rh BMP-2 soaked Absorbable Collagen Sponge ACS for

rh. BMP-2 soaked Absorbable Collagen Sponge (ACS) for the treatment of Open Tibial Shaft Fractures Clinical Summary Points to consider

Clinical Summary of International Investigation

Clinical Review Study Design Effectiveness Safety

Clinical Review STUDY DESIGN • • Confounding Variables Patient Assessment Study Endpoints Data Analysis

$Confounding Variables • Technique for IM nail insertion • All fracture types considered equal$

Confounding Variables • Technique for IM nail insertion • All fracture types considered equal • Isolated fractures grouped with multiple injuries • Gustilo I ≠ Gustilo IIIA • Not all patients received a full sponge to the fracture site • Large centers ≠ small centers

Distribution of Patients by Country # sites Control . 75 mg/ml 1. 5 mg/ml Total Australia 6 18 18 12 48 Belgium/ Netherlands 8 8 12 12 32 Canada 4 5 10 15 30 France 4 8 6 11 25 Germany 12 35 22 26 83 Israel 4 5 13 10 28 Finland/ Norway 3 13 8 8 29 So. Africa 5 47 50 41 138 U. K. 3 11 12 14 37 TOTALS 49 150 151 149 450

Clinical Aspects of Pooling • Multiple centers • Over 50% of patients from 2 countries • Different experiences: • Few patients from many sites pooled with many patients from few sites • Different philosophies regarding optimal treatment : reamed vs. unreamed nails • Different interpretations of healing, delayed healing • Cultural and Geographic differences • Applicability to US trauma populations

Assessment Methods Clinical ØFracture site tenderness ØRadiographic union ØWeight bearing status Independent Radiographic

$Patient Assessment PAIN • No scale used for comparison • Differentiation between fracture site$

Patient Assessment PAIN • No scale used for comparison • Differentiation between fracture site tenderness and soft tissue injury difficult

$Criteria For Radiographic Union for Independent Radiology Panel A fracture was considered united when:$

Criteria For Radiographic Union for Independent Radiology Panel A fracture was considered united when: • 3 of 4 cortices demonstrated cortical bridging and/or complete disappearance of fracture lines • This definition includes: • 3 of 4 cortices demonstrate bridging • 3 of 4 cortices demonstrated disappearance of fracture lines • 2 of 4 cortices demonstrate cortical bridging and at least 1 of the remaining 2 cortices demonstrate disappearance of fracture lines • 1 of 4 cortices demonstrated cortical bridging and at least 2 of the remaining 3 cortices demonstrated disappearance of fracture lines

Independent and Investigator Review Gustillo Grade IIIA Unreamed locked nail 0. 75 mg/ml rh. BMP

$Definitions Healed Fracture: • Absence of tenderness upon manual palpation of the fracture site$

Definitions Healed Fracture: • Absence of tenderness upon manual palpation of the fracture site • Radiographic fracture union as assessed by the investigator • Full weight bearing status

$Definitions Delayed Union: “A fracture is considered a delayed union if insufficient fracture healing$

Definitions Delayed Union: “A fracture is considered a delayed union if insufficient fracture healing was observed as determined by the investigators radiographic and clinical assessment”

Number of Patients with SI recommended & Patients meeting criteria of Delayed union SOC 41 0. 75 28 Total w/SI # of criteria met 3 criteria 12 18 1. 5 24 Total 93 7 24 (26%) 2 criteria 19 18 11 48 (52%) 1 criterion 10 5 6 21 (23%)

Primary Endpoint Secondary Interventions : How was the decision made?

Study Design: Control Group What is the standard of Care? • Depends on ØFracture type ØInjury severity ØBone loss ØContamination ØConcomitant injuries Different prognoses for different types

Relevance of Endpoints Primary Endpoint • Rate of Secondary Interventions – – Recommended & Performed Recommended & Not Performed Not Recommended but Performed Self Dynamizations (screw breakages) Secondary Endpoints • Healing rate at 6 months • 50% probability of healing • CCRE

What is important? • How many healed? • What is the incidence of nonunion? • What are the complications and incidence? • Incidence of Infection? • Time to healing for majority of the patients?

Combined Clinical & Radiographic Endpoint (CCRE) • Independent review paired with investigator review • Clinical assessment compared to purely radiographic assessment • Patients with SI evaluated differently than patients without SI

Treatment of Missing Data • Inconsistent • Three examples

The Dilemma • Investigators unblinded • Investigators determined pain, weight bearing status and radiologic healing • Investigators determined when to perform secondary intervention • The CCRE is 50% dependent on investigators determination

The Dilemma • No time course/interval to delineate “delayed healing” from “healing” • No radiographic/clinical criteria to separate healing fracture vs. delayed healing • How patients with delayed healing fractures were recommended for secondary intervention is imprecise. • Extent to which all the investigators used the same criteria for determining a secondary intervention is unknown

Results Effectiveness • • • Primary Endpoint Rate of Fracture Healing Time to event Analysis Probability of 50% healing Nonunion

Primary Endpoint Standard 0. 75 mg/ml of Care rh. BMP-2/ ACS group 1. 5 mg/ml rh. BMP-2/ ACS group SI 66 (44%) 51 (35%) 38 (26%) Recommended & Performed 38 (25%) 25 (17%) 19 (13%) Exclude Self dynamization 31% 25% 21% No SI 84 (56%) 98 (65%) 111 (74%)

Rate of Fracture Healing Investigator 26 wks 39 wks 50 wks Radiology Panel SOC 0. 75 1. 5 mg/ml 36 % 40 % 55 % 20 % 25 % 33 % 48 % 51 % 64 % 38 % 41 % 50 % 51 % 58 % 70 % 47 % 52 % 64 %

Probability of a Fracture Healing by Investigator

Time To Healing by Investigator Treatment Days to Healing Probability of Healing Standard of Care 141 25% 184 50% 275 75% 128 25% 187 50% 273 75% 102 25% 147 50% 273 75% 0. 75 mg/ml rh. BMP 2/ACS 1. 5 mg/ml rh. BMP 2/ACS

Time to Independent Radiographic Assessment of Fracture Union

Time to Fracture Healing by Independent Radiology Panel Treatment Days to Healing Probability of Healing Standard of Care 189 25% 275 50% 351 75% 186 25% 272 50% 351 75% 182 25% 271 50% 353 75% 0. 75 mg/ml rh. BMP 2/ACS 1. 5 mg/ml rh. BMP 2/ACS

Nonunion SOC Overall 12 months Patients w/ SI 0. 75 mg/ml 1. 5 mg/ml Inv Rad 49% 53% 42% 48% 30% 38% 12% 9% 13%

Time to Fracture healing by Investigator Assessment : Patients with Secondary Interventions

Safety Serology • Anti-rh. BMP antibodies • Anti Type I Bovine collagen antibodies Hardware failure Laboratory Results Heterotopic Ossification Infection

Serology Antibodies formed to: Standard of 0. 75 mg/ml Care rh. BMP 2/ACS 1. 5 mg/ml rh. BMP 2/ACS rh. BMP-2 1 (1%) 3 (2%) 9 (6%) Bovine Type I Collagen 9 (6%) 22 (16%) 29 (20%)

Hardware Failure Total Patients SOC 32 (22%) . 75 mg/ml 25 (17%) 1. 3 mg/ml 16 (11%) Total Events 48 33 25 Nail breakage 0 2 0 Screw Breakage 48 31 24 Pts reamed Nail screw breakage 7/39 (18%) 8/48 (17%) 2/59 (3. 4%) Pts unreamed nail screw breakage 25/108 17/97 14/86 (23%) (18%) (16%)

Laboratory results Liver function Tests elevated in rh. BMP -2 treated groups • Elevated Amylase • Hypomagnesemia

Heterotopic Ossification Patients SOC 0. 75 mg/ml 1. 5 mg/ml 4 5 8 Hypertrophic Callus 3 2 1 Soft tissue callus 0 1 2 HO of tibia fracture 0 1 0 Synostosis 1 0 1

Infection Rate SOC . 75 mg/ml 1. 5 mg/ml Total 30% 23% 25% Leg/Tibia 29% 25% 24% Gustilo IIIA/B 18% 20% 48% 17% 19% 30% 19% 20% 33%

CONCLUSIONS • • • Definitions for assessment unclear Assessments based on investigators Clinical relevance of endpoints Control group as standard of care an issue Pooling across different sites and applicability to US population an issue • Outcomes Interpretations differ • Safety questions