granules (1-2 mm particle size). β-TCP is a bioresorbable osteoconductive scaffold. Components are packaged separately: β-TCP in PETG/PE laminate cup (gamma sterilized); rhPDGF-BB in Type I borosilicate glass vials (EO sterilized). Mixed intraoperatively. No software or electronic components.

Indications for Use

Indicated for patients ≥18 years old requiring ankle or hindfoot arthrodesis due to arthritis, avascular necrosis, joint instability, deformity, or congenital defect, where supplemental graft is required. Contraindicated in patients with hypersensitivity to product components or yeast-derived products, active cancer, skeletal immaturity, pregnancy, active infection at the operative site, metabolic bone disorders (excluding osteoporosis/diabetes), or where soft tissue coverage is not achievable.

Regulatory Classification

Identification

The device acts as a bone void filler and consists of a filling material and a synthetic peptide. It is intended to fill traumatic or surgically-created defects that are not intrinsic to the bony structure of the extremities, spine or pelvis. It will be packed or placed into the defect and will resorb over time to be replaced by bone.

Reference Devices

• GEM 21S® (P040013)
• Regranex® gel

Submission Summary (Full Text)

{0}

PROPOSED SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)

I. GENERAL INFORMATION

Device Generic Name: Bone grafting material containing a therapeutic biologic

Device Trade Name: Augment® Bone Graft

Applicant's Name and Address: Wright Medical Technology, Inc.
BioMimetic Therapeutics, LLC
389 Nichol Mill Lane
Franklin, Tennessee 37067

Date of Panel Recommendation: May 12, 2011

Premarket Approval Application (PMA) Number: P100006

Date of FDA Notice of Approval: September 1, 2015

II. INDICATIONS FOR USE

Augment® Bone Graft is indicated for use as an alternative to autograft in arthrodesis (i.e., surgical fusion procedures) of the ankle (tibiotalar joint) and/or hindfoot (including subtalar, talonavicular, and calcaneocuboid joints, alone or in combination), due to osteoarthritis, post-traumatic arthritis, rheumatoid arthritis, psoriatic arthritis, avascular necrosis, joint instability, joint deformity, congenital defect, or joint arthropathy in patients with preoperative or intraoperative evidence indicating the need for supplemental graft material.

III. CONTRAINDICATIONS

Augment® Bone Graft should not:

- be used in patients who have a known hypersensitivity to any of the components of the product or are allergic to yeast-derived products.
- be used in patients with active cancer.
- be used in patients who are skeletally immature (<18 years of age or no radiographic evidence of closure of epiphyses).
- be used in pregnant women. The potential effects of rhPDGF-BB on the human fetus have not been evaluated.
- be implanted in patients with an active infection at the operative site.
- be used in situations where soft tissue coverage is not achievable.

Augment® Bone Graft SSED
Page 1 of 51

{1}

- be used in patients with metabolic disorders known to adversely affect the skeleton (e.g. renal osteodystrophy or hypercalcemia), other than primary osteoporosis or diabetes.
- be used as a substitute for structural graft.

## IV. WARNINGS AND PRECAUTIONS

The warnings and precautions can be found in the Augment® Bone Graft labeling.

## V. DEVICE DESCRIPTION

Augment® Bone Graft is a combination device/drug indicated for use as an alternative to autograft in arthrodesis (i.e., fusion procedures) of the ankle and/or hindfoot indicating the need for supplemental graft material. The use of Augment® Bone Graft eliminates the need for a second surgery to harvest autologous bone, thereby avoiding donor site morbidity which may occur (e.g., pain, infection, etc.).

Augment® Bone Graft combines recombinant human platelet-derived growth factor B homodimer (rhPDGF-BB) with a bioresorbable synthetic bone matrix (beta-tricalcium phosphate or β-TCP). The rhPDGF-BB is chemotactic for fibroblasts, neutrophils, and monocytes, cell types important for the early phases of tissue repair. The rhPDGF-BB is mitogenic for fibroblasts, osteoblasts, chondrocytes, and mesenchymal stem cells, which are important for later-stage tissue formation. The rhPDGF-BB functions as a chemo-attractant and mitogen for cells involved in wound healing and through its promotion of angiogenesis at the site of healing. The β-TCP acts as bone void filler to prevent soft tissue from collapsing into the void. When the β-TCP is placed near a viable host bone, it acts as a scaffold for new bone growth (osteoconductive).

These two components are packaged together and are physically combined immediately prior to use as follows:

- 1.5, 3, 6, or 9cc of synthetic β-TCP in granule form (nominal particle size 1 to 2 mm) provided in a 44mm PETG/PE laminate cup closed with a heat-sealed PET/PE/Foil laminate lid. The β-TCP cup is packaged into a PETG tray closed with a heat-sealed coated Tyvek® lid. The finished component (cup/tray subassembly) is terminally sterilized by gamma irradiation.
- 1.5, 3, 6, or 9 mL of rhPDGF-BB (0.3 mg/mL in 20mM USP sodium acetate buffer, pH 6.0) is aseptically filled into a single or multiple 3 mL USP Type I borosilicate glass vial(s) (Kimble or Schott) with a 13 mm chlorobutyl rubber stopper (West Pharma S2-F451 4432/50 Gray Butyl Rubber with B2-40 Coating) and a flip-off tear cap. The rhPDGF-BB vial is packaged into a PETG tray closed with a heat-sealed coated Tyvek® lid. The finished component (vial/tray subassembly) is terminally sterilized by ethylene oxide gas.

The surgeon estimates the volume of graft material required, which is dependent on the size of the voids to be filled, and selects the appropriate number of kits for the surgery. Equal volumes

Augment® Bone Graft SSED
Page 2 of 51

{2}

of rhPDGF-BB and $\beta$ -TCP are mixed in a sterile surgical bowl. Any remaining material is discarded.

Figure 1: Augment® Bone Graft

The two sub-assemblies of equal size are included in each kit, along with the package insert, and are packed together in an SBS carton as the final product (Figure 2).

Figure 2. Augment® Bone Graft 9cc final kit (finished product) showing the assembly of the $\beta$ -TCP device component (filled cup in a tray) and the rhPDGF-BB drug component (filled vial in a tray). The package insert is placed in the flipside of the carton when pulled open.

# VI. ALTERNATIVE PRACTICES AND PROCEDURES

Currently, there are no generally accepted alternatives to autograft for arthrodesis procedures of the ankle or hindfoot. Ankle and hindfoot arthrodesis procedures, such as triple arthrodesis

Augment® Bone Graft SSED

{3}

(three hindfoot articulations), double arthrodesis (two hindfoot articulations), subtalar arthrodesis, talonavicular arthrodesis, calcaneocuboid arthrodesis, and ankle arthrodesis involve the same treatment principles of creating a peri-articular osteotomy, stabilizing the joint with rigid fixation, placing harvested autograft bone, and following standard post-operative regimen of physical therapy and gradually increasing load on the fusion.

One of the most widely used options for bone graft is autologous bone, because there is no risk of cross contamination with autologous bone in contrast to allografts. However, clinical difficulties have been associated with autograft. Most of these difficulties result from the harvest of the bone graft, including increased operative time, hospital stay and cost, increased blood loss, post-operative pain, risk of infection, and/or fracture.

Other reported complications associated with autograft include a potential nidus for infection associated with avascular bone, limited tissue supply, and variability in cellular activity of the bone graft (Younger et al. 1989¹). In addition to these complications, limitation exists in the amount of bone graft that may be harvested. The morbidity associated with autograft and its limited amount available to be harvested has directed surgeons to look for a better alternative for a chemotactic, mitogenic, and angiogenic bone graft substitute to accelerate healing (St. John et al. 2003²).

## VII. MARKETING HISTORY

BioMimetic Therapeutics, LLC, received approval from Health Canada (2009) to begin marketing Augment® Bone Graft in Canada for ankle and hindfoot arthrodesis indications. Approval was granted from Australia (2011), New Zealand (2011), and Mexico (2013) to market Augment® Bone Graft for ankle and hindfoot indications. The components of Augment® Bone Graft (rhPDGF-BB and β-TCP) have been independently marketed in the United States, Canada, and internationally. Neither Augment® Bone Graft, nor any of its separate components, have ever been withdrawn from a market for any safety or effectiveness reason.

In addition, a chemically identical combination of β-TCP and rhPDGF-BB was developed by BioMimetic Therapeutics, LLC, under the trade name GEM 21S®. This product was approved in 2005, under PMA #P040013, to treat patients who have dental bone defects due to periodontal disease. This product is currently marketed in the United States and Canada by Osteohealth Company, a Division of Luitpold Pharmaceuticals, Inc., and over 200,000 units have been implanted. It has not been withdrawn from any market. The data from this PMA formed part of the basis for the IDE approval of Augment® Bone Graft.

## VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH

Patients may experience any of the following adverse events that have been reported in the literature with regard to the use of autograft or bone graft substitute products: swelling, pain, bleeding, hematoma, superficial or deep wound infection, cellulitis, wound dehiscence, incomplete or lack of osseous ingrowth, transient hypercalcemia, neuralgia and loss of sensation

Augment® Bone Graft SSED
Page 4 of 51

{4}

locally and peripherally, and anaphylaxis. Occurrence of one or more of these conditions may require an additional surgical procedure and may also require removal of the grafting material.

Below is a list of the potential adverse effects (e.g., complications) associated with the use of the Augment® Bone Graft identified from the Augment® Bone Graft clinical trial results and published scientific literature including: (1) those associated with any surgical procedure; (2) those associated with ankle and hindfoot arthrodesis surgery; and (3) those associated with bone graft substitute products for use in ankle and hindfoot arthrodesis, such as Augment® Bone Graft. In addition to the risks listed below, there is also the risk that surgery may not be effective in relieving symptoms, or may cause worsening of symptoms. Additional surgery may be required to correct some of the adverse effects and may also require removal of the grafting material.

1. Risks associated with any surgical procedure include: infection; pneumonia; atelectasis; septicemia; injury to blood vessels; soft tissue damage; phlebitis, thromboembolus, or pulmonary embolus; hemorrhage; respiratory distress; pulmonary edema; reactions to the drugs or anesthetic agent used during and after surgery; reactions to transfused blood; failure of the tissue to heal properly (e.g., hematoma, seroma, dehiscence, etc.) which may require drainage, aspiration, or debridement or other intervention; incisional pain; heart attack; stroke; and death.
2. Risks associated with ankle and hindfoot arthrodesis surgery with or without the use of graft material include: swelling; bleeding; hematoma; superficial or deep wound infection; cellulitis; wound dehiscence; transient hypercalcemia; neuralgia and loss of sensation locally and peripherally; anaphylaxis; incomplete or lack of osseous ingrowth, postoperative muscle and tissue pain; surgery may not reduce the preoperative pain experienced; pain and discomfort associated with the presence of implants used to aid in the arthrodesis surgery or reaction to the metal used in the implant, as well as the cutting and healing of tissues; the ankle and/or hindfoot may undergo adverse changes or deterioration including loss of height, and/or reduction, or malalignment, and another surgery may be required; and adverse bone/implant interface reaction.
3. Risks associated with bone graft substitute products, including Augment® Bone Graft, include non-unions, allergies or immunogenic response to implant materials, hypersensitivity, migration of the graft material into the surrounding soft tissue, musculoskeletal and connective tissue disorders, nervous system disorders, arthralgia, pain in extremities, and infections.

Adverse events that occurred in the Augment® Bone Graft clinical trial are listed in Section IX.

IX. SUMMARY OF NON-CLINICAL STUDIES

Pre-clinical studies were conducted to assess the safety and effectiveness of the individual and combined components contained in Augment® Bone Graft. BioMimetic Therapeutics, LLC conducted an extensive preclinical evaluation of rhPDGF-BB in combination with β-TCP

Augment® Bone Graft SSED
Page 5 of 51

{5}

matrices to provide sufficient data supporting its safe and effective use in treating orthopedic indications.

The structure, biology, and function of the rhPDGF-BB protein has been characterized and reported in the scientific literature. It has been reported that rhPDGF-BB functions to stimulate wound healing by attracting healing cells to wound sites, inducing them to proliferate, and supporting neovascularization to help establish an adequate blood supply. There is no evidence of systemic toxicity, acute or chronic, associated with rhPDGF-BB.

The efficacy of rhPDGF-BB combined with β-TCP for stimulating bone repair was evaluated in a variety of animal preclinical studies. Several animal models of bone repair demonstrated the local pharmacological activity and safety of rhPDGF-BB in combination with β-TCP and other matrix materials.

To evaluate the biocompatibility, toxicology, stability, and ADME/pharmacokinetics of rhPDGF-BB alone, and in combination with β-TCP, a series of studies were performed. The sponsor conducted a panel of biocompatibility/toxicology studies in compliance with ISO 10993 and USP guidelines. These studies evaluated β-TCP alone or in combination with rhPDGF-BB, and on β-TCP from several sources with or without rhPDGF-BB. The totality of the data from the biocompatibility studies demonstrated that rhPDGF-BB combined with β-TCP is non-toxic and biocompatible, which helps to define the safety profile for Augment® Bone Graft. In addition, a repeat-dose toxicity study to evaluate bone tissue responses to rhPDGF-BB in rats and an acute toxicity study to evaluate systemic toxicity following intravenous administration of rhPDGF-BB in rats were also performed. The results from these studies showed no signs of toxicity for rhPDGF-BB administered either locally or intravenously in animal models.

## Stability

Primary analytical data of an on-going stability study were provided supporting label storage claims and drug product expiration dating for 36 months. The study was performed on six (6) lots of 0.3 mg/mL rhPDGF-BB (three lots each of 1.5 mL and 3.0 mL in 3 mL vials). The study assessed stability of the samples stored under real-time conditions (2°C - 8°C) and accelerated aging room temperature conditions (22°C - 27°C). All drug product stored at 2°C - 8°C met the specification limits for all evaluated attributes through the 36 month time point. The rhPDGF-BB met the stability criteria under accelerated aging conditions for a duration corresponding to 36 months of real-time storage at 2°C - 8°C.

## Chronic Toxicity and Carcinogenicity

A study was conducted to evaluate the chronic toxicity and carcinogenicity of rhPDGF-BB mixed with β-tricalcium phosphate (β-TCP) matrix (Augment® Bone Graft) implanted in a rat model. Three hundred (150 male/150 female) Sprague-Dawley rats were randomly distributed into three groups (test article, control article, and sham surgery). Test or control article were implanted adjacent to the femur underneath the overlying muscle. The test article dose administered was 30 μg of rhPDGF-BB, which is approximately four times the maximum clinical dose. Animals were treated on day 0 and euthanized after 30, 180, or 365 days. Both

Augment® Bone Graft SSED
Page 6 of 51

{6}

macroscopic and microscopic evaluations were performed to evaluate toxicity and tumor incidence. Serum was collected for hematology, coagulation, and clinical chemistry determinations. Bone marrow was also collected from all animals at all time-points. Additionally, anti-PDGF-BB antibody formation was determined using ELISA.

No treatment-related mortality or effects on the clinical condition of the rats were observed. No remarkable article-related changes in body weight or body weight gain were observed. No significant changes in urinalysis parameters across treatment groups and gender were observed. Similarly, no significant changes in bone marrow parameters across treatment groups and gender were observed. In general, minor changes in clinical chemistry were observed, but due to the sporadic changes across treatment groups, gender, and time point, the changes were not attributed to treatment with rhPDGF-BB.

There were no test article-related microscopic findings on days 30, 180, or 365 of the study. On day 30, minimal foreign body granulomas were present at the implant site within the skeletal muscle across groups (control article, test article, and sham surgery) in the majority of animals examined. These granulomas contained material consistent with surgical sutures and were considered a result of the operative procedure. Minimal to mild granulation tissue was noted at the implant site in animals from the control and the test article groups (no animals from the sham surgery group displayed this change) on days 30, 180, and 365. This finding was likely related to the $\beta$ -TCP matrix and was not considered rhPDGF-BB related. Mild hyperostosis was present on the periosteal surface of the femur where the granulation tissue was present at days 180 and 365, but not at day 30. It consisted of well-differentiated bone that was indistinguishable from the normal femoral bone. This hyperostosis was considered secondary to a local effect of the $\beta$ -TCP matrix on the periosteal surface of the femur.

An adenocarcinoma of the mammary gland was noted in one of the 10 females of the test article (rhPDGF-BB) group on day 180. This finding was considered incidental, based on its unique occurrence and the absence of any hyperplastic changes noted in this group. There were no test article-related neoplastic microscopic observations noted in either sex on day 365. Serum test article antibody analysis of 590 samples found one sample to be positive for anti-rhPDGF-BB antibodies. This animal was treated with the control article, i.e., a false positive.

The results of this study demonstrated that implantation of the test article was not associated with any unexpected mortality, clinical findings, or changes in body weight or food consumption. In addition, implantation of the test article was not associated with any treatment-related changes in hematology, coagulation, clinical chemistry, or bone marrow parameters. Upon necropsy and histopathologic evaluation, no differences were noted in tissue response between the sham or control treated animals and the test article implanted animals. The results of this study demonstrate that implantation of the test article did not result in any toxicity or tumorigenicity and, in addition, demonstrate that the test article was biocompatible.

## Pharmacokinetics

Augment® Bone Graft SSED
Page 7 of 51

{7}

Two animal studies were presented that provide data characterizing the pharmacokinetics of $^{125}\mathrm{I}$-rhPDGF-BB, its metabolism and excretion, and tissue distribution in a rat model. Both studies indicated that rhPDGF-BB is cleared rapidly from the blood (mainly in the urine), with lesser amounts eliminated in the feces following intravenous administration. There is limited systemic exposure to the protein following intramuscular implantation of $^{125}\mathrm{I}$-rhPDGF-BB combined with Augment® Bone Graft's $\beta$-TCP and clearance is again mainly in the urine. Overall, the toxicology and pharmacokinetic data demonstrated that rhPDGF-BB combined with Augment® Bone Graft's $\beta$-TCP does not lead to any signs of acute or chronic toxicity and the protein is eliminated rapidly from the body following administration with limited systemic exposure. However, while the evaluations provide information regarding the relative clearance rates of injected or implanted rhPDGF-BB, they only approximate how the product will be implanted in ankle or hindfoot bone grafting procedures.

The characterization of the release kinetics, biological potency, and biochemical integrity of rhPDGF-BB combined with $\beta$-TCP from different sources was also studied. Both in vivo and in vitro preclinical data demonstrated that rhPDGF-BB is released rapidly from Augment® Bone Graft's $\beta$-TCP and other sources of $\beta$-TCP in a similar fashion. The protein retains its biological potency and is biochemically intact following release from $\beta$-TCP matrices as determined by in vitro, cell-based analyses. Thus, the data support Augment® Bone Graft's $\beta$-TCP as an appropriate matrix as the device component of this combination product.

## Reproductive Development/Teratology

A reproductive development/teratology study in female Sprague-Dawley rats was conducted at two dose levels (1 times and 10 times the maximum single, clinical dose) repeated daily over 21 days starting on day zero of gestation, i.e., the low dose group received 21 times (21 days of 1x dose amounts) and the high dose group 210 (21 days of 10x dose amounts) times the maximum clinical dose during the extended period of administration. The rats were divided into three groups: 22 rats served as the control group; 22 rats received $0.04\mathrm{mg / ml}$ rhPDGF-BB $40~\mu \mathrm{g / kg / day}$ ("the lose dose group"); and 22 rats received $0.4\mathrm{mg / ml}$ rhPDGF-BB $400~\mu \mathrm{g / kg / day}$ ("the high dose group"). Detailed examinations, which included measurement of body weight and food consumption, were performed on gestation days 0, 3, 6, 9, 12, 15, 18, and 21. Maternal blood sampling was performed on gestation days 0 and 21 and fetal blood sampling was performed on day 21 of gestation.

The fetuses were evaluated macro- and microscopically for visceral and skeletal development anomalies. No visceral or skeletal anomalies were observed in the control or the low dose rhPDGF-BB groups; no visceral anomalies were found in the first generation fetuses. There were indications of somewhat accelerated ossification of the interparietal and hyoid bones and slight (not significant) increase in the presence of a rudimentary 14th rib in first generation fetuses from maternal rats (dams) receiving rhPDGF-BB. The low dose group had a higher incidence of incomplete ossification of the hyoid bone, while the high dose group had lower incidence of incomplete ossification of the interparietal bone. However, this finding was not dose-dependent; the incidence of incomplete ossification of the hyoid bone in the high dose group was not different from the control group. These observations were made in comparison to the control group.

Augment® Bone Graft SSED
Page 8 of 51

{8}

No detectable amount of rhPDGF-BB was found in the maternal and fetal plasma samples. Anti-rhPDGF-BB antibodies were detected in one out of 15 dam pretreatment samples analyzed. All dam and fetal post-treatment samples analyzed were negative for antibodies to rhPDGF-BB.

The administration of rhPDGF-BB at 0.040 and 0.40 mg/kg/day, by intravenous injection, to female rats from Gestation Days 0 to 20 resulted in neither maternal toxicity, nor adverse effects on embryo-fetal development, in this study. Based on these results, 0.40 mg/kg/day (i.e., the highest dose tested in this study) was the no-observed-effect-level for maternal toxicity and the no observed-adverse-effect-level for embryo-fetal development.

Table 1 summarizes the biocompatibility studies and Table 2 summarizes the preclinical animal studies performed. See below.

Table 1: Biocompatibility Studies

| Test | Standard | Methods | Result | Pass |
| --- | --- | --- | --- | --- |
| Genotoxicity of β-TCP | ISO 10993 Part3 | Reverse Mutation Assay using NaCl and CSO Extracts (Salmonella typhimurium and Escherichia coli) | Nonmutagenic to Salmonella typhimurium tester strains TA98, TA100, TA1535, and TA1537, and to Escherichia coli strain WP2 | Pass |
| Intracutaneous Reactivity of β-TCP | ISO 10993 Part 11 | Intracutaneous Injection Test using NaCl and CSO Extracts in New Zealand White Rabbits | Negligible irritant | Pass |
| Sensitization of β-TCP | ISO 10993 Part 10 | Klingman Maximization Test using NaCl and CSO Extracts in Guinea Pigs | Grade I Reaction (0% sensitization) | Pass |
| Cytotoxicity of β-TCP | ISO 10993 Part 5 | L929 MEM Elution Test | Non-cytotoxic. Grade 2 Reaction (slight) | Pass |
| Intramuscular Reactivity of β-TCP | ISO 10993 Part 6 | Intramuscular Implantation Test (4 Week Implantation) in New Zealand White Rabbits | Moderate Bioreactivity Rating of 9.5-15.0 | Pass |
| Acute Toxicity of β-TCP | ISO 10993 Part 11 | Systemic Injection Test in Swiss Albino Mice | No biological response when compared to control. | Pass |
| Genotoxicity of Augment® Bone Graft Bone Graft | ISO 10993 Part3 | Reverse Mutation Assay using NaCl and CSO Extracts (Salmonella typhimurium and Escherichia coli) | Nonmutagenic to Salmonella typhimurium tester strains TA98, TA100, TA1535, and TA1537, and to Escherichia coli strain WP2 | Pass |
| Intracutaneous Reactivity of Augment® Bone Graft | ISO 10993 Part 11 | Intracutaneous Injection Test using NaCl and CSO Extracts in New Zealand White Rabbits | Negligible irritant | Pass |
| Sensitization of Augment® Bone Graft | ISO 10993 Part 10 | Klingman Maximization Test using NaCl and CSO Extracts in Guinea Pigs | Grade I Reaction (0% sensitization) | Pass |
| Cytotoxicity of Augment® Bone Graft | ISO 10993 Part 5 | L929 MEM Elution Test | Non-cytotoxic. Grade 1 Reaction (slight) | Pass |
| Intramuscular Reactivity of Augment® Bone Graft | ISO 10993 Part 6 | Intramuscular Implantation Test (4 Week Implantation) in New Zealand White Rabbits | Bioreactivity Rating of 0.5 | Pass |
| Acute Toxicity of Augment® Bone Graft | ISO 10993 Part 11 | Systemic Injection Test in Swiss Albino Mice | No biological response when compared to a control. | Pass |

Augment® Bone Graft SSED

{9}

Table 2: Pre-Clinical Animal Studies

| Test | Methods | Result |
| --- | --- | --- |
| Acute Toxicity of rhPDGF-BB Following Intravenous Administration in Rats (Acute, single dose Systemic Toxicity) | Single IV injection of 0.2 and 4.0 mg/kg of rhPDGF-BB; follow-up 2 and 15 days The high dose (4,000 μg/kg), was approximately 100 times the maximum clinical dose (39 μg/kg) | • Intravenous rhPDGF-BB at 0.15 or 3.0 mg/mL did not elicit significant toxicity in rats • rhPDGF-BB has a high margin of safety in this assay when administered intravenously |
| Evaluation of the chronic toxicity and carcinogenicity of recombinant human platelet-derived growth factor-BB (rhPDGF-BB) mixed with β-tricalcium phosphate (β-TCP) matrix (Augment® Bone Graft) implanted in a rat model (Chronic Local Toxicity and carcinogenicity) | Single implantation of 0.175 mg/kg of Augment® Bone Graft bone graft; follow-up 30, 180 and 365 days | • No evidence of carcinogenicity • No treatment-related mortality or effects on body weight, hematology, coagulation, clinical chemistry, bone marrow parameters, or histopathology • No differences in local tissue response between groups • No anti-PDGF-BB antibodies seen in the Augment® Bone Graft-treated group |
| Bone Response to Intramuscular Injections of rhPDGF-BB (Acute Local Toxicity; Repeated dose) | Repeated IM injections in muscle next to femur; doses of 13.9; 41.2 and 138.8 μg/kg/day every other day for 2 weeks; follow-up 2 and 6 weeks | • Effects at the high dose for both soft tissue and bone were consistent with the mechanism of action of PDGF • The responses were transient and not present 6 weeks after the last dose |
| Recombinant Human Platelet-derived Growth Factor-BB (rhPDGF-BB): an Intravenous Injection Teratology Study in the Rat (Reproductive and developmental Toxicity) | Repeated IV injections daily on gestation days 0-020; assessment of maternal and fetal toxicity | • No maternal toxicity • No adverse effects on embryo-fetal development • Minor transitory increases in the rate of ossification in the high dose fetuses. • No detectable neutralizing antibodies against rhPDGF-BB • NOAEL for maternal toxicity is 400 μg/kg/day • NOAEL for embryo fetal development is 400 μg/kg/day |
| Bacterial Mutagenicity Test - AMES Assay (Genotoxicity) | • Potential of rhPDGF-BB to induce: · histidine (His) reversion in S. typhimurium · tryptophan reversion in E. coli • Six dose levels with the top dose tested at 10 mg/mL (1.0 mg/plate) | The highest dose tested: 10 mg/mL (1.0 mg/plate) rhPDGF-BB was non-mutagenic |

Augment® Bone Graft SSED

{10}

Augment® Bone Graft SSED
Page 11 of 51
| Test | Methods | Result |
| --- | --- | --- |
| Fracture Healing of the Tibia in Geriatric-Osteoporotic Rats | • Treatment groups (n=110 rats):
• 1.0 mg/mL rhPDGF-BB + collagen/β-TCP
• 0.3 mg/mL rhPDGF-BB + collagen/β-TCP
• sodium acetate + collagen/β-TCP
• untreated fracture
• Model:
• Unilateral fracture model (Einhorn model)
• Outcomes:
• Cellular proliferation – 4 days post-fracture (n=26)
• Biomechanical testing – 6 and 8 weeks post fracture (n=66)
• Histomorphometry – 12 weeks post-fracture (n=18) | • At five weeks, the mechanical strength of Augment® Bone Graft-treated tibias was not different from the non-fractured contralateral tibia
• There were no untoward tissue responses |
| Diabetic Rat Fracture Model | • Treatment groups (n=110 rats):
• 1.0 mg/mL rhPDGF-BB + collagen/ β-TCP
• 0.3 mg/mL rhPDGF-BB + collagen/ β-TCP
• sodium acetate + collagen/ β-TCP
• untreated fracture
• Model:
• Unilateral fracture model (Einhorn model)
• Outcomes:
• Cellular proliferation – 4 days post-fracture (n=26)
• Biomechanical testing – 6 and 8 weeks post fracture (n=66)
• Histomorphometry – 12 weeks post-fracture (n=18) | • Augment® Bone Graft treatment in diabetic rats resulted in:
• Increased cell proliferation at 4 days
• Increased biomechanical strength as early as 6 weeks
• Increased bone content of the fracture callus at 12 weeks
• No evidence of either abnormal or ectopic bone formation |
| Partial arthrodesis of the carpus in dogs | • Treatment groups (n=30 dogs):
• Autologous bone graft
• rhPDGF-BB + Autologous bone graft
• β-TCP
• rhPDGF-BB + β-TCP
• Collagen/β-TCP
• rhPDGF-BB + Collagen/β-TCP
• Model:
• Two-level arthrodesis of the carpus
• Outcomes:
• Manual palpation – 5 and 12 weeks post-surgery (n=7; n=3)
• Radiograph – 5 and 12 weeks post-surgery (n=7; n=3)
• Histology – 12 weeks post-surgery (n=13) | • Addition of rhPDGF-BB to the graft materials increased the number and extent of fused joints compared to the materials alone
• New bone formed was normal
• No evidence of ectopic bone formation
• No evidence of acute or chronic toxicity |

{11}

Augment® Bone Graft SSED
Page 12 of 51

| Test | Methods | Result |
| --- | --- | --- |
| Biological Assessment of a Bone Repair Model (Rabbit Tibial Osteotomy) | • Unicortical 5-mm osteotomies
• Treated with cylinders of:
• β-TCP
• β-TCP + 25 μg rhPDGF-BB
• β-TCP+ 75 μg rhPDGF-BB
• Histomorphometric analysis at 4 and 8 weeks post-implantation | • Residual β-TCP detected at four weeks was significantly reduced by eight weeks
• no statistically significant differences among the treatment groups |
| Vertical ridge bone augmentation in a dog model | • Critical sized chronic defect
• Treated with implants of deproteinized cancellous block
• ± rhPDGF-BB
• Evaluation 4 months post-op | • Matrix with rhPDGF-BB created 1 cm vertical bone growth as compared with matrix alone
• Histology analysis confirmed accelerated bone remodelling with rhPDGF-BB treated sites |
| Tissue distribution and mass balance of radioactivity in Sprague-Dawley rats following an intravenous injection of ^{125}I-rhPDGF-BB | • Single IV Injection 0.31 mg/kg
• Whole body autoradiography, blood, urine, feces and cage residue collected for radioactivity analysis at different time points up to 7 days | • rhPDGF-BB is widely distributed and cleared rapidly from the circulation.
• T_{max}: 30 min
• 18% of C_{max} at 4 hours
• Radioactivity was excreted in urine and feces primarily as unbound ^{125}I with smaller amounts of bound ^{125}I also excreted in feces |
| Pharmacokinetics of radioactivity in Sprague-Dawley rats following intravenous administration or intramuscular implantation of ^{125}I-labeled recombinant human Platelet-Derived Growth Factor-BB (rhPDGF-BB) combined with β-Tricalcium Phosphate (β-TCP) | • Single IV Injection 0.31 mg/kg
• Single IM Implantation adjacent to femur 0.29 mg/kg
• blood, urine, feces and cage residue collected for radioactivity analysis at different time points up to 7 days | • The systemic bioavailability of the test article was similar by both routes of administration
• rhPDGF-BB is rapidly released from the β-TCP matrix over the 24 hours following IM implantation, and is nearly depleted from the implant site by 168 hours post-dose:
• T_{max}: 8 hours
• t_{½}: 30.3 hours
• 3% of C_{max} at 168 hours |
| Evaluation of In Vivo Release of ^{125}I-Recombinant Human Platelet-Derived Growth Factor (^{125}I-rhPDGF-BB) from β-TCP/COLLAGEN and β-TCP Matrices Implanted in a Rat Calvarial Bone Defect | • Single Implantation in Calvarial defect
• β-TCP: 56 μg/kg
• β-TCP /Collagen: 112 μg/kg
• Radioactivity at implantation site measured at different intervals up to 7 days | • Rapid release of 50% in the first 60 minutes and 80% in the first 24 hours
• Only 2% of input counts at 7 days
• No differences between groups |
| Evaluation of In Vivo Release of ^{125}I-Recombinant Human Platelet-Derived Growth Factor (^{125}I-rhPDGF-BB) from β-TCP (β-Tricalcium Phosphate) and β-TCP/Human Bone Allograft Matrices Implanted in a Rat Calvarial Bone Defect | • Single Implantation in Calvarial defect
• β-TCP (1000 – 2000 μm): 182.4 μg/kg
• β-TCP (250 – 1000 μm): 187.5 μg/kg
• Allograft+β-TCP (250 - 1000 μm): 237.6 μg/kg
• Radioactivity at implantation site measured at different intervals up to 3 days | • Rapid release of 50% in the first 30 minutes
• Only 10% of the initial radioactivity was present at the implantation site at 72 hours (3 days)
• No differences among groups |

{12}

Summary of Human Pharmacokinetic Study

BioMimetic Therapeutics, LLC, conducted a clinical study to evaluate the pharmacokinetic profile of the rhPDGF-BB in Augment compared to autograft control subjects when implanted in the human hindfoot or ankle. A total of 11 subjects were treated: 4 subjects received standard rigid fixation plus autologous bone graft and 7 subjects received standard rigid fixation plus Augment Bone Graft. Blood samples were collected from each subject prior to treatment and at 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, 1 day, 2 days, 3 days, and 7 days. The blood samples were processed to obtain serum, which was frozen and stored until analysis of the PDGF-BB concentration.

Serum PDGF-BB levels after the administration of 6-9 cc of Augment used in this study fell within the PDGF-BB concentration range of the autograft control subjects receiving comparable volumes of autologous bone graft. Seventeen of the 119 serum samples tested showed quantifiable levels of PDGF-BB (above 7.8 ng/mL). The 17 samples with quantifiable levels of PDGF-BB were found in three subjects; two of three subjects received autograft. The data suggest a low systemic exposure to rhPDGF-BB following one time implantation of Augment (up to 2.7 mg of rhPDGF-BB) in the human hindfoot or ankle and may be supportive of the safety of Augment. Caution should be taken when interpreting these data because the assay for measuring rhPDGF-BB in human serum has not been fully validated.

## X. SUMMARY OF PRIMARY CLINICAL STUDY

The data used as the basis for this PMA approval came from a prospective, randomized, controlled, non-inferiority, multi-center clinical trial to evaluate the safety and effectiveness of Augment® Bone Graft compared to autologous bone graft in bone grafting procedures. The trial enrolled subjects requiring a bone grafting procedure in either the ankle or hindfoot that necessitated the placement of bone graft. The trial was performed in the United States and Canada under IDE # G050118. A summary of the clinical trial is presented below.

## A. Study Design

Subjects were treated between April 2007 and January 2009. The database for this PMA reflected data collected through January 2010, and included 434 subjects (414 subjects with surgery). There were 37 sites in the US and Canada.

The clinical trial was a prospective, randomized, concurrently controlled, non-inferiority, multicenter clinical trial to evaluate the safety and effectiveness of Augment® Bone Graft compared to autograft as the concurrent control to fill bony defects used in conjunction with supplemental hardware in the treatment of ankle and hindfoot arthrodesis procedures.

This was a frequentist non-inferiority study with a pre-specified primary endpoint of proportion of patients with fusion and a non-inferiority margin of 10%. The statistical model for this endpoint was two independent binomial proportions.

Augment® Bone Graft SSED

{13}

Fusion was assessed using CT imaging for the full complement of joints, defined as evaluating all joints, and was classified as a success only if all treated joints possessed at least 50% bridging osseous bone. An alternative approach analyzed each joint separately.

Joints were assessed via plain films in four imaging planes (medial, lateral, anterior/superior, posterior/inferior). Each plane was classified as fused, not fused, or not evaluable; a joint had to be classified as fused by three of these images to be called fused. An alternative approach required only two of the images to be classified as fused.

Letting $p_{\text{Augment® Bone Graft}}$ and $p_{\text{Autograft}}$ represent the proportions with 24-week fusion success for the Augment® and autograft bone groups, respectively, and $\delta = 0.10$ being the non-inferiority margin, the statistical hypotheses for the pre-specified primary endpoint were:

$$
\begin{array}{l}
H_0: p_{\text{Augment® Bone Graft}} - p_{\text{autograft}} \leq -\delta \\
H_a: p_{\text{Augment® Bone Graft}} - p_{\text{autograft}} > -\delta \\
\end{array}
$$

These statistical hypotheses were assessed via one-sided 95% confidence intervals on the difference in the proportion of responders in the Augment®-treated group minus the proportion of responders in the autograft-treated group.

Quality-of-life and functional outcomes scores which were studied included SF-12, Foot Function Index (FFI), and AOFAS Ankle-Hindfoot questionnaires. In addition, there were three VAS pain assessments: bone graft harvest site pain, fusion site pain, and pain with weight-bearing. By definition, Augment® Bone Graft subjects had no harvest site pain and were classified as 0 for that aspect. Asymptotic methods were used to calculate all $p$-values.

Importantly, because the assessment of radiographic fusion for Augment® Bone Graft patients at 24 weeks was inconclusive, demonstration of effectiveness relied primarily on the clinical secondary endpoints and, in particular, the endpoint of weight bearing pain. Please refer to the Effectiveness Results section for a more detailed discussion of this issue.

The study sample size calculation was based on testing the original primary endpoint of fusion and assumed a one-sided 0.05 significance level, 2:1 randomization, 10% non-inferiority margin, and an expected 24-week fusion rate of 85% for both treatment groups. The resulting sample size was 238 Augment® Bone Graft and 119 bone graft, yielding 80% power. The proposed sample size was increased to 396 to account for an anticipated 10% dropout.

Upon confirmation of eligibility, patients were randomized into one of two treatment groups: standard rigid fixation plus Augment® Bone Graft or standard rigid fixation plus autologous bone graft (autograft). Randomization was conducted at a 2:1 ratio of Augment® Bone Graft to autograft. The subjects then underwent an ankle/hindfoot bone grafting procedure using an open surgical technique with supplemental bone graft or Augment® Bone Graft, according to the randomization assignment. The bone grafting construct required adequate reduction and

Augment® Bone Graft SSED
Page 14 of 51

{14}

stabilization with rigid fixation intra-operatively to meet final study eligibility. Both treatment groups were immobilized according to standardized operative and post-operative protocols.

The investigators, who were fellowship trained and Board Certified ankle and hindfoot surgeons, performed clinical and radiographic assessments (as required by protocol) to monitor healing/union status. A masked independent radiographic assessment was performed by a designated fellowship trained and Board Certified musculoskeletal radiologist who assessed radiographic parameters for fusion.

All enrolled subjects were to be monitored over a 12-month period to evaluate for clinical and safety outcomes, including incidence of loss of reduction, infection, non-union, need for revision surgery, and associated complications with ankle and hindfoot arthrodesis procedures, in addition to the occurrence of other adverse device effects. A Data Safety Monitoring Board (DSMB) supervised the reporting of unanticipated, device-related, and serious adverse events.

1. Clinical Inclusion and Exclusion Criteria

Enrollment in the Augment® Bone Graft IDE clinical trial was limited to patients who met the following inclusion criteria.

a. The patient signed the IRB-approved Informed Consent Form specific to this study prior to enrollment.
b. The patient had a bone defect (surgically created osseous defects or osseous defects resulting from pathology or traumatic injury to the bone) in the ankle or hindfoot requiring fusion using open surgical technique with supplemental bone graft/substitute, requiring one of the following procedures:

i. Ankle joint fusion (tibiotalar fusion)
ii. Subtalar fusion
iii. Calcaneocuboid fusion
iv. Talonavicular fusion
v. Double fusions
vi. Triple fusions

c. The fusion site was able to be rigidly stabilized with no more than 3 screws across the fusion site. Supplemental pins may have been used. Supplemental screws external to the fusion site(s) were also allowed.
d. The patient was independent, ambulatory, and could comply with all post-operative evaluations and visits.
e. The patient was at least 18 years of age and considered to be skeletally mature.

Patients were not permitted to enroll in the Augment® Bone Graft IDE clinical trial if they met any of the following exclusion criteria.

a. The patient had undergone previous fusion surgery of the joints to be fused.
b. The fusion site required plate fixation*, more than 3 screws across the fusion site to achieve rigid fixation, or more than 3 kits (9 cc) of graft material.

Augment® Bone Graft SSED
Page 15 of 51

{15}

c. There was radiographic evidence of bone cysts, segmental defects or growth plate fracture around the fusion site that may negatively impact bony fusion.

d. The patient currently had untreated malignant neoplasm(s) at the surgical site, or was currently undergoing radio- or chemotherapy.

e. The patient had a pre-existing sensory impairment (e.g., diabetes with baseline sensory impairment) which limited the ability to perform objective functional measurements and may have placed patients at risk for complications. For the purpose of this protocol, diabetics that were not sensitive to the 5.07 monofilament (Semmes-Weinstein) were to be excluded.

f. The patient had a metabolic disorder known to adversely affect the skeleton, other than primary osteoporosis or diabetes (e.g., renal osteodystrophy or hypercalcemia).

g. The patient used chronic medications known to affect the skeleton (e.g., glucocorticoid usage > 10 mg/day). Non-steroidal anti-inflammatory drug (NSAID) use was excluded during the first 6 weeks post-operatively.

h. The patient had a pre-fusion neuromuscular or musculoskeletal deficiency which limited the ability to perform objective functional measurements.

i. The patient was physically or mentally compromised (e.g., currently being treated for a psychiatric disorder, senile dementia, Alzheimer’s disease, etc.) to the extent that the investigator judged the patient to be unable or unlikely to remain compliant.

j. The patient had an allergy to yeast-derived products.

k. The patient had received an investigational therapy or approved therapy for investigational use within 30 days of surgery or during the follow-up phase of this study.

l. The patient was a prisoner, known or suspected to be transient, or had a history of drug/alcohol abuse within the 12 months prior to screening for study entry.

m. The patient was pregnant or a female intending to become pregnant during the study period. A urine pregnancy test was to be administered within 21 days of the surgical visit to any female unless she was post-menopausal, had been sterilized, or was practicing a medically-accepted method of contraception.

n. The patient was deemed morbidly obese (body mass index [BMI] > 45 kg/m²).

* Screw fixation was used in all study subjects to reduce the number of variables.

2. Follow-up Schedule

All patients were scheduled to return for follow up examinations at 7-21 days post-operative, 6 weeks (±1 week), 9 weeks (±1 week), 12 weeks (±1 week), 16 weeks (±1 week), 24 weeks (±2 weeks), 36 weeks (±2 weeks), and 52 weeks (±2 weeks).

Clinical assessments included the primary effectiveness variable of bridging bone at the involved level, in addition to secondary pain/disability status, general health status, and graft site pain. The protocol also included measurements of antibodies for anti-rhPDGF-BB screening in the Augment® Bone Graft group at baseline (prior to grafting procedure), visit 3 (day 7-21), visit 4 (week 6), visit 6 (week 12), and visit 8 (week 24). The objective parameters measured preoperatively and postoperatively during the study are included in Table 3 below. Adverse events and complications were recorded at all visits.

Augment® Bone Graft SSED
Page 16 of 51

{16}

The key time-points are shown below in the tables summarizing safety and effectiveness.

Table 3. Schedule of Study Assessments

| Procedure | Screening | Surgery | Post-Treatment Follow-up Evaluation | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | Visit 1 | Visit 2 | Visit 3 | Visit 4 | Visit 5 | Visit 6 | Visit 7 | Visit 8 | Visit 9 | Visit 10 |
| | Within 21 Days of Surgery | Day 0 | Day 7-21 | Wk 6 ±7 Days | Wk 9 ±7 Days | Wk 12 ±7 Days | Wk 16 ±7 Days | Wk 24 ±14 Days | Wk 36 ±14 Days | Wk 52 ±14 Days |
| Informed Consent | X1 | | | | | | | | | |
| Urine Pregnancy Test (if applicable) | X | | | | | | | | | |
| Medical History / Non-Union Risk Factors | X | | | | | | | | | |
| Physical Examination of Foot / Ankle Region | X | X | X | X | X | X | X | X | X | X |
| Eligibility Criteria Verification | X | X | | | | | | | | |
| Identification of Target Bone Defect | X | | | | | | | | | |
| Serum for rhPDGF-BB Ab Testing | X2 | | X | X | | X | | X | | |
| Patient Randomization | | X3 | | | | | | | | |
| Intraoperative Report | | X | | | | | | | | |
| Volume of Graft Material Placed (if applicable) | | X | | | | | | | | |
| Physical Therapy | | | | X | X | X | X | X | X | |
| Radiographic Outcomes | X4 | X | X | X | X | X | X | X | X | X |
| CT Scans5 | | | | | X | | X | X | X | |
| Clinical / Functional Assessments6 | X | | | X | X | X | X | X | X | X |
| Pain Assessments | X | X | X | X | X | X | X | X | X | X |
| Quality-of-Life Assessments | X | | | X | | X | X | X | X | X |
| Adverse Events /Complications | | X | X | X | X | X | X | X | X | X7 |
| Concomitant Medications Review | X | X | X | X | X | X | X | X | X | X |
| Ab = antibody; CT = computed tomography1 Must have occurred prior to any study-specific procedures.2 Peripheral blood sample for antibody evaluation was to be taken at baseline (prior to grafting), Day 7-21, Week 6, Week 12, and Week 24. For patients who tested positive for antibodies to rhPDGF-BB, additional serum samples were to be requested in order to monitor patients until antibody titers returned to baseline. Patients testing positive for anti-rhPDGF-BB antibodies were to be tested for neutralizing activity.3 Interactive web randomization within 48 hours of scheduled surgery.4 Pre-operative radiographic films may have been taken within 6 months of surgery. Radiologic assessments including osseous bridging across subchondral surfaces (primary endpoint for union), callus formation, % osseous bridging, and heterotopic bone formation were to be used to assess overall fusion site healing. Plain film radiographs were to be the primary source of data for clinical assessment of fusion.5 CT scans (0.5-0.7 mm thickness at 0.2-0.3mm intervals, pitch of 0.7, and kVp of 130-140) were to be taken at Week 9, 16, 24, and 36. A baseline CT scan may have been taken to confirm that there were no radiographic signs of cysts that would exclude the patient. CT scans were to be assessed for radiographic union by independent radiologist.6 VAS pain assessment, SF-12 quality-of-life assessment and functional assessments include range of motion, and weight bearing (Foot Function Index and AOFAS Hindfoot / Ankle scale).7 Non-unions (therapeutic failures) after 12-month follow-up were to be collected | | | | | | | | | | |

Augment® Bone Graft SSED

{17}

Augment® Bone Graft SSED
Page 18 of 51

3. Clinical Endpoints

The following clinical, functional, radiologic, and safety endpoints were used to evaluate the risk-benefit profile of Augment® Bone Graft:

Safety Endpoints:
- Presence of treatment emergent adverse events (TEAEs)
- Secondary Surgeries
- Serum sample analysis for presence of anti-rhPDGF-BB antibodies.

Clinical / Functional Endpoints:
- Pain on weight bearing
- Fusion site pain
- Foot Function Index (FFI)
- American Orthopaedic Foot and Ankle Society (AOFAS) Score
- SF-12 (PCS)
- Non-union rate
- Graft harvest site pain

Radiologic Endpoint:
- ≥50% osseous bridging across the joint space as determined using computed tomography (CT) images at 24 weeks post-operative

Clinical and Functional Endpoints Used to Determine Individual Success by FDA
- Pain on weight bearing
- FFI
- AOFAS Score

B. Accountability of PMA Cohort

There were three patient populations separately accounted for: Intent to Treat (ITT), modified Intent to Treat (mITT), and Safety or “All Treated.” The ITT population consisted of 434 patients. Of these, 285 were implanted with Augment® Bone Graft and 149 received autograft. The mITT population, submitted as the primary effectiveness analysis for the radiographic evaluation of bridging bone, consisted of 397 patients (414 patients in the Safety, or “All Treated”, group minus an additional 17 subjects excluded post-operatively) divided into 260 with Augment® Bone Graft and 137 with autograft. (Figure 3)

Efficacy results, based on the radiographic interpretation of bridging bone with a “full complement of joints” analysis, was assessed on the mITT population, which included 397 evaluable patients (597 joints) who were considered by the investigator as eligible, properly randomized, and received treatment in accordance with the clinical trial protocol (“per protocol population”). Of these, 260 subjects (394 joints) received Augment® Bone Graft and 137 subjects (203 joints) received autograft. However, because interpretation of the primary radiographic endpoint for the study was inconclusive due to the radiographic attenuation of the Augment® Bone Graft implant, it was necessary to consider success or failure of the product in terms of success or failure of the individual with clinical outcome measures of pain and function

{18}

(VAS on weight bearing, FFI, and AOFAS score). Therefore, in order to determine individual success, those individuals who had medically relevant protocol deviations and missing data were not considered in this analysis (Figure 3).

Figure 3: Subject Accounting Tree

The "All Treated" population consisted of 414 randomized and treated subjects (272 Augment® Bone Graft and 142 autograft). The mITT population consists of 397 subjects, as 17 subjects were removed due to post-operative exclusions. This group of 17 included subjects who required hardware prohibited by the protocol (5), had an excluded joint fused (9), had chronic steroid use (1), were weight bearing at two weeks (1), and required more than the allowed amount of graft material (1). Of the 397 per protocol subjects, 260 were implanted with Augment® Bone Graft and 137 received the autograft control treatment. This group was used to assess the success or failure of the study's radiographic primary endpoint for bridging bone at 24 weeks. Regarding the assessment for Individual Patient Success, overall there were 301 individuals for VAS on weight bearing (204 Augment® Bone Graft and 97 autograft), and 332

Augment® Bone Graft SSED

{19}

for the FFI and AOFAS functional outcome measures (223 Augment® Bone Graft and 109 autograft).

Patient accounting at the week 24 time point of the primary endpoint is presented in Table 4.

Table 4: 24 Week Data Accounting for Augment® Bone Graft

| Parameter | Augment® Bone Graft | Autograft |
| --- | --- | --- |
| Randomized | 285 | 149 |
| Withdrawn prior to treatment | 13 | 7 |
| Subjects Treated | 272 | 142 |
| Failed Study Inclusion Criteria | 12 | 5 |
| Per protocol cohort | 260 | 137 |
| Death | 1^{1} | 0 |
| Device Related Serious Adverse Event (SAE) | 1 | 0 |
| Secondary Procedures Among treated^{2} | 5 | 4 |
| Medically Relevant Protocol Deviations^{3} | 38 | 29 |

1. Death due to pulmonary embolus was not evaluated by the investigator as being related to Augment® Bone Graft, but was likely procedure related.
2. Subjects with secondary procedures due to non-union, delayed union, or application of bone stimulator to prevent one of these outcomes prior to Week 24.
3. Subjects who were considered as individual failures due to protocol deviations that materially impacted the outcome assessments of pain and function.

Augment® Bone Graft SSED

{20}

# C. Study Population Demographics and Baseline Parameters for "All Treated" Patients

Subject demographics are summarized in Table 5.

Table 5: Demographic & Clinical Characteristics at Baseline - "All Treated" Population

| | Augment® Bone Graft (n=272) | | | Autograft (n=142) | | |
| --- | --- | --- | --- | --- | --- | --- |
| Gender | | | | | | |
| Male | 129 (47.4%) | | | 81 (57.0%) | | |
| Female | 143(52.6%) | | | 61 (43.0%) | | |
| Affected Ankle/Hindfoot | | | | | | |
| Right | 135 (49.6%) | | | 77 (54.2%) | | |
| Left | 128 (47.1%) | | | 56 (39.4%) | | |
| Bilateral | 9 (3.1%) | | | 9 (6.3%) | | |
| Arthrodesis Procedure Performed | | | | | | |
| Ankle | 102 (37.5%) | | | 53 (37.3%) | | |
| Subtalar | 68 (25.0%) | | | 38 (26.7%) | | |
| Calcaneocuboid | 3 (1.1%) | | | 0 (0.0%) | | |
| Talonavicular | 15 (5.5%) | | | 9 (6.3%) | | |
| Double arthrodesis | 23(8.5%) | | | 12 (8.5%) | | |
| Triple arthrodesis | 61 (22.4%) | | | 30 (21.1%) | | |
| Surgery Site | | | | | | |
| Hindfoot | 170 (62.5%) | | | 88 (62.0%) | | |
| Ankle | 102 (37.5%) | | | 54 (38.0%) | | |
| Description of Injury/Deformity | | | | | | |
| Primary Arthritis | 91 (33.5%) | | | 56 (39.4%) | | |
| Rheumatoid Arthritis | 23 (8.5%) | | | 5 (3.5%) | | |
| Post-traumatic injury/deformity | 135 (49.6%) | | | 63 (44.4%) | | |
| Ankylosing spondylitis | 0 (0.0%) | | | 0 (0.0%) | | |
| Non-Specified | 23 (8.5%) | | | 18 (12.8%) | | |
| Comorbidities | | | | | | |
| Smoking history within last 5 years | 66 (24.3%) | | | 33 (23.2%) | | |
| Obesity (BMI >= 30 kg/m2) | 125 (46.0%) | | | 77 (54.2%) | | |
| Previous revision surgery1 | 63 (23.2%) | | | 32 (22.5%) | | |
| Diabetes history (type 1 or 2) | 31 (11.4%) | | | 19 (13.4%) | | |
| Other Factors | n | Mean | SD | n | Mean | SD |
| Age at surgery (years) | 272 | 55.9 | 14.5 | 142 | 57.6 | 13.4 |
| BMI (kg/m2) | 272 | 0.5 | 0.5 | 142 | 0.5 | 0.5 |
| Age of injury (weeks) | 170 | 266.6 | 468.8 | 88 | 325.5 | 464.5 |
| Baseline Functional Status | | | | | | |
| Foot Function Index (FFI) Total | 272 | 51.8 | 18.7 | 142 | 48.8 | 18.4 |
| AOFAS Total | 272 | 39.7 | 17.9 | 142 | 40.8 | 18.3 |
| SF12 PCS (Physical) | 272 | 30.9 | 9.0 | 142 | 31.5 | 9.3 |
| VAS - Fusion site pain | 242 | 52.9 | 29.3 | 128 | 49.3 | 28.0 |
| VAS- Weight bearing pain | 240 | 67.8 | 26.2 | 125 | 65.5 | 23.7 |

Note: Percent values are based on the number of treated subjects (N=414).
This includes any surgery at the revision site(s).

Augment® Bone Graft SSED

{21}

As part of FDA's review, the sponsor provided information on how patients were enrolled based on the absence or presence of a "bone defect." The sponsor assessed baseline radiographs for the presence of 16 parameters, which physicians would use to indicate the need for bone graft in ankle and hindfoot arthrodesis surgery as described in a survey article by Baumhauer, et al. The results of this review are included in Table 6.

Table 6: Radiographic Assessment of the Need for Graft Material

| Radiographic Parameters Observed Indicating Need for Graft Material | n | % |
| --- | --- | --- |
| Total number subjects with evaluable radiograph at baseline | 400 | 100.0 |
| Convexity/concavity mismatch of the articulating surfaces of the joint | 394 | 98.5 |
| Large surface areas to be fused | 374 | 93.5 |
| Irregular bony surfaces of joints to be fused | 285 | 71.2 |
| Evidence of potential incongruous apposition | 247 | 61.8 |
| Intra-articular deformity | 206 | 51.5 |
| Joint malalignment | 194 | 48.5 |
| Subchondral cysts | 143 | 35.8 |
| Radiographic evidence of bone loss | 125 | 31.3 |
| More than one joint to be fused | 119 | 29.8 |
| Osteoporosis or post-traumatic with subchondral collapse | 89 | 22.3 |
| Osseous defects resulting from pathology or traumatic injury to the bone | 64 | 16.0 |
| Extra-articular deformity | 49 | 12.3 |
| Bony step-offs | 19 | 4.8 |
| Prior adjacent joint fusions | 18 | 4.5 |
| Avascular necrosis (AVN) | 2 | 0.5 |
| | | |
| At least one radiologic parameter | 400 | 100.0 |
| At least two radiologic parameters | 396 | 99.0 |
| At least three radiologic parameters | 368 | 92.0 |
| At least four radiologic parameters | 332 | 83.0 |
| At least five radiologic parameters | 275 | 68.8 |

Of the 400 subjects with an evaluable baseline radiograph, 400 (100%) demonstrated at least 1 of the 16 radiographic findings that required bone graft to treat the subject. Three-hundred ninety six (99.0%) demonstrated at least 2 such findings, 368 (92.0%) demonstrated at least 3, and 332 (83.0%) demonstrated at least 4 radiographic findings.

The most common parameters observed (those seen in at least $50\%$ of subjects) were convexity/concavity mismatch of the articular surfaces $(394, 98.5\%)$ , large surface area to be fused $(374, 93.5\%)$ , irregular bony surfaces of the joint to be fused $(285, 71.3\%)$ , evidence of potential incongruous apposition $(247, 61.8\%)$ , and intra-articular deformity $(206, 51.5\%)$ .

Augment® Bone Graft SSED

{22}

D. Safety and Effectiveness Results

Safety Results

The safety of the investigational product was assessed using a separate analysis population at 24 weeks and was not part of the primary study endpoint. Safety was assessed by evaluating graft harvest site pain scores as the primary safety endpoint, and operating room time and surgical wound infection rate as secondary safety endpoints. Safety was also evaluated based on the nature and frequency of adverse events which occurred in the Augment® Bone Graft group, as compared to those that occurred in the autograft group.

Reported adverse events were classified as systemic and product-specific. The MedDRA was used to classify systemic adverse events. Product-specific complications were collected according to seven subgroups pre-defined by the sponsor's protocol: 1) "Pre-treatment signs and symptoms"; 2) "Treatment Emergent Adverse Events" (TEAEs) defined as AEs reported on or after the day of surgery; 3) "Complications" defined as complications associated with surgical procedures, a subset of the TEAEs; 4) "Serious Complications"; 5) Infections; 6) Related TEAEs; and 7) Serious TEAEs.

All Adverse Events

The adverse events, as shown in the tables below, are reported from the "Safety Population" which included 272 Augment® Bone Graft patients and 142 autograft control patients enrolled in the multi-center clinical study. Adverse event rates presented are based on the number of patients having at least one occurrence for a particular adverse event divided by the total number of patients in that treatment group.

A total of 212 (77.9%) of Augment® Bone Graft patients had at least one adverse event within 52 weeks versus 105 (73.9%) autograft control patients. A total of 657 events were reported in the Augment® Bone Graft patients and 316 events were reported in the controls. The 24-week data analysis was used as the primary effectiveness endpoint. The summary of AEs by System Organ Classification (SOC) and Preferred Term (PT) in either treatment group is provided in Table 7.

Table 7– Adverse Events Summary by MedDRA SOC and PT

| System Organ Class Preferred Term | All Patients (N=414) | | Augment® Bone Graft (N=272) | | Autologous Bone Graft (N=142) | |
| --- | --- | --- | --- | --- | --- | --- |
| | Subjects | Events | Subjects | Events | Subjects | Events |
| Any Adverse Event | 317 (76.6%) | 973 | 212 (77.9%) | 657 | 105 (73.9%) | 316 |
| Blood and lymphatic system disorders | 2 (0.5%) | 2 | 1 (0.4%) | 1 | 1 (0.7%) | 1 |
| Cardiac disorders | 9 (2.2%) | 10 | 3 (1.1%) | 3 | 6 (4.2%) | 7 |
| Congenital, familial and genetic disorders | 2 (0.5%) | 2 | 1 (0.4%) | 1 | 1 (0.7%) | 1 |
| Ear and labyrinth disorders | 3 (0.7%) | 3 | 1 (0.4%) | 1 | 2 (1.4%) | 2 |
| Endocrine disorders | 2 (0.5%) | 3 | 2 (0.7%) | 3 | 0 (0.0%) | 0 |

Augment® Bone Graft SSED
Page 23 of 51

{23}

| Eye disorders | 5 (1.2%) | 6 | 2 (0.7%) | 3 | 3 (2.1%) | 3 |
| --- | --- | --- | --- | --- | --- | --- |
| Gastrointestinal disorders | 52 (12.6%) | 66 | 35 (12.9%) | 45 | 17 (12.0%) | 21 |
| General disorders and administration site conditions | 56 (13.5%) | 61 | 37 (13.6%) | 40 | 19 (13.4%) | 21 |
| Hepatobiliary disorders | 1 (0.2%) | 1 | 1 (0.4%) | 1 | 0 (0.0%) | 0 |
| Immune system disorders | 12 (2.9%) | 13 | 10 (3.7%) | 11 | 2 (1.4%) | 2 |
| Infections and infestations | 89 (21.5%) | 121 | 61 (22.4%) | 86 | 28 (19.7%) | 35 |
| Injury, poisoning and procedural complications | 104 (25.1%) | 125 | 67 (24.6%) | 82 | 37 (26.1%) | 43 |
| Medical device pain | 21 (5.1%) | 21 | 14 (5.1%) | 14 | 7 (4.9%) | 7 |
| Investigations | 9 (2.2%) | 9 | 6 (2.2%) | 6 | 3 (2.1%) | 3 |
| Metabolism and nutrition disorders | 8 (1.9%) | 9 | 4 (1.5%) | 5 | 4 (2.8%) | 4 |
| Musculoskeletal and connective tissue disorders | 166 (40.1%) | 276 | 117 (43.0%) | 193 | 49 (34.5%) | 83 |
| Arthralgia | 53 (12.8%) | 63 | 38 (14.0%) | 46 | 15 (10.6%) | 17 |
| Pain in extremity | 69 (16.7%) | 80 | 48 (17.6%) | 56 | 21 (14.8%) | 24 |
| Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 7 (1.7%) | 7 | 5 (1.8%) | 5 | 2 (1.4%) | 2 |
| Nervous system disorders | 58 (14.0%) | 65 | 43 (15.8%) | 49 | 15 (10.6%) | 16 |
| Psychiatric disorders | 16 (3.9%) | 18 | 11 (4.0%) | 13 | 5 (3.5%) | 5 |
| Renal and urinary disorders | 28 (6.8%) | 29 | 17 (6.3%) | 17 | 11 (7.7%) | 12 |
| Reproductive system and breast disorders | 3 (0.7%) | 3 | 1 (0.4%) | 1 | 2 (1.4%) | 2 |
| Respiratory, thoracic and mediastinal disorders | 25 (6.0%) | 30 | 14 (5.1%) | 15 | 11 (7.7%) | 15 |
| Skin and subcutaneous tissue disorders | 61 (14.7%) | 69 | 41 (15.1%) | 47 | 20 (14.1%) | 22 |
| Surgical and medical procedures | 14 (3.4%) | 16 | 9 (3.3%) | 9 | 5 (3.5%) | 7 |
| Vascular disorders | 27 (6.5%) | 29 | 18 (6.6%) | 20 | 9 (6.3%) | 9 |

* Serious Adverse Events are defined by FDA's Medwatch Adverse Event program as any death, any life-threatening event (i.e., an event that placed the patient, in the view of the investigator, at immediate risk of death from the event as it occurred; this does not include an event that, had it occurred in a more severe form, might have caused death), any event that required or prolonged in-patient hospitalization, any event that resulted in persistent or significant disability/incapacity, any congenital anomaly/birth defect diagnosed in a child of a patient who participated in this study following the study procedure, any other medically important events that in the opinion of the investigator may have jeopardized the patient or may have required intervention to prevent one of the other outcomes listed above, or any serious problem associated with the device that related to the rights, safety or welfare of study patients.

There are five categories of adverse events in which the Augment® Bone Graft group is greater than or equal to two percentage points higher than the autograft control group: immune system disorders (3.7% vs 1.4%); musculoskeletal and connective tissue disorders (43.0% vs 34.5%); arthralgia (14.0% vs 10.6%); pain in extremity (17.6% vs 14.8%); and nervous system disorders (15.8% vs 10.6%). There are two categories of adverse events in which the autograft control group had a higher rate by two percentage points or more than the Augment® Bone Graft group: cardiac disorders (4.2% vs. 1.1%); and respiratory, thoracic and mediastinal disorders (7.7% vs. 5.1%). The correlation of high rates of pain measured as adverse events with secondary outcome measures for product effectiveness is unclear. Infections and infestations are higher in the investigational group by over 2 percentage points (22.4% vs. 19.7%). Although infections and infestations rates are similar rates, these rates are clinically concerning for hind foot and ankle arthrodesis. No inferential statistical comparison of adverse events between investigational and autograft control groups was performed.

Augment® Bone Graft SSED
Page 24 of 51

{24}

There were 1.8% (5/272) of adverse events in Augment® Bone Graft patients categorized as Neoplasms (i.e., cancers, either benign or malignant) and 1.4% (2/142) in autograft control patients. Please see the section “Observed Cancer Incidence in Pivotal Trials” for additional details.

## Serious Adverse Events

Table 8 summarizes the SAEs by treatment group and System Organ Class (SOC) for each visit and in total during the study for all 414 randomized and treated subjects (272 Augment® Bone Graft and 142 autograft subjects).

Table 8: SAEs by Treatment Group and System Organ Class (SOC)

| System Organ Class | Visit | | | | | | | | | | | | | | | Total events | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | Week 3 | | Week 6 | | Week 9 | | Week 12 | | Week 16 | | Week 24 | | Week 36 | | Week 52 | | |
| | (Day 1 to 31) | | (Day 32 to 52) | | (Day 53 to 73) | | (Day 74 to 97) | | (Day 98 to 139) | | (Day 140 to 209) | | (Day 210 to 307) | | (Day 308 to 421) | | |
| | I | C | I | C | I | C | I | C | I | C | I | C | I | C | I | C | I |
| Cardiac disorders | | 1 | | | | 2 | 1 | | | | | 1 | | | | 1 | 4 |
| Congenital, familial and genetic disorders | | | | | | | | | | | | 1 | | | | 0 | 1 |
| Gastrointestinal disorders | | 1 | 1 | | | | 1 | | | 1 | | | 1 | | 1 | 1 | 4 |
| General disorders and administration site conditions | 1 | | | | 1 | | | | 1 | | | | 2 | | 1 | | 6 |
| Infections and infestations | | | | | | | 2 | 1 | 2 | 3 | 1 | 1 | | | 2 | | 7 |
| Injury, poisoning and procedural complications | | 1 | | 1 | 1 | | | | | | 1 | | 1 | | | 3 | 2 |
| Investigations | | | | | | | | | | | 1 | | | | | 1 | 0 |
| Musculoskeletal and connective tissue disorders | 1 | | 1 | | | | | | | | 3 | 2 | 2 | | | 7 | 2 |
| Neoplasms benign, malignant and unspecified (incl cyst | | 1 | | | | | | | 1 | | 1 | | 1 | 1 | | 3 | 2 |
| Nervous system disorders | 1 | | | | | | | | | | 1 | | | | | 2 | 0 |
| Psychiatric disorders | | | | | | | | | | 1 | | | 1 | | | 1 | 1 |
| Respiratory, thoracic and mediastinal disorders | 1 | 2 | | | | | | | | | | 1 | | 1 | | 1 | 4 |
| Surgical and medical procedures | | | | | | | | | 1 | | | | | | | 1 | 0 |
| Vascular disorders | 2 | 2 | | 1 | 1 | 1 | 2 | | 1 | | 1 | | | | 1 | 1 | 8 |

Augment® Bone Graft SSED
Page 25 of 51

{25}

| Total events | 6 | 8 | 2 | 2 | 3 | 3 | 6 | 1 | 6 | 5 | 9 | 6 | 8 | 2 | 5 | 2 | 45 | 29 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |

Note:
I= Investigational (Augment), C=Control (autograft)

As the above table shows, a total of 74 SAEs occurred during the study, of which 45 were in Augment® Bone Graft subjects and 29 were in autograft subjects. Twenty eight of the 272 (10.3%) Augment® Bone Graft subjects experienced the 45 SAEs in that treatment group while 21 of the 142 (14.8%) autograft subjects experienced the 29 SAEs in that treatment group.

There were three Augment® Bone Graft subjects who were withdrawn from the study due to SAEs. There were no autograft control subjects withdrawn due to SAEs. The three Augment® Bone Graft subjects withdrew for the following reasons: one infection was noted during surgery and the fusion procedure was not performed and no graft material was implanted, one death due to pulmonary embolism, and one bilateral MRSA infection of both knees.

Nine surgical wound infections were classified as SAEs: four infections in Augment® Bone Graft subjects and five infections in autograft control subjects.

All but five of the subjects (three Augment® Bone Graft and two autograft control subjects) who experienced SAEs were reported as recovered/resolved.

The table of treatment-emergent SAEs by SOC and PT is presented in Table 9 below. Defined: WHO Grade 3 or 4; N=272 Augment® Bone Graft patients and N=142 autograft controls.

Table 9: Treatment-Emergent SAEs by SOC and PT

| System Organ Class (Preferred Term) | All Patients (N=414) | | Augment® Bone Graft (N=272) | | Autologous Bone Graft (N=142) | |
| --- | --- | --- | --- | --- | --- | --- |
| | Subjects | Events | Subjects | Events | Subjects | Events |
| Any Adverse Event | 49(11.8%) | 74 | 28(10.3%) | 45 | 21(14.8%) | 29 |
| | | | | | | |
| Cardiac disorders | 4( 1.0%) | 5 | 1( 0.4%) | 1 | 3( 2.1%) | 4 |
| Acute myocardial infarction | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| Atrial flutter | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |
| AV block complete | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| Cardiac failure congestive | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| Myocardial infarction | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| Congenital, familial and genetic | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| Congenital foot malformation | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| Gastrointestinal disorders | 6( 1.4%) | 7 | 3( 1.1%) | 4 | 3( 2.1%) | 3 |
| Gastritis | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |
| Gastrointestinal haemorrhage | 4( 1.0%) | 5 | 2( 0.7%) | 3 | 2( 1.4%) | 2 |
| Megacolon | 1( 0.2%) | 1 | 0( 0.0%) | 0 | 1( 0.7%) | 1 |
| General disorders and administration site conditions | 6( 1.4%) | 6 | 6( 2.2%) | 6 | 0( 0.0%) | 0 |
| Chest pain | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |
| Impaired healing | 2( 0.5%) | 2 | 2( 0.7%) | 2 | 0( 0.0%) | 0 |
| Non-cardiac chest pain | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |
| Pyrexia | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |
| Cardiac chest pain | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |
| Infections and infestations | 9( 2.2%) | 12 | 5( 1.8%) | 7 | 4( 2.8%) | 5 |
| Cellulitis | 1( 0.2%) | 1 | 1( 0.4%) | 1 | 0( 0.0%) | 0 |

Augment® Bone Graft SSED
Page 26 of 51

{26}

| Clostridium difficile colitis | 1 (0.2%) | 1 | 0 (0.0%) | 0 | 1 (0.7%) | 1 |
| --- | --- | --- | --- | --- | --- | --- |
| Infection | 1(0.2%) | 1 | 1 (0.4%) | 1 | 0 (0.0%) | 0 |
| Osteomyelitis | 3(0.7%) | 3 | 1 (0.4%) | 1 | 2 (1.4%) | 2 |
| Pneumonia | 2(0.5%) | 2 | 2 (0.7%) | 2 | 0 (0.0%) | 0 |
| Postoperative wound infection | 2(0.5%) | 2 | 1 (0.4%) | 1 | 1 (0.7%) | 1 |
| Staphylococcal infection | 2(0.5%) | 2 | 1 (0.4%) | 1 | 1 (0.7%) | 1 |
| Injury, poisoning and procedural complications | 5 (1.2%) | 5 | 3(1.1%) | 3 | 2(1.4%) | 2 |
| Device related infection | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Medical device complication | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Overdose | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Postoperative wound infection | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Wound infection staphylococcal | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Investigations | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Prothrombin level abnormal | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Musculoskeletal and connective tissue | 9 (2.2%) | 9 | 7(2.6%) | 7 | 2(1.4%) | 2 |
| Foot fracture | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Joint instability | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Joint range of motion decreased | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Muscle strain | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Osteoarthritis | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Osteoporosis | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Pain in extremity | 3 (0.7%) | 3 | 3(1.1%) | 3 | 0(0.0%) | 0 |
| Neoplasms benign, malignant and unspecified (incl cyst) | 5 (1.2%) | 5 | 3(1.1%) | 3 | 2(1.4%) | 2 |
| Endometrial cancer | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Lung neoplasm malignant | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Prostate cancer | 2 (0.5%) | 2 | 2(0.7%) | 2 | 0(0.0%) | 0 |
| Renal cell carcinoma stage unspecified | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Nervous system disorders | 2 (0.5%) | 2 | 2(0.7%) | 2 | 0(0.0%) | 0 |
| Cerebrovascular accident | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Convulsion | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Psychiatric disorders | 2 (0.5%) | 2 | 1(0.4%) | 1 | 1(0.7%) | 1 |
| Alcohol withdrawal syndrome | 1 (0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Suicide attempt | 1 (0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Respiratory, thoracic and mediastinal | 4(1.0%) | 5 | 1(0.4%) | 1 | 3(2.1%) | 4 |
| Atelectasis | 1(0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Chronic obstructive pulmonary disease | 1(0.2%) | 2 | 0(0.0%) | 0 | 1(0.7%) | 2 |
| Hypoxia | 1(0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Pulmonary embolism | 1(0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Surgical and medical procedures | 1(0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Osteotomy | 1(0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |
| Vascular disorders | 12(2.9%) | 13 | 7(2.6%) | 8 | 5(3.5%) | 5 |
| Aortic stenosis | 1(0.2%) | 1 | 0(0.0%) | 0 | 1(0.7%) | 1 |
| Deep vein thrombosis | 7(1.7%) | 7 | 4(1.5%) | 4 | 3(2.1%) | 3 |
| Pulmonary embolism | 4(1.0%) | 4 | 3(1.1%) | 3 | 1(0.7%) | 1 |
| Thrombosis | 1(0.2%) | 1 | 1(0.4%) | 1 | 0(0.0%) | 0 |

Highlighted=higher by >1 percentage point in the investigational group (yellow) and autograft control group (blue). See AE discussion above for further information.

There were eight categories where rates differed by greater than one percentage point:

- Augment® Bone Graft higher: general disorders and administrative site conditions, which included chest pain, non-cardiac chest pain, cardiac chest pain, impaired healing and pyrexia (2.2% vs 0%); and musculoskeletal and connective tissue disorders, which includes foot fracture, joint instability, joint range of motion decreased, and muscle strain (2.6% vs 1.4%); and pain in extremity (1.1% vs 0%).

Augment® Bone Graft SSED
Page 27 of 51

{27}

- Autograft control higher: cardiac disorders (2.1% vs 0.4%); GI disorders (2.1% vs 1.1%); infections and infestations (2.8% vs 1.8%); osteomyelitis (1.4% vs 0.4%); and respiratory, thoracic and mediastinal disorders (2.1% vs 0.4%).

The median time to first SAE was 109 days for Augment® Bone Graft and 64 days for autograft $(p = 0.249$ based on Wilcoxon-Gehan test statistic of 1.33 for comparing equality of time to first SAE in subjects who experienced an SAE). The analysis did not detect any significant differences between the treatment groups in the number of subjects with SAEs, or the time to first SAE in those subjects experiencing an SAE.

# Detailed Information on Specific Adverse Event Categories

# Infection Rates

Table 10 summarizes the incidence of fusion-related, procedure-related, graft harvest site related, and other-site related infections by treatment group at 24, 36, and 52 weeks after implantation of the graft material.

Table 10: Incidence of Fusion-Related, Procedure, Graft Harvest Site, and Other Site Related Infections by Treatment Groups at 24, 36, and 52 Weeks After Implantation

| Site | Total events through week 24 | | Total Patients through week 24 (%) | | p-value | Total events through week 52 | | Total Patients through week 52 (%) | | p-value |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | I | C | I | C | | I | C | I | C | |
| Fusion site related* | 19 | 17 | 15(5.5) | 12(8.5) | 0.2951 | 24 | 17 | 19(7.0) | 12(8.5) | 0.6944 |
| Procedure site related* | 24 | 19 | 20(7.4) | 14(9.9) | 0.4509 | 29 | 20 | 24(8.8) | 15(10.6) | 0.5968 |
| Graft harvest site related* | 0 | 2 | | 2(1.4) | 0.1171 | | 2 | | 2(1.4) | 0.1171 |
| Other site related* | 52 | 18 | 33(12.1) | 16(11.3) | 0.8734 | 65 | 25 | 40(14.7) | 20(14.1) | >0.999 |
| All Infections* | 78 | 38 | 55(20.2) | 31(21.8) | 0.7035 | 96 | 46 | 66(24.3) | 36(25.4) | 0.8111 |

Note:
I-Investigational (Augment, $N = 272$ ), C-Control (Autologous Bone Graft, $N = 142$ )
p-value of two-tailed Exact Fisher's test using patient counts.
* a patient may have more than one infection event and an infection event may be classified into more than one site category.

As table 10 above shows, fusion site related, procedure site related, and graft harvest site related infections demonstrated no significant differences between treatment groups. Augment® Bone Graft reported a higher infection rate for “other” site related infections (e.g., sinus infection, thrush, UTI), which were not related to arthrodesis surgery. When looking at all infections, infection and infestation rates between the two groups were similar (Augment® Bone Graft, $20.2\%$ and autograft control, $18.3\%$ ). Although infections and infestations rates are similar rates, these rates are clinically concerning for hind foot and ankle arthrodesis.

Augment® Bone Graft SSED

{28}

Graft Harvest Site Pain

Subjects in the autologous bone graft group report clinically significant pain at the graft harvest site (≥ 20 mm) on VAS at and after the week 24 visit: 12.4% of autologous bone graft subjects at week 24 and 8.8% at week 52. A breakdown of the different anatomical areas from which graft material was obtained showed that the iliac crest with a higher morbidity constituted only 11.7% of all site materials used. Distal tibia (16.1%) and calcaneous (13.9%) were also used. The remaining autograft subjects u…