Product: (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 2 of 17187

2. SYNOPSIS Name of Sponsor: Inc., Thousand Oaks, CA Name of Finished Products: Ganitumab and conatumumab Name of Active Ingredients: AMG 479 and AMG 655 Title of Study: A Phase 2, Randomized, Double-blind, Placebo-controlled Study Evaluating the Safety and Efficacy of FOLFIRI in Combination With AMG 479 or AMG 655 Versus FOLFIRI for the Second-line Treatment of KRAS-mutant Metastatic Colorectal Carcinoma Investigators and Study Centers: This study was conducted at a total of 49 centers in 10 countries (France, Hong Kong, Hungary, India, Italy, Poland, Russian Federation, Singapore, Spain, and United States). Study centers and investigators are listed in Appendix 4. Publications: None Study Period: This clinical study report includes results from the date of the first subject enrolled (31 March 2009) to the primary analysis data cutoff date (31 January 2011). Results after the primary analysis data cutoff date will be reported separately. Development Phase: 2

Introduction and Objectives: Ganitumab, a fully human monoclonal IgG1 antibody against human insulin-like growth factor receptor type 1 (IGF-1R), is being developed as an anticancer drug to provide inhibitory effects on tumor growth and invasion in combination with standard cancer therapy, molecularly targeted therapy, or both. Ganitumab exerts its antitumor activity by blocking ligand binding (insulin-like growth factor-1 [IGF-1] and insulin-like growth factor-2 [IGF-2]) and inducing receptor internalization and degradation without cross-reacting with the insulin receptor. Evidence from nonclinical xenograft models of colorectal cancer shows that ganitumab is active as monotherapy and in combination with and other targeted agents commonly used in colon cancer treatment (eg, irinotecan, 5-fluorouracil [5-FU], and ). Conatumumab, a fully human agonist (immunoglobulin type G1) targeting human death-receptor 5 (DR5), is being developed as an anticancer drug. Conatumumab mimics endogenous (TNF)-related -inducing ligand (TRAIL), binds DR5, and induces apoptosis in sensitive cells. Conatumumab demonstrates dose-dependent activity against colorectal cancer xenografts in vivo. The in vivo activity of conatumumab also is significantly enhanced when used in combination with 5-FU and irinotecan. Approved The primary objective was to estimate the treatment effect on progression-free survival (PFS) of conatumumab plus irinotecan/leucovorin/5-fluorouracil (5-FU) (FOLFIRI) or ganitumab plus FOLFIRI relative to FOLFIRI alone when administered as a second-line treatment for subjects with metastatic colorectal cancer (mCRC) whose tumors express mutant Kirsten rat sarcoma virus oncogene homolog (KRAS). The secondary objectives were: • To estimate treatment effect on overall survival, objective response rate (complete response + partial response), rates of disease control (ie, complete response + partial response + stable disease), duration of response, and time to response • To evaluate the incidence of adverse events and significant laboratory abnormalities and the incidence of anti-conatumumab or anti-ganitumab antibody formation • To evaluate treatment effect on patient-reported outcomes (PROs) using the European Organization for Research and Treatment of Cancer (EORTC) quality of life questionnaire-core questionnaire (QLQ-C30) and the Functional Assessment of Cancer Therapy (FACT)/National Cancer Comprehensive Network (NCCN)-Colorectal Symptom Index (FCSI) • To evaluate pharmacokinetics of conatumumab, ganitumab, and FOLFIRI components (irinotecan and 5-FU), and estimate the impact of administration of conatumumab and ganitumab on the pharmacokinetics of FOLFIRI components (irinotecan and 5-FU)

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 3 of 17187

• To correlate treatment outcomes with tumor tissue analysis of somatic gene mutations that regulate the IGF-1R and death receptor (DR) pathways and other genes that are known to be involved in tumorigenesis

Methodology: This is an ongoing, phase 2, multicenter, randomized, double-blind, double-dummy, placebo-controlled study evaluating the safety and efficacy of FOLFIRI in combination with ganitumab or conatumumab relative to FOLFIRI alone for the second-line treatment of mutant KRAS mCRC. Eligible subjects were randomized in a 1:1:1 ratio to second-line therapy consisting of either: • Arm A: Ganitumab-placebo plus conatumumab 10 mg/kg in combination with FOLFIRI every 14 days • Arm B: Ganitumab 12 mg/kg plus conatumumab-placebo in combination with FOLFIRI every 14 days • Arm C: Ganitumab-placebo plus conatumumab-placebo in combination with FOLFIRI every 14 days Randomization was stratified by Eastern Cooperative Oncology Group (ECOG) performance status (0 versus 1) and previous exposure to anti-vascular endothelial growth factor (VEGF) therapy in first-line treatment (yes versus no). Cycles of chemotherapy plus investigational products (ganitumab, conatumumab, ganitumab-placebo, and conatumumab-placebo) were to be administered every 14 days (± 3 days) (not including the time to recover from toxicities). Subjects were permitted to receive protocol-specified therapy until disease progression (by modified Response Evaluation Criteria in Solid Tumors [RECIST version 1.0] guidelines or clinical progression), unacceptable toxicities, withdrawal of consent, death, or until 36 months after the last subject had been randomized. In the event the administration of any component of the FOLFIRI chemotherapy was discontinued for any reason other than disease progression, the other components of the FOLFIRI chemotherapy were to continue along with the investigational products. Subjects who became intolerant to all components of FOLFIRI chemotherapy were to continue investigational products until disease progression (radiographic per RECIST or clinical progression), unacceptable toxicities, withdrawal of consent, death, or until 36 months after the last subject has been randomized. Subjects who became intolerant to investigational products were to continue to

receive FOLFIRI at the discretion of the investigator until disease progression (radiographic per Approved RECIST or clinical progression), unacceptable toxicities, withdrawal of consent, death, or until 36 months after the last subject had been randomized. Subjects who remained on protocol-specified therapy 36 months after the last subject was randomized could be eligible for continued treatment with open-label ganitumab or conatumumab by protocol extension or as provided for by the local country’s regulatory mechanism. Subjects who discontinued protocol-specified therapy for disease progression or other reasons (except death) were followed for safety for 30 (+3) days, and for antibodies and pharmacokinetic assessment of ganitumab and conatumumab for 30 (+3) days and 60 (+14) days after the last study administration of protocol-specified therapy. Subjects were followed for survival every 3 months (±14 days) starting from the date of the last administration of protocol-specified therapy until death or 36 months after last subject had been randomized, whichever was earlier. Any subject who discontinued protocol-specified therapy before radiographic disease progression or death continued to have radiological imaging performed at week 6, week 12, and every 8 weeks (± 7 days) thereafter during the long-term follow-up period. This procedure was used to assess disease status until disease progression (radiographic per RECIST or clinical progression), start of a new treatment, death, withdrawal of consent, administrative decision, or the end of the study, whichever was earlier. An internal Amgen data review team (DRT) was established to review safety and pharmacokinetic data. Safety interim reviews were planned to occur after the first 18 and 36 subjects were randomized, had received at least 1 dose of investigational products (ganitumab, conatumumab, ganitumab-placebo, or conatumumab-placebo) in combination with FOLFIRI, and had the opportunity to complete 2 treatment cycles. Based on the length of time

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 4 of 17187 between the second DRT review and the planned primary analysis for the study, the DRT conducted an additional safety review after the 100th subject received treatment.

Number of Subjects Planned: Approximately 150 subjects Number of Subjects Enrolled: 155 subjects Diagnosis and Main Criteria for Eligibility: The main inclusion criteria for subjects in this study included men or women ≥ 18 years of age with histologically confirmed adenocarcinoma of the colon or rectum with metastatic disease; mutant KRAS tumor status confirmed by central laboratory assessment; 1 prior anticancer therapy regimen for metastatic disease consisting of the combination of a fluoropyrimidine and oxaliplatin-based chemotherapy with or without anti-VEGF therapy and documented disease progression ≤ 6 months after the last dose of this prior anticancer therapy; measurable or nonmeasurable disease according to modified RECIST; ECOG performance status of 0 or 1; and adequate hematologic, renal, and hepatic function. A complete list of inclusion/exclusion criteria is provided in Section 7.5. Investigational Product, Dose and Mode of Administration, Manufacturing Batch Number: Ganitumab, ganitumab-placebo, conatumumab, and conatumumab-placebo were considered investigational products in this study. Ganitumab and conatumumab were provided as a sterile, clear, colorless protein solution. Each glass vial contained 3.0 mL of investigational product with a concentration of 30 mg/mL (90 mg/vial). The listings of lot numbers by subject for all investigational products are provided in Appendix 18.

The dose of ganitumab was 12 mg/kg once every 2 weeks (Q2W) and the dose of conatumumab was 10 mg/kg Q2W. Investigational products were administered sequentially by intravenous (IV) infusion on day 1 of each cycle just before the administration of FOLFIRI chemotherapy. Reference Therapy, Dose and Mode of Administration, Manufacturing Batch Number: The FOLFIRI chemotherapy regimen (irinotecan with infusional 5-FU and leucovorin) was administered on day 1 of each treatment cycle after the administration of investigational products. Irinotecan (180 mg/m2) was administered over approximately 90 minutes on day 1 of each cycle. Leucovorin (400 mg/m2) was administered over approximately 2 hours during the irinotecan infusion but without mixing (lines could be connected at the IV port closest to the subject), immediately followed by a 5-FU bolus (2,400 mg/m2) administered over approximately 2 to 4 minutes and a 5-FU continuous IV infusion administered via ambulatory pump over approximately 46 to 48 hours. Lot numbers were not collected.

Duration of Treatment: The expected median length of treatment was approximately 4 months. Approved Subjects were treated with second-line treatment on study until one of the following occurred: disease progression (by modified RECIST criteria or clinical progression), unacceptable toxicities, withdrawal of consent, death, or until 36 months after the last subject had been randomized. Subjects were followed up for survival for up to 36 months from the date that the last subject had been randomized. The expected maximum duration of the study was approximately 56 months (ie, from the first subject randomized until approximately 36 months from the last subject randomized).

Study Endpoints: Primary Endpoint • Progression-free survival (defined as the number of days from the date of randomization to the first observation of disease progression [per modified RECIST or clinical progression, whichever occurred first], or death due to any cause, or date of censoring)

Secondary Endpoints Efficacy • Overall survival (defined as the number of days from the date of randomization to death from any cause, or if applicable, date of censoring) • Objective response rate (defined as proportion of subjects with either a confirmed complete response or a confirmed partial response as determined by modified RECIST criteria and as evaluated by the investigator)

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 5 of 17187

• Rate of disease control (defined as the number of subjects with a best overall response of stable disease or better [based on modified RECIST] divided by the number of subjects in the full analysis set with baseline measurable disease) • Duration of response (defined as the number of days between the date of the first tumor response assessment with an outcome indicating an objective response through to the subsequent date of progression [based on modified RECIST, or clinical progression, whichever occurred first] or death due to any cause, or where applicable, date of censoring) • Time to response (defined as time from the date of randomization to the first observation of an objective response in the subset of subjects who responded) Pharmacokinetic

• Pharmacokinetic parameters (observed maximum concentration [Cmax] and observed minimum concentration [Cmin]) for conatumumab and ganitumab

• Pharmacokinetic parameters (Cmax, Cmin, and area under the curve [AUC]) for irinotecan and SN-38 • Pharmacokinetic parameters (concentrations over an approximately 48-hour infusion period) for 5-FU Patient-reported Outcome (PROs) • Changes (improvement or worsening) in PROs as assessed using the EORTC QLQ-C30 and the FCSI Safety • Incidence of adverse events and significant laboratory abnormalities • Incidence of anti-conatumumab or anti-ganitumab antibody formation Biomarker • Tumor tissue analysis of somatic gene mutations that function to regulate the IGF-1R and/or apoptosis pathways and other genes that are known to be involved in the development and progression of solid tumors and correlation with treatment outcomes

Statistical Methods: The primary efficacy analysis was event-driven and was to occur when 100 subjects had experienced a PFS event (radiological disease progression or clinical Approved progression, whichever occurred first) or death. The statistical analyses were descriptive and noninferential. The efficacy analysis was conducted on the full analysis set. For PFS and overall survival, the proportionality between treatment arms was assessed via examination of Kaplan-Meier survival and log(-log) survival curves. The primary method of analysis used a stratified Cox’s proportional hazards model, and the estimated hazard ratio was presented with associated 80% and 95% confidence intervals (CIs) for ganitumab relative to placebo and conatumumab relative to placebo. Objective response rate and disease control rate were summarized descriptively with corresponding 95% CIs using the method described by Clopper and Pearson (1934). Duration of response was summarized for subjects with an objective response using Kaplan-Meier time-to-event curves with 95% CI estimates. Time to response was summarized using summary statistics. Serum ganitumab and conatumumab, and plasma 5-FU, irinotecan, and SN-38 (the active metabolite of irinotecan) concentrations were summarized with descriptive statistics.

The effect of FOLFIRI on ganitumab or conatumumab pharmacokinetics was investigated by comparing the concentration exposures in the current study to those in the respective first-in-human study for each investigational product. For PROs, the scores generated from the EORTC QLQ-C30 and FCSI were summarized descriptively for the following scales, as well as change from baseline, at each scheduled assessment by treatment arm: the 5 functional scales, 3 multi-item symptom scales,

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 6 of 17187

6 single-item symptoms scales, the health-related quality of life (HRQOL) scale of the EORTC QLQ-C30, and the FCSI score. Safety endpoints were analyzed using the safety analysis set, which included all randomized subjects who received at least 1 dose of ganitumab, conatumumab, or placebo; subjects in this analysis set were analyzed according to the treatment received. Adverse events were tabulated by system organ class and preferred term using the Medical Dictionary for Regulatory Activities (MedDRA) version 13.1. The subject incidence of all treatment-emergent adverse events, serious adverse events, adverse events leading to withdrawal of investigational product, and fatal adverse events was summarized. Adverse events of interest were summarized separately for ganitumab (drug-related hepatotoxicity, hyperglycemia, infusion reaction, neutropenia, rash, thrombocytopenia, venous thromboembolic events, and sensorineural hearing loss) and for conatumumab (hypomagnesemia, hyponatremia, infusion reactions, and venous thromboembolic events). Clinical laboratory parameters and vital signs were summarized using descriptive statistics or shift tables. The incidence of subjects developing anti-ganitumab and anti-conatumumab antibodies was determined.

Summary of Results: Subject Disposition: A total of 155 subjects were randomized into the study. Of these subjects, 52 were randomized to receive ganitumab, 51 subjects were randomized to receive conatumumab, and 52 were randomized to receive placebo. As of the primary analysis cutoff date, 75%, 71%, and 73% of the subjects in the ganitumab, conatumumab, and placebo treatment groups, respectively, discontinued investigational product. The 2 most common reasons for withdrawal from investigational product were disease progression (ganitumab: 58%, conatumumab: 49%, placebo: 63%) and adverse event (10%, 12%, 2%). Forty-two percent, 37%, and 42% of the subjects in the ganitumab, conatumumab, and placebo treatment groups, respectively, discontinued the study. The 2 most common reasons for study withdrawal were death (ganitumab: 31%, conatumumab: 33%, placebo: 35%) and full consent withdrawn (8%, 4%, 4%). Baseline Demographics: Sex: 74 men (48%); 81 women (52%) Age: mean (standard deviation [SD]) 57.5 (10.5) years (range: 28 to 81) Ethnicity/Race: 119 (77%) white; 9 (6%) black; 25 (16%) Asian; 2 (1%) Hispanic or Latino Approved Efficacy Results: All efficacy evaluations are presented separately for ganitumab and conatumumab. Ganitumab: The primary analysis of PFS, specified as a stratified Cox model (stratification factors: ECOG performance status and prior anti-VEGF exposure) comparing ganitumab with placebo, provided a hazard ratio estimate (95% CI) of 1.01 (0.61, 1.66) (p = 0.998, stratified log-rank test). The median PFS time was 4.5 months in the ganitumab treatment group and 4.6 months in the placebo treatment group. Results of the secondary efficacy and sensitivity analyses were consistent with the primary analysis. The overall survival data are immature, with 69% of the subjects in the ganitumab treatment group and 65% of the subjects in the placebo treatment group censored at the time of the primary analysis. The overall survival unstratified hazard ratio was 0.92 (95% CI: 0.46, 1.84). The median overall survival was 12.2 months in the ganitumab treatment group and 9.6 months in the placebo treatment group. The confirmed objective response rate was 8% (4 subjects) in the ganitumab treatment group and 2% (1 subject) in the placebo treatment group (all partial responses). The disease control rate was similar between the ganitumab and placebo treatment groups at week 6 (67% versus 65%), week 12 (45% versus 47%), and week 18 (27% versus 27%). Of the 4 subjects in the ganitumab treatment group who had a confirmed response, the median duration of response was 5.6 months and the median time to response was 4.6 months. The duration of response and the time to response was 5.2 months and 1.4 months, respectively, for the 1 subject in the placebo treatment group who had a confirmed response.

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 7 of 17187

A higher proportion of subjects in ganitumab treatment group (6 of 52 subjects) had a decrease in the sum of the longest diameters (SLD) of 30% to 50% compared with the placebo treatment group (1 of 52 subjects). A higher proportion of subjects in the placebo treatment group (16 of 52 subjects) had an increase in SLD of > 0% compared with the ganitumab treatment group (9 of 52 subjects). No other differences in subsets of decreases in SLD were observed between treatment groups. Conatumumab: The primary analysis of PFS, specified as a stratified Cox model (stratification factors: ECOG performance status and prior anti-VEGF exposure) comparing conatumumab with placebo, provided a hazard ratio estimate (95% CI) of 0.69 (0.41, 1.14) (p = 0.147, stratified log-rank test). The median PFS time was 6.5 months in the conatumumab treatment group and 4.6 months in the placebo treatment group. Results of the secondary efficacy and sensitivity analyses were consistent with the primary analysis. The overall survival data are immature, with 67% of the subjects in the conatumumab treatment group and 65% of the subjects in the placebo treatment group censored at the time of the primary analysis. The overall survival unstratified hazard ratio was 0.84 (95% CI: 0.43, 1.64) (p = 0.610). The median overall survival was 12.3 months in the conatumumab treatment group and 9.6 months in the placebo treatment group. The confirmed objective response rate was 14% (7 subjects) in the conatumumab treatment group and 2% (1 subject) in the placebo treatment group (all partial responses). The disease control rate was similar between the conatumumab and placebo treatment groups at week 6 (69% versus 65%) and week 12 (53% versus 47%). At week 18, the disease control rate was higher in the conatumumab treatment group (37%) compared with the placebo treatment group (27%). Of the 7 subjects in the conatumumab treatment group who had a confirmed response, the median duration of response was 5.3 months and the median time to response was 2.6 months. The duration of response and the time to response was 5.2 months and 1.4 months, respectively, for the 1 subject in the placebo treatment group who had a confirmed response. A higher proportion of subjects in conatumumab treatment group (5 of 51 subjects) had a decrease in SLD of 30% to 50% compared with the placebo treatment group (1 of 52 subjects). A higher proportion of subjects in the placebo treatment group (16 of 52 subjects) had an increase in SLD of > 0% compared with the conatumumab treatment group (12 of 51 subjects). No other differences in subsets of decreases in SLD were observed between treatment groups. Pharmacokinetic Results: Pharmacokinetic assessments were performed for ganitumab, conatumumab, irinotecan, SN-38, and 5-FU. Approved

Ganitumab: Following 12 mg/kg Q2W IV infusion, the mean (SD) preinfusion concentration (Cmin) in the fifth cycle (steady state) was 33.5 (16.5) µg/mL. Additional analysis was performed to compare the exposures over cycles with historical data of ganitumab monotherapy under the same regimen in subjects with advanced solid tumors (Study 20050118). The exposure levels were found to be comparable with or without coadministration of FOLFIRI. Conatumumab: Following 10 mg/kg Q2W IV infusion, the mean (SD) preinfusion concentration (Cmin) in the fifth cycle (steady state) was 92.4 (29.4) µg/mL. Additional analysis was performed to compare the exposures over cycles with historical data of conatumumab monotherapy under the same regimen in subjects with advanced solid tumors (Study 20050171). The exposure levels were found to be comparable with or without coadministration of FOLFIRI. Irinotecan and SN-38 (the active metabolite): Following IV infusion of 180 mg/m2 irinotecan, mean (SD) irinotecan area under the plasma concentration-time curve from time 0 to infinity (AUC0-inf) values were 12000 (2420), 14300 (3020), and 12400 (1500) ng•hr/mL in the ganitumab, conatumumab, and placebo treatment groups, respectively. The mean SN-38 AUC0-inf (SD) values were 354 (35), 345 (116), and 364 (124) ng•hr/mL, respectively. Because comparable irinotecan and SN-38 exposures were observed in the different treatment groups, coadministration with ganitumab or conatumumab did not appear to affect the pharmacokinetics of irinotecan or its metabolite SN-38. 5-FU: Following the protocol-specified dosing regimen, highly variable individual 5-FU concentrations were observed during the 48-hour infusion period. Nevertheless, the concentration ranges in ganitumab, conatumumab, and placebo treatment groups were comparable, and the median 5-FU concentrations at 24 hours (C24hr) were 490, 371, 450 ng/mL,

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 8 of 17187 respectively, in cycle 1 and 453, 475, and 389 ng/mL, respectively, in cycle 3. These results indicate that coadministration with ganitumab or conatumumab did not appear to affect the pharmacokinetics of 5-FU. Patient-reported Outcomes Results: The ganitumab, conatumumab, and placebo treatment groups were generally similar at baseline for each EORTC QLQ-C30 and FCSI parameter. Results for the EORTC QLQ-C30 and FCSI analyses did not demonstrate any treatment-related differences during the course of the study. Safety Results: All safety evaluations are presented separately for ganitumab and conatumumab. Ganitumab: The overall incidence of adverse events was similar for the ganitumab and placebo treatment groups (92% and 96%, respectively). The most common adverse events (≥ 20% of the subjects in either treatment group) were diarrhea (ganitumab: 43%, placebo: 41%), neutropenia (47%, 37%), nausea (35%, 33%), fatigue (29%, 22%), vomiting (27%, 27%), anemia (27%, 10%), thrombocytopenia (24%, 6%), stomatitis (20%, 20%), decreased appetite (20%, 16%), alopecia (18%, 22%), asthenia (16%, 24%), and constipation (14%, 33%). The subject incidence of grade ≥ 3 treatment-emergent adverse events was higher in the ganitumab treatment group (55%) compared with the placebo treatment group (47%). The most common grade ≥ 3 adverse events (occurring in ≥ 2 subjects in either treatment group) were neutropenia (ganitumab: 25%, placebo: 18%), diarrhea (2%, 10%), intestinal obstruction (6%, 8%), abdominal pain (4% each), vomiting (2%, 6%), anemia (8%, 4%), fatigue (2%, 4%), asthenia

(2%, 6%), hyperglycemia (8%, 4%), leukopenia (6%, 0%), and mucosal inflammation (0%, 4%). The difference between treatment groups in grade ≥ 3 adverse events is mainly driven by hematologic adverse events (ie, leukopenia and anemia), and adverse events reflecting the known potential or identified risks of ganitumab (ie, neutropenia and hyperglycemia). No new safety signals appear to be reflected in this difference. The subject incidence of serious adverse events was higher in the ganitumab treatment group (31%) compared with the placebo treatment group (24%). The most common serious adverse events (≥ 2 subjects in either treatment group) were intestinal obstruction (ganitumab: 6%, placebo: 4%), abdominal pain (4%, 2%), diarrhea (2%, 6%), vomiting (2%, 4%), and mucosal inflammation (0%, 4%). The difference between treatment groups in the incidence rate of serious adverse events is mainly driven by gastrointestinal events of abdominal pain and intestinal obstruction. These serious events of intestinal obstruction were due to either disease progression or intra-abdominal adhesions and were not reported to be related to ganitumab. Approved Two subjects in the ganitumab treatment group experienced fatal treatment-emergent adverse events. The fatal adverse event for Subject was rectal cancer (this subject’s primary cause of death is listed as disease progression), and for Subject was diabetic ketoacidosis; both events occurred within 30 days after the last dose of investigational product and were not reported to be related to investigational product. Two subjects (4%) in the ganitumab treatment group and 1 subject (2%) in the placebo treatment group withdrew from the study due to an adverse event. A higher proportion of subjects in the ganitumab treatment group (14%) experienced adverse events that led to discontinuation of any protocol specified therapy compared with subjects in the placebo treatment group (8%). Overall, median post-treatment values were generally within normal range and/or were similar to baseline for most of the clinical laboratory parameters. Despite never having received a dose of ganitumab, 1 subject in the placebo treatment group and 1 subject in the conatumumab treatment group, who were negative at baseline, tested positive for anti-ganitumab antibodies during the treatment period. In addition, 1 subject in the ganitumab treatment group, who was negative at baseline, tested positive for anti-ganitumab binding antibodies during the treatment period. No neutralizing antibodies were detected in any of the subjects. Anti-ganitumab antibodies detected at baseline or during treatment with placebo or conatumumab are likely to represent cross-reactive antibodies that are present as a consequence of molecular mimicry. Conatumumab: The overall incidence of adverse events was similar for the conatumumab and placebo treatment groups (96% and 96%, respectively). The most common adverse events (≥ 20% of the subjects in either treatment group) were diarrhea (conatumumab: 58%, placebo:

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 9 of 17187

41%), neutropenia (54%, 37%), nausea (36%, 33%), fatigue (22%, 22%), vomiting (28%, 27%), anemia (20%, 10%), stomatitis (8%, 20%), decreased appetite (24%, 16%), alopecia (26%, 22%), asthenia (30%, 24%), abdominal pain (24%, 16%), and constipation (16%, 33%). The subject incidence of grade ≥ 3 treatment-emergent adverse events was higher in the conatumumab treatment group (72%) compared with the placebo treatment group (47%). The most common grade ≥ 3 adverse events (occurring in ≥ 2 subjects in either treatment group) were neutropenia (conatumumab: 30%, placebo: 18%), diarrhea (18%, 10%), intestinal obstruction (4%, 8%), abdominal pain (8%, 4%), vomiting (8%, 6%), anemia (4% each), fatigue (8%, 4%), asthenia (6% each), hyperglycemia (0%, 4%), dehydration (6%, 2%), decreased appetite (8%, 0%), weight decreased (6%, 0%), increased aspartate aminotransferase (AST) (4%, 0%), mucosal inflammation (2%, 4%), abdominal distension (4%, 0%), hypertension (4%, 0%), nausea (4%, 0%), and pyrexia (4%, 0%). The higher incidence of neutropenia, diarrhea, abdominal pain, and fatigue in the conatumumab group accounts for most of the overall difference between treatment groups in grade ≥ 3 adverse events. The subject incidence of serious adverse events was higher in the conatumumab treatment group (40%) compared with the placebo treatment group (24%). The most common serious adverse events (≥ 2 subjects in either treatment group) were diarrhea (conatumumab: 10%, placebo: 6%), intestinal obstruction (4%, 4%), vomiting (6%, 4%), pyrexia (8%, 0%), dehydration (6%, 0%), and chest pain (4%, 0%). The higher incidence of diarrhea, pyrexia, and dehydration in the conatumumab group accounts for most of the overall difference between treatment groups in serious adverse events. Two subjects in the conatumumab treatment group experienced treatment-emergent fatal adverse events. The fatal adverse event for Subject was colorectal cancer (this subject’s primary cause of death is listed as disease progression), and the fatal adverse event for Subject was “death” (primary cause of death was “other” with further information noted as unknown death upon arrival at the hospital). One subject (2%) in the conatumumab treatment group and 1 subject (2%) in the placebo treatment group withdrew from the study due to an adverse event. A higher proportion of subjects in the conatumumab treatment group (16%) experienced adverse events that led to discontinuation of any protocol specified therapy compared with subjects in the placebo treatment group (8%). Overall, median post-treatment values were generally within normal range and/or were similar to baseline for most of the clinical laboratory parameters. One subject (3%) in the conatumumab treatment group, who was negative at baseline, tested positive for anti-conatumumab binding and Approved neutralizing antibodies during the treatment period.

Conclusions: This was a well-controlled study. Demographics and baseline disease characteristics were balanced across treatment groups and representative of a second-line mCRC population. The addition of ganitumab to FOLFIRI as second-line treatment did not appear to improve PFS in subjects with mutant KRAS mCRC over FOLFIRI alone. The hazard ratio point estimate of 1.01 (95% Cl: 0.61, 1.66) for ganitumab plus FOLFIRI compared with FOLFIRI alone suggests no effect of ganitumab on PFS; however, the data could be consistent with either substantial benefit (hazard ratio = 0.61) or harm (hazard ratio = 1.66). Because of the low proportion of death events, the overall survival data are immature. The confirmed objective response rate was higher in the ganitumab treatment group compared with the placebo treatment group and the time to response also was longer. This study showed a higher incidence of certain serious adverse events in the ganitumab treatment group as compared with the placebo treatment group, which were mainly reflective of gastrointestinal system organ class events and disease progression. In addition, the grade ≥ 3 adverse events, which were numerically higher in the ganitumab treatment group vs the placebo treatment group, were consistent with the known safety profile observed to date with ganitumab, including events identified as events of interest (neutropenia and hyperglycemia). No new safety risks associated with ganitumab treatment were identified.

Product: Ganitumab (AMG 479) and Conatumumab (AMG 655) Clinical Study Report: 20060579 Date: 15 October 2011 Page 10 of 17187

The addition of conatumumab to FOLFIRI as second-line treatment resulted in PFS that was numerically higher than treatment with FOLFIRI alone in subjects with mutant KRAS mCRC. The hazard ratio estimate was 0.69 (95% Cl: 0.41, 1.14) in favor of conatumumab treatment group, suggesting that the addition of conatumumab to FOLFIRI may improve outcome in this patient population and additional studies may be warranted. Because of the low proportion of death events, the overall survival data are immature. The confirmed objective response rate was higher in the conatumumab treatment group compared with the placebo treatment group. This study showed higher incidences of grade ≥ 3 adverse events of neutropenia, diarrhea, abdominal pain, and fatigue and higher incidences of serious adverse events of diarrhea, pyrexia, and dehydration in the conatumumab treatment group versus the placebo treatment group. A review of the safety data observed in this study compared with the safety data in the monotherapy setting and with chemotherapy backbone (FOLFIRI) suggests that these events most likely reflect the adverse events associated with the chemotherapy backbone. FOLFIRI coadministration did not appear to affect the pharmacokinetics of ganitumab and conatumumab. Likewise, ganitumab or conatumumab coadministration did not appear to affect pharmacokinetics of irinotecan and 5-FU.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: fol Eligible subjectswererandomized ina1:1:1ratiotosecond conatumumab relativetoFOLFIRI aloneforthesecond study evaluatingthesafety Me Explorator      The secondaryobjectiveswere: ( cancer (mCRC)whosetumorsexpressmutant alone whenadministeredasasecond survival (PFS)ofconatumumabplusFOLFIRIorganitumabrelativeto Objectives: Development Phase: 04 Study Period: Un (France, Investigator(s) andStudyCenter(s): the S Safety andEfficacyofFOLFIRIinCombination Title ofStudy: Name ofA Name ofFinishedProduct: Name ofSponsor: SYNOPSIS This C Ganitumab (GAN)+FforSecond Publication(s): KRAS ancer (mCRC)[poster].ASCOGastrointestinalCancersSy lowing ited States).Studycentersandinvestigatorsarelistedin June 2012. thodology: and otherge regulate the To correlatetreatmentoutcomeswithtumortissueanalysis ofsomaticgenemutationsthat ganitumab onthep (irinotecan and5 To evaluatepharmacokineticsofconatumumab,ganitumab, andFOLFIRIcomponents (FCSI) (FACT)/ core questionnaire(EORTCQLQ Organization forResearchandTreatmentofCancer(EORTC)qualitylifequestionnaire To evaluate incidence ofanti To evaluatetheincidenceofadverseeventsandsignificantlaboratoryabnormalities duration ofresponse,andtimetoresponse response +partialresponse),ratesofdiseasecontrol(ie To estimatetreatmenteffectonoverallsur was econd 11 February2013 ). : Hong Kong,Hungary,India,Italy,Poland,RussianFederation,Singapore,Spain,and a phase2,multicenter,randomized, double Ganitumab (AMG479)andConatumumab655) y objectivesarelocatedintheclinicalprotocol( . ctive Ingredient: - Nati Theprimary line Treatmentof Thefirst

APhase2,Randomized,Double CohnAL,TaberneroJ,Maureletal.Conatumumab(CON)+FOLFIRI(F)or onal CancerComprehensiveNetwork( treatment effectonpatient insulin nes thatareknowntobeinvolvedintumorigenesis

- Am - conatumumab oranti FU) andestimatetheimpactofadministrationconatumumab and 2 - harmacokinetics ofFOLFIRIcomponents(irinotecan and 5 like growthfactorreceptortype1( gen, Inc.,ThousandOaks,CA subject objectivew andefficacy AMG479and655 Ganitu KRAS was enrolled - line TreatmentofMutant(MT) - ma mu as toestimatethetreatmenteffectonprogression- - - C30) andthe Thisstudywasconductedat line treatmentforsubjectswithmetastaticcolorectal ofFOLFIRIincombination withganitumabor b andconatumumab tant MetastaticColorectalCarcinoma - - ganitumab antibodyformation. reported outcomes(PRO 31 March2009 vival (OS),objectiveresponser - . ty Wi pe Kirstenratsarcomavirusoncogenehomolog - blind, Placebo- th AMG479or655VersusFOLFIRIfor Functional AssessmentofCancerTherapy - blind, double NCCN - line treatmentof Appendix 1 IGF - line therapyconsistingofonethe , complete+partial+stabledisease), Appendix 2 ; mp ) - the lastsubject’sfinalvisitwas 1R - Colorectal Sy osium, 2012. controlled StudyEvaluatingthe KRAS ) - and deathreceptorpathway dumm . s 49 ). ) usingtheEuropean Metastatic Colorectal . centers

KRAS

y, placebo

mp ate Abstract 534. - mutant m in tom Index ( - Page 2of4990 complete 10 FU) - controlled countrie . free CRC s . - s

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis)  Date: Product: study report. (ganitumab, conatumumab,ganitumab 36 pharmacokinetic data.Safetyinterimreviewswere An internalAmgendatareviewteam(DRT)wasestablished toreviewsaf consent, administrativedecision,ortheendofstudy, whicheverwasearlier. (radiographic perRECISTorclinicalprogression),start ofanewtreatment,death,withdrawal follow   analysis data primary The primary conducted anadditionalsafetyreviewafterthe100th subje between thesecondDRTreviewandplannedprimary analysisforthestudy,DRT FOLFIRI performed atweek6,12,andevery8weeks(± Solid Tumors[RECIST protocol Cy in firstline( status (0vs1)andpreviousexposuretoanti Randomization wasstratifiedbyEasternCooperativeOncologyGroup(ECOG)performance before radi randomized, whicheverwasearlier.Anysubjectwhodiscontinuedprotocol adm Subjects werefollowedforsurvivale study administrationofprotocol assessment of (except death)werefo Subjects whodiscontinuedprotocol regulator ganitumab or after thelastsubjectwasrandomizedcouldbeeligibleforcontinu subject hadbeenrandomized.Subjectswhoremainedo progression), unacceptabletoxicities,withdrawalofconsent,death,or36 discretion oftheinvestigatoruntildiseaseprogression(radiographicperRECISTorclinical became intoleranttoinvestigationalproductswerecontinuereceiveFOLFIRIatthe of consent,death progression (radiographicperRECISTorclinicalprogression),unacceptabletoxicities,withdrawal com were any reasonotherthandiseaseprogression,thecomponentsofFOLFIRIchemotherapy event theadministrationofanycomponentFOLFIRIchemotherapywasdiscontinuedfor withdrawal ofconsent,death,or36monthsafterthelastsubjecthad (± ganitumab 3 cles ofche subjects wererandomized Arm A:Ganitumab Arm C:Ganitumab 14 Arm B:Ganitumab12mg/kgplusconatumumab 14 14 ponents of inistration ofprotocol day to continuealongwiththeinvestigational - up period 11 February2013 day day day analysisclinicalstudyreport dated15October2011.Thecurrentreportsummarizes final - s) (notincludingthetimetorecoverfromtoxicities).Subjectswerepermittedreceive , specified therapyuntildiseaseprogression(bymodifiedResponseEvaluationCriteriain y m and hadtheopportunitytocomplete2treatmentcycles. Ganitumab (AMG479)andConatumumab655) - ographic diseaseprogressionordeathcontinuedtohaveradiologicalim s s s placebo, andconatumumab- yes vsno).

echanism. analy mo (all data) conatumumab by ganitumab and FOLFIRI . Thisprocedurewasused therapy plusinvestigationalproducts(ga , or36monthsafterthelastsubjectha sis data llowed forsafety . ] - - placebo plusconatumumab10mg/kginco version 1.0 placebo plusconatumumab Table chemotherapy weretocontinueinvestigationalproductsuntildisease - , specified th using adatacutoffdate , had receivedatleast1doseofinvestigationalproducts conatumumab 1 - extension protocolorasprovidedforbythelocalcountry’s specified therapy. provides asummaryofthedata summarizedineach ] guidelinesorclinicalprogression),unacceptabletoxicities, - specified therapyfordiseaseprogressionorotherreasons very 3months( erapy untildeathor36monthsafterlastsubjectwas - for30(+ placebo, orconatumumab placebo) weretobeadm to assessdiseasestatusuntilprogre - vascular for products. Subjectswhobecameintoleranttoall 3) daysandforantibodiespharmacokinetic 30 (+ of 31 planned - ± placebo incombi -placebo incomb -placebo 14 days)startingfromthedateoflast 7 da endothelial growthfactor( 3) day d January 2011 n protocol nitumab, conatumumab, been randomized.Subjectswho ys ct receivedtreatment. ) thereafterduringthelong to occurafterthefirst18and s and60(+ inistered ever e -placebo) incombinationwith -placebo) d treatmentwithopen - Based onthelengthoftime mb specified therapy36 been randomized.Inthe ainwt FOLFIRIevery nation with ination withFOLFIRIever ination withFOLFIRIever , were presentedinthe 14) daysafterthelast ety and

mo - specified therapy nths afterthelast y 14da VEGF Page 3of4990 aging clinical ys ) - term therapy - mo ssion label nths y y

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Number ofSubjectsEnrolled: Number ofSubjectsPlanned: CSR =clinicalstudyreport; Irinotecan (180mg/m adm FOLFIRI chemotherapyregimen (irinotecanwithinfusional5 Reference Therapy,Dose andModeofA infusion onday1ofeachcycle justbeforetheadministrationofFOLFIRIchemotherapy. was 10mg/kgQ2 The doseofganitumabwas12mg/kgonceevery2weeks (Q2 products areprovidedin investigational productsinthisstudy. Ganitumab, ganitumab Investigational Product,DoseandModeofAdministration, Manufacturing clinical studyreport com ECOG performancestatusof0or1;andadequatehematologic, renal,andhepaticfunction. prior anticancertherapy;measurableornonmeasurable diseaseaccordingtomodifiedRECIST; anti the combinationofafluoropyrimidineandoxaliplatin laborator colon orrectumwithmetastaticdisease;mutant included menorwomen≥18yearsofagewithhistologicallyconfirmedadenocarcinomathe Dia Biom Antibody PK PRO me Safety interest, Safety tim Efficacy Efficacy Demographic andbaselinecharacteristics Subject d Data T PRO survival, ECOG=EasternCooperativeOncologyGroup, gnosis andMainCriteriaforEligibility: - e toresponse plete listofinclusion/exclusion criteriaisprovidedin VEGF therapyanddocumenteddiseaseprogressio d inistered onday1ofeach treatment cycleaftertheadministrationofinvestigational products. ications arker = 11 February2013 ype - - patient y assessment;1prioranticancertherapyregimenformetastaticdiseaseconsistingof - - ECOG, vitalsigns,ECGs,concomitant adverse events laboratory evaluation Ganitumab (AMG479)andConatumumab655) isposition overall response,durationof OS Table , -reported outcome,PK PFS W. . 1

2 Investigational productswereadministe . ) wasadministeredoverapproximately 9 - placebo, conatumumab,andconatumumab Data Summarizedinthe Listing 14 , a – = nodataanalysisperformed dverse eventsof s Approximately150subjects 155subjects - 1.1, pharmacokinetics. = Clinical Study The listingsoflotnumbersbysubjectforall Study final analysis( dministration, Manufacturing Batch Themaininclusioncriteriaforsubjectsinthisstudy 2 KRAS 0060579 Reports - Primary based chemotherapywithorwithout ECG ; OS=overallsurvival; FA tumor status (31 January2011 n s ). A ection 7.5 ≤ 6monthsafterthelastdoseofthis = Data C nalysis CSR electrocardiogram, Prim - FU andleucovorin)was red sequentiallybyintravenous andFinal W) 0 minutesonday1ofeach cycle. X X X X X X X X X X ary utoff) andthedoseofconatumumab - placebo wereconsider

of theprimaryanalysis confirmed bycentral

PF An = S aly Batch Final Analysis progression sis Number: investigational Page 4of4990 CSR X X X X X – – – – – Number ed -free The (IV) A :

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Study Endpoints randomized. were followedforsurvivalupto36monthsfromthedatethatlastsubjecthad withdrawal ofcon disease progression(bymodifiedRECISTcriteriaorclinicalprogression),unacceptabletoxicities, Subjects weretreatedwithsecond-  Pharmacokinetic Endpoints      Secondary Endpoints  Primar study report. presented intheprimaryanaly provided intheprotocollocated The followingisafulllistofendpointsfromthe Duration ofTreatment: over Leucovorin (400mg/m 4 immediately followedbya5- infusion butwithoutmixing(linescouldbeconnectedattheIVportclosesttosubject),   mi Pharmacokinetic parameters(observedmaximumconcentration [C analysis clinicalstudyreport) an objectiveresponseinthesubsetofs Ti (presented intheprimaryanalysisclinicalstudyreport) whichever occurredfirst] subsequent dateofprogression[basedonmodifiedRECISTorclinicalprogression, response assessmentwithanoutcomeindicatingobjectivethroughtothe Duration ofresponse(define study report) full analysissetwithbaselinemeasurabledisease) stable diseaseorbetter[basedonmodifiedRECIST]dividedbythenumberofsubjectsin Rate ofdiseasecontrol(definedasthenumbersubjectswithabestoverallresponse evaluated bytheinvestigator) response oraconfirmedpartialasdeterminedbymodifiedRECISTcriteriaand Objective responserate(definedasproportionofsubjectswitheitheraconfirme any cause,orifapplicable,dateofcensoring) O whichever occurredfirst],ordeathduetoanycause,dateo the firstobservationofdiseaseprogression[permodifiedRECISTorclinical Progression mi for 5 Pharmacokinetic parameters (concentrationsoveranapproximately48 SN Pharmacoki nutes anda5 approximately 46to48hours. verall survival(definedasthenumberofdaysfromdaterandomizationtodeath me nim 11 February2013 y Endpoint - 38 - toresponse(definedastimefromthedateofrandomization tothefirstobservationof FU um Ganitumab (AMG479)andConatumumab655) concentration[C netic parameters(C - free survival(definedasthenumberofdaysfromdaterandomization sent, death,or36monthsafterthelastsubjecthadbeenrandomized. : -

FU continuousIV 2 ) wasadministeredoverapproximately2hoursduringtheirinotecan Theexpectedmedianlength FU , min sis clinicalstudyreportandarenotincludedinthepresent or bolus ( d asthenumberofdays betweenthedateoffirsttumor ]) forconatumumabandganitumab death duetoanycause,orwhereapplicable,dateofcensoring) (presented intheprimaryanalysisclinicalstudyreport) Appendix 1 line treatmentonstudyuntiloneofthefollowingoccurred: max Lot numberswerenotcollected. infusion , C 400 mg/m min ubjects whoresponded) , andareaunderthecurve[AUC])foririnotecan (2400 mg/m ). Someendpoints,asindicatedbelow,were study 2 ) administered (additionalexplorator of (presented intheprimaryanalysisclinical 2 treatment wasapproximately4 ) adm inistered over approximately2to f censoring) (presented intheprimary max via ambulator

] andobserved y endpointsare - hour infusionperiod) progression, Page 5of4990 d com been y pum Subjects mo plete nths. from p to

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: occurred first)ordeath. experienced aPFSevent(radiologicaldiseaseprogressionorclinical The primary described below. me described previouslyintheprimary  Biomark   Safety  Patient- Final resultsfromkey Statistical Methods the pri arm, 51to A totalof155subjectswere randomizedtothestudy:52subjectsganitumab treatment Subject report, andcomparisonshavebeenmadebacktothe prima me Serum ganitumabandconatumumabplasma5 (CIs) forganitum the estimatedhazardratiowaspresentedwith curves. treatment armswasassessedviaexamina analysis wasconductedonthefullset. conatumumab treatmentarm, and30subjectsintheplacebotreatmentarmwere ongoinginthe Summary ofResults: was determined. by analysis setwereanalyzedaccordingtothetreatmentreceived.Adverseeventstabulated subjects whoreceivedatleast1doseofganitumab,conatumumab,orplacebo;inthis Safety endpointswereanalyzedusingthesafetyanalysisset,whichincludedallrand study tothoseintherespectivefirst- pharmacok tables. Theincidenceofsub events). Clinicallaboratoryparametersweresummarized usingdescriptivestatisticsorshift conatumumab (hypomagnesemia,hyponatremia,infusion reactions,andvenousthromboembolic thrombocy (drug (MedDRA) version1 s thodologies tabolite ofirinotecan)concentrationsweresummarizedwithdescriptivestatistics. progression ofsolidtumorsandcorrelationwithtreatmentoutcomes apoptosis pathwaysandothergenesthatareknowntobeinvolvedinthedevelopment Tu Incidence ofanti Incidence ofadverseeventsandsignificantlaboratoryabnormalities the FCSI Changes (improvementorworsening)inPROsas primary ys - related hepatotoxicity,hyperglycemia,infusionreaction, neutropenia,rash, ma tem organclassandpreferredtermusingtheMedicalD 11 February2013 mo reported Outcome er h rmr ehdo nlssue stratifiedCox’s The primarymethodofanalysisuseda Disposition ry r tissueanalysisofsomaticgenemutationsthatfunctiontoregulatetheIGF Ganitumab (AMG479)andConatumumab655) topenia, venousthromboembolicevent the conatumumabtreatment arm,and52totheplacebotreatmentarm. inetics wasinvestigatedbycomparingtheconcentrationexposuresincurrent

analysisclinicalstudyreport) analy efficacy (presented intheprimary for the ab relativetoplaceboand sis : , analy 5.0 - The : conatumumab oranti 30 efficacy key . Adverseeventsofinterestweresummarizedseparatelyforganitumab subjects intheganitumab treatment arm,32subjectsinthe The statisticalanaly methodologies foralloftheefficacy efficacy sis wasevent jects developinganti andsafetyanalysesarepresentedinthepresentclinical study TheeffectofFOLFIRIonganitumaborconatumumab andsafetyanaly analysisclinicalstudyreport in - hum analysisclinicalstudyreport) - driven andwastooccurwhen - tion of an studyforeachinvestigationalproduct. ganitumab antibodyformation conatumumab relativetoplacebo. ses weredescriptiveand associated 80%and95%confidenceintervals For PFSand - s, andsensorineuralhearingloss)for Kaplan ganitumab andanti ses presentedinthepresentreport - FU, irinotecan,andSN assessed usingtheEORTCQLQ - Meier survivalandlog( ry andsafetyanalyseshavebeen analysisreportwhereapplicable. OS ictionary forRegulatoryActivities P , dated 15October2011.The roportional , theproportionalitybetween - pro noninferential. conatumumab antibodies 100 (pres gression, whichever

- subjects had 38 (theactive H ented inthe azards model,and - log) survival Page 6of4990 At thetimeof

om - The efficacy 1R and/or are - iz C30 and ed

.

Approved the finalanalysis.TheKaplan p- (y analysis, stratifiedbyECOGperformancestatus 4.5 monthsforganitumabversus4.8placebo.Thehazardratio(95%CI)atthefinal progression, radiologicalordeath) and 50(96%)subjectsintheplacebotreatmentarmexperienced In thefinalanalysis,perinvestigatorreview,49(94%)subjectsinganitumabtreatmentarm Progression Ganitumab Efficacy Efficacy Results: Subject Demographics 12. As ofthefinalanalysis, received irinotecan ganitumab armneverreceivedleucovorinand5- received investigationalproduct,1subjectintheconatumumabarmand2subjects Of thesubjectsrandomizedto the conatumumabarm,and35[67%]subjectsinplac discontinuing thestudy discontinued thestudy Synopsis ClinicalStudyReport:20060579(FinalAnalysis) 67%, and85%; were 79%,71%,and85%,respectively; for overall, themostcommonreasonfordiscontinuationoftreatmentwasdiseaseprogression: study Date: Product: es value 3 investigational product Ethnicity (Race): Ag Sex: versus no), was1.11(0.73,1.67) versus no), for ( Table 14 11 February2013 e: resultsbelowarepresentedseparatelyforganitumabandconatumab. of conatumumab, and9. 74 men(48%);81women(52% me 0.632 Ganitumab (AMG479)andConatumumab655) - fre an (standarddeviation[SD e Survival - and . 1. Table 2 , ( Table 14 for primary analysis[P ( 119 (77%)white;9 (6%) black;25(16%)Asian;2(1%)HispanicorLatino was death irinotecan the med Table 14 2 : , the compares thePFSresultsfromprimary From - 1.2 6 each ofthetreatments,1subjectinganitumabarmnever pe theprimaryanaly ian num for placebo( - , FA rcentage , thepercentage 1.1, FA (39 [75%]subjectsintheganitumabarm,40[78%] - Meier plotforPFSispr ). f

or ganitumabcomparedwithplacebo Wi A] ber ofmonthsonstudywas9.6 ]) 57.5(10.5)years(range:28to81) for ) . Inthefinalanalysis,mostcommonreasonfor ) ) s intheganitumab,conatumumab,andplaceboarms th respectto . Atthetimeoffinalanalysis,alls leucovorin and5 Table 14 ( FU, and1subject Table 14 (0 sis clinicalstudyreport: s were77%, versus 1) - 1.3 all oftheprotocol ebo arm). , FA - 4.3 ovided in - FU , FA ). and prioranti 67% , thepercentage PFS ) . ThemedianPFStimewas

in theganitumabarmnever , and Figure analysiswiththosefrom events for ganitumab, - 85%.

specified treatments, - VEGF exposure 1 , witha . (clinical s were79%, ubjects had Page 7of4990

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source: e d c b a Q2W Note: The Censored Un Stratified 95% CI 80% CI Unstratified 95% CI 80% CI Stratified Min, Max Q1, Q3(K 95% CI(K 80% CI(K Median (K Progression Events The hazardratioisobtainedfromtheCoxProportionalHazard Model.Ahazardratio<1.0indica A normalscor Stratification factorisperprimary poolingstrategy: Progression Events areclinicalprogressions,radiological e longer progression lower averageeventrateand alongerprogression due toanycause,dividedby(365.25/12). assessment ofdiseaseprogression(asclassifiedbymodified RECIST Oncology Group;VEGF Ma xposure (y s tratified = x =every2weeks; 11 February2013 a maximum; RECIST=ResponseEvaluationCriteriainSolid Tumors; ECOG=EasternCooperative Table 14 Normal p- Normal p- - c h l n (%) og-r value value - Ganitumab (AMG479)andConatumumab655) - - - otherapy is - azard M) M) M) M) n (%) -free survivaltimeiscalculatedasthenumberofdaysfrom randomization es or l h - og-r free survival azard e <0indicatesfewerthanexpected eventsforAMG ank d,e e -4.1 s s Table

ank r n core core atio -free survivaltime. o) t est r (Primary analysis); . atio FOLFIRI. t K-M =Kaplan c,d est 2. = c b vascular endothelialgrowthfactor

(months) An alysis ofProgression -Meier estimate;CI= c Cotherapy AMG 12 .,11.7 0.0, (Q (N 4.0, 2.5, 3.0, 34 (65) 18 (35) Table 14 2 4.5 mg = (Full A W) Prim 5.3 8.3 6.6 52 479 /k + (0.61, (0.72, (0.61, (0.73, ) g ary - 0.944 0.998 0.003 -4. 0.070 0.98 1.01 nalysis Set) -free survivaltimeforAMG Analy ECOG p 3 1.59) 1.34) 1.66) 1.40) (F Cotherapy Placebo + inal analysis) .,12.2 0.0, (N 4.4, 1.8, 3.1, -f 36 (69) 16 (31) sis onfidence ree Survival or deaths. 4.6 5 = erformance . 5.1 6.6 6.5 2) 479 i nterval relative toplaceboand, or clinicalprogression)death status (0or1)andp – Cotherapy AMG 479 12 .,22.2 0.0, (Q2 (N 4.0, 2.4, 2.8, 49 (94) Ganitumab ; Min=minimum; 3 (6) 4.5 mg = 479 W)

5.3 7.7 6.2 52) Final Analysis /k + (0.66, (0.75, (0.73, (0.84, relative toplacebo. g - 0.916 0.632 0.479 0.106 0.98 1.11 to thefirst 1.46) 1.27) 1.67) 1.45) Page 8of4990 Cotherapy Placebo + rior anti .,16.4 0.7, (N 4.5, 1.9, 4.4, 50 (96) therefore 2 (4) 4.8 = tes a 6.5 6.7 6.5 52 -VEGF ) , a

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: The finalanalysispiecewise 3 > placebo treatmentarmhaddied the finalanalysis,39(75%)subjectsinganitumabtreatmentarmand36 In Overall Survival Figure analysis arecomparedwiththoseofthe num tim com 0 ganitumab and12.0monthsforplacebo;the Output: g14-04_002_km_pfs.cgm(DateGenerated:21AUG12:11:01:51)SourceData:adam.asleff Program: /statistics/amg655/crc/20060579/analysis/part2_final/graphs/g_kmplot_eff.sas or 1andprioranti the finalanalysis es mo   pared withplacebo Subjects atrisk: erically

Progression-Free Survival 1 Figure for ganitumab and placebo at thefinalanalysiswere for ganitumabandplacebo 11 February2013 nths) showed 2. 0.0 0.2 0.4 0.6 0.8 1.0 For thetimeperiod unstratified hazardratio(95%CI)of0.84(0 treatment arm For thetimeperiod unstratified hazardratio(95%CI)of1.27(0.65,2.47) treatment arm

Ganitumab (AMG479)andConatumumab655) , 15 93 42 71 510 15 16 27 28 34 35 39 50 51 24 83 02 91 12 12 13 22 19 24 22 34 30 36 31 38 38 51 48 52 52 9 8 7 6 5 4 3 2 1 0 than attheprimary Placebo +Cotherapy AMG 65510mg/kg(Q2W . Kaplan - of VEGF exposure the following( OS and and , withap - Me , thedataweremorematurethaninprimary 34 [ 16 [ of of ier Pl mo 0 to > 65% 31 analy del 3 - ( value of0. Table 14 %] ) +Cotherapy ot ofProgression mo 3 Table 14 for PFS ] mo subjects witheventsintheplacebotreatmentarm subjects witheventsintheplacebotreatmentarm): (y nths ( sis. ThefinalKaplan es versus no), was1.27 es versusno), (Full A 0 0 0 0 0 0 0 0 0 1 1 1 1 2 3 6 6 9 nth final analysisin - 0 0 0 1 1 1 1 1 1 1 1 1 2 2 5 7 8 s ( 4.1 30 [ 342 using acut - 4.4 hazard ratio 19 [ 01 21 41 61 81 02 22 425 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 0 1 1 1 2 2 2 2 3 3 4 4 4 5 7 9 , FA nalysis Set) 58%] . Overallsurvivalresultsfromtheprimary , FA Time (Months) 37 ) . Themedian .51, 1.39) %] ): - subjects witheventsin free SurvivalperInvestigatorReview

off subjects witheventsintheganitumab Table (95 point of AMG 47912mg/kg(Q2W - mo Meier plot % CI) re sim 3 (0.78, 2.06) and showthatthemedian OS 3 mont , stratified byECOGstatusof ilar was 12.4 for to eachother hs OS analy

for ganitumab the ganitumab (0 to3 (69%) subjectsinthe is providedin ) +Cotherapy months for sis o Page 9of4990 and f , OS ): . Asof

an an OS

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source: e d c b a Q2W Note: The Unstratified 95% CI 80% C Stratified Min, Max Q1, Q3(K 95% CI(K 80% CI(K Median (K Overall s Events Censored Normal Un Stratified 95% CI 80% CI The hazardratioisobtainedfromtheCoxProportionalHazard Model.Ahazardratio<1.0indicatesa A normalscore<0indicates fewer thanexpectedeventsforAMG Stratification factor Overall survivaltimeiscalculatedasthenumberofdaysfrom randomizationtodeathdueanycause, Events aredeaths. longer overallsurvivaltime. VEGF exposure(yesorno). lower averageeventrateandalongeroverallsurvivaltime forAMG divided by(365.25/12). ECOG Min p- p- Normal s value value tratified =every2weeks; = 11 February2013 I a minimum; Max=maximum; Table 14 = - urvival s e d, e c h l n (%) Eastern CooperativeOncologyGroup;VEGF=vascularendothelial growthfactor. s og-r

- Ganitumab (AMG479)andConatumumab655) - - - core otherapy is - azard M) core M) M) M) n (%) h l og-r azard ank b -4.20 ank (months) r atio s are t est r atio t FOLFIRI. K-M =Kaplan (Primary analy Table c,d est per primarypoolingstrategy c 3. 12

+ Cotherapy An - mg AMG 479 -Meier estimate;CI= c .,15.1 0.0, (N = datanotavailableinprimaryreport 8.6, 9.9, 6.0, 16 (31) 36 (69) sis); alysis ofOverallSurvival 12.2 /k = g (Q2 Prim NE NE NE Table 14 52) (Full A (0.46, (0.59, ary - W) 0.822 0.225 0.92 Analy – – – nalysis Set)

-4.1 1.84) 1.45) : ECOGperformancestatus(0or1)andprioranti- Cotherapy Plac .,15.3 8.7, .,20.2 0.1, (F (N 8.0, 6.7, 18 (35) 34 (65) sis inal analysis) 9 = onfidence ebo + .6 NE NE 52) 479 – 479 i 12 nterval. NE= Ganitumab relative toplaceboand, + Cotherapy mg relative toplacebo. AMG 479 for comparisons .,15.4 8.0, 14.9 9.4, .,31.7 0.1, 15.6 5.5, (N 39 (75) 13 (25) 12.4 /k = g (Q2 52)

Final Analysis (0.75, (0.88, (0.78, (0.92, n ot W) 0.473 0.717 0.342 0.950 1.18 1.27 e

stimable Page 10of4990 1.87) 1.59) 2.06) 1.74) ; Cotherapy Placebo + .,17.0 8.5, 14.4 9.5, .,37.9 1.5, 24.4 7.4, (N 36 (69) 16 (31) therefore 12.0 = ; 52) , a

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: following The piecewisemodelfor for PFSisprovidedinFigure results fromtheprimary of 0. 6.4 progression, radiologicalordeath) ar In thefinalanalysis,perinvestigatorreview,4 Progression Conatumumab (y final analysis,stratifiedbyECOGperformancestatus Output: g14-04_001_km_os.cgm(DateGenerated:21AUG12:11:01:26)SourceData:adam.asleff Program: /statistics/amg655/crc/20060579/analysis/part2_final/graphs/g_kmplot_eff.sas es versusno) m and50(96%)subjectsintheplacebotreatmentarmexperiencedPFSevents mo   Subjects atrisk 177, whichwasconsistentwiththatofthepri Overall Survival 11 February2013 nths for 0.0 0.2 0.4 0.6 0.8 1.0 For the treatment arm an For thetimeperiod unstratified hazardratio(95%CI)of1.32(0.58,3.01) ( Table 14 Ganitumab (AMG479)andConatumumab655) unstratified hazardratio(95%CI)of 25 14 34 93 12 42 22 01 71 11 10 10 11 10 14 10 17 11 18 15 20 18 16 21 20 17 22 20 19 23 21 22 24 22 24 28 27 24 31 30 28 35 36 31 39 39 31 41 39 35 43 41 42 48 42 45 51 43 49 52 45 49 52 45 50 47 52 51 51 - 9 8 7 6 5 4 3 2 1 0 free Survival iue2 Figure Placebo +Cotherapy AMG 65510mg/kg(Q2W conatumumab , was time period - 4. 0.75 2 . Kaplan , FA and analy OS (0. ): 10 [ of of using acutoffpointof6months(0toand> 49, 1. sis withthosefromthefinalanaly versus 4.8monthsforplacebo.Thehazardratio 0 to6months > 6mo 19%

- 1. Me ) +Cotherapy 01 21 41 61 81 02 22 42 62 82 03 23 43 63 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 14 ] ier PlotofOverallSurvival subjects witheventsintheplacebotreatmentarm) ) for nths conatumumab ( 26 [ ( 13 [ 7 50% (9 0 0 0 0 0 0 0 1 2 2 2 3 3 4 4 5 6 6 7 7 7 9 9 1.13 ( ( Table 14 Time (Months) 25% ma 2 %) subjectsinthe ] ry subjects witheventsinbothtreatmentarms) 0. ] (0 versus1) analysis. subjects witheventsintheganitumab 65, 1.95 compared withplacebo,ap- - AMG 47912mg/kg(Q2W 4.3 0 1 1 1 1 1 1 1 1 3 4 4 4 4 5 7 7 8 0 0 0 0 1 1 1 1 1 1 2 3 4 4 5 6 9 , FA sis. ) Table (Full A ). ThemedianPFStimewas and prioranti ThefinalKaplan conatumumab 4 nalysis Set) compares thePFS

6 mo ) +Cotherapy (95% CI)atthe - nths) showedthe VEGF exposure Page 11of4990 (clinical - treatment Meier plot : an value

: .

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: d c b a Q2W Note: ThecotherapyisFOLFIRI. Source: e 95% CI 80% CI Stratified Min, Max Q1, Q3(K 95% CI(K 80% CI(K Median (K (m Progression Events Censored Normal Un Stratified 95% CI 80% CI Unstratified Stratificationfactorisperprimary poolingstrategy: The hazardratioisobtainedfrom theCoxProportionalHazardModel.Ahazardratio< 1.0 indicatesa Progression Events areclinicalprogressions,radiological A normalscore<0indicates fewer thanexpecte lower averageeventrateand due toanycause,dividedby(365.25/12). assessment ofdisease VEGF Oncology Group;VEGF Ma longer progression p- p- Normal onths) s va value tratified = x =every2weeks; lue 11 February2013 a e maximum; RECIST=ResponseEvaluationCriteriainSolid Tumors;ECOG=EasternCooperative Table 14 - xposure (y s e d,e h l n (%) s og-r

- Ganitumab (AMG479)andConatumumab655) - - - core - azard M) core M) M) M) n (%) -free survivaltimeiscalculatedasthenumberofdaysfrom randomization tothefirst -f l h og-r azard ree ank Table -4.1 ank r s atio es or -free survivaltime. t urvival est r (Primary analysis); atio t K-M =Kaplan 4 c,d est . A progression (asclassifiedbymodifiedRECISTorclinicalprogression) ordeath n = c o) b vascular endothelialgrowthfactor nalysis ofProgression . a longerprogression -Meier estimate;CI= c Cotherapy AMG 655 10 .,21.3 0.0, (Q2 (N 4.7, 2.7, 4.4, 33 (65) 18 (35) Table 14 6.5 mg 51) = (Full A W) Prim 6.7 8.3 7.0 /k + (0.45, (0.53, (0.41, (0.49, g ary d eventsforAMG655relative toplaceboand - - 0.196 0.147 -4.3 1.293 1.452 0.73 0.69 nalysis Set) -free survivaltimeforAMG Analy ECOG p 1.18) 1.00) 1.14) 0.96) (F -f Cotherapy Placebo + inal analysis) .,12.2 0.0, ree Survival (N 4.4, 1.8, 3.1, 36 (69) 16 (31) sis onfidence or deaths. 4.6 = . erformance 5.1 6.6 6.5 52) i nterval – status (0or1)andp status Conatumumab Cotherapy AMG 655 10 .,24.1 0.3, (Q2 (N 4.7, 2.7, 4.4, 47 (92) ; Min=minimum; 4 (8) 6.4 mg 51) = 655 W)

6.7 8.3 6.9 Final Analysis /k + (0.49, (0.56, (0. (0.57, relative toplacebo. g - - 49, 0.137 0.177 1.488 1.350 0.73 0.75 Page 12of4990 1.10) 0.96) 1.14) 0.99) Cotherapy Placebo + .,16.4 0.7, rior anti (N 4.5, 1.9, 4.4, , 50 (96) therefore 2 (4) 4.8 = 6. 6.7 6.5 52) - 5 , a

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) 3 > The finalanalysispiecewise Date: Product: primary tim analysis arecomparedwiththoseofthefinalinTable com of 0or1andprioranti conatumumab the placebotreatmentarmhaddied( the finalanalysis, In thefinalanalysisof Overa es for mo   pared withplacebo, ll Survival 11 February2013 nths) showedthefollowing analysis.ThefinalKaplan treatment arm For thetimeperiod treatment armand For thetimeperiod unstratified hazardratio(9 unstratified hazardratio(95%CI)of0.65(0.39,1.06) conatumumab Ganitumab (AMG479)andConatumumab655) and 12.0monthsforplacebo;thehazardratio(95%CI),stratifiedbyECOGstatus 41 ( 80 - OS and VEGF exposure and placeboatthefinalanalysisw %) subjectsinthe , thedataweremorematurethaninprimary 16 [ 34 [ of of mo a p 0 to3months( > 3months( 31% 65% del forPFSusingacutoffpointof3months(0to - value of0. ( 5% CI)of0.95(0.47,1.92) - Table 14 Meier plotfor ] ] subjects witheventsintheplacebotreatmentarm): subjects witheventsintheplacebotreatmentarm): Table 14 (y es versusno) conatumumab 32 [ 867 - 4.4, FA 15 [ - 4.1 63% . Overallsurvivalresultsfromtheprimary 29% OS , FA ] ) subjects witheventsinthe : is providedinFigure ] ). Themedian subjects witheventsinthe , was treatment armand36 ere 0.96 closer 5 and showthatthemedian (0. , 61, OS num 1.53) for was 12.4 analy

2. erically sis of conatumumab (69%) subjectsin Page 13of4990 , conatumumab conatumumab than atthe mo nd OS nths for . Asof

an an OS

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source: e d c b a Q2W Note: The Unstratified 95% CI 80% CI Stratified Min, Max Q1, Q3(K 95% CI(K 80% CI(K Median (K Overall s Events Censored Normal Un Stratifi 95% CI 80% CI Stratificationfactorisperprimarypoolingstrategy The hazardratioisobtainedfromtheCoxProportionalHazard Model.Ahazardratio<1.0indicatesa A normalscore<0indicates fewer thanexpectedeventsforAMG Overall survivaltimeiscalculate Events aredeaths. longer overallsurvivaltime. VEGF exposure(yesorno). lower averageeventrateandalongeroverallsurvival divided by(365.25/12). ECOG Min p- p- Normal s value value tratified =every2weeks; = ed 11 February2013 a minimum; Max=maximum; Table 14 = - urvival s e d, e c h l n (%) Eastern CooperativeOncologyGroup;VEGF=vascularendothelial growthfactor. s og-r

- Ganitumab (AMG479)andConatumumab655) - - - core otherapy is - azard M) core M) M) M) n (%) h l og-r azard ank b -4.20 ank (months) r atio t est Table r atio t FOLFIRI. K-M =Kaplan (Primary analysis); c,d est c 5 . 10 An d asthenumberofdaysfromrandomizationtodeathdue toanycause, + Cotherapy - mg 10.2, 10.4, 10.6, alysis ofOverallSurvival AMG 655 -Meier estimate;CI= c .,22.4 0.0, (N = datanotavailableinprimaryreportforcomparisons 17 (33) 34 (67) 12.3 /k = g (Q2 Prim 17.3 17.3 12.4 Table 14 51) (Full A (0.43, (0.54, ary - W) 0.610 0.509 0.84 Analy – – – nalysis Set)

: ECOGperformancestatus(0or1)andprioranti -4.1 1.64) 1.30) Cotherapy Placebo + time forAMG .,15.3 8.7, .,20.2 0.1, (F (N 8.0, 6.7, 18 (35) 34 (65) sis inal analysis) 9.6 = onfidence NE NE 52) – 655 Conatumumab 655 i 10 nter relative top + Cotherapy mg relative toplacebo. 10.9, 11.0, val. NE= AMG 655 .,34.0 1.1, 19.6 9.6, (N 41 (80) 10 (20) 12.4 /k = g (Q2 17.0 14.4 51)

Final Analysis (0.64, (0.75, (0.61, (0.71, lacebo and n ot W) - 0.989 0.014 0.867 0.167 1.00 0.96 e

stimable Page 14of4990 1.58) 1.35) 1.53) 1.30) ; Cotherapy Placebo + .,17.0 8.5, 14.4 9.5, .,37.9 1.5, 7.4, (N 36 (69) 16 (31) , therefore 12.0 = ; 24.4 52) - , a

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) following ( The piecewisemodelfor Date: Product: the finalanalysiswasasfollows( Extent ofexposuretothecotherapiesinganitumabandplacebotreatment 12.00 4.2 and4.9,respectively;themedianaverageganitumabdosedeliveredtosu num (range: 2to54)insubjectstheplacebotreatmentarm( 13 As ofthefinalanalysis,s Extent ofExposure Ganitumab below. forganitumab andconatumumab Extent ofexposureandsafetydataaresummarizedseparately Safety Results: were asfollows( Investigational productdosemodificationsmadeintheganitumabandplacebotreatmentarms infusions (range:2to60)ofinvestigationalproductcomparedwithamedian18          ber ofmonthsontreatmentforsubjectsintheganitumabandplaceboarmswas mg 11 February2013 treatment arm For treatment armand For thetimeperiod unstratified hazardratio(95%CI)of0.71(0.27,1.87) dose intensitywas0.89 average dosedeliveredwas 1 1 Irinot dose intensitywas0.93and0.89 average dosedeliveredwas200.73and199.61mg/m 1 0 to30)and10.0(range: was5.5(range: Leucovorin: mediannumberofinfusions event and“other A 16 subjects, A hazard ratio(95%CI)of1.1 me 4.9; medianaveragedose delivered 10.0 (range:1to27),respectively; mediannumberofmonthsontreatmentwas 3.7and 5- 5- dose intensi average dosedeliveredwas176.05and176.09mg/m 9 (protocol change, 7 A /k t least1dosewithheld(andmostcommonreasonforwithholding) t least1dosedelay(andmostcommonreasonfordelay) t least1dosechange(andm to to to subjects, subjects, Table 14 FU FU Ganitumab (AMG479)andConatumumab655) g/infusion; andthemedianrelativedoseintensitywas0.96forganitumab. dian relativedose intensitywas0.88and 0.86 the timeperiodof 27), respectively;mediannumberofmonthsontreatment was3.7and4.9;median 27), respectively;mediannumberofmonthsontreatment was4. 27), respectively;mediannumberofmonthsontreatment was3.7and4.9;median ecan: mediannumberofinfusionswas6.0(range:0to30)and10.0 c b ontinuous IVInfusion:median numberofinfusionswas5.0(range: olus: mediannumberofinfusionswas5.5(range: 5 subjects, Tabl - spe - 17% (adverseevent, 13% (weightchange,4subjects, 4. ty 31%) e 14 cified adverseevent, 2 was0.92and0.91 , FA and ,” ubjects intheganitumabtreatmentarmreceivedamedianof OS - 5.6 5 ): 10 [ 26 [ of 10%) subjects, using acutoffpointof6months(0toand> , FA 0 to6months( > 6months 19% 50% and 0.8 Table 14 ): ] ] 1 subjects witheventsintheplacebotreatmentarm): subjects intheplacebotreatmentarm):an 389.67 (0.6 10% each) ost commonreasonforthechange) 3 5 subjects, 6 - ( 12 subjects, 5.2 , 1. 34 [ and was 2333.27and2333.58 mg/m 7 [14% , FA 8 67% 5) 388.93 through ] ] 10%) subjects witheventsintheconatumumab 8%) subjects witheventsintheconatumumab 23%) mg ; ; placebo: Table 14 placebo: Table 14 /m ; placebo: 2 2 2 /infusion; andme /infusion; andmedianrelative /infusion; andmedianrelative 1 : - 5.1 11 subjects, - to 30)and ganitumab: 5.5 9 subjects, , FA , FA 27 subjects,

6 : ). Themedian : ): mo ganitumab: 2 2 ganitumab: bjects was /infusion; and arms atthetimeof and 4.9;median 9 nths) showedthe dian relative 18% (adv .0 (range: unstratified 25 subjects, 22% (wei Page 15of4990 1 53% (

to 30)and an erse ght “other 48% ,”

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Based onfinalan All Treatment Safety Results Cotherapy dosemodificationsaresummarizedin Date: Product: consistent withthoseinthefinalanalysis. leukopenia (15%,4%),andchills(13%,6%).Resultsfromtheprimary neutropenia (52%,41%,respectively),anemia(27%,16%),thromb that werereportedin≥ treatment armwasneutropenia(gani In thefinalanalysis, primary events reportedmostfrequently( treatment 94% ; placebo: 98%( 11 February2013 analysisresultswiththefinalanaly - em Ganitumab (AMG479)andConatumumab655) - ergent adverseeventwas emergent AdverseEvents aly sis the adverseeventreportedinmorethanhalfofsubjectsganitumab Table 14 5% moresubjectsintheganitumabarmthanplaceboincluded results, - 6.4 the overallpercentageofsubjects ≥ , FA 10% ofsubjectsineithertreatmentarm)andcomparesthe tumab: 52% ). comparable Table sis results. 6 pre Table 14 ; placebo: 41%).Adverseeventsin sents thetreatment between the - 5.7 , FA 2 treatmentarms: reporting atleast1 ocy through topenia (23%,6%), analysisweregenerally - em

Table 14 ergent adverse Page 16of4990 - ganitumab: 5.10 Table , FA 6 .

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source: Note: A Note: Thecotherapyis Preferred Term Oedema Pain Anxiety Dry Hy Mucosal Headache Hy Intestinal Abdominal p Epistaxis Chills Dizziness Back Leukopenia Py Asthenia Decreased Abdominal p Stom Constipation Alopecia Thrombocy Anaemia Fatigue Vom Nausea Diarrhoea Neutropenia least S the finalanalysis ubjects reportingat rexia pokalaemia perglycaemia s iting kin atitis 1 adverseevent dverse eventswerecodedusing 11 February2013 p Table 14 ain i Ganitumab (AMG479)andConatumumab655) p nflammat o eripheral bstruction topenia a ain ain ppetite Table -6.4 and gradedusing u pper (Primary analysis); ion FOLFIRI. 6.

Most FrequentlyReported Cotherapy AMG 479 12 (Q2 (N 0(0 0(0 0(9 11(22) 10(19) 10(20) 10 (20) 4(7 4(7 5(9 14(27) 20(39) 10 (20) 23(45) 15(29) 21(41) 15(29) 19(37) 50(98) 23(44) 11(22) 27(52) 14(27) 12 (24) 17(33) 14 (27) 49(94) 21(41) 15 (29) 19(37) 14 (27) 18 (35) 49(96) 22 (43) 24 (47) 47 (92) 5 (10) 7 (14) 5 (10) 8 (16) 8 (16) 8 (16) 9 (18) 7 (14) 9 (18) 1 (2) 3 (6) 3 (6) 2 (4) 2 (4) 2 (4) 2 (4) 3 (6) 3 (6) 4 (8) 2 (4) Common TerminologyCriteria forAdverseEvents(CTCAE). mg = W) Prim 51) MedDRA Table 14 /k (Safety A + g Ad ary vers Analy -6.4 version 13.1fortheprimaryanalysis andversion nalysis Set) e Events Cotherapy Placebo + (Final analys (N 7(3 0(9 19(37) 13(25) 12 (24) 10(19) 11(21) 17 (33) 11 (22) sis 5 (10) 5 (10) 5 (10) 5 (10) 7 (14) 6 (12) 5 (10) 8 (16) 8 (16) 5 (10) 4 (8) 4 (8) 3 (6) 4 (8) 3 (6) 3 (6) 4 (8) 1 (2) 3 (6) 2 (4) = ( ≥ 51) 10%) is) Treatment Cotherapy AMG 479 12 (Q2 (N 12 (23) 0(9 10(20) 10 (19) 10 (19) 14 (27) 6 (12) 8 (15) 5 (10) 7 (13) 6 (12) 8 (15) 8 (15) 1 (2) 4 (8) 2 (4) 2 (4) 2 (4) 2 (4) 3 (6) 4 (8) 4 (8) 2 (4) mg = W) 52) -e /k Final Analysis

+ g mergent Page 17of4990 Cotherapy Placebo + (N 14 (27) 10 (20) 13 ( 7 (14) 5 (10) 8 (16) 5 (10) 5 (10) 5 (10) 7 (14) 5 (10) 5 (10) 8 (16) 9 (18) 8 (16) 5 (10) 3 (6) 3 (6) 3 (6) 4 (8) 2 (4) 51) = 25) 15 .0 for

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: thrombocy Source: Note: Adverseeventswerecoded using Note: Thecotherapyis 65% forganitumaband57%placebo( The overallincidenceofgrade Grade 20%) (40% emergent treatment arm) The mostfrequentlyreportedtreatment- ganitumab treatmentarmand90%ofsubjectsintheplacebotreatm investigational productor Treatment in eithertreatmentarmincluded obstruction (8% from boththeprimary Table diarrhea, whichwasreportedin2%ofsubjectsforganitumabversus12%placebo. placebo treatmentarm (22%);allothereventswerereportedatsimilarfrequenciesexcept was reportedinmorethan5%subjectstheganitumabtreatmentarm(33%) Mucosal Fatigue Diarrhoea Asthenia Vom Hy Leukopenia Abdominal p Intestinal Anaemia Neutropenia adverse event least 1grade3,4,or5 S Preferred Term the finalanalysis ubjects reportingat perglycaemia . ; Table iting 7 39%), nausea(33% ≥ 3T 11 February2013 Table 14 summarizes themostfrequentlyreport - i adverse - topenia (21% Ganitumab (AMG479)andConatumumab655) nflammation o em reatment bstruction 7. ain ergent adverseeventsthatwereconsideredbytheinvestigatortoberelated

were -6.13 ; Most FrequentlyReported 10%), anddiarrhea(2% and gradedusing - event presentation:neutropenia(ganitumab:48% (Primary analysis); - generally FOLFIRI. analysisandthefinalanalysis. emergent AdverseEvents ; 6%), alopecia(17% chemotherapy administrationwerereportedin83%ofsubjectsthe ; 37%) Cotherapy consistent withthosereportedmostfrequentlyintheall ≥ AMG 479 12 (Q2 (N 8(5 24(47) 13 (25) 28 (55) neutropenia (ganitumab:33% 3 treatment , 3 (6) 0 (0) 1 (2) 4 (8) 1 (2) 1 (2) 1 (2) 3 (6) 2 (4) 4 (8) CTCAE. mg vom = W) Prim 51) M /k (Safety A + Table 14 edDRA version13.1fortheprimary analysisandversion iting (23% g ≥2 related adverseevents(≥ Ad ar ; Table 14 12%). y verse Events - ( Analy ; em ≥ 25%), asthenia(12% -6.9 ed 5%) Grade3,4,and5 nalysis Set) ergent adverseeventsinthefinalanalysiswas Cotherapy Placebo + ; The (N 24%), fatigue(21% grade (Final analysis) sis 9 (18) 5 (10) - 4 (8) 2 (4) 3 (6) 2 (4) 2 (4) 3 (6) 0 (0) 2 (4) 2 (4) 6.9 = grade ≥ 51) , FA ≥ 3 ). Thosereportedin treatment 3 ; adverse eventof placebo: 22%),intestinal Cotherapy AMG 479 12 (Q2 (N ; 0% ofsubjectsineither 7(3 11(22) 17 (33) 34 ent arm( 22%), andstomatitis(17% 4 (8) 0 (0) 2 (4) 3 (6) 1 (2) 1 (2) 2 (4) 3 (6) 3 (6) 4 (8) ; mg - ; = 22%), anemia(21% em Treatment W) (65) placebo: 39%),diarrhea 52) /k Final Analysis + ergent adverseevents

g Table 14 ≥ 10%ofsubjects neutropenia -e Page 18of4990 mergent Cotherapy Placebo + (N 29 (57) - 5 (10) 6 (12) - 3 (6) 3 (6) 3 (6) 3 (6) 3 (6) 0 (0) 2 (4) 3 (6) 6.5 treatment 51) = 15 , FA ; 12%), .0 ). for ; -

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source: Note: Adverseeventswerecodedusing Note: Thecotherapyis arms for ( com As ofthefinalanalysis,overallincidencetreatment- Serious Treatment provided inAppendix Detailed narrativesforsubjectswithseriousadverseeventsreportedthroughoutthestudyare arm fromtheprimary Table treatment arm 36 At thefinalanalysis,atotalof39(75%)subjectsin Deaths nervous system (serious)( grade For ganitumab,theevents includedgrade4acute from thestudy the As ofthefinalanalysis,deathsfor3subjectsin ganitumabtreatmentarmresultedfrom progression. placebo treatmentarmdiedfroma considered toberelatedeitherinvestigationalproduct orchemotherapy. aftertheprimaryanalysis).None obstruction (thiseventreported ( represents anadditional23 ganitumab (4% At thetimeoffinalanalysis, theincidenceoftreatment- Treatment provided inAppendix Detailed narrativesforsubje Table 14 Table Preferred Term Intestinal Diarrhoea adverse event least 1serious S the finalanalysis ubjec ( t 71%) parable betweentheganitumabandplacebotreatmentarms(33%31%,respectively) reatment 8 1 thrombocytopenia(nonserious); forplacebo,theeventwasgrade3metastases tocentral 11 February2013 14- Table 14 ts reportingat summarizes theseriousadverseeventsreportedin≥5%ofsubjectseithertreatment each - subjects intheplacebotreatmentarmdied 6.14 6.45, PA - Ganitumab (AMG479)andConatumumab655) o emergent AdverseEventsLeadingtoStudyWithdrawal orDiscontinuationofTherapy bstruction - em of theseriousadverseevents Table , FA were was unchangedfromtheprimary -6.14 [ 2 subjects and gradedusing ergent adverseevents - ). ). Formostofthesesubjects emergent AdverseEvents diarrhea (2% (Primary analysis); analysisandthefinalanalysis. FOLFIRI. The frequenciesweresimilarbetweentheganitumabandplacebotreatment 5. 5 8. .

Most FrequentlyReported Listing 14 ] ) andplacebo(2%[ deaths and18deaths,respectively,sincetheprimaryanalysis cts withfataladverseeventsreportedthroughoutthe studyare Cotherapy AMG 479 12 (Q2 (N 6(1 2(4 7(3 16(31) 17(33) 12(24) 16 (31) 3 (6) 1 (2) mg = CTCAE. and W) Prim 51) Serious /k - MedDRA version13.1for + 6.3, FA (Safety A g Table 14 6%, respectively)andintestinalobstruction(8%,6%). ary of rectal cancer,diabeticketoacidosis,andintestinal Analy . The ). - Ad em -6.14 1 subject Cotherapy nalysis Set) Placebo + verse Events analysi , (N ergent adverseevent. sis overall, theprimary 2 (4) 3 (6) events reportedin cholangitis (seriousadverse event)and (Final analysis) = in thestudy 51) ( ≥ ganitumab treatmentarmand s ( ] 5%) ) treatmentarms em Table 14 em the primaryanalysisandversion ergent seriousadverseeventswas Treatment ergent adv ( Cotherapy of the AMG 479 Listing 14 12 (Q2 (N - 6.15, PA 4 (8) 1 (2) cause ≥ 5%ofsubjectsineither mg = W) se 52) -emergent /k erse events Final Analysis +

leading towithdraw g events - No subjectsinthe of deathwasdisease ; Table14 6.2 , FA Page 19of4990 were Cotherapy Placebo + ), which (N in the - 3 (6) 3 (6) 6.1, FA = 51) 15 .0 al for ) .

.

Approved

event ofhypoalbuminemiaintheganitumabarm( An overviewoftheadverseeventsinterestfromprimary search termsforthedrug (narrow terms) placebo treatmentarm neutropenia of interestthatweregrade hy Hepatic disorders,venousthromboembolicevents,sensorineuralhearingloss,and arm thantheplacebo(neutropenia:52%,41%,respectively;thrombocy ganitumab armatthefinalanaly ( com By narrow searchstrategyforeacheventofinterest. the primary hearing loss. disorders, rash,infusionreaction,neutropenia,venousthromboembolicevents,andsensorineural Adverse eventsofinterestforganitumabincludehypergly Adverse EventsofInterest congestive cardiacfa to discontinuationofallprotocol acute cholangitisandthrombocytopeniaalsoledtowithdrawalfromthestudy); primary 3 protocol As ofthefi abdom ketoacidosis; intheplaceboarm,theseeventsincludedpancytopenia,cardiacfailurecongestive, cholangitis acut ganitumab armincludedthrombocy Synopsis ClinicalStudyReport:20060579(FinalAnalysis) obstruction, keratitis,colonic discontinuation ofallprotocol ( The summaryoftreatment Date: Product: 5 ganitumab treatmentarm(thesamenumberofsubjectsasin product atthefinalanalysischanged Table Table (6%) subjectsintheplacebotreatmentarm (10%) perglycemia occur thetimeoffinal parable 11 February2013 inal pain,diarrhea,andmetastasestocentralnervouss 14- 14- analysis - subjects intheplacebotreatmentarm(1additionalsubjectsinceprimary specified therapyoccurredin5(10%)subjectstheganitumabtreatmentarmand 6.47, 6.45, FA Ganitumab (AMG479)andConatumumab655) nal analysis,treatment andfinalanalyses , between theganitumabandplacebotreatmentarms of Se is e, hyperbilirubinemia,alanineaminotransferase which FA arch strategies ( provided inTable Listing 14 ) ) . . red atgenerallysimilarfrequenciesinthe2treatmentarms. ilure, andpanc Neutropenia andthrombocytopeniawerereportedmostofteninthe Adverse eventsleadingtodiscontinuationofinvestigationalproductinthe 33% ofsubjectsintheganitumabtreatmentarm analysis, theoverallfrequencyofadverseeventsinterestwas report - related hepaticdisorderseventofi - ≥ em - 6.8, PA; 3 ed - specified therapyintheganitumabtreatmentarmwereintestinal ; f ergent adverseeventsleadingtodiscontinuat occurred in - or thisclinicalstudyreport,thefocusisonoutputusing sis, andtheyoccurredathigherfrequenciesintheganitumab such using broadandnarrowsearchterms specified therapyintheplacebotreatmentarmwerediarrhea, - em topenia, keratitis,colonicobstruction,intestinal 9. ytopenia. Table 14 ergent adverseeventsleadingtodiscontinuationofall little

events acute cholangit from theprimary ≤ 2 subjectsineacheventofinterestexceptfor . ; theseresultsrepresentlittlechange - A m 6.4 6 Table 14 edical reviewwasalsoperformedofthebroad , FA ) . Theadverseeventsleadingto is, cemia, thrombocy - analy 6.17, FA and thrombocytopenia(theeventsof nterest, whichrevealedagrade th analysisandthefinalanalysi ys e primary (ALT) sis: 7(13%)subjectsinthe tem. (65% each) ). increased, were and 22%ofsubjectsinthe analysis)and

topenia, hepatic im ion ofinvestigational topenia: 23%,6%). plemented the eventsleading Page 20of4990 and Adverse events from the analy diabetic at both sis) s 3

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source: Note: Adverseeventswerecodedusing Note: Thecot The incidenceoftreatment Clinical Laborat [all [all wereelevatedvalues] in theganitum analytes the finalanalysis arm]) (8%, 2%,respectively[decreased valueswiththeexceptionof1elevatedvalue intheplacebo treatment values]) forhematology Category Event ofInterest a Thrombocy Rash Neutropenia Treatment Infusion reaction Hy hearing loss S events thromboembolic Venous Drug interest adverse eventof reporting any S infusion reaction hepatic disorders the finalanalysis ensorineural ubjects decreased values] perglycaemia - ( 11 February2013 related Table Table 14 ; thoseanalyteswithahigherincidenceofgrade - em Ganitumab (AMG479)andConatumumab655) Table - herapy is related topeni ergent values( 14 ab armversusplaceboincludedalkalinephosphatase -6.60 ory Evaluations - 7.2, FA and gradedusing was 9. IncidenceofTreatment (Primary analysis); FOLFIRI. 12 (24) 19(37) 15(29) 26 (51) ) forchemistry similar betweentheganitumabandplacebotreatment arms 5 (10) 5 (10) 5 (10) 7 (14) mg Total 6 0 0 0 6 0 2 0(0) 0(0) 1(2) 2 (4) 2(4) 0(0) 3(6) 3(6) 2(4) 0(0) 2(4) 0(0) 1(2) 0(0) 1(2) 3 (6) 2(4) 2 (4) , Theonlyanalytewithahigher incidenceofgrade AMG 47912 ; Cotherapy ) Listing 14 /k (N , lipase 33 (65) - g (Q2 n (%) em ≥ = 5%) intheplacebo ergent grade≥ 51) Prim Grade 2 6 0(0) 3(6) 1 (2) 2(4) 4(8) 0(0) 2(4) 3(6) 3(6) 0 (0) 3 (6) 0 ( ( 7 1(2) 4(8) 2 (4) W)  CTCAE. (10%, 4% - 3 0) 7.1, FA + ary andabsoluteneutrophilcount (31%,24% MedDRA version13.1fortheprimaryanalysisand Narrow SearchTerms (Safety A Table 14 A naly 7 (14) Total 2 0(0) 1 (2) ; Cotherapy Listing 14 Placebo + [all elevatedvalues]) (N sis 2 3 chemistryandhematologylaboratoryvaluesasof -6. - n (%) 8 (55) emergent A nalysis Set) = 47 arm versus 51) Grade 9 (18) (Final analysis) 0 4 0(0) 2(4) 0 (0)  - 3 7.2, FA ≥ 3 treatment dverse EventsofInterest 7(2 7(3 1(1 11(22) 21(41) 17(33) 27 (52) 12 (23) 5 (10) 5 (10) mg Total the ganitumabarmwaspotassium ). AMG 47912 Cotherapy 4 ( 6 (4) 2 13) /k (N , and 4( 34 g (Q2 n (%) = (12% 65) 52) Grade sodium - 2 6 0(0) 3(6) 1 (2) 0(0) 3(6) 0 (0) 0 0 0(0) 0(0) 0 (0) Final Analysis em W) 

3 + and ergent values ≥ 3 (10%, 2% [all decr 6%, respectively 8 (16) 5 (10) Total Page 21of4990 for mostofthe Cotherapy 1(2) 12) Placebo + (N 3( 33 n (%) eased = 65) (≥ 51) Grade 15 0 (0) 3 (6)  5%) .0 3 for

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Table Change frombaselineinhematologyanalytesforthefinalanaly Table Change frombaselineinchemistry between 2 analysis comparedwiththefinalanalysis.Theresultsweregenerallyconsistentbetween Table Note: Note: Thecotherapyis Source: ALT =alanineaminotransferase;ASTaspartate NCI=NationalCancerInstitute; Category Hematology Chemistry CTCAE =CommonTerminology CriteriaforAdverseEvents. analy CTCAE Based onNCI- Analyte count Absolute neutrophil Total bilirubin Sodium Potassium Phosphorus Magnesium Lipase Glucose Calcium Bicarbonate AST Am phosphatase Alkaline Albumin ALT Wh Platelets Ly Hemoglobin 14- 14- 10 11 February2013 ses, withthe Table 14 mp yl ite bloodcells the primary presents thefrequenciesofgrade 7.17, FA 7.6, Ganitumab (AMG479)andConatumumab655) ase hoc Table FA ytes 7.80 -7.79 andTable14- through and 10.

e FOLFIRI. (18%) xception of Table 14 IncidenceofTreatm Table 14 vers and the ion 3.0. Cotherapy AMG 479 12 (Q2 (N - 4(7 0(0 6(1 12(24) 16(31) 10(20) 14 (27) 7.25, FA 5 (10) 5 (10) 5 (10) 6 (12) 9 (18) a decreasein 1 (2) 1 (2) 1 (2) 0 (0) 4 (8) 2 (4) 1 (2) 2 (4) 0 (0) 1 (2) 2 (4) 3 (6) 3 (6) mg - analytesforthefinalanalysisaresummarizedin = 7.16, FA W) final Prim 51) /k (Safety A (Primary analysis); + g (10%) ary through ≥ Analy and 3 chemistry ent analy nalysis Set) grade Cotherapy Placebo + Table 14 (N - Table 14 sis emergent NCICTCA 1 (2) 0 (0) 2 (4) 0 (0) 0 (0) 2 (4) 3 (6) 0 (0) 0 (0) 0 (0) 0 (0) 2 (4) 0 (0) 0 ( 4 (8) 1 (2) 3 (6) 2 (4) = si ≥ 0) s. 51) Table 14 3 andhematologyvaluesfortheprimary ly - 7.18, FA - mp 7.37, FA hoc - sis aresummarizedin 7.2 Cotherapy ytes intheganitumabarm AMG 479 12 (Q2 (N (Final analysis) through . 5 (10) 5 (10) 6 (12) 6 (12) 5 (10) 1 (2) 1 (2) 1 (2) 0 (0) 4 (8) 2 (4) 1 (2) 2 (4) 0 (0) 1 (2) 2 (4) 3 (6) 4 (8) mg = W) 52) E /k Final Analysis

+ Grade g Table 14 Page 22of4990 3 ≥ Cotherapy Placebo + - (N 7.24, FA 1 (2) 4 (8) 1 (2) 0 (0) 3 0 (0) 0 (0) 1 (2) 0 (0) 0 (0) 0 (0) 1 (2) 2 (4) 3 (6) 1 (2) 3 (6) 4 (8) 4 (8) = (6) 51) .

.

Approved

Listing hepatitis/toxic hepatitisandmetastas individual Extent ofExposure Conatumumab ( cotherapies wasday44. last doseofganitumabwasadministeredonday30(cycle2 day, radiological day 66(cons to investigationalproduct( hy points duringthestudy. highest value 277.0 ALT and hyperbilirubinemia(notrelated ganitumab asaresultof2events me ≤ Synopsis ClinicalStudyReport:20060579(FinalAnalysis) [AST] liver analyteabnormalitiesthatmetthecriteriaforHy’sLaw(ALToraspartateaminotransferase At Date: Product: was 4. num 14 As ofthefinalanalysis,subjectsin 10 (range: arms wereasfollows( Investigational productdosemodificationsmadeinthe Extent ofexposuretothecotherapiesin tim Listing 2 perbilirubinemia werereportedduringthestudy,someofwhichconsideredtoberelated .00 mg/kg/infusion;andthemedianrelativedoseintensitywas0. the finalanalysis, e ofthefin nuin (range: 2to infusions t thesecriteria. x ULN)       , AST,andtotalbilirubin ber ofmonthsontreatmentforsubjectsinthe µ ≥ 5 11 February2013 14 mo 14 3 11 13 At least1dosechange(andmostcommonreasonforthechange): 67% ( At least1dosedelay(andmostcommonreasonfordelay): event and“other,” 10 At least1dosewithheld(andmostcommonreasonfor withholding): 1 5- dose intensitywas0.85 average dosedeliveredwas 17 1 1 Leucovorin: mediannumberofinfusionswas7.0 Irinotecan: mediannumber ofinfusionswas dose intensitywas0.85 average dosed and 4.9,respectively;themedianaverage 2 to54)insubjectstheplacebotreatmentarm( - x to to to criteria at FU 4.2, PA). - Ganitumab (AMG479)andConatumumab655) l/L, respectively 6.2, subjects, subjects, subjects, ( the upperlimitofnormal[ Table 14 27), respectively; mediannumberofmonth 27), respectively;mediannumber ofmonthsontreatmentwas4. 27), respectively;mediannumberofmonthsontreatment was idered notrelatedtoinvestigationalproduct) al analy b (not meeting “other olus: mediannumberofinfusions was7.0 FA disease ).

Subject ,” mu Values forALT,AST,totalbilirubin,andalkalinephosphataseeachmet 1 subjectintheganitumabtreatmentarm(Subject 22% (weightch 25 20 - sis wasasfollows( 7.40, FA 21 Table 14 ltiple % ( % (adverseevent, elivered was200. Multiple adverseeventsofALTincreased,ASTand subjects, progression wasdocumentedforthissubject(Listing14 92 The subjectdiedduetodiseaseprogressiononday198 ); onthisday,alkalinephosphatase,at559 Listing 14 5 adverse event and weight change, adverse eventand Hy ) ofinvestigationalproductcomparedwithamedian18infusions tim subjects, for thissubject ’s Lawcriteria ; - e pointsduringthestudy Listing 14 5.6, FA and 0.91 and 0.89 of 41%) ) - on day58 ange, ALT increased 6.4 es totheliverandbiliar ULN 10% each) , a conatumumab ): 7.43 ; , FA Table 14 placebo: - 26 ], totalbilirubin 7.3, FA 6 subjects, 5 subjects, ). occurred onstudyday74(1371U/L,669and ). conatumumab and 176.09mg/m and 199.61mg/m

. Alkaline phospha This subjectdiscontinuedtreatmentwith An ). - 5.2, FA conatumumab 27 subjects, conatumumab event ofhyperbilirubinemiawasreported ( No subjectsintheplacebotreatmentgroup considered rel 7 12 treatment armreceivedamedianof 10%) .0 (range: conatumumab s ontreatment was . T %) > that me (range: (range: through Table 14 , ; placebo: 2 and placebotreatment he highestvalues , had apriormedicalhistoryof y tract(Listing x day 5 subjects, 2 2 ULN /infusion; andmedianrelative /infusion; andmedianrelative tase was>2 53% (“other,” 1),and t seriouscriteria 1 to46 and placebotreatm dose deliveredtosubjectswas 0 to 0 to Table 14 ated toinvestigationalproduct) - , 88 5.1, FA and alkalinephosphatase conatumumab: U/L (grade2),wasalsothe 9 subjects, and placebotreatment for 46) and9.0(range: 46) and10.0(range: ) and10.0(range: the lastdoseof 10%

conatumumab 14.2.2, PA; ). Themedian 4.5 4.5 - 5.5, FA conatumumab: x 5 conatumumab: 16 subjects, (all grade4) each ULN atmosttime and 4.9;median and and 4.9;median Page 23of4990 ; thefollowing - 18% (adverse 1.7, arms atthe ) exhibited ); 4.9; median ): 34 subjects, ent arm placebo: P A) . . 31%) for The s the on

.

Approved

from tract infection(10%,0%) 2%), hypomagnesemia(12%,hypertension(10%,depressionand Synopsis ClinicalStudyReport:20060579(FinalAnalysis) 1 em Based onfinalanaly All Treatment Date: Product: neut the diarrhea (59% conatumumab In thefinalanalysis,adverseevent results withthefinalanalysisresults. mo arms ( Cotherapy dosemodificationsaresummarizedin Safety Results 6%), st frequently( ergent adverseeventwas conatumumab  ropenia (5 the finalanalysis. Table 14 rash 11 February2013 4.9; medianaveragedosedeliveredwas2221.03 5- dose intensitywas0.80 average dosedeliveredwas me 10.0 (range: FU Ganitumab (AMG479)andConatumumab655) dian relativedoseintensitywas0.81 (1 - 6 c emergent AdverseEvents ; 3 %, ontinuous IVInfusion:mediannumberofinfusionswas treatment arm - 45%) %, 41%),anemia(2 6.4, FA ≥ 6 10% ofsubjectsineithertreatmentarm)andcomparestheprimary arm thantheplacebo %), chills(1 sis results,theoverallpercentageofsubjectsreportingatleast1treatment . Adverseeventsin 1 ). to 27),respectively;mediannumberofmonthsontreatmentwas . Resultsfromtheprimary Table the same(98%)inbothconatumumaban were 4 11 and 0.83 %, 6%) s neutropenia ( presents thetreatment 4 reported inmorethanhalfofthesubjects %, 16%), decreased appetite(24%,18%),pyrexia 364.65 , dehydration(14%,6%),peripheralneuropathy Table and 388.93mg/m were 11 and 0.86 cona Table 14 that werereportedin≥ diarrhea (59% analysisweregenerallyconsistentwiththose tumu - and 2333.58mg/m -5 ma em .7, FA b ergent adverseeventsreported 2 : /infusion; andmedianrelative 53%; and through 45%, respectively) placebo: 7 .0 (range:

5% moresubjectsin d placebo Table 14 2 /infusion; and 41%) Page 24of4990 0 to - 5.10, FA treatment analy and (2 46 4.5 urinary 2 ) and %, sis and .

-

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Note: Adverseeventswerecoded using Note: ThecotherapyisFOLFIRI. Source: Preferred Term Hy Dy We Insomnia Headache Stom Hy Cough Hy Neuropathy p Dehy Chills Constipation Rash Mucosal Py Fatigue Decreased Abdominal p Alopecia Anaemia Vom Asthenia Nausea Neutropenia Diarrhoea least 1adverseevent S the finalanalysisandgraded using ubjects reportingat rexia pertension spnoea pomagnesaemia pokalaemia Table ight i ting dration atitis 11 February2013 Table 14 d ecreased i Ganitumab (AMG479)andConatumumab655) nflammation 11. a ain ppetite eriph

-6.4 Most FrequentlyReported( (Primary analysis); eral Cotherapy AMG 10 (Q2 (N 1(2 1(2 1(2 13(25) 11(22) 13(25) 14(27) 11(22) 12(24) 14(27) 20(39) 14(27) 21(41) 11 (22) 11(22) 16(31) 23(45) 12 (24) 19(37) 50(98) 12 (24) 14(27) 27(53) 13 (26) 12(24) 30(59) 10 (20) 17(33) 50(98) 14 (28) 19(37) 15 (30) 21(41) 18 (36) 49(96) 27 (54) 29 (58) 48 (96) 5 (10) 5 (10) 5 (10) 6 (12) 6 (12) 6 (12) 7 (14) 7 (14) 7 (14) 8 (16) 7 (14) 9 (18) 9 (18) CTCAE 4 (8) 4 (8) 4 (8) mg 50) = W) Prim MedDRA version Table 14 655 (Safety A /k + g . ary Analy -6.4 ≥ nalysis Set) 10%) Cotherapy Placebo + (Final analysis) (N 10 (20) 17 (33) sis 6 (12) 7 (14) 5 (10) 8 (16) 8 (16) 5 (10) 1 (2) 2 (4) 3 (6) 4 (8) 0 (0) 3 (6) 4 (8) 1 (2) 2 (4) 3 (6) 3 (6) 13.1 fortheprimaryanalysisand = Treatment 51) - emergent A Cotherapy AMG 655 10 (Q2 (N 11 (22) 10(20) 12 (24) 12 (24) 12 (24) 5 (10) 5 (10) 5 (10) 5 (10) 5 (10) 5 (10) 6 (12) 6 (12) 6 (12) 7 (14) 7 (14) 7 (14) 7 (14) 8 (16) 9 (18) mg = W) 51) /k Final Analysis

+ g dverse Events Page 25of4990 version Cotherapy Placebo + (N 11 (22) 19 (37) 10 (20) 7 (14) 5 (10) 8 (16) 9 (18) 8 (16) Page 1of2 1 (2) 3 (6) 3 (6) 4 (8) 1 (2) 4 (8) 1 (2) 3 (6) 3 (6) 3 (6) = 51) 15.0 for

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Note: Adverseeventswerecodedusing Note: ThecotherapyisFOLFIRI. Source analysis andthefinalanalysis. mo 4 events werereportedatsimilarfrequenciesexcept diarrhea (20%,12%),decreasedap treatment armthantheplaceboarm: grade Treatment 76% for The overallincidenceofgrade Grade 16%), andstomatitis(10% inv diarrhea (20% subjects ineithertreatmentarmincludedneutropenia ( 25%), fatigue(22% placebo: 39%),diarrhea( the all- ( conatumumab subjects in either treatment arm) were generally subjects ineithertreatmentarm)were Table Table Dry Hy Anxiety Pain Intestinal Oedema Back Abdominal p Urinar Depression Preferred Term % ofsubjectsfor the finalanalysisandgradedusing estigational productorchemotherapyadministrationwerereportedin8 st frequentlyreported perglycaemia s 3 ≥ : Table14 kin treatment 11 February2013 p ≥ 14- y 11. ain conatumumab 3 Treatment t ract adverse event 6.5, - Ganitumab (AMG479)andConatumumab655) p o em

eripheral bstruction Most ain i ; ergent adverseeventsthatwereconsideredbytheinvestigatortoberelated nfection FA treatment armand90%ofsubjectsintheplacebo 12%), fatigue(10% -6.4 - em u ). Themostfrequentlyreportedtreatment conatumumab pper ; Frequently Reported( (Primary analysis); ergent- 22%), asthenia(25% - emergent AdverseEvents and 57%forplacebo( s werereportedinmorethan5%subjectsthe conatumumab grade 51 ; adverse 20%) %; mg AMG 65510 + Cotherapy ≥ ≥ 39%), nausea( /k 3 treatment (N=50) . 5 (10) 3 treatment CTCAE versus 1 2 (4) 2 (4) 3 (6) 3 (6) 3 (6) 3 (6) 3 (6) 4 (8) 2 (4) g (Q2 ; - petite (8%,0%),andweightdecreased(6%,0%) 6% event presentation:neutropenia( Prim (Safety MedDRA version Table 14 ), and W) . ary ; 0

22%), anemia(24% % ofsubjectsforplacebo.Table Analy - A - -6.4 em ≥ Table 14 em intestinal obstruction( 35 10% nalysis Set) neutropenia ergent adverseeven Cotherapy consistent withthosereportedmostfrequentlyin Placebo + ergent adverseeventsfromboththeprimary (Final analysis) %; sis (N=51) 5 (10) 5 (10) 5 (10) 5 (10) intestinal obstruction 3 (6) 4 (8) 4 (8) 0 (0) 0 (0) 2 (4) ) 13.1 37%) Treatment - conatumumab 6.9, FA for theprimaryanalysisand , vomi - related adverseevents( ( 9 and 29% ). Thosereportedin ; 12%), decreasedappetite(20% ting (24% mg AMG 65510 + Cotherapy - emergent 4%; /k (N=51) 5 (10) 5 (10) ts inthefinalanalysiswas 2 (4) 2 (4) 3 (6) 3 3 (6) 3 (6) 4 (8) 2 (4) : g (Q2 conatumumab (6) 22%, respectively), 29%; , whichwasreportedin 10%). Final Analysis

8 ; % ofsubjectsinthe 24%), alopecia(24% W) placebo: 22%), 12

Ad summarizesthe Four Page 26of4990 verse Events version ≥ 10%of Cotherapy Placebo + : ; allother 2 ≥ (N=51) 5 (10) 5 (10) 5 (10) 5 (10) 7 (14) 8 (16) 5 (10) 5 (10) Page 2of 51%; 0 (0) 1 (2) 0% of 15.0 for

; ;

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Note: Adverseeventswerecodedusing Note: ThecotherapyisFOLFIRI. Source: As ofthefinalanalysis,overallincidencetreatment Serious Treatment conatumumab armthantheplaceboarm:py higher in provided inAppendix Detailed narrativesforsubjectswithseriousadverse events reportedthroughoutthestudyare subjects ineithertreatmentarmfromtheprimaryanalysis andthefinalanalysis. neutropenia (6%,0%).Table ( Table Preferred Term Hy Mucosal Intestinal Anaemia We Dehy Decreased Asthenia Vom Abdominal p Fatigue Diarrhoea Neutropenia adverse event least 1grade3,4,or5 S the finalanalysisandgradedusing ubjects reportingat perglycaemia Table ight iting dration 11 February2013 14- Table 14 d the ecreased i 6.14, Ganitumab (AMG479)andConatumumab655) nflammation o 12. bstruction a conatumumab treatmentarm(43%)thanthe ain pp -6.13 FA etite Most FrequentlyReported( - ). Thefrequencies for 2seriousadverseeventswere>5% higherinthe emergent AdverseEvents (Primary analysis); 5 . 13 Cotherapy AMG 655 10 (Q2 (N 6(2 4(7 9(6 29(57) 39(76) 24(47) 15 (30) 36 (72) summarizes theserious CTCAE 9 (18) 3 (6) 4 (8) 0 ( 1 (2) 2 (4) 2 (4) 3 (6) 4 (8) 3 (6) 4 (8) 4 (8) mg = W) MedDRA version 0) Prim 50) (Safety A /k Table 14 + g Ad . ary verse Events Analy rexia (6% -6.9 ≥ nalysis Set) 5%) Grade3,4,and5 Cotherapy Placebo + (N (Final analysis) sis 5 (10) 9 (18) 1 (2) 3 (6) 2 (4) 2 (4) 4 (8) 2 (4) 0 (0) 0 (0) 3 (6) 2 (4) 2 (4) 13.1 fortheprimaryanalysisand = 51) and - adverse eventsreportedin em placebo treatmentarm(31%) 0%, respectively)andfebrile ergent seriousadverseeventswas Cotherapy AMG 655 10 (Q2 (N 5(9 11(22) 10 (20) 15 (29) 5 (10) 3 (6) 4 (8) 0 (0) 1 (2) 2 (4) 2 (4) 3 (6) 4 (8) 4 (8) 4 (8) mg = W) Treatment 51)

Final Analysis /k + g Page 27of4990 -emergent version Cotherapy Placebo + (N ≥ 5%of 6 (12) 5 1 (2) 3 (6) 3 (6) 3 (6) 3 (6) 0 (0) 0 (0) 3 (6) 2 (4) 3 (6) = (10) 15.0 for 51)

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source Note: Adverseeventswerecodedusing Note: ThecotherapyisFOLFIRI. 36 At thefinalanalysis,atotalof41 Deaths from treatment As ofthefinalanalysis,deathsfor3subjectsin progression. adm sy anem Subject (this eventreportedaftertheprimary ( represents anadditional24 Listing chemotherapy Table in eachtreatmentarm) conatumumab At thetimeoffinalanalysis, theincidenceoftreatment- Treatment treatment considered toberelatedanyprotocol provided inAppendix Detailed narrativesforsubjectswithfataladverseevents r Table Intestinal Dehy Febrile Py Vom Diarrhoea event least 1seriousadverse S Preferred Term the finalanalysisandgradedusing mp ubjects reportingat (71%) subjectsintheplacebotreatment rexia inistration ofconatumumab toms, andn ia iting 14- : Table14 dration 11 February2013 14- 14 who n 6.1, FA - - eutropenia 6.45, PA - em 4.2, PA Ganitumab (AMG479)andConatumumab655) o emergent AdverseEventsLeadingtoStudyWithdrawal orDiscontinuationofTherapy bstruction

was founddeadathome,causeunknown.Thereno ergent adverseeventsintheplacebotreatmentarm. Table . T - and placebotreatmentarms leadingtowithdrawalfromthestudy(2% em -6.14 ). For o otherinformationwasavailable he other2fatal ) . Thedeathwas ergent adverseevents ). Formostofthesesubjects , a (Primary analysis); 13. 5. was conatumumab

Most FrequentlyReported( unchangedfromtheprimary analy deaths and18deaths,respectively,sincetheprimaryanalysis Cotherapy AMG 655 10 (Q2 (N ( 0(0 2(4 2(3 16(31) 22(43) 12(24) 20 (40) 5 (10) Listing 14 ( events CTCAE 3 (6) 1 (2) 4 (8) 3 (6) 2 (4) 80%) mg = W) Serious A considered toberelatedinvestigationalproduct Prim analy MedDRA version 50) /k (Safety A , theevent Table 14 + g withapasthistoryofmassivelivermetastasesand - subjects inthe . specified treatment (co ary of sis). - 6.1, FA arm lorectal cancerandneoplasmmalignant) Analy colo -6.14 dverse Events nalysis Set) T , had rectal cancer, Cotherapy Placebo + a grade was overall, theprimary he event (N . sis ; 0 (0) 0 (0) 0 (0) 2 (4) 3 (6) 2 (4) (Final analysis) Listing 14 T = 13.1 fortheprimaryanalysisand died he deatho 51) conatumumab conatumumab ≥ 5%) in thestudy eported throughoutthestudyare of “dea em . 5 neoplasm There werenofatal - Treatment ergent adverseeventsinthe sis (Table14 6.2, FA death, and ccurred 5daysafterthe th” Cotherapy AMG 655 10 (Q2 (N 5 (10) causeofdeathwasdisease 2 (4) 3 (6) 3 (6) 3 (6) 2 (4) was ( ; Listing14 mg treatment armand = treatment armresulted Listing 14 W) prior complaintof 51) ma -emergent /k Final Analysis

+ reported for neoplasm malignant g - lignant 6.15, PA - - 6.2, FA Page 28of4990 2.2, PA version (serious) Cotherapy Placebo + ; (N [ were not 1 subject and 0 (0) 0 (0) 0 (0) 2 (4) 3 (6) 3 (6) = ), which last ; 51) 15.0 fo ; for ] r

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) ( placebo, theeventwasgrade3metastasestocentralnervoussystem(serious) Date: Product: protocol primary As ofthefinalanalysis,treatment The summaryoftreatment provided inTable An overviewoftheadverseeventsinterestfrom primary occurred in≤ hy Listing 3 protocol [ neutropenia, hypocalcemia,andneoplas loss ofconsciousness decreased, dehydration(2subjects),hypoalbuminemia,hypocalcemia,neoplasmmalignant,and produc similar frequenciesinthe2treatmentarms.Adverseeventsofinterestthatwere am arm ( fi Hy between the By for eacheventofinterest. analyses; forthisclinicalstudyreport,thefocusisonoutputusingnarrowsearchstrategy strategies usingbroadandnarrowsearchtermswereimplementedatboththeprimary hy Adverse eventsofinterestforconatumumab Adverse EventsofInterest congestive cardiacfailure(theaddition protocol neoplasm malignant[ 3 increased, thrombocytopenia,febrileneutropenia, sy these eventsincludedpancytopenia,abdominalpain,diarrhea,metastasestocentralnervous vom conatumumab armincludedfebrileneutropenia,thrombocytopenia,abdominalpain,diarrhea, ( subjects intheplacebotreatmentarm(1additionalsubjectsinceprimaryan conatumumab Listing Listing Table nal analysis,andtheyoccurredathigherfrequenciesinthe (6%) subjectsintheplacebotreatmentarm;theseresultsreprese additional eventssincetheprimary stem poalbuminemia, lossofconsciousness,infusion- pomagnesemia, hyponatremia,infusionreaction,and thetimeoffin pomagnesemia ylase andlipase,hyponatremia, iting, asthenia,death,infusion- hy 11 February2013 t atthefinalanalysis 14- 14 , andcardiacfailurecongestive(theadditionaleventsincetheprimary analysis pom 14- 14 - - - specified therapyinthe specified therapyoccurredin specified therapyintheplacebotreatmentarmwerediarrhea,panc - 6.45, FA - 6.8, PA Ganitumab (AMG479)andConatumumab655) 6.3, FA 6.3, agnesemia conatumumab 2 subjectsineacheventofinterest. treatment arm( FA of 3subjects ; 14. ]). ( ) ). and Listing 14 both oftheseeventswereduetodiseaseprogression al analysis,theoverallfrequencyofadverseeventsinterestwassimilar Listing 14 Adverse eventsleadingtodiscontinuationofinvestigationalproductinthe (the 3additionaleventssincetheprimary : infusion reaction 12 - em % ( increased 18%) andplacebo( - 6.3, FA and in theconatumumabtreatmentarm ergent adverseeventsleadingtodiscontinuationofinvestigational 3 additionalsubjectssince - 6.3, FA conatumumab - em 2 related reaction,blood %, respectively; analy ). Theadverseeventsleadingtodiscontinuationofall ergent adverseeventsleadingtodiscontinuationofall 9 (18 from theprimaryanalysis:11 al eventsincetheprimary ]) venous thromboemboliceventsoccurr were reportedmostoftenintheconatumumab ; theeventsleadingto m ma sis were %) subjectsinthe include increased lignant [ treatme 16% death related reaction,bloodalkalinephosphatase febrile neutropenia,hypocalcemia,and inf ) treatmentarms( Listing 14 usion reaction nt armwere , venous thromboembolicevents. and neoplasmmalignant the primary alkaline phosphataseincreased,weight conatumumab conatumumab analysisandthefinalis serum discontinuation ofall analy - analy 6.3, FA ( T nt anincrease hy able 14 ( analysis)and5(10%) : 22%) subjectsinthe sis werefebrile am pocalcemia, sis 12%, 6%). Table 14 ]) ; intheplaceboarm, yl

[ ). Listing 14 ase andlipase, - arm thantheplacebo ytopenia 6.46, FA treatment armand analysis ed atgenerally alysis) - 6.4 Page 29of4990 grade from the Increased (the 9 - , and , FA 6.3, FA ; andfinal arm 3 ≥ ). Search atthe ]).

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Source Note: Adverseeventswerecodedusing Note: ThecotherapyisFOLFIRI. The incidenceoftreatment Clinical LaboratoryEvaluations Table Change frombaselineinhematologyanalyt Table Change frombaselineinchemistry analysis comparedwiththefinalanalysis.Theresults were ( the the analytes; the finalanalysiswassimilarbetween Table Table Category Event ofInterest events thromboembolic Venous Tre Infusion reaction Hy Hy am Increased s interest adverse eventof Subjects reportingany infusion reaction the finalanalysisandgradedusing conatumumab ponatremia pomagnesemia ylase andlipase atm 14- 14- 15 : Table14 11 February2013 14- ent presents thefrequenciesofgrade 7.17, FA 7.6, 7.2, Table Ganitumab (AMG479)andConatumumab655) - related erum one FA FA -6.62 through ). 14. analy and arm versusplacebo( (Primary analysis); I ncidence ofTreatment Table 14 te Table 14 had a>5% mg 5 (10) 6 (12) - Total 6 2 2 2 6 2 6 2(4) 3(6) 1(2) 3(6) 1(2) 1(2) 1(2) 3 (6) 6 2 2 0 6 2 0 0(0) 0(0) 0(0) 1(2) 1(2) 3(6) 0(0) 0(0) 1(2) 0(0) 1(2) 4(8) 0(0) 1(2) 2(4) 1(2) 3 (6) 3(6) 0(0) 1(2) 0 (0) 3(6) 1(2) 3 (6) em AMG 65510 Cotherapy /k (N 4(8 10(20) 14 (28) ergent grade≥ g (Q2 n (%) - 7.25, FA = CTCAE - Prim analytesforthefinalanalysisaresummarizedin 50) 7.16, FA Grade 4 6 0(0) 3(6) 0 2 (4) 2(4) 1 (2) MedDRA version W)  Narrow SearchTerms (Safety A higher incidenceofgrade Table 14 3 ary + ly . m through oauua andplacebotreatmentarmsformostof conatumumab Analy phocy es forthefinalanaly ≥ and Total 3 chemistryandhematologyvaluesfortheprimary 3 chemistryandhematolo -6.4 Cotherapy - Placebo + nalysis Set) emergent A (N sis Table 14 tes n (%) 9 Table 14 = (Final analysis) : 13.1 forth 51) Grade 14%  (0) 3 - 7.18, FA - and 7.37, FA consistent betweenthe2 dverse EventsofInterest e primaryanalysisand mg 6 (12) 6 (12) Total sis aresummarizedin 8 ≥ AMG 655 %, respectively) Cotherapy 3 treatment /k (N 9 (18) through g (Q2 n (%) . = 51) Grade gy Final Analysis 4 6 0(0) 3(6) 0(0) 2 (4) 1(2) 1 (2) W)

 10 laboratoryvaluesasof 3 Table 14 + - emergent valuesin Page 30of4990 Total version Cotherapy Placebo + (N - 7.24, FA 8 (16) analy n (%) = 51) 15.0 for Grade ses.  3 .

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Note: BasedonNCI Note: ThecotherapyisFOLFIRI. exhibited abnormalitiesthatmetthesecriteriaondifferent days( At Source: ALT =alanineaminotransferase;ASTaspartateNCINationalCancerInstitute; Listing of extensivemetastaticdiseaseincludingliver,ly One subjectinthe liver analyteabnormalitiesthatmetthecriteriaforHy’s Lawonthesameday( ( during thestudy, × 2 ≤ grade point. T criteria atmultipletimepointsduringthestudy serious adverseeventofgrade 4febrileneutropeniaonday736 Listing Category Chemistry Hematology CTCAE =CommonTerminologyCriteriaforAdverseEvents. the Wh Platelets Ly Hemoglobin coun Absolute neutrophil Total bilirubin Sodium Potassium Phosphorus Magnesium Lipase Glucose Calcium Bicarbonate AST Am phosphatase Alkaline Albumin ALT Analyte ULN atmosttimepoints. 3) final analysis,nosubjectsineitherthe 11 February2013 14.4.2, Table 14 mp 14 he highesttotalbilirubinforthissubjectoccurredonstudy day7 , an ite bloodcells yl t ase - Ganitumab (AMG479)andConatumumab655) hoc 6.4, FA d Table the highestASToccurredon day727(99U/L,grade ytes PA). Valuesfortotalbilirub 7.80 -7.79 andTable14- none -CTCAE Version3 ). conatumumab 15. This subject was treated for 46 cycles and was discontinuedduetoa This subjectwastreatedfor 46cyclesandwas of whichwereconsideredto berelatedtoinvestigationalproduct IncidenceofTreatment Cotherapy AMG 655 10 (Q2 Multiple adverseeventsof (N 4(8 10(20) 14 (28) 5 (10) 8 (16) 1 (2) 2 (4) 2 (4) 0 (0) 4 (8) 2 (4) 1 (2) 4 (8) 0 (0) 1 (2) 0 (0) 2 (4) 2 (4) 1 (2) 1 (2) 2 (4) treatment arm,Subject mg .0. = W) Prim 50) /k (Safety A (Primary analysis); + g ary in andalkalinephosphataseeachmet Analy conatumumab , andASTmettheindividualcriteriaatonetime nalysis Set) Cotherapy Placebo + mp (N - sis emergent NCICTCA 4 (8) 1 (2) 3 (6) 2 (4) 1 (2) 0 (0) 2 (4) 0 (0) 0 (0) 2 (4) 3 (6) 0 (0) 0 (0) 0 (0) 0 (0) 2 (4) 0 (0) 0 ( h node, = 0) 51) Table 14 blood bilirubinincreased and or placebotreatmentarmsexhibited - 7.2 Listing 14 lung parenchy ( Listing 14 Cotherapy AMG 655 10 (Q2 2) (N (Final analysis) 4(7 12(24) 14 (27) 6 (12) 7 (14) 6 , 1 (2) 2 (4) 3 (6) 0 (0) 2 (4) 1 (2) 4 (8) 0 (0) 1 (2) 0 (0) 2 (4) 2 (4) 2 (4) 1 (2) 2 (4) . mg

with apriormedicalhistor = ( W) 12) Alkaline phosphatasewas 27 51) E Grade /k Final Analysis + -

g 7.3, FA ( - 82.1 6.4, FA Table 14 ma the individual µ 3 ≥ l m Page 31of4990 ; mo ). Cotherapy were reported Placebo + (N etastases, l/L, 4 (8) 4 (8) 1 (2) 3 (6) 1 (2) 1 (2) 4 (8) 1 (2) 0 (0) 2 (4) 3 ( 0 (0) 0 (0) 1 (2) 0 (0) 3 (6) 0 (0) 0 (0) - 7.40, FA = 6) 51) y ).

.

Approved

placebo treatmentarmtestedpositiveforpre treatment arm,5(10%)subjectsintheconatumumaband7(15%)in were similartothosedescribedintheprimary Synopsis ClinicalStudyReport:20060579(FinalAnalysis) ( Final analysisresults Ganitumab Antibody results Antibody Results Other Evaluations Date: Product: ( Addition pre assessed byC Table Ganitumab end Ganitumab conatumumab, irinotecan,SN ganitumab Pharmacokinetics resultsforirinotecan,itsmetaboliteSN Pharmacokinetics resultsforganitumabandconatumumab for pre conatumumab treatmentarmand2(4%)subjectsintheganitumabt described intheprimaryanalysis:2(4%)subjects antibodies weresimilartothose Final analysisresultsoftheincidencesubjectstestingpositiveforanti Co cro with placebo,despiteneverhavingreceivedadoseofganitumab,arelikelytorepresent anti Overall, noneoftheseadverseeventswerelikelytohaveresultedfromthepresence day 1).On124,thesubjectexperiencedneutropenia;nofurthereventswerereported. subject experiencedfatigue.Sub follow Subject this time,adverseeventsincludedhypot (ganitumab) testedpositivefortheanti antibodies (non placebo treatmentarmwhowerenegativeatbaselinetestedpositivefor cy Pharmacokinetic Results experienced bythis me anti (cy positive for conatumumab treatmentarm Table Appendix 7 cle 1an cle 5 natumumab ss dical reviewoftheadverseeventsforthissubjectindicatedthat - - - infusion concentration(C ganitumab conatumumab antibodies - - reactive 16. up visit).Onday - 11 February2013 14- existing anti , al detailsareprovidedinthe day d cycle3were186and227 8.2, Ganitumab (AMG479)andConatumumab655) , Wi anti conatumumab ). 1)( th theQ2 FA antibodies thatarepresentasaconsequenceofmolecularmimicry. max - - - antibodies. conatumumab antibodiesandneutralizingpostbaseline of neutralizing) duringthetreatmentperiod( below arepresentedseparatelyforganitumabandconatumab Listing 14 ). In , betweency - infusion (C - : (placebo)testedpositiveonday113(cy conatumumab subject wereunlikely of theincidencesubjectstestingpositiveforanti W addition, 1 113, adverseeventsincludedmucositisandnausea.Onday48,the regimen, themeanend , andplacebogroups -

7.5, FA Anti min max (Subject - cles (s 38, and5 ) forcy - ection 11.3.10of ) andpre ganitumab antibodiesdetectedatbaselineorduringtreatment ject subject intheganitumabtreatmentarmand2subjects ) 1 and3,wasapproximately1.2 antibodies ( . As describedintheprimaryanalysisclinicalstudyreport,a pharmacokinetics clinicalstudyreport cles 3,5,9,13,and17rangedfrom29.9to54.8μg/mL. μg/mL, respectively. - ganitumab antibodiesonday51(safetyfollow - ension, anemia,andhypoalbuminemia. FU) arepresentedin - to havebeena infusion (C - analysis:4(8%)subjectsintheganitumab existing anti Table 14 .

- (placebo)te Individual concentrationdata(ganitumab, of the primaryanalysisclinicalstudyreport) ) - who testednegativeatbaseline infusion (C min ) valuesthroughcycle17arelistedin - 8.1, FA - result ofthepresence - Listing 14 38, and5 ganitumab antibodies are presentedsepar The accumulationofganitumab,as cle 9 sted positiveonday134(cycle Append max ). Inaddition,1 , ) concentrationsobservedin day - fold inthisstudy. - - 7.5, FA FU are reatment armtestedpositive 1)andday154(safety ix 7 - - anti ganitumab antibodies conatumumab .

c ): Subject presented - ontribution ganitumab . subject inthe ately Page 32of4990 on day tested below. - The mean of up for visit). At 93 .

9,

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: ( Pre = Source: PKS/ Pre = belowthelimitofquantificationforconatumumab(0.0299μg/mL)weresettozero. Concentrations Additional detailsareprovidedinthe (C of steady state(cy 178 and216μg/mL,respectively. Table Conatumumab end Conatumumab Source: PKS/Study2006 Note: Allconcentrationsarereportedinμg/mL. Note: Allconcentrationsarereportedinμg/mL Appendix 7 Range SD Mean N Range SD Mean N Concentrations A (Version 3) (Version 3) min C Table max ) concentrationobservedincycles3,5,9,13,and17rangedfrom77.0to136μg/mL. p p Table 17. re re 11 February2013 between cycles1and3 Concentrations A -infusion; EOI=endofinfusion;SDstandarddeviation. -infusion; EOI=end ofinfusion;SD=standarddeviation. Ganitumab (AMG479)andConatumumab655) The meanend 17. (0.00 - (0.00 - ). Study20060579 finalanalysismapping 102) 141) 474. 13.8 49.0 14.7 2.20 046. 846. 953. 8351.6 38.3 36.6 29.5 62.1 28.4 63.3 20.4 2.94 Pre Pre 16. 84 23 71 110 11 18 27 34 32 40 48 84 92 61 13 14 26 28 29 41 48 DescriptiveStatisticsofConatumumab cle 5)cannotbeassesseddueto Cy Cy DescriptiveStatisticsofGanitumabEnd cle 1 cle 1 - of fter (54.3 - - 0579 finalanalysismapping/BaseScenario/DescStatsw/excl/AMG479/655 (101 - infusion (C 298) 355) EOI EOI 186 178 Intravenous Every 2WeeksinCombinationWithFOLFIRI - of fter - infusion (C Intravenous was approximately1.2 (5.43 - (18.4 - 64.1) 144) 29.9 77.0 in CombinationWithFOLFIRI Pre Pre max The accumulationof ) andpredose(C Cy Cy pharmacokinetics clinicalstudyreport Ad cle 3 cle 3 max ministration of12 . (28.7 - ) concentration (112 - 698) 345) EOI 227 107 EOI 216 Ad /Base S ministration of1 the Cy Cy (6.56 - (49.8 - min preset samplingscheme.Theaccumulation 79.7) 494. 8454.8 23.1 48.4 21.1 41.0 18.2 34.9 145) 94.0 cenario/Desc Statsw/excl/AMG479/655 - r r r Pre Pre Pre Pre r r r Pre Pre Pre Pre fold inthisst cle 5 cle 5 ) valuesthroughcycle17arelistedin conatumumab bevdi yl n yl 3were s observedincycle1and End mg/kg Ganitumab - of - of - Cy Cy infusion andPre (3.26 - (34.0 - 0 91.7) - 193) 2 3 136 130 120 infusion andPre mg/kg cle 9 cle 9 udy between cycl . The meanpre

Conatumumab c Cy Cy ontribution (12.2 - (67.4 - 86.8) 182) cle 13 cle 13 Ev Page 33of4990 - infusion ery 2Weeks 1 e - infusion - Cy Cy and infusion (23.0 - (87.2 - 96.3) 231) 24.4 cle 17 cle 17

7

.

Approved

Synopsis ClinicalStudyReport:20060579(FinalAnalysis) Date: Product: Reference List: clinical studyreport. final analys observed withthecombinationofganitumabandFOLFIRIorconatumumab over FOLFIRIalone. second Based onthefinalstudyresults,additi Following IVinfusionof180mg/m Irinotecan and5 (GAN) +F Cohn AL,TaberneroJ,Maureletal.Conatumumab(CON)+FOLFIRI(F)orGanitumab 17, treatment groups, 1950 (m (ie, me cy groups, respectively. 320, 287,and307ng/mLincy The median5 conatumumab, andplacebogroups Conclusions: ( Appendix cle 1and361,441,389ng/mLi CRC) [poster].ASCOGastrointestinalCancersSymposium,2012.Abstract534. 400 an ( SN (626), 1910(388),and1860(522)ng/mLinganitumab,conatumumab,placebo

11 February2013 - Additional detailsareprovidedinthe SD) AUC - (3610), and14 38) hadmean(SD)C line treatmentinsubjectswithmutant Ganitumab (AMG479)andConatumumab655) 7 for Second e ). s - in FU concentrationvaluesat3hourswere34 in

t - his studyaregenerallyconsistentwiththose f FU respectively. values of374(126),340(121),and432(149)ng•hr/mLintheganitumab, Based onthefinalanaly - The medianconcentrationsat24hourswere449,400,and466 , line TreatmentofMutant(MT)KRASMetastaticColorectalCancer 100 (3120)ng•hr/mL,respectively. max cle 3fortheganitumab,conatumumab,andplacebotreatment Mean (SD)irinotecanAUC values of23.6(9.51),21.0(13.0),and27.1(9.13)ng/mL 2 irinotecan, mean(SD)irinotecanC , respectively n cycle3forthe on ofganitumaborconatumumabtoFOLFIRIas pharmacokinetics clinicalstudyreportc sis safetydata,nonewsignals KRAS . mCRC 3 treatment groups,respectively. 3, 282,and374ng/mLincycle Th in did notappeartoimprovePFSorOS f values were13 e metaboliteofirinotecan drawn in Theover max the primaryanalysis all conclusionsfrom

values were , 500 Page 34of4990 (3390), ontribution were ng/mL in . 1 and

the

.

Approved